Document zzopKJ61BKazkn57xdoNQqaqB

RECEIVED JUN 2 8 1982 RESEARCH TECHNIQUES AND METHODS FOR THE DETECTION AND PREVENTION OF CARCINOGENESIS IN THE INDUSTRIAL WORKER GRANT REFERENCE NO, VC 7,0 UNIVERSITY OF LOUISVILLE HEALTH SCIENCES CENTER SCHOOL OF MEDICINE GRADUATE SCHOOL CANCER CENTER REPORT FOR THE CHEMICAL MANUFACTURERS ASSOCIATION (FORMERLY MANUFACTURING CHEMISTS ASSOCIATION) CMA 003421 INVESTIGATORS CARLO H. TAMBURRO, M.D., PRINCIPAL INVESTIGATOR CHARLES E. KUPCHELLA, PH.D., CO-PRINCIPAL INVESTIGATOR JOHN L. WONG, PH.D., CO-PRINCIPAL INVESTIGATOR PROGRAM INVESTIGATORS G.H. BARROWS, M.D. J. T. DU, PH.D. E. ESPINOSA, M.D. R.C. FELDHOFF, PH.D. H.P. FORTWENGLER, M.S. G.R. SCHRODT, M.D. G. SONNENFELD, PH.D. U.N. STREIPS, PH.D. M.T. TSENG, PH.D. W.O. WADDELL, M.D. 11 CMA 003423 CMA 003424 CONTRIBUTING INVESTIGATORS G.H. BARROWS, M.D., ASSISTANT PROFESSOR DEPARTMENT OF PATHOLOGY JOHN L. CREECH, JR., M.D., CLINICAL ASSISTANT PROFESSOR DEPARTMENTS OF SURGERY AND COMMUNITY HEALTH MEDICAL DEPARTMENT, B.F. GOODRICH COMPANY J. T. DU, PH.D., RESEARCH ASSOCIATE LIVER RESEARCH CENTER DEPARTMENT OF MEDICINE E. ESPINOSA, M.D., PROFESSOR DEPARTMENT OF PATHOLOGY R.C. FELDHOFF, PH.D., ASSISTANT PROFESSOR DEPARTMENT OF BIOCHEMISTRY H.P. FORTWENGLER, JR., M.S., RESEARCH ASSOCIATE LIVER RESEARCH CENTER DEPARTMENT OF MEDICINE RICHARD A. GREENBERG, PH.D., PROFESSOR AND ACTING CHAIRMAN DEPARTMENT OF COMMUNITY HEALTH CHARLES E. KUPCHELLA, PH.D., ASSOCIATE PROFESSOR ASSOCIATE DIRECTOR OF CANCER CENTER; ^PROFESSOR AND CHAIRMAN DEPARTMENT OF BIOLOGICAL SCIENCES MURRAY STATE UNIVERSITY GARY LISS, M.D., M.P.H., CLINICAL INSTRUCTOR DEPARTMENT OF COMMUNITY HEALTH NIOSH/ROBERT A. TAFT LABORATORY; CINCINNATI, OHIO CAROLYN MARLOWE, B.A., RESEARCH ASSOCIATE DEPARTMENT OF PHARMACOLOGY & TOXICOLOGY ^presently 11 i CMA 003425 CONTRIBUTING INVESTIGATORS (Continued) LASLO MAKK, M.D., CLINICAL INSTRUCTOR DEPARTMENTS OF PATHOLOGY AND COMMUNITY HEALTH DIRECTOR, DEPARTMENT OF PATHOLOGY ST. ANTHONY HOSPITAL HANS POPPER, M.D., PROFESSOR STRATTON RESEARCH LABORATORY FOR THE STUDY OF LIVER DISEASE MOUNT SINAI SCHOOL OF MEDICINE NEW YORK, NEW YORK JOHN P. SANDOZ, M.S., INSTRUCTOR DEPARTMENT OF COMMUNITY HEALTH G.R. SCHRODT, M.D., PROFESSOR AND CHAIRMAN DEPARTMENT OF PATHOLOGY G. SONNENFELD, PH.D., ASSISTANT PROFESSOR DEPARTMENT OF MICROBIOLOGY & IMMUNOLOGY U.N. STREIPS, PH.D., ASSOCIATE PROFESSOR DEPARTMENT OF MICROBIOLOGY & IMMUNOLOGY C. H. TAMBURRO, M.D., PROFESSOR LIVER RESEARCH CENTER DEPARTMENTS OF MEDICINE AND COMMUNITY HEALTH M.T. TSENG, PH.D., ASSOCIATE PROFESSOR DEPARTMENT OF ANATOMY W.J. WADDELL, M.D., PROFESSOR AND CHAIRMAN DEPARTMENT OF PHARMACOLOGY AND TOXICOLOGY RAYA WARICK, PH.D., RESEARCH ASSOCIATE CANCER CENTER JOHN L. WONG, PH.D., PROFESSOR, DEPARTMENT OF CHEMISTRY TV CMA 003426 CONTRIBUTING INVESTIGATORS (Continued) LASLO MAKK, M.O., CLINICAL INSTRUCTOR DEPARTMENTS OF PATHOLOGY AND COMMUNITY HEALTH DIRECTOR, DEPARTMENT OF PATHOLOGY ST. ANTHONY HOSPITAL HANS POPPER, M.O., PROFESSOR STRATTON RESEARCH LABORATORY FOR THE STUDY OF LIVER DISEASE MOUNT SINAI SCHOOL OF MEDICINE NEW YORK, NEW YORK JOHN P. SANDOZ, M.S., INSTRUCTOR DEPARTMENT OF COMMUNITY HEALTH G.R. SCHRODT, M.D., PROFESSOR AND CHAIRMAN DEPARTMENT OF PATHOLOGY G. SONNENFELD, PH.D., ASSISTANT PROFESSOR DEPARTMENT OF MICROBIOLOGY & IMMUNOLOGY U.N. STREIPS, PH.D., ASSOCIATE PROFESSOR DEPARTMENT OF MICROBIOLOGY & IMMUNOLOGY C. H. TAMBURRO, M.D., PROFESSOR LIVER RESEARCH CENTER DEPARTMENTS OF MEDICINE AND COMMUNITY HEALTH M.T. TSENG, PH.D., ASSOCIATE PROFESSOR DEPARTMENT OF ANATOMY W.J. WADDELL, M.D., PROFESSOR AND CHAIRMAN DEPARTMENT OF PHARMACOLOGY AND TOXICOLOGY RAYA WARICK, PH.D., RESEARCH ASSOCIATE CANCER CENTER JOHN L. WONG, PH.D., PROFESSOR, DEPARTMENT OF CHEMISTRY iv CMA 03427 RESEARCH TECHNIQUES AND METHODS FOR THE DETECTION AND PREVENTION OF CARCINOGENESIS IN THE INDUSTRIAL WORKER Program Final Report Table of Contents Investigators ................................................................................................................. ii Contributors ...................................................................................................... iii I. Introduction .................................................................................................................... 1 II. Summaries of Research Programs ............................................................................. 3 III. Research Programs and Results ............................................................................. 16 A. Studies of Human Immunological Systems in the Detection of Vinyl Chloride and Other Chemical Injury; Investigators - H.P. Fortwengler, Jr., and C.H. Tamburro..................................................................... 17 Al. Evaluation of Immunocompetence of Workers Chronically Exposed to Vinyl Chloride .................................................................... 17 A2, Assessment of the Leukocyte Adherence Inhibition Test for the Detection of Angiosarcoma of the Liver....................................... 21 A3. Assessment of Vinyl Chloride-Induced Angiosarcoma Antigen for the Detection of Latent Angiosarcoma in Exposed Workers ......................................................................................... 22 A4. Study of Human Leukocyte Antigen (HLA) Frequencies in Chemically-Exposed Workers .................................................................... 25 AS. Identification of the Endothelial Cell as the Cell of Origin for Vinyl Chloride-Induced Angiosarcoma of the Liver.................................................................................................................... 30 B. Study of Hepatic Biochemical and Enzymatic Systems for the Identification of Vinyl Chloride Chemical Injury and Cancer Development: Animal and Human Studies; Investigators - J.T. Du, M.T. Tseng, and C.H. Tamburro.................................................................... 34 v CMA 003428 ANIMAL STUDIES: Bl. Characterization of Hepatic Enzyme Changes in Rats with Prolonged Vinyl Chloride Exposure: Decreased Glucose-6-Phosphatase Activity ............................................................ 34 B2. Morphological Alterations in Livers of Rats Exposed to Vinyl Chloride................................................................................................................ 34 B3. Alterations in the Oxidizing and Detoxifying Systems of Rat Liver After Prolonged Exposure to Vinyl Chloride ..................... 38 B4. Oxidative and Detoxifying Ability of Liver Mesenchymal versus Parenchymal Cells in the Metabolism of Xenobiotics ................. 41 HUMAN STUDIES: B5. The Effectiveness of Indocyanine Green (ICG) Clearances in the Detection of Liver Injury.............................................................43 B6. The Assessment of Bile Acids versus Indocyanine Green (ICG) Clearances in the Detection of Chemical Liver Injury Chemical in Exposed Humans.......................................................................................... 46 Study of Glycosaminoglycan Changes in the Detection of Hepatic Fibrotic Injury in Chemical Exposure and Hepatic Cancer Development; Investigator - C. E. Kupchella and R. Warick ... 51 ANIMAL STUDIES: Cl. The Study of Glycosaminoglycan Changes in Livers of Animals Bearing Metastasizing Hepatomas ....................................................... 51 C2. Significance of Tissue Glycosaminoglycan Elevations in Hepatic Necrosis and Regeneration ................................................... 51 HUMAN STUDIES: C3. Characterization of Glycosaminoglycan Patterns in Humans with Angiosarcoma.......................................................................................... 51 C4. Evaluation of Glycosaminoglycan Analysis of Urines of Chemically-Exposed Workers .................................................................... 51 C5. Characterization of Glycosaminoglycan Patterns in the Identification of Chemical and Nonchemical Injury of the Liver................................................................ 51 vi CMA 003429 ANIMAL STUDIES: 31. Characterization of Hepatic Enzyme Changes in Rats with Prolonged Vinv1 Color ice E cosure; Cec^easao G1ucose-6-Pnospnatase activity ........................................................... 32. Morphological Alterations in Livers of Rats Exposed to Vinyl Chloride .......................................................................................................... 34 B3. Alterations in the Oxidizing and Detoxifying Systems of Rat Liver After Prolonged Exposure to Vinyl Chloride ..................... 33 B4. Oxidative and Detoxifying Ability of Liver Mesenchymal versus Parenchymal Cells in the Metabolism of Xenobiotics ................. 4] HUMAN STUDIES: B5. The Effectiveness of Indocyanine Green (ICG) Clearances in the Detection of Liver Injury ....................................................... 43 86. The Assessment of Bile Acids versus Indocyanine Green (ICG) Clearances in the Detection of Chemical Liver Injury Chemical in Exposed Humans ..................................................................................... 46 C. Study of Glycosaminoglycan Changes in the Detection of Hepatic Fibrotic Injury in Chemical Exposure and Hepatic Cancer Development; Investigator - C. E. Kupchella and R. Warick . . . 51 ANIMAL STUDIES: Cl. The Study of Glycosaminoglycan Changes in Livers of Animals Bearing Metastasizing Hepatomas ....................................................... 51 C2. Significance of Tissue Glycosaminoglycan Elevations in Hepatic Necrosis and Regeneration ................................................... 51 HUMAN STUOIES: C3. Characterization of Glycosaminoglycan Patterns in Humans with Angiosarcoma ..................................................................................... 51 C4. Evaluation of Glycosaminoglycan Analysis of Urines of Chemically-Exposed Workers .................................................................... 51 C5. Characterization of Glycosaminoglycan Patterns In the Identification of Chemical and Nonchemical Injury of the Liver ............................................................ .................................................. 51 CMA 003430 0. Histologic and Morphometric Analysis: A Means of Assessing Hepatic Injury in Chemical Workers; Investigators - G.H. Barrows, G.R. Schrodt, and C.H. Tamburro.................................. 56 01. Morphometric Assessment of Histological Lesions Character istic of Vinyl Chloride Injury ......................................................... 56 D2. Computer-Assisted Morphometric Analysis as a Rapid Means of Determining Collagen Content ................................................................ 56 D3. Development and Assessment of the Morphometric Method of Analysis of Collagen Content from Human Liver Biopsies Relationship to Age...................................................................................... 56 D4. Light Microscopic Assessment of Various Histological Lesions Found in Liver Biopsy Tissue Obtained from Vinyl Chloride Workers .................................................................................. 56 E. Studies of Vinyl Monomer Chemicals and Their Metabolites Using Chemical Structure and Synthesis in the Determination of Toxicity of These Agents; Investigator - J.L. Wong .......................... 67 El. Synthesis, Purification, and Utilization of Vinyl Chloride Metabolites in Mutagenicity and Carcinogenicity Studies . . 68 E2. Detoxification Studies of Vinyl Chloride and Its Metabolites....................................................................................................... 70 E3. Vinyl Chloride MetaboliteDetection--Chloroacetic Acid ... 72 F. Assessment of Assays for the Carcinogenic Potential of Industrial Chemicals Using Prokaryotic and Eukaryotic Systems; Investigators - U.N. Streipsand G. Sonnenfeld......................................... 74 FI. Further Development of Bacterial Systems for Testing Carcinogenicity and Mutagenicity of Chemical Agents Used in the Manufacturing of Vinyl Chloride and Synthetic Rubber....................................................................................................................74 F2. Preliminary Studies on the Use of Interferon Induction as an Indicator of Mutagenicity and Carcinogenicity of Chemicals........................................................................................................... 76 G. The Study of Liver Tissue Antigens and Antibodies in the Detection of Vinyl Chloride Injury; Investigator E. Espinosa............................................................................................................... 80 Gl. The Study of Tissue Antigens From Liver Tumors, Angiosar coma, and Hepatomas......................................................................................80 vi i CMA 003431 G2. Circulating Antigens and Autoantibodies in Vinyl ChlorideAssociated Liver Disease......................................................................... H. Use of Isolated Mammalian Liver Cells for the Study of Chemical Monomer Metabolism; Investigator - R.C. Feldhoff .............................. 87 HI. Isolation of Mammalian Liver Cells for the Study of the Metabolism of Chemical Monomers ........................................................ 87 H2. The In Vitro Study of Albumin Synthesis by Liver from Human Biopsies................................................................................................... 87 I. The Study of Tissue Disposition of Industrial Chemicals: The Vinyl Chloride Example; Investigator - W.J. Waddell and C. Marlowe............................................................................................................92 II. The Use of Whole Body Autoradiography in the Specific Tissues. Localization of Accumulated and Retained Industrial Chemicals and their Metabolites .... 92 IV. Lists of Publications, Abstracts, Preprints, and Publications in Preparation.........................................................................................99 A. Publications.....................................................'...................................................100 B. Abstracts.................................................................................................................. 102 C. Preprints ............................................................................................................... D. Publications in Preparation Titles ............................................................ 105 V. Appendix (copies of IV., A, B, C)...........................................................................106 vi i i 003432 G2. Circulating Antigens and Autoantibodies in Vinyl ChlorideAssociated Liver Disease............................................................................. 0 H. Use of Isolated Mammalian Liver Cells for the Study of Chemical Monomer Metabolism; Investigator - R.C.reldhoff ................................. 7 Hi. Isolation of Mammalian Liver Cells for the Study of the Metabolism of Chemical Monomers ................................................... 87 H2. Tne In Vitro Study of Albumin Synthesis by Liver from Human Biopsies.................................................................................................. S7 I. The Study of Tissue Disposition of Industrial Chemicals: The Vinyl Chloride Example; Investigator - W.J. Waddell ana C. Marlowe........................................................................................................... 92 II. The Use of Whole Body Autoradiography in the Specific Tissues. Localization of Accumulated and Retained Industrial Chemicals andtheirMetabolites .... 92 IV. Lists of Publications, Abstracts, Preprints, and Publications in Preparation.......................................1..................................................................... 99 A. Publications................................................... ;.................................................. iqo B. Abstracts..................................................................................................................102 C. Preprints.................................................................................................................. 104 D. Publications in Preparation Titles ............................................................. 105 V. Appendix (copies of IV., A, B, C) .................................................................... 106 vi i i CMA 003433 INTRODUCTION In 1974 Dr. John L. Creech, Jr., a surgeon and the plant physician for B.F. Goodrich's Chemical Plant on Bells Lane in Louisville, Kentucky, identi fied the increased occurrence of hepatic angiosarcoma in workers involved in polyvinyl chloride manufacturing. This discovery led to the implementation of an industrial cancer control, detection and prevention program at the Louis ville B.F. Goodrich Plant involving approximately 1,800 workers: 1,200 were active employees and 600 were previously employed individuals. The University of Louisville, in cooperation with the B.F. Goodrich Company, developed and implemented the program and has continued to follow this cohort of workers clinically and epidemiologically. A proposal submitted to the Chemical Manufacturers Association, formerly the Manufacturing Chemists Association, to foster scientific research in better technical methods for the detection and prevention of chemical injury and carcinogenesis in industrial workers was awarded in 1976. These investi gations covered nine general areas of interest and involved both animal and human studies. The studies included: (A) the use of the human immunological system for the detection of injury from vinyl chloride and other chemicals, (B) the assessment of hepatic biochemical enzymatic systems to detect and characterize vinyl chloride and other chemical injury, (C) the use of tissue and urinary glycosaminoglycan changes for early detection and diagnosis of chemical injury including cancer development, (D) the histological evaluation of liver tissue from chemical workers and the assessment of its collagen content as an indicator of chronic latent injury, (E) synthesis and analysis of vinyl chloride intermediate and end products to further understand its biological metabolism and removal, (F) the identification and assessment of cell assays to determine the carcinogenic potential of industrial chemicals, (G) the evaluation of tissue antigenic systems for detecting chemicallyinduced carcinogenesis, (H) exploratory studies for the use of isolated human liver cells to determine individual ability to detoxify chemicals, and (I) the further assessment of vinyl chloride metabolism by the use of whole body autoradiographic technique. Each of these areas of study were directed toward four major objectives: (1) to better characterize and understand the pathogenesis of chemicallyinduced injury, especially vinyl chloride injury, (2) to characterize the mechanisms of chemical injury and the body's biological defense to this injury using animal studies, (3) to evaluate and assess old and new biochemical parameters for the detection and verification of chemical injury in humans, and (4) to develop new techniques that could be applied to the problem of identifying occupationally-related chemical injury for purposes of preven tion. Many of the original objectives of the individual studies have been changed or modified because of developments and discoveries in both the animal 1 CMA 003434 2 program and the human medical surveillance program which were ongoing concomitantly during the research periods of this study. This has led to a large body of literature which has changed the medical approach to the screening, identification and verification of occupationally-related chemical injury. Although the contractual funding of this grant was only three years, each year was funded separately with varying time intervals between the funding from year to year so that almost six years have evolved between the initiation of these studies and this final report. This report, therefore, summarizes the work accomplished during this six-year interval and the related work which is still ongoing. The investigators' reports vary in the degree of detail depending upon whether their work is in the process of being published or has already been published. Work not yet published is described in greater detail; published work is accompanied by a copy of the actual article(s) attached in the Appendix. The investigators wish to express their appreciation to the B.F. Goodrich Company and especially to the entire management of its Bells Lane, Louisville plant. The continued cooperation and help of the plant managers, Philip Lawrence, Gabriel Le Febvre, Edward L. Beeler and especially Raymond Pruitt, Personnel Manager, along with Dr. John Creech, Jr., and the medical department staff, requires special acknowledgement. The investigators also wish to thank their laboratory and research staff for their devoted work, and to thank Dr. Harold Boyer, Vice President for Health Affairs, and Dr. Joseph X. Musacchia, Dean of the Graduate School, for their administrative efforts, and especially, Mrs. Vicky Strong for her patient secretarial support. A special thanks is given to Dr. Kathleen O'Connell without whose help and support completion of the periodic and final reports would have been much more difficult. Finally, and very importantly, the participation and cooperation of the entire work force of the B.F. Goodrich 8ells Lane Plant, and their representative union$--the Distillery Workers Union, the International Brotherhood of Electrical Workers #369, the International Association of Machinist Workers #681, and Pipefitters Union Local #522--is acknowledged with appreciation. This work would have been impossible without the active involvement and help of the workers. The objectives of this task were all directed toward improving and maintaining the best possible work environment. Again, a thank you to all who have contributed to this effort. 003435 2 program ana me i'.u'an r a a i c a 1 surveillance program which were ongoing concomitantly curing me research periods of this study. This has led to a 1orge body of lite-ature which has changed the medical approach to the screening, identification and verification of occupationally-related chemical injury. Although the contractual funding of this grant was only three years, each year was funded separately with varying time intervals between the funding from year to year so that almost six years have evolved between the initiation of these studies and this final report. This report, therefore, summarizes the work accomplished during this six-year interval and the related work which is still ongoing. The investigators' reports vary in the degree of detail depending upon wnether their work is in tne process of being published or has already been published. Work not yet published is described in greater detail; published work is accompanied by a copy of the actual article(s) attached in the Appendix. The investigators wish to express their appreciation to the B.F. Goodrich Company and especially to the entire management of its Bells Lane, Louisville plant. The continued cooperation and help of the plant managers, Philip Lawrence, Gabriel Le Febvre, Edward L. Beeler and especially Raymond Pruitt, Personnel Manager, along with Dr. John Creech, Jr., and the medical department staff, requires special acknowledgement. The investigators also wish to thank their laboratory and research staff for their devoted work, and to thank Dr. Harold Boyer, Vice President for Health Affairs, and Dr. Joseph X. Musacchia, Dean of the Graduate School, for their administrative efforts, and especially, Mrs. Vicky Strong for her patient secretarial support. A special thanks is given to Dr. Kathleen O'Connell without whose help and support completion of the periodic and final reports would have been much more difficult. Finally, and very importantly, the participation and cooperation of the entire work force of the B.F. Goodrich Bells Lane Plant, and their representative unions--the Distillery Workers Union, the International Brotherhood of Electrical Workers #369, the International Association of Machinist Workers #681, and Pipefitters Union Local #522--is acknowledged with appreciation. This work would have been Impossible without the active involvement and help of the workers. The objectives of this task were all directed toward Improving and maintaining the best possible work environment. Again, a thank you to all who have contributed to this effort. CMA 003436 SUMMARIES OF RESEARCH PROGRAMS 3 CMA 003437 PROGRAM A STUDIES OF HUMAN IMMUNOLOGICAL SYSTEMS IN THE DETECTION OF VINYL CHLORIDE AND OTHER CHEMICAL INJURY; Investigators - H. P. Fortwengler, Jr., and C. H. Tamburro Al. Evaluation of Immunocompetence of Humans Chronically Exposed to Vinyl Chloride The scientific literature is replete with the demonstrations of immunodepression as a manifestation of medical disease, especially cancer. Lymphocytes (T cells) can be cytotoxic to human tumor cells and are often found decreased or poorly functioning in cancer patients resulting in various degrees of immunodepression. The immunocompetence of a cohort of chemical workers was studied to determine if there was any evidence that (a) varying degrees of prolonged exposure to vinyl monomers was associated with immu nological suppression, (b) the immune response was impaired in workers with exposure-related liver injury including angiosarcoma, and (c) exposure caused changes in immunological function which were identifiable in the pre-cancerous stages. The immunocompetence of 75 employees with demonstrated liver disease and 225 individuals without clinical or laboratory evidence of disease was studied. No significant clinical or laboratory evidence of impaired immuno competence was found in the immunological parameters studied, nor were differences seen among the various subpopulations of workers. A2. Assessment of the Leukocyte Adherence Inhibition Test for the Detection of Angiosarcoma of the Liver Halliday et al. (1974) and Thompson (1976) had reported the development of a leukocyte adherence inhibition test which had identified the clinical stages in colon cancer. Similar findings were reported in individuals with hepatic cancer. This study was undertaken to determine the feasibility of applying this newly developed technique to screen vinyl chloride-exposed workers for angiosarcoma of the liver. It was assessed in a number of chemical and nonchemical workers who had cancer. This test system lacked sufficient reproducibility to be considered for clinical use. A3. Assessment of Vinyl Chloride-Induced Angiosarcoma Antigen for the Detection of Latent Angiosarcoma in Exposed Workers New antigens arise from tumors formed as a response to carcinogens. Their presence in methylcholanthrene-induced sarcomas was discovered by Foley in 1953. This discovery in mice was verified and extended by Prehn and Main (1957) to conclude that there were antigens particular to and specific for tumor tissue. Subsequent evidence of tumor antigen was found in humans by the Hellstroms, Vankey, Halliday, Maluish, Thompson, and others. The majority of 4 CMA 003438 evidence suggested that the tumor antigens found were distinctive for each histological type of tumor. Studies were undertaken to determine whether angiosarcoma tumor antigens would provide specific immune reactions that could be utilized as a specific test for vinyl chloride-induced tumor development. Lymphocytes from normal and liver diseased individuals were isolated and then grown in the presence of a liver reagent prepared from either normal indi viduals or individuals who had angiosarcoma. Reactivity of lymphocytes was assessed by determining the incorporation of 3H-thymidine into stimulated cultures as compared to unstimulated cultures. A comparison study was per formed between vinyl chloride plant workers and nonchemical plant workers utilizing reactions of lymphocytes to normal and angiosarcoma liver reagents. The lymphocytic response was also studied in vinyl chloride-exposed and non vinyl chloride-exposed workers with and without evidence of liver disease. Initially, specific reactivity was seen only among vinyl chloride workers. Although these inital results were provocative, subsequent studies with large control populations demonstrated similar reactivities. The lack of further occurrences of angiosarcoma in the cohort worker population prevented further characterization of the tumor antigen and restudy of the earlier findings. A4. Study of Histocompatibility Leukocyte Antigen (HLA) Frequencies in Chemically-Exposed Workers This study involved the search for HLA tissue types which may identify individuals susceptible to chemically-induced injury. In 1976 Mulvih^i suggested genotyping as a means of screening potential employees for antilB subtypes which may predispose them to neoplastic development after occupa-^ tional exposure. Itwas prompted by the suggested increased occurrence of HLA-B27 antigen in workers with occupational asbestosis. The HLA frequencies in polyvinyl chloride manufacturing workers were compiled to determine if any increased genotype occurrences were associated with angiosarcoma or other chemically-related liver disease. Nine hundred individuals were eligible for this study: 538 individuals were HLA tissue typed and 30 different HLA-A, HLA-B, and HLA-C antigens were assessed. Preliminary comparisons were made with the HLA frequencies in vinyl chloride workers with and without liver disease, and later in those with liver disease of chemical vs. nonchemical origin. HLA frequencies were also analyzed in those individuals with and without biochemical evidence of liver injury. HLA-B15 was found to occur with a greater frequency among chemical workers with liver disease than in "control" and "normal" populations. The HLA-B15 occurrence was greater among those with chemical liver injury, although it was not statistically significant. Two other HLA markers occurred with unusual lower frequency. A5. Identification of the Endothelial Cell as the Cell of Origin For Vinyl Chloride Angiosarcoma of the Liver There has been considerable disagreement as to the cell of origin for liver angiosarcoma, a nonparenchymal cell malignancy of the liver. This is of CMA 003439 5 evidence suggested that the turner antigens found were distinctive for each histological type of tumor. Studies were undertaken to determine whether angiosarcoma tumor antigens would provide specific immune reactions that could be utilized as a specific test for vinyl chloride-induced tumor development. Lymphocytes from normal and liver diseased individuals were isolated and then grown in the presence of a liver reagent prepared from either normal indi viduals or individuals who had angiosarcoma. Reactivity of lymphocytes was assessed by determining the incorporation of ^-thymidine into stimulated cultures as compared to unstimulated cultures. A comparison study was per formed between vinyl chloride plant workers and nonchemical plant workers utilizing reactions of lymphocytes to normal and angiosarcoma liver reagents. The lymphocytic response was also studied in vinyl chloride-exposed and non vinyl chloride-exposed workers with and without evidence of liver disease. Initially, specific reactivity was seen only among vinyl chloride workers. Although these inital results were provocative, subsequent studies with large control populations demonstrated similar reactivities. The lack of further occurrences of angiosarcoma in the cohort worker population prevented further characterization of the tumor antigen and restudy of the earlier findings. A4. Study of Histocompatibility Leukocyte Antigen (HLA) Frequencies in Chemically-Exposed Workers This study involved the search for HLA tissue types which may identify individuals susceptible to chemically-induced injury. In 1976 Mulvihill suggested genotyping as a means of screening potential employees for antigen subtypes which may predispose them to neoplastic development after occupa tional exposure. It was prompted by the suggested increased occurrence of HLA-B27 antigen in workers with occupational asbestosis. The HLA frequencies in polyvinyl chloride manufacturing workers were compiled to determine if any increased genotype occurrences were associated with angiosarcoma or other chemically-related liver disease. Nine hundred individuals were eligible for this study: 538 individuals were HLA tissue typed and 30 different HLA-A, HLA-B, and HLA-C antigens were assessed. Preliminary comparisons were made with the HLA frequencies in vinyl chloride workers with and without liver disease, and later in those with liver disease of chemical vs. nonchemical origin. HLA frequencies were also analyzed in those individuals with and without biochemical evidence of liver injury. HLA-B15 was found to occur with a greater frequency among chemical workers with liver disease than in "control" and "normal" populations. The HLA-B15 occurrence was greater among those with chemical liver injury, although It was not statistically significant. Two other HLA markers occurred with unusual lower frequency. A5. Identification of the Endothelial Cell as the Cell of Origin For Vinyl Chloride Angiosarcoma of the Liver There has been considerable disagreement as to the cell of origin for liver angiosarcoma, a nonparenchyma1 cell malignancy of the liver. This is of CMA 003440 6 great importance since the identification of the type of cell responsible for malignant transformation could provide the basis for a better understanding of chemically-induced carcinogenesis. This was particularly relevant to vinyl chloride-induced liver cancer since vinyl chloride is predominantly metabo lized and detoxified by hepatocytes and produces a malignancy of the liver that is nonparenchymal in origin. Considerable controversy arose as to whether the cell of origin was a macrophage (Kupffer cell), a fibroblast, or an endothelial vascular lining cell. The recent identification of Factor VIII production by endothelial lining cells led us to determine whether liver angiosarcoma tumor contained increased amounts of coagulation Factor VIII. Factor VIII is a large protein, often called antihemophilic factor, found differentially in endothelial cells, platelets, and megakaryocytes. The studies described herein demonstrated Factor VIII presence in endothelial lining cells of arterial and venous vessels in normal liver, as well as in the cells of angiosarcoma. Histological verfication of the cell type was deter mined by light microscopy of the same tissue sections of the tumor. This verified the endothelial cell as the cell of origin in angiosarcoma. PROGRAM B STUDY OF HEPATIC BIOCHEMICAL AND ENZYMATIC SYSTEMS FOR THE IDENTIFICATION OF VINYL CHLORIDE CHEMICAL INJURY AND CANCER DEVELOPMENT: ANIMAL AND HUMAN STUDIES; Investigators - J. T. Du, M. T. Tseng, and C. H. Tamburro Bl. Characterization of Hepatic Enzyme Changes in Rats With Prolonged Vinyl Chloride Exposure: Decreased Glucose-6-Phosphatase Activity Considerable concern was initially raised concerning the best technique for screening individuals with chemical exposure. At present, federal testing requirements recommend enzyme studies related to liver parenchymal cells. However, clinical evidence indicated that not only were the hepatocytes injured but so were Kupffer and endothelial cells, and that the endothelial cell or Kupffer cell was the malignant cell of origin. In addition, studies to determine the sequential enzymatic changes which occurred with prolonged continuous vinyl chloride exposure similar to that seen with the worker population were needed. Most studies at this time had been conducted with short term exposures. The early animal studies included assessment of the standard clinical, biochemical, and enzymatic studies and later explored the enzymatic changes occurring in various subcellular organelles. These included (a) markers for microsomal enzymes related to the metabolism of vinyl chlo ride, -450, NADPH-cytochrome reductase, and mixed function oxidase, (b) cyto solic enzymes related to glutathione metabolism and glutathione content, (c) mitochondrial marker cytochrome-C-oxidase, (d) cytosolic enzymes glucose-6-phosphatase, glucose-6-phosphate dehydrogenase related to pentose phosphate pathway, and (e) nucleic acid synthesis. In addition, protein synthesis was evaluated by in-vitro incorporation of 3H-leucine. These CMA 003441 7 studies were conducted in sequential experiments exposing Sprague-Dawley rats to 10,000 ppm vinyl chloride for a period of up to 300 exposure hours. These studies demonstrate various adaptive changes in liver parenchymal cells that were not identifiable by conventional clinical laboratory test (CCLT). These data illustrate the limited usefulness of CCLT in identifying early liver cell changes even during high levels of exposure. B2. Morphological Alterations in Livers of Rats Exposed to Vinyl Chloride Sequential morphological assessments were performed in the rats from both chronic exposure experiments utilizing light and electron microscopy. Light microscopic findings included an increase in liver cell polyploidy, double nucleated cells, and areas of focal hepatocellular hyperplasia without evidence of cellular injury or increased fibrosis. Electron microscopic changes included proliferation of smooth endoplasmic reticulum without evidence of other subcellular organelles, or increased collagen deposit in the Space of Oisse. B3. Alterations of Oxidizing and Detoxifying Systems of Rat Liver After Prolonged Exposure to Vinyl Chloride A second long-term study, based partly on the results from the B1 study, was conducted to determine the adaptive capability of the detoxifying systM^ of liver cells. Glutathione reductase activity, glutathione content, glu^p thione epoxide-S-transferase (GEST; Glutathione transferase E), and glutathione aralkyl-S-transferase (GAST; glutathione transferases A&B) were measured in rats exposed to vinyl chloride and to chloroethanol, a vinyl chloride metabolite. A third set of studies was conducted to compare these detoxifying enzymes in rats exposed to vinyl chloride and excessive alcohol consumption. These studies illustrated the adaptability of liver cells to prolonged exposure to xenobiotlcs. Liver cells showed increases in reduced glutathione and glutathione reductase activity and subsequent increases in GEST and GAST, reflecting a progressive increase in the liver cells' capacity for detoxification of vinyl chloride's toxic intermediates. Concomitant alcohol consumption appears to interfere with this detoxifying capability and appears to place the hepatocyte at greater risk of malignant transformation. 84. Oxidative and Detoxifying Ability of the Liver Mesenchymal vs. Parenchymal Cells in the Metabolism of Xenoblotics Another aspect of the prolonged vinyl chloride exposure studies was to determine the oxidative and detoxifying capability of various liver cells. In collaboration with Dr. Feldhoff, the oxidative and detoxifying ability of the liver mesenchymal versus parenchymal cells was studied by determining 450, glutathione reductase, GEST, and GAST in liver hepatocytes and endothelial cells. CMA 003442 stjoias were conductec in secuer.nal experiments exoosing Spreoue-Oex'e ' -?-.3 to 10,000 ppm vinyl chloriae for a period of up to 300 exposure hours. ' These studies demonstrate various adaptive changes in liver parenchymal cells that were not identifiable oy conventional clinical laboratory test (CCLT). These data illustrate the limited usefulness of CCLT in identifying early liver cell changes even during high levels of exposure. B2. Morphological Alterations in Livers of Rats Exposed to Vinyl Chloride Sequential morphological assessments were performed in the rats from both chronic exposure experiments utilizing light and electron microscopy. Light microscopic findings included an increase in liver cell polyploidy, double nucleated cells, and areas of focal hepatocellular hyperplasia without evidence of cellular injury or increased fibrosis. Electron microscopic changes included proliferation of smooth endoplasmic reticulum without evidence of other subcellular organelles, or increased collagen deposit in the Space of Oisse. B3. Alterations of Oxidizing and Oetoxifying Systems of Rat Liver After Prolonged Exposure to Vinyl Chloride A second long-term study, based partly on the results from the B1 study, was conducted to determine the adaptive capability of the detoxifying systems of liver cells. Glutathione reductase activity, glutathione content, gluta thione epoxide-S-transferase (GEST; Glutathione transferase E), and glutathione aralkyl-S-transferase (GAST; glutathione transferases A&8) were measured in rats exposed to vinyl chloride and to chloroethanol, a vinyl chloride metabolite. A third set of studies was conducted to compare these detoxifying enzymes In rats exposed to vinyl chloride and excessive alcohol consumption. These studies illustrated the adaptability of liver cells to prolonged exposure to xenobiotics. Liver cells showed increases in reduced glutathione and glutathione reductase activity and subsequent increases in GEST and GAST, reflecting a progressive increase in the liver cells' capacity for detoxification of vinyl chloride's toxic intermediates. Concomitant alcohol consumption appears to interfere with this detoxifying capability and appears to place the hepatocyte at greater risk of malignant transformation. 84. Oxidative and Oetoxifying Ability of the Liver Mesenchymal vs. Parenchymal Cells in the Metabolism of Xenobiotics Another aspect of- the prolonged vinyl chloride exposure studies was to determine the oxidative and detoxifying capability of various liver cells. In collaboration with Or. Feldhoff, the oxidative and detoxifying ability of the liver mesenchymal versus parenchymal cells was studied by determining 450, glutathione reductase, GEST, and GAST in liver hepatocytes and endothelial cells. CMA 003443 8 The mesenchymal cells were shown to be capable of oxidizing xenobiotics which require P^-450 and mixed function oxidases, although this capability was less than that of the parenchymal cells by a 70 to 1 ratio. GEST and GAST activity were also present but in the ratio of 1 to 65, and 1 to 500, respec tively. These results demonstrate that the nonhepatocytic liver cells have the capability to activate vinyl chloride and other similar xenobiotics, al though to a lesser degree than hepatocytes. However, their ability to de toxify vinyl chloride metabolites is significantly less and may contribute to the reason for nonmesenchymal cell malignant transformation. B5. Effectiveness of Indocyanine Green (ICG) Clearances in the Detection of Liver Injury A chemically-exposed worker cohort of almost 1,000 individuals were medically screened to determine the effectiveness of ICG clearances versus standard biochemical studies of the liver to identify latent hepatic injury. Alanine aminotransferase (ALT/SGPT), aspartic aminotransferase (AST/SG0T), gamna glutamyl transpeptidase (GGPT), alkaline phosphatase (AP), and total bilirubin (TB) were assessed in contrast to 3 different dose levels of ICG. Positive predictive values as well as the specificity and sensitivity of each of the screening tests were determined in "normal" and "abnormal" worker populations as well as in those with hepatic disease and nonhepatic disease. Chemical and nonchemical liver injury were also determined in analysis of their relationship of work histories to the biochemical abnormalities. These studies demonstrated that ALT/SGPT was the screening test with the highest sensitivity and specificity among federally-required studies. ICG clearance, even at the low dose, clearly remained the best overall screening test for the detection of subclinical hepatic disease. B6. Assessment of Bile Acids vs. Indocyanine Green (ICG) Clearances in The Detection of Liver Injury Due to Chemical Exposure in The Human Population Fasting serum bile acid levels and ICG clearances were performed for 400 employees with varying degrees of chemical exposure as well as biochemical injury related to vinyl chloride exposure. Bile acid levels, ICG clearances and standard enzymatic biochemical tests were assessed regarding their ability to identify normal individuals, individuals with liver disease of nonchemical origin, and individuals with liver disease with chemical origin, correctly. These.more sensitive studies were to identify early chemical injury; six other enzyme studies were also conducted regarding more specific enzymes for hepa tocellular injury, such as sorbitol dehydrogenase. One thousand seven hundred and sixty human seras taken from 900 exposed workers- were assessed and the results compared to the vinyl chloride exposure. Fasting serum bile acids demonstrated a high sensitivity separating normal versus abnormal individuals with early or latent liver dysfunction, particularly chemically-induced types of injury. Fasting serum bile acids 003444 CV&- 9 hold promise as a potential detector of early liver injury, whether chemi or not. Bile acids have the advantage of being natural biological substances which may be taken orally for screening clearance studies. PROGRAM C STUDY OF GLYCOSAMINOGLYCAN CHANGES IN THE EARLY DETECTION OF HEPATIC FIBROTIC INJURY IN CHEMICAL EXPOSURE AND HEPATIC CANCER DEVELOPMENT; Investigators - C. E. Kupchella and R. Warick ANIMAL STUDIES: Cl. The Study of Glycosaminoglycan Changes in Livers of Animals Bearing Metastasizing Hepatomas C2. Significance of Tissue Glycosaminoglycan Elevations in Hepatic Necrosis and Regeneration Animal studies to determine the sequence of GAG changes in experi mentally-induced fibrotic liver injury, and the relationship between GAG patterns in tumor tissue and in urine were studied in animals with fast vs. slow growing, metastasizing vs. nonmetastasizing, chemically-induced and transplantable liver cells tumors. The experimental animal studies identified heparan sulfate ( a type of GAG) elevation in hepatic tissue undergoing fibrotic changes and that the increased levels were reflected in the urine. Both heparan sulfate and hyaluronic acid levels were 3-4 times higher in experimentally transplanted liver tumors than in normal controls. Urinary excretion reflects both tumor GAG composition and size. Livers from animals bearing metastasizing hepatomas had a 10-fold great concentration of nonsulfated neutral uronic acid positive material than animals bearing nonmetastasizing hepatomas. Finally, hepatic necrosis was shown to be accompanied by significant tissue GAG elevation; hepatic regeneration was not. HUMAN STUDIES: C3. Characterization of Glycosaminoglycan Patterns in Humans with Angiosarcoma C4. Evaluation of Glycosaminoglycan Analysis of Urines of Chemically-Exposed Workers CMA 003445 9 hO promise as a pctenti al oetectcr of early liver injury, wnether chemical ", ^ i* dC 1 O 5 T, 0'< e the advantage of using natural biological suostances wmcri may be ta*.en orally for screening clearance studies. PROGRAM C STUDY OF GLYCOSAMINQGLYCAN CHANGES IN THE EARLY DETECTION OF HEPATIC FIBROTIC INJURY IN CHEMICAL EXPOSURE AND HEPATIC CANCER DEVELOPMENT; Investigators - C. E. Kupchella ana R. Warick ANIMAL STUDIES: Cl. The Study of Glycosaminoglycan Changes in Livers of Animals Bearing Metastasizing Hepatomas C2. Significance of Tissue Glycosaminoglycan Elevations in Hepatic Necrosis and Regeneration Animal studies to determine the sequence of GAG changes in experi mentally-induced fibrotic liver injury, and the relationship between GAG patterns in tumor tissue and in urine were studied in animals with fast vs. slow growing, metastasizing vs. nonmetastasizing, chemically-induced and transplantable liver cells tumors. The experimental animal studies identified heparan sulfate ( a type of GAG) elevation in hepatic tissue undergoing fibrotic changes and that the increased levels were reflected in the urine. Both heparan sulfate and hyaluronic acid levels were 3-4 times higher in experimentally- transplanted liver tumors than in normal controls. Urinary excretion reflects both tumor GAG composition and size. Livers from animals bearing metastasizing hepatomas had a 10-fold great concentration of nonsulfated neutral uronic acid positive material than animals bearing nonmetastasizing hepatomas. Finally, hepatic necrosis was shown to be accompanied by significant tissue GAG elevation; hepatic regeneration was not. HUMAN STUOIES: C3. Characterization of Glycosaminoglycan Patterns in Humans with Angiosarcoma C4. Evaluation of Glycosaminoglycan Analysis of Urines of Chemically-Exposed Workers CMA 003446 10 C5. Characterization of Glycosaminoglycan Patterns in the Identification of Chemical and Nonchemical Injury of the Liver The study of glycosaminoglycans, glycoproteins involved in wound healing and scar formation, as potential indicators of chemical injury, fibrosis, and cancer development was conducted in a large cohort of exposed workers. Analy sis of GAG in tissue and urine was performed in individuals with angiosarcoma and heptocellular tumors, in individuals with chemical and nonchemical liver injury, and in individuals with chemical liver abnormalities and in normal controls. The studies indicated that human hepatic angiosarcoma and fibrotic liver disease are accompanied by elevated tissue GAGs, that the tumor tissue is different from the adjacent fibrotic areas of the liver, and that patients with angiosarcoma and primary hepatocellular carcinoma (hepatoma) have charac teristic urinary GAG patterns. These patterns are not found in "normal" indi viduals. Urinary GAG determinations give a better indication of liver disease than ultrasound or radioisotopic scanning, and although they are not as sensi tive as transaminases or ICG clearances, fractionated GAG analysis appears to differentiate between active and inactive liver disease. PROGRAM D HISTOLOGIC AND MORPHOMETRIC ANALYSIS: A MEANS OF ASSESSING HEPATIC INJURY IN CHEMICAL WORKERS; Investigators - G. H. Barrows, G. R. Schrodt, and C. H. Tamburro D1. Systematic Assessment of Histological Lesions Characteristic of Vinyl Chloride or Vinyl Monomer Injury D2. Computer-Assisted Morphometric Analysis as a Means of Determining Collagen Content 03. Development and Assessment of Morphometric Method of Analysis of Collagen Content from Liver Biopsies - Normal Occurrence with Age Vinyl monomer injury has been associated with various hepatic histo logical abnormalities. These have included subcapsular, portal, and perisinusoidal fibrosis, as well as hyperplasia of the hepatocytes and nonparenchymal (sinusoidal) cells. The ability to assess light microscopic determina tions of fibrosis (liver tissue collagen content) using computer-assisted morphometric analysis quantitatively, was delineated. The methodologies determined the normal presence of collagen and the effect of age. Prospective analysis to determine the ability of specific histological findings to cor rectly identify vinyl monomer exposure was also done. Both manual and computer-assisted morphometric assessment of normal collagen content identified the distribution between perisinusoldal. CMA 003447 11 periportal, and pericentral areas in 4 different anatomical locations of ^e human liver of 4 different age groups. These studies demonstrated that the collagen content of the liver varies with age, increasing after the fourth decade. The collagen distribution shows an increased desposition in the perisinusoidal and midzonal regions with age. The computerized morphometric readout demonstrated that the collagen estimates vary from 2-6% of the total liver globules. (This does not include capsular fibrosis or collagen support tissue about major vessels.) 04. Light Microscopic Assessment of Liver Tissue Obtained from Vinyl Chloride Workers Characterizing Various Histological Lesions Light microscopic assessment of liver tissue obtained from vinyl chloride and nonvinyl chloride workers illustrated that focal hepatocellular hyper plasia and focal mixed hyperplasia, sinusoidal dilitation and focal areas of increased reticulum characterized chemical injury could be identified in a prospective blind study and that those individuals with these identified lesions had the highest correlation with their vinyl chloride accumulative exposure ranked months; this correlation was not seen with other chemicals within their exposure environment. PROGRAM E STUDIES OF VINYL MONOMER CHEMICALS AND THEIR METABOLITES USING CHEMICAL STRUCTURE AND SYNTHESIS IN THE DETERMINATION OF TOXICITY OF THESE AGENTS; Investigator - J. L. Wong El. Synthesis, Purification, and Utilization of Vinyl Chloride Metabolites in Mutagenicity and Carcinogenicity Studies E2. Detoxification Studies of Vinyl Chloride and Its Metabolites E3. Vinyl Chloride Metabolite Detection--Chloroacetic Acid In order to further elucidate the intermediate metabolism of vinyl chloride, especially with regard to the carcinogenic potential of its inter mediate metabol ites--chlorooxirane and chloroacetaldehyde-- laboratory syn thesis of these two putative metabolites was performed. The materials were utilized to further study the detoxification of these agents and their effect on bacterial systems. Additional studies were conducted to determine the ability to identify chloroacetic acid by mass spectrometry and gas chromatog raphy. These studies illustrated that although chlorooxirane and chloroacetalde hyde eventually give the same final product, the rate of reaction and the CMA 003448 periportal, and pericentral areas in 4 different anatomical locations of the numan liver of 4 different age groups. These studies demonstrated that the collagen content of the liver varies with age, increasing after the fourth decade. The collagen distribution shows an increased desposition in the perisinusoidal and midzonal regions with age. The computerized morphometric readout demonstrated that the collagen estimates vary from 2-6* of the total liver globules. (This does not include capsular fibrosis or collagen support tissue about major vessels.) D4. Light Microscopic Assessment of Liver Tissue Obtained from Vinyl Chloride Workers Characterizing Various Histological Lesions Light microscopic assessment of liver tissue obtained from vinyl chloride and nonvinyl chloride workers illustrated that focal hepatocellular hyper plasia and focal mixed hyperplasia, sinusoidal di1itation and focal areas of increased reticulum characterized chemical injury could be identified in a prospective blind study and that those individuals with these identified lesions had the highest correlation with their vinyl chloride accumulative exposure ranked months; this correlation was not seen with other chemicals within their exposure environment. PROGRAM E STUDIES OF VINYL MONOMER CHEMICALS AND THEIR METABOLITES USING CHEMICAL STRUCTURE AND SYNTHESIS IN THE DETERMINATION OF TOXICITY OF THESE AGENTS; Investigator - J. L. Wong El. Synthesis, Purification, and Utilization of Vinyl Chloride Metabolites In Mutagenicity and Carcinogenicity Studies E2. Detoxification Studies of Vinyl Chloride and Its Metabolites E3. Vinyl Chloride Metabolite Detection--Chioroacetic Acid In order to further elucidate the intermediate metabolism of vinyl chloride; especially with regard to the carcinogenic potential of its inter mediate metabolites--chlorooxirane and chloroacetaldehyde--laboratory syn thesis of these two putative metabolites was performed. The materials were utilized to further study the detoxification of these agents and their effect on bacterial systems. Additional studies were conducted to determine the ability to identify chloroacetic acid by masis spectrometry and gas chromatog raphy. These studies illustrated that although chlorooxirane and chloroacetalde hyde eventually give the same final product, the rate of reaction and the CMA 003449 12 intermediates in the two reactions are different. Chlorooxirane conjugates instantaneously with the sulfhydral compounds, while chloroacetaldehyde takes about 2 1/2 hours for a comparative reaction. Although both routes are con verged to yield the cysteine-S-acetaldehyde conjugate, the reaction rates are vastly different, one taking minutes, the other hours to complete. Chloroacetic acid is identifiable by mass soectrometry in liquid solutions, solu tions as small as 0.Q2M (2 mg/ml), while gas chromatography sensitivity was limited to concentrations of 0.2 to 2 mg/ml H2O (200 to 2,000 ppm). Use of a porous polymer solid support and formic acid is expected to improve sensi tivity by 100-fold. Combination of GC-MS techniques appear to be the most promising approaches to the detection of end metabolic products and biological tissue. PROGRAM F ASSESSMENT OF ASSAYS FOR THE CARCINOGENIC POTENTIAL OF INDUSTRIAL CHEMICALS USING PROKARYOTIC AND EUKARYOTIC SYSTEMS; Investigators - U.N. Streips and G. Sonnenfeld FI. Further Development of Bacterial Systems for Testing Carcinogenicity and Mutagenicity of Chemical Agents Used in the Manufacturing of Vinyl Chloride and Synthetic Rubber Application of mutagenic assays for the determination of mutagenic potential of a list of environmental chemicals was conducted using improved modifications of their bacterial system. In addition, a new screening technique for carcinogenesis was developed involving the inhibition of interferon induction. This was applied to several carcinogens and mutagens and compared to the effectiveness of the standard viral bacteriological assays. These studies have demonstrated improved techniques for rapid screening of large numbers of chemical formulations being utilized within an environmental region. F2. Preliminary Studies on the Use of Interferon Induction as an Indicator of Mutagenicity and Carcinogenicity of Chemicals The new method assessing the mutagenic/carcinogenic capabilities of chemicals using an interferon assay is reported as well as related research regarding the controlled process of cell division in bacteria and how these findings relate to mammalian cells. Interferon induction inhibition in contrast to the Ames bacterial assay demonstrated the ability to separate more accurately agents with more carcinogenic potential related to human occurrences. CMA 003450 13 PROGRAM G THE STUOY OF LIVER TISSUE ANTIGENS AND ANTIBODIES IN THE DETECTION OF VINYL CHLORIDE INJURY; Investigator - E. Espinosa Gl. The Study of Tissue Antigens From Liver Tumors, Angiosarcoma, and Hepatomas G2. CirculatingAntigens and Autoantibodies in Vinyl Chloride-Associated Liver Disease The approach to the use of tissue antigens produced by neoplasms and identified in serum, either as the antigens or autoantibodies, were studied in angiosarcoma and in chemically-induced liver cancers. Several normal tissue antigens, a tumor-associated protein antigen, and a glycoprotein liver antigen absent in individuals with angiosarcoma were identified and characterized. The finding of a missing antigen in vinyl chloride-related angiosarcoma stimulated studies of antigenic deletion in chemically-Induced hepatoma and cultured human liver carcinoma cells. Two liver antigens were found to be absent. One of these antigens was shown to be normally present in other tissues (kidney and spleen) in addition to the liver. The second antigen was detected only in the liver and was found to be unrelated to the liver-specific F antigen. Liver-specific F antigen was also studied as a possible sensij^te indicator of chemically-induced liver tumor. F antigen appeared to be ab^^ in fast growing hepatoma 7777 and undetectable or very low in the slow growing hepatomas. The level of the F antigen did not appear to correlate with the rate of growth of these tumors; however, an increased concentration of F antigen appears related to those tumors with the highest metastatic characteristics. Studies of cultured human cancer carcinoma cells also noted a deficiency in liver-specific F antigen similar to that of the experimental hepatoma 7777. These data support the clinical observation that tissue antigens appear to be more useful in treatment and follow-up care, while antigenic deletions may prove more important for screening and early detection. PROGRAM H USE OF ISOLATED MAMMALIAN LIVER CELLS FOR THE STUDY OF CHEMICAL MONOMER METABOLISM; Investigator - R. C. Feldhoff HI. Isolation of Mammalian Liver Cells for the Study of the Metabolism of Chemical Monomers H2. The In Vitro Study of Albumin Synthesis by Liver from Human Biopsies CMA 003451 PROGRAM G THE STUDY OF LIVER TISSUE ANTIGENS AND ANTIBODIES IN THE DETECTION OF VINYL CHLORIDE INJURY; Investigator - E. Espinosa Gl. The Study of Tissue Antigens From Liver Tutors, Angiosarcoma, and Hepatomas G2. Circulating Antigens and Autoantibodies in Vinyl Chloride-Associated Liver Disease The approach to the use of tissue antigens produced by neoplasms and identified in serum, either as the antigens or autoantibodies, were studied in angiosarcoma and in chemically-induced liver cancers. Several normal tissue antigens, a tumor-associated protein antigen, and a glycoprotein liver antigen absent in individuals with angiosarcoma were identified and characterized. The finding of a missing antigen in vinyl chloride-related angiosarcoma stimulated studies of antigenic deletion in chemically-induced hepatoma and cultured human liver carcinoma cells. Two liver antigens were found to be absent. One of these antigens was shown to be normally present in other tissues (kidney and spleen) in addition to the liver. The second antigen was detected only in the liver and was found to be unrelated to the liver-specific F antigen. Liver-specific F antigen was also studied as a possible sensitive indicator of chemically-induced liver tumor. F antigen appeared to be absent in fast growing hepatoma 7777 and undetectable or very low in the slow growing hepatomas. The level of the F antigen did not appear to correlate with the rate of growth of these tumors; however, an increased concentration of F antigen appears related to those tumors with the highest metastatic characteristics. Studies of cultured human cancer carcinoma cells also noted a deficiency in liver-specific F antigen similar to that of the experimental hepatoma 7777. These data support the clinical observation that tissue antigens appear to be more useful in treatment and follow-up care, while antigenic deletions may prove more important for screening and early detection. PROGRAM H USE OF ISOLATED MAMMALIAN LIVER CELLS FOR THE STUDY OF CHEMICAL MONOMER METABOLISM; Investigator - R. C. Feldhoff Hi. Isolation of Maranallan Liver Cells for the Study of the Metabolism of Chemical Monomers H2. The In Vitro Study of Albumin Synthesis by Liver from Human Biopsies CMA 003452 14 Presently, the rat has provided the most useful means of studying hepatic metabolism in mammalian cells. The intact liver, however, consists of at least four cell types which differ in their metabolic and functional characteristics. These studies were directed toward the development of clinically utilizable techniques to study human liver biopsy material in relationship to its ability to synthesize, retain, and secrete protein. Improved techniques were also utilized for isolation of hepatic paren chymal and nonparenchymal cells so that nearly homogeneous populations of cells could be studied for their oxidizing and detoxifying capabilities. Preliminary results of this late addition to our research techniques have proven encouraging in the adaptability of highly sophisticated laboratory techniques to the study of human tissue in an in vitro system. It has demon strated that animal and human tissue obtained~5y a clinical biopsy technique are viable in an _i_n vitro system, being able to perform normal functions relative to albumin synthesis. The technique for the study of this tissue in an in vitro system may be applicable to human tissue from individuals with varying degrees and types of cellular injury. This technique would provide the ability to determine the oxidizing and detoxifying capability of liver cells to various xenobiotics in a quantitative fashion. Although very futuristic in its approach, it appears accomplishable since the present devel oped methodology is clinically applicable. PROGRAM I THE STUDY OF TISSUE DISPOSITION OF INDUSTRIAL CHEMICALS: THE VINYL CHLORIDE EXAMPLE; Investigators - W. J. Waddell and C. Marlowe II. The Use of Whole Body Autoradiography in Specific Tissues. Localization of Accumulated and Retained Industrial Chemicals and Their Metabolites The need for methods which more accurately identify the sites of locali zation of chemical agents and their metabolites has obvious importance. The use of whole-body autoradiography to identify the location of radioactively tagged chemicals was studied with regard to ^C-vinyl chloride. Whole-body saggital sections of mice exposed for three hours in '4C-vinyl chloride demonstrated that the highest levels of radioactivity in mice sacrificed at 20 minutes and one hour after removal from the vinyl chloride environment were observed in the liver, pancreas, kidney, intestinal contents, urine and bile. Concentrations of metabolites in the organs of excretion decreased continually over a 24-hour period. After nine hours removal from the vinyl chloride environment, Harder's gland, epithelium of the esophagus and intestine, and sublingual glands retained the highest levels of radioactivity, while moderate concentrations were seen in the liver, kidney, and intestinal contents. At 24 hours, the primary organs of retention were the thymus, Harder's gland, liver, and esophagus and intestinal epithelium. The high concentration of CMA 003453 15 nonvolatile metabolites of vinyl chloride retained in the thymus after 24 hours suggested that there was covalent binding of these metabolites to the molecules in the thymus. Possible interactions of the thymus may reflect a dual mechanism of carcinogenic action, one related to tissue damage in the liver, the other to a suppressed immune surveillance system. CMA 003454 15 nonvolatile of '/''r1''' on ^ O'* ^ is retailed in the of'."',js if*-?'- ~ hours suggested that there was covalent binding cf these metabolites to t^e molecules in the thymus. Doss'tla interactions of the thymus may reflect a dual mechanism of carcinogenic action, one related to tissue damage in the liver, the other to a suppressed immune surveillance system. CMA 003455 RESEARCH PROGRAMS AND RESULTS 16 MA 003456 PROGRAM A STUDIES OF HUMAN IMMUMOLOGICAL SYSTEMS IN THE DETECTION OF VINYL CHLORIDE AND OTHER CHEMICAL INJURIES; Investigators - H.P. Fortwengler, Jr., and C. H. Tamburro. Al. Evaluation of Immunocompetence of Workers Chronically Exposed to Vinyl Chloride Background The study of the human immunological systems as a means of detecting chemical injury was initiated to determine whether or not the body's ability to recognize substances that were foreign would make specific responses that could be identified clinically. The immune response can be divided loosely into two types: antibody formation and cell-mediated responses. Our initial studies were conducted to determine whether the mediators of immunity, i.e., lymphocytes, could be identified as having any immune defects which could be related to chemical exposure and be an early signal of an impaired immune system which might interfere with the body's normal defenses. These antibody forming cells constitute the humoral part of the human immune response. The lymphocytes are divided into at least two classes or types called B cells and T cells. The B cell lymphocytes are responsible for the synthesis and secre tion of antibody molecules, while the T cells are responsible for helping B cells. In addition, the T cells are also able to carry out a whole series of reactions in their own right. Generally these reactions involve tissue destruction, and since antibodies are usually not involved, they are called cell-mediated reactions. There has been recent thinking that the body's defense against tumors is a result of a tumor's having a unique antigen which is therefore recognized as foreign. The immune system constantly responds against these new antigens, and in this manner, there is an ongoing immune surveillance against tumors. Failure to react with the new antigen results in cancers, according to this theory. Objective 1. To determine the immune competence of chemical workers with prolonged exposure to vinyl chloride monomers 2. To compare their iirmune response to those workers who had developed liver injury, including angiosarcoma. 17 CMA 003457 18 Research Results An evaluation of immunocompetence was carried on in individuals with documented chemical hepatic injury and in those without biochemical or clinical evidence of liver injury. A second comparison was conducted between workers with high versus low exposure to vinyl chloride, utilizing our cumula tive exposure rank months (CERM) ranking. Tests for immunocompetence included enumeration of lymphocytes, determination of certain lymphocyte subpopula tions, and response of lymphocyte to nonspecific antigens as a test of lympho cytic function. Results of lymphocyte response to stimulation with nonspecific antigens Phytohemagglutinin (PHA), Concanavallin A (Con-A), and Pokeweed mitogen (PWM) are shown in Figure 1. LYMPHOCYTE RESPONSE OF VINYL CHLORIDE WORKERS (non-specific antigens) 300 - 250 _ 200 150 . 100 50- I! 1: i VINYL CHLORIDE EXPOSURE 4L,... above mm :::::: below hedian cik IT |I 1 1::: ;IJi PHYT0HEJ1AGGUJTININ CONCANAVALLIN A POKEVEED RITOGEN Figure 1 Response of the other iinnune parameters in high and low vinyl chloride exposure are shown in Table 1. No statistically significant difference was found between workers with high and low vinyl chloride exposure, or between unexposed invidivduals and chemical workers (Figure 2). CMA 003458 18 Research Results An evaluation of immunocompetence was carried on in individuals with documented chemical hepatic injury and in those without biochemical or clinical evidence of liver injury. A second comparison was conducted between workers with high versus low exposure to vinyl chloride, utilizing our cumula tive exposure rank months (CERM) ranking. Tests for immunocompetence included enumeration of lymphocytes, determination of certain lymphocyte subpopula tions* and response of lymphocyte to nonspecific antigens as a test of lympho cytic function. Results of lymphocyte response to stimulation with nonspecific antigens Phytohemagglutinin (PHA), Concanaval1 in A (Con-A), and Pokeweed mitogen (PWM) are shown in Figure 1. LYMPHOCYTE RESPONSE OF VINYL CHLORIDE WORKERS (NON-SPECIFIC ANTIC-rlo) 300 - 250 - 200 5 T: Ij VINYL CHLdRtdC EXPOSURE ABOVE .tDIAR (N-5A) :::::: eelcw ediar okr) I )r i 100 - 11 50 - PHYTOHEMGCLUTINIR Fisurc X UMCANAVALUN A pokeweed MTOGEX Response of the other irnnune parameters In high and low vinyl chloride exposure are shown in Table 1. No statistically significant difference was found between workers with high and low vinyl chloride exposure* or between unexposed invidivduals and chemical workers (Figure 2). CMA 003459 TABLE 1 IMMUNE PARAMETERS OF VC WORKERS ACCORDING TO EXPOSURE TEST STREPTOCOCCAL ANTIGEN STIMULATION ABOVE MEDIAN EXPOSURE 191**31** Cn-52) BELOW MEDIAN EXPOSURE 186+39 U-29) streptolysin 0 stimulation 8*2 (n*53) 2*1+12 Cn--24) PRO STIMULATION 29+8 (n*99) 13+6 Cn-23) VARIDASE STIMULATION 29S CIA-5'4) 41+12 Cn-23) CANDIDA STIMULATION 11+3 (n-53) 6+2 Cn--23) ABSOLUTE LYMPHOCYTE COUNT 2355132 (n*5Z) 2316+198 Cn--24) T-CELL ROSETTES AT 4C (absolute) 1503+191 (nSO) 1529+132 (n--24) T-CELL ROSETTES AT 37C (absolute) `stimulation INOtX "SEN 1227+80 (n-59) 1269+123 (n-29) STATISTICAL SIGNIFICANCE NONE NONE NONE NONE NONE NONE NONE NONE Mitogen induced lymphocyte transformation in VC workers and unexposed individuals VC workers In-78) HI Unexposed individuals (n = 2l) 19 CMA 003460 20 No statistical differences were seen in these same immune parameters between workers with or without liver injury (Tables 2 and 3). TABLE 2 irWUNE PARAMETERS OF VC WORKERS WITH AND WITHOUT LIVER DISEASE TEST PHYTOWCMAGGLUTININ STIMULATION LIVER DISEASE 253*:35** (n-25) NO LIVER DISEASE 174:35 (n-48) CONCANAVALLIN A STIMULATION 195:36 <n-2S) 162:29 Cn*48> POXEWfED MITOGEN stimulation 100:12 (w-25) 80:13 (n-4S) STREPTOCOCCAL ANTIGEN STIMULATION 208:42 <n-24> 181+32 (n-49) STREPTOLYSIN 0 STIMULATION 8*2 <n-25> 16+6 (h+49) PPD STIMULATION 18:6 (n25) 30:9 <n-45) varidase stimulation 36:9 Cn-25) 31+5 <n--50) CANDIDA STIMULATION 14+5 (n*25) 7:1 (n-49) `stimulation inmx "UM TABLE 3 IMMUNE PARAMETERS OF VC WORKERS WITH AND WITHOUT LIVER DISEASE TEST WHITE BLOOD CELL COUNT LYMPHOCYTE COUNT (percent) LIVER DISEASE 6900:500* (n-25) NO LIVER DISEASE 6650+380 (n-50) 36+2 (n-2S) 38+2 (n-48) absolute lymphocyte count 2350:200 (n-25) t-cell rosettes at 4e (percent) 67+3 <n-25) t-cill rosettes at 4c (absolute) 1600:100 (n-25) t-cell rosettes at 33c (percent) 55+2 (n-25) 2300:290 (n-48) 62:2 (n-49) 1500:100 (n*49) 52:2 (n-47) t-cell rosettes at 33c (absolute) 1300+100 (n-25) 1200+100 (n-47) Tables 4 and 5 show the results of these immunological parameters examined in old and young workers. Older individuals demonstrated significant differences in their total white count (Pc.01), percentage of lymphocytes (Pc.001), and lymphocyte reactivity to Streptolysin 0 (Pc.05) and PPD (tuberculin) (Pc.05) stimulation. We believe these findings to be age related with no relationship to the presence of liver disease or exposure to vinyl chloride. No other statistically significant differences were found between the various groups. There Is no evidence from this data that immunological abnormalities result from chronic exposure to various levels of vinyl chloride. 003461 20 No statistical differences were seen in these same immune parameters between workers with or without liver injury (Tables 2 and 3). TABLE 2 im PARAMETERS 3f % WORKERS WITH AHD WITHOUT LIVER DISEASE TEST PWTTOHWOCLUTtHIH STIMULATION LIVER DISEASE 253*155** (*-25> HO LIVER DISEASE 17"i35 (*48) CONCANAVALLJN A STIMULATION 195i3S (*-25) 162129 (**8) POKCNfCD MITOGEN STIMULATION 103*12 (*-25) 80113 (*43) STMPTOCOCCAL ANTIGEN STIMULATION 208*<2 (**24) 181132 (*-49) ST*erot,YSiN o STIMULATION 8*2 (**25) 16i6 (*49) PAD STIMULATION 18*6 (*-25) 30*9 (*45) VAN(DASC STIMULATION 36*9 (*-25) 31i5 (*-60) CANDIDA STIMULATION 14*5 (*-25) 7*1 (*49) 'STIMULATION \H1XX TABLE 3 imUNE PARAMETERS OF VC 'WORKERS WITH AND WITHOUT LIVER DISEASE TEST HITE 3(.000 CELL COUNT LIVER DISEASE 69001500* (*-25) HO LIVER DISEASE S6501380 (*-50) LT.-PUOCYTE COUNT (percent) 36*2 (n-25) AESOLJTE LYMPHOCYTE COUNT 2353*200 (*-25) t-cell rosettes at 4c (percent) 67*3 (*25) t-cell rosettes at 4c (aisoluti) 1600*100 (*-25) T-CEU ROSETTES AT 33C (percent) 55*2 (*-25) t-cill rosettes at 33c (arsolute) 1330*100 (*25) 38i2 (*-48) 2300*200 (*48) 52*2 (*49) ISOOilOO (*-M9) 52*2 (*47) 12001100 (**47) Tables 4 and 5 show the results of these imnunological parameters examined in old and young workers. Older individuals demonstrated significant differences in their total white count (Pc.01), percentage of lynjjhocytes (Pc.001), and lymphocyte reactivity to Streptolysin 0 (Pc.05) and PPD (tuberculin) (Pc.05) stimulation. We believe these findings to be age related with no relationship to the presence of liver disease or exposure to vinyl chloride. No other statistically significant differences were found between the various groups. There Is no evidence from this data that immunological abnormalities result from chronic exposure to various levels of vinyl chloride. CMA 003462 TABLE A IltIUNE PARAMETERS OF VC WORKERS OLD VERSES YOUNG TEST phytohemagglutinin stimulation cqncanavallin a STIMULATION POKEWEED MITOGEN stimulation STREPTOCOCCAL ANTIGEN STIMULATION STREPTOLYSIN 0 stimulation ppo stimulation VARICASE STIMULATION CANDIDA STIMULATION OLD 140**20** (n-57) 160*25 (n-37) 90115 C n-37) 135t35 (n-3S) 5*2 (n-38) 36+12 (n-39) 25*6 (h*38) 9*3 (h*37> YOUNG 210t32 (n-90) 18825 (n-90) 90i23 (n-90) 195*35 (n-38) 21*7 r*.os (n-33) 13*5 P*,Q5 (m*35) 39*9 <n*59> 9+3 (n*39> `STIMULATION index "SEM 21 TABLE 5 I (TONE PARAffTERS OF VC WORKERS OLD VERSES YOUNG TEST WHITE BLOOD CELL COUNT OLD 7150+400* Ck-38) LYMPHOCYTE COUNT (percent) 33+2 (n-36) ABSOLUTE LYMPHOCYTE COUNT 2300*200 (n-36) T-CELL ROSETTES AT ^C (percent) 64+2 (n-36) t-cell rosettes at 4c (absolute) 1930+150 (n-39) T-CELL ROSETTES AT 33c (percent) 52+2 (n-36) T-CELL ROSETTES AT 33c (absolute) `sen 1100+100 (n-39> YOUNG 6150+300 "`.01 (n*90) 91+2 w < ,001 (n-90) 2900+100 (n-90) 69+2 (n-90) 1600ilOO (n-90) 53+2 (n-90) 1300+100 (n-90) A2. Assessment of the Leukocyte Adherence Inhibition Test for the Detection of Angiosarcoma of the Liver Background Hal 1 id ay et al. (1974) reported the development of a leukocyte adherence inhibition test which identified the preclinical stages of colon cancer. Similar findings were reported in patients with liver cancer--the primary hepatocellular type. The need for a more specific screening test, especially for those who might be developing angiosarcoma that was not clinically de tectable, led us to assess this test in our cohort population. Objective 1. To verify the studies of Halliday and Maluisb regarding the use of the leukocyte adherence inhibition test in the identification of carcinomas of the colon and its possible adaptation to vinyl chloride-induced angiosarcoma. CMA 003463 22 Research Results Dr. Maluish came from Australia and spent a number of days helping us develop and further validate the methodology used. Despite multiple attempts we were not able to validate the reproducibility of the leukocyte adherence inhibition test, neither in colon cancer nor in liver angiosarcoma patients. Unfortunately, we must conclude that the leukocyte inhibition adherence test is not sufficiently reproducible to be used as a reliable indicator of the presence of tumors. A3. Assessment of Vinyl Chloride-Induced Angiosarcoma Antigen for the Detection of Latent Angiosarcoma in Exposed Workers Background New antigens arise on tumors formed as a response to carcinogens. Their presence on induced sarcomas was discovered by Foley (1953). This discovery in mice was verified and extended by Prehn and Maine (1957) who concluded that there were antigens particular to and specific for tumor tissue. Subse quently, evidence for specific tumor antigens was found in humans by Hellstroms, Vankey, Halliday and Maulish, Thompson, and others. The majority of the evidence suggests that tumor antigens found were distinctive for each histological type of tumor. This concept was applied to our cohort population for primary liver cancer. Objective 1. To determine whether the human body mounts an inmune reaction to developing cancer cells that can be identified by specific immune reactions with tumor-specific antigen. Research Results Lynphocytes (the cells responsible for immunity) were obtained from individual workers, isolated, and grown in the presence of liver reagents prepared from either normal individuals or individuals who had had angiosar coma. A positive reaction was identified by incorporation of ^-thymidine into stimulated lymphocyte cultures as compared to unstimulated (control) cultures. Comparison of responses to a panel of tissue extracts (Table 6) by lymphocytes from vinyl chloride workers and normal non-chemical plant workers indicated that many unexposed individuals have reactivity to these tissue antigens. Lynphocyte reactivity against angiosarcoma tissue alone occurred more frequently among vinyl chloride workers (Table 7). CMA 003464 22 Research Results Or. Maluish cane fron Australia and spent a number of days haloing us develop and further validate the methodology used. Despite multiple attempts we were not able to validate the reproducibi 1 ity of the leukocyte adherence inhibition test, neither in colon cancer nor in liver angiosarcoma patients. Unfortunately, we must conclude that the leukocyte inhibition adherence test is not sufficiently reproducible to be used as a reliable indicator of the presence of tumors. A3. Assessment of Vinyl Chloride-Induced Angiosarcoma Antigen for the Detection of Latent Angiosarcoma in Exposed Workers Background New antigens arise on tumors formed as a response to carcinogens. Their presence on induced sarcomas was discovered by Foley (1953). This discovery in mice was verified and extended by Prehn and Maine (1957) who concluded that there were antigens particular to and specific for tumor tissue. Subse quently, evidence for specific tumor antigens was found in humans by Hellstroms, Vankey, Halliday and Maulish, Thompson, and others. The majority of the evidence suggests that tumor antigens found were distinctive for each histological type of tumor. This concept was applied to our cohort population for primary liver cancer. Objective 1. To determine whether the human body mounts an immune reaction to developing cancer cells that can be identified by specific immune reactions with tumor-specific antigen. Research Results Lymphocytes (the cells responsible for immunity) were obtained from individual workers, isolated, and grown in the presence of liver reagents prepared from either normal individuals or individuals who had had angiosar coma. A positive reaction was identified by incorporation of ^H-thymidine into stimulated lymphocyte cultures as compared to unstimulated (control) cultures. Comparison of responses to a panel of tissue extracts (Table 6) by lymphocytes from vinyl chloride workers and normal non-chemical plant workers indicated that many unexposed individuals have reactivity to these tissue antigens. Lymphocyte reactivity against angiosarcoma tissue alone occurred more frequently among vinyl chloride workers (Table 7). CMA 003465 TABLE 6 TISSUE EXTRACT PANEL extract ANGIO TUMOR NORMAL LIVER HEPATOMA LIVER ANGIO "NORMAL" LIVER* NORMAL KIDNEY** NUMBER IN PANEL 3 3 1 2 2 `"NORMAL" AREAS OF LIVER TISSUE FROM ANGIOSARCOMA LIVERS. `OBTAINED FROM THE SAME DONORS AS NORMAL LIVERS. 23 TABLE 7 LYMPHOCYTE REACTIVITY TO TUMOR AND NORMAL TISSUE PANEL LYM'HOCYTES REACTIVE AGAINST ANGIO LIVER ALONE NORMAL AND ANGIO LIVER NORMAL. LIVER ALONE NONREACTIVE TOTAL VINYL CHLORIDE WORKERS N PERCENTAGE NON-VINYL CHLORIDE WORKERS N PERCENTAGE INTERPRETATION OF RESULTS 6 16 7 IS 5 13 20 53 38 100 00 8 57 17 5 36 m 100 INDIVIDUALS WITH POSSIBU SPECIFIC ANTI-TIMOR REACTIVITIES INDIVIDUALS WITH NONSPE CIFIC REACTIVITIES MASKING ANY POSSIBLE SPECIFIC REACTIVITIES INDIVIDUALS WITH NONSPECIFIC REACTIVITIES INDIVIDUALS WITH NO TISSUE REACTIVITIES CMA 003466 24 Lymphocytic reaction to normal and angiosarcoma liver reagents w studied in workers with high and low exposure (above or below the median exposures for the plant) to vinyl chloride (Figure 3). No statistical dif ference was noted between the reactivities of these two groups. LYMPHOCYTE RESPONSE OF VINYL CHLORIDE WORKERS (TISSUE ANTIGENS) VINYL CHLORIDE EXPOSURE i ABOVE MEDIAN (N-54) :::::: below median in-24) I I Il ANGIOSARCOMA ANTIGEN EXTRACT LIVER ANTIGEN EXTRACT Figure 3 Similar studies in workers with and without liver disease also failed show significant differences in their lymphocytic reactivity to ti reagents (Table 8A). TABLE 8A LYMPHOCYTE REACTIVITY TO NORMAL AND ANGIOSARCOMA TUMOR TISSUES IN VINYL CHLORIDE WORKERS WITH AND WITHOUT LIVER DISEASE LIVER NORMAL TISSUE TUMOR TISSUE LIVER DISEASE NORMAL 2.4 t 0.3* 2.6 + 0.4 1.5 + 0.2 1.8 + 0.2 'STIMULATION index + SEM CMA 003467 Lymphocytic reaction to normal and angiosarcoma liver reagents were studied in workers with nigh and low exposure (above or below the median exposures for the plant) to vinyl chloride (Figure 3). No statistical dif ference was noted between the reactivities of these two groups. LYMPHOCYTE RESPONSE OF VINYL CHLORIDE WORKERS (TISSUE ANTIGENS) VINYL CHLORIDE EXPOSURE J ABOVE YEDIAN (N-54) :: BELOW hEDIAN (N-2N) J I 3= I I Figure 3 ANGIGSARCC.TA ANTIGEN EXTRACT LIVER ANTIGEN EXTRACT Similar studies in workers with and without liver disease also failed to show significant differences in their lymphocytic reactivity to tissue reagents (Table 8A). TABLE 8A LYMPHOCYTE REACTIVITY TO NORMAL AND ANGIOSARCOMA TUMOR TISSUES IN VINYL CHLORIDE WORKERS WITH AND WITHOUT LIVER DISEASE LIVER NORMAL TISSUE TUMOR TISSUE LIVER DISEASE NORMAL 2.4 + 0.3* 2.6 0.4 1.5 0.2 1.8 i 0.2 `STIMULATION INDEX SEM < CMA 003468 25 On the chance that non-specific reactions might be obscuring the specific reactions, workers with vinyl chloride-suspected disease (working in a non-vinyl chloride environment--Pallet Plant) were compared to workers without known liver disease (working at the Main Plant). TABLE 8B [N VITRO LYMPHOCYTE STIMULATION BY ANGIOSARCOMATOUS LIVER EXTRACT SUBJECTS HUMBER TESTED NUMBER OF SUBJECTS REACTIVE AGAINST: NORMAL LIVER ONLY NORMAL & ANGIO ANGIO ONLY NON-PALLET PLANT WORKERS 16 1 11 PALLET PLANT WORKERS 18 TOTAL 34 *ONE INDIVIDUAL HAD ANGIOSARCOMA 1 2 0 2* 13 At first, this appeared to be a specific reaction to the angiosarcoma tumor extract, i.e., negative reactions to normal liver with positive reactions to tumor antigen. Initially, this type of reactivity occurred only among the vinyl chloride workers. Subsequent studies among additional nonchemical workers demonstrated similar reactivities to these liver reagents. Although these results were provocative, the lack of further occurrences of angio sarcoma in our worker population prevented us from further characterizing this (these) possible angiosarcoma antigen(s) and having the opporunity to verify these observations. A4. Study of Human Leukocyte Antigen (HLA) Frequencies in Chemically-Exposed Workers Background The major histocompatibility complex of the immune system, HLA, is com posed of multiallelic genes located on one region of chromosome 6 in the human and on chromosome 17 in the mouse. The genes in this region control a wide variety of surface antigens and lymphoid functions. This major histocompati bility complex is intimately related to the understanding of transplantations, graft versus host reactions, interactions of B and T Cells, genetic control of immune responses and several other phenomena in the immunological system. Certain genetic traits appear to be controlled in animals by the genes of this complex. These include transplantation antigens, cell-mediated lympholysis. 003469 CMA 26 ,# target antigens, immune response, tumor virus susceptibility, liver cycl adenosine monophosphate levels, hybrid resistance, and T-cell: B-cell inter actions. HLA antigens are found on almost all cells of the body except red blood cells. The major histocompatibility complexes in man are identified by letters (A, B, C, D) and numbers (A10, B12, C3), An increased incidence of certain HLA antigens had been shown to "Be associated with susceptibility to certain diseases. Mulvihill (1976) proposed that screening for these genetic markers to identify abnormal genotypes in potential employees should be done so that individuals who might be predisposed to neoplasia after occupational exposure would be separated from those from normal genotypes. The first occupational disease correlation reported was HLA-B27 antigen which was found with. increased frequency in workers suspected of having occupational asbestosis. The possibility that the histocompatible complex could be used to identify individuals at increased risk of developing chemical injury led to its study in vinyl chloride workers. Objective 1. To determine the HLA frequency in vinyl chloride workers 2. To compare the HLA frequency in vinyl chloride workers with and without liver disease. In order to deal with the statistical problem of multiple comparisons due to the large number of antigens to be studied, potential marke for liver disease were identified for study: A-9, B-15 and B-17. Research Results HLA determinations were done using the Terasaki microdroplet lymphocyte cytotoxicity test. HLA typing was performed on 538 of our chemical cohort population. Tissue typing included 11 HLA-A antigens, 15 HLA-B antigens and 4 HLA-C antigens. Frequency distributions were determined in our "standard" (healthy population) and compared with the "normal" groups studied by Scott et al. (1977) and the World Health Organization. The comparison of these fre quencies are found in Tables 9 and 10. A preliminary analysis of these HLA frequencies included the comparison of a subcohort population consisting of individuals identified through the medical screening program as having liver disease. These individuals had been transferred from the main chemical plant to an ancillary plant making wooden shipping pallets, referred to subsequently as "Pallet Plant Cohort". As seen in Table 10 this cohort had increased frequency of HLA-B 15 (12 percent-normal population versus 28 percent-Pallet Plant Cohort). CMA 003470 target antigens, immune 'esocrsa, ~^cr /irus susceotibil ity, liver cyciic aaenosine monopnosphate ieveis, ru-tr^a resistance, and T-cell: B-cell interacticns. HLA antigens are fourc m almost ail cs'ls of t^e body except red blood cells. The major histocompatibility complexes in man are identified by letters (A, B, C, D) and numbers (A10, 312, C3). An^ increased incidence of certain HLA antigens had been shown to 'lie associated* with susceptibility to certain diseases. Mulvihill (1576) proposed that screening for these genetic markers to identify abnormal genotypes in potential employees should be done so that indiviauals who might be predisposed to neoplasia after occupational exposure would be separated from those from normal genotypes. The first occupational disease correlation reported was HLA-827 antigen which was found with increased frequency in workers suspected of having occupational . asbestosis. The possibility that the histocompatible complex could be used to identify individuals at increased risk of developing chemical injury led to its study in vinyl cnloriGe workers. Objective 1. To determine the HLA frequency in vinyl chloride workers 2. To compare the HLA frequency in vinyl chloride Workers with and without liver disease. In order to deal with the statistical problem of multiple comparisons due to the large number of antigens to be studied, potential markers for liver disease were identified for study: A-9, 8-15 and 8-17. Research Results HLA determinations were done using the Terasaki microdroplet lymphocyte cytotoxicity test. HLA typing was performed on 538 of our chemical cohort population. Tissue typing included 11 HLA-A antigens, 15 HLA-8 antigens and 4 HLA-C antigens. Frequency distributions were determined in our "standard" (healthy population) and compared with the "normal" groups studied by Scott et al. (1977) and the World Health Organization. The comparison of these fre quencies are found in Tables 9 and 10. A preliminary analysis of these HLA frequencies included the comparison of a subcohort population consisting of individuals identified through the medical screening program as having liver disease. These individuals had been transferred from the main chemical plant to an ancillary plant making wooden shipping pallets, referred to subsequently as "Pallet Plant Cohort". As seen in Table 10 this cohort had increased frequency of HLA-B15 (12 percent-normal population versus 28 percent-pallet Plant Cohort). CMA 003471 HtA-A ANTIGENS A1 A3 A3 A9 A10 All A38 A29 AW30 AV31 AI32 Blank T43LE 9 HLA-A FREQUENCIES HEALTHY CONTROLS 34 31 23 16 T 13 7 9 Np** m NO NO WHO 1974 WORKSHOP 33 49 22 17 13 a n 7 S 7 a NO PALLET PLANT<arc) 46 37 14 11 16 7 IS 0 4 4 7 14 NON PALLET PLANT (BFG> 30 52 31 32 6 11 13 8 6 4 7 1* N Total Frequency NOT DON* 900 160% S03 170% 24 200% 319 300% TABLE 10 HLA-B FREQUENCIES HLA-B ANTIGENS 43 B7 B8 B12 13 14 B13 V16 17 11 W21 122 27 W3S 40 Blank HEALTHY CONTROLS 10 31 27 30 3 5 10 NO 6 NO 2 7 5 n NO N Total Fr*q*ncy 903 133% WHO 197S WORKSHOP 11 23 20 24 11 7 12 7 9 4 5 4 17 12 NO 303 174% PALLET PLANT/BFO U 14 11 21 0 11 29 7 14 4 4 7 7 11 12 1# 24 201% NON PALLET PLANT/8F0 0 24 30 31 5 4 12 4 12 4 A 11 17 1 13 410 2011 CMA 003472 28 Table 11 illustrates the frequency of HLA-815 among the entire worker cohort at the main chemical plant. This cohort was subdivided into those with liver disease versus normal individuals based on medical screening studies and liver biopsy. HLA-B15 occurred in 13 percent of the normal group versus 17 percent of the liver diseased group. TABLE 11 DISTRIBUTION OF HLA-815 ANTIGEN AMONG CHEMICAL WORKERS WITH AND WITHOUT LIVER DISEASE (MAIN PLANT COHORT) Clinical Diagnosis NORMAL N=509 LIVER DISEASE 64 3 427 15 Total C) 491 (13) 18 (17) To determine if this increased occurrence of HLA-B15 was due to chemi cally-related liver injury and not due to incidental (non-occupational) liver disease, both the pallet plant and the main plant cohorts with liver disease were subclassified into those with chemical liver injury (CLI), and non chemical liver disease (LD). The frequency of HLA-B15 in these cohorts is shown in Tables 12 and 13. Table 14 illustrates the occurrence of HLA-B15 in the entire plant cohort subclassified into those without liver disease (NORMAL), those with liver disease, all types (BOTH). Those with liver disease are further subclassified into those with chemical liver injury (CLI) and those with liver disease, non-chemical in origin (LD). CMA 003473 28 Table II illustrates the frecuency of HiA-315 among the aotir? cohort at the main chemical plant. This cohort was subdivided into those with liver disease versus normal individuals based on medical screening studies and liver biopsy. HLA-B15 occurred in 13 percent of the normal group versus 17 percent of the liver diseased group. TABLE 11 DISTRIBUTION OF HLA-B15 ANTIGEM AMONG CHEMICAL WORKERS WITH AND WITHOUT LIVER DISEASE (MAIN PLANT COHORT) Clinical Diagnosis normal N-509 liver disease 64 B15 3 427 15 Total (X) 491 (13) 18 (17) To determine if this increased occurrence of HLA-815 was due to chemi cally-related liver Injury and not due to incidental (non-occupational) liver disease, both the pallet plant and the main plant cohorts with liver disease were subclassified Into those with chemical liver injury (CLI), and non chemical. liver disease (ID). The frequency of HLA-815 in these cohorts is shown In Tables 12 and 13. Tdtle 14 Illustrates the occurrence of HLA-815 in the entire plant cohort subclassified Into those without liver disease (NORMAL), those with liver disease* all types (BOTH). Those with liver disease are further subclassified into those with chemical liver injury (CLI) and those with liver' disease, non-chemical in origin (LO). CMA 003474 TABLE 12 DISTRIBUTION OF HLA-B15 ANTIGEN AMONG WORKERS WITH AND WITHOUT CHEMICAL RELATED LIVER DISEASE (PALLET PLANT COHORT) Clinical Diagnosis 29 TABLE 13 DISTRIBUTION OF HLA-B1S ANTIGEN AMONG WORKERS WITH AND WITHOUT CHEMICAL RELATED LIVER DISEASE (MAIN PLANT COHORT) Clinical Diagnosis ___ NflRMAl 1 3 Total A (Z) (25) Cl I 4 9 13 (31) N*29 LD 3 9 12 (25) MQRHfll 65 430 Total 495 <Z) (15) N=526 CLI LD 03 2 13 15 IT (0) (19) TABLE 14 OCCURRENCE OF HlA-315 AMONG CHEMICAL WORKERS WITH AND WITHOUT CHEMICAL LIVER INJURY Clinical Diagnosis NflRMAl 65 430 Total 495 (Z) (13) ri i 4 11 15 (27) N-538 LD BOTH 6 10 22 33 28 43 (21) (23) CMA 003475 30 A5. Identification of the Endothelial Cell as the Cell of Origin for Vinyl Chloride-Induced Angiosarcoma of the Liver Background A search for evidence of vinyl chloride-induced tumor antigen has led to the finding that these tumors have an increased concentration of antigenic coagulation Factor VIII. Factor VIII is a known marker for the vascular lining cells of the endothelial type. Further experiments in animals indicated that the normal endothelial cell found lining the liver sinusoids has little if any Factor VIII fluorescence. Endothelial cells found lining larger vessels on the other hand demonstrated a striking fluorescence, as did experimentally transplanted mouse angiosarcomas. Conversely, mouse hepatic Kupffer cells, a second type of hepatic lining cell distinguished in histo logical cross section by engorgement with carbon particles, failed to demon strate positive fluorescence. An increase in antigenic Factor VIII in hepatic angiosarcoma endothelial cells as compared to the normal, led us to believe that these cells have increased production or storage capacity for antigenic Factor VIII. Objective 1. To study the Factor VIII fluorescence in human angiosarcomas, primary hepatocellular carcinomas, and normal livers. Research Results Examination of angiosarcoma liver demonstrated a significant fluorescence due to Factor VIII content in the malignant cells of the liver sinusoids. An intermittent pattern of fluorescence of the sinusoidal lining cells was seen. These cells were histologically identified as angiosarcomatous on hematoxylin and eosin ( H & E) staining of adjacent tissue. This type fluorescence was not seen in individuals with primary hepatocellular carinomas and a different pattern was seen in normal livers. These cells and their fluorescent pattern are illustrated in Figures 4 and 5. Figure 4 illustrates the H & E stained cells and Figure 5 their Factor VIII fluorescence. 003476 20 A5. Iaentification of tr.e EnGothelial Ceil as the Cell of Origin for Vinyl Chloride-Inaucea Angiosarcoma of the Liver Background A search for evidence of vinyl chloride-induced tumor antigen has led to the finding that these tumors have an increased concentration of antigenic coagulation Factor VIII. Factor VIII is a known marker for the vascular lining cells of the endothelial type. Further experiments in animals indicated that the normal endothelial cell found lining the liver sinusoids has little if any Factor VIII fluorescence. Endothelial cells found lining larger vessels on the other hand demonstrated a striking fluorescence, as did experimentally transplanted mouse angiosarcomas. Conversely, mouse hepatic Kupffer cells, a second type of hepatic lining cell distinguished in histo logical cross section by engorgement with carbon particles, failed to demon strate positive fluorescence. An increase in antigenic Factor VIII in hepatic angiosarcoma endothelial cells as compared to the normal, led us to believe that these cells have increased production or storage capacity for antigenic Factor VIII. Objective 1. To study the Factor VIII fluorescence in human angiosarcomas, primary hepatocellular carcinomas, and normal livers. Research Results Examination of angiosarcoma liver demonstrated a significant fluorescence due to Factor VIII content in the malignant cells of the liver sinusoids. An intermittent pattern of fluorescence of the sinusoidal lining cells was seen. These cells were histologically identified as angiosarcomatous on hematoxylin and eosin ( H & E) staining of adjacent tissue. This type fluorescence was not seen in individuals with primary hepatocellular carinomas and a different pattern was seen in normal livers. These cells and their fluorescent pattern are illustrated in Figures 4 and 5. Figure 4 illustrates the H & E stained cells and Figure 5 their Factor VIII fluorescence. 003477 CMA 31 Relevance to Industry These rather extensive, detailed studies of a well-defined human population provide very valuable data regarding the use of immunological screening tests for the detection and identification of chemically-related injury. (1) There appears to be no significant evidence that the immunodepression or change in the antibody responding components or the cell-mediated responding components of the human immune system is affected by vinyl chloride exposure. Therefore, studying immunological parameters of type 8 or T cells will not provide any useful information in the early stages of disease even with excessive exposure. CMA 003478 32 (2) The use of tumor antigens for the detection of developing angio sarcoma, although a provocative possibility, is far too non-specific in its present stage of development to provide any reasonably useful means of detecting early cancer development. Until there is better evidence that tumor antigens provide sufficient specificity to prevent cross reactivity with other tissue materials, this approach will not appear to be a clinically useful one. (3) The present data suggest that the HLA-B15 antigen may reflect an increased susceptibility to liver injury, especially of chemically-induced origin. Confirmation of these findings and the determination of the effec tiveness of this marker in screening those with increased susceptibility to chemical liver injury must yet be determined by prospective study. (4) The leukocyte adherence inhibition test, unfortunately, was not adequate in its technical development nor sufficiently reproducible to be useful as a screening test for either colon cancer or liver angiosarcoma. All of the above immunological studies, with the possible exception of HLA marker, have provided little supportive evidence that chemical injury interferes with human immunocompetence as determined by our present methods of study. Therefore, studies of the immunological system for screening purposes appear to be of little, if any, value in the early identification of chemical injury or disease. (5) The identification of Factor VIII production by human and ani|^ angiosarcoma tumors provides further evidence for the need to develp screen^P methods to detect endothelial rather than hepatocytic cell injury or dysfunction since these are the cells of origin that malignantly transform when exposed to vinyl chloride and other agents such as arsenic. References Foley, E.J. (1953) Antigenic properties of methylcholanthrene-induced tumors in mice of the strain of origin. Cancer Research 13:835-837. Halliday, W.J., Halliday, J.W., Campbell, C.B. et al. (1974) Specific immunodiagnosis of hepatocellular carcinoma by leukocyte adherence inhibition. British Medical Journal, 18,:349-352. Hellstrom, I., Hellstrom, K.E., Sjogren, H.O. and Warner, G.A. (1971) Demonstration of cell-mediated immunity to human neoplasms of various histological types. International Journal of Cancer, 7,:1-16. Mulvihill, J.J. (1976) Host factors in human lung tumors: An example of co-genetics in oncology. Journal National Cancer Institute, 57:3-6. CMA 003479 to (2) The usa of tu'or \~z: ^ n 3 fen !8 cetaction of dsvs 1 ooi no anglesarcoma, da 1i ~c.Tti Ccu`sjg^rf,i da PpTrCcvv oC cC acl a,, v: ,ts pp0oSs5 s1 0ib' ii lity, is r'ar too non-specific m its preseant sttaigge go:f cavveellccpp--xient to provide any reasonably useful means cf cetecting early cancer development. Until there is better evidence that tumor antigens provide sufficient specificity to prevent cross reactivity with other tissue materials, this approach will not appear to be a clinically useful one. (3) The present data suggest that the HLA-815 antigen may reflect an increased susceptibility to liver injury, especially of chemically-induced origin. Confirmation of these findings and the determination of the effec tiveness of this marker in screening those with increased susceptibility to chemical liver injury must yet be determined by prospective study. (4) The leukocyte adherence inhibition test, unfortunately, was not adequate in its technical development nor sufficiently reproducible to be useful as a screening test for either colon cancer or liver angiosarcoma. All of the above immunological studies, with the possible exception of HLA marker, have provided little supportive evidence that chemical injury interferes with human immunocompetence as determined by our present methods of study. Therefore, studies of the immunological system for screening purposes appear to be of little, if any, value in the early identification of chemical injury or disease. (5) The identification of Factor VIII production by human and animal angiosarcoma tumors provides further evidence for the need to develp screening methods to detect endothelial rather than hepatocytic cell injury or dysfunction since these are the cells of origin that malignantly transform when exposed to vinyl chloride and other agents such as arsenic. References Foley, E.J. (1953) Antigenic properties of methylcholanthrene-induced tumors in mice of the strain of origin. Cancer Research 13:835-837. Halliday, W.J., Hal 1iday, J.W., Campbell, C.B. et al. (1974) Specific inmunodiagnosis of hepatocellular carcinoma by leukocyte adherence inhibition. British Medical Journal, 18,;349-352. Hellstrom, I., Hellstrom, K.E., Sjogren, H.O. and Warner, G.A. (1971) Demonstration of cell-mediated immunity to human neoplasms of various histological types. International Journal of Cancer. 7,:1-16, Mulvihill, J.J. (1976) Host factors in human lung tumors: An example of co-genetics in oncology. Journal National Cancer Institute, 57:3-6. CMA 003480 33 Prehn, R.T. and Maine, J.M. (1957) Immunity to methylcholthrene-induced sarcomas. Journal of the National Cancer Institute, 16:No. 6, 769- 775. ------------------------------------------------------------------------- ~ Scott, B.B., Rajah, S.M. and losowsky, M.S. (1977) Histocompatibility antigens in chronic liver disease. Gastroenterology, 72:112-125. Terasaki, P.I., Bernoco, 0., Park, M.S., et al. (1978) Microdroplet testing for HLA-A, -B, -C, and -D antigens. American Journal of Clinical Pathology, _2:103--119. Thompson, D.M.P., Gold, P., Freedman, S.O. and Shuster, J. (1976) The isolation and characterization of tumor-specific antigens of rodent and human tumors. Cancer Research, 36:3518-3525. Vankey, F., Stjernsward, J. and Nilsonne, U. (1971b) Cellular immunity to human sarcoma. Journal of the National Cancer Institute, 46:1145- 1151. -------------------------------------------------------------------- " World Health Organization Report (1977) Histocompatibility testing. Bodmer, w. (ed.), Munksgaard, Copenhagen. CMA 003481 PROGRAM B STUDY OF HEPATIC BIOCHEMICAL AND ENZYMATIC SYSTEMS FOR THE IDENTIFICATION OF VINYL CHLORIDE AND OTHER CHEMICAL INJURY IN CANCER DEVELOPMENT: ANIMAL AND HUMAN STUDIES; Investigators - U.T. Du, C.H, Tamburro and M.T. Tseng ANIMAL STUDIES: 81. Characterization Of Hepatic Enzyme Changes In Rats With Prolonged Vinyl Chloride Exposure: Decreased Glucose-6-Phosphatase Activity 82. Morphological Alterations In Livers Of Rats Exposed To Vinyl Chloride Background Vinyl chloride has been shown to induce an identical malignancy in both animals and man. Vinyl chloride-induced liver cancer in animals, therefore, provided an excellent model for the study of chemically-induced carcino genesis. The rat model provides the opportunity to further elucidate the mechanism and the pathogenesis of this particular chemical agent, whose major oxidation occurs in the hepatocytes (primary liver cells) while inducing a malignant transformation in an adjacent, different cell type (endothelial lining cell of the sinusoids). Vinyl chloride had been shown to cause angiosarcoma and primary hepatocellular carcinoma in laboratory animals (Maltoni, 1975, 1976), and angiosarcoma in man (Creech and Johnson, 1974). However, little was known regarding the biochemical changes that vinyl chloride mediated to induce hepatocellular injury and subsequent malignant transformation of the mesenchymal cells. Enzymatic alterations in primary liver parenchymal cell cancers (hepato cellular carcinoma, hepatoma) have been studied extensively. Weber and Convery (1966) and Weber and Lea (1967) proposed that there was a biochemical pattern which could be related to the biological behavior of neoplastic cells. Using a spectrum of hepatomas with different growth rates, a basis for underlying neoplastic cell transformation was proposed as being related to the cell's progressive alterations in the molecular pattern which led to progressive increases in the growth rate of the tumors. Figure 1 outlines the pathways. 34 CMA. 003482 FIGURE 1 WEBER'S MOLECULAR CORRELATION CONCEPT OF NEOPLASIA 35 Key enzymes of glycolysis were shown to increase while key enzymes in gluconeogenesis decreased with tumor growth. In addition, the pentose phosphate biosynthetic pathway, glucose-6-phosphate dehydrogenase (Weber and Morris, 1963) and transaldolase (Heinrich et al., 1974) were shown to be increased in all hepatoma studies. Furthermore, the activity of phosphoribosylpyrophosphate (PRPP) synthetase and glutamine PRPP aminotransferase, the first two enzymes channeling ribose-5-phosphate into purine, DNA and RNA synthesis were also Increased. The PRPP sythetase was Increased in rapidly growing hepatomas and the transferase in all the hepatomas Irrespective of growth rates (Weber et al., 1975). Decreased enzyme activity was observed at an early stage of carcinogenesis before the morphological signs of a tumor appeared in the liver, and the activity of enzymes dependent upon the rate of hepatoma proliferation correlated with progressive malignant transformation. These findings and the lack of any long-term sequential studies of animals exposed to vinyl chloride directed us to our first three objectives. Objectives 1. To determine the sequential hepatic enzymatic changes in animals chronically exposed to vinyl chloride and their correlation to con ventional clinical tests of liver function. CMA 003483 FIGURE 1 WE3ER ' S MOLECULAR CORRELATION CONCEPT OF NEOPLASIA 35 < - - -- aiucoMioofiMiis Key enzymes of glycolysis were shown to increase while key enzymes in gluconeogenesis decreased with tumor growth. In addition, the pentose phosphate biosynthetic pathway, glucose-6-phosphate dehydrogenase (Weber and Morris, 1963) and transaldolase (Heinrich et al., 1974) were shown to be increased in all hepatoma studies. Furthermore, the activity of phosphoribosylpyrophosphate (PRPP) synthetase and glutamine PRPP aminotransferase, the first two enzymes channeling ribose-5-phosphate into purine, DNA and RNA synthesis were also Increased. The PRPP sythetase was increased in rapidly growing hepatomas and the transferase in all the hepatomas irrespective of growth rates (Weber et al., 1975). Decreased enzyme activity was observed at an early stage of carcinogenesis before the morphological signs of a tumor appeared in the liver, and the activity of enzymes dependent upon the rate of hepatoma proliferation correlated with progressive malignant transformation. These findings and the lack of any long-term sequential studies of animals exposed to vinyl chloride directed us to our first three objectives. Objectives 1. To determine the sequential hepatic enzymatic changes in animals chronically exposed to vinyl chloride and their correlation to con ventional clinical tests of liver function. CMA 003484 36 2. To determine if these sequential enzymatic patterns, especially with regard to carbohydrate metabolism and nucleic acid synthesis, would correlate with vinyl chloride-induced liver injury and ultimate malignant transformation. 3. To correlate the sequential, morphological changes with these en zymatic findings. Research Results 1. Characterization of Hepatic Enzyme Changes Sprague-Dawley adult male rats were chronically exposed to vinyl chloride. The initial studies characterized the metabolic alteration in the liver by studying various organelle enzyme markers which included glucose-e- phosphatase (microsomal), cytochrome oxidase (mitochondrial) and glucose-6- phosphatase dehydrogenase (cytosol). In addition, standard serum biochemical clinical liver studies of the liver (liver "function" tests) were performed including aspartate aminotransferase (SGOT), alanine aminotransferase (SGPT), alkaline phosphatase, lactic acid dehydrogenase, total protein, and bili rubin. Additional studies of subcellular enzymes related to nucleic acid synthesis--glutamine PRPP aminotransferase, and glycolysis--phosphofructo- kinase were done. vivo protein synthesis was assessed by ^H-leucine incorporation. The results of these studies illustrated that rats exposed to high chronic doses of vinyl chloride had a 25% lower activity of glucose-6-phosphatase, a key enzyme for gluconeogenesis after about 71 hours of exposure and a 50-100% increase in glucose-6-phosphate dehydrogenase, a rate limiting enzyme in the pentose pathway, after 84 hours of exposure. These studies demonstrated that (a) metabolic alterations in the liver cell are induced by vinyl chloride exposure, (b) gluconeogenesis is altered as reflected by a lower glucose-6-phosphatase and a higher glucose-6-phosphatase dehydrogenase activity, and (c) that these changes occur in a fashion similar to the lower gluconeogenic and higher pentose shunt activity of hepatomas. Although this could result in an enhanced production of PRPP and ultimately more nucleic acid synthesis, our studies of PRPP did not document this as a subsequent finding. These enzymatic changes, although similar to the biochemical changes seen in rapidly growing primary hepatocellular carcinomas, most likely reflect liver cell adaptation to vinyl chloride exposure but do not appear to significantly increase DNA synthesis. In addition, these changes were sufficiently limited so as not to be reflected by circulating enzyme levels traditionally used for the detection of .clinical hepatocellular injury (Figure 2). CMA 003485 CONVENTIONAL CLINICAL BIOCHEMICAL STUDIES IN RATS EXPOSED TO VINYL CHLORIDE SGPT IU/I ' ____ ___ CONTROL T i VC TREATED T L SOOT IU/I 37 LDH IU/I TOTAL PROTON GM/dt FIGURE 2 14 Im Mtm 41 lira HOURS OP VINYL CHLORIDE EXPOSURE 7IW* Kln 2. Morphological Alterations: Light and Electron Microscopic Assessment Sequential morphological assessment was performed in both of the chronic experiments utilizing light and electron microscopy. The enzymatic findings discussed in B1 and subsequently in B3 were associated with light microscopic findings which showed a marked increase in liver cell polyploidy, double nucleated cells and areas of focal hepato cellular hyperplasia. There was no evidence of hepatocellular injury, necrosis, or Increased fibrosis. The morphological changes seen, however, reflect an increased liver cell regeneration which was not measurable by the techniques used to measure DNA synthesis in our experiments. The electron microscopic changes demonstrated a proliferation of the smooth endoplasmic reticulum without evidences of changes in the hepatocytic cell, cell membrane, nucleic acid or rough endoplasmic reticulum structures. CMA 003486 CONVENTIONAL CLINICAL BIOCHEMICAL STUDIES IN RATS EXPOSED TO VINYL CHLORIDE SOFT IU/| SOOT Ol/I ir .... .. I* T* " ~ 1 in in T i T i CONTROL I VC TREATED i T TT 1 --t T I T* r i* 1 LOH 1U/I tee > 10* T T T* I T* i 1 TOTAL PROTCIN GM/dl ' T :----------1_______ t T* i Mh* T 1 41 hn FIGURE 2 HOURS Of VINYL CHLORIDE EXPOSURE T l T 71 In min 2. Morphological Alterations: Light and Electron Microscopic Assessment Sequential morphological assessment was performed in both of the chronic experiments utilizing light and electron microscopy. The enzymatic findings discussed in B1 and subsequently in B3 were associated with light microscopic findings which showed a marked Increase in liver cell polyploidy, double nucleated cells and areas of focal hepato cellular hyperplasia. There was no evidence of hepatocellular injury, necrosis, or increased fibrosis. The morphological changes seen, however, reflect an Increased liver cell regeneration which.was not measurable by the techniques used to measure DNA synthesis in our experiments. The electron microscopic changes demonstrated a proliferation of the smooth endoplasmic reticulum without evidences of changes In the hepatocytic cell, cell membrane, nucleic acid or rough endoplasmic reticulum structures. CMA 00348^ 38 There was no evidence of increased collagen deposition in the space of Disse which has been reported by others during terminal or final stages of angio sarcoma development in animals. B3. Alterations in the Oxidizing and Detoxifying System of the Rat Liver After Prolonged Exposure to Vinyl Chloride. Background Vinyl chloride metabolism had been shown to be blocked at low dosage levels (50 ppm) by inhibiting the alcohol dehydrogenase system and at higher levels (200 ppm) by metabolism via the microsomal mixed function oxidase (MFO) system (Hefner et al., 1975, Bolt et al., 1975). Vinyl chloride's major metabolites, chlorooxirane and chloroacetaldehyde, have been identified as probable ultimate carcinogens (Barbin et al, 1975; Van Duuren, 1975; and Jaeger et al., 1974). Further studies (Elmore et al., 1976; Malaveille et al., 1975; McCann et al., 1975) demonstrated the mutagenic activity of chloro oxirane and chloroacetaldehyde on microorganisms and the nonmutagenicity of chloroethanol and chloroacetic acid, other metabolites of vinyl chloride. Detoxification of these metabolites was reflected by a reduction in non protein sulfhydryl content of the liver (Hefner, 1975) and tracer studies showing major urinary metabolites of vinyl chloride to be N-acetyl-S-hydroxyethylcysteine and thiodiglycolic acid (Watanabe et al., 1976). The proposed metabolic fate of vinyl chloride is illustrated: Cl CH - CHj (v c) ^ LIVER MFO Cl CH-CH, V . (chlorooxirane) DETOXIFICATION WITH GLUTATHIONE CJ ch2ch2oh +GSH GS CH2CH2OH (chloroethanol) ^^ +GSH CJ CH2CHO gs2ch2cho (chloroacetaldehyde) I Cl CHjCOOH N*Ac-S-(2-HYDROXY ethyl) cysteine GS CH^COOH THIODIGLYCOLIC ACID (CHLOROACETIC ACID) In these experiments, the enzymes studied related to vinyl chloride oxidation, included -450, NADPH-cytochrome-C-reductase and mixed function oxidase. Those enzymes related to detoxification include non-protein sulfhydryl content (NPSC), glutathione content (6SH), glutathione reductase (GR), glutathione-epoxide-S-transferase (GEST, glutathione-E-transferase), and glutathione-aralkyl-S-transferase (GAST, glutathione A&B transferases). CMA 003488 39 Objective 1. To determine the sequential changes of key hepatic oxidizing and detoxifying enzymes with prolonged vinyl chloride exposure. Research Results The oxidizing and detoxifying capability of rat liver cells during prolonged exposure to chronic high doses of vinyl chloride demonstrated a 25-50% higher activity of glutathione reductase, the enzyme which regenerates reduced glutathione to detoxify the metabolites of vinyl chloride. Although our acute and single vinyl chloride exposure studies demonstrated a decrease in the nonprotein sulfhydryl content (48%) similar to that reported by Hefner (1975), repeated or chronic exposure to higher levels showed a progressive increase which was statistically significant after 70 hours of exposure (Figure 3). Glutathione-epoxide-S-transferase (GEST) and glutathione-aralkyl-S-transferase (GAST), the enzymes capable of conjugating the toxic vinyl chloride metabolites, were significantly elevated following vinyl chloride exposure; glutathione reductase (GR), the enzyme to regenerate glutathione, was also elevated significantly following vinyl chloride exposure (Figure 4). o NORMAL a VC EXP AIR EXP. CONTROLS RANGE a NORMAL * VC Exp * HR EXP CONTROLS UJ* C<O a O a c ma ha ai E UOJ' c E X CO e3n* oo o FIGURE j 0 FIGURE 4 2 (70) 4 (14Q) 6 (210) TIME IN WEEKS (HRS) CMA 003489 39 Objective 1. To determine the sequential changes of key hepatic oxidizing and detoxifying enzymes with prolonged vinyl chloride exposure. Research Results The oxidizing and detoxifying capability of rat liver cells during prolonged exposure to chronic high doses of vinyl chloride demonstrated a 25-50* higher activity of glutathione reductase, the enzyme which regenerates reduced glutathione to detoxify the metabolites of vinyl chloride. Although our acute and single vinyl chloride exposure studies demonstrated a decrease in the nonprotein sulfhydryl content (48%) similar to that reported by Hefner (1975), repeated or chronic exposure to higher levels showed a progressive increase which was statistically significant after 70 hours of exposure (Figure 3). Glutathione-epoxide-S-transferase (GEST) and glutathione-aralkyl-S-transferase (GAST), the enzymes capable of conjugating the toxic vinyl chloride metabolites, were significantly elevated following vinyl chloride exposure; glutathione reductase (GR), the enzyme to regenerate glutathione, was also elevated significantly following vinyl chloride exposure (Figure 4). o no* ye exp. *i* exp. CQNr*oi* .WSJ.. O NOftffAL 4 VC UK * Alft tXP CONTROLS IU* </) I<o 3 e 9 i. o U0j e i <n 3 o| FIGURE 3 FIGURE 4 CMA 003490 40 The oxidizing capability as reflected by cytochrome -450, the major protein involved in vinyl chloride metabolism, however, was reduced signifi cantly only after prolonged vinyl chloride exposure (Figure 3), supporting reports in vivo (Reynolds et al., 1975) and in vitro (Guengerich and Stricklan<37 1977; Ivanetich et al., 1977) that vinyl chloride metabolites destroyed P-450. No differences we*"e found in the conventional clinical serum biochemicaT tests or in other cellular enzyme studies (cytochrome-C oxidase, NADPH-cytochrome-C-reductase and mixed function oxidase). These changes seem to reflect early hepatic cell adaptation to chronic vinyl chloride exposure by increasing their capacity for detoxification and decreasing their capability to further metabolize vinyl chloride into its toxic intermediates--both provide a means of natural defense to a toxic substance. This entire detoxifying sequence is shown in Figure 5 as a composite of the detoxifying enzyme changes over a prolonged period of exposure. FIGURE 5 It is interesting to note that there is an earlier increase in the GEST trans ferase than the GAST transferases with prolonged exposure. This may reflect an earlier (in time) need for adaptation to excessive production of the epoxide and subsequently the accumulation of the aldehyde metabolite. Whether this reflects the maximum adaptability as indicated by a leveling off of enzyme after 200 hours, or whether these changes reflect an adaptation to CMA 003491 41 prevent cellular injury as well as prevent primary hepatocellular malignant transformation would require even longer sequential studies for confirmation; these were beyond our time constraints. B4. Oxidizing and Detoxifying Ability of Liver Mesenchymal or Parenchymal Cells in the Metabolism of Xenobiotics. Background Even though the malignant transformation induced by vinyl chloride occurred in the mesenchymal cells of the liver (endothelial cells), all meta bolic studies up to this time had been directed at the liver as a whole or to the parenchymal cell--the hepatocyte. There had been no differentiation between the oxidizing capability of the hepatocytes that demonstrated injury and the endothelial cells which were malignantly transformed. Little was also known regarding the variable ability of these two cell types in the removal of the ultimate carcinogens or active metabolites. Therefore, our next set of studies was directed toward the ability of two different cell types to metabolize vinyl chloride. Objective 1. To study the comparative ability of the hepatocytes versus the mes chymal cells to metabolize vinyl chloride and determine the detoxi fying capability of the various liver cell types. Research Results Enzymatic studies of the detoxifying activities of the subcellular fraction of liver cells, both hepatocytic and mesenchymal included the study of GEST, GAST, GR, as well as -450 and mixed function oxidase capability of the various cell fractions. The effectiveness of the isolation of the hepatocytes and mesenchymal cells was confirmed by studying pyruvate kinase activity. Mesenchymal cells were shown to have about half the GEST activity, 7% of the GAST activity, less than half of the microsomal function oxidase activity, and 73% of the GR capability compared to the hepatocytes. The activity per million cells Indicated that the nonhepatocytes had about 1/70 MFO activity, 1/65 GEST activity, 1/500 GAST activity, and 1/45 GR activity. These results demonstrate that the nonhepatocyte has significantly less capability in both the activation and detoxification of vinyl chloride and other xenobiotics. The data supports the hypothesis that the endothelial cells have a more limited capability to detoxify vinyl chloride metabolites and that this limited capability provides a greater possibility for DNA injury and subsequent malignant transformation. Again, unfortunately much longer CMA 003492 41 prevent cellular Injury as well as crsvenc cri-arv reoacocellular malignant transformation woula require even longer sequential studies for confirmation; tnese were beyond our time constraints. B4. Oxidizing and Detoxifying Ability of Liver Mesenchymal or Parenchymal Cells in the Metabolism of Xenobiotics. Background Even though the malignant transformation induced by vinyl chloride occurred in the mesenchymal cells of the liver (endothelial cells), all meta bolic studies up to this time had been directed at the liver as a whole or to the parenchymal cell--the hepatocyte. There had been no differentiation between the oxidizing capability of the hepatocytes that demonstrated injury and the endothelial cells which were malignantly transformed. Little was also known regarding the variable ability of these two cell types in the removal of the ultimate carcinogens or active metabolites. Therefore, our next set of studies was directed toward the ability of two different cell types to metabolize vinyl chloride. Objective 1. To study the comparative ability of the hepatocytes versus the mesen chymal cells to metabolize vinyl chloride and determine the detoxi fying capability of the various liver cell types. Research Results Enzymatic studies of the detoxifying activities of the subcellular fraction of liver cells, both hepatocytic and mesenchymal included the study of GEST, GAST, GR, as well as -450 and mixed function oxidase capability of the various cell fractions. "The effectiveness of the isolation of the hepatocytes and mesenchymal cells was confirmed by studying pyruvate kinase activity. Mesenchymal cells were shown to have about half the GEST activity, 7% of the GAST activity, less than half of the microsomal function oxidase activity, and 73% of the GR capability compared to the hepatocytes. The activity per million cells indicated that the nonhepatocytes had about 1/70 MFO activity, 1/65 GEST activity, 1/500 GAST activity, and 1/45 GR activity. These results demonstrate that the nonhepatocyte has significantly less capability in both the activation and detoxification of vinyl chloride and other xenobiotics. The data supports the hypothesis that the. endothelial cells have a more limited capability to detoxify vinyl chloride metabolites and that this limited capability provides a greater possibility for DNA injury and subsequent malignant transformation. Again, unfortunately much longer CMA 003493 42 experimental studies would be needed to eventually validate this hypothesis (Du et al., preprint attached in Appendix). Relevance to Industry (1) These chronic sequential studies are highly relevant to the chemical industry's environmental health problems. The sequential, biochemical changes that occur in the carbohydrate metabolism of the liver hepatocytes reflect adaptation of the liver cell to chemical exposure. This is further reflected by the histological changes which show polyploidy, double nucleated cells and areas of focal hyperplasia. As the clinical studies in this report will subsequently show, these changes are characteristic of chemical injury but do not necessarily reflect premalignancy. This would be consistent with the enzymatic studies failing to demonstrate any evidence of increased nucleic acid synthesis which is seen in the premalignant hepatoma studies of Weber and others. The important fact is that these changes cannot be detected by the standard federally-required biochemical studies, because these are directed solely at the hepatocyte. The hepatocyte may become injured but it does not usually malignantly transform. (2) The oxidative and detoxifying results of this study demonstrate that the hepatocyte progressively adapts its ability to remove and detoxify the active metabolites. This occurs without evidences of hepatocellular injury, at least up to what would be equivalent to 6-20% of the working years of a chemical worker. The studies also emphasize the fact that at very high exposures, the detoxifying system is adaptive and the oxidizing system's capacity is reduced, thereby providing a relative defense against continued formation of toxic metabolites. (3) The massive ability of the hepatocyte to oxidize as well as detoxify the active metabolites of vinyl chloride can well explain why there is no hepatocellular carcinoma development in humans with vinyl chloride exposure. A markedly reduced ability of the nonhepatocyte to detoxify the metabolites while still having the capacity to oxidize vinyl chloride into its active metabolites would support the higher probability of this cell becoming malignantly transformed. Since all clinical screening studies are directed toward the hepatocyte, these animal studies emphasize the insensitiveness of clinical biochemical studies to correctly identify early injury in the more susceptible cell, i.e., endothelial cell. More sensitive indicators of nonhepatocytlc cell injury or dysfunction are needed for the detection of early chemical injury. These indicators must take into account the function of both types of liver cells. (4) The morphological findings described in the animal experiments con firm earlier observations regarding the earliest histological manifestations of vinyl chloride injury. They support our clinical observations that these lesions can be used to identify chemical injury. Program D, on the morphormetric and histological analysis details the clinical results found in humans. 003494 OAK 'ft HUMAN STUDIES: 43 B5. Effectiveness of Indocyanine Green Clearance in the Detection of Liver Injury. Background The increasing concern over the potential health hazard of synthetic chemicals in the occupational environment has led the government to require, and many industries to establish, medical laboratory screening programs for their employees. The primary objectives of these screening programs is to identify potential work-related disease, so that intervention may reduce disability, morbidity, and mortality. Most screening studies are directed toward the detection of abnormalities in certain body systems. The specific tests used are most often selected on the basis of medical experience in symp tomatic or hospitalized populations. Prior experiences utilizing non specific, multi-phasic health screening have not proven to be effective in this area. The effectiveness of standard medical testing used predominately in asymptomatic individuals with subclinical low-grade disease--both work and non-work related--had not been determined. This study assessed the effec tiveness of using clearance studies of anionic dyes in the detection of liver injury in such individuals. Objective 1. To determine the effectiveness of ICG clearances in detecting hepatic injury. 2. To compare ICG sensitivity and specificity to standard clinical biochemical screening in the detection of liver injury. Research Results The study group (cohort) consisted of 969 male employees who were tested between June 1, 1976 and May 31, 1977 with standard biochemical studies of the liver which included alanine aminotransferase (ALT/SGPT), aspartic aminotrans ferase (AST/SGOT), gamma glutamyl transpeptidase (GGTP), alkaline phosphatase (AP), and total bilirubin (TB). In addition, these employees were also screened by ICG clearances at the 0.5, 2.5, and 5.0 mg/kg doses, and by two more specific liver enzyme tests (isocltrlc dehydrogenase [ICO] and sorbitol dehydrogenase [SDH]). This main cohort was divided into a "standard" and a "non-standard" population for purposes of analysis. The division was based upon the best medical opinion. The "standard" group demonstrated no clinical evidence of significant medical disease-occupational or non-occupational in origin, and "non-standard" included all the others. A subcohort of 120 individuals who had undergone liver biopsies for medical reasons, whether related to work or not, were also studied. This subcohort population was divided into those individuals with and without CMA 003495 43 35. Effectiveness of Indocyanine Green Clearance in the Oetection of Liver Injury. Background The increasing concern over the potential health hazard of synthetic chemicals in the occupational environment has led the government to require, and many industries to establish, medical laboratory screening programs for their employees. The primary objectives of these screening programs is to identify potential work-related disease, so that intervention may reduce disability, morbidity, and mortality. Most screening studies are directed toward the detection of abnormalities in certain body systems. The specific tests used are most often selected on the basis of medical experience in symDtomatic or hospitalized populations. Prior experiences utilizing non specific, multi-phasic health screening have not proven to be effective in this area. The effectiveness of standard medical testing used predominately in asymptomatic individuals with subclinical low-grade disease--both work and non-work related--had not been determined. This study assessed the effec tiveness of using clearance studies of anionic dyes in the detection of liver injury in such individuals. Objective 1. To determine the effectiveness of ICG clearances in detecting hepatic injury. 2. To compare ICG sensitivity and specificity to standard clinical biochemical screening in the detection of liver injury. Research Results The study group (cohort) consisted of 969 male employees who were tested between June 1, 1976 and May 31, 1977 with standard biochemical studies of the liver which included alanine aminotransferase (ALT/SGPT), aspartic aminotrans ferase (AST/SGOT), gamma glutamyl transpeptidase (GGTP), alkaline phosphatase (AP), and total bilirubin (TB). In addition, these employees were also screened by ICG clearances at the 0.5, 2.5, and 5.0 mg/kg doses, and by two more specific liver enzyme tests (isocitric dehydrogenase [ICO] and sorbitol dehydrogenase [SDH]). This main cohort was divided into a "standard" and a "non-standard" population for purposes of analysis. The division was based upon the best medical opinion. The "standard" group demonstrated no clinical evidence of significant medical disease--occupationa1 or non-occupational in origin, and "non-standard" included all the others. A subcohort of 120 individuals who had undergone liver biopsies for medical reasons, whether related to work or not, were also studied. This subcohort population was divided into those individuals with and without CMA 003496 44 histological evidence of liver injury, and the former further subdivided into those with and without histological characteristics of chemical injury (see Program D2). All employees had individual work histories and rank-ordered exposure indices for 22 different chemicals used within the work place. The GGTP provided the highest positive predictive value as a screening test for non-standard individuals (population with medical disorder). It also provided the highest sensitivity and specificity sum among the 5 most frequently abnormal tests (AST/SGOT, ALT/SGPT, AP, ICG); 2 other tests also had high predictive values--the indirect bilirubin, due to the high number of congenital indirect hyperbilirubinemic individuals, and the triglyceride levels, most likely reflecting differences in age, weight, and diabetic status in the non-standard population. The ICG clearance (0.5 mg/kg) espressed as mean t 1/2 in minutes (min.) demonstrated differences not only between standard and non-standard population but also differences between those with high and low mean exposure to vinyl chloride (Table 1). TABLE 1 INDOCYANINE CLEARANCE IN A COHORT INDUSTRIAL POPULATION MEAN SUBGROUP NO. MIN. 9 OX 100X STANDARD 662 2.9 2.0*3.8 1.8-4.8 LOW VC EXPOSURE 453 2.9 1.8-4.0 1.7-4,3 NON-STANDARD HIGH VC EXPOSURE 257 466 3.4 2.1-5,6 2.0-6.6 3.1 2.1-4.4 1.9-4.8 The biochemical tests which best correlated with the presence of hepatic disease in the chemical workers with clinical evidence of liver disease were ALT (SGPT), GGTP, alkaline phosphatase, and ICG clearance (i.e. these were the tests with the highest sensitivity in detecting latent liver disease). Sensi tivity alone, however, is not an adequate indicator of a test's screening value, especially when used in asymptomatic individuals. More appropriate evaluation of the screening abilfty of the tests requires that both sensi tivity and specificity be determined, i.e., identify the diseased individuals and correctly exclude the non-diseased individuals. The sensitivity and CMA 003497 45 specificity of the 4 most sensitive biochemical screening tests and the wu clearances (all doses) demonstrated that G6TP although the most sensitive is the least specific; AP has the highest specificity but the lowest sensitivi ty. ICG clearances, even at the low dose, clearly remain the best test, i.e., in combined sensitivity and specificity for screening and detecting individu als with subclinical liver disease (Table 2 a,b,c). TABLE 2 ABC SENSITIVITY ANO SPECIFICITY OF INDOCYANINE GREEN CLEARANCE SENSITIVITY AND SPECIFICITY OF INDOCYANINE GREEN CLEARANCE NO. MTItNTJ 1500 ICG 2.5 mo/kc BUT Radical Asukmmt ICG 0.5 mo/ius <+> (-> (<>) 203 140 () 38 1923 TOTAL 516 2011 2439 <+> But (+) Medical Auusment (-) 45 14 (-) 28 396 JCMITIVITT; 55.81 IP1CIFICITY: 95.6Z scnmviTV; 45/71 61.61 wcipicity: 396/410 96.51 Total 73 410 483 SENSITIVITY AND SPECIFICITY OF INDOCYANINE GREEN CLEARANCE ICG 5.0 ne/no 111 Btir (> 78 U. s MtDICAL Amumuit (-) 4_ a Scmitivity: 78/84 - 92.91 SftciPicrrr: 18/42 90.51 Inin. go 126 In addition ICG clearance demonstrates excellent correlation with the histo logical presence of liver disease and liver cancer (Figure 6). LIVER HISTOLOGY AND DYE CLEARANCE AMONO VINYL CHLORIDE (VC) WORKERS FIGURE 6 AVIMOC VC UPOMi IQU The subcohort blopsied population was additionally studied with regard to the sensitivity and specificity of the screening tests, the specific histo logical interpretation of the liver biopsy, and the relationship to vinyl chloride exposure. All blopsied individuals were subclassified into three groups: those with histological evidence consistent with chemical liver CMA 003498 45 scacificity of the 4 most sensitive biochemical screening tests and the ICG clearances (ail doses) demonstrated that GGTP although the most sensitive is tne leas: specific; AP has the highest specificity but the lowest sensitivi ty. ICG clearances, even at the low dose, clearly remain the best test, i.e., in combined sensitivity and specificity for screening and detecting individu als with subclinical liver disease (Table 2 a,b,c). TABLE 2 ABC SENSITIVITY AMO SPECIFICITY OF INDOCYANINE GREEN CLEARANCE SENSITIVITY AND SPECIFICITY OF IHtOCYAHI.il GREEN CLEARANCE MO. at:ents 1500 ICG 2.3 .c/a Jest Aid:cal Assessment ICG 0.5 "S/<3 (*> (-> <*> 203 140 <-) 33 1923 TOTAL 516 2011 2459 (-) 3eet (*) 'lt CAL Assessment t-) 45 14 (-) 23 596 SENSITIvtTT: 55.2 seecieicitt: 95.61 stMiTtvirr: 45/71 61.61 sMeinem; 596/41Q * 96.51 Total 73 410 433 SENSITIVITY JHD SPECIFICITY OF INDOCYANINE GREEN CLEARANCE ICS 5.0 ms/tt 8cst ftSIGU. 1Cl (> ;g LI 6 Assessment (.) 1j_ & ScNtiTivtrr; 73/W 92.91 SeiEteirr: 33/42 90.51 lAIAi. 34 52 12S In addition ICG clearance demonstrates excellent correlation with the histo logical presence of liver disease and liver cancer (Figure 6). LIVER HISTOLOGY AND DYE CLEARANCE AMONG VINYL CHLORIDE (VC) WORKERS FIGURE 6 The subcohort biopsled population was additionally studied with regard to the sensitivity and specificity of the screening tests, the specific histo logical interpretation of the liver biopsy, and the relationship to vinyl chloride exposure. All biopsled individuals were subclassified into three groups: those with histological evidence consistent with chemical liver CMA 003499 46 injury, those with histological evidence of liver disease without evidence of chemical injury, and those with normal liver biopsies. Each group was then evaluated regarding the sensitivity and specificity of the biochemical tests and their total exposure to vinyl chloride based on a rank ordered scale. Analysis of the subcohort biopsied group's work histories demonstrated that those individuals with chemical liver injury had the highest average rating for vinyl chloride exposure and were best identified by screening with ICG clearance tests. B6. The Assessment of Bile Acids Vs. Indocyanine Green (ICG) Clearances in the Detection of Liver Injury in a Chemically-Exposed Worker Population Background Since highly sensitive and specific screening tests for the detection of latent hepatic injury were needed, studies were conducted to determine whether bile acid clearances could be utilized to detect occupationally-related hepatic injury. Bile acids are naturally occurring steroids which are synthesized and removed solely by the liver. These substances, in contrast to synthetic anionic dyes such as indocyanine green, have been reported to provide equal if not better, sensitivity in the detection of liver disease. The development of radioimmunoassays for bile acids provided an economical means of determining whether these natural substances could be used as effective screening measures in subclinical liver disease. Studies were conducted to determine the relative effectiveness of bile acid levels vs. ICG clearances in the detection of subclinical liver disease. Objective 1. To determine the effectiveness of bile acid clearance in the detection of subclinical liver injury 2. To compare bile acids to ICG clearance in identifying chemical and non-chemical liver injury. Research Results A subcohort (64) of the worker population who had liver biopsies done for medical reasons ana had both ICG clearances and serum bile acids--cholylglycine (CG) and conjugates of cholic acid (CCA)--performed, were studied. These individuals were subdivided on the basis of the histological features of their biopsies into individuals with chemical liver injury (CLI), individuals with non-chemical liver disease (NCLD), and individuals with no histological abnormalities (NBX). The mean plus or minus S.E.M. for ICG, CG and CCA in the CLI, NCLD, NBX, and a non-biopsied normal group are shown in Table 2. CMA 003500 TABLE 2 CORRELATION OF LIVER BIOPSY WITH BILE ACID LEVELS AND ICG CLEARANCE ICG(tij) (0.5 tng/kg) CG (ug/dl) CCA (ug/di) CL I 4.2 + 0,6 95.2 28.3 89.7 i 29.3 NCLD 3,2 i 0.1 27.3 t 4.4 25.3 + 4.5 NBX 3.3 + 0,2 34.60+ 7,1 52.6 + 26.6 NON BIOPS I ED NORMAL 3.1 t 0.01 14.9 + 0.9 18.7 t 1.2 47 Analysis of variance (on log transform data) showed significant differences for ICG clearance for the three biopsied groups and for all four groups. Values for the CLD were significantly different from the ICGs for the normal, but showed no significant difference between the NCLD and the NBX group. Analysis of variance for ICG clearance was not as discriminating for the 3 biopsied groups. ICG clearance gave a better separation (sensitivi^^ of the normal vs. abnormal at the 95 percentile level for the not^B population in the serum bile acids. This preliminary analysis, therefore, shows that fasting serum bile acids hold promise as a potential detector of early liver dysfunction, particularly chemically-induced, in those individuals who are asymptomatic. This test should be further investigated to determine how this natural substance might be more effectively utilized as a screening technique. Bile acids can be taken orally for clearance study and are not limited to intravenous route as is the case for synthetic anionic dyes such as ICG, Studies in this area continued to be pursued in similar populations. Relevance to Industry These clinical study results provide the best scientific basis upon which to recommend effective medical screening tests for the detection of occupationally-related injury to the liver. The results also provide sound scientific reasons for recommending modification of presently required federal testing in environments utilizing potentially hepatotoxic agents. Although synthetic anionic dye presently provides the best means for detecting early latent liver injury, it is limited by the need to be injected (an invasive procedure). Although reactions to the injection of this dye have been minimal, it does present barriers to voluntary worker compliance. Therefore, development of a screening test which can effectively determine CMA 003501 5^ pi -- ^ CORRELATION OF LIiEF 2ICPSY WITH BILE ACID LEVELS AMD ICG CLEARANCE ICG(ek) (0.5 ng/kg) CG (u*/<U) CCA (ug/dl) CL I 4,2 0,6 95.2 28.3 39.7 29.3 MCLD 3.2 i 0.1 27,3 + 4.4 25.3 + 4.5 NBX 3.3 i 0.2 34.60+ 7.1 52.6 + 26.6 NON BIOPSIED NORMAL 3.1 t 0.01 14.9 0,9 18.7 + 1.2 47 Analysis of variance (on log transform data) showed significant differences for ICG clearance for the three biopsied groups and for all four groups. Values for the CLD were significantly different from the ICGs for the normal, but showed no significant difference between the NCLD and the NBX group. Analysis of variance for ICG clearance was not as discriminating for the 3 biopsied groups. ICG clearance gave a better separation (sensitivity) of the normal vs. abnormal at the 95 percentile level for the normal population in the serum bile acids. This preliminary analysis, therefore, shows that fasting serum bile acids hold promise as a potential detector of early liver dysfunction, particularly chemically-induced. In those individuals who are asymptomatic. This test should be further investigated to determine how this natural substance might be more effectively utilized as a screening technique. Bile acids can be taken orally for clearance study and are not limited to intravenous route as is the case for synthetic anionic dyes such as IC&. Studies in this area continued to be pursued in similar populations. Relevance to Industry These clinical study results provide the best scientific basis upon which to reconmend effective medical screening tests for the detection of occupationally-related Injury to the liver. The results also provide sound scientific reasons for recommending modification of presently required federal testing In environments utilizing potentially hepatotoxlc agents. Although synthetic anionic dye presently provides the best means for detecting early latent liver injury, it is limited by the need to be injected (an Invasive procedure). Although reactions to the injection of this dye have been minimal, it does present barriers to voluntary worker compliance. Therefore, development of a screening test which can effectively determine CMA 003502 48 overall hepatic function, can be given orally, and is a nonsynthetic, normal biological constituent approaches the most ideal test for screening. Bile acid clearance could be the least invasive and therefore the most acceptable method to test asymptomatic individuals undergoing screening. The bile acid studies described above are the basis for recommending its use in future screening. The standardization of these bile acid test techniques, however, requires further careful validation so that specific recommendations regarding their use will provide the most accurate delineation between those with disease and without disease and those with occupationally-related disorders versus naturally-occurring ones. References Barbin, A., Bresh, H., Croisy, A., Jacquignon, P., Malaveille, C., Malaveille, C., Montesano, R., and Bartsch, H. (1975). Livermi crosome-mediated formation of alkylating agents from vinyl bromide and vinyl chloride. Biochem. Biophys. Res. Commun. 67, 596-603. Bolt, H.M., Kappus, H., Butcher, A., and Bolt, W. (1975). Metabolism of vinyl chloride. Lancet 1, 1425. Creech, J.L., and Johnson, M.N. (1974). Angiosarcoma of liver in the manufacture of polyvinyl chloride. jJ. Qccup. Med. 16, 150-151. Creech, J.L., Makk, L., Whelan, J.G., Or., and Tamburro, C.H. Hepatotoxicity among polyvinyl chloride production workers during first year of surveillance program. Gastroenterology 67, 786 (1974). Du, J.T., and Tamburro, C.H. (1976). Decreased glucose-6-phosphatase activity in liver in vinyl chloride exposed rats. Fed. Proc. 35, 329. Du, O.T., ana Tamburro,C.H. (1978). Elevated glutathione content, glutathione-S-transferase and glutathione reductase in liver of rats exposed to vinyl chloride. Fed. Proc. 37, 1545. Du, J.T., Sandoz, J.P., Tseng, M.T., and Tamburro, C.H. (1979). Biochemical alterations in livers of rats exposed to vinyl chloride. jJ. Toxicol. Environ. Health 5, 1119-1132. Elmore, J.D., Wong, J.L., Laumbach, A.D., and Streips, U.N. (1976). Vinyl chloride mutagenicity via the metabolites chlorooxirane and chloroacetaldehyde monomer qydrate. Biochem. Biophys. Acta 442, 405-419. ' Fortwengler, H.P., and Tamburro, C.H. (1975) Use of dye clearance in the detection of hepatocellular injury among vinyl chloride workers. Clin. Res. 23, 264A. Green, T., and Hathway, D.E. (1977). The chemistry and biogenesis of S-containing metabolites of vinyl chloride in rats. Chem. Biol. Interact. 17, 137-150. CMA 003503 49 Greenberg, R.A., and Tamburro, C.H. (1981). Exposure indices for epi demiological surveillance of carcinogenic agents in an industrial chemical environment. .J. Occup. Med., 23. 353-358. Greim, H., Bonse, G., Radwan, Z., Reichert, D., and Henschler, D. (1975). Mutagenicity in vitro and potential carcinogenicity of chlorinated ethylenes as a function of metabolic oxirane formation. Biochem. Pharmacol. 24, 2013-2017. Guengerich, F.P. and Strickland, T.W. (1977). Metabolism of vinyl chloride: Destruction of the heme of highly purified liver microsomal cytochrome P-450 by a metabolite. Mol. Pharmacol. 13, 993-1004. Hefner, R.E., Watanabe, P.G., and Gehring, P.J. (1975). Preliminary studies of the fate of inhaled vinyl chloride monomer in rats. Ann. N.Y. Acad. Sci. 246, 135-148. Heinrich, P.C., Morris, H.P., and Weber, G. (1974). Increased phosphoribosylpyrophosphate synthetase activity in rapidly growing hepatomas. FEBS Lett. 42, 145-148. Ivanetich, K.M., Aronson, I., and Katz, I.D. (1977). The interaction of vinyl chloride with rat hepatic microsomal cytochrome P-450 in vitro. Biochem. Biophys. Res. Commun. 74, 1411-1418. Jaeger, R.J., Connolly, R.B., and Murphy, S.O. (1974a). Effect of 18 h^T fast and glutathione depletion on 1, 1-dichloroethylene-induced hepatotoxicity and lethality in rats. Exp. Mol. Pathol. 20, 187-198. Jaeger, R.J., Reynolds, E.S., Connolly, R.B., Moslen, M.T., Szabo, S., and Murphy, S.D. (1974b). Acute hepatic injury by vinyl chloride in rats pretreated with phenobarbital. Nature (London) 252, 724-726. Kappus, H., Bolt, H.M., Buchter, A., and Bolt, W. (1976). Liver microsomal uptake of 14C-vinyl chloride and transformation to protein alkylatinq metabolites in vitro. ToxicoL. Appl. Pharmacol. 37, 461-471. Malaveille, C., 8artsch, H., Barbin, A. Camus, A.M., and Montesano, R. (1975). Mutagenicity of vinyl chloride, chloroethylene oxide, chloroacetaldehyde and chloroethanol. Biochem. Biophys. Res. Commun. 63, 363-370. Maltoni, C., Lefemine, G., Chieco, P. and Carretti, D. (1974). Vinyl chloride carcinogenesis: current results and prospective. Medicine Del. Lavoro. 65, 421-444. Maltoni, C., and Lefemine, C. (1975). Carcinogenicity bioassays of vinyl chloride: Current results. Ann. N.Y. Acad. Sci. 246, 195-218. CMA 003504 49 Greenberg, R.A., and Tambur'-o, C.H. (1531). Exposure indices for epi demiological surveillance of carcinogenic agents in an industrial chemical environment. 2. Occup. Med., 23. 353-358. Greim, H., Bonse, G., Radwan, Z., Reichert, 0., and Hensehler, 0. (1975). Mutagenicity in vitro and potential carcinogenicity of chlorinated ethylenes as a function of metabolic oxirane formation. Biochem. Pharmacol. 24, 2013-2017. Guengerich, F.P. and Strickland, T.W. (1977). Metabolism of vinyl chloride: Destruction of the heme of highly purified liver microsomal cytochrome P-450 by a metabolite. Mol. Pharmacol. 13. 993-1004. ------------------------------ Hefner, R.E., Watanabe, P.G., and Gehring, P.J. (1975). Preliminary studies of the fate of inhaled vinyl chloride monomer in rats. Ann. N.Y. Acad. Sci. 246, 135-148. Heinrich, P.C., Morris, H.P., and Weber, G. (1974). Increased phos- phoribosylpyrophosphate synthetase activity in rapidly growing hepatomas. FEBS Lett. 2, 145-148. Ivanetich, K.M., Aronson, I., and Katz, I.D. (1977). The interaction of vinyl chloride with rat hepatic microsomal cytochrome P-450 in vitro. Biochem. Biophys. Res. Common. 74, 1411-1418. Jaeger, R.J., Connolly, R.B., and Murphy, S.D. (1974a). Effect of 18 hr. fast and glutathione depletion on 1, 1-dichloroethylene-induced hepatotoxlcity and lethality in rats. Exp. Mol. Pathol. 20, 187-198. Jaeger, R.J., Reynolds, E.S., Connolly, R.B., Moslen, M.T., Szabo, S., and Murphy, S.D. (1974b). Acute hepatic injury by vinyl chloride in rats pretreated with phenobarbital. Nature (London) 252, 724-726. Kappus, H., Bolt, H.M., Buchter, A., and Bolt, W. (1976). Liver microsomal uptake of ^C-vinyl chloride and transformation to protein alkylating metabolites in vitro. ToxicoL. Appl. Pharmacol. 37, 461-471. -------------- ---------------------------------------------------- Malaveille.C., Bartsch, H., Barbin, A. Camus, A.M., and Montesano, R. (1975). Mutagenicity of vinyl chloride, chloroethylene oxide, chloroacetaldehyde and chloroethanol. Biochem. Biophys. Res. Commun. 63, 363-370. Maltoni, C., Lefemine, G., Chieco, P. and Carretti, D. (1974). Vinyl chloride carcinogenesis: current results and prospective. Medicine Del. Lavoro. 65, 421-444. Maltoni, C., and Lefemine, C. (1975). Carcinogenicity bioassays of vinyl chloride: Current results. Ann. N.Y. Acad. Sci. 246, 195-218. CMA 003505 50 McCann, J., Simmon, V., Streitwieser, 0., and Ames, 8.N. (1975). Muta genicity of chloroacetaldehyde, a possible metabolic product of 1,2dichloroethane, chloroethanol, vinyl chloride and cyclophosphamide. Proc. Nat. Acad. Sci. USA 72_, 3190-3193. Reynolds, E.S., Moslen, M.T., Szabo, $., and Jaeger, R.J. (1975). Vinyl chloride-induced deactivation of cytochrome P-450 and other components of the liver mixed function oxidase system: An in vivo study. Res. Commun. Chem. Pathol. Pharmacol. 12, 685-694. Tamburro, C.H., Makk, L., and Popper, H. (1979). Early hepatic histological alterations among chemical (vinyl monomer) workers. Gastroenterology 77, A33. VanDuuren, B.L. (1975). Possible mechanism of carcinogenic action of vinyl chloride. Ann. N.Y. Acad. Sci. 246, 258-267. Watanabe, P.G., Hefner, R.E., Jr., and Gehring, P.J. (1976a). Vinyl chloride-induced depression of hepatic non-protein sulfhydryl content and effects of bromosulphalein (BSP) clearance in rats. Toxicol. 6, 1-8. ------------- Watanabe, P.G., McGowan, G.R., and Gehring, P.J. (1976b). Fate of 14C-vinyl chloride after single oral administration in rats. Toxicol. Appl. Pharmacol. 36, 339-352. Watanabe, P.G., Zempel, J.A., Pegg, D.G., and Gehring, P.J. (1978). Hepatic macromolecular binding following exposure to vinyl chloride. Toxicol. Appl. Pharmacol. 44, 571-579. Weber, G. (1974). The molecular correlation concept, hi The Molecular Biology of Cancer, ed. H. Busch, pp. 487-521. New York:Academic. Weber, G. and Convery, H.J.H. (1966). Insulin: Inducer of glucose-ephosphate dehydrogenase. Life Sci. 5, 1139-1146. Weber, G. and Lea, M.A. (1967). The molecular correlation concept. In Methods in Cancer Research, ed. H. Busch, vol. 2, pp. 523-578. Hew York:Academic. CMA 003506 CMA 003507 / PROGRAM C STUDY OF GLYCOSAMINOGLYCAN CHANGES IN THE DETECTION OF HEPATIC FIBROTIC INJURY IN CHEMICAL EXPOSURE AND HEPATIC CANCER DEVELOPMENT; Investigators - C.E. Kupchella and R. Warick ANIMAL STUDIES: Cl. The Study of Glycosaminoglycan Changes in Livers of Animals Bearing Metastasizing Hepatomas C2. Significance of Tissue Glycosaminoglycan Elevations in Hepatic Fibrosis, Necrosis, and Regeneration HUMAN STUDIES: C3. Characterization of Glycosaminoglycan Patterns in Humans with Angiosarcoma C5. Characterization of Glycosaminoglycan Patterns in the Identification of Chemical and Nonchemical Injury of the Liver C4. Evaluation of Glycosaminoglycan Analysis of Urines of Chemically- Exposed Workers Background It is well established that glycosaminoglycans (GAGs) are involved in wound healing and scar formation. Although it is less certain what role they play, it is known that certain GAGs are elevated in malignant tumors including hepatic tumors, and it has been postulated that GAGs may be impor- tant determinants of tumor-cell properties. A number of laboratories, including ours, have established that urinary GAG excretions may serve as markers in the pathogenesis of chemical injury, fibrosis, and cancer. Despite the fact that urinary GAG analyses have long been used clinically to detect and diagnose genetically-determined metabolic disorders of GAG metabolism, a 51 CMA 003508 52 systematic evaluation of the usefulness of urinary GAG patterns in the detec tion and diagnosis of acute and/or chronic, necrotic and/or fibrotic liver injury--or cancer--has never been made. Studies were begun to determine the practical utility of tissue and urinary GAG analysis in the detection and diagnosis of chemically-induced liver injury and cancers. Objective In Animals: 1. To determine the relationship between GAG patterns in tumor tissue (and urine) in animals with fast versus slow growing and in metasta sizing versus non-metastasizing, chemically-induced, transplantable, hepatocellular tumors. (As an initial means of exploring the func tional role of GAGs known to be elevated in hepatic cancer). 2. To determine the sequence of GAG changes (in tissue, urine and blood) associated with the onset of experimentally-induced fibrotic injury with special attention to any change coinciding with the transition from reversible to irreversible fibrosis. 3. To determine the degree to which chemically-induced necrosis of the liver results in altered urinary GAG excretion and confirm changes seen in tissue in earlier studies. In Humans: 4. To compare the relative accuracy of the conmon methods of evaluating urinary GAGs and find the least expensive and least time-consuming method of urinary GAG analysis able to give good specificity and sen sitivity. 5. To repeat in a double-blind clinical trial the evaluation of the ability of urinary GAG analysis to correctly identify active liver disease. 6. To evaluate urinary GAG patterns in groups of alcoholics with active liver injury over a time period. 7. To compare liver tissue and urinary GAG patterns in patients (a) with alcohol-injured livers, (b) hepatic angiosarcoma, (c) primary hepatocellular tumors, (d) livers with metastatic cancer, and (e) chemically-induced hepatitis and cirrhosis. Research Results In Animals: Livers of animals bearing metastasizing hepatoma (5123tc) have 10-fold greater concentrations of a non-sulfated, neutral, uronic acid-positive CMA 003509 52 stsmatic evaluation of the usefulness of urinary GAG patterns in the detecon and diagnosis or acute and/or chronic, necrotic and/or fibrotic liver njury--or cancer--has never been made. Studies were begun to determine the ractical utility of tissue and urinary GAG analysis in the detection and iagnosis of chemically-induced liver injury and cancers. Objective In Animals: 1. To determine the relationship between GAG patterns in tumor tissue (and urine) in animals with fast versus slow growing and in metasta sizing versus non-metastasizing, chemically-induced, transplantable, hepatocellular tumors. (As an initial means of exploring the func tional role of GAGs known to be elevated in hepatic cancer). 2. To determine the sequence of GAG changes (in tissue, urine and blood) associated with the onset of experimentally-induced fibrotic injury with special attention to any change coinciding with the transition from reversible to irreversible fibrosis. 3. To determine the degree to which chemically-induced necrosis of the liver results in altered urinary GAG excretion and confirm changes seen in tissue in earlier studies. In Humans: 4. To compare the relative accuracy of the common methods of evaluating urinary GAGs and find the least expensive and least time-consuming method of urinary GAG analysis able to give good specificity and sen sitivity. 5. To repeat in a double-blind clinical trial the evaluation of the ability of urinary GAG analysis to correctly identify active liver disease. 6. To evaluate urinary GAG patterns in groups of alcoholics with active liver Injury over a time period. 7. To compare liver tissue and urinary GAG patterns in patients (a) with alcohol-injured livers, (b) hepatic angiosarcoma, (c) primary hepatocellular tumors, (d) livers with metastatic cancer, and (e) chemically-induced hepatitis and cirrhosis. Research Results In Animals: Livers of animals bearing metastasizing hepatoma (5123tc) have 10-fold greater concentrations of a non-sulfated, neutral, uronic acid-positive CMA 003510 53 material than is found in the livers of animals bearing two others, non-metastasizing hepatomas (4). Heparan sulfate and hyaluronic acid levels--but not heparin--are 3-4 times higher in experimentally transplanted hepatomas than in normal liver and urinary excretion reflects both the tumor GAG composition and the size of tumors (4, 8). Hepatic necrosis is accompanied by significant tissue GAG elevations but hepatic regeneration is not (5). Heparan sulfate (a type of GAG) is elevated in hepatic tissue undergoing experimentally induced fibrosis and heparan sulfate is elevated in the urine of experimental animals (3). In Humans; Human hepatic angiosarcoma and fibrotic liver disease are accompanied by elevated tissue GAGs (1). The GAGs in the angiosarcomatous tumor tissue are different from those in fibrotic tissue adjacent to the tumor (2). Angiosarcoma and hepatoma patients have characteristic urinary GAG patterns--patterns not found in normal controls (2). Gross (non-fractionated) urinary GAG determinations give a better indica tion of liver disease than ultrasound analysis. Exacting urinary GAG analysis (fractionated) is able to differentiated active from inactive liver disease (7). Table 1 illustrates this for chrondrotin sulfate fraction. TABLE 1 PRELIMINARY EVALUATION OF THE USE OF AN EXCLUSIVELY CH0NM01TIN SULFATE EXCRETION PATTERN AS A SCREENING TEST FOR VINYL CHLORIDE - EXPOSURE - ASSOCIATED LIVER INJURY TEST* RESULT VINYL CHLORIDE EXPOSURE PLUS NON-ANGIOSARCOMA LIVE* INJURY 4, 7J 2 29 sun 9 32 SENSITIVITY - 77.71 <951 LIMITS 0.0X - 97,21) SPECIFICITY * 90.SI (951 LIMITS 751 - 981) urine fractionation in HMICM the chonoroitin sulfate fraction is POSITIVE RUT THE HYALURONIC AC 10 *N0 HEPARIN FRACTION* ARE IOTH NEGATIVE. CMA 003511 54 Relevance to Industry These studies address the need for useful screening tests for liver injury--urine tests of the type evaluated obviously fit the ideal of being non-invasive, have no associated morbidity/mortality, and do not require "time off" from work to perform. They could provide useful screening tests for early injury and for dif ferentiating active versus inactive disease. GAG analysis may help assess methods of therapeutic intervention, i.e., the elucidation of the role of the GAGs in the pathogenesis of fibrotic liver disease and the identification of strategies by which fibrogenesis can be blocked and/or reversed. Finally, they could provide useful means of identifying individuals at risk of chemical injury--by further characterization of GAG changes, one may be able to identify specifc injury due to chronic alcohol, chemical or other exposure agents in individuals with liver impairment. References 1. Kupchella, C. 0., and Tamburro, C. H., (1978). Urinary and Tissue Glycosaminoglycans patterns in angiosarcoma and other vinyl chloride exposure associated liver injury in: Detection and Prevention of Cancer. H.E. Neiburgs, Ed., Part 1, VFTTT71iar^B| Dekker, Inc., New York. 2. Curran, K. L., Kupchella, C. E., and Tamburro, C.H., (1977). Urinary Glycosaminoqlycan Patterns in Angiosarcoma of the Liver. Cancer 40:3050-3053. 3. Kupchella, C. E., Jarvis, J. 0., Curran, K. L. and Tamburro, C. H., (1977). Tissue and urinary glycosaminoglycans (GAG) changes in hepatic fibrosis. Presented at the meeting of the American Association of Study of Liver Disease. Chicago, Illinois, November 1, 1977. Gastroenterology 73(5):1229. 4. Kupchella, C. E., Drake, E., Curran, K. L., Kennedy, J. and Tamburro, C. H., (1979). Tissue and Urinary Glycosaminoglycans in Transplantable Hepatomas. Gastroenterology 75(5):972. 5. Kupchella, C. E., Seeskas, E., Kennedy, J. S. and Espinosa, E., (1979). Glycosaminoglycan Changes Associated with Hepatic Tumors: The Contributions of Regeneration and Necrosis. Clinical Research 27(2):389. 003*12 54 Relevance to Inaustry These studies address the need for useful screening tests for liver injury--urine tests of the type evaluated obviously fit the ideal of being non-invasive, have no associated morbidity/mortality, and do not require "time off" from work to perform. They could provide useful screening tests for early injury and for dif ferentiating active versus inactive disease. GAG analysis may help assess methods of therapeutic intervention, i.e., the elucidation of the role of the GAGs in the pathogenesis of fibrotic liver disease and the identification of strategies by which fibrogenesis can be blocked and/or reversed. Finally, they could provide useful means of identifying individuals at risk of chemical injury--by further characterization of GAG changes, one may be able to identify specifc injury due to chronic alcohol, chemical or other exposure agents in individuals with liver impairment. References 1. Kupchella, C. D., and Tamburro, C. H., (1978). Urinary and Tissue Glycosaminoglycans patterns in angiosarcoma and other vinyl chloride exposure associated liver injury in: Detection and Prevention of Cancer. H.E. Neiburgs, Ed., Part 1, Vo 1. 1, Marcel Dekker, Inc., New York. 2. Curran, K. L., Kupchella, C. E., and Tamburro, C.H., (1977). Urinary Glycosaminoglycan Patterns in Angiosarcoma of the Liver. Cancer 40:3050-3053. ---------- 3. Kupchella, C. E., Jarvis, J. 0., Curran, K. L. and Tamburro, C. H., (1977). Tissue and urinary glycosaminoglycans (GAG) changes in hepatic fibrosis. Presented at the meeting of the American Association of Study of Liver Disease. Chicago, Illinois, November 1, 1977. Gastroenterology 73(5):1229. 4. Kupchella, C. E., Drake, E., Curran, K. L., Kennedy, J. and Tamburro, C. H., (1979). Tissue and Urinary Glycosaminoglycans in Transplantable Hepatomas. Gastroenterology 75(5):972. 5. Kupchella, C. E., Seeskas, E., Kennedy, J. S. and Espinosa, E., (1979). Glycosaminoglycan Changes Associated with Hepatic Tumors: The Contributions of Regeneration and Necrosis. Clinical Research 27(2):389. CMA 003513 55 6. Greenberg, R. A., and Tamburro, C. H. (with C. E. Kupchella et al.), (1978). Early Detection of Disease in Individuals Exposed to Vinyl Chloride. Presented at the 1978 Annual Meeting of the American Public Health Association, San Diego, CA. 7. Curran, K. L., Kupchella, C. E., Sandoz, J. and Tamburro, C. H., (1979). Urinary Glycosaminoglycan Patterns in Human Hepatic Angiosarcoma, Hepatoma and in Workers at Risk with Angiosarcoma. Gastroenterology 75(5):959. 8. Kupchella, C. E., Drake, E., Curran, K. L., Kennedy, J., and Warick, R., (1981). Tissue and Urinary Glycosaminoglycan Patterns in Three Types of Morris Hepatomas. Cancer Research 4(2):419-424. -------------------------- 9. Tamburro, E. H., Kupchella, C. E., Taylor, K., Landau, E., Green berg, R., Maricq, H., Whelan, J., and Seifter, J., (1981). Screening for the Early Detection of Disease in Individuals Exposed to Vinyl Chloride. Environmental Protection Agency Technical Report 560/6-81-002, U.5. ePA, Washington, D.C. 10. Espinosa, E., Caple, $., Kupchella, C. E., Chia, S., (1979). Two liver antigens undetectable in a fast frowing line of trans planted hepatomas (Morris Hepatoma 7777). Federation of American Societies for Experimental Biology (Houston, Texas). Presented at Dallas, Texas, April 7, 1979; Fed. Proc. 38:1069. CMA 003514 PROGRAM 0 HISTOLOGIC AND MORPHOMETRIC ANALYSIS: A MEANS OF ACCESSING HISTOLOGICAL HEPATIC INJURY IN CHEMICAL WORKERS; Investigators - G.H. Barrows, R. Schrodt, and C.H. Tamburro Dl. Morphometric Assessment of Histological Lesions Characteristic of Vinyl Chloride Injury D2. Computer-Assisted Morphometric Analysis as a Rapid Means of Determining Collagen Content D3. Development and Assessment of Morphometric Method of Analysis of Collagen Content from Human Liver Biopsies - Relationship to Age DA. Light Microscopic Assessment of Various Histological Lesions Found in Liver Biopsy Tissue Obtained from Vinyl Chloride Workers Background Observations made on liver biopsy specimens obtained from vinyl monomer polymerization workers have identified the association of periportal, subcapsular, and mid-zonal fibrosis, and activated sinusoidal lining cells with vinyl chloride exposure. Many have believed these findings to be early indicators of potentially dangerous chemical exposure since these histological findings are characteristic patterns in the terminal or late stages of disease. The early stages of increased collagen deposition are often difficult to identify due to a lack of data concerning the normal collagen content and normal variation of collagen in the sinusoidal spaces. Other types of chemical workers' injuries, such as copper smelters' arsenic-induced injury, are also associated with increased collagen formation especially in the perisinusoidal areas (1-3, 6-7). In addition, there had been no systematic analysis of the collagen content of the liver's parenchyma based on age. Only subjective data was available regarding "normal" increases in the quantity of fibrous tissue within the aging liver. Therefore, objective criteria were needed to determine (a) whether the increased collagen deposition in indus trial workers could be used as an indicator of developing cancer, and (b) whether this increased collagen could be identified solely on tissue obtained by liver biopsy. 56 CMA 003515 57 Industrial vinyl monomer exposure, especially vinyl chloride, has befJT associated with various hepatic histological abnormalities. These have included hyperplasia of both the hepatocytes and sinusoidal cells. Early histological studies, mainly on autopsy material, have shown focal mixed hyperplasia (hyperplasia of the hepatocytes or sinusoidal cells) to be an early histological alteration associated with vinyl chloride exposure (4,7,8). A double-blind histological assessment was conducted on the liver biopsies of some 120 exposed chemical workers to substantiate this observation and determine its potential use in medical surveillance. Assessment of the degree and duration of the vinyl chloride exposure was determined utilizing rank ordered exposure indices as reported by Greenburg and Tamburro (9). Objectives 1. Develop a clinically usable method for morphometric analysis of collagen content of liver biopsies. 2. Determine whether computer-assisted morphometric analysis could be used as the standard means of determining collagen content from human liver biopsies. 3. Morphometic assessment of normal collagen content and changes associ ated with increasing age. 4. Review and analyze light microscopic findings of liver tis^^ obtained from vinyl chloride workers to determine whether specifl) histological lesions were sufficiently characteristic to be identr^ fied and associated with vinyl chloride exposure. Research Results (1) Initially, a manual calculator-assisted approach was used for quanitation of trichrome stainable collagen in liver biopsies. Randomly selected areas of the liver biopsy were photographed. These were projected in an 16 x 16 cm area on a Hewlett-Packard 9864A digitizing plate and area cal culated using a Hewlett-Packard 9815 calculator. Measurements of collagen content and distribution of the perisinusoidal spaces were made from a ran domly selected number of complete grid sections in the various sublobular zones; they were prenumbered from left to right as shown in Figure 1A. The areas were measured in micrographs by superimposing an array of sampling points (Figure IB) and measuring the number of those points which overlaid specific areas. Two randomly selected grid areas from 3 histological, (peri portal, mid-zonal, central) and 4 anatomical locations by 2 methods were studied from 2 to 3 individuals in each bracket. Portal to central distance was quantitated and divided into 3 sections: pericentral, midzonal, and periportal. A trained observer then manually CMA 003516 57 Industrial vinyl rrcnc~er exposure, especially vinyl chloride. has been associated ,vi:n various r.spatic nistological abnormal ities. Tnese nave included nyperplasia of ootn the hepatocytes and sinusoidal cells. Early histological studies, mainly on autopsy material, have shown focal mixed nyperplasia (hyperplasia of the hepatocytes or sinusoidal cells) to be an early histological alteration associated with vinyl chloride exposure (4,7,8). A double-blind histological assessment was conducted on the liver biopsies of some 120 exposed chemical workers to substantiate this observation and determine its potential use in medical surveillance. Assessment of the degree and duration of the vinyl chloride exposure was determined utilizing rank ordered exposure indices as reported by Greenburg and Tamburro (9). Objectives 1. Develop a clinically usable method for morphometric analysis of collagen content of liver biopsies. 2. Determine whether computer-assisted morphometric analysis could be used as the standard means of determining collagen content from human liver biopsies. 3. Morphometic assessment of normal collagen content and changes associ ated with increasing age. 4. Review and analyze light microscopic findings of liver tissue obtained from vinyl chloride workers to determine whether specific histological lesions were sufficiently characteristic to be identi fied and associated with vinyl chloride exposure. Research Results (1) Initially, a manual calculator-assisted approach was used for quanitation of trichrome stainable " collagen in liver biopsies. Randomly selected areas of the liver biopsy were photographed. These were projected in an 16 x 16 cm area on a Hewlett-Packard 9864A digitizing plate and area cal culated using a Hewlett-Packard 9815 calculator. Measurements of collagen content and distribution of the perisinusoidal spaces were made from a ran domly selected number of complete grid sections in the various sublobular zones; they were prenumbered from left to right as shown in Figure 1A. The areas were measured in micrographs by superimposing an array of sampling points (Figure IB) and measuring the number of those points which overlaid specific areas. Two randomly selected grid areas from 3 histological, (peri portal, mid-zonal, central) and 4 anatomical locations by 2 methods were studied from 2 to 3 Individuals in each bracket. Portal to central distance was quantitated and divided Into 3 sections: pericentral, midzonal, and periportal. A trained observer then manually CMA 003517 58 outlined all areas staining as collagen and the total sinusoidal area (Figure 2). This was repeated for a total of 3 areas in each biopsy sample. These randomly selected areas provided sufficient sampling to obtain a reproducible estimate of collagen content of various areas of liver lobule and were highly reproducible (Table 1). Area Percent 2. 03 4.30 6.10 14.70 ftEPRODUCIBIUTY Digitizer Inter assay Intra Assay 0. 40 0.19 a, 55 1.27 0. 018 0. 052 0.054 0. 6075 Square - 0.27 0.35 1.10 2.95 CMA 003518 59 (2) Later developments for computer quantitation promise to facilitate the execution of morphometric collagen determination. In this approach a video microscopic image is digitized to 4 bit precision, and held in digital video storage (Morphometrix-156 Image Analyzer). This digital image (Figure 3A, 3B) contains all density information on a 256 x 512 matrix. The amount of staining of a particular density can be rapidly quantitated by counting the percentage of a particular density which occupies an image. This approach allows rapid quantitation of the total biopsy and enables study of a large number of samples. A second advantage is the adaptability of this technique for 3-dimensional reconstruction and quantitation of liver fibrosis volume. Figure 3 illustrates a computer printout of a histological area and a drawing of the area morphometrically analyzed. FIGURE 3a Density printout of Figure !, with 10;1 reduction (iret in square corresponds to digital printout IS). CMA 003519 59 (2) Later developments for computer quantitation promise to facilitate the execution of morphometric collagen determination. In this approach a video microscopic image is digitized to 4 bit precision, and held in digital video storage (Morphometrix-156 Image Analyzer). This digital image (Figure 3A, 3B) contains all density information on a 256 x 512 matrix. The amount of staining of a particular density can be rapidly quantitated by counting the percentage of a particular density which occupies an image. This approach allows rapid quantitation of the total biopsy and enables study of a large number of samples. A second advantage is the adaptability of this technique for 3-dimensional reconstruction and quantitation of liver fibrosis volume. Figure 3 illustrates a computer printout of a histological area and a drawing of the area morphometrically analyzed. (araa la touara eomsponda to digital printout *). CMA 003520 60 FIGURE 3b Diglul dentltle* of retlculln stained liver biopsy. (3) Histological tissue was obtained for the study of liver collagen content at various ages from individuals dying of sudden death without historical, clinical or autopsy evidence of hepatobiliary disease, chronic congestive heart failure, excessive alcohol consumption, exposure to hepatoxins or infectious agents known to produce long term hepatic injury, drug abuse, or prolonged medical use of drugs known to be hepatoxic or potentially hepatoxic. The groups were divided into four age brackets: 16-30, 31-45, 46-60, 60 and older. Autopsy tissue was obtained from areas of the liver most frequently obtained in living patients by (A) percutaneous needle, (R, L) open wedge, and (D) transjugular biopsy. Needle and wedge samples were obtained from 4 areas of the liver as illustrated in Figure 4. A wtdgt needle SITES Anterior oxillary Right lobe Left lobe Deep peri-venow* FIGURE 4 CMA 003521 61 Two hundred and fifty three light micrographs derived from 30 tissue bl^lre from 8 normal individuals were studied. A Hewlett Packard 9864-A digitizer microcomputer was used to quantitate areas of trichrome stainable collagen. Collagen estimates varied from .2 to 6.lit (mean = 1.25) with considerable variation even in different biopsies from the same patient. Subcapsular biopsies have more collagen than deep biopsies; however, deep biopsies have more variation in collagen content. Collagen content appears to increase with age and more in midzonal regions than in portal areas (Table 2). TABLE 2 PERCENT COLLAGEN RATIO IN THE VARIOUS LOBULE ZONES AGE PERICENTRAL PERIPORTAL PERICENTRAL MIDZONAL MIDZONAL PERIPORTAL IS 11.8: 1 8.6: 1 1.4: 1 23 2.8: 1 1: 1 2.4: 1 40 2.8: 1 1: 1 2.3: 1 50 1.3: 1 1: 1 1.9: 1 51 1: 1 1: 1 1: 1 m This increase appears to become detectable in the 4th and 5th deca (Figure 5). HEPATIC COLLAGEN CONTENT WITH AGE figure 5 ZO AO AGE GROUPS 50 CMA 003522 61 Two hundred and fifty three light micrographs derived from 30 tissue docks from 3 normal individuals `.are stvOiao, r\ newlett Packard 9364-A digitizer microcomputer was used to quantitate areas of trichrome stainable collagen. Collagen estimates varied from .2 to 6.1* (mean = 1.25) with considerable variation even in different biopsies from the same patient. Subcapsular biopsies have more collagen than deep biopsies; however, deep biopsies have more variation in collagen content. Collagen content appears to increase with age and more in mldzonal regions than in portal areas (Table 2). TABLE 2 PERCENT COLLAGEN RATIO IN THE VARIOUS LOBULE ZONES Aoc PER iCENTRAL PERIPORTAL MiOZONAL MIDZCNAL 15 11.3: 1 3.6: 1 23 2.3: 1 1: 1 AO 2.3: 1 1: 1 50 1.3: 1 1: 1 51 1: 1 1: 1 PERICENTRAL PERIPORTAL 1.4: 1 2.4: 1 2.3: 1 1.9: 1 1: 1 This increase appears to become detectable in the 4th and 5th decades (Figure 5). u* < LU 3 at < < 2o- HEPATIC COLLAGEN CONTENT WITH AGE ... . .............. figure 5 20 UQ ACE GROUPS SO CMA 003523 62 Differences in central, mid-zonal and portal collagen vary from 0.49 - 0.63 percent in the youngest age group and from 4 - 5.35 percent in the older age group. This histological data is consistent with the ICG clearances in normal adults which show a progressive decrease withr age. The decreased functional capacity of normal aging liver may be due to an increased collagen deposition. (4) Light microscopic assessment of liver tissue obtained from vinyl chloride workers with varying histological lesions was done. In order to substantiate the ability to identify early histological alterations indicative of chemical exposure, 93 liver biopsies from 78 individuals were investigated duplicatively in a double blind manner. These histological lesions included focal hepatocellular hyperplasia (Figure 6), focal mixed hyperplasia, sinusoidal dilitation (Figure 7), focal areas of increased reticulum (Figure 8), and focal subcapsular fibrosis (Figure 9). The progressive development of vinyl chloride liver lesions are illustrated in Table 3. Thirty-five of these individuals were exposed chemical workers with hepatic screening test abnormalities, and 13 were exposed workers without hepatic screening test abnormalities who had had liver biopsies for non-liver related medical reasons. A group of 30 individuals who were not chemical workers, who had liver biopsies for nonhepatic-related Illness during the same period of time at the same hospital, were compared. Twenty-three (48%) of the exposed workers had hepatic lesions consistent with exposure, 17 (35%) had only focal hepatocellular hyperplasia, and 6 (13%) had focal mixed hyperplasia or more advanced lesions. CLINICAL FIN0IN6S TABLE 3 PROBABLE PHASE DEVELOPMENT OF VINYL CHLORIDE PRE-CANCEROUS LESIONS NONE BIOCHEMICAL PORTAL SPLENOABNORMAL KYPER- MEGALY TENSION HEPATIC SCAN ABNORMAL KERATIS PELIOSIS y / CANCER HISTOLOGICAL NORMAL FOCAL PERI- FOCAL PORTAL AND- FHfQHHL HEBMCYTIC SINUSOOAL MIXED CAPSULAR HYPERPLASIA FIBROSIS HYPERPLASIA FIBROSIS SINUSOIDAL caL DYSPLASIA ANGIO SARCOMA CMA 003524 63 friiuguurve. 6w LFOCAL HYPERPLASIA OF HEPATOCYTES WITH GREAT VARIATIONS IN THEIR SIZE AS WELL AS THEIR NUCLEI/XTHE LATTER ARE FRE QUENTLY DOUBLE., POLYCHROMATOPHILIC OR VACUOLATED, H&E, I50x. FIGURE 8 'FOCAL INCREASE OF RETICULIN FRAMEWORK. SILVER IMPREGNA TION, lOOx. FIGURE HYPERPLASIA OF SINUSOIDAL CELL^7 (CURVED ARROWS) AND HEPATOCYTES 1 STRAIGHT ARROWS 100x3 ,,!XED HYPERPLASIA FIGURE FOCAL CAPSULAR ANO SUBCAPSULAR FIBROSIS, ANILINE BLUE, 6Gx. CMA 003525 FIGURE 6 lFOCAL hyperplasia of hepatocytes with GREAT VARIATIONS IN THEIR SIZE AS WELL AS THEIR NUCLEI.XTHE LATTER ARE FRE QUENTLY DOUBLE, POLYCHROMATOPHILIC OR vacuolated, h&e, I50x. CUsUKt O T^?C"ArLjoINSCRiE,AsS,LEv"OF RETfcCULIN * r;-. HYPERPLASIA OF SINUSOIDAL CELL^7 (CURVED ARROWS) AND HEPATOCYTES (STRAIGHT ARROWS^HlE, IflOxi ^ HYPERPUSIA FIGURE --MBMBII n i y-enrti CAPSULAR ANO SUBCAPSULAR FIBROSIS. ' ANILINE BLUE, 6Qx. CMA 003526 64 In contrast only 5 of the comparison group had similar findings. Four (13%) had focal hepatocellular hyperplasia, and 1 (3%) had focal mixed hyperplasia and sinusoidal dilitation. On subsequent biopsy this individual was found to have angiosarcoma and a history of using hair spray containing vinyl chloride as a propellent. Ten individuals had 28 biopsies reviewed double blindly, and 10 individuals had 23 readings of the same biopsy; 21 of 23 (91%) duplicate readings and 27 of 28 (96%) multiple biopsy readings in the same individuals were identical. Only 18% of the individuals (3 of 17) had either duplicate and/or multiple biopsy readings which disagreed with their prior biopsy assessment. Focal hepatocytic hyperplasia, in addition to the previously described mixed hyperplasia, appeared to be the earliest identifiable change consistent with chemical exposure. These lesions can be consistently identified and are useful in screening chemical workers for evidence of chemical exposure. Chemical exposure histories to 20 different occupational chemicals were analyzed with regard to the histological findings in the liver biopsy. Portal tract fibrosis, disruption of the limiting plate, and changes in the limiting plate of the portal area had a significant correlation in individuals identi fied as having histological features of chemical liver injury in contrast to those who had nonchemical liver disease. A larger percentage of individuals with histological evidence of chemical injury had higher levels of exposure to vinyl chloride (Figure 10). Changes in the sinusoidal areas also demonstrated the highest correlation with vinyl chloride's cumulative exposure rank months, but not with the cumulative exposure rank months of acrylonitrile, a com parison chemical. INJURY-EXPOSURE CORRELATION CLI-CHEMICAL LIVER INJURY LO- LIVER DISEASE SO-STANOARD-HORM 60 < O average vc EXPOSURE 12 3 44 RATINGS ("J!) FIGURE 10 CROUP RATINGS CMA 003527 65 Relevance to Industry These studies demonstrate some highly important and critical information regarding documentation of chemical exposure-induced liver injury: (1) Rapid methods of quantitatively assessing total collagen content in human liver tissue have been applied. These can provide both retrospective as well as prospective analysis. (2) Morphometric analysis of collagen can be performed for clinical pur poses from human liver biopsies. This can be automated by using a computer for large volume work. (3) The identification of increased liver collagen content with age demon strates the need for better standards of "normality" before one can properly determine whether increased liver collagen is a reflection of chronic injury that may be work-related. (4) There are histologically identifiable lesions that are consistent with chemical injury; these lesions can be separated from those histological findings of nonchemical liver injury. These characteristic chemical liver injury lesions are correlated with vinyl monomers, most significantly vinyl chloride. Therefore, at the present state of knowledge, one can reasonably determine the presence or absence of significant chemical exposure from th^ examination of human liver tissue. The present studies have not demonstrate^ any spontaneous progression in the earliest lesions, i.e., focal hepat^^ cellular hyperplasia when found alone. . The present opinion is that lesion may simply reflect cellular adaptation to chronic exposure while mixed hyperplasia may indicate a higher risk of future cancer development. References 1. Creech, J.L., Jr. and Johnson, M.N. (1974) Angiosarcoma of Liver in the Manufacture of Polyvinyl Chloride. J. Occup. Med. 16:150. 2. Falk, J., Creech, J.L., Jr., Heath, C.W., Jr., Johnson, M.N. and Key, M.M. (1974) Hepatic Disease Among Workers at a Vinyl Chloride Polymerization Plant. JAMA 230:59. 3. Makk, L., Creech, J.L., Jr., Whelan, J.G. and Johnson, M.N. (1974) Liver Damage and Angiosarcoma in Vinyl Chloride Workers. JAMA 230:64. 4. Popper, H. and Thomas, L.B. (1975) Alterations of Liver and Spleen Among Workers Exposed to Vinyl Chloride. Ann. N.Y. Acad. Sci. 246:172-193. 5. Makk, L., Delmore, F., Creech, J.L., Jr., et ah (1975) Clinical and Morphological Features of Hepatic Angiosarcoma in Vinyl Chloride Workers. Cancer 37:149-163. CMA 003528 CIO Relevance to Industry These studies demonstrate some hignly important and critical infc^mat';regarding documentation of chemical exposure-induced liver injury: (1) Rapid methods of quantitatively assessing total collagen content in human liver tissue have been applied. These can provide both retrospective as well as prospective analysis. (2) Morphometric analysis of collagen can be performed for clinical pur poses from human liver biopsies. This can be automated by using a computer for large volume work. (3) The identification of increased liver collagen content with age demon strates the need for better standards of "normality" before one can properly determine whether increased liver collagen is a reflection of chronic injury that may be work-related. (4) There are histologically identifiable lesions that are consistent with chemical injury; these lesions can be separated from those histological findings of nonchemical liver injury. These characteristic chemical liver injury lesions are correlated with vinyl monomers, most significantly vinyl chloride. Therefore, at the present state of knowledge, one can reasonably determine the presence or absence of significant chemical exposure from the examination of human liver tissue. The present studies have not demonstrated any spontaneous progression in the earliest lesions, i.e., focal hepato cellular hyperplasia when found alone. . The present opinion is that lesion may simply reflect cellular adaptation to chronic exposure while mixed hyperplasia may indicate a higher risk of future cancer development. References 1. Creech, J.L., Jr. and Johnson, M.N. (1974) Angiosarcoma of Liver in the Manufacture of Polyvinyl Chloride. J. Occup. Med. J6-.150. 2. Falk, J., Creech, J.L., Jr., Heath, C.W., Jr., Johnson, M.N. and Key, M.M. (1974) Hepatic Disease Among Workers at a Vinyl Chloride Polymerization Plant. JAMA 230:59. 3. Makk, L., Creech, J.L., Jr., Whelan, J.G. and Johnson, M.N. (1974) Liver Damage and Angiosarcoma in Vinyl Chloride Workers. JAMA 230:64. 4. Popper, H. and Thomas, L.B. (1975) Alterations of Liver and Spleen Among Workers Exposed to Vinyl Chloride. Ann. N.Y. Acad. Sci. 246:172-193. ---------------------------------- 5. Makk, L., Delmore, F., Creech, J.L., Jr., et aL (1975) Clinical and Morphological Features of Hepatic Angiosarcoma in Vinyl Chloride Workers. Cancer 37:149-163. CMA 003529 66 6. Creech, J.L., Jr., Makk, L., Whelan, J.G. and Tamburro, C.H. (1974) Hepatotoxicity Among Vinyl Chloride (PVC) Production Workers During First Year of Surveillance Program. Gastro. 64:786. 7. Thomas, L.B. and Popper, H. (1975) Pathology of Angiosarcoma of the Liver among Vinyl Chloride-Polyvinyl Chloride Workers. Ann. N.V. Acad. Sci. 246:268-277. ------- 8. Gedigk, P., Muller, R., and Bechtelsheimer, H. (1975) Morphology of Liver Damage among Polyvinyl Chloride Worders. A Report of 51 Cases. Ann. N.Y. Acad. Sci. 246:278-285, 1975. 9. Greenberg, R. and Tamburro, C.H. (1981) Exposure Indices for Epidemiological Surveillance of Carcinogenic Agents in an Industrial Chemical Environment. J. Occup. Med. 23:353-358. CMA 003530 PROGRAM E STUDIES OF VINYL MONOMER CHEMICALS AND THEIR METABOLITES USING CHEMICAL STRUCTURE AND SYNTHESIS IN THE DETERMINATION OF TOXICITY OF THESE AGENTS; Investigator - J. L. Wong El. Synthesis, Purification, and Utilization of Vinyl Chloride Metabolites in Mutagenicity and Carcinogenicity Studies E2. Detoxification Studies of Vinyl Chloride and Its Metabolites E3. Vinyl Chloride Metabolite Detection--Chloroacetic Acid Background The mutagenicity of vinyl chloride and its potential metabolites was studied in detail by a combination of chemical synthetic and microbiological assay techniques. We prepared and characterized the potential putative metabolites (chlorooxirane and four forms of chloroacetaldehyde), and Dr. Uldis Streips of Microbiology conducted the testing of these compounds in the Bacillus and Salmonella systems. Since all literature methods for the preparation of chlorooxirane result in mixtures with contaminants such as chlorine, hydrogen chloride, ethylene oxide, polymers, etc., it was necessary to develop a method to prepare pure chlorooxirane. In order to elucidate the intermediary metabolism of vinyl chloride, i.e., the metabolic activation and inactivation steps, to determine the significant end products of vinyl chloride in body tissues and fluids, the fate of chlorooxirane, the immediate 450 product of vinyl chloride, was studied using synthetic chlorooxirane to react with non-protein sulfhydryls. Another prominent reactive property of chlorooxirane is its facile rearrangement to chloroacetaldehyde (CAA). In dimethylformamide at 25*C, it has a half-life of "0.5 min in rearranging to CAA. Therefore, the mechanism of deactivation of CAA by cysteine was studied. The plausible cyclic product--3L-carboxy-2,3-dihydro-l,4-thiazine--may be the precursor of the urinary metabolites S-hydroxyethylcysteine and S-carboxymethylcysteine. 67 CMA 003531 68 As a putative metabolite, CAA is well-known to react with adenine and cytosine nucleotides. We will demonstrate that CAA can also react with guanine derivatives at body temperature and physiological pH to yield covalent products. These products will alter the N-H hydrogen-bonding sites in the nucleic acid bonds, hence their formation in a biological host may be consequential in inducing the onset of neoplasia. Since previous animal studies have not demonstrated chloroacetic acid in the urines of animals at low exposure levels but have been reported in humans at higher exposure levels, we intend to develop analytical methodologies which will help to verify this observation. Such studies may lead to a method of warning of environmental exposure to humans. The first target metabolite for analysis was chloroacetic acid. Objective 1. To study the chemistry of vinyl chloride toxicity by identifying the intermediates of vinyl chloride metabolism and their reactions with cytoplasmic chemicals. 2. To determine the putative actions of the primary metabolites with regard to: a) the chemical reactions of metabolites, chlorooxirane (COR) and chloroacetaldehyde (CAA) with sulfhydryls (detoxification). b) reactions on nucleic acid constituents (mutagenesis and carcinogenesis). 3. To develop analytical methods to detect a target metabolite--chloro acetic acid. El. Synthesis, Purification, and Utilization of Vinyl Chloride Metabolites in Mutagenicity and Carcinogenicity Studies The detection study of the putative action of vinyl chloride is based on the hypothesis that such action comes from the modification of the nucleic acid materials by the primary metabolites COR and CAA. Although CAA is long known to react with nucleic acid bases such as cytosine and adenine to form the etheno derivatives, little is known about the reaction of CAA on the most reactive base, guanine. Our study of CAA with guanine base using analytical tools such as HPLC, GC-MS and FT-NMR have shown that the guanosine reaction is enormously complicated. The hetero-bifunctional alkylating structures of 2-chloroacetaldehyde are shown below. CMA 003532 HI Cl-CHrC-OH OH HH CI-CHr^-O-C-CHj-Cl OH OH CICHj 69 HH Q CDrC-O-C-COj Cl OD OD' H Cl-CHiC-OCHaCH 6chch;- Studles of their reactivities have necessitated: a. Synthesis of etheno-modified components of RNA and DNA-- angularetheno-guanine, 1inear-etheno-guanine, linear-etheno-guanosine, linear-ethenodeoxy-guanosine, etheno-adenine, etheno-adenosine, etheno-deoxy-adenosine, etheno-cytosine, etheno-cytidine, and etheno-deoxy-cytidine. b. Synthesis of acidic hydrolytic products of etheno-modified bases-- 2-amlnobiimidazole, 2-formamidobiimdazole, El-D-angular-etheno-guanine, El-D-1 inear-etheno-guanine. c. Synthesis of alkaline decomposition products of etheno-modif ied nucleosides--2-amino-biimidazole nucleoside and 2-formamido-biimidazole nucle oside. d. Synthesis of the deuterated analogues for detailed structural study--8-D-etheno-adenine, El-D-etheno-cytosine, El-D-angular-etheno-guanine, El-D- etheno-adenine, and El-0-1inear-etheno-guanine. The kinetics of guanosine with CAA at wide range pH value have been studied with careful pH control. The pK vs. pH plot showed that there are two plateaus, pHl and pH8, and there is minimum at pH 4.5. From the product profiles, two different mechanisms can be suggested (Figure 1). The putative action of vinyl chloride in respect to the physicochemical properties of the etheno derivatives of cytidlne, adenosine, and guanosine is further studied. The *3C-FTNMR spectroscopic properties are particularly revealing. 8y means of deuteration and proton undercoupling technique in acquiring the carbon-13 spectra, the various carbon chemical shifts are assigned and the ^3C-1H coupling constants determined. From the latter data, the conformational preferences of the sugar moiety in the modified nucleosides are deduced as follows: (1) anti conformation for ethenocytidine CMA 003533 oH Hi Hi c: -CKs-C-0-C-CH,-a OH OH C1CH, O^O CICH^O^CHiCl HH a cdj-c-o-C-CD2 ci 00 60' H Cl-CHjC-OCHjCHj OCHjCHj' Studies of their reactivities have necessitated: a. Synthesis of etheno-modified components of RNA and DNA-- angularetheno-guanine, 1 inear-etheno-guanir.e, 1 i near-etheno-guanosine, 1 inear-ethenodeoxy-guanosine, etheno-adenine, atheno-adenosine, etheno-deoxy-adenosine, etheno-cytosine, etheno-cytidine, and etheno-deoxy-cytidine. b. Synthesis of acidic hydrolytic products of etheno-modified bases-- 2-aminobiimidazole, 2-formamidobiimdazole, l-D-angular-etheno-guanine, and E1-0-1inear-etheno-guanine. c. Synthesis of alkaline decomposition products of etheno-modified nucleosides--2-amino-biimidazole nucleoside and 2-formamido-biimidazole nucle oside. d. Synthesis of the deuterated analogues for detailed structural study--8-0-etheno-adenine, E1-0-etheno-cytosine, E1-0-angu1ar-etheno-gu anine, El--0-- etheno-adenine, and El-0-1inear-etheno-guanine. The kinetics of guanosine with CAA at wide range pH value have been studied with careful pH control. The pK vs. pH plot showed that there are two plateaus, pHl and pH8, and there is minimum at pH 4.5. From the product profiles, two different mechanisms can be suggested (Figure 1). Th* putative action of vinyl chloride in respect to the physicochemical properties of the etheno derivatives of cytldine, adenosine, and guanosine is further studied. The spectroscopic properties are particularly revealing. By means of deuteration and proton undercoupling technique in acquiring the carbon-13 spectra, the various carbon chemical shifts are assigned and the ^c_1h coupling constants determined. From the latter data, the conformational preferences of the sugar moiety in the modified nucleosides are deduced as follows: (1) anti conformation for ethenocytidine CMA 003534  ftlbmeyl S'-monooiiosBhaec rldosyl 3' ,3'-cyclic monphoBnata rtteayl 70 and (2) syn for etheno-adenosine. Since anti conformation of nucleosides is necessary for the formation and stability of the helix, the influence of the etheno bridge on the adenine nucleus in altering the ribosyl group to syn will impose considerable stress on the DNA chain which might result in biological damages. E2. Detoxification Studies of Vinyl Chloride and Its Metabolites A previous attempt by Gothe et al. to trap the putative metabolites COR and CAA in vitro from vinyl chloride with 3,4-dichlorobenzenethlol has led to the identification of the product as 3,4-dichlorophenylthloacetaldehyde. It was interpreted to be indicative of the formation of either COR or CAA. We have defined this experiment further. Thus, chlorooxirane in organic medium reacts slowly with 3,4-dichlorobenzenethlol to form 2-(3,4-d1chlorophenylthio) acetaldehydes as 80 percent#of the products. This reaction at room temperature takes about 7 days. At 60*C the reaction is completed in 21 hours giving the same products. (In both cases some disulfide Cl2PhrS-S-Ph-Cl2 and other unidentified polymeric materials are observed.) The identity of the product was established by comparison of gas chromatography retention time with an authentic example, by PMR and by mass spectrum. In aqueous acetone (2:1) or aqueous acetonitrile (3:1), the reaction is much faster. As pH - 4, 55 percent conversion is observed in 15 min. which is the lifetime of COR under CMA 003535 71 these conditions, and 80 percent conversion at pH = 7. The sole produW identified from these reactions by high pressure liquid chromatography is the aldehyde indicated. On the other hand, chloroacetaldehyde in CHCI3 gives another addition product, which was identified by PMR and IR and its facile reversion to the starting materials as 3,4-Cl2Ph-S-CH(0H)CH2Cl. In aqueous acetone or aqueous acetonitrile, CAA did not react with the thiol within 1/2 hour, indicating its lower reactivity to the aromatic SH group compared to COR. These results, combined with the in vitro experiment by Gothe, have confirmed that chlorooxirane is an obligatory intermediate in the metabolism of vinyl chloride. The detoxification of chlorooxirane was also studied in aqueous media at different pH's. N-Acetylcysteine was used as a typical cellular sulfydryl compound involved in detoxification. In aqueous solutions at pH4 and 7, their reaction was extremely fast at room temperature*, yielding N-acetylcysteine-Sacetaldehyde. It was identified by -*-H and "C NMR and characterized as the 2,4-dinitrophenylhydrazone derivative. This aldehyde is probably the precursor of the urinary metabolites, S-2-hydroxyethylcysteine and thiodiglycolic acid. The reaction of chloroacetaldehyde with N-acetylcysteine under controlled pH conditions in aqueous medium at 0*C produced an intermediate compound. Upon warming up the reaction mixture to room temperature, our previously identified thiazene was isolated as the final product. In order to elucidate the stepwise formation of the thiazene, N-acetylcysteine methyl ester was allowed to react with chloroacetaldehyde in chloroform at 0*C. The init product, plausibly the hemithioacetal H^CO-CO-CH^NHCOI^Jf^S-CHOH-CHgCl, eliminated HC1 upon neutralization with aqueous sodium hydroxide to produce the corresponding epoxide H3C0C0CH(NHC0CH3)CH2$-CH0-CH2. This structural assignment is supported by its PMR spectrum. Furthermore, when this epoxide was extracted into chloroform, it rearranged to the S-acetaldehyde H3COCOCH(NHCOCH3)CH2-S-CH2CHO, identified by its pmr spectrum and by comparison with that formed from the reaction of N-acetylcysteine with chlorooxirane. Even though chlorooxirane and chloroacetaldehyde eventually give the same final product with N-acetylcysteine, the rate of reaction and the inter mediates in the two reactions are different. The chlorooxirane conjugates instantaneously with the sulfhydryl compound, while chloroacetaldehyde takes about 2 1/2 hours for a comparable reaction. The comparative study of the reaction of chlorooxirane and chloroacetal dehyde with sulfhydryl compounds can now be summarized in the following reaction pathways with cysteines (unmodified, N-acetyl, and N-acety1-methyl ester): CMA 003536 7' - -^ i -____-- ^ i i . .''>1 = ' ,, r.. I* : T , i'Z T 'a,71 a.Is; llair.yde ir; In , - ^ -y ae c; n v* 3" i ar* 1 5 `Q ar,c:r,ar adai 7iGr. p r c c u c l wni cn rtdS "I -*cil u ' T 1 cu Z j . V t duu i% dfiG 113 raci le reversion to the starting materials as 3,4--CI^Pn--S-CH(GH) In aqueous acetone or aqueous acetonitrile, CAA did not react with the thiol within 1/2 hour, indicating its lower reactivity to the aromatic 3H group comoared to CCR. These results, corbired vith the in vitro t/.pirim, "j Gothe, have confirmed that chlorooxirane is an obligatory intermediate in the metabolism of vinyl chloride. diffeTrheentdeptoHix'si.ficaNtio-Anceotyflccyhsltoerinoeoxiwraans e was used aiso studied in aqueous media at as a typical cellular sulfydryl compound involved in detoxification. In aqueous solutions at pH4 and 7, their reaction was extremely fast at room t&fperacura yielaing !-acetyicysts'ne-S- acetaldehyae. It was identified by *H and `^C NMR and characterized as the 2,4-dinitrophenylhydrazone derivative. Tnis aloehyoe is probaDly the precursor of the urinary metabolites, S-2-hydroxyethylcysteine and thiodiglycolic acid. The reaction of chloroacetaldehyde with N-acetylcysteine under controlled pH conditions in aqueous medium at 0*C produced an intermediate compound. Upon warming up the reaction mixture to room temperature, our previously identified thiazene was isolated as the final product. In order to elucidate the stepwise formation of the thiazene, N-acetylcysteine methyl ester was allowed to react with chloroacetaldehyde in chloroform at 0*C. The initial product, plausibly the hemithioacetal H3C0-C0-CH-(NHC0CH3)CH2S-CH0H-CH2C1, eliminated HC1 upon neutralization with aqueous sodium hydroxide to produce the corresponding epoxide H3C0C0CH(NHC0CH3)CH2S-CH0-CH2. This structural assignment is supported by its PMR spectrum. Furthermore, when this epoxide was extracted into chloroform, it rearranged to the S-acetaldehyde H3COCOCH- (NHCOCH3)CH2-S-CH2CHO, identified by its pmr spectrum and by comparison with that formed from the reaction of N-acetylcysteine with chlorooxirane. Even though chlorooxirane and chloroacetaldehyde eventually give the same final product with N-acetylcysteine, the rate of reaction and the inter mediates in the two reactions are different. The chlorooxirane conjugates instantaneously with the sulfhydryl compound, while chloroacetaldehyde takes about Z 1/2 hours for a conparable reaction. The conparative study of the reaction of chlorooxirane and chloroacetal dehyde with sulfhydryl compounds can now be summarized in the following reaction pathways with cysteines (unmodified, N-acetyl, and N-acetyl-methyl ester): CMA 003537 72 X-NH-CH-COOR h2sh Cl-CH2CHO X-NH-CH-COOR ch2-s-ch2cho XI coop X-NH-CH-COOR -------------CH2-S-CH-CHpCl 6h X=H, Ac R=H, Me X-NH-CH-COOR CH2S-CH-pH2 Although both routes are converged to yield the cysteine S-acetaldehyde con jugate, their reaction rates are vastly different. In general, chlorooxirane completes its reaction within an hour while chloroacetaldehyde requires over night. These chemical observations can be applied in (1) specific assays for the two putative metabolites in cellular studies of vinyl chloride, and (2) detoxification mechanism of cells exposed to the vinyl carcinogen. E3. Vinyl Chloride Metabolite Detection - Chloroacetic Acid The methodology includes gas chromatography and mass spectrometry. Gas chromatography results are as follows. Solid supports containing Carbowax and FFAP liquid phases have been used for direct analysis of carboxylic acids. Accordingly, 20 percent FFAP on chromosorb W passes chloroacetic acid at 200* with modest tailing of the peak. However, with this simple column, sensi tivity is limited to concentration ranges of 0.2 - 2 mg/ml HgO (200 - 2,000 ppm). Application of the improvements reported by Ackman (use of porous poly mer solid support and use of formic acid in helium carrier gas) is expected to improve sensitivity by 100 fold. Identification of chloroacetic acid by mass spectrometry was carried out. Chloroacetic acid can be identified by mass spectroscopy of water solutions as dilute as 0.02M (2mg/ml). The 2 doublets of peaks at m/e 49, 51 (35qch i and 37c1XH2+,) and 50,52 (^cio^t and 37cicH3?) are distinctive because rela tive intensities within each doublet reflect the 76:24 isotopic ratio of 35ci:37ci. Thus, the GC-MS combination technique should be most useful in the detection of the above. CMA 003538 73 Relevance of Industry Broadly speaking, this is a study for early detection and prevention of industrial cancers. Our chemical methodologies (synthesis, structure, and analysis), aplied as an integral part of the multidisciplinary approach, contribute to elucidate specific molecular events in the effects of vinyl monomers on industrial workers. This information will form a rational basis for safer use of chemicals and design of preventive measures. Our molecular studies also provide the opportunity to develop useful marker(s) in the form of metabolites in the pathogenesis of chemical injury. References Elmore, J. D., Wong, J. L., Laumbach, A. D. and Streips, U. N. (1976) Vinyl chloride mutagenicity via the metabolites chlorooxirane and chloroacetaldehyde monomer hydrate. Biochem. Biophys. Acta, 442, 405. Laumbach, A. D., Lee, S., Wong, J. L. and Streips, U. N. (1977) Studies on the mutagenicity of vinyl chloride metabolites and related chemicals. Prev. Detect. Cancer (Proc. Int. Symp), 3rd 1^, 155. Laumbach, A. D., Streips, U. N. and Wong, J. L. (1978) Chloroacetal dehyde- induced damage to Bacillus subtil is. Abs. Ann. Mtq. ASM, p. 125. Laumbach, A. 0., Streips, U. N. and Wong, J. L. (1979) Chloroacetal dehyde- induced damage to Bacillus subtills, Abs. Twelfth Internat. Congress Microbiol. Joseph, 0. T. and Wong, J. L. (1982) Model study of vinyl chloride metabolism sulfate reactions with chlorooxirane, Carcinoqenesis (in preparation). CMA 003539 PROGRAM F ASSESSMENT OF ASSAYS FOR THE CARCINOGENIC POTENTIAL OF INDUSTRIAL CHEMICALS USING PROKARYTOIC AND EUKARYOTIC SYSTEMS. Investigators; U.N. Streips and G, Sonnenfeld FI. Bacterial Assays for Testing Carcinogenicity and Mutagenicity of Indus trial Chemicals. Background Our laboratory had instituted Salmonel1 a reversion, Bacillus subtilis repair, and Bacillus subtilis forward mutations assays for screening the mutagenic potential of chemicals (1-3). For that reason we became an integral part, first, for B. F. Goodrich, then for the Manufacturing Chemists Association sponsored projects on industrial chemical risk assessment. Objective 1. To screen a series of industrial and environmental - health-related chemicals for mutagenic potential. 2. To initiate with Dr. G. Sonnenfeld a rapid mammalian assay to better predict human-related bacterial chemical activity and carcinogenic potential. Research Results In our previous progress reports we have documented the work which delineated the involvement of chloroacetaldehyde and chlorooxirane as the probable active components of vinyl chloride-mediated carcinogenesis. In addition, we screened a multitude of other chemicals generated by Dr. John Wong (our collaborator in this study) or by other investigators in the university or industry (1-3). This is documented in Table 1. Recently, in collaboration with Dr. Gerald Sonnenfeld, a new, exciting, potential screen for carcinogens was developed (4,5). This involved the inhibition of inter feron induction and was specific for several carcinogens (including chloro acetaldehyde) but not for non-carcinogenic analogues (including some muta gens). It should be illustrative of some of the molecular mechanisms of carcinogenesis attack on susceptible cells. We continue to develop this system even though our support from CMA has ended (6-8).. 74 CMA 003540 A screening laboratory (using microbial assays) for institutional service purposes is being maintained. We have proposed to the state of Kentucky that a regional chemical testing laboratory be funded to serve the industry and university needs in this part of the Commonwealth. That proposal is still pending. TABLE 1 siMun or ninittt substances tested to* hotacenicitt Salmonella 1. + 2a Benzoyl peroxide 3* N-icecoxy^N-phenylacatimlde 4* N-eyelobuCftnecarboxyloxy*N-phanylcaciftidt 3, ble-cyclobuttn* carboxyl peroxide 6, Aflatoxlo 7. Chloroeehanol 6, Beazo(*)pyrtne 9. 4-nltrO qwlnollne-loxtde 10. 11. Eplchlofohydrln 12. Qilorooxirane 13. Butane dlepoxld* 14. Styrene oxide 13. 3*4 epoxide butene NR NR NR NR 44+ NR 44 +++ 444 44 444 44 44 4 16. blx(Bece-ehloroechyl)phenyl phosphate 17. Bet* chloroechyl phosphite (bis cyelohexylamina Alt) u. Urn| iiplretee from smoker* with cancer 19. Lunp iipinces from smokers without cancer NR 4 4 W - mm react lorn * * mutstmlc +++ - extremely mutatmmlc - - **t4mIU* M**e*ielty ++ - stronuly Mjtefmmle Subclll* NK NR NR nr nr ++ nr nr HI Hi To further elucidate molecular mechanisms of carcinogenesis, in collaboration with John Wong, we initltated a study of DNA-interaction with carcinogens, and with Ron Doyle, control of cell division processes in bacteria. Progress in this area has allowed us to secure a grant from the National Science Foundation to carry on with the work. Soon, we will be able to describe the cellular controls which maintain proper cell division in bacteria, then relate these findings to manmallan cells* At that time, we will be able to examine carcinogen attack on these processes and determine which events lead to loss of cell division control and neoplastic growth. CMA 003541 n scrs;n 'c i .,,'v s purposes is oeir.i ns-rra''-so. = ve a regional chenica' :3S-.'i"g ' " ' >* 1 " university neeas in phis pan of tne pending. TMi , ^ ^ ^ ^ /*> Commonwealth. :^ ` St^t3 ^ V* ' * O That proposa p'< v *r ar is still TABLE 1 s-jkiak-t of ?f*T'.;ii;fr ins:i,ic;s nsrvo -o. "yncKitcirr 1. *ti-UfoSenjoyL-cyclaDucanGir!>ony! peroxide * 2. Senjovl ptr3*'.4 3. .X-cvelaegtanecae^oicy Loxy-S-o^s-tv i ice-.a-.id* 5. 5i*-cye+25ucane carsoicv; perjxid* 6. Aflacaxin SX >> SX VS 7. Chlocoethenol S- lartao^aipyrt-'e NX 9, A*-i;ro quino. ln-L*oxid* 10. Oiloroaeecaidehyde 11. Splchlorehydrln 12. Qiloreexiran* ++ 13. IvtiM dlipaildt 14. Styrene avid# 13. 1,4 epoxide butene +* r U- hi*(8eca-cnlofeethyl)Phenyl phosphate 17. Seta ehloroetnrl pho*phca (m* cvclohexvlanine s*it) 18. Ung a*plr*ca* free awoke r* with cancer NX NX If. Irimtfi aaplrata* from twoker* without cancer : " * ~ "" * - iwctiMlc * - turwlf <<(< - - Wr^nin MiMKlilty +* - itranitlf ainiMlc NX NX v? NX NR ++ NX * M. * ++ NR NX NX NX NX To further elucidate molecular mechanisms of carcinogenesis, in collaboration with John Wong, we inititated a study of DNA-Interaction with carcinogens, and with Ron Doyle, control of cell division processes in bacteria. Progress in this area has allowed us to secure a grant from the National Science Foundation to carry on with the work. Soon, we will be able to describe the cellular controls which maintain proper cell division in bacteria, then relate these findings to mammalian cells. At that time, we will be able to examine carcinogen attack on these processes and determine which events lead to loss of cell division control and neoplastic growth. 0035^2 76 Support from the CMA has been instrumental in generating and maintaining research activity in our laboratory relating to carcinogenesis and cell cycle events. We deeply appreciate this support. Relevance to Industry With the tremendous number of chemical formulations generated yearly, a battery of rapid screens are crucial for examining the environmental and health-related impact of these chemicals. A regional laboratory can do such screening effectively and economically as shown by the results previously described. In addition we have demonstrated that such research also leads to innovation in terms of new assays (the interferon assay), and related research (the studies on cell division). Industrial support also has led to other granting. Valuable mutagenesis information has been provided to industry regarding crucial chemicals. In turn, industry has provided us with the opportunity to expand and solidify our research efforts. The interaction has been optimal. F2. Preliminary Studies on the Use of Inhibition of Interferon Induction as an Indicator of Mutagenicity and Carcinogenicity of Chemicals Background Interferon, which was originally described as an antiviral agent, has since been shown to have several other activities, including an antitumor activity (9). Clinical trials are now in progress to determine the efficacy of interferon as an anti-cancer agent (10). We became interested in the interactions of carcinogens, or cancercausing substances, and interferon. Early studies by DeMaeyer and others suggested that there was indeed an interaction between carcinogens and interferon induction (11-13). Pretreatment of rat embryo fibroblasts with the carcinogens triethyleneamine, 4-nitroquinoline-N-oxide, benzo-(a)-pyrene, and 3-methylcholanthrene all inhibited the induction of alpha-beta interferon by viruses, without affecting the gross viability of the cultures or the replication of the viruses (11-13). Several poorly or nonearcinogenic analogs, such as benzo-(e)-pyrene, had no effect on the production of interferon (11-13). The effect was observed after only one day of incubation of very small amounts of carcinogen with the tissue cultures. In addition, Hahon and co- workers have shown that incubation of cell cultures with various forms of aflatoxin also inhibited the induction of interferon in the cultures (14). In this case, the degree of inhibition correlated with the in vivo carcinogenic potential of the aflatoxin form (14). Asbestos fibers a!T3 coal dust have also been reported to inhibit interferon induction in tissue culture (15,16). In vivo, injection of methylcholanthrene into mice inhibited the induction oT alpha-beta interferon (17). These results created an interest in our laboratory to further study this phenomenon. CMA 003543 77 Objective 1. To repeat the early studies to determine if carcinogen pretreatment inhibited interferon induction. 2. To begin studies to determine if effects on interferon induction could be used as a reliable indicator of carcinogenic potential of chemicals. 3. To study the mechanisms of the observed effects. Research Results Our first step was to establish the system in a mouse culture model, since this was more appropriate to current interferon technology. Mouse embryo tissue cultures were prepared, treated with carcinogen or analogue for 24 hr., washed with fetal bovine serum-containing medium and with serum-free medium to help remove carcinogen, challenged with an interferon inducer (either Newcastle disease virus or poly I:C), and then assayed for antiviral activity. Earlier data was able to be duplicated and those findings extended. Several known carcinogens including 7,12-dimethylbenz-(a)-anthracene, benzo-(a)-pyrene, 2-aminofluroene, aflatoxin-B] and the #4 fraction of tobacco smoke condensate all inhibited interferon induction (5). Styrene oxide, an important industrial chemical which was positive in the Ames Salmo nella assay but negative in all Bacillus assays done to date (11), Se inhibited interferon induction. Ethyl methanesulfonate (EMS) and me^P methanesulfonate, differ with respect to their carcinogenic potential as determined by tumor induction (18,19). MMS is a mutagen and highly carcino genic, while EMS is a potent mutagen but rarely carcinogenic. In our assay, MMS strongly inhibited alpha/beta interferon induction, while EMS had no effect. Additional studies in our laboratory extended these findings to the potent carcinogen chloroacetaldehyde and its noncarcinogenic analogues chloroethanol and chloroacetic acid (4). Only chloroacetaldehyde inhibited inter feron induction. An additional finding of interest was the apparent requirement for activation of carcinogens before they could inhibit interferon induction. In a preliminary study, we were able to show that addition of reduced glutathione to our cultures blocked the inhibition of Interferon induction by benzo-(a)pyrene (4). Reduced glutathione- can block the active metabolites of carcinogens (20), which suggested that these active metabolites were required for the effects of benzo-(a)-pyrene on interferon induction. Relevance to Industry We believe these studies to be highly relevant to the chemical industry because: 1) the studies yielded information on the effects of toxic chemicals on the interferon system, an important body defense against cancer, and 2) the studies may eventually lead to a new assay system for discriminating between carcinogenic and nonearcinogenic chemicals. CMA 003544 77 i i - ' ~ ~ ' -' - ' -,-^an '' = tr*eatme" t i pi n i u) j i n c ^ r r c r c ^ n uuC D i o n . 2. To begin studies to determine if effects on interferon induction could be used as a reliable indicator of carcinogenic potential of C'fi 3'7i i r i 1 S 3. To study the mechanisms of the observed effects. Research Results Our first step was to establish the system in a mouse culture model, since snis was more appropriate to current irtarfercn tec.nnology. Mouse embryo tissue cultures were prepared, treated with carcinogen or analogue for 24 hr., washea with fetal Devine serum-containing medium and with serum-free medium to help remove carcinogen, challenged with an interferon inducer (either Newcastle disease virus or poly I:C), and then assayed for antiviral activity. Earlier data was able to be duplicated and those findings extended. Several known carcinogens including 7,12-dimethylbenz-(a)-anthracene, benzo-(a)-pyrene, 2-aminofluroene, aflatoxin-Bi and the #4 fraction of tobacco smoke condensate all inhibited interferon induction (5). Styrene oxide, an important industrial chemical which was positive in the Ames Salmo nella assay but negative in all Baci 11 us assays done to date (11), also inhibited interferon induction. Ethyl methanesulfonate (EMS) and methyl methanesulfonate, differ with respect to their carcinogenic potential as determined by tumor induction (13,19). MMS is a mutagen and highly carcino genic, while EMS is a potent mutagen but rarely carcinogenic. In our assay, MMS strongly inhibited alpha/beta interferon induction, while EMS had no effect. Additional studies in our laboratory extended these findings to the potent carcinogen chloroacetaldehyde and its noncarcinogenic analogues chloroethanol and chloroacetic acid (4). Only chloroacetaldehyde inhibited inter feron induction. An additional finding of interest was the apparent requirement for activation of carcinogens before they could inhibit interferon Induction. In a preliminary study, we were able to show that addition of reduced glutathione to our cultures blocked the inhibition of interferon induction by benzo-(a)pyrene (4). Reduced glutathione can block the active metabolites of carcinogens (20), which suggested that these active metabolites were required for the effects of benzo-(a)-pyrene on interferon induction. Relevance to Industry We believe these studies to be highly relevant to the chemical industry because: 1) the studies yielded information on the effects of toxic chemicals on the interferon system, an important body defense against cancer, and 2) the studies may eventually lead to a new assay system for discriminating between carcinogenic and noncarcinogenic chemicals. CMA 003545 78 References 1. Horowitz, S., Doyle, R.O., and U.N. Streips. (1978) Restricted chromosome-membrane association in a stable L-form of Bacillus subti1 is. In Transformation-1978. Proceedings of the Fourth European Meeting on Bacterial Transformation and Transfection, pp. 245-251. 2. Horowitz, $., Doyle, R.J., Young, F.E., and U.N. Streips. (1979) Selective association of the chromosome with membrane in a stable L-form of Bacillus subtilis. J. Bacteriol. 138:915-922. 3. Streips, U.N., Horowitz, S., and R.J. Doyle. (1980) Genetic analysis of DNA-surface interactions in Bacillus subtilis. Microbiology 1980, ASM Press, pp. 284-287. In 4. Sonnenfeld, G., Barnes, M.C., Schooler, J., and U.N. Streips. (1980) Inhibition of interferon induction as a secreen for the carcino genic potential of chemicals. J[ Interferon: Properties and Clinical Uses. Leland-Fikes Foundation Press, Dallas, pp. 589. 5. Barnes, M.D., Streips, U.N., and G. Sonnenfeld. (1981) Effect of carcinogens and analogs on interferon induction. Oncology, 38:98-101. 6. Streips, U.N., Laumbach, A.D., and R.E. Yasbin. (1981) Bacterial mutation monitors for active metabolites of clinical carcinogens: Bac i11 us subtilis assays for mutation and repair In Microbial Testers for Chemical Carcinogenesis, I.C. Felkner, e3., Marcel Dekker, New York, pp. 131-145. 7. Tamburro, C.H., Wong, J.L., and U.N. Streips. (1982) Approaches to occupational cancers. In Clinical Medicine for the Occupational Physician, Alderman, an<T"Hanley (eds.), Marcel Dekker, New York, pp. 255-295. 8. Clarke, J.S., Streips, U.N., Hoffman, J.L., Fok, F.F., and J.A. Yankeelove. (1980) J. Toxicology and Environmental Health (submitted). 9. Gresser, I. Cell. Immunol., (1977) Commentary on the varied biologic effects of interferon 34:406-416. 10. Merigan, T.C. (1982) Interferon Therapy in Human Viral Infections and Malignant Disease, pp. 88-90, Stiehm, E.R. Moderator, Ann. Intern. Med. 96:80-93. ^ 003546 79 11. DeMaeyer, E., and DeMaeyer-Guignard, J. (1964) Inhibition by 3-methylcholanthrene of interferon formation in rat embryo cells infected with Sindbis virus. J. Natl. Cancer Inst. 32:1317. 12. DeMaeyer-Guignard, J., and DeMaeyer, E. (1965) Effects of carcinogenic and noncarcinogenic hydrocarbons on interferon synthesis and virus plaque development. J. Natl. Cancer Inst. 34:264. ------------------------------------ 13. DeMaeyer-Guignard, J., and DeMaeyer, E. (1967) Inhibition of interferon synthesis by triethylenemelamine and 4-nitroquinoline-N-oxide. Arch. Gesamte Virusforch. 22:61-68. 14. Hahon, N., et al. (1979) Aflatoxin inhibition of viral interferon induction. Antimicrob. Agents Chemother. 16:277-282. 15. Hahon, N. (1974) Depression of viral interferon induction in cell monolayers by coal dust. Br. J. Indus. Med. 21:201-208. 16. Hahon, N., and Eckert, H.L. (1976) Depression of viral interferon induction in cell monolayers by asbestos. Environ. Res. 11:52-65. 17. Waker, A., Feller, H., and Gericke, Da Hemmung. (1971) Der inter feron-induction durch 20-methylcholanthrene in vivo. Naturwissenschaften 58:274-275. 18. Roe, F.L.C., et al. (1963) Tests for carcinogenesis using newborn mice: 1,2-benzanthracene, 2-napthylamine, 2-naphthyl hydroxyl amine and ethyl methanesulfonate. Br. J. Cancer 17:255-260. 19. Casto, B.C., et. al. (1977) Development of a focus assay for trans formation of hamster cells vn v~>tr0 by chemical carcinogens. Cancer Res. 37:3508-3515. 20. Chasseaud, L.F. (1979) Glutathione (reduced) and glutathione-Stransferase block the carcinogenic event by trapping the ultimate metabolites. Adv. Cancer Res. 29:176-255. CMA 003547 / Ci 11 infected witn SincDis /irus. j. Natl. Cancer Inst. 32.: 13i7. 12. DeMaeyer-Guignard, J., and DeMaeyer, E. (1965) Effects of carcinogenic and noncarcinogenic hydrocarbons on interferon synthesis and virus plaque development. J. Natl. Cancer Inst. 34:264. ------------------------------------ 13. DeMaeyer-Guignard, J., and DeMaeyer, E. (1967) Inhibition of interferon synthesis by triethylenemelamine and 4-nitroquinoline-N-oxide. Arch. Gesamte Virusforch. 22:61-68. 14. Hahon, N., et al. (1979) Aflatoxin inhibition of viral interferon induction. Antlmicrcb. Agents Chamother. 16:277-282. 15. Hahon, N. (1974) Depression of viral interferon induction in cell monolayers by coal dust. 8r. 3. Indus. Med. 3J_:201-208. 16. Hahon, N., and Eckert, H.l. (1976) Depression of viral interferon induction in cell monolayers by asbestos. Environ. Res. 11:52-65. 17. Waker, A., Feller, H., and Gericke, Da Hemmung. (1971) Der inter feron-induction durch 20-methylcholanthrene in vivo. Naturwissenschaften 58:274-275. 18. Roe, F.L.C., et al. (1963) Tests for carcinogenesis using newborn mice: 1,2-benzanthracene, 2-napthylamine, 2-naphthylhydroxylamine and ethyl methanesulfonate. Br. J, Cancer 17:255-260. 19. Casto, B.C., et. al. (1977) Development of a focus assay for trans formation of hamster cells ^n vitro by chemical carcinogens. Cancer Res. 37:3508-3515. 20. Chasseaud, L.F. (1979) Glutathione (reduced) and glutathione-Stransferase block the carcinogenic event by trapping the ultimate metabolites. Adv. Cancer Res. 29:176-255. CMA 003548 PROGRAM G THE STUDY OF TISSUE ANTIGENS AND ANTIBODIES IN THE DETECTION OF VINYL CHLORIDE INJURY; Enrique Espinosa, M.D. Gl. The Study of Tissue Antigens from Liver Tumors, Angiosarcoma, and Hepatomas. G2. Circulating Antigens and Autoantibodies in Vinyl Chloride-Associated Liver Disease Background In hepatic fibrosis and angiosarcoma associated with vinyl chloride exposure of industrial workers, manifestations of the disease could not be detected in most cases until the process was far advanced (1). Normal values of liver function tests were reported in a case with significant vinyl chloride hepatic fibrosis (2), and only a small percentage of workers of a plant unit where 7 cases of liver angiosarcoma were diagnosed had abnormal blood screening tests (3). Thus, conventional liver function tests do not appear to be sensitive indicators of vinyl chloride liver disease. Develop ment of more sensitive and specific methods for detecting the disease in early stages would be of great importance. An approach to this may be provided by antigenic studies in view of the tissue antigenic modifications which are known to occur in neoplasia. For example, in human carcinoma, loss of antigens have been reported in squamous cell carcinoma (4,5) and in ovarian carcinoma (6,7), loss of the ABH blood group isoantigens in some solid tumors (8,9), and of HL-A isoantigen in lymphoma (10). Also, tumor-specific trans plantation antigens have been demonstrated in a number of experimentally induced tumors (11-14) as well as tumor-associated antigen in spontaneous tumors in man (e.g. 15-18). In such an antigenic study of liver angiosarcoma the question arises whether the fibrotic and angiosarcomatous livers contain antigens that are quantitatively or qualitatively different from those present in normal tissue and whether such changes could stimulate an immunologic response. If such hypothetical changes are demonstrable and proved to be specific, they may be of use in diagnosis of vinyl chloride related liver disease. Objective 1. To test for liver antigenic changes in a) angiosarcomatous liver tissue, b) chemically-induced hepatomas, and c) cultured human liver carcinoma cells. 80 CMA 003549 81 2. To test for abnormal serum antigens and autoantibodies in patients with histories of vinyl chloride exposure including individuals with liver angiosarcoma, liver dysfunction with fibrosis, and with normal liver function tests. Antigens studied included liver specific antigen (LSA) (19), and F-antigen (20,21), tissue antigens of wide organ distribution (22,23), and bile antigens (24). Autoantibodies were to nuclei, mitochondria and smooth muscle. Patients' sera were also tested by immunofluorescence for possible reactivity with rats exposed to vinyl chloride (25). Research Results 1. Liver antigenic changes a) angiosarcomatous liver tissue Liver angiosarcomatous tissue was analyzed in this work for presence of neoantigens, normal tissue antigens and tumor bound imminoglobulins. Several normal tissue antigens were found by immunodiffusion to be present in the tumor. These included the previously described liver-specific antigen, (LSA) (19) and antigens which are shared with several tissues (22,23). In addition, a tumor associated protein antigen and a glycoprotein liver antigen found to be absent in angiosarcoma were identified and characterized (25). Tissue distribution of liver angiosarcoma related antigen is shown in Table 1. Preparations containing tumor associated antigen were found to inhibit lymphocyte transformation tests in vinyl chloride workers suggesting that these workers have a decreased lymphocyte response to these preparations (26). Further, antigen coagulation Factor VIII was detected in the angio sarcoma cells, pointing to an endothelial cell origin of the tumor (25). Finally, irmiunoglobulin G bound to the angiosarcomatous tissue was demon strated by immunofluorescence and elution experiments suggesting antibody stimulation by the tumor (25). TABLE 1 TISSU* DISTPI POTION OP LIVE* ANOIOSARCOMA-ftELATID ANTIGEN TISSUI EXTRACT U10 AS ASSOMBNT A NOtOSARCOMA- RttLATBD A NT10EN* Live* Md*p Spleen Leap Atl*n|t*#*f ti itivm i tbeorbeO till IpepblttseO llnii extreeie* 100 i4 tested iiiltcirciiM iilriei. f illl| It4|a(t4 bp a. CMA 003550 82 b) chemically-induced hepatomas Our finding of an antigen missing in VC-related liver angiosarcoma stimulated further studies of antigenic deletion in chemically-induced hepatomas and in cultured human liver carcinoma cells. In studies performed with the fast growing and undifferentiated chemically-induced Morris hepatoma 7777, 2 liver antigens were found to be absent. These antigens were characterized and partially isolated. In studies of their occurrence in other tissues, one of these antigens (Antigen I) was shown to be present in kidney and spleen in addition to liver. The second antigen (Antigen II) was detected only in liver. Antigen II was found unrelated to liver-specific F-antigen, differing in a number of properties and in immunologic reactivity. In studies of their subcellular distribution in normal liver. Antigen I appeared localized in cytosol (54%) and mitochondrial (38%) fractions. Antigen II was about equally distributed in cytosol, mitochondria and nuclei fractions with little amounts in microsomes. Antigen I has a electrophoretic mobility in immunoelectrophoresis close to that of serum gamma-globulins and Antigen II to that of serum alpha-globulins. The two antigens were completely inactivated with Pronase indicating that both antigens are proteins or protein associated. Both antigens were relatively thermolabile; they were partially inactivated following incubation at 56C and completely inactivated at higher temperatures. Both antigens were completely inactivated when incubated it^H buffer lower than 3.5. In Sephadex-G200 gel filtration. Antigen I behflu like a protein of approximately 51,000 Daltons, using as standards s^m albumin, ovalbumin, chymotrypsinogen and ribonuclease. The molecular size of Antigen II (determined on a Bio-gel A5m column) was approximately 240,000 Daltons, using aldolase, catalase and ferritin as markers. The two antigens were found in the more differentiated and slower growing hepatomas 5123tc and 9618A at about the same concentration as normal liver. The fact that hepatoma 7777 is the fastest growing and least differentiated of the tumors studied suggests a possible functional relationship between the absent antigens and these properties (28). These antigenic deletions may be used as indicators in the early detection of liver tumors and in the evaluation of the rate of growth, histologic differentiation and metastatic properties of such hepatomas. Another liver constituent which may serve as a sensitive indicator of chemically-induced liver tumors, liver-specific F-antigen, was studied. In studies on the behavior of this antigen in Morris hepatomas, it was found that different types of these chemically-induced tumors have quite different levels of F-antigen. F-antigen appeared to be absent in the fast growing hepatoma 7777. In the slow growing hepatoma 9618A, the concentration was very low ranging from less than 2% to 10% of the normal liver concentration. The medium growing hepatoma, 5123tc, highly metastatic, had about twice the concentration as CMA 003551 nan a , - r s; - j n cultured f zn l i ^ z n i c ^ 3 i w c i c n in Cfififfl < C 3. \ i y - 1HGUC60 1 iver car cinema calls. In studies performed witn the fast growing and undifferentiated chemical'v-1rdvced '*c,"r'5 ^ectfcmi 7777> 2 'i''ar ant'-gans //ere found do absent. These antigens were characterized and partially isolated. In studies of their occurrence in other tissues, one of these antigens (Antigen I) was snown to be present in kidney and spleen in addition to liver. The second antigen (Antigen II) was detected only in liver. Antigen II was found unrelated to liver-specific F-antigen, differing in a number of properties and in immunologic reactivity. In studies of cneir suoceiiular distribution in normal liver. Antigen I appeared localized in CjCosoi (54*) ana mitochondrial (38%) fractions. Antigen II was about eaually distributed in cytosol, mitochondria and nuclei fractions with little amounts in microsomes. Antigen I has a electrophoretic mobility in immunoelectrooho'-esis close to that of serum gaimta-globul ins and Antigen II to that of serum alpha-globulins. The two antigens were completely inactivated with Pronese indicating that both antigens are proteins or protein associated. Both antigens were relatively thermolabile; they were partially inactivated following incubation at 56C and completely inactivated at higher temperatures. Both antigens were completely inactivated when incubated in pH buffer lower than 3.5. In Sephadex-6200 gel filtration. Antigen I behaved like a protein of approximately 51,000 Daltons, using as standards serum albumin, ovalbumin, chymotrypsinogen and ribonuclease. The molecular size of Antigen II (determined on a Bio-gel A5m column) was approximately 240,000 Daltons, using aldolase, catalase and ferritin as markers. The two antigens were found in the more differentiated and slower growing hepatomas 5123tc and 9618A at about the same concentration as normal liver. The fact that hepatoma 7777 is the fastest growing and least differentiated of the tumors studied suggests a possible functional relationship between the absent antigens and these properties (28). These aatlgenlc deletions may be used as Indicators In the early detection of liver tumors and in the evaluation of the rate of growth, histologic differentiation and metastatic properties of such hepatomas. Another liver constituent which may serve as a sensitive indicator of chemically-induced liver tumors, liver-specific F-antigen, was studied. In studies on the behavior of this antigen in Morris hepatomas, it was found that different types of these chemically-induced tumors have quite different levels of F-antigen. F-antigen appeared to be absent in the fast growing hepatoma 7777. In the slow growing hepatoma 9618A, the concentration was very low ranging from less than 2% to 10% of the normal liver concentration. The medium growing hepatoma, 5123tc, highly metastatic, had about twice the concentration as CMA 003552 83 normal liver. F-antigen of hepatoma 5123tc and of normal liver were found localized in the cytosol subcellular fraction and were determined to be immunologically identical and to have equivalent electrophoretic mobility and molecular weight (29). Since the antigen was undetectable in the fast growing hepatoma and undetectable or very low in the slow hepatoma, the level of F-antigen does not appear to correlate with the rate of growth of these tumors. A possible relationship between metastatic properties and F-antigen is now being considered because the hepatoma with the increased concentration of F-antigen was by far the most highly metastatic. This may prove useful in treatment of tumors. c) cultured human hepatoma cells In studies on cultured human liver carcinoma cells (after establishing optimal conditions required for the maintenance in serum free media of PLC/PRF/5 human liver carcinoma cells) it was determined that these hepatoma cells, similar to the experimental Morris hepatoma 7777, are deficient in liver-specific F-antigen. Nevertheless, these cells, like normal liver cells, produce serum albumin, fibrinogen, transferrin, alpha-1 antitrypsin and alpha-2 macroglobulin as shown by immunodiffusion (30,31) and immunofluorescence (32). These data add further support to the clinical observation that tissue antigens are more useful for treatment and follow-up care than screening and early detection, and that antigenic deletions may prove useful in early screening. 2. Abnormal Serum Antigens and Autoantibodies Serum autoantibodies to nuclei,mitochondria and smooth muscle were negative in all patients examined. In addition, serum from these patients did not show reactivity with liver from rats exposed to vinyl chloride. Liver-specific antigen LSA (19), bile antigens (24) and other tissue antigens (22, 23) associated with liver damage were not detected in these patients (25). Relevance to Industry These studies are relevant to chemical industry in view of the evidence provided shotting Important changes in the levels of angiosarcomatous liverantigenic coiponents In association with vinyl chloride exposure. Deletion of liver antigens were also demonstrated In chemically-induced hepatoma. In addition, these observations may eventually prove useful in the development of new approaches for screening of chemical-associated tumors. References 1. Heath, C.W., Jr., Falk, H., and Creech, J.L. (1975) Characteristics of cases of angiosarcoma of the liver among vinyl chloride workers in the United States. Ann. N.Y. Scl., 246:231. CMA 003553 84 .2 Berk, P.D., Martin, J.F., and Waggoner, J.G. (1975) Persistence of vinyl chloride-indeuced liver injury after cessation of exposure. Ann. N.Y. Acad. Sci., 246:70. 3. Wyatt, R.H., Kotchen, J.M., Hochstrasser, O.L., Buchanan, J.W., Jr., Campbell, D.R., Slaughter, J.C., and Doll, A.H. (1975) An epidemiologic study of blood screening tests and illness histories among chemical workers involved in the manufacture of polyvinyl chloride. Ann. N.Y. Acad. Sci., 246:80. 4. Nairn, R.C., Richmond, H.G., McEntegart, M.G., and Fothergill, J.E. (1960) Immunological difference between normal and malignant cells. Br. Med. J., 2^:1355. 5. Carruthers, C., and Baumler, A. (1965) Immunochemical staining with fluorescein-labeled antibodies as an aid in the study of skin cancer formation. J. Natl. Cancer Inst., 34:191. .6 Burton, R.M., Hope, N.O., Beyerle, M.P. and Espinosa, E. (1977) Tissue antigens in ovarian carcinoma. Oncology 34:146. 7. Burton, R.M., Hope, N.O., Beyerle, M.P. and Espinosa, E. (1978) Gewegsanti gene bein ovarialkarzinom. Onkologle 1:75. .8 Kay, H.E.M. (1957) A and B antigens in normal and malignant cells Br. J. Cancer, 11:409. 9. Davidson, I., and Ni, L.Y. (1969) Loss of isoantigens A, B and H in carcinoma of the lung. Am. J. Pathol., 57:307. .10 Seigler, H.F., Kremer, W.B., Metzgar, R.S., Ward, F.E., Haung, A.T., and Amos, D.B. (1971) HL-A antigenic loss in malignant trans formation. J. Natl. Cancer Inst., 46:577. .11 Cryan, W.S., Hide, R.M., and Garb, S. (1966) Demonstration by gel diffusion of antiqen in spontaneous mouse tumors. Cancer Res., 26:145, . .12 Heppner, G.H., and Pierce, G. (1969) In vitro demonstration of tumor-specific antigens in spontaneous mammary tumors in mice. Internat. J. Cancer, 4^:212. 13. Isojima, S., Yagi, Y., and Pressman, D. (1969) Antigens common to rat hepatoma induced with 2-acetyl aminofluorene. Cancer Res., 29:140. 14. Kahan, B.D., Holmes, E.C., Reisfeld, R.A., and Morton, D.L. (1969) Water soluble guinea pig transplantation antigen from carcinogeninduced sarcomas. J. Immunol., 102:28. CMA 003554 ?1 2, i.-wcsure. * ** a r> 3 ^ . 5'ter cessation *0. 3. Wyatt, R.H., Kptchen, J.M., Hochstrasser, O.L., Buchanan, J.W., Jr., Campbell, O.R., Slaughter, J.C., and Doll, A.H. (1975) An epidemiologic study of blood screening tests and illness histories among chemical workers involved in the manufacture of polyvinyl chloride. Ann. N.Y. Acad. Sci., 246:80. 4. Nairn, R.C., Ricfimond, H.G., McEntegart, M.G., and Fothergill, J.E. (1960) Immunological difference between normal and malignant cells. Br. Med. J., 2:1355. -mler, A. (1365) Immunochemical staining with fluorescein-iabelea antioodies as an aid in the study of skin cancer formation. J. Nat 1. Cancer Inst.. 34:191. 6. 8urton, R.M., Hope, N.J., Beyerle, M.P. and Espinosa, E. (1977) Tissue antigens in ovarian carcinoma. Oncology 34:146. 7. Burton, R.M., Hope, N.O., Beyerle, M.P. and Espinosa, E. (1978) Gewegsantigene bein ovarialkarzinom. Onkoloqie 1:75. 8. Kay, H.E.M. (1957) A and B antigens in normal and malignant cells, 8r. J. Cancer, 11:409. 9. Davidson, I., and Ni, L.Y. (1969) Loss of isoantigens A, B and H in carcinoma of the lung. Am. J. Pathol., 57:307. 10. Seigler, H.F., Kremer, W.B., Metzgar, R.5., Ward, F.E., Haung, A.T., and Amos, D.B. (1971) HL-A antigenic loss in malignant trans formation. J. Natl. Cancer Inst., 46:577. 11. Cryan, W.S., Hide, R.M., and Garb, S. (1966) Demonstration by gel diffusion of antigen in spontaneous mouse tumors. Cancer Res., 26:145, . ---------------- 12* Heppner, G.H., and Pierce, G. (1969) In vitro demonstration of tumor-specific antigens in spontaneous mammary tumors in mice. Internat. J. Cancer, ^:212. 13. Isojlma, S., Yagi, Y., and Pressman, 0. (1969) Antigens conrnon to rat hepatoma induced with 2-acetylaminofluorene. Cancer Res., 29:140. 14. Kahan, B.D., Holmes, E.C., Reisfeld, R.A., and Morton, D.L. (1969) Water soluble guinea pig transplantation antigen from carcinogeninduced sarcomas. J. Immunol., 102:28. CMA 003555 85 15. Hughes, L.E., and Litton, B. (1964) Antigenic properties of human tumors: Delayed cutaneous hypersensitivity reactions. Br. Med. J. J_:209. ------------ 16. Itakura, K. (1963) Studies on human cancer antigens by gel diffusion methods. Gann, 54:93. 17. McKenna, J.M., Sanderson, R.P., and Blakemore, W.S. (1962) Extrac tion of distinct antigens from neoplastic tissue. Science, ,135:370. ------------ 18. Burton, R.M., McGrew, T.L., Barrow, G.H., Beyerle, M.P., Fort- wengler, P.H., Day, T.G., Kuhns, S.H., and Espinosa, E. (1979) Occurence of a thermostable antigen of ovarian carcinoma in normal tissues and secretions. Cancer, 43:2385. 19. Espinosa, E. (1973) Circulating tissue antigens. II. Studies on an organ-sepcific antigen of human liver. Lab. Invest., 29:556. 20. Espinsoa, E. (1977) F-antigen in liver injury. Gastroenterology, 72:985. ------------------------ 21. Espinosa, E. and Caple S. (1981) Extrahepatic immunofluorescent reactivity of antisera detecting liver F antigen. Fed. Proc., 40:764. ----- ------- 22. Espinsoa, E. (1974) Circulating tissue antigens. I. Tissue antigens in serum of patients with diseases involving injury of the liver and of other organs. Clin. Exp. Immunol., 16:153. 23. Espinosa, E. (1976) On two tissue antigens detected in pathologic sera. Lab. Invest., 34:8. 24. Espinosa, E. (1976) Circulating tissue antigens. III. Identifica tion and characterization of antigens of limited and of wide body distribution in human gallbladder bile. Presence of serum of patients with acute hepatitis. Clin. Exp. Immunol., 25:410. 25. Espinosa, E. (1976) Inmunopathologic observations in liver angio sarcoma. jn Nieburgs, H.E. (ed.). Third international symposium on detection and prevention of cancer proceedings, New York, Marcel Oekker, Inc. 26. Fortwengler, H.P., Dever, M.E., Tamburro, C.H. and Espinosa, E. (1978) Lymphocyte transformation tests in vinyl chloride (VC) workers. Fed. Proc., 37:362. 27. Fortwengler, H.P., Jr., Jones, D., Espinosa, E. and Tamburro, C.H. (1981) Evidence for endothelial cell origin of vinyl chloride induced hepatic angiosarcoma. Gastroenterology, 80:1415. 003556 CMA 86 28. Espinosa, E., Caple, S., Kuchella, C. and Chia, S. (1979) Two liver antigens undetectable in a fast growing line of transplanted hepatoma (Morris Hepatoma 7777). Fed. Proc., 38:1069. 29. Espinosa, E., Chia, S., Caple, S. and Kupchella, C. (1979) Liverspecific antigen in transplantable hepatomas having different growth rates. Fed. Proc., 38, 1069. 30. Johnston, P.B., Espinosa, E., Chia, S. and Caple S. (1979) Proper ties of 14 week matintenance cultures of PLC/PRF/5 cells. In vitro, 15:227. 31. Espinosa, E., Johnston, P.B., El-Naggar, M.M. and Caple, S. (1980) Serum and liver antigens in cell sheets and culture fluid of PLC/PRF/5 human hepatoma cells. Lab. Invest., 42:18. 32. Espinosa, E., Johnston, P.B., and Caple, S. (1980) Immunofluorescent localization of plasma proteins in cultured human hepatoma cells. Fed. Proc., 39:777. CMA 003557 PROGRAM H THE USE OF ISOLATED MAMMALIAN LIVER CELLS FOR THE STUDY OF CHEMICAL MONOMER METABOLISM; Investigator, R. C. Feldhoff HI. Isolation of Mammalian Liver Cells for the Study of the Metabolism of Chemical Monomers H2. The In Vitro Study of Albumin Synthesis by Liver from Human Biopsies Background Presently there is no simple or even complex method to assess the functional capability of the human liver to handle exposure to various chemicals and/or their metabolites. In general, the rat has proven to be the most useful experimental animal for studying hepatic metabolism. Perfused liver slices and Isolated hepatocyte systems have been developed to better study the interrelationship and to control the myriad processes of liver metabolism. In the perfused liver the architecture of the organ is preserved but separated from other body influences. Newer in situ techniques allow the liver tissue to never be without a supply of oxygenated"red cells. The same technique can also be used to prepare isolated liver cells enzymatically. The intact liver consists of at least four cell types, whereas isolated liver cells are nearly homogeneous in population. These cell suspensions provide technical advantages in that a number of conditions can be tested at one time and many uniform serial samples can be collected. Preliminary investigations were therefore done to determine whether the human liver biopsy material can function in a manner equivalent to liver slices and whether these tissues would demonstrate similar findings to those of isolated cell suspensions. If this technique were made functionally feasible It could then be used to study the metabolism of xenobiotlcs using viable human liver tissue in an jn^ vitro manner. Objective 1. To determine if an in vitro adaptation could be developed which would utilize animal/human biopsy material for assessing the functional capability of human liver cells. Protein metabolism has been used as the standard indicator and criteria for comparison 87 CMA 003558 88 2. To determine the usefulness of this technique in human bio material obtained for routine medical purposes via percutaneous transvenous approaches. Research Results Our first step was to verify the clinical and biochemical conditions needed for handling human liver biopsy material obtained during routine medi cal procedures. Appropriate assay conditions are being determined to assess the functional capability of the liver at the time of the biopsy to synthesize retained and secreted protein. At the same time, the biopsy's histopathology will be assessed and correlated with the biochemical data. The technique used to study both protein synthesis and secretion is shown in Figure 1. The his topathology is illustrated in Figure 2. IN VITRO TECHNIQUE FOR STUDY OF HUMAN LIVER TISSUE A - RETAINED PROTEIN B - SECRETORY PROTEIN C LIGHT MICROSCOPY LIVER BIOPSY FIGURE 1 M 0*rw* DIVIDED LIVER BIOPSY otf*MQft* 0*tm*f+P*m*lttArivM0*0**** r* mn*>*f***?** JvT0c jmtkvsii 9f em far**. FIGURE 2 rU-j7 Attmo*p**a*1** ****** CMA 003559 i.2 ^ _ -__ .. .i _ ^ tl. . ^ m _ Research Results Our first step was to verify the clinical and biochemical conditions needed for handling human liver biopsy material obtained during routine medi cal procedures. Approoriate assay conditions are being determined to assess the functional capability of the liver at the time of the biopsy to synthesize retained and secreted protein. At the same time, the biopsy's histopathology will be assessed and correlated with the biochemical data. The technique used to study both protein synthesis and secretion is shown in Figure 1. The his topathology is illustrated in Figure 2. IS VTTsn TfoimgiT FOR STu'OV OF HONAN LIVER TISSUE CUT A - RETAINED PROTEIN 3 - SECRETORY PROTEIN C LIGHT MICROSCOPY LIVER BIOPSY FIGURE 1 * ************* DIVIDED LIVER BIOPSY 0m0e7m*a* f****** r* me emrwn*. mt0>0m0$+*f) n><* CtuMM ifi.'vfft;Ji jm<i FIGURE 2 CMA 003560 89 The initial experimental studies utilizing liver perfusion and the isola tion of hepatocytes techniques provided the hepatocytic and nonparenchymal cells for Du and Tamburro's study of the oxidizing and detoxifying capabili ties of these cells. Subsequent experiments have been used to quantitate the intracellular albumin* synthesis, the rate cf a&fimjn secretion by the l&e of specific antibody frcii ons av a 11 ab lefor human >and rati albumin, initial ^_se&ttlU are^Uustrated itu.the rocket iikwimoe^trop^^ frtmmg liver"biopsy study (FigureS) and the aatinoacTd analysis is seen in Figure & The technique for liver cell identification after cell separation by the perfusion technique is illustrated in Figure 5. % prV*( ....:.... . .;.....;qj f | r.p ftf*# ",v /J- ,, ; *>. **, life \ .. .V'V'irS^ a *.'?> &S- -; ' -s*-4^-f 5 *"P n ep. 3"^"iai n' * l 'l *! ** #"*. , ,, w^- _ 9o 9o--o9 S90O 9000 00 096966 6'#o"o 66 0 9 ' `j'jS*. FI d"tCr ,31* CMA 003561 90 FIGURE 4 MARKER CELL TYPE dna/rna ALL CELLS KC EC FIGURE 5 FB Relevance to Industry Although this study was a last year addition to our initial proposal, it has provided the preliminary exploration into the adaptation of highly sophisticated laboratory techniques to human tissue in an in vitro system. While the work of these studies is still developmental, it prov\des promise for future clinical use. Such use could include determination of an indi vidual liver's ability to oxidize and detoxify xenobiotics. The degree and nature of liver cellular injury or impairment could be determined in a quantitative biochemical fashion, and could be directly correlated with CMA 003562 > ''iM - ,u 90 1 : : I i I, FIGURE 4 - j |l = 1'11 j'j 1"1 TECHNIQUE FOR HEPATIC CELL TYPE IDENTIFICATION MARKER CELL TYPE CMA 003563 91 morphological findings. These techniques could be also used to determine the degree of biochemical adaptations of the human tissue in individuals who have been exposed to xenobiotics. Finally, these methods could, on sequential biopsy, provide clinical information regarding whether these alterations persist or revert back tp.a normal state* ^ 6, ^ -v . , "'is", .. t .. . vr Although: this a. very- _futm-tstio attem&t to- edap* presently, developf^ methodology, it does accurately reflect the direction in Which* cHnicaT research should be directed in order to better assess the human capability of * sustaining environmental changes. References 1. Ou, J. and Tamburro, C.H. (1980) Oxidative and detoxifying ability of liver mesenchymal vs. parenchymal cells in the metabolism of xenobiotics. Gastroenterology 79, 1013. 2. Exton, J.H. (1975) The perfusion rat liver.; Methods in Enzymoloqy 37, 25-37. ------------------------ 3. Feldhoff, R.C., Taylor, J.M., and Jefferson, L.S. (1977) Synthesis and secretion of albumin in vivo, in perfused liver and Isolated hepatocytes: Effects of" fiypophysectomy and growth hormone treatment. J. Biol. Chem. 252, 3611-16. 4. Feldhoff, R.C., Taylor, J.M., and Jefferson, L.S. (1977) Albumin synthesis by isolated rat hepatocytes. Effects of amino acids, bovine albumin, rat serum and hormones. Fed. Eur. Bioch. Soc. (Abstract). 5. Ledden, O.J., Feldhoff, R.C. and Gray, R.O. (1980) Purification and characterization of human albumin fragments. Fed. Proc. 39, 1676 Leevy, C .^11683}- Lnyitro studles-of heg#l^oaA* synthesis .t%l; perciatmeopljfejf " biopsy , swcins frfpt : 3; tahj-folfc- Metfm 76^*700^* v-'v.,!' - - ^ *"*?'*.-`-4^; ,, ... ... - ,, Wl > '-r- ?'* J'vi> 7. Peters, tei^r aSTSi ^Serum; albumjh. ln*The f lasmjrFrotetns, 2r' 1 ?ed., Putnam* F.t*. ed., ifol. 'F^ pp."T33*181, Academic Press, MY. .* "**1-'' -a-" . ^ - 8. ,, T.,f jw. (1986) Albumin: An overview and bibliography. Miles Laboratories, Inc., Elkhart, IN. 9. WheTan, J.F., Jr., Creech,. J*, Jr. and Tamburro, C.H. (1981) The adequacy and safety of transvenous hepatfc biopsy in complicatedhepatic disease. Clinical Research 29 313A. CMA 003564 PROGRAM I THE STUDY OF TISSUE DISPOSITION OF INDUSTRIAL CHEMICALS: THE VINYL CHLORIDE EXAMPLE; Investigators - W.J. Waddell and C. Marlowe II. The Use of Whole Body Autoradiography in Specific Tissues. Localization of Accumulated and Retained Industrial Chemicals and Their Metabolites Background Vinyl chloride has been classified by regulatory agencies as a carcinogen. Chronic exposure of workers to this carcinogen has resulted in angiosarcoma of the liver, as well as other tumors (Creech and Johnson, 1974). Several types of tumor also are observed after vinyl chloride exposure to animals (Maltoni and Lefemine, 1974). How or why these various body sites developed tumors was unclear. The need for methods which more accurately identify the sites of localization of chemical agents and their metabolites has obvious importance. The use of whole body autoradiography to identify the ultimate location of radioactively tagged chemicals was studied utilizing 14C-vinyl chloride. Objective 1. To determine the sites of localization and retention of the metabolites of vinyl chloride in mice. 2. Identification of the tissues of localization should help elucidate the mechanism of carcinogenic action of vinyl chloride. Research Results The sites of localization of radioactivity in- tissues of the mouse following exposure to ^C-vinyl chloride were studied by the technique of whole-body autoradiography (Waddell and Marlowe, 1977). Male CD-I mice purchased from Charles River, weighing 25-31g., were exposed to 'x-vinyl chloride in room air in a sealed chamber for 3 hours. The radioactive vinyl chloride was synthesized from '^C-ethylene dichloride (New England Nuclear, Lot <*1194-143, specific activity 3.2 mCi/mmole) immediately prior to exposure by the method of Wagner et al. (1975). The initial concentration of the vinyl chloride was 50 ppm with a specific activity of 1.84 mCi/mmole. At the end of the 3 hour exposure of the mice to '^C-vinyl chloride-the mice were removed to room air. Twenty minutes, 1, 3, 9 or 24 hours after removal of the mice to 92 CMA 003565 93 room air, a mouse was briefly anesthesized with ether and sacrificed freezing in a dry ice/hexane bath at -75*C. The preparation of the radioactive vinyl chloride, exposure of the mice to this isotopic gas, and the sacrifice of the mice were done in collaboration with Dr. William T. Stott at Dow Chemical in Midland, Michigan. All remaining procedures were performed at the University of Louisville. Whole-body sagittal sections of the mice, 20v- and 40y- thick were taken onto Scotch tape at -20*C. After freeze-drying, these sections were placed against Kodak AA X-ray film and allowed to expose in light-tight containers at -14*C for 6 to 39 weeks. The sites of localization of all nonvolatile metabo lites of vinyl chloride were visualized in the developed X-ray film (auto radiograph) and reveals the j_n vivo disposition of these metabolites at the time of sacrifice of the moused These procedures for whole-body autoradi ography, first described by Ullberg in 1954 do not allow thawing or contact with any solvents. Thus, there is no translocation nor loss of radio activity. A complete description of these procedures have been published previously (Waddell and Marlowe, 1977). Representative autoradiographs from each mouse were used as negatives to produce the prints included with this report. Therefore, white areas in the prints represent the sites of radioactive accumulation of the nonvolatile metabolites of 14C-vinyl chloride. The highest levels of radioactivity in the mice sacrificed 20 minutes 1 hour after removal from the ^C-vinyl chloride environment were obser^J in liver, pancreas, kidney, intestinal contents, urine and bile (Figures 1 and 2). Additional sites of localization of radioactivity in the 1-hour animal were thymus, thyroid and seromucous glands (Figure 2). The concentration of metabolites in the organs of excretion decreased continually over the 24 hour period (Figures 3-5). By 9 hours after removal from the vinyl chloride, thymus, Harder's gland, epithelium of the esophagus and intestine, urine and sublingual gland retain the highest levels of radioactivity; moderate concentrations of radioactivity were observed in liver, kidney and intestinal contents (Figure 4). After 24 hours the primary organs of retentions of radioactivity were thymus and Harder's gland; some radioactivity could be visualized in liver and in esophageal and intestinal epithelium (Figure 5). Figure 6 shows the relative concentration of nonvolatile metabolites in the cortex and medulla of the thymus at the earliest and latest time intervals studied. Twenty minutes after removal from the vinyl chloride, the concentration of radioactivity in the thymus was only slightly increased above that of blood (Figure 1 and 6); however, by 1 hour the thymus showed the highest uptake of radioactivity in the body (Figure 2) and remained the highest throughout the 24 hour period (Figures 2-6). CMA 003566 93 v* ' -\ :"';fly ares:'53:zed with ether 1:0 3 ssr f ces .. .e oath at -75'c. 7 - or :,ii radioac:ive vinyl chloride, exposure of the mice to this isc::oic gas, and the sacrif te of the mice were done in collaboration with Or. 'william T. Stott at Dow Che 1 ical in Midland, Michigan, All remaining procedures were performed at the University of Louisville. Wnoie-oody sagittal |ections of the mice, 20u- and 40u- thick were taken onto Scotch tape at -20*C. After freeze-drying, these sections were placed against Kodak AA X-ray film and allowed to expose in light-tight containers at -14*C for 6 to 39 weeks. The sites of localization of all nonvolatile metabo lites of vinyl chloride were visualized in the developed X-ray film (auto radiograph) and reveals the jji vivo disposition of these metabolites at the of !ce of tbs moused These procedures for whole-body autoradi. Wy1 , nrst described by Ullberg in 1954 do not allow thawing or contact with any solvents. Thus, there is no translocation nor loss of radio activity. A complete description of these procedures have been published previously (Waddell and Marlowe, 1977). Representative autoradiographs from each mouse were used as negatives to produce the prints included with this report. Therefore, white areas in the prints represent the sites of radioactive accumulation of the nonvolatile metabolites of '4C-vinyl chloride. The highest levels of radioactivity in the mice sacrificed 20 minutes and 1 hour after removal from the ^C-vinyl chloride environment were observed in liver, pancreas, kidney, intestinal contents, urine and bile (Figures 1 and 2). Additional sites of localization of radioactivity in the 1-hour animal were thymus, thyroid and seromucous glands (Figure 2). The concentration of metabolites in the organs of excretion decreased continually over the 24 hour period (Figures 3-5). By 9 hours after removal from the vinyl chloride, thymus, Harder's gland, epithelium of the esophagus and intestine, urine and sublingual gland retain the highest levels of radioactivity; moderate concentrations of radioactivity were observed in liver, kidney and intestinal contents (Figure 4). After 24 hours the primary organs of retentions of radioactivity were thymus and Harder's gland; some radioactivity could be visualized in liver and in esophageal and intestinal epithelium (Figure 5). Figure 6 shows the relative concentration of nonvolatile metabolites in the cortex and medulla of the thymus at the earliest and latest time intervals studied. Twenty minutes after removal from the vinyl chloride, the concentration of radioactivity in the thymus was only slightly increased above that of blood (Figure 1 and 6); however, by 1 hour the thymus showed the highest uptake of radioactivity in the body (Figure 2) and remained the highest throughout the 24 hour period (Figures 2-6). CMA 003567 ,4C-VINYL CHLORIDE; 20 MIN AFTER REMOVAL 94 BRAIN THYMUS LUNG LIVER ADRENAL KIDNEY PANCREAS NASAL EPITHELIUM BLOOD CONTENTS OF STOMACH AND INTESTINE Figure 1: A print of a whole-body autoradiograph from a male CD-I mouse which was exposed for 3 hours to ^C-vinyl chloride and then frozen 20 minutes after removal from the vinyl chloride environment. White areas correspond to radioactivity. ,4C-V1NYL CHLORIDE; I HR AFTER REMOVAL SEROMUCOUS GLAND THYROID BLOOD LUNG LIVER PANCREAS URINE MINOR AND MAJOR SALIVARY GLANDS THYMUS BILE CONTENTS OF INTESTINE Figure 2: A print of a whole-body autoradiograph from a male CD-- 1 mouse which was exposed for 3 hours to ^C-vinyl chloride and then frozen 1 hour after removal from the vinyl chloride environment White areas correspond to radioactivity. CMA. 003568 I * f i4C-VINYL CHLORIDE; 3 HR AFTER REMOVAL 1 HARDER'S GLAND PARATHYROID BLOOD PANCREAS KIDNEY URINE 95 SEROMUCOUS GLAND THYMUS LIVER CONTENTS OF STOMACH AND INTESTINE Figure 3: A print of a whole-body autoradiograph from a male CD-I mouse which was exposed for 3 hours to ^C-vinyl chloride and then frozen 3 hours after removal from the vinyl chloride environment. White areas correspond to radioactivity. ,4C-VINYL CHLORIDE; 9 HR AFTER REMOVAL HARDER'S GLAND SUBLINGUAL GLAND LIVER SPLEEN KIDNEY PANCREAS LINGUAL MUCOSA THYMUS BLOOD GASTRIC MUCOSA WALL OF INTESTINE URINE Figure 4: A print of a whole-budy nuiorudi o.,raph f: win j :n<iJe CD-L mouse wirlcii vjs exposed lor 3 hours* to i4-vinyl chloride and then iruzou v .tours after removal from the vinyl chloride env irontpent. White areas correspond to rad tonet ivLty - CMA 003569 C-ViN'YL if'\r* ! . ----------- .' O~ Vi-Ji1.W* 1mIWr i'5uQr*iTt.u..v..O..--', '--r' -a--> V'' i'"* -: v r- ,Y -- Y1 :-! ? ' -" SEROMUCOUS GLAND THYMUS LIVER CONTENTS OF STOMACH AND INTESTINE Figure 3: A print of a whale-body autoradiograph from a male CD-I mouse which was exposed for j hours to '`C-vinyl chloride and then frozen 3 hours after removal from the vinyl chloride environment. White areas correspond to radioactivity. ,4C-VINYL CHLORIDE; 9 HR AFTER REMOVAL HARDER'S GLAND SUBLINGUAL GLAND LIVER SPLEEN KIDNEY PANCREAS LINGUAL MUCOSA THYMUS BLOOD GASTRIC MUCOSA WALL OF INTESTINE URINE Figure 4: A print of a whole-body autoradiograph fr^m j uiaJe CD-I mouse wiricu was exposed for 3 hours to ^C-vinyl chloride and then frozen 9 hours after removal from the vinyl chloride environment. White area correspond to radioactivity. CMA 003570 C-VINYL CHLORIDE; 24 HR AFTER REMOVAL HARDER S GLAND BLOOD KIDNEY PANCREAS VIBRtSSAL FOLLICLES THYMUS LIVER Figure 5: k print of a whole-body autoradiograph from a male CD-I mouse which was exposed for 3 hours to '"C-vinvl chloride and then frozen 24 hours after removal from the vinyl chloride environment. White areas correspond to radioactivity. I4C-VINYL CHLORIDE Figure 6: Prints of the thymus areas of whole-body autoradiographs from male CD-I mice which were exposed to ^C-vinyl chLorido Cor 3 hours and then frozen 20 minutes or 2^ hours after removal from the vLnyi chloride ei vtronruenc. White areas correspond to radioactivity. Note the high retention of the nonvolatile metabolites in the cortex of the thymus after 2$ hours. CMA 003571 97 The sites of localization of the nonvolatile metabolites of vinyl chlor in the mouse are similar to those reported for the rat (Ouprat et al. 1977). The high concentration of nonvolatile metabolites of vinyl chloride retained in the thymus after 24 hours suggests that there is covalent binding of these metabolites to molecules in the thymus. It is possible that these molecular interactions in the thymus could have an effect on the immune system. The high concentration in the thymus may be involved in the stimulation of the immune system in mice seen by Sharma and Gehring (1979) and in the reduction in peripheral T-lymphocytes seen in man by Ward et al. (1976). Vinyl chloride may exert its carcinogenic action by a dual mechanism. It may suppress the immune surveillance system mediated by the thymus and concurrently damage several tissues including the liver. Relevance to Industry Whole-body autoradiographic studies on the biological disposition of industrial chemicals offers the most thorough approach to the study of the interaction of these chemicals with tissues in the body. Knowledge of these tissue interactions allows predictions of potential toxicity or, conversely, to a lack of toxic effect if the chemical is rapidly and completely eliminated (Waddell et al., 1977). When chemicals are retained in specific tissues as in the case with vinyl chloride, the tissue of retention gives an indication of possible mechanisms of toxic action. Vinyl chloride appears to be having a toxic effect on the immune system; the direct evidence for an interaction of the chemical with this system was not available until these studies were dor^p References Creech,J.L. and Johnson, M.N. (1974) Angiosarcoma of liver in the manufacture of polyvinyl chloride. Journal of Occupational Medicine 16: 150. ------------------------- ---------------------------- Duprat, P., Fabry, J.P., Gradiski, 0. and Magadur, J.L. (1977) Metabolic approach to industrial poisoning: blood kinetics and distribution of '^C-vinyl chloride monomer (V.C.M.) Acta Pharmacology and Toxicoloqy 41:142. Maltoni, S. and Lefemlne, G. (1974) Carcinogenicity bioassays of vinyl chloride. Research plan and early results. Environmental Research 7: 387. Sharma, R.P. and Gehring, P.J. (1979) Immunologic effects of vinyl chlo ride in mice. Annals New York Academy of Science. 320:551. Ullberg, S. (1954) Studies on the distribution and fate of S^-labelled benzylpenicillin in the body. Acta Radiology, Supplment 118:1. CMA 003572 .WU 5 t c ^ i-,: , r 2C V - , r '*'*'>!* V \ S* pr ' * l C,2" * i -* ;e 11 an concentration v*'*/1. "r V:f iC2 ''i z~e cnyr^us after a4 riCurs suggests tnat tnsrg i s ccva lent pinoing cf tnese Tstabolites to molecules in the thymus. It is possible that these molecular interactions in the thymus cculd have an effect on the immune system. The high concentration in the thymus may be involved in the stimulation of the i-mtjre svstem in mice seen bv Sharma and Gehrim (19791 and i^ the v*ed,jct';'' in peripheral T-lymphocytes seen in man by Ward et al. (1976). Vinyl chloride may exert its carcinogenic action by a dual mechanism. It may suppress the immune surveillance system mediated by the thymus and concurrently damage several tissues including the liver. '.I'hole-fcody autoradiographic studies on the biological disposition of industrial chemicals offers the most thorough approach to the study of the interaction of these chemicals with tissues in the body. Knowledge of these tissue interactions allows predictions of potential toxicity or, conversely, to a lack of toxic effect if the chemical is rapidly and completely eliminated (Waddell et al., 1977). When chemicals are retained in specific tissues as in the case with vinyl chloride, the tissue of retention gives an indication of possible mechanisms of toxic action. Vinyl chloride appears to be having a toxic effect on the immune system; the direct evidence for an interaction of the chemical with this system was not available until these studies were done. References Creech,J.L. and Johnson, M.N. (1974) Angiosarcoma of liver in the manufacture of polyvinyl chloride. Journal of Occupational Medicine _T6; 150. ------------------------- ---------------------------- Ouprat, P., Fabry, J.P., Gradlski, 0. and Magadur, J.L. (1977) Metabolic approach to industrial poisoning: blood kinetics and distribution of `^C-vinyl chloride monomer (V.C.M.) Acta Pharmacoloqy and Toxicology 41:142. -------------------------- Maltoni, S. and Lefemine, G. (1974) Carcinogenicity bioassays of vinyl chloride. Research plan and early results. Environmental Research 7: 387. Sharma, R.P. and Gehring, P.J. (1979) Immunologic effects of vinyl chlo ride in mice. Annals New York Academy of Science. 320:551. Ullberg, S. (1954) Studies on the distribution and fate of S-^-labelled benzylpenicillin in the body. Acta Radiology, Supplment 118:1. CMA 003573 98 Waddell, W.J. and Marlowe, C. (1977) Autoradiography. IN: Garrett, E.R. and Hirtz, J.L. (eds.), Drug Fate and Metabolism: Methods and Techniques. Vol. 1, pp. 1-25. New York: Marcel Dekker. Waddell, W.J., Marlowe, C., Miripol, J.E., and Gravin, P.J. (1977) The distribution in mice of intravenously administered plasma solutions of [l^C]Di-2-Ethylhexyl Phthalate Determined by Whole-Body Autoradiography. Toxicology and Applied Pharmacology 39:339-353. Wagner, E.R., Muelder, W.M., Watanabek, P.G., Hefner, R.E., Jr., Braun, W.H. and Gehring. P.J. (1975) A gas chromatographic method for the preparation of i^c-labelled vinyl chloride. Journal of Labelled Compounds XI (4):535. Ward, A.M., Udnoon.S., Watkins, J., Walker, A.E. and Darke, C.S. (1976) Immunological mechanisms in the pathogenesis of vinyl chloride disease. British Medical Journal, 1:936. CMA 003574 LISTS OF PUBLICATIONS, ABSTRACTS, PREPRINTS AND PUBLICATIONS IN PREPARATION 99 CMA 003575 100 PUBLICATIONS 1. Fortwengler, Jr., H.P., Jones, D., Espinsoa, E. and Tamburro, C.H. (1981) Evidence of Endothelial Cell Origin of Vinyl Chloride-Induced Hepatic Angiosarcoma. Gastroenterology, 80:1415-1419. 2. Ou, J.T., Sandoz, J.P., Tseng, M.T. and Tamburro, C.H. (1979) Biochemical Alterations in Livers of Rats Exposed to Vinyl Chloride. Journal of Toxicology and Environmental Health, J5:1119-1132. 3. Tamburro, C.J. (1978) The Hepatic Role in Carcinogenesis and Its Early Detection--The Vinyl Chloride Model. The Yale Journal of Biology and Medicine, Inc., 51:67-80. 4. Du, J.T., Tseng, M.T. and Tamburro, C.H, (1982) The Effect of Repeated Vinyl Chloride Exposure on Rat Hepatic Metabolizing Enzymes. Toxicology and Applied Pharmacology, 62:1-10. 5. Tamburro, C.H, and Greenberg, R.A. (1981) Effectiveness of Federally Required Medical Laboratory Screening in the Detection of Chemical Liver Injury. Environmental Health Perspectives, 41:117-122. 6. Curran, K.L., Kupchella, C.E. and Tamburro, C.H. (1977) Urinary Glyco- saminoglycan Patterns in Angiosarcoma of the Liver. Cancer, 40: 3050-3053. ---------- " 7. Kupchella, C.E. and Tamburro, C.H. (1978) Urinary and Tissue Glycosamino- glycan Patterns in Hepatic Angiosarcoma. Detection and Prevention of Cancer, H.E. Neiburgs, Ed., Part 1, Vol i, Marcel uekxer, inc.. New York. 8. Kupchella, C.E., Drake, E.E., Kennedy, J., Curran, K.L., Warick, R, and Morris, H.P. (1981) Tissue and Urinary Glycosaminoglycan Patterns Associated with a Fast, an Intermediate, and a Slow-growing Morris Hepatoma. Cancer Research, 41:419-424. 9. Tamburro, C.H., Kupchella, C.E. and Greenberg, R.A., et al. (1981) Screening for the Early Detection of Disease in Individuals Exposed .to Vinyl Chloride. U.S. Environmental Protection Agency, 560/6-81- 002 10. Laumbach, A.D., Lee, S., Wong, J. and Streips, U.N. (1976) Studies on the Mutagenicity of Vinyl Chloride Metabolites and Related Chemidals. Prev., Detect. Cancer (Proc. Int. Symp.) 3rd, Vol. 1:155-170. 11. Elmore, J., Wong, H., Laumbach, A.O. and Streips, U.N. (1976) Vinyl Chloride Mutagenicity and Carcinogenicity via the Metabolites Chlorooxirane and Chloroacetaldehyde Monomer Hydrate. 8iochem. Biophys. Acta 442:405. CMA 003576 101 12. Streips, U.N., Laumbach, A.D. and Yasbin, R.E. (1981) Bacterial Mutation Monitor for Active Metabolites of Chemical Carcinogens: B. Subtil is Assays for Mutation and INA Repair, _In Microbial Testers^ T Cecil Felkner, Ed., Vo 1. 5, Marcel Dekker, Inc., New York. 13. Horowitz, S., Doyle, R.J., Young, F.E. and Streips, U.N. (1981) Selective Association of the Chromosome with Membrane in a Stable L-Form of Bacillus Subtil is. Journal of Bacteriology, 138:915-922. 14. Streips, U.N., Horowitz, S. and Doyle, R.J. (1980) Genetic Analysis of DNA-Surface Interactions in Bacillus Subtilis. Microbiology:284-287. 15. Horowitz, S., Doyle, R.J. and Streips, U.N. (1978) Restricted ChromosomeMembrane Association in a Stable L-form of Bacillus Subtilis. In Transformation. 1978. Proceedings of the Fourth International Meeting on Bacterial Transformation. 16. Sonnenfeld, G., Barnes, M.C., Schooler, J. and Streips, U.N. (1980) Inhibition of Interferon Induction as a Screen for the Carcinogenic Potential of Chemicals, jji Interferon: Properties and Clinical Uses. A. Khan, N.O. Hill and G.L. Dorn, Eds. pp. 589-598, Wadley Institutes of Molecular Medicine, Dallas, Texas. 17. Barnes, M.C., Streips, U.N. and Sonnenfeld, G. (1981) Effect of Car cinogens and Analogs on Interferon Induction. Oncology 38:98-101 18. Espinosa, E. (1976) Immunopatho logic observations in liver angiosarcoma. In Prevention and Detection of Cancer, Vol. 1, Neiburgs, H.E., Ed. pp. 927-36, Marcel Dekker, Inc., New York. 19. Espinosa, E. Circulating Tissue Antigens. (1976) Identification and Characterization of Antigens of Limited and of Wide Body Distribution in Human Gall Bladder Bile. Clin. Exp. Immunol., 25:410-417. CMA 003577 .12 Streips, U.N.. Lsumbach. A,n_ v?s:in, r> .E. ',1931) 3 i' Mutation Monitor for Active Metabolites of Chemical Cerci" 5. Subti 1 is Assays for vcfaficn and A Repair, '-c . u j a Testers. I. Cecil Felkner, Ed., Vol. 5, Marcel Dekker, Inc , New York. 13. Horowitz, S., Doyle, R.J., Young, F.E. and Streips, U.N. (1981) Selective Association of tne Chromosome with Membrane in a Stable L-Form of Bacillus Subtilis. Journal of Bacteriology, 138:915-922. 14. Streips, U.N., Horowitz, S. and Doyle, R.J. (1980) Genetic Analysis of DNA-Surface Interactions in Bacillus Subtilis. Microbiology:284-287. 15. Horowitz, S., Doyle, R.J. and Streips, U.N. (1978) Restricted Chrririosp'Tio,. Membrane Association in a Stable L-form of Bacillus Subtilis. In Transformation. 1978. Proceedings of the Fourth Intern ationTT Meeting on Bacterial Transformation. 16. Sonnenfeld, G., Barnes, M.C., Schooler, 0. and Streips, U.N. (1980) Inhibition of Interferon Induction as a Screen for the Carcinogenic Potential of Chemicals, _I_n Interferon: Properties and Clinical Uses. A. Khan, N.Q. Hill and G.L. Dorn, Eds. pp. 589-598, Wadley Institutes of Molecular Medicine, Dallas, Texas. 17. Barnes, M.C., Streips, U.N. and Sonnenfeld, G. (1981) Effect of Car cinogens and Analogs on Interferon Induction. Oncology 38:98-101. 18. Espinosa, E. (1976) Immunopathologic observations in liver angiosarcoma. hi Prevention and Detection of Cancer, Vol. 1, Neiburgs, H.E., Ed. pp. 927-36, Marcel Dekker, Inc., New York. 19. Espinosa, E. Circulating Tissue Antigens. (1976) Identification and Characterization of Antigens of Limited and of Wide Body Distribution in Human Gall Bladder Bile. Clin. Exp. Immunol.. 25:410-417. --------------- ------------------ CMA 003578 102 ABSTRACTS 1. Fortwengler, H.P., Dever, M.E., Tamburro, C.H., and Espinosa, E. (1978). Lymphocyte transformation tests in vinyl chloride workers. Fed. Proc., 37: 362. ---------------- 2. Fortwengler, H.P., Jones, D., Tamburro, C.H. and Espinosa, E. (1979). Factor VIII content as evidence for endothelial origin of vinyl chloride associated liver angiosarcoma. Fed. Proc., 38:999. 3. Fortwengler, H.P. and Tamburro, C.H. (1976). Use of dye clearance in the detection of hepatocellular injury among vinyl chloride workers. Clin. Res., 23:264A. 4. Du, J.T. and Tamburro, C.H. (1976). Decreased glucose-6-phosphatase activity in liver in vinyl chloride exposed rats. Fed. Proc., 35:1422. 5. Du, J.T. and Tamburro, C.H. (1978). Elevated glutathione content, glutathione-S-transferase and glutathione reductase in liver of rats exposed to vinyl chloride. Fed. Proc., 37:1545. 6. Liss, G. and Tamburro, C.H. (1982). Serum bile acids in screening for chemical hepatoxicity. The Toxicologist, July 1982. 7. Kupchella, C.E., Jarvis, J.O., Curran, K.L., Greenberg, R.A. and Tamburro, C.H. (1977). Tissue and urinary glycosaminoglycan changes in hepatic fibrosis. Gastroenterology, 73:1299. 8. Kupchella, C.E. and Tamburro, C.H. (1977). Urinary chrondroitin sulfate fraction patterns in hepatic angiosarcoma. Clin. Res., 25:35A. 9. Kupchella, C.E. and Tamburro, C.H. (1977). Urinary glycosaminoglycan excretion patterns in chemically induced liver injury and cancer. Clin. Res., 25, 329. 10. Kupchella, C.E. and Curran, K.L., Drake, E., Kennedy, J. and Tamburro, C.H. (1978). Tissue and urinary glycosaminoglycans in transplantable hepatomas. Gastroenterology, 75:972. 11. Curran, K.L., Kupchella, C.E., Sandoz, J. and Tamburro, C.H. (1978). Urinary glycosaminoglycan patterns in human hepatic angiosarcoma, hepatoma, and in workers at risk for angiosarcoma. Gastroenterology, 75:959. 12. Kupchella, C.E., Secskas, E.M., Kenndy, J.S., and Espinosa, E. (1979). Glycosaminoglycan changes associated with hepatic tumors: The contributions of regerneration and necrosis. Clin. Res., 27:389. CMA 00351$ 103 13. Greenberg, R.A. and Tamburro, C.H. (1978). Early detection of disease in individuals exposed to vinyl chloride. American Public Health Association Ann. Mtg. 14. Barrows, G.H., Joyce, M.J., Schrodt, G.R., Greenberg, R.A, and Tamburro, C.H. (1979). Computer-assisted morphologic quantitation of collagen in human liver biopsies. Laboratory Investigations, 40:3. 15. Tamburro, C.H., Makk, L. and Popper, H. (1979). Early hepatic histological alterations among chemical (vinyl monomer) workers. Gastroenterology, 77:A33. 16. Laumbach, A.D., Streips, U.N. and Wong, J.L. (1978). Chloroacetaldehyde-induced damage to Bacillus subtilis. Ann. Mtg. Amer. Soc. Microbiol., p. 125. 17. Streips, U.N. and Wong, J.L. (1978). Chioroacetaldehyde-induced damage to Bacillus subtilis. XII Intern. Congress Microbiol., p. 105. 18. Sonnenfeld, G., Barnes, M.C. and Streips, U.N. (1979). Inhibition of interferon induction as a screen for the carcinogenic potential of chemicals. J. Clin. Hematol. and Oncol., 9:291. 19. Johnston, P.B., Espinosa, E., Chia, S. and Caple, S. Properties of 14 week maintenance cultures of PLC/PRF/5 cells. In Vitro, J_5:227. 20. Espinosa, E., Chia, S., Caple, S. and Kupchella, C.E. (197^ Liver-specific F antigen in transplantable hepatomas having different growth rates. Fed. Proc., 38:1069. 21. Espinosa, E., Caple, S., Kupchella, C.E. and Chia, S. (1979). Two liver antigens undetectable in a fast growing line of transplanted hepatoma (Morris Hepatoma 7777). Fed. Proc., 1069. 22. Du, J.T. and Tamburro, C.H. (1980). Oxidative and detoxifying ability of liver mesenchymal vs. parenchymal cells in the metabolism of xenobiotlcs. Gastroenterology, 79:1013. 23. Liss, G. and Tamburro, C.H. (1982). Toxic carcinogenic and safe exposure levels in vinyl chloride-induced hepatic angiosarcoma. Toxicologist, July 1982. CMA 003580 orsancs.'g, !V . * a.3 i: r~ i r p 14 Sarro vs, S.H., Joyce, M.J., Scnrodt, G.R., Greanoerg, R.A. and Tamburro, C.H. (1979). Computer-assisted morphologic quantitation of collagen in human liver biopsies. Laboratory Investigations, 40:3, 15. Tamburro, C.H., MakK, L. and Popper, h. (1979). Early hepatic histological alterations among chemical (vinyl monomer) workers. Gastroenterology, _77_:A33. 16. Laumbach, A.O., Streips, U.N. and Wong, J.L. (1978). Chloroacetal- dehyde-induced damage to Bacillus s u b ti1is. Ann. Mtc. a ^ See tcrou 1 o i j o ' c5 17. Streips, U.N. ana wong, J.L. (197). Chlcroacetalaehyde-induced damage to Bacillus subtilis. XII Intern. Congress Microbiol., p. 105. 18. Sonnenfeld, G., Barnes, M.C. a-d Streips, U.ll. ;T979). Inhibition of interferon induction as a screen for the carcinogenic potential of chemicals. J. Clin. Hematol. and Oncol., 9:291. 19. Johnston, P.B., Espinosa, E., Chia, S. and Caple, S. Properties of 14 week maintenance cultures of PLC/PRF/5 cells. In Vitro, JJ5:227. 20. Espinosa, E., Chia, S., Caple, S. and Kupchella, C.E. (1979), Liver-specific F antigen in transplantable hepatomas having different growth rates. Fed. Proc., 38:1069. 21. Espinosa, E., Caple, S., Kupchella, C.E. and Chia, S. (1979). Two liver antigens undetectable in a fast growing line of transplanted hepatoma (Morris Hepatoma 7777). Fed. Proc., 1069. 22. Du, J.T. and Tamburro, C.H. (1980). Oxidative and detoxifying ability of liver mesenchymal vs. parenchymal cells in the metabolism of xenobiotics. Gastroenterology, 79:1013. 23. Liss, G. and Tamburro, C.H. (1982). Toxic carcinogenic and safe exposure levels in vinyl chloride-induced hepatic angiosarcoma. Toxicologist, July 1982. ------------- ------ 003581CM&- 104 PREPRINTS 1. Tamburro, C.H., Makk, L. and Popper, H. Early Hepatic Histological Alterations among Chemical (Vinyl Monomer) Workers. 2. Du, J.T., Eades, D.S., and Tamburro, C.H. Oxidative and GlutathioneRelated Detoxifying Enzyme Capabilities in Hepatocytes and Nonhepatocytes of Rat Liver. 3. Kupchella, C.E., Greenberg, R.A. , Warick, R.A. and Tamburro, C.H. Preliminary Assessment of the Usefulness of Urinary Total Glycosaminoglycan. CMA 003582 105 PUBLICATIONS IN PREPARATION TITLES 1. Tamburro, C.H., Miller, B. and Greenberg, R.A. Specificity of ICG Clearance TEst of Hepatotoxicity. Sensitivity and 2. Tamburro, C.H., Miller, B. and Chan, C. Indocyanine Green (ICG) Clearance Test Safety and Toxicity. 3. Fortwengler, P. and Tamburro, C.H. Use of HLA Tissue Typing in the Identification of Chemical Injury in Vinyl Monomer-Exposed Workers. 4. Fortwengler, P. and Tamburro, C.H. Immunocompetence of Humans Chemically Exposed to Finyl Monomer Chemical. 5. Barrow, G., Schrodt, G.R. and Tamburro, C.H. Collagen Changes in Normal Aging Liver. CMA 003583 ^ ^ i IU N S T'* 1. Tambu'"rc, I.-,. ''"e'-, i-i Specificity of ICG Clearance TEst of nepatotoxicity. Ce-;- ai -: -._, ;rd 2. Tamburro, C.H., Miller, 8. and Chan, C. Indocyanine Green (ICG) Clearance Test Safety and Toxicity. 3. Fortwengler, P. and Tamburro, C.H. Use of HLA Tissue Typing in the Identification of Chemical Injury in Vinyl Monomer-Exposed Workers. 4. Fortwengler, P. and Tamburro, C.H. Immunocompetence of Humans Chemically Exposed to Finyl Monomer Chemical. 5. Barrow, ., Sen,root, G.T. and Tamo.jrro, C.H. Collagen Chances in 'iormal Aging Liver. CMA 003584 APPENDIX 106 CMA 003585 PUBLICATIONS CMA 003586 GASTROENTEROLOGY 1981;80:1415-19 1 Evidence for Endothelial Cell Origin of Vinyl Chloride-Induced Hepatic Angiosarcoma H. PHILIP FORTWENGLER, JR., DOUGLAS JONES, ENRIQUE ESPINOSA, and CARLO H. TAMBURRO Department of Medicine, Division of Digestive Diseases and Nutrition, Department of Pathology and Cancer Center, University of Louisville School of Medicine, Louisville, Kentucky Previous reports of hepatic angiosarcoma have not clearly defined the cellular type from which this tu mor arises, as evidenced by the terminology of endo thelioma, Kupffer cell sarcoma, endothelial cell sar coma, and hemangioendothelial sarcoma, etc., which have been used interchangeably. In addition, there has been no consensus on the separate entity of Kupffer and sinusoidal endothelial cells. In the work presented here, evidence for the endothelial cell origin of this tumor is provided by the demon stration of factor VIII, a known endothelial cell marker, in the tumor cells. Fluorescence due to the presence of factor VIII appeared intense in the tu mor sinusoidal cells of all four vinyl chloride-associ ated angiosarcomas studied, whereas normal liver sinusoidal lining cells showed negligible fluores cence. Hepatic angiosarcoma has been associated with ex posure to thorotrast (1), arsenic (2), androgenic-ana bolic steroids (3), and vinyl chloride (4). However, most cases are of undetermined origin. Regardless of etiology, earlier studies (5-8) made no clear dis tinction between endothelial and Kupffer cells, and gave various histologic designations to this type of tumor, e.g., hemangioblastoma, Kupffer cell sar- Received April 22. I960. Accepted {anuary 12.1981. Address requests for reprints to: Carlo R Tamburro, M.D.. Di vision of Digestive Diseases and Nutrition. Department of Medi cine. Health Science! and Cancer Center, University of Louisville. Louisville, Kentucky 40292. This investigation was supported by a Research Grant from the Manufacturing Chemiata Association. This work was presented in part at tha Federation of American Societies for Experimental Biology meeting in Dallaa, Texas in April. 1979. The technical assistance of Mr. Larry Wilder ia greatly ac knowledged. We would like to thank Dr. Hans Popper for his re view of the histology and hia suggestions. <0 1981 by the American Gastroenterological Association 0010-5085/81/061415-05*02.50 coma, angiosarcoma, and endothelioma (9), thus raising the question of its true cellular origin. This question is of particular current interest in view of recent findings by electron microscopy which clearly support a distinct origin for Kupffer and endothelial cells (10), as well as indicating the ab sence of endothelial cell to Kupffer cell transition (IllFactor VIII, a coagulation factor, was shown by Hoyer et al. (12), to be present in endothelial cells, megakaryocytes, and platelets. In the work pre sented here, using immunochemical evidence, the presence of abundant factor VIII in proliferating si nusoidal lining cells of angiosarcomatpus liver tissue is demonstrated; it supports the endothelial cell ori gin of vinyl chloride-associated angiosarcoma. Materials and Methods Hepatic tissue studies were conducted fag eight in dividuals--three with hepatic angiosarcoma related to heavy vinyl chloride exposure as production workers, one hepatic angiosarcoma associated with vinyl chloride used as hair spray propellant, and four normal subjects (trau matic death with no liver injury) who were used as con trols. Reagents. Purified human factor VIII waa pro vided by the Louisville Red Cross Blood Center. Corre sponding antiserum prepared in rabbits was from Behring Diagnostics (American Hoechst Corp., Somerville, N.J.). A single precipitin line in immunoelectrophoresis was seen with this material when tested against human plasma and purified fector VIII. Fluorescein-conjugated goat anti serum to rabbit IgG (GARI) was obtained from Meloy Lab oratories, Inc. (Springfield, Va.). Tissue specimens. Both normal and hepatic angio sarcoma tissue were obtained at postmortem examination within 30 min to 6 h of death. Portions of normal and tu mor tissues were processed for standard histologic and immunofluorescent staining. Immunofluorescence technique. Frozen sections (8 CMA 003587 1418 FORTWENGLER ET AL. jum) were cut on a cryostat, air dried, fixed for 5 min in cold acetone and rehydrated in phosphate-buffered saline (PBS). The sections were then treated for 30 min with anti serum to factor VIII, washed with PBS, and stained with fluorescein-conjugated GARI for 45 min. After three 5-min washes m PBS, the sections were examined by immunofluorescent microscopy utilizing a Vanox microscope (Olympus Corporation of America), and an HB0200 UV il luminator with a 3-mm BF 12 exciter filter and a G530 bar rier filter. Control sections were incubated with nonimmunized rabbit serum. Specificity of factor VIII staining was demonstrated by inhibition of the staining reaction after absorption of the anti-factor VIII rabbit serum with the purified factor VIII and by the negative staining reaction found using nonimmune rabbit serum. Results Factor VIII immunofluorescence of endo thelial cells in human umbilical cord shows a ho mogeneous continuous staining pattern (Figure 1). This preparation served as a positive control. Sec tions from normal liver did not display any specific factor VIII immunofluorescence in the sinusoidal GASTROENTEROLOGY Vol. 80, No, S Figure 1. Umbilical cord, endothelial lining cells (arrows) show ing factor VIII related immunofluorescence in white (X 100). Figure 2. Fluorescent staining of normal liver showing minimal granular punctate factor VIII fluorescence in hepatic si nusoids (white arrows) (x 100). areas except for occasional spotty granular areas of minimal intensity along the sinusoidal borders (Fig ure 2). Only the linings of the hepatic arteries and veins consistently gave intense staining reactions. In contrast, vinyl chloride hepatic angiosarcoma tumor tissue demonstrated intense fluorescence occurring in the proliferating cells lining the enlarged sinu soids (Figure 3). These areas of specific immuno fluorescence were limited to the sinusoidal spaces, were multifocal and occasionally occupied up to 20% of the tissue specimen. There was no demonstrable difference among the four angiosarcoma specimens as to the frequency of stained cells or the pattern of specific fluorescence. Adjacent tissue stained with H & E documents angiosarcomatous cells within the enlarged sinusoids lining the hepatocytic cords (Fig ure 4). Fluorescence was present in the cytoplasm and cell surface but not in the nuclei of the sinusoidal cells as illustrated in Figure 5. This distinguishable pattern of factor VIII immunofluorescent staining was seen in all the angiosarcomas examined and in none of the normal hepatic tissue. CMA 003588 |un.' 19S1 CELLULAR 0RIC1N OF HEPATIC ANGIOSARCOMA 1417 described it as composed of cells closely resembling Kupffer cells and indicated that some tumor cells were phagocytic. Nevertheless, the tumor was dc^fl nated "endothelial cell sarcoma of the liver." Lat^ in reporting a tumor, Baker et al. (16), made a dis tinction between the apparent maturation stages of sinusoidal lining cells, i.e. "Kupffer cells" and "nor mal resting sinusoidal endothelia." He noted that the tumor cells were phagocytic and therefore, de scribed the tumor as originating from Kupffer colls, as did Burston (17). Edmonson's classical description of this tumor (9) makes no distinction between endo thelial and Kupffer cells, but describes it as com posed of malignant endothelial cells. In the initial report of arsenic-associated "hemangioendothelial sarcoma" of the liver due to ingestion of Fowler's solution, Regelson et al. (18), indicated the presence of neoplastic endothelial cells but did not differentiate Kupffer from endothelial cells. On the other hand, Blackwell et al. (19), made no dis- Figure 3. A section of liver angiosarcoma tissue exhibiting one of the multifocal areas displaying marked fluorescence in the proliferating sinusoidal cella (white arrows), with some nonspecific background staining of hepatocytea (X 200). Discussion Descriptions and definitions of hepatic angio sarcoma have varied in the past and'have relied pre dominantly on morphologic criteria for identifying the various sinusoidal cells. Some investigators have proposed that both Kupffer and sinusoidal endothelial cells are transitory in respect to each other and that Kupffer cells merely represent activated endothelial cells (5,13). In con trast, others support the view that the Kupffer cell is independent in nature and represents a hepatic macrophage (14,15). As early as 1939, Miller considered the Kupffer cell as a class of endothelial cells and distinguished them as "endothelial cells of Kupffer (13)." He sug gested the term endothelioma or endothelioblastoma for the tumors arising from Kupffer cells and heman gioendotheliomas or hemangioendothelioblastoma for those derived from vascular endothelium. In the first communication of a thorotrast-associated ma lignant hepatic vascular tumor, McMahon ct al. (1). Figura 4. A frozen section of hepatic angiosarcoma tissue dem onstrating individual and small groups of proliferating endothelial cella in dilated sinusoids (block arrows) (H & E X 400). cm 03S89 cLscrioeu t`. as comuosi d of dos:,.v :'.--TrT'c huprror ,,o!l3 ur.j .nciicated ;hut some tumor '."'Is sinusoidal lining cells, i.e. "Kupffer cells" and "nor mal resting sinusoidal endothelia." He noted that the tumor cells were phagocytic and therefore, de scribed the tumor as originating from Kupffer cells, as did 3-.ru,cn (./), fdmc,-jcn ^ c,ussiuui of this tumor (9) makes no distinction between endo thelial and Kupffer cells, but describes it as com posed of malignant endothelial cells. In the initial report of arsenic-associated "hemangioendothelial sarcoma" of the liver due to ingestion of Fowler's solution, Regelson et al. (!8), indicated the presence iff neeplust.c endothelial ceils cut did not differentiate Kupffer from endothelial cells. On the other hand, Blackwell et al. (19), made no dis- Figure 3. A section of liver angiosarcoma tissue exhibiting one of the multifocal areas displaying marked fluorescence in the proliferating sinusoidal cells (white arrows), with some nonspecific background staining of hepatocytes (x 200). Discussion Descriptions and definitions of hepatic angio sarcoma have varied in the past and have relied pre dominantly on morphologic criteria for identifying the various sinusoidal cells. Some investigators have proposed that both Kupffor and sinusoidal endothelial cells are transitory in respect to each other and that Kupffer cells merely represent activated endothelial cells (5,13). In con trast, others support the view that the Kupffer cell is independent in nature and represents a hepatic macrophage (14,15). As early as 1939, Miller considered the Kupffer cell as a class of endothelial cells and distinguished them as "endothelial cells of Kupffer (13)." He sug gested the term endothelioma or endothelioblastoma for the tumors arising from Kupffer cells and heman gioendotheliomas or hemangioendothelioblastoma for those derived from vascular endothelium. In the first communication of a thorotrast-associated ma lignant hepatic vascular tumor, McMahon ct al. (1), Figur* 4. A fatten section o( hepatic angioaarcoraa tiaaue dem onstrating individual and final] group* of proliferating endothelial cell* in dilated tinuioidi (black arrow*) (H A E x 400). CMA 003590 1418 FORTWENGLER ET AL GASTROENTEROLOGY Vol. 80, No. 8 Kupffer cells which arc peroxidase positive and phagocytic. In the present work we have shown by an iramunofluorescent procedure that Factor VIII is present in the proliferating cells of vinyl chloride-associated angiosarcoma, thus confirming their endothelial cell origin. Our observation that normal hepatic tissue displays intense factor VIII fluorescence in cross sections of vessels, but limited fluorescence of the si nusoidal endothelial cells in comparison to angio sarcomatous tissue, suggests that the aberrant cells may have an increased production or storage capac ity for factor VIII or a decreased transport from the cell. In addition, staining for factor VIII may provide a means for identification of vascular cell in volvement in hepatic tumors and could be used to identify specific endothelial cell changes and their relationship to chemical biotransformation in both animals and humans. In view of these findings, we suggest that the term "endothelial cell angiosarcoma" be used as an ap propriate term for this tumor. References Figure 5. High-powered frozen section of angiosarcomatous tis sue showing factor Vttl immunofluorescence of indi vidual proliferating cells lining hepatic cords (white ar rows). Fluorescent staining of cytoplasm but not cell nucleus is seen in one of the cells (x 1000). tinction between these cells but called the tumor Kupffer cell sarcoma. In more recent reports involving vinyl chloride-as sociated angiosarcoma, the question of cellular ori gin of the tumor persists. This malignancy has been variously described as pleomorphic malignant endo thelial cells (9), malignant spherical and spindle cells (20), "tumor cells" (21), sinusoidal cells (22). and as a tumor believed to be malignant Kupffer cells (23). Makk et al. (24) and Popper et al. (25) depicted this tumor as arising from sinusoidal endothelial cells. In comparing the pathology of liver angiosarcoma due to vinyl chloride, thorotrast, and arsenic. Popper et al. (25) concluded from light and electron micros copy that the tumors were probably endothelial and suggested confirmation by use of histologic markers. An appropriate marker which concentrates in the endothelial cells is provided by coagulation factor VUI which differentially appears in the endothelial cells, megakaryocytes and platelets (12) and not in 1. MacMahon HE. Murphy AS, Bate* MI. Endothelial-cell sar coma of liver following thorotraat injection*. Am J Pathol 1947:23:585. 2. Roth F. The sequelae of chronic arsenic poisoning in moselle vintners. Ger Med Mon 1957:2:172. 3. Falk H, Popper H, Thomas LB. et al. Hepatic angiosarcoma associated with androgenic-anabolic steroids. Lancet 1979;ii:ll2Q. 4. Creech JL Johnson MN. Angiosarcoma of the liver in the manufacture of polyvinyl chloride.) Occup Med 1974:16:150. 5. Gibbs WN, Connor CR, Hutchison HE. Malignant hemangio endothelioma associated with thrombocytopenia. ) Pathol Bact 1960,-92:207. 6. MacSween RNM, Vetters JM. Ross SK, et al. Haemangiocndothelial sarcoma of the liver. J Pathol 1973;109:39. 7. Stout AP. Hemangio-endothelioma: a tumor of blood vessels featuring vascular endothelial cell*. Ann Surg 1943:118:445. 8. Greenberg M. Kupffer cell sarcoma of the liver. Report of 2 cases in South African blacks. SA Med I 1977,52:244. 9. Edmonson HA. Tumors of the liver and intrahepatic bile ducts. Atlas of Tumor Pathology, Section 7, Fascile 25. Wash ington, D.C.. Armed Force* Institute of Pathology, 1958:13945. 10. Van Furth R, et al. The bone marrow origin of Kupffep cells. In: Wisse E. Knook DL eds. Kupffer cells and other liver si nusoidal cell*. Amsterdam: Elaevier/North Holland, 1977:471. 11. Wisse E, Knook DL The investigation of sinusoidal cells: a now approach to the study of liver function. Prog Liver Dis 1979:6:153. 12. Hoyer LW, de loa Santos RP. Hoyer JR. Antihemophilic factor antigen. Localization in endothelial cells by immunofluorcscent microscopy. J Clin Invest 1973:53:2737. 13. Miller |K. Primary sarcoma of the liver endothelioblastoma. Am | Surg 1939:45:459. 14. Wisse E. Ultrastructure and function of Kupffer cells and other sinusoidal cells in the liver. In: Wisse E. Knook OL. eds. CMA 003591 fune 1961 CELLULAR ORIGIN OF HEPATIC ANGIOSARCOMA 1417 described it as composed of cells closely resembling Kupffer cells and indicated that some tumor csfl were phagocytic. Nevertheless, the tumor was deJJ nated "endothelial cell sarcoma of the liver." Later, in reporting a tumor, Baker et al. (16), made a dis tinction between the apparent maturation stages of sinusoidal lining cells, i.e. "Kupffer cells" and "nor mal resting sinusoidal endothelia." He noted that the tumor cells were phagocytic and therefore, de scribed the tumor as originating from Kupffer cells, as did Burston (17). Edmonson's classical description of this tumor (9) makes no distinction between endo thelial and Kupffer cells, but describes it as com posed of malignant endothelial cells. In the initial report of arsenic-associated "hemangioendothelial sarcoma" of the liver due to ingestion of Fowler's solution, Regelson et al. (18), indicated the presence of neoplastic endothelial cells but did not differentiate Kupffer from endothelial cells. On the other hand, Blackwell et al. (19), made no dis- Figure 3. A section of liver angiosarcoma tissue exhibiting one of the multifocal areas displaying marked fluorescence m the proliferating sinusoidal cells (white arrows), with some nonspecific background staining of hepatocytes (x 200). Discussion Descriptions and definitions of hepatic angio sarcoma have varied in the past and have relied pre dominantly on morphologic criteria for identifying the various sinusoidal cells. Some investigators have proposed that both Kupffer and sinusoidal endothelial cells are transitory in respect to each other and that Kupffer cells merely represent activated endothelial cells (5,13). In con trast. others support the view that the Kupffer cell is independent in nature and represents a hepatic macrophage (14,15). As early as 1939, Miller considered the Kupffer cell as a class of endothelial cells and distinguished them as "endothelial cells of Kupffer (13)." He sug gested the term endothelioma or endothelioblastoma for the tumors arising from Kupffer cells and heman gioendotheliomas or hemangioendothelioblastoma for those derived from vascular endothelium. In the first communication of a thorotrast-associated ma lignant hepatic vascular tumor, McMahon et al. (1), Figure 4. A frozen section of hepatic angiosarcoma tissue dem onstrating individual and small groups of proliferating endothelial cells in dilated sinusoids (black arrows) (H 4 E x 400). CMA 003592 J .OJl CELLO LA A OKiCIA 'JF HEP A i IC A.VjIGE \ SO'A cede:.sea ;t as composed or ce.is coiuy rosem.ciing K'.cC'-r cells tied indicated `ha; -crr.i' -r> \ nateG "I'niju'r.pi: <i i,en -.arccm.n ot tne ever ' Lr; in rcDurlira \ ccr,-, A -rc: *f }' i ' i> -- i - , tinction between the apparent maturation stages of sinusoidal lining cells, i.e. ' Kupffer cells ' and ' nor mal resting sinusoidal endotheha." He noted that the tumor cells were phagocytic and therefore, de scribed the tumor as originating from Kuprfer cells, as did Burston (17). Edmonson's classical description of this tumor (9) makes no distinction between endo thelial and Kupffer cells, but describes it as com posed of malignant endothelial cells. In the initial report of arsenic-associated hemangioendothelial sarcoma" of the liver due to ingestion of Fowlers solution. Reclsr>n et al. riai, mdici'M the presence of neoplastic endothelial cells but did not differentiate Kupffer from endothelial cells. On the other hand. Blackwell et al. (19). made no dis- Figure 3. A section of liver angiosarcoma tissue exhibiting one of the multifocal areas displaying marked fluorescence in the proliferating sinusoidal cells (white arrows), with 3ome nonspecific background staining of hepatocytes (X 200). Discussion Descriptions and definitions of hepatic angio sarcoma have varied in the past and have relied pre dominantly on morphologic criteria for identifying the various sinusoidal ceils. Some investigators have proposed that both Kupffer and sinusoidal endothelial cells are transitory in respect to each other and that Kupffer cells merely represent activated endothelial ceils (5,13). In con trast. others support the view that the Kupffer cell is independent in nature and represents a hepatic macrophage (14,15). As early as 1939, Miller considered the Kupffer cell as a class of endothelial cells and distinguished them as "endothelial cells of Kupffer (13)." He sug gested the term endothelioma or endothelioblastoma for the tumors arising from Kupffer cells and heman gioendotheliomas or hemangioendothelioblastoma for those derived from vascular endothelium. In the first communication of a thorotrast-associated ma lignant hepatic vascular tumor. McMahon et al. (1), Figure 4. A frozen section of hepatic angiosarcoma tissue dem onstrating individual and small group# of proliferating endothelial cells in dileted sinusoids (black arrows) (H 4 E x 400). CMA. 003593 |uni: 1981 CELLULAR ORIGIN OF HEPATIC ANGIOSARCOMA 1419 Kupffer colls and other liver sinusoidal cells, Amsterdam Elsevier/North Holland. 1977:33 15. Naito M, Wisse E. Observations on the fine structure and cytochemistry of sinusoidal cells in fetal and neonatal rat liver. In: Wisse E, Knook DL, eds. Kupffer cells and other liver sinusoidal cells. Amsterdam: Elsevier/North Holland. 1977.497. 10. Baker H de C, Paget GE, Davson J. Hemangioendotheliomas (Kupffer-cell sarcoma) of the liver. J Pathol Bact 1956:72:173. 17. Burston (. Kupffer cell sarcoma. Cancer 1958:11:798. 18. Regelson W. Kim U. Ospino J. et al. Hemangioendothelial sar coma of liver from chronic arsenic intoxification by Fowler s solution. Cancer 1968:21:514. 19. Blackwell JB, Joske RA. Kupffer cell sarcoma. Dig Dis 1970:15 133. 20. Lee FI, Harry DS. Angiosarcoma of the liver in a vinyl chlo ride worker. Lancet 1974;i:1316. 21 Pollard SM. Millward-Sadler GH. Malignant haemangioendothelioma involving the liver. | Clin Pathol 1974:27,214, 22 Thomas LB, Popper H, Berk PD, et al. Vinyl chloride-induced liver disease. N Engl | Med 1975:292.17. 23 Smith PM, Williams Mf, Evans DMD. Hepatic angiosarcoma in vinyl chloride workers. Bull NY Acad Med 1976:52:447. 24. Makk L. Delmore F. Creech JL Jr. et al. Clinical and morpho logic features of hepatic angiosarcoma in vinyl chloride workers. Cancer 1978:37:149. 25. Popper H, Thomas LB, Telles N, et al. Development of hepatic angiosarcoma in man induced by vinyl chloride, thorotrast, and arsenic. Am | Pathol 1978:92:349. ' *C CMA 003594 2 BIOCHEMICAL ALTERATIONS IN LIVERS OF RATS EXPOSED TO VINYL CHLORIDE Julie T. Du, John P. Sandoz, Michael T. Tseng, Carlo H. Tamburro Division of Digestive Diseases and Nutrition, Department of Medicine, Department of Anatomy, and Cancer Center, University of Louisville School of Medicine, Louisville, Kentucky Sprague-Dawlty rati were exposed to vinyl chloride to determine the earliest sequential biochemical changes occurring with liver injury before angiosarcoma development. Activity of glucose-6-phosphatase, a hey gluconeogenic enzyme in the liver microsomal fraction, decreased 25% with respect to controls after 70 h of exposure, Glucose-6-phosphote dehydrogenase activity Increased twofold after more than 100 h of exposure. Nonprotein sullhydryl levels (glutathione and(or cysteine) showed a slight but progressive elevation, whereas glutathione reductase activity increased 50-60Jl during exposure to vinyl chloride, NAOPH-cytochrome c reductase and mixed function oxidase were unchanged in the same microsomal fraction. There were no changes In seven conventional clinical biochemical liver ttsts or in four other markers of liver mitochondrial, cytosol, and microsomal function. No significant histological changes were found on light microscopic examination during this exposure period. However, with electron microscopy, dilation of rough endoplasmic reticulum was seen in the animals txposed far more than 137 h. These enzymatic changes are considered to reflect early hepatocellular adaptation to vinyl chloride exposure with very mild or limited hepatocellular Injury In Its earliest stage. INTRODUCTION Vinyl chloride has been shown to induce tumors, including angio sarcoma, in laboratory animals (Maltoni and Lefemine, 1975; Viola et al., 1971) and angiosarcoma in humans (Creech and Johnson, 1974). At high concentrations, vinyl chloride is believed to be metabolized by the microsomal mixed function oxidase (MFO) system of the liver to toxic metabolites (Bolt et al., 1975; Hefner et al., 1975; Johnson, 1967; Watanabe et al., 1976b, 1976c). Animals pretreated with phenobarbital, an inducer of We wish to express our sincere thanks to the people in the B, F. Goodrich Plant to Louisville.^ for their cooperation in the exposure studies, Or. R. A~Greenberg for help with statistics, andTrfs. Ruth Shelton tor technical assistance. This work was supported by a grant from the Manufacturing ChcmHts Association, Washing ton, D.C, Requests for reprints should be sent to Carlo H. Tamburro, Division of Digestive Diseases and Nutrition, Health Sciences Center, MDR 535, University of Louisville, Louisville, Kentucky 40232. 1119 Journal of Toxicology and Environmental Health, J:1119-1132, 1919 Copyright O 1979 by Hemisphere Publishing Corporation 0099-41 Oa/79/051119-1452.25 CMA 003595 1120 J.T. OU 6T Al. MFO, and exposed to 5% vinyl chloride (50,000 ppm) had increased serum alanine aminotransferase, a conventional biochemical indicator of liver injury, and increased scrum sorbitol dehydrogenase, a liver specific enzyme (jaeger et al., 1974). However, there have been no sequential enzymatic studies to determine what progressive metabolic alterations of the hepatocyte occur during vinyl chloride exposure preceding the development of angiosarcoma. Although the hepatocyte is the major cell for oxidation and. probably detoxification of vinyl chloride and its metabolites, angiosarcoma develops not in the hepatocyte but in adjacent mesenchymal cells. During this time, the hepatocytes undergo changes (focal nodular hyperplasia) suggesting focal regeneration, probably from limited hepatocellular injury (Tamburro et al., 1979). The purpose of our study was to characterize the earliest sequential enzymatic changes that occurred in vinyl chloride exposure to elucidate the hepatocytes' role in vinyl chloride-induced injury and angiosarcoma develop* ment. Portions of this study have been presented elsewhere (Du and Tamburro, 1976). METHODS Animals and Experimental Design Sprague*Dawley male rats (300-500 g) were randomly assigned before each experiment to either a group to be exposed to vinyl chloride gas (15,000 ppm) or a nonexposed control group kept outside the chamber in the animal room. Animals were feef standard laboratory chow pellets ad libitum. Four sequential experiments were done. In experiments A and B, rats were exposed 2-4 h/d, 4 d/wk, for 1-2 wk with total accumulated exposure times of 14, 28, and 42 h. In experiment C, rats were exposed 4-8 h/d, 5 d/wk, for 2-3 wk with total accumulated exposure periods of 71 and 103 h. In experiment D, rats were exposed 6-8 h/d, 5 d/wk, for 3-4 wk with' total accumulated exposure periods of 84 and 137 h. The animals were exposed in a modified chamber consisting of an airtight vat that had been used for manufacturing polyvinyl chloride. An aliquot of vinyl chloride (185 g) was added to the vat per 4*d period to make an average concentration of 15*000 4000 ppm. The air was constantly circulated by a stirrer. The chamber volume was 4400 I (1100 gal), so the' respiration of the rats had a negligible effect on the composition of the chamber's atmosphere. Experiments with an added control group exposed only to air in an identical modified vat showed no differences from the controls kept in the animalTOom (Table 1). All animals were anesthetized with ether, had blood drawn by cardiac puncture, and were sacrificed at the same time of day (within 3 h after final exposure) to avoid diurnal effects. CMA 03596 VINYL CHLORIDE EFFECTS ON RAT LIVER 1121 TABLE 1. Compirhon of Hepatic Enzyme Activity in Animal Control Group* Based in Animal Quartan and in the Exposure Chamber Air exposed Enlyme Animal room Chamber^ Glucoj*6-phojphit dehydrogenase Glutathione reductase 12,82 t 1.32 4.77 t 0.27** 11.06 i 1.78 5.28 t 0.40 Value* are means a 5EM (rt 6). "Kept in chamber 7 h/d, 5 d/wk, for 6 wk; tout, 210 h. Not significant. "Not significant. Sample Preparation Liver was excised rapidly and was immediately rinsed in ice-cold 0.15 M KCI with 0.02 M Tris buffer, pH 7.4. A portion of the tissue was homogenized with 9 volumes of the cold KCI-Tris buffer in a PotterElvehjem homogenizer. Each sample was prepared from a single organ and kept at 4WC during preparation. Remaining liver was frozen rapidly and stored at --20C. For assays with frozen tissue, livers from control and experimental rats were frozen in an identical manner for the same length of time. Subcellular Fractionation and Biochemical Determination Homogenate was centrifuged at 600 and 8000 Xg for 10 min and at 100,000 Xg for 60 min to obtain the nuclear, mitochondrial, microsomal, and cytosol fractions, respectively, by the differential technique of Schneider and Hogeboom (1950). Centrifugation was at 4C in a Sorvall refrigerated RC5 supercentrifugc with a fixed angle rotor and a Beckman model L-5 ultracentrifuge with a swinging bucket rotor. Enzyme activities and cytochrome P-450 were determined in the isolated subcellular frac tions: cytochrome oxidase in the mitochondrial fraction; NADPHcytochome c reductase, mixed function oxidase, cytochrome P-450, and glucose-6-phosphatase in the microsomal fraction; and glutathione reductase and giucose-6-phosphate dehydrogenase in the 100,000 X g supernatant fraction. Cytochrome P-450, NADPH-cytochrome c reductase, and cytochrome oxidase were determined in fresh samples; all other determinations were in ^ freshly isolated fractions from frozen tissue.-Glucose-6-phosphatase was estimated by measuring release of inorganic phosphate (Harper, 1965). Cytochrome P-450 was estimated by the maximum absorption difference between the dithionite-reduced cytochrome and its CO complex (Omura and Sato, 1964). The P-450 concentration was calculated by using 91,000 CMA 003597 V'! VI r.2i 1. "i* rtspiiic r:yme Ac;,\n(i; in Animal Control Groups Based in Animal Quartan and in the Exposure Chamber0 Air expoied Errlyme AAimil r^om GIucow-6-phoiphiu dchydrogenu* Glu richion* rtductue 12.82 t 1.32e 4.77 0.27^ 11.06 i 1.7$c J.2S i 0.40rf Rvalue* tft meins t SEM (n m s), "Kept in chamber 7 h/d, 5 d/wk, for 6 wk ; total, 21 Oh. Not significant. "Not significant. Sample Preparation Liver was excised rapidly and was immediately rinsed in ice-cold 0.15 M KCI with 0.02 M Tris buffer, pH 7.4. A portion of the tissue was homogenized with 9 volumes of the cold KCl-Tris buffer in a PotterIvehjem homogeni2er. Each sample was prepared from a single organ and kept at 4C during preparation. Remaining liver was frozen rapidly and stored at --20C. For assays with frozen tissue, livers from control and experimental rats were frozen in an identical manner for the same length of time. Subcellular Fractionation and Biochemical Determination Homogenate was centrifuged at 600 and 8000 Xg for 10 min and at 100,000 Xg for 60 min to obtain the nuclear, mitochondrial, microsomal, and cytosol fractions, respectively, by the differential technique of Schneider and Hogcboom (1950). Centrifugation was at 4C in a Sorvall refrigerated RC5 supercentrifugc with a fixed angle rotor and a Beckman model L-5 ultracentrifuge with a swinging bucket rotor. Enzyme activities and cytochrome P-450 were determined in the isolated subcellular frac tions: cytochrome oxidase in the mitochondrial fraction; NADPHcytochome c reductase, mixed function oxidase, cytochrome P-450, and glucose-6-phosphatase in the microsomal fraction; and glutathione reductase and glucose-6-phosphate dehydrogenase in the 100,000 X g supernatant fraction. Cytochrome P-450, NADPH-cytochrome c reductase, and cytochrome oxidase were determined in fresh samples; all other determinations were in freshly isolated fractions from frozen tissue. ~Glucose-_6-phosphatase was estimated by measuring release of inorganic phosphate (Harper, 1965). Cytochrome P-450 was estimated by the maximum absorption difference between the dithionite-reduced cytochrome and its CO complex (Omura and Sato, 1964). The P-450 concentration was calculated by using 91,000 CMA 003598 1122 J.T. DU ETAL. M~x cm'1 as the extinction coefficient for the increase in peak height between 490 and 450 nm. Cytochrome oxidase was measured as described by Wharton and Tzagoloff (1967), and NADPH-cytochrome c reductase according to Degroot and Dunn (1964). Mixed function oxidase was determined according to Holtzman et al. (1968) by measuring the hydroxylation of analine. Nonprotein sulfhydryl was estimated by the method of Sediak and Lindsay (1968). Rate of oxidation of NADPH by oxidized glutathione was used as a measure of enzymatic activity of glutathione reductase (Carlberg and Mannervik, 1975). Glucose-6phosphate dehydrogenase was measured spectrophotometricafly, as the rate of NADPH formation (Lohr and Waller, 1965). Assays were conducted under conditions of linearity with respect to both rime and protein. Protein content was determined by the method of Lowry et al. (1951). Conventional clinical analyses of serum included aspartate aminotransferase (AST, SGOT), alanine aminotransferase (ALT, SGPT), alkaline phosphatase, bilirubin,albumin, cholesterol, and triglyceride, determined by the technicon sequential multiple analyzer computer (SMAC) system. Materials The NADPH, cytochrome c, and oxidized glutathione were obtained from Sigma Chemical Co., St. Louis, Mo. Analine and other chemicals were reagent grade. Double-distilled water was used throughout. Light and Electron Microscopy Small strips of liver were removed under ether anesthesia and , immersed immediately in icc-cold 3% glutaraldchyde (pH 7,4). Tissues were sliced into small cubes and fixed for 2 h at 4"C. Subsequently, samples were washed overnight in phosphate buffer and postfixed in osmium tetroxide for 1 h before being dehydrated in ascending alcohol and embedded in Epon. Tissue blocks were polymerized at 60oC for 2 d. Thin sections were cut with a diamond knife and stained with uranyl acetate and lead citrate before examination on a Philips 300 electron microscope. For ultrastructural analysis, control rats and rats exposed for 42 and 137 h to vinyl chloride were reviewed. For light microscopy, a block of tissue was fixed in buffered formalin and processed routinely for paraffin embedding. Sections 6 pm thick were stained with hematoxylin and eosin before examination. RESULTS The protein contents (milligrams per gram of liver) of the subcellular fractions in control and vinyl chloride-exposed groups were the same throughout the exposure; therefore, the enzymatic results were expressed as micromoles converted per minute per milligram, of subcellular protein. The statistical analysis (sec below) showed certain significant 003599 CMA VINYL CHLORIDE EFFECTS ON RAT LIVER 1123 enzymatic differences between the exposed and control groups in glucosc6-phosphatasc, glutathione reductase, and glucose-6-phosphatc dehydro genase. Glucose-6-phosphatase. No significant differences in glucose-6-phosphatase activity were detected between the two groups up to 42 h; however, after 71 h, the glucose-6-phosphatase activity was significantly less in the exposed than in the control animals. Figure 1 a shows the 25% decrease in mean activity of glucose-6-phosphatase after 71 h; this activity remained significantly lower up to 137 h. Figure 1 b shows the 95% confidence intervals for D, the mean difference between the exposed group and the control groups, in each experiment. (> B .j . .4 fxpriinftt 4 ft < CspOftUr# 'MV* MM* M,t01 I hr*) (ft) 0 FIGURE 1. (e) Composite curve of group mean* of specific activity of glueow-6-phosphatasc with respect to exposure time; (b) 95% confidence interval* for D, the mean difference between the exposed and control groups. There it no significant difference between the exposed and control groups until after 42 h of exposure. After 71 h the mean level of the exposed group U significantly less than that of the control group. - CMA 003600 enzymatic differences between the exposed and control groups in glucose- 6-phospfutasc, glutathione reductase, and glucose-6-phosphatc dehydro genase. Glucose-6-phosphatase. No significant differences in glucose-6-phosP- -c`.:vi:y wsrp cist-acted bet./cen the two groups up to 42 h; however, after 71 h, the glucose-6-phospbatase activity was significantly less in the exposed than in the control animals. Figure 1 a shows the 25% decrease in mean activity of g!ucase-6-phosphatasc after 71 h; this activity remained significantly lower up to 137 h. Figure 16 shows the 95% confidence intervals for D, the mean difference between the exposed group and the control groups, in each experiment al . . i lipawr* f hr*J i VM1 () c >V* W" FIGURE 1. () Composite cum of (roup mans of specific activity of glucoK-$-phosphatase with respect to exposure time; (b) 95% confidence Intervals for O, the mean difference between the exposed and control (roups. There is no significant difference -between the exposed and control (roups until after 42 h of exposure. After 71 h the mean level of the expoied (roup Is significantly less than that of the control group. .. CMA 003601 1124 I, r. OU ET At. Glutathione Reductase. Figure 2, a and b, illustrates the significant differences (p<0.05) between the exposed and control groups in all 4 experiments except at 71 h of exposure (p<0.06). The glutathione reductase level was 25% greater in exposed than in nonexposed animals up to 71 h and approximately 50% greater after 84 and 137 h. Glucose-6-phosphate Dehydrogenase. There was no significant differ ence between exposed and control groups from 14 to 42 h and at 71 and 84 h. However, after 103 and 137 h of exposure there was a statistically significant difference as shown in Fig. 3, a and b. After about 100 h of exposure, the mean glucose-6*phosphate dehydrogenase level in exposed animals was about twice that in nonexposed animals. While these differences were occurring, none of the conventional biochemical liver function tests (e.g., serum aminotransferases, alkaline phosphatase) showed any significant change, nor were there any changes in cytochrome oxidase, mixed function oxidase, P-450, or NADPH* cytochrome c reductase. FIGURE 2. (e) Composite curve of group means of specific activity of glutathione reductase with respect to exposure time; (b1 95% confidence intervals for D, th* mean difference between the exposed and control froup. The mean value for the exposed group is significantly greater Due that for the controls throughout the entire experiment {ft < 0.05) except at 71 h of exposure. CMA 003602 VINYL CHLORIDE EFFECTS ON RAT LIVER ms (*) FIGURE 3. (a) Composite curve of (roup owns of specific activity of glucott'6-phosphate dehydrogenate with reaped to exposure time; (6) 9SK confidence Intervals for D, the mean difference between the exposed and control (roups. The mean value for the exposed (roup Is not significantly different from that for the control group until after 44 h of exposure. The concentration of reduced glutathione in the livers of rats decreased to 5296 of the control value after a single 2-h exposure to 15,000 ppm vinyl chloride (Table 2). However, the reduced glutathione concentration appeared to be the same or slightly elevated when the rats underwent multiple exposures. This elevation was not significant at the 596 level. The vinyl chloride-exposed rats suffered a weight loss of 4-1396 and the control group gained 2-1096 (Table 3). All rats survived to the termination of the experiment without noticeable ill effects. 1 Statistical Analysis of Results The 9596 confidence interval for group mean differences in each enzyme study was based on the error mean square from a two-factor analysis of variance with interaction,, the two main effects being exposure time in hours and cxposurc-nonexposurc. This analysis was performed for 003603 n:3 Exposure Tim* \hfsJ Exptn'mMf a Expour* **M* C ",** o %* FIGURE 3. (0) Composite curve of iroup main* of specific activity of gtucoM-d-pttosplutc dafcydroganas* with raspaet to exposure time; (6) 95* confidant* Intervait for D, the mam difference batwaan tha exposed and control groups, The mean value for the exposed group Is not significantly different front that for the control group until after 14 h of axposura. The concentration of reduced glutathione in the livers of rats decreased to 52% of the control value after a single 2-h exposure to 15,000 ppm vinyl chloride (Table 2). However, the reduced glutathione concentration appeared to be the same or slightly elevated when the rats underwent multiple exposures. This elevation was not significant at the 5% level. The vinyl chloride-exposed rats suffered a weight loss of 4-13% and the control group gained 2-10% (Table 3). All rats survived to the termination of the experiment without noticeable ill effects. a**'*'--" Statistical Analysis of Results The 95% confidence interval for group mean differences in each enzyme study was based on the error mean square from a two-factor analysis of variance with interaction,, the two main effects being exposure time in hours and cxposurc-nonexposurc. This analysis was performed for CMA 003604 1126 J.T. DU ETAL. TABLE 2. Effect of Vinyl Chloride Exposure on Concentration of Liver Nonprotein Sulfhydryl Compound In Rati Exposure time (h) 2** 14e 28 42 71 84 103 137 Concentration ratio (exposed/con troll 0.52 (3)* 1.02 (6) 0.96 (6) 0.82 (6) 1-21 (5) 1.07 (3) 1-49 (6) 1.29 (3) `'single exposure. ^Number of animals in control or experimental group is shown in parentheses. `Multiple exposures (4-8 h/d, 4-5 d/wk) for 14-137 h. TABLE 3. Body Weights of Rats before and after Vinyl Chloride Exposure* Accumulated xpoHire (h) 14 28 42 71 14 103 137 Time Before After Before After Before After Before After Before After Before Affter Before After Control 442 e 6.2* (3)e 454 a 9.5 (3) 436 e 9.2 (3) 445 * 12.1 (3) 433 1 5.2 (3) 455 * 6.4 (3) 437.6 12.2 (5) 447 i 19.4 (5) 412.7*7.9(3) 455.0 * 6.8 (3) 453 12.9 (6) 465 e 16.9 (6) 404.7 a 1.2 (3) 425,0 a 5.2 (3) Change (*> 3 +2 5 2 +10 , 3 +5 Vinyl chlorhleexposad Chang* (*) 453 a 1.5 (3) 421 a 4.1 (3) . 448*6.0(3) 408 a 3.0 (3) 450 * 4.3(3) 410* 2.8 (3) 422.7 * 9.2 (S) 385.0*4.2(5) 425 * 17.4 (3) 399 * 72 (3) 441.5 a 6.8 (6) 397.0 * 5.7 (6) 415.0 a 15.7(3) 398 a 78 (3) -7 -9 -9 -13 -6 -10 -4 'Multiple, exposures to 1JK vinyl chloride for 14-137 h, 4-1 h/d, 4-5 d/wk. * Results are expressed as mean a SEM. `Number of animals It given in parentheses. CMA 003605 VINYL CHLORIDE EFFECTS ON RAT LIVER 1127 each of the four experiments. Overall significant (p < 0.05) differences between the exposed and control groups are found when the 95% confidence intervals (O values) do not contain zero. In Fig. 1 tr, there is no significant difference (/? > 0.05) between the 2 groups in experiments A and B because a D value of zero is within these two confidence intervals. However, for experiments C and D, the con fidence intervals do not contain a D value of zero, indicating a significant difference (p<0.05) between the exposed and control groups. When the D value is less than zero (as in Fig. 1 b), the experimental group is significantly losver than the control throughout the experiment; when the D value is greater than zero (as in Figs. 2 and 3), the experimental group is significantly higher than the control. Further, in experiments C and D of Fig. Id, a single confidence interval is shown even though there are two time periods of exposure each. This results from the nonsignificance of the interaction term in the analysis of variance, meaning that the observed difference at 71 h of exposure is not statistically different from the difference at 103 h. Similar conclusions hold for experiment D in Fig. 16. Figure 2b shows 2 confidence intervals for the set of experiments C at 71 and 103 h because the interaction of the time and exposure factors was significant; that is, the difference between the exposed and control groups after 103 h of exposure was significantly greater than the difference after 71 h. This is also reflected by the differences of the mean values at 71 and 103 h as shown in Fig. 2a. For the same reasons, there are also two confidence intervals for the C (71 and 103 h) and D (84 and 137 h) sets of experiments in Fig. 36. Morphological Studies Morphological light microscopic studies were performed in a blind (coded) and randomized fashion and did not show any evidence of hepatocellular injury or changes usually seen in the latter stages of vinyl chloride exposure. Besides some variations in glycogen content, no difference in the fine structure of the hepatocytes was observed between controls (Fig. 4) and those exposed to vinyl chloride for 42 h. Dilation of rough endoplasmic reticulum was observed in a small number of hepatocytes after 137 h of vinyl chloride exposure (Fig. 5). Although no noticeable change in the amount of smooth endoplasmic reticulum was associated with this change, a concomitant increase in cytoplasmic density was evident in these cells. Another type of lesion found in other hepatocytes was characterized by the presence of small patches of clear spaces, which tended to aggregate near the cell periphery (Fig. 6). Such lesions usually affect the adjacent cell equally. No other cell type appeared to be affected by vinyl chloride exposure in this study. CMA 003606 VINYL CHLORIOS E'FSCTS ON RAT L1V0R n?7 each of the four experiments. Overall significant (p<0.05) differences between the exposed and control groups are found when the 95% confidence intervals (D values) do not contain zero. In rig. Ya, there is no significant difference (p>0.05) between the 2 groups in experiments A and B because a D value of zero is within these two confidence intervals. However, for experiments C and D, the con fidence intervals do not contain a D value of zero, indicating a significant difference (p < 0.05) between the exposed and control groups. When the D value is less than zero (as in Fig. 16), the experimental group is significantly lower than the control throughout the experiment; when the D value is greater than zero (as in Figs. 2 and 3), the experimental group is significantly higher than the control. Further, in experiments C and D of Fig. 16, a single confidence interval is shown even though there are two time periods of exposure each. This results from the nonsignificance of the interaction term in the analysis of variance, meaning that the observed difference at 71 h of exposure is not statistically different from the difference at 103 h. Similar conclusions hold for experiment D in Fig. 16. Figure 26 shows 2 confidence intervals for the set of experiments C at 71 and 103 h because the interaction of the time and exposure factors was significant; that is, the difference between the exposed and control groups after 103 h of exposure was significantly greater than the difference after 71 h. This is also reflected by the differences of the mean values at 71 and 103 h as shown in Fig. 7a. For the same reasons, there are also two confidence intervals for the C (71 and 103 h) and D (S4 and 137 h) sets of experiments in Fig. 36. Morphological Studies Morphological light microscopic studies were performed in a blind (coded) and randomized fashion and did not show any evidence of hepatocellular injury or changes usually seen in the latter stages of vinyl chloride exposure. Besides some variations in glycogen content, no difference in the fine structure of the hepatocytes was observed between controls (Fig. 4) and those exposed to vinyl chloride for 42 h. Dilation of rough endoplasmic reticulum was observed in a small number of hepatocytes after 137 h of vinyl chloride exposure (Fig. 5). Although no noticeable change in the amount of smooth endoplasmic reticulum was associated with this change, a concomitant increase in cytoplasmic density was evident in these cells. Another type of lesion found in other hepatocytes was characterized by the presence of small patches of clear spaces, which tended to aggregate' near the ceil periphery (Fig. 6). Such lesions usually affect the adjacent ceil equally. No other ceil type appeared to be affected by vinyl chloride exposure in this study. CMA 003607 mi ,vs'3#a J. T. DU ET AL. FIGURE 4. Portion of a hepatocyte from a control rat. A good complement of mitochondria and rough endoplatmic reticulum Is shown (XI 2,000}. FIGURE 5. Dilation of RER (arrows) shown in hepatocyto 137 h after exposure to 13,000 ppm vinyl chloride during 2-3 wk (X 23,000). FIGURE 6. Subplasmalemmal lesions () in hepatocytes 137 h after exposure to 15,000 ppm vinyl chloride during 2-3 wk. The' lesions are present in two adjacent cells and the dear spaces seem to coalesce (X 12,000), ' DISCUSSION Decreased glucose-6-phosphatase activity and increased glucose6-phosphate dehydrogenase and glutathione reductase activity In rat liver after exposure to vinyl chloride (as in our studies) may be the biochemical alterations indicative of early liver injury, adaptation to increased, detoxi fication activity, or preparation for increased nucleic acid synthesis. These differences were observed before any abnormalities were detectable from conventional liver function tests, such as serum aminotransferases, or from changes in other subccllular organelle markers, such as mitochondrial cytochrome oxidase activity. In primary hepatocellular cancer (hepatomas), the activity of key enzymes for gluconeogcnesis (glucose-6-phosphatasc, etc.), decreased with increased rate of tumor growth (Weber, 1974; Weber and Convery, 1966; Weber and Lea, 1967). In contrast, two enzymes for the pentose phosphate pathway, glucosc-6*phosphate dehydrogenase (Sclmcci and CMA 003608 VINYL CHLORIDE EFFECTS ON RAT LIVER 1129 Weber, 1976; Weber and Morris, 1963) and transaldolasc (Heinrich ct al., 1974) , increased in all hepatomas. Further, the activity of gluclosc-6phosphatase decreased before and during the development of hepatomas when carcinogens such as nitrosamine and dimethylaminoazobenzene were fed to rats (Isok and Teras, 1973; Weber and Cantero, 1955). Whether our similar findings in this study are indicators of eventual cancer development (angiosarcoma) is not yet known. We used a shorter exposure period in order to identify the biochemical changes that would best reflect the morphological and cellular changes anticipated on the basis of previous human and animal studies. No attempt was made to determine what effects these shorter exposure periods would induce with long-term observation. This is now in progress. However, Maltoni and Lefemine (1975) showed that of 69 Sprague-Dawley rats exposed to 10,000 ppm vinyl chloride, 16 (26%) developed Zymbal gland carcinomas after 50 wk, 5 (8%) developed nephroblastomas after 59 wk, and 9 (15%) developed angiosarcomas after 64 wk. The vinyl chloride-exposed groups lost weight (4-13%) whereas the control group gained weight (2-10%) during the entire experimental period. However, the differences in the three enzymes cannot be accounted for by lack of dietary food intake in the experimental group, because fasting increases glucose-6-phosphatase (Ashmore et al., 1954) and decreases glucose-6-phosphate dehydrogenase (Winberry and Holtcn, 1977), and we found decreased glutathione reductase and reduced glutathione content in fasted animals (unpublished data). Rats exposed to a high dose (5%) of vinyl chloride (Reynolds et al., 1975) had decreased mixed function oxidase activity. An in vitro study (Ivanetich ct al., 1977) showed that the metabolites of vinyl chloride from the microsomal enzyme system decreased the levels of cytochrome P-450. In our studies; the cytochrome P-450 content and the activity of mixed function oxidase and NADPH-cytochrome c reductase in rats repeatedly exposed to 1-2% vinyl chloride did not show any such changes. Sub sequent experiments in which rats were exposed to 28,000 ppm vinyl chloride for 70-210 h in 2-6 wk showed a significant decrease of cytochrome P-450 concentration (in preparation). Differences in dose level or experimental design may account for these variations. Drew et al. (1975) found that the activity of mixed function oxidase depended on the time interval between cessation of exposure to vinyl chloride and sacrifice of animals. The main detoxification route for vinyl chloride metabolism is thought to be conjugation with glutathione (Johnson, 1967); sulfur-containing urinary metabolites of radioactively labeled vinyl chloride have been found in rats' urine (Green and Hathway, 1975, 1977; Watanabe et al., 1976b, 1976c). We found that under the present experimental conditions the non- protein sulfhydryl content of liver (glutathione and/or cysteine) tended to 003609 ViN VL j E.-Vicrj > >-i - c.-. WHber, 1975; W'ebsr arid Morris, 1963) aod transaldolasc (He;nr;ch et al,, 1974), increased in all hepatomas. Further, the activity of g!uclosc-5- phosphatase decreased before and during the development of hepatomas when carcinogens such as nitrosamme and dimcthylaminoacobeniene were fed to rats (Isok and Teras, 1973; Weber and Cantero, 1955). Whether our similar findings in this study are indicators of eventual cancer development (angiosarcoma) is not yet known. We used a shorter exposure period in order to identify the biochemical changes that would best reflect the morphological and cellular changes anticipated on the basis of previous human and animal studies. No attempt was made to determine what thrss s^ertsr ^xpesurs p?r*cs wcj!j v/j*^ observation. This is now in progress. However, .Maitor.i and Lsferuu.s (1975) showed that of 69 Sprague-Caw lay rats exposed to 10,009 ppm vinyl chloride, 16 (26%) developed Zymbai gland carcinomas after 50 wk, 5 (S%) developed nephroblastomas after 59 wk, and 9 (15%) developed angiosarcomas after 64 wk. The vinyl chloride-exposed groups lost weight (4-13%) whereas the control group gained weight (2-10%) during the entire experimental period. However, the differences in the three enzymes cannot be accounted for by lack of dietary food intake in the experimental group, because fasting increases glucose-6-phosphatase (Ashmore et al., 1954) and decreases glucose*6-phosphate dehydrogenase (Winberry and Hoi ten, 1977), and we found decreased glutathione reductase and reduced glutathione content in fasted animals (unpublished data). Rats exposed to a high dose (5%) of vinyl chloride (Reynolds et al., 1975) had decreased mixed function oxidase activity. An in vitro study (Ivanetich et al., 1977) showed that the metabolites of vinyl chloride from the microsomal enzyme system decreased the levels of cytochrome P-450. In our studies; the cytochrome P-450 content and the activity of mixed function oxidase and NAOPH-cytochrome c reductase in rats repeatedly exposed to 1-296 vinyl chloride did not show any such changes. Sub sequent experiments in which rats were exposed to 28,000 ppm vinyl chloride for 70-210 h in 2-6 wk showed a significant decrease of cytochrome P-450 concentration (in preparation). Differences in dose level or experimental design may account for these variations. Drew et al. (1975) found that the activity of mixed function oxidase depended on the time interval between cessation of exposure to vinyl chloride and sacrifice of animals. The main detoxification route for vinyl chloride metabolism is thought to be conjugation with glutathione (Johnson, 1967); sulfur-containing urinary metabolites of radioactiveiy labeled vinyl chloride have been found in rats' urine (Green and Hath way, 1975, 1977; Watanabe et al., 1976b, 1976c). We found that under the present experimental conditions- the non- protein sulfhydryl content of liver (glutathione and/or cysteine) tended to CMA 003610 1130 J. T. OU ET AL. increase in rats repeatedly exposed to vinyl chloride (Tabic 2). Although the difference was not significant in this experiment, subsequent exposure of rats to 28,000 ppm for 70, 140, and 210 h in 2, 4, and 6 wk led to a significant elevation of the nonprotein sulfhydryl content in liver (Du and Tamburro, 1978). Watanabe et ai. (1976a) and Johnson (1967) found a depletion of nonprotein sulfhydryl content in rats after only a single exposure to vinyl chloride or chloroethanol. Hefner et al. (1975) found that a single vinyl chloride exposure reduced the glutathione content by one-half, but the decrease became smaller and even unnoticeable after repeated exposure to vinyl chloride. Fiala et al. (1976) showed that the reduced glutathione content increased after prolonged exposure of rats to various chemical carcinogens. These results are thought to reflect the exposed animals' attempt to make more glutathione to meet the unusually strong demand for detoxification. Glutathione reductase was elevated in our system after rats were exposed to vinyl chloride for 42 h. This enzyme generates reduced glutathione from its oxidized form as a compensatory mechanism to maintain the level of glutathione, and the simultaneous elevation of glucose-6-phosphate dehydrogenase regenerates NADPH, which can be used as a cofactor for various synthetic pathways including nucleic acid synthesis. Elevated glutathione reductase activity was found in rats with primary hepatocellular cancer induced by diethylnitrosamine (Pinto and Bartley, 1973). This consistent increase in glutathione reductase activity after exposure to vinyl chloride suggests that it may be one of the earliest biochemical manifestations of exposure and injury. These metabolic changes could also play a role in the early changes observed by light and electron microscopy. Hepatic lesions such as dilation of smooth endoplasmic reticulum and loss of microvilli were reported in mice as early as 1 mo after vinyl chloride exposure (Schaffner ct al., 1976). In contrast, we observed dilation of rough endoplasmic reticulum and patchy, lesions near the plasmalemma. The dilated rough endoplasmic reticulum could be related to the increased enzyme synthesis (i.e, glucose6-phosphate dehydrogenase and glutathione reductase) induced by vinyl chloride. The absence of smooth endoplasmic reticulum proliferation in our animals suggests a comparatively milder effect in our short-term study. In a subsequent study where long-term effects of vinyl chloride exposure was assessed, dilation of smooth endoplasmic reticulum in hepatocytcs was observed (in preparation). The nature of the patchy lesions found near the plasmalemma remains to be established. The location of the lesion, however, suggests that some toxic agents may be entering or exiting the hepatocytcs in these sites. We believe these early enzymatic changes reflect adaptation of the liver cell to early mild injury. The alterations in gfuconeogenesis and in the pentose phosphate shunt appear to indicate the liver's adaptation while undergoing repetitive and prolonged exposure to the mildly toxic chemical CMA 003611 VINYL CHLORIDE EFFECTS ON RAT LIVER 1131 vinyl chloride, which, when inadequately metabolized, produces inter* mediates that lead to the formation of cancer. REFERENCES Ashmore, )., Hastings, A. B., and Nesbctt, F. B. 1934. The effect of diabetes and fasting on liver glucose-6-phosphatase. Proc. Natl, Acad. Sd. U.S.A. 40:673-678. Bolt, H, M.( Kappus, H., Buchter, A., and Bolt. W, 1975. Metabolism of vinyl chloride. Lanctt 1:1423. Carlberg, I. and Mannervik, B. 1975. Purification and characterization of the flavoenzyme glutathione reductase from rat liver. /. Biol. Cham. 250:5475-5480. Creech, J. L. and Johnson, M. N. 1974. Angiosarcoma of liver In the manufacture of polyvinyl chloride. A Occup. Mid. 16:150-151. Oegroot, L. I. and Dunn, A. 0. 1964. Electron-transport enzymes of calf thyroid. Blochim. Biophys. Acta 92:205-222. Drew, R, T., Harper, C, Gupta, B. N., and Talley, F. A. 1975. Effects of vinyl chloride exposures to rats pretreated with phenobarbital. Environ. Health Ptnpcct. 11:235-242. Du, J. T. and Tamburro, C H. 1976. Decreased glucose-6-phosphatase activity In liver In vinyl chloride exposed rats. Fid. Proc. 35:329. Du, J. T. and Tamburro, C H. 1978. Elevated glutathione content, glutathione-5-transferase and glutathione reductase in liver of rau exposed to vinyl chloride. Fid. Proc. 37:1545. Flala, S., Mohindru, A., Kettering, W. G., Fiala. A. E., and Morris, H. P. 1976. Glutathione and gamma glutamyl transpeptidase in rat liver during chemical carcinogenesis. A Natl. Conor Inst. 57:591-598. Green, T. and Hathway, D. E. 1975. The biological fate in rats of vinyl chloride In relation to Its oncogenicity. Chtm, Biol, tntiraet, 11:545-562. Green, T. and Hathway, D. E. 1977. The chemistry and biogenesis of 5-containing metabolites of vinyl chloride In rats. Chtm, Biol. Inttroct. 17:137-150. Harper, A. . 1965. Glucose-6-phosphause. In Methods of Enzymatic Analysis, ed. H. U. Bergmeyer, pp. 788-792. New York: Academic. Hefner, R. E., Watanabe, P. G., and Gchring. P. J. 1975. Preliminary studies of the fate of Inhaled vinyl chloride monomer in rats. Ann. N.Y. Acad. 5c/. 246:135-148. Heinrich, P. C, Morris, H. P., and Weber, G. 1974. Increased phosphoribosylpyrophoaphate synthetase activity in rapidly growing hepatomas. FEBS Lift. 42:145-148. Hoitzman, ). L, Gram, T. E., Gigon, P. L., and Gillette, |. R. 1968: The distribution of the components of mixed-function oxidase between the rough and the smooth endoplasmic reticulum of liver cells. Blochim. A 110:407-412. Isok, M. . and Tens, L. E. 1973, Glucose-6-phosphatase activity fn liver carcinogenesis and in transplantable hepatoma In mice. Vopr. Mid. Kh/m. 19:568-570. Ivanetich, K. M., Aronson, I., and Kata, I. D. 1977. The interaction of vinyl chloride with nt hepatic microsomal cytochrome P-450 In vitro. Blochim. Biophys. Bis. Common. 74:1411-1418. jaeger, R. J., Reynolds, E. S., Connolly, R. B., Moslen, M. T., Szabo, A., and Murphy, S. M. 1974. Acute hepatic injury by vinyl chloride in rats pretreated with phenobarbltal. Nature (Land.) 252:724-726. Johnson, M. K. 1967. Metabolism of chloroethanol In the rat. Blochim. Pharmacol. 16:185-199. ^ Lohr, G. W. and Waller, H. D. 1965. Glucose-6-phosphate dehydrogenase. In Methods of Enzymatic Analysis, ed, H. U. Bergmeyer, pp. 744-751. New York: Academic. Lowry, 0. H., Rosenbrough, N. J., Farr, A. L,, and Randall, R. J. 1951. Protein measurement with the Foltn phenol reagent. A Biot. Chtm. 193:265-275. Maltonl, C. and Lefemlne, G. 1975. Carcinogenicity bioassays of vinyl chloride: Current results. Ann. N.Y. Acad. Sd. 246:195-218. `/IMVL CHLD'.I vinyl chloride, which, when inadequately metabolized, produces inter mediates that lead to the formation of cancer. REFERENCES Ashmore, J., Hastings, A. 8., ind Neibett, F. B, 19j4, The effect of diabetes and fasiing on liver |)ueoe-6*pf)OSpbataie. Proc. Noll. Acad. Sci. US.A. 40-673-578, Bolt, H. M.f Kappus, H., Buchtsf, A., and Bolt, W. 197S. Metabolism of vinyl chloride, Lancet 1:1425. Carlbctj. I. >i-d 3. 1975. ?u- glulathloni -sductas; o~i nt liver. J. B:p, Oeeeh, J. L. and Johnson. M, N. 1974. Angiosarcoma of liver in the manufacture of polyvinyl chloride./. Occ-jp. Mid. 16:150-151. Degroot, L. J. and Dunn, A. 0. 1964. Electron-transport enrymes of calf thyroid. Biochim. Biophyt. Acta 92:205-232. Drew, R. T., Harper, C, Gupta, 8. N., and Talley, F. A. 1975. Effects of vinyl chloride exposures to rats pretreated with phenobarbital. Environ. Htilth Ptripcct. 11:235-242. Du, J. T. and Tamburro, C. H. 1 973. Decreased glucose-6-phosphaiase activity in liver In vinyl chloride exposed rats. Ftd. Proc. 35:329. Du, J. T. and Tamburro, C. H. 1973. Elevated glutathione content, glutathione-5-transftraie end glutathione reductase in liver of rats exposed to vinyl chloride. Ftd. Proc, 37:1545. Flala, S., Mohlndru, A., Kettering, \V. G., Fiala. A. E., and Morris, H. P. 1976. Glutathione and gamma glutamyl transpeptidase in rat liver during chemical carcinogenesis. /. Natl. Omar Intt. 57:591-598. Green, T. and Hathway, D. E. 1975. The biological fate in rats of vinyl chloride In relation to Its oncogenicity. Chtm. Biot. imtratt, 11:545-562. Green, T. and Hathway, D. E. 1977. The chemistry and biogenesis of 5-containlng metabolites of vinyl chloride in rats. Chtm. Blot. Inttroct. 17:137--150. Harper, A. E. 1965. Glucose-6-phosphatase. In Mtthods of Enzymatic Analytic, ed. H. U. Bcrgmeyer, pp. 788-792. New York: Academic. Hefner, R. ., Waunabe, P. G., and Gehring, P. I. 1975. Preliminary studies of tho fate of Inhaled vinyl chloride monomer In rats. Ann, N.Y. Acad. Sci. 246:135-148. Heinrich, P. C, Mortis, H. P., and Weber, G. 1974. Increased phosphoribosylpyrophotphate synthetase activity in rapidly growing hepatomas. FEBS Lett 42:145-1*8. Holtxman, ). l_. Gram, T. E., Ci;;on, P. L., and Gillette, ). R. 1968. The distribution of the components of mixed-function oxidase between the rough and the smooth endoplasmic reticulum of liver cells. Blocltttn. J. 110:407-412. Isofc, M. E. and Taras, U. E. 1973. Glucose-6-phosphatase activity to liver carcinogenesis and in transplantable hepatoma In mice. Vopr. Med. Khlm. 19:568-570. Imnetlch, K. M,, Aronson, I., and Kao, I. D. 1977. The interaction of vinyl chloride with rat hepatic microsomal cytochrome P-450 In vitro. tdochim. Biophyt. Bn. Common. 74:1411-1418. jaepr, (L Reynolds, . S., Connolly, R. B., Moslen, M. T., Stabo, A., and Murphy, S. M. 1974. Acuta hepatic injury by vinyl chloride In rats pretreated with phenobarbital. Nature (Land.) 232:724-726. Johnson, M. K. 1967. Metabolism of chloroethanol in the rat. Biochtm. Pharmacol, 16:185-199. Lohr, G. W. and Waller, H. 0. 1965. Glucose-6-phosphate dehydrogenase. In Mtthodt at Enzymatic Anolydt, ed. H. U. Bergmeyer, pp. 744-751. New York: Academic. Lowry, O. H,, Roscnbrough, N. )., Farr, A. L., and Randall, R. ). 1931. Protein measurement with the Foiln phenol reagent. /. Biol. Cham. 193:265-275. Maltoni, C. and Lefemine, G. 1975. Carcinogenicity bioasscya of vinyl chloride: Current results. Ann. N.Y. Acad. Sci. 246:195-218. CMA 003613 1132 J. T. DU T AL. Omurj, T, 2nd Sato, R. 1964. The carbon monoxide-binding pigment of liver mierosomes. I. Evidence for It* hemoprotein nature. /. Biol. Chem. 239:2370-2378. Pinto, R. E. and Bartley, W. 1973. Glutathione reductase and glutathione peroxidase activities in hepatomou* liver* of rat* treated with diethyl nitrosamine. FEBS Lett. 32:307-309. Reynolds, E. 5., Moslen, M. T., Szabo, S., and Jaeger, R. J. 1975. Vinyl chloride-induced deactivation of cytochrome P-450 and other component* of the liver mixed function oxidase system: An in vivo study. Res. Common. Chem. Pathol. Pharmacol. 12:685-694. Schaffner, F., Popper, H., and Selikoff, I. j. 1976. Initial features of vinyl chloride hepatic injury. Gastroenterology 71 :A35 (abstr.). Schneider, W. C. and Hogeboom, G. H. 1950. Intracellular distribution of enzymes. V. Further studies on the distribution of cytochrome c in rat liver homogenates. /. Biol. Chem. 183:123-12S. Sedlak, J, and Lindsay, R. H. 1968. Estimation of total protein bound and non-protein sulfhydryl groups in tissue with Ellman's reagent. Anal. Bioehem. 25:192-205. Selmeci, L, E. and Weber, G. 1976. Increased glucose-6-phosphate dehydrogenase concentration in hepatoma 3924A: Enzymic and immunological evidence. FEBS Lett. 61:63-67. Tamburro, C. M., Makk, L., and Popper, H. 1979. Early hepatic histological alterations among chemical (vinyl monomer) workers. Gastroenterology, in press. Viola, P. L., Bigotti, A., and Caputo, A. 1971. Oncogenic response of rat skin, lungs-, and bones to vinyl chloride. Cancer Res. 31:516-522. Watanabc, P. G.., Hefner, R. E., Jr., and Gehring, P. J. 1976a. Vinyl chloride-induced depression of hepatic non-protein sulfhydryl content and effects on bromosulphalein (BSP) clearance in rats. Toxicology 6:1-8. Watanabe, P. G., McGowan, G. R., and Gehring, P. M. 1976b. Fate of ,4C-vlnyl chloride after single oral administration in rats. Toxicol. Appl. Pharmacol. 36:339-352, Watanabe, P. G., McGowan, G. R,, Madrid, E. O., and Gehring, P. J. 1976c. Fate of "C-vfnyl chloride following inhalation exposure in rats. Toxicol. Appl. Pharmacol. 37:49-50. Weber, G. 1974. The molecular correlation concept. In The Molecular Biology of Cancer, ed. H. Busch, pp. 487-521. New York: Academic. Weber, G. and Cantero, A. 1955. Glucose-6-phosphatase activity In normal, precancerous, and neoplastic tissues. Cancer Res. 15:105-108. Weber, G. and Convery, H. J. H. 1966. Insulin: Inducer of glucose-6-phospltate dehydrogenase. Life Sci. 5:1139-1146. Weber, G. and Lea, M. A. 1967. The molecular correlation concept. In Methods In Cancer Research, ed, H. Busch, vol. 2, pp. 523-578. New York: Academic. Weber, G. and Morris, H. P. 1963. Comparative biochemistry of hepatomas. Ilf. Carbohydrate enzyme* in liver tumors of different growth rates. Cancer Res. 23:987-994. Wharton, O. C. and Tzagoloff, A. 1967. Cytochrome oxidase from beef heart mitochondria. In Methods In Entymology, eds. R. W. Estabrook and M: E. Pullman, vol. 10, pp. 245-250. New York: Academic. Wlnberry, L and Hoiten, O. 1977. Rat liver giucose-6-phosphai dehydrogenase. Dietary regulation of die rate of synthesis. /. Biol. Chem. 252:7796-7801. Receded March 30, 1979 Accepted Jury 23, 1979 CMA 003614 THE YALE JOURNAL OF BIOLOGY AND MEDICINE 31 (1978), 67-80 The Hepatic Role in Carcinogenesis and Its Early Detection--The Vinyl Chloride Model*2 CARLO H. TAMBURRO University of Louisville School of Medicine, Louisville, Kentucky Received October 17. 1977 The liver's role in vinyl chloride toxicity and carcinogenicity n providing a better understanding of the chemical carcinogenesis mechanism. A variety of both malignant and benign hepatic tumors has been demonstrated with prolonged exposure to vinyl chloride. The multi-system involvment of this carcinogen and toxin has provided a model for the study of chemical carcinogenesis common to both man and animal. Clinical studies have shown the usefulness of biochemical, radioisotopic, and radiological studies in the detection of toxic and carcinogenic lesions. Animal studies have demonstrated the biochemical metab olism by the liver of vinyl chloride-produced intermediates which are mutagenic in bacterial systems and may be the ultimate carcinogens. Hepatic subcellular enxyme studies prove preliminary evidence of cellular adaptation and increased detoxification. Disruption of this oxidization and detoxification balance may be the key to the malignant transformation of cells. A working hypothesis is presented which may explain the metabolism of vinyl chloride into mutagenic intermediates by the liver cell and the develop ment of malignant transformation by extra hepatic sinusoidal lining cells, lung cells, and brain tissue. INTRODUCTION Currently there is a growing concern that chemical compounds may be respon sible for most human cancer through environmental contact. Although 100,000 to 200,000 new chemicals are introduced into industry each year, little is known about their effects. These compounds are primarily synthetics and thus not natural to the environment. The view that industrial chemicals may be latent carcinogenic hazards has again been brought into sharp focus by the discovery ofvinyl chloride* induced angiosarcoma. Vinyl chloride (CHz = CH-C1 monochloroethyene, a gas) is the basic molecule or monomer ofpolyvinyl chloride and its co-polymers and one of the most important organic intermediates in the plastics industry. The resulting plastic resin, polyvinyl chloride, is used in innumerable consumer and industrial products, such as containers, wrapping film, electrical insulation, pipelines, credit cards, etc. Until recently (early 1970's) vinyl chloride was regarded as being rela tively non-toxic [1,2], Initially this opinion seemed to be supported by the facts that it was used transiently as an anaesthetic and had been used commercially in indus try for many years [3]. The direct information on the toxicity of vinyl chloride to man was obtained from experiments by research workers on themselves, from the evaluation of its suit- 67 `Rxtion* ofthis work were supported by a front from Manufacturing Chemists Association and by National Cancer Institutes Contract #NOl-CN-5J212. 'This article is the thirteenth in a series entitled, "Seminars on Liver Disease/1 that have been presented as part of the Training Program in Liver Disease at the Veterans Administration Hospital, West Haven, Connecticut- Dr. Harold O- Conn, Professor of Medicine, Yale University School of Medicine, and Director of the Training Piogram in Liver Disease, is guest editor. Please address reprint requests to: Carlo H. Tamburro, MD. Professor of Medicine. Chief. Digestive Diseases and Nutrition Division, Director, Vinyl Chloride Project. 511 South Floyd Street, P.O. Box 35200, Louisville. KY 40232 0044-0086/78/5101-0067 $01.40 Copyright 1978 by Th* Yale Journal of Biology and Medicine, Inc. All rights of reproduction in any form reserved. CMA 003615 68 CARLO H. TAMBURRO ability as an anaesthetic [4, 5], and from study of the cases of vinyl chloride poison ing contracted during industrial use [6, 7, 8, 9]. Sporadic studies with vinyl chloride polymerization workers demonstrated vary ing degrees of hepatic biochemical derangements, hematological abnormalities, and skin changes [9]. Acute short term exposures led to disturbances of the central nervous system, cardiac arrythmias, severe irritation of the mucosal membrane of the eyes and the respiratory tract, and in some cases--severe pulmonary edema with obstruction of the liver and kidneys. Chronic inhalation trials in animals clearly showed that vinyl chloride was toxic to the liver and kidneys as well as irrita ting to the mucosal membranes and the lungs. Microscopic examination of liver sections showed degeneration of the central lobules while the damage to the kidneys chiefly involved the tubule and the interstitial lining somewhat like that of carbon tetrachloride damage [10, 11, 12]. BACKGROUND The literature, however, contained virtually no information on damage to man after chronic exposure until Filatova et al. reported disturbances of the blood vessels and nerves in individuals exposed with 20-300 ppm of vinyl chloride on a continuous basis [13]. Cordier et al. [14] and Wilson et al. [15] were the first to re port the hitherto unrecognized disorder termed occupational acroosteoiysis (AOL) which included the symptoms oftenderness ofthe fingertips, gradual destruction of bony integrity of the fingers and a Raynaud's-like phenomena. Studies were initiated in animals to reproduce the acroosteoiysis. In 1971, Violi et al. [ 16] while exposing animals to 30,000 ppm to induce acroosteoiysis, accidentally discovered cancer. Maltoni et al. [17] while studying various levels of exposure demonstrated that angiosarcoma occured at 250 ppm; the Manufacturing Chemists Association's studies [18] demonstrated these liver cancers even at 50 ppm. At the same time. Dr. John Creech, at the B.F. Goodrich Chemical Company Plant in Louisville, Kentucky, discovered an hepatic angiosarcoma in one employee. Creech, recalling an earlier hepatic angiosarcoma at the plant, reviewed the medical histories of previous employees. Four additional angiosarcomas were found, further supporting the connection between vinyl chloride exposure and tumor development. Measures to control the levels of vinyl chloride exposure were then instituted following federal regulation. THE CHEMICAL Vinyl chloride's chemical structure is a double-bonded, 2-carbon halogenated hydrocarbon which has structural similarities to tri-chloroethylene, an inhalation anaesthetic. As previously noted, vinyl chloride was once considered as an anaesthetic but was discarded because it caused myocardial irritability. Some ofits important physical properties include a low boiling point, a high specific gravity, a low solubility in water, and a half-life in air which ranges from 3-20 hours. Knowledge of these physical properties may be important in determining how this agent produces a cumulative effect in the environment which ultimately leads to cancer formation. Vinyl chloride is both toxic and carcinogenic, as recognized by the wide variety of associated disorders which have been found among vinyl chloride polymeri zation workers and vinyl chloride-exposed animals. A yet incomplete list of these CMA 003616 VINYL CHLORIDE--RELATED CARCINOGENESIS TABLE I Occupational Vinyl Chlonde Exposure Associated Disorders 1. Thrombocytopenia 2. Reticulocytosis 3. Splenomegaly 4. Hepatic fibrosis 3. Scleroderma-like skin changes 6. Acro-osteolysis 7. Raynaud's phenomenon 8. Leukopenia 9. Hepatomegaly 10. Pulmonary functional impairment 11. Angiosarcoma 12. Cardiac arrhythmia 13. Nephroblastomas* 14. Zymbai gland carcinomas* 15. Large cell lung cancer 16. Brain cancer In rats only 69 associated disorders is shown in Table 1. Animal and epidemiological studies indicate the probability that cancer induction at other sites is also directly attributed to prolonged and excessive vinyl chloride exposure [19], In order to develop effective methods ofprevention, accurate knowledge of the pathogenesis of this environmental chemical in ultimately producing its most destructive effect--cancer--is needed. EXPERIMENTAL ANIMAL STUDIES--TUMOR FORMATION The carcinogenicity of vinyl chloride has been demonstrated in a variety of animals as well as in humans at exposure levels that vary from 50-30,000 ppm. As illustrated in Table 2, a variety of tumor types have been found in rats, mice, and hamsters. An extensive list ofbenign tumors have also been reported in mice, rats, and hamsters exposed to vinyl chloride. Maltoni's group has now demonstrated primary liver cell cancers in exposed newborn rats. Direct hepatocellular injury as well as pulmonary, mucosal and skin injuries have been shown in directly exposed animals. Pretreatment with many agents increases the toxicity ofvinyl chloride; they include phenobarbital, ethanol, polychlorinated biphenyls and pesticides such as hexachlorobenzene [20]. This relationship to vinyl chloride's ability to induce cancer is under study, particularly in view of the industrial environment which allows exposure to many chemicals to occur con currently. HUMAN STUDIES Epidemiological studies in the human strongly suggest that exposure beyond 10 years is associated with increased cancer mortality, mainly digestive system cancers, primarily hepatic [21,22]. There also appears to be a higher incidence of large cell carcinomas of the lung, brain glioblastoma multifonns, and lymphomas [23,24], although there is some disagreement as to the interpretation of this aspect of the epidemiological data. CMA 003617 '>'NVL CHLORIDE--RELATED C kRCISOOENESIS T*BLE I Ck-cupauorul Vim! Ci jnje Exposure Aiio.'i-it-J C\i^>rJcr> I, TVcmbocyiopema 2 Reiiculoeyiosis 3. Splenomegaly e. Hepauc fibrosu J, Sderodentia-hke skin changes 6. Acro-oateolycu 7. Raynaud's phenomenon 8. Leukopenia 9 Hepatomegaly 10. Pulmonary functional impairment 11. Angiosarcoma 12. Caidiac arrhythmia 13 Nephroblastomas* 14, to/ iuj4u -Jw^-tTurn^i * 13. Large ce.l lung cancer 16 Brain cancer *In rats only 69 associated disorders is shown in Table 1. Animal and epidemiological studies indicate the probability that cancer induction at other sites is also directly attributed to prolonged and excessive vinyl chloride exposure [19]. In order to develop effective methods of prevention, accurate knowledge of the pathogenesis of this environmental chemical in ultimately producing its most destructive effect--cancer--is needed. EXPERIMENTAL ANIMAL STUDIES--TUMOR FORMATION The carcinogenicity of vinyl chloride has been demonstrated in a variety of animals as well as in humans at exposure levels that vary from 50-30,000 ppm. As illustrated in Table 2, a variety of tumor types have been found in rats, mice, and hamsters. An extensive list of benign tumors have also been reported in mice, rats, and hamsters exposed to vinyl chloride. Maltoni's group has now demonstrated primary liver cell cancers in exposed newborn rats. Direct hepatocellular injury as well as pulmonary, mucosal and skin injuries have been shown in directly exposed animals. Pretreatment with many agents increases the toxicity of vinyl chloride; they include phenobarbital, ethanol, polychlorinated biphenyls and pesticides such as hexachlorobenzene [20]. This relationship to vinyl chloride's ability to induce cancer is under study, particularly in view of the industrial environment which allows exposure to many chemicals to occur con currently. HUMAN STUDIES Epidemiological studies in the human strongly suggest that exposure beyond 10 years is associated with increased cancer mortality, mainly digestive system cancers, primarily hepatic [21, 22], There also appears to be a higher incidence of large cell carcinomas of the lung, brain glioblastoma multifonns, and lymphomas [23,24], although there is some disagreement as to the interpretation of this aspect of the epidemiological data. CMA 003618 70 CARLO H. TAMBURRO TABLE 2 Carcinogenicity of Vinyl Chlonde (Exposure- 50-10,000 ppm) TumorTypc Species I Liver--angiosarcoma 2. Liver--hepatocellular carcinoma 3. Lung--adenocarcinoma 4. Lung--largo cell carcinoma 5. Mammary adenocarcinoma 6. Zymbal gland tumors 7. Nephroblastoma 8. Osteochondromas 9. Skin epitheliomas 10. Melanomas 11. Glioblastoma multiforme 12. Lymphoma Adult humans* rats, mice and hamsters Newborn rats Rats and mice Humans* Mice Rats Rats Rats Hamsters Hamsters Humans* Humans* *Strongly suggested epidemiologically The multisystem involvement of this carcinogenic and toxic chemical is further illustrated in man. Early physical findings of vinyl chloride-injury include hepato megaly, portal hypertension, possible mild systemic pulmonary hypertension, bilateral midzonal pleural thickening of the lung, and splenomegaly with and without increased portal pressure [25]. These anatomical and physiological findings most frequently occur in the absence of the traditional clinical biochemical derangement of the liver. Screening studies of vinyl chloride workers during the past two and a half years have clearly illustrated the irregular and often delayed appearance ofabnormalities in aspartate and alanine aminotransferases (SGOT and SGPT), alkaline phosphatase as well as other hepatocellular enzymes. Gamma glutamic transpeptidase (GGTP), believed to be a more sensitive indicator of hepatocellular injury, has proven to have too high a false-positive rate to warrant its use in the screens for hepatocellular injury. Sorbitol dehydrogenase (SDH) studies indicate that this enzyme, which is liver tissue specific, is too insensitive for primary screening but is useful for confirmatory testing. Anionic dye clearance studies (Indocyanine Green, ICG) have demonstrated the highest specificity and sensitivity of all the primary screening procedures tested when performed at the 5 mg/kg dose level. At the traditional 0.5 mg/kg level, it has the same effectiveness as the aminotransferases and alkaline phosphatase com bined. The frequency of abnormal ICG dye clearances increases with prolonged exposure to vinyl chloride as illustrated in Fig. 1 and correlates well with the cumulative exposure to vinyl chloride as well as the histological evidence of hepatocellular injury. These functional studies for detecting hepatocellular injury do not, however, identify the cause. Clinical studies have illustrated the usefulness ofradioisotopic liver-spleen scans as a primary screening procedure for anatomical lesions of the liver and spleen. This procedure has proven to be the single most reliable method oftumor detection. Sixteen of the 19 individuals with anatomical lesions were detected by liver-spleen scan. In contrast, only 32 of 950 normal individuals had scan abnormalities which further diagnostic studies proved incorrect. This method provides an 84% sensitiv ity and 97% specificity, with only a 3% false-positive rate. CM& 003619 VINYL CHLORIDE--RELATED CARCINOGENESIS 71 Diagnostic angiographic studies of these radioisotopic abnormalities in 80 individuals have demonstrated 3 major lesions. The first is peliosis hepatis, illustrated in Fig. 2. These lesions are usually numerous involving the entire liver, and have a diffuse stain throughout the nodules which persists into the late venous phase without central hypovascularity [26]. A second, similar lesion has been discovered in individuals with splenomegaly, and named lienai peliosis (Fig. 3). These splenic lesions demonstrate a shortened celiac artery to portal vein circulation time, normal portal vein diameters, and increased spleen size. This has been found only in individuals with long-term vinyl chloride exposure. The final lesion is that of angiosarcoma (Fig. 4). This tumor has characteristic angiographic features of central hypovascularity, midarterial puddling, and a prolonged peripheral tumor stain which continues up to 30-36 seconds after injection. These characteristic findings have allowed differentiation from other primary hepatocellular cancers, benign tumors and benign vascular lesions [26]. These angiographic lesions have been pathologically confirmed with the additional histological finding including peliosis hepatis, sinusoidal dilatation, and activated sinusoidal cells with increased deposits ofcollagen in the sinusoidal space ofDisse. Exploratory wedge biopsies have in addition demonstrated increased subcapsular fibrosis with subcapsular bile duct proliferation plus the often described portal fibrosis [27]. VINYL CHLORIDE METABOLISM AND CARCINOGENESIS Present biochemical knowledge indicates that vinyl chloride is most likely metabolized by the liver in a three step process [28], At concentrations less than 50 ppm, vinyl chloride is metabolized by the alcoholic dehydrogenase system into chloroacetaldehyde and monochloroacetic acid, CI-CH-CHj--CI-CHi-CHi-OH alcohol CI-CHj-CHO--ClCHa-COOH dehydrogenase An alternative pathway which appears to become operative at 220 ppm is oxidation by the peroxidase-catalase system. Cl-CHi-CHi-OH fe0* .CICHj-CHiOOH--------- CICHi-CHO catalase Tom M*. % FIG. 1. Frequency ol abnormal indocyanine green dye clearance unoq| vinyl chloride worteri uulizingO.J ms/Y* and J.0m*/k*do*ea- 8 i: 0 0 al CMA 003620 Diagnostic siuu;;s o: tnesa rauiooctcpic abnorma.i;iai '.i 20 individuals have demonstrated 3 major lesions. The first is peliosis hepaos, illustrated in Fig. 2. These lesions are usually numerous involving the entire liver, and have a diffuse stain throughout the nodules which persists into the late venous phase without central hypovascularity [26]. A second, similar lesion has been discovered in individuals with splenomegaly, and named lienal peliosis (Fig. 3). These splenic lesions demonstrate a shortened celiac artery to portal vein circulation time, normal portal vein diameters, and increased spleen size. This has been found only in individuals with long-term vinyl chloride exposure. The final lesion is that of angiosarcoma (Fig. 4). This tumor has characteristic angiographic features of central hypovascularity, midarterial puddling, and a prolonged peripheral tumor stain which continues up to 30-36 seconds after injection. These characteristic findings have allowed differentiation from oinsr primary hepatocellular cancers, benign tumors and benign vascular lesions [26]. These angiographic lesions have been pathologically confirmed with the additional histological finding including peliosis hepam, sinusoidal dilatation, and activated sinusoidal cells with increased deposits of collagen in the sinusoidal space of Disse. Exploratory w edge biopsies have in addition demonstrated increased subcapsular fibrosis with subcapsular bile duct proliferation plus the often described portal fibrosis [27]. VINYL CHLORIDE METABOLISM AND CARCINOGENESIS Present biochemical knowledge indicates that vinyl chloride is most likely metabolized by the liver in a three step process [28]. At concentrations less than 50 ppm, vinyl chloride is metabolized by the alcoholic dehydrogenase system into chloroacetaldehyde and monochloroacetic acid, Cl-CH-CHi--Cl-CHi-CHi-OH---- alcoho1 .Cl-CHi-CHO--ClCHj-COOH dehydrogenase An alternative pathway which appears to become operative at 220 ppm is oxidation by the peroxidase-catalase system. CI-CHi-CHi-OH - Hl* .CiCHa-CHiOOH-------- CJCHi-CHO catalase iwaa aa. V, I | BBS ---/w */* */ V* PIG. 1. Frequency of abnormal rtopmiioo |tmb dya etaanaca aamni viayl chlooda wwtari uoliza0.S ma/kf and3.0aa/kfSows. CMA 003621 72 CARLO H. TAM BURRO FIG. 2. Hepatic arteriogram: Venoui phase. Changes of peliasis hepatia are present throughout the left lobe. The multiple nodular stains represent pelioais hepatia lesions (arrows) ranging from 2-3 mm Id 2 cm in size. FIG. 3. Splenic arteriogram: Ve nous phase (15 seconds) There are 3-4 circular and oval stains (ar rows) in the superior and inferiorlateral portions of the spleen. Normat pancreatic stain occurs just above the splenic vein. CMA 003622 VINYL CHLORIDE--RELATED CARCINOGENESIS 73 FK3. 4. Hepatic vtcrioenun: Venous phase. At approximssaty 14-11 ssconds die peripteral stain is idcntiflMl (arrows). lasting through th entire phase. Scattered areas at puddling at* also present in and around the area of central hypovaacuiarity. In this case chloroacetaldehyde is again formed. At higher levels oxidation appears to be by the mixed function oxidase system, forming chloroethylene oxide which spontaneously rearranges to form chloroacetaldehyde which then can be further oxidized to form monochloroacetic acid. Cl-CH-CHi oxidase +CL-CH Cl-CHs-CHO ClCHr-COOH As illustrated in Table 3, vinyl chloride oxidation intermediates chloroethanol and chloroacetaldehyde, at low doses, are most likely detoxified via the glutathi one-cysteine conjugation system. This system, however, is saturable and at higher levels vinyl chloride is excreted via the lungs [28]. It appears that at higher vinyl chloride levels increased amounts of chloroethanol and chloroacetaldehyde are further oxidized to chloroacetic acid which is excreted in the urine. This is further supported by the absence ofchloroacetic acid in urines ofrats exposed to low, short term levels of vinyl chloride but found in urine of rats exposed to 5,000 ppm for an extended tune and reported in workers exposed to levels greater than 250 ppm for a prolonged time [29, 30]. Elmore et al., utilizing a modified Ames system and pure synthesized vinyl chloride intermediates, has demonstrated that vinyl chloride, chloroethanol, and CMA 003623 - fc---- . ^y - . * FIG. 4. Hepatic arteriogram: Venous phase. At approaimawly 14-15 seconds the peripheral stain is identified (arrows), lasting through the entire phase. Scattered areas of puddling are also present in and around the area of central hypovaacularity. In this case chioroacetaldehyde is again formed. At higher levels oxidation appears to be by the mixed function oxidase system, forming chloroethylene oxide which spontaneously rearranges to form chioroacetaldehyde which then can be further oxidized to form monochloroacetic acid. oxidase a-CH-CHj hCL-CH O-CHs-CHO--sCICHr-COOH As illustrated in Table 3, vinyl chloride oxidation intermediates chloroethanol and chioroacetaldehyde, at low doses, are most likely detoxified via the glutathi one-cysteine conjugation system. This system, however, is saturable and at higher levels vinyl chloride is excreted via the lungs [28]. It appears that at higher vinyl chloride levels increased amounts of chloroethanol and chioroacetaldehyde are fUrther oxidized to chloroacetic acid which is excreted in the urine. This is further supported by the absence ofchloroacetic acid in urines ofrats exposed to low, short term levels of vinyl chloride but found in urine of rats exposed to 5,000 ppm for an extended time and reported in workers exposed to levels greater than 250 ppm for a prolonged time [29, 30). Elmore et al., utilizing a modified Ames system and pure synthesized vinyl chloride intermediates, has demonstrated that vinyl chloride, chloroethanol, and CMA 003624 74 Cl CH - CHj (VC) Liver MFO Cl CH-CHi O (ChJorooxirane) CARLO H. TAMBURRO TABLE 3 Proposed Metabolic Fate of Vinyl Chloride Cl CHj CHj OH (Chloroethanol) I +GSH Cl CHiCHO---------(ChloroacetaJdehyde) -GS CHj CHO Cl CHi COOH (Chloroacetic acid) aCS CHi COOH 1 TtuodiflycoUc acid chloroacetic acid are not mutagenic in bacteriological systems (31], This may indicate that the vinyl chloride monomer is neither hepatotoxic nor carcinogenic until it has been metabolized to its intermediate forms by the liver and/or other tissues. Alternatively, chloroethanol--the most transportable of the vinyl chloride metabolites--may be transferred or diffused to adjacent cells, such as the sinusoi dal lining cells, where it could be converted to chloroacetaldehyde but less likely to be detoxified or further oxidized. Since vinyl chloride appears to bind the serum albumin, it may, itself be transported to and oxidized by extra-hepatic cells which are unable to fully oxidize or completely detoxify its metabolites, and thereby lead to molecular DNA injury and cancer formation at distant tissue sites. These quandaries led to further study of the hepatochemical changes in rats undergoing progressively increased exposure to vinyl chloride. Subcellular en zymes and metabolites were studied in animals exposed to from 10-20,000 ppm vinyl chloride, ranging from 14-137 hours. Microsomal enzymes, including P-430, NADPH cytochrome c reductase, and mixed function oxidases were studied. Determinations of cytochrome c oxidase as the mitochondrial, tritiated-leucine incorporation as the protein synthesis and glucose-6-phosphatase as the carbo hydrate metabolism markers were also done. Glutathione and glutathione reduc tase as oxidative and detoxification markers were determined in addition to the conventional clinical biochemical studies which included the aspartate (SGOT) and alanine aminotransferase (SGPT), alkaline phosphatase, bilirubin, lactic acid dehydrogenase (LDH), total protein, albumin, cholesterol, and triglycerides. During the entire 137 hours of exposure there were no significant changes in the mitochondrial and the microsomal enzymes, the tritiated-leucine incorporation, or in the glutathione content. There was however, after 71 hours, a rise in the gluta thione reductase and a concomitant fall in giucose-6-phosphate. This occurred without any histologically discernible changes in'the hepatocytes by light micro scopy nor any significant changes in the conventional clinical biochemical studies. CMA 003625 VINYL CHLORIDE--RELATED CARCINOGENESIS 75 The discovery of a decreased glucose-6-phosphatase after "simulated" chronic exposure led to the study of enzymes in the pentose phosphate shunt pathway. Weber and Lea [32] had found similar changes for primary hepatocellular neo plasms, demonstrating that in a rapidly developing primary hepatocellular tumor, there is decreased gluconeogenesis with a reduction in the glucose-6-phosphatase, followed by an increase in glucose-6-phosphate dehydrogenase and transaldolase. These biochemical changes were also followed by an increase in purine biosynthe sis (increased phosphoribosylpyrophosphate aminotransferase (PRPP) and in creases in the production of ATP and GTP leading to increased nucleic acid synthesis. Vinyl chloride-exposed animals showed no significant changes in the glucose-6- phosphatase dehydrogenase activity during the initial 84 hours of exposure. However, after 103 hours, there was significant increase in glucose-6-phosphatase dehydrogenase. Studies of PRPP, at least up to 137 hours, have as yet shown no significant changes. Studies are now underway using animals exposed to 130 to 250 hours to determine if the biochemistry in vinyl chloride injury is similar to that in primary hepatocellular tumors. This, however, does not explain why the hepatocyte, which is the primary cell for oxidizing and detoxifying vinyl chloride, is not the primary target for cancer trans formation. How do the hepatocytic biochemical changes, seen in the early phase of high vinyl chloride exposure, relate to the later morphological changes that occur in the adjacent sinusoidal cells? MORPHOLOGICAL FINDINGS Electron microscopic examination of liver sections of mice exposed from 1 to 6 months to 2,500-6.000 ppm vinyl chloride, for 5 hours/day, 5 days/week--a level known to induce angiosarcoma [33]---have demonstrated hepatocellular changes as early as one month. These changes included hypertrophy of the smooth endo plasmic reticulum (believed to reflect vinyl chloride metabolism) and, plasma membrane loss of microvilli with invaginations--possibly reflecting the movement of injurious metabolites across the membrane and out of the cell, allowing the metabolites to be picked up by the sinusoidal cells [34]. The gimmniHai cell reactions were multicellular. Increasing numbers and sizes of lipocytes were seen with little fibrosis. Macrophages were seen filled with phago somes, sometimes containing long needle-like crystals. Although there were many mononuclear cells present, the main abnormalities were seen in the endothelial lining cells. In the early stages they are larger and thicker--possibly swollen. Later, they became bulky and in places, multi-layered containing increased organelles, especially mitochondria and endoplasmic reticulum. Later disruptions in the sinu soidal walls seen were consistent with beginning peliosis hepatis. The lining cells, probably the precursors of angiosarcoma, often resembled fibroblasts. However, their endoplasmic reticulum did not contain any collagen components. These observations by Schaffner et at. [34] give support to the suggestion that metabolites of vinyl chloride produced in the hepatocytes may be transported through the plasma membrane and enter sinusoidal lining cells, eventually leading to angio sarcoma. Attempts at screening for vinyl chloride hepatic injury might be better aimed at the endothelial cells and the hepatic sinusoidal circulation rather than the hepatocytes. Our work in humans has identified similar findings. One major difference at CMA 003626 Web^r and Lsa [32] had found similar changes for primary hepatocellular neo plasms, demonstrating that in a rapidly developing primary hepatocellular tumor, there is decreased gluconeogenesis with a reduction in the glucose-6-phosphatase, followed by an increase in gluco'e-6-phcsphate dehydrogenase and transaldclase. These biocnemicaJ cnanges were also followed oy an increase in purine Diosynthe* sis (increased phosphoribosyipvrophosphate aminotransferase (FRPP) and in creases in the production of ATP and GTP leading to increased nucleic acid synthesis. Vinyl chloride-exposed animals showed no significant changes in the glucose-6phosphatase dehydrogenase activity during the initial 84 hours of exposure. However, after 103 hours, there was significant Increase in slucose-6-phosphatase significant changes. Studies are now underway using animals exposed to 130 to 250 hours to determine if the biochemistry in vinyl chloride injury is similar to that in primary hepatocellular tumors. This, however, does not explain why the hepatocyte, which is the primary cell for oxidizing and detoxifying vinyl chloride, is not the primary target for cancer trans formation. How do the hepatocytic biochemical changes, seen in the early phase of high vinyl chloride exposure, relate to the later morphological changes that occur in the adjacent sinusoidal cells? 1 MORPHOLOGICAL FINDINGS Electron microscopic examination of liver sections of mice exposed from 1 to 6 months to 2,500-6.000 ppm vinyl chloride, for 5 hours/day, 5 days/week--a level known to induce angiosarcoma [33]--have demonstrated hepatocellular changes as early as one month. These changes included hypertrophy of the smooth endo plasmic reticulum (believed to reflect vinyl chloride metabolism) and, plasma membrane loss of microvilli with invaginations--possibly reflecting the movement of injurious metabolites across the membrane and out of the cell, allowing the metabolites to be picked up by the sinusoidal cells [34], The sinusoidal cell reactions were multicellular. Increasing numbers and sizes of lipocytes were seen with little fibrosis. Macrophages were seen fitted with phago somes, sometimes containing long needle-like crystals. Although there were many mononuclear cells present, the main abnormalities were seen in the endothelial lining cells. In the early stages they are larger and thicker--possibly swollen. Later, they became bulky and in places, multi-layered containing increased organelles, especially mitochondria and endoplasmic reticulum. Later disruptions in the sinu soidal walls seen were consistent with beginning pdiosis hepatis. The lining ceils, probably the precursors of angiosarcoma, often resembled fibroblasts. However, their endoplasmic reticulum did not contain any collagen components. These observations by Schaffher et al. [34] give support to the suggestion that metabolites of vinyl chloride produced in the hepatocytes may be transported through the plasma membrane and enter sinusoidal lining cells, eventually leading to angio sarcoma. Attempts at screening for vinyl chloride hepatic injury might be better aimed at the endothelial ceils and the hepatic sinusoidal circulation rather than the hepatocytes. Our work in humans has identified similar findings. One major difference at CMA 003627 76 CARLO H. TAMBURRO present is an increased collagen deposition, characteristic of human vinyl chloride injury and likely species specific. Light microscopic studies utilizing special stains on hepatic tissue from individuals with extensive exposure to vinyl chloride but without clinical biochemical hepatic abnormalities have shown distinctive midzonal increased deposition of collagen in the space of Disse [35], Routine light microscopic studies using hematoxylin and eosin failed to easily demonstrate this midzonal increased collagen. The increased deposition along the hepatic cell surface is associated with larger sinusoidal space and activation of the sinusoidal lining cells illustrated by increased nuclear size and cytoplasmic content. The in creased collagen deposition, when studied electron microscopically, demonstrates compression of the hepatocytes by the collagen bundles which initially give the ap pearance of m/ra-hepatoceliular collagen bundles as illustrated in Fig. 5. The strands of collagen appear to compress the hepatocytes causing cords of hepato cytes to be broken and to coalesce with adjoining sinusoids eventually leading to peliosis hepatis-like lesions. These observations led us to the study ofthe proteroglycan role in collagen form ation in vinyl chloride-exposed workers. It had been suggested in the literature that glycosaminoglycans in blood and/or urine might be useful as means ofearly cancer detection since a number of studies had demonstrated the production of sulfated glycosaminoglycans with malignant states. Pathologists have often used this feature as a diagnostic aid in characterizing malignant vascular tumors of the skin PIC. S. Election microscopy showing collagen (CB) bundles (arrows) invaginating into the hepatocyte. giving the appearance of inter-hepatocytic collagen. N nucleus; S - sinusoidal space; IM - invagination into the cell membrane (imaU arrows). CMA 003628 VINYL CHLORIDE--RELATED CARCINOGENESIS 77 (36). Others have noted a strong positive Alcian blue glycosaminoglycan staining reaction in human angiosarcoma tissue [37], This suggested that quantitative and qualitative determinations of glycosaminoglycan production in individuals with neoplasm, either by serum or urine, might be used to identify those at high risk or as an early indicator of neoplastic formation. The feasibility of glycosaminoglycan "spot test'' for vinyl chloride production workers made this an attractive possibil ity for mass screening. Urinary glycosaminoglycans, measured as uronic acid, were studied in individu als with alcoholic cirrhosis, viral hepatitis, secondary liver metastasis, hepatic angiosarcoma and normal controls. The percentage oftotal glycosaminoglycans that was dialyzable and the percent age ofunfractionated total that appeared in the hyaluronic acid, chondroitin sulfate, or in the heparin fractions was similar for all groups. However the distribution of positive fractions varied with the different groups studied. Seven of the nine vinyl chloride-exposed individuals, other than those with angiosarcoma, had glycosaminoglycan positive chondroitin sulfate fractions with negative hyaluronic acid and heparin fractions whereas this occurred in only 3 of the 32 urines from other hepatic diseases [38]. In addition, study of the total tissue glycosaminoglycan levels in angiosarcoma tumors and fibrotic tissue adjacent to the tumor demonstrated that tumor tissue itself had higher levels of hyaluronic acid and heparin fractions as compared to the non-tumor adjacent tissue which had higher levels ofchondroitin sulfate fractions. A similar relationship was found in cirrhotic liver tissue and normal controls. This data conforms to reports by others that both hepatic connective tissue disorder [39,40,41,42,43] and hepatic cancer [44] result in increased hepatic glycosaminoglycan levels. It may be significant that the angiosarcoma patient has half the urinary glycosaminoglycan excretion of patients with liver metastasis and that analysts ofangiosarcoma tumor tissue exhibits halfthe glycosaminoglycan content reported by Kojima et al. [44] for hepatocellular carcinoma. The increases in liver and urinary glycosaminoglycans may well reflect the importance of these substan ces in the process of fibrogenesis and tumor growth. Although no significant differences were found in total glycosaminoglycans ofvinyl chloride-exposed indi viduals with associated liver injury, there was a significant difference in the excre tion patterns of these individuals. Seventy-eight percent of them had positive chondroitin sulfate fractions in contrast to only nine percent of the non-exposed liver injury cases. Thus the change in the glycosaminoglycan excretion pattern in individuals with pre-cancerous injuries may be of significant prognostic and diagnostic importance [43], The urinary glycosaminoglycan excretion patterns in an angiosarcoma patient 3 nwnths to 2 weeks prior to death demonstrated an increase in the urinary chondroi tin sulfate fraction with a change in its composition as the disease progressed. During tHU time, the chondroitin sulfate composition showed acontinuous increase in the ratio of the 1.25 M NaCl to the 1.5 M NaCl fractions. This was due to an increase in the 1.25 Meluate and a decrease in the 1.5 Meluate fraction and was 2.3 times greater than the controls. In the most advanced stage ofthe angiosarcoma the ratio increased to 13.7 times greater [46]. These very preliminary studies would suggest that alterations in the ratios of these fractions* compositions may be useful in evaluating the severity and subsequent progression ofdisease. Early lesions may produce small changes in the ratio which would become more pronounced as the disease worsened. The CMA 003629 rr5]. Of.h?rs have ac:ed a s r -s n r aijco'rjT'.inoalycan staining qualitative determinations of glycosaminoglycan production in individuals with neoplasm, either by serum or urine, might be used to identify those at high risk or as an early indicator of neoplastic formation. The feasibility of glycosaminoglycan "spot test" for vinyl chloride production workers made this an attractive possibility for mass screening. Urinary glycosaminoglycans, measured as uronic acid, were studied in individu als with alcoholic cirrhosis, viral hepatitis, secondary liver metastasis, hepatic angiosarcoma and normal controls. The percentage oftotal glycosaminoglycans that was dialyzable and the percent age ofunfractionated total that appeared in the hyaluronic acid, chondroitin sulfate, or in the heparin fractions was similar for all groups. However the distribution of positive fractions varied with the different groups studied. Seven of the nine vinyl chloride-exposed individuals, other than those with angiosarcoma, had glycosaminoglycan positive chondroitin sulfate fractions with negative hyaluronic acid and heparin fractions whereas this occurred in only 3 of the 32 urines from other hepatic diseases [38]. In addition, study of the total tissue glycosaminoglycan levels in angiosarcoma tumors and fibrotic tissue adjacent to the tumor demonstrated that tumor tissue itself had higher levels of hyaluronic acid and heparin fractions as compared to the non-tumor adjacent tissue which had higher levels ofchondroitin sulfate fractions. A similar relationship was found in cirrhotic liver tissue and normal controls. This data conforms to reports by others that both hepatic connective tissue disorder [39,40,41,42,43] and hepatic cancer [44] result in increased hepatic glycosamino glycan levels. It may be significant that the angiosarcoma patient has half the urinary glycosaminoglycan excretion of patients with liver metastasis and that analysis of angiosarcoma tumor tissue exhibits halfthe glycosaminoglycan content repotted by Kojima et al. [44] for hepatocellular carcinoma. The increases in liver and urinary glycosaminoglycans may well reflect the importance of these substan ces in the process of fibrogenesis and tumor growth. Although no significant differences were found in total glycosaminoglycans ofvinyl chloride-exposed indi viduals with associated liver injury, there was a significant difference in the excre tion patterns of these individuals. Seventy-eight percent of them had positive chondroitin sulfate fractions in contrast to only nine percent of the noo-exposed liver injury cases. Thus the change in the glycosaminoglycan excretion pattern in individuals with pre-cancerous injuries may be of significant prognostic and diagnostic importance [45], The urinary glycosaminoglycan excretion patterns in an angiosarcoma patient 3 months to 2 weeks prior to death demonstrated an increase in the urinary chondroitin mlht* fraction with a change in its composition as the disease progressed. During this time, the chondroitin sulfate composition showed acontinuous increase in the ratio of the 1.23 M NaCl to the 1.3 M NaCl fractions. This was due to an increase in the 1.23 Meluate and a decrease in the 1.3 Meluate fraction and was 2.3 times greater than the controls. In the most advanced stage ofthe angiosarcoma the ratio increased to 13.7 times greater [46]. These very preliminary studies would suggest that alterations in the ratios of fractions' compositions may be useful in evaluating the severity and subsequent progression ofdisease. Early lesions may produce small changes in the ratio which would become more pronounced as the disease worsened. The CMA 003630 78 CARLO H. TAMBURRO determination of this ratio might also be useful in identifying those individuals with significant injury or continuing progression of disease despite changes in the environment. Finally therapeutic measures for intervention might be better evalu ated for their effectiveness in arresting cancer development as reflected by the glycosaminoglycan changes which in turn reflect changes in collagen formation. SUMMARY AND HYPOTHESIS Vinyl chloride appears to enter the body through the respiratory tract, the skin, or by swallowing; it is absorbed and transported to the liver by the systemic and portal circulatory systems. In the liver, the primary site of metabolism, it is oxidized by various enzyme systems: alcohol dehydrogenase at lower exposure levels; the peroxidase-catalase system at intermediate levels; and mixed function oxidases at higher levels. At the higher levels, the mixed function oxidase system transforms vinyl chloride to chlorooxiranes which are then spontaneously trans formed into choroethanol and chloroacetaldehyde. These two intermediate metabolites, chloroethanol and chloroacetaldehyde, are detoxified by conjugation with glutathione and cysteine-SH groups and are excreted in the urine. At even higher doses increasing amounts ofthe chloroacetal dehyde are further oxidized to chloroacetic acid and excreted as an end product in the urine. However, when chloroacetaldehyde and/or the chlorooxiranes exceed the detoxification threshold of the hepatocyte, this leads to hepatocellular toxicity and/or stimulation ofthe sinusoidal cells. This acute event in turn acts as a stimulat ing mechanism for increased collagen deposition in the space of Disse and sinusoidal areas as shown by electron and light microscopy studies. The increase in the collagen depostion in sinusoidal spaces then leads to disruption of hepatic cell surface function, disruption of the hepatic cords, coalition ofthe sinusoidal spaces and eventual peliosis hepatis. These lesions alternately lead to sufficient vascular dysfunction to add further to the biochemical manifestation of hepatic cellular injury. Since it is highly unlikely that the unstable chlorooxiranes are able to be trans ported to adjacent cells and that the chloroacetaldehyde would most likely be conjugated or detoxified within the hepatocyte, an intermediate form, such as chloroethanol, which is transportable from the hepatocyte, may then move on to the adjacent sinusoidal lining ceils, or possibly even further to other extrahepatic tissue. At these extrahepatic sites, an intermediate, such as chloroethanol, may then be converted to chloroacetaldehyde. The extrahepatic tissue sites are most likely unable to further convert the chloroacetaldehyde to chloroacetic acid nor to detoxify it sufficiently, if at all, by their own detoxification systems. This would allow a longer contact period with the cell's DNA. In addition many of the extra hepatic cells normally are regenerating at fester rates than hepatocytes, thus increasing the possibility of DNA derangement and ultimate carcinogenesis. Alternatively, vinyl chloride itself may be taken up by extrahepatic tissue, oxidized but incompletely detoxified, allowing the cell itselfto become susceptible to direct DNA injury. In this or similar manner, chemical metabolites may induce injury to the DNA in rapidly replicating cells at sites beyond the liver, thus accounting for other cancers developing with vinyl chloride. The recent work by Maltoni's group, showing that exposure of newborn rats to the same dose of vinyl chloride as adult rats, results in primary hepatocellular carcinoma (40-45%) rather than in angiosarcoma (8-12%), lends further support to VINYL CHLORIDE--RELATED CARCINOGENESIS 79 the concept that the hepatocyte's ability to resist cancer transformation is dependent upon its ability to detoxify the mutagenic/carcinogenic metabolite of vinyl chloride. This review of our present knowledge of vinyl chloride injury and cancer formation in man is, at best, a very rough hypothetical outline. With continued investigation and study it will allow us to more accurately and completely fill in the missing pieces ofthis fascinating puzzle, thus leading us to a better understanding of the pathogenesis of chemically induced cancer in the biologically complex human system. REFERENCES 1. Elkin* HB: The Chemistry of Industrial Toxicology. Second Edition. New York, New York, John Wiley A Son* Inc. 1950 2. Oettel H: In Ullmann'S Enzyfclopsdie der Techmchen Chemie Third Edition, 5:4*9. Munchen-Bcriin, Urban and Schwarzenbcrf, 1954. 3. Kunststoffkomission dcs Bundesgesundhertsautes (Piastres Committee of the Federal German Ministry of Health), Bundcsgenundhcrlsblatt, 8:369, 1975 4. von Ottingcn WF: The Halogenated Aliphatic. Olefinic. Cyclic, Aromatic, and Aliphatic-Aromatic Hydrocarbons Including the Halogenated Insecticides. Their Toxicity and Potential Danger*. (PHS Publication No. 414). Wash ington. D.C., Government Printing Office, 1955 5. Osier RH, Can CT, Krantz JC, et al: Anesthesia XXVII. Narcosis with vinyl chloride. 8:359-61,1947 6. Schottek W: The toxicity of vinyl chloride. Chcm Techn 21:706-711, 1969 7. Gauvain S: Vinyl chloride. Proe Roy Soc Med, 69:275-310. 1975 8. Braun P. Druckman E. Eds: Public-health rounds at the Harvard School of Public Health. Vinyl chloride: Can the worker be protected? New Eng 1 Med 294:653-657, 1976 9. Sclikoff U, Hammond EC. Eds: Toxicity of vinyl chlonde--polyvinyl chloride. Ann NY Acad Sci 246:1-337.1975 10. Irish DD: In Aliphatic halogenated hydrocarbons. Industrial Hygiene and Toxicology. Second Edition. Edited by FA Patty. New York. Intencience Publishers, 1963. 2:1241-1332 11. Mastromatteo E, Fisher M. Christie H, et ai: Acute inhalation toxicity of vinyl chloride to laboratory animal*. Amer Ind Hyg Assoc J 21:394-396. I960 12. Lester D. Greenberg LA, Adam* WR: Effects of single and repeated exposures of humans and rats to vinyl chloride, Amer Ind Hyg Assoc J 24:265-275. 1963 13. Filatova VS, Balakhonova 1, Gtonsberg ES: Hygienic characteristics of vinyl chloride production. Gig Tr Prof Zabol 2:6, 1958 14. Coidier JM, Fievcz C. Lt Fiver, et al: Acroosteolysc et lesions cutanies assoeiie* clwz deux ouvrien. affectes au nettoyaga d'autoclaves. Med Trav 4:14-19, 1966 15. Wilson RH, McCormick WF, Tatum CF. et al: Occupational acrooitcolysis: Report of 31 cases. JAMA 201:577- 561, 1967 16. Violi PL. Bigotti A. Capulo A: Oncogenic response of rat skin, lungs, and bones to vinyl chloride. Cancer Rea 31:516-522. 1971 17. Maitom C. LefemineGL: Carcinogenicity bioassaya of vinyl chloride I. Research plan and early result*. Environ Res 7:3*7-405, 1974 II. Kepiinger ML, Goode JW, Gordon DE. et ai: Interim results of exposure of rats, hamaten, and mice to vinyl chloride. Ann NY Acad Sci 246:219-224. 1975 19. Nicholson WJ. Hammomi EC, Scidman H. et ai: Mortality experience of a cohort of vinyl chloride--polyvinyl chloride workers. Ana NY Acad Sci 246:225-230, 1975 20. Star Series: Scientific and technical assessment report on vinyl chloride and polyvinyl chloride. EPA 600/6-73. 004:48, 1973 21. Tabershaw IR, Gaffey WR: Mortality study atworkers in ihe manufacture ofvinyl chloride and polymer* ofvinyl ddoride. J Occup Med 16:509-318, 1974 22. Tabcrshaw-Cooper Associates Inc: Supplementary epidemiological study of vinyl chloride workers I. Manuf Chcm Assoc 5:1-30. 1975 23. Duck BW, Taylor KJW, Williams DMJ- Mortality study of workers in a polyvinyl chloride production plant. Lancet it: 1197-99. 1975 24. Wxxweiller RJ. Stringer W, Wagoner JK. et al: Neoplastic risk among workets exposed to vinyl chloride. Ann NY Acad Sci 271:40-8, 1976 VINYL CHLORIDE--DELATED C s.tCINGGENESiS '9 the concept mm ir.e ..ep^n'v.,1 ie s jCuil,. .0 r.j.jI jL.'.csr ifcn.i.Ci:..Ltt.on is dependent upon :t5 cb.;.;.- :o -su-wl:";.* the mutu::.T:c,carc:r;ccer.:c rr.ctcboi/.e vf vinyl chloride. This review of our present Itnowledgc of vinyl chloride injury and cancer formation in man is, at best, a very' rough hypothetical outline. With continued investigation and study it will allow us tc more accurately and completely fill in the missing pieces of this fascinating puzzle, thus leading us to a better understanding of the pathogenesis of chemically induced cancer in the biologically complex human system. REFERENCES 1. Elkins HB: The Chemistry of Industrial Toxicology. Second Edition. New York, New York, John Wiley a Son* Inc, 1950 2. Oettel H. In Ullmann's EnzykJopidie JerTechnichenChemie Third Edition. 5 489 Munchen-Berlin. Urban and Scnwarzzncerg, i554. 3. Kunststotlkomisiion aes Bundesgesunahensautes i Plastics Commiitee of the Federal German Ministry of Health), Bandesienur.Jherlshbn, 5,389, `T'J 4. vonOtttngen WF: The Haiogenated Aliphatic. Olefinic. Cyclic, Aromatic, and Aliphatic-Aromatic Hydrocarbons Including the Haiogenated Insecticides, Their Toxicity and Potential Dangers. iPHS Publication No. 4M). Wash* ington. D.C., Government Printing Office, 1955 3. OsterRH. CarrCT. KrantlJC, et aJ: Anesiheiia XXVII. Narcosis with vinyl chloride. > 359-61,1947 6. Schottek W: The toxicity of vinyl chlonde. Chem Teehn 21:708-711, 1969 7. Gauvain S. Vinyl chlonde. Proc Roy Soc Med. 89*275-310. 1973 5. Braun P, Druckman E, Eds: Public-health rounds at the Harvard School of Public Health. Vinyl chlonde: Can the worker be protected? New Eng J Med 294.633-657, 1976 9. ScldtoffU, Hammond EC, Eds: Toxicity of vinyl chlonde--polyvinyl chlonde. Ana NY Acad Sci 246:1-337.1975 10. Irish DO: In Aliphatic haiogenated hydrocarbons. Industrial Hygiene and Toxicology. Second Edition. Edited by FA Fatty. New York. Imerscicnce Publishers. 1963, 2:1241-1331 11. Mastroinatteo E. Fisher M. Christie H, ct al: Acute inhalation toxicity of vinyl chlonde la laboratory animats. Amer Ind Hyg Assoc J 21:394-398. I960 12. Lester D, Greenberg LA. Adams WR: Effects of single and repeated exposures of humans and rats to vinyl chloride. Amer Ind Hyg Assoc J 24*265-275. 1963 13. Filatova VS. Balakhonova I. Orensberg ES. Hygienic characteristics of vinyl chloride production. GigTr Prof Zaboi 2:6. 1951 14. Cordier JM, Fievcz C. L4 F*v*r. et al: Acroosteoiyse t lesions cutance* associccs ehez deux ouvriars. affteres au nenoyaga dsutoclaves. Med Trsv 4:14-19, 1966 13. Wilson RH. McCormick WF. Tatum CF. ct al: Occupational acroosteolysis; Report of 31 cases. JAMA 201:577- 3*1, 1967 16. Viok PL. Bigottl A. Caputo A: Oncogenic respoose of rat skin, lungs, and bona* to vuiyl chlonde Cancer Re* 31:316-322. 1971 17. Mahon C, LefemdeGL: Carcinogenicity bioassays of vinyl chloride I. Research plan and early results. Environ bs 7:317-403, 1974 It. Kephager ML, Oooda JW, Gotdoo DE. et al: Interim results of exposure of mi*, hamsters, and mice to vinyl AM NY Acad 3d 246:219-224. 1973 Ml NkMMB WJ, Hammond EC. Sealman H, et al: Mortality experience of a cohort of vinyl chloride--polyvinyl chloride workers. Am NY Acad Sd 246:223-230, 1975 Ml tar Series: * w-m. iM technical assessment report on vinyl chloride and polyvinyl chloride. EPA 6<*V6-75004:46. 1973 21. Tlfmillair IT fldfir H' MrnF*i--tr~*-- ------- *~~ ~ii--ft--hit -f-inyl -tilnrirti ini r~lyinin nf inyl chloride. J Occup Med 14:306-3 II. 1974 22. Taberthaw-Cooper Associates Inc: Suppicmentary epidemidofical study of vinyl chloride workers I. Manuf Chem Asaoc 5:1-30, 1973 23. Duck BW, Taylor KJW. Williams DMJ: Mortality study of workers in a polyvinyl chloride production plant. Lancet sat 1197-99. 1973 24. WaxseedIsrRJ. Stringer W, WagonerJK. et al: Neoplaaiic risk amongwortenexpoeadto vinyl chloride. Am NY Acad Sd 271:40-1, 1976 CMA 003633 80 CARLO H. TAMBURRO 25. Tamburro, CH: Unpublished results 26. Whelan JC. Creech JL, Tamburro CH: Angiographic and radionuclide characteristics of hepatic angiosarcoma found in vinyl chloride workers. Radiology 118 549-557, 1976 27. Popper H. Thomas LB: Alterations of liver and spleen among workers exposed to vinyl chloride. Ann NY Acad Sci 246:172-193. 1975 28. Hefner RE Jr., Watanabe PG, Gehring PJ: Preliminary studies of the fate of inhaled vinyl chloride monomer in rats. Ann NY Acad Set 246:135-148. 1975 29. Watanabe PG, McGowan GR, Gehring PJ: Fate of "C- vinyl chloride after single oral adminstration in rats. Toxicol Appl Pharmacol 36:339-352. 1976 30. McGowan GR. Watanabe PG, Gehring PJ: Vinyl chloride urinary metabolites: Isolation and identification. Personal Communication, 1977 31. Elmore JD. Wong JL, Laumbach AD. et al: Vinyl chloride mutagenicity via the metabolite* chlorooxirwie and chloracetaldchyde monomer hydrate. Biochim Biophys Acta 442:405-419, 1976 32. Weber G, Lea MA: The molecular correlation concept, in Methods in Cancer Research. Edited by NH Busch. NY, Academic Press Inc, 2:523-578. 1967 33. Maltoni C, Lefemine G: Carcinogenicity bioassays of vinyl chloride: current results. Ann NY Acad Sci 246:193- 219, 1975 34. Schaffner F, Popper H, Selikoff U. et al: Initial features ofvinyl chloride hepatic injury. Gastroenterology ii:(No. 5) A35/928, 1976 35. Schrodt R, Tamburro CH: Unpublished data 36. Girard D, Johnston WC. Grahm JH: Cutaneous angiosarcoma. Cancer 25:868-83, 1970 37. Barr R. Bower M: -Letters" JAMA 231 (9):9|4', 1975 38. Curran KL. Kupchella CE, Tamburro CH: Urinary glycosaminoglycan patterns in angiosarcoma trf the liver. Cancer 40:3050-53. 1977 39. Galambos JT, Shapira R: Natural history of hepatitis: IV glycosaminoglycuronans and collagen in the hepatic connective tissue. J Clin Inves 52 (111:2932-62, 1973 40. Koizumi T, Nakamura N, Abe H; Changes in acid mucopolysaccharide in the liver in hepatic fibrosis. Biochim Biophys Acta 148:749-56, 1967 41. Kojina J: Studies on the metabolism of hepatic connective tissue in fibrosis of the liver. Med J Osaka Univ 16:419-29, 1964 42. Rubin E: Autoradiographic characterization of sulfated acid mucopolysaccharides in experimental cirrhosis. J Histochem Cytochem 14:688-89. 1966 43. Patrick RS, Kennedy JS: The synthesis ofthe sulfated mucopolysaccharides at sites ofhepttic fibrosis is induced by carbon tetrachloride, amyloidosis, and the implantation of catgut. J Pathol Bacterid 88:549-53, 1964 44. Kojinia J, Kanalani M. Ohmori K: The glycosaminoglycans in human hepatic cancer. Cancer Res 35 (3):J42-J7, 1973 45. Kupchella CE, Tamburro CH:t Urinary glycosaminoglycan excretion patterns in chemically induced liver Injury and cancer. Clin Res 25:329, 1977 46. Kupchella CE, Tamburro CH: Urinary and tissue glycosaminoglycan pattern* in hepatic angiosarcoma. In Pro ceedings ofthe III International Symposium on Detection and Prevention ofCancer. Editedby HE Niebergs. NY, Marcel Dckker Inc, 1977 Carlo H. Tamburro, M.D. Cancer Center and Digestive Diseases and Nutrition Division Department of Medicine University of Louisville School of Medicine Louisville, Kentucky 40201 CMA 003634 TOXICOLOGY AND APPLIED PHARMACOLOGY 62. 1-10 (1982) 4 The Effect of Repeated Vinyl Chloride Exposure on Rat Hepatic Metabolizing Enzymes1 Julie T. Du,2 Michael T. Tseng, and Carlo H. Tamburro3 Liver Restarch Center, Division of Digestive Diseases and Nutrition, Departments of Medicine and Anatomy, and Regional Cancer Center, University of Louisville School of Medicine, Louisville. Kentucky 40292 Received July 17, 1980; accepted September 12. 1981 The Effect of Repeated Vinyl Chloride Exposure on Rat Hepatic Metabolizing Enzymes. Du, J. T., Tseng, M, T,, and Tamburro, C. H. (1982). Toxicol. Appl Pharmacol 62, 110. Sprague-Dawley rats were exposed to 2.8% vinyl chloride for 2 (70 hr), 4 (140 hr), and 6 (210 hr) weeks to determine the sequential biochemical changes related to the oxidation and detoxification ability of hepatic tissue. Glutathione-5-transferase(s) activity using 1,2* epoxy-(p-nitrophenoxy)propane and p-nitrobenzyl chloride as substrates was elevated 17 to 24. 28, and 35 to 42% after 2, 4, and 6 weeks of exposure, respectively, suggesting enzymeis) induction. Reduced glutathione, the major substrate required to conjugate the toxic metabolites of vinyl chloride, was also consistently elevated. Similarly, the activity of glutathione reductase, the enzyme necessary for the regeneration of reduced glutathione from its oxidized form, was also increased following vinyl chloride exposure. Cytochromes P-4S0, the major protein involved with vinyl chloride metabolism, was reduced after vinyl chloride exposure, confirming reports of others that vinyl chloride metabolites destroy P-4J0. No abnormalities of standard clinical biochemical blood tests of liver function were found during 6 weeks of vinyl chloride exposure. The only consistent ultrastructural modification was the dilation of endoplasmic reticulum. The biochemical and ultrastructural alterations could reflect early hepatocellular adaptation to vinyl chloride exposure. Vinyl chloride, at high concentrations, has been shown to be carcinogenic in both lab oratory animals (Maltoni and Lefemine, 1975; Viola et al., 1971) and man (Creech and Johnson, 1974). Present data support the metabolism of vinyl chloride by hepatic microsomal mixed-function oxidase system into toxic intermediates, chloroethylene ox ide (Bolt et at., 1975; Hefner et at., 1975; Kappus et at., 1976) and chloroacetaldehyde 1 This work was supported by a grant from the Man ufacturing Chemists Association, Washington, D C, Portions of this study have been presented (Fed. Proc. 37, 1545, 1978). 1 Present address: Clement Associates, Inc., 1010 Wisconsin Avenue, N.W., Suite 660, Washington, D.C. 20007. 1 Address requests for reprints to: Carlo H. Tam burro, University of Louisville. (Gross and Freiberg, 1969). These two in termediates are considered to be the ultimate carcinogens (Barbin et at., 1975; Jaeger et at., 1974b; Van Duuren, 1975), to be mu tagenic in bacterial systems (Elmore et at., 1976; Greim et at., 1975; Malaveille et at., 1975; McCann et at., 19-75), to act as an alkylating agent by reacting with adenosine (Barbin et at., 1975) and cytidine (Laib and Bolt, 1978), and to bind with protein (Bolt et at., 1976; Kappus et at., 1976; Watanabe et at., 1978). Detoxification of these metab olites occurs mainly by conjugation with glu tathione and is catalyzed by hepatic gluta thione transferases; the conjugates are excreted in the urine as substituted cysteine derivatives (Watanabe et al., 1976b,c; Green and Hathway, 1975, 1977), Chloroacetal dehyde can be further oxidized to chloro- I 0041-008X/82/01000I-10502,00/0 Copyright e 1**2 by ArsStlMt Pm. Inc. All nfhu oT reproduce ID uy form mud. CMA 003635 2 DU. TSENG. AND TAMBURRO acetic acid (Hefner et al., 1975). These data are compiled in a metabolic scheme in Fig. 1 as an updated hypothesized metabolic fate of vinyl chloride in the adult rat. The me tabolism of chlorocthylene oxide via epoxide hydratase is not listed in the scheme because its product has not been identified. There have been a few in vivo studies con cerning the effects of vinyl chloride exposure on hepatic concentrations of glutathione (Hefner et al., 1975; Watanabee/oA, 1976c; Du and Tamburro, 1978), cytochromes P~ 450 (Reynolds et al., 1975) and on mixedfunction oxidase activity (Drew et al., 1975; Reynolds et al., 1975). However, informa tion about the sequential alterations in he patic oxidation and detoxification of vinyl chloride, during prolonged exposure, simu lating the occurrence in workers, is still lack ing. It was reported previously that the en zymatic changes in rat liver following prolonged exposure to vinyl chloride were similar to those found in rat hepatoma (Du and Tamburro, 1976; Du et al., 1979). The sequential biochemical changes related to the hepatic oxidation and detoxification of vinyl chloride following prolonged exposure are reported here. METHODS Animals and experimental design. Eight- to tenweek-old Sprague-Dawley male rats ( -- 300 g), sup plied by Laboratory Supply of Indianapolis. Indiana, were randomized prior to the experiment into three group*: a vinyl chloride-expoted group and the air-exposed group housed in identical chambers and a second control group housed in the University's Central Animal Care Center. The exposure level was 28,000 ppm vinyl chloride, 7 hr/day, S days/week for 2, 4, and 6 weeks. The exposure chambers were 4400-liter airtight vat*. Vinyl chloride (~300 to 340 g) was added to the vat to give a time-weighted average concentration of 28,000 1000 ppm. The chamber air was changed daily and the vinyl chloride concentration was determined by gas chromatography. The air was constantly circulated by a stirrer. The rats' respirations had negligible effect on the composition of the chamber's atmosphere because of the chamber's large volume. Animals were fed on standard laboratory chow pellets ad libitum. All animals were anesthetized with ether, blood was drawn from the inferior vena cava, and the animals were (V c) Cl CH * CH2 2* Cl CM2CH2QH (chloroethanol) DETOXIFICATION WITH GLUTATHIONE (chlorooxirane) !Cl ch-ch2 GS CHpCHpOH t k Cl CH2CH0 (*gsh) mm . GAST GS CH2CH0 (chloaoacetaldehyde) I Cl cii2cooh N-Ac-S-(2-hydroxy ethyl) cysteine GS CH2C00H THIODIGL'fCOLlC AC IQ (CHLOAOACtTIC ACID) - Fto. 1. The proposed metabolic fate of vinyl chloride. (GSH, glutathione; MFO, mixed-function oxidase; VC, vinyl chloride; GEST. glutathione 5-epoxide transferase; and GAST, glutathione 5-al dehyde transferase). CMA 003636 VINYL CHLORIDE EFFECT ON ENZYMES 3 killed approximately 20 hr after exposure at I 00 pm each day. NADPH. glutathione, and glutathione disulfide were obtained from Sigma Chemical Company. St Louis. Missouri, 1,2-epoxy-3-(/>mtrophenoxy)propane was purchased from Eastman Kodak Company, Rochester. New York, p-mtrobenayl chloride was obtained from Matheson, Coleman and Beil. East Rutherford, New Jersey; benzphetamine was donated by the Upjohn Company, Kalamazoo, Michigan. Double-distilled water was used throughout. Sample preparation and biochemical determination. Homogenates and subcellular fractions were prepared as described previously (Du et al., 1979). Each sample was prepared from a single organ and kept at 4C during preparation. The remaining liver was frozen rapidly in liquid nitrogen and stored at -70C. Cytochromes P450 concentrations were determined in the frozen mi crosomal fractions the day following sacrifice. The glu tathione (GSH) concentration, as well as glutathione5-transferase. glutathione reductase, and mixed-func tion oxidase activities were determined in the freshly fractionated frozen liver. For the assays using frozen tissue, the livers from control and experimental rats were frozen in an identical manner for the same length of time. Cytochromes P-450 concentration (Omura and Sato. 1964), nonprotein sulfhydryl content (Sedlak and Lindsay, 1968), and glutathione reductase activity (Carlberg and Mannervik, 1975) were determined in the microsomal or cytosol fractions by methods de scribed previously (Du et ai., 1979). Glutathione-5-transferase activity was determined using the 100,000 X g supernatant fraction. 1.2-Epoxy3-(p-nitrophenoxy)propane and p-nitrobenzy! chloride were the substrates for glutathione-5-epoxide transfer ase and glutathione-S-araikyl transferase (GAST), re spectively. Enzyme activity was determined as described by others (Habig et al,, 1974; Kaplowitz et al., 1975). All assays were linear functions of protein concentration and timed for at least 2 min. Solutions of l,2-epoxy-3(p-nitrophenoxy)propane and p-mtrobenzyl chloride were prepared in absolute ethanol; the final ethanol con centration in the incubation mixture was 0.5%. Mixedfunction oxidase activity was estimated in the micro somal fraction by measuring NADPH disappearance in the NADPH-dependent demethylation reaction of benzphetamine (Lu et al., 1972). The protein content was determined by the method of Lowry et al. (1951). The serum clinical liver tests including aspartate ami notransferase, alanine aminotransferase, alkaline phos phatase. bilirubin, cholesterol, and triglyceride were determined by Technicon sequential multiple analyzer computer (SMAC) system. Light and electron microscopy. Small strips of liver were removed under ether anesthesia, sliced into small cubes, placed immediately in ice-cold 1% osmium tetroxide (pH 7.4), and fixed for 2 hr at 4"C. Subse quently, samples were washed overnight in phosphate buffer, dehydrated in ascending alcohol, and embedded in Epon. Tissue blocks were polymerized at 60C for 2 days. Thin sections were cut with a diamond knife and stained with uranyl acetate and lead citrate before ex amination on a Philips 300 electron microscope. For ultrasiructural analysis, three rats randomly selected from controls and groups exposed for 2, 4, and 6 weeks to v;ny| chloride were studied. For light microscopy, a block of tissue was fixed m buffered formalin and processed routinely for paraffin embedding. Sections 6 urn thick were stained with he matoxylin and eosin. Statistical analysis. Analysis of variance was per formed for the various groups at the different time pe riods and multiple comparisons were performed based on the results of the analysis of variance. RESULTS The protein content (mg protein/g liver) in the subcellular fractions in both control and vinyl chloride-exposed groups was the same throughout the exposure (data not shown); the enzymatic results, therefore, are expressed as micromoles of substrate con verted per minute per milligram of protein. There were no statistical differences be tween the normal and air-exposed groups in the glutathione and cytochromes P-450 con tents or in any of the enzyme activities. The nonprotcin sulfhydryl content (Table 1) was significantly elevated from 26 to 54% at 2, 4, and 6 weeks in the vinyl chloride-exposed group compared to both control groups. Al though the nonprotein sulfhydryl content in creased with exposure, the increases were not statistically significant. Glutathione re ductase activity (Table 1) in the exposed group was increased by 53 to 77% at all three time periods. The increase in glutathione reductase was the same at 2 and 4 weeks of exposure but showed a further significant increase after 6 weeks of exposure. Glutathione-S-epoxidc transferase (GEST, Table 1) and glutathione-S-aralkyl transferase (GAST, Table I) activities were signifi cantly higher than controls after 6 weeks of exposure, 37 and 45%, respectively. The cy tochromes P-450 content, on the other hand. CMA 003637 ' ; Vi l CHLORIDE EFFECT ON ENZYMES 1 ..i.sz :.:ja ijy 20 h- v. ^ n \ DPH, glutathione, and jluui '..or.e Jibuittue -ere t-am Sigma C'remtcai C.msanv. Si. Louu. Missouri. 1 2-eDox>-3-(/?-nitrophenoxy Ipropane was purchased from Eastman Kodak Company, Rochester, New York, ^-mtrobeniyl chloride was obtained from Mjtneson, Coleman and 8ell. East Rutherford, New Jersey, benzphetamme was donated bv the Upjohn Company, Kalamazoo, Michigan. Double-distilled water was used throughout. Sample preparation and biochemical determination. Homogenates and subcellular fractions were prepared as described previously lDu et at.. 1979) Each sample was prepared from a single organ and kept at a^C during preparation. The remaining liver was frozen rapidly in liquid nitrogen and stored at -70SC Cytochromes P- j rations were ueiermir.cci m ;ne trozen mi- crowomai fractions the day following sacrifice. The glu tathione IGSH) concentration, as well as glutathione- 5-transferase. glutathione reductase, and mixed-func tion oxidase activities were determined in the freshly fractionated frozen liver. For the assays using frozen tissue, the livers from control and experimental rata were frozen in an identical manner for the same length of time. Cytochromes P-450 concentration (Omura and Sato. 1964). nonprotein sulfhydryl content (Sedlak and Lindsay, 1968), and glutathione reductase activity (Cartberg and Mannervik, 1975) were determined in the microeomal or cytosol fractions by methods de scribed previously (Du et al,, 1979). Glutathione-5-transferaae activity was determined using the 100,000 X g supernatant fraction. 1,2-Epoxy?-(p-nitrophenoxy)propane and p-nitrobenzyl chloride were the substrate* for glutathione-S-epoxtdc transfer ase and glutathione-5-aralkyl transferase (GAST), re spectively. Enzyme activity was determined as described by others (Habig et al., 1974; Kapiowitz et ai, 1975). All assays were linear functions of protein concentration and timed for at least 2 min. Solutions oTI,2-epoxy-3(p-nitrophenoxy)propane and p-nitrobenzyl chloride were prepared in absolute ethanol; the final ethanol con centration in tha incubation mixture was 0.5%. Mixedfunction oxidase activity was estimated in the micro somal fraction by measuring NADPH disappearance in the NADPH-dependeni demethyiatioa reaction of hantphenmma (La et ai, 1972). The protein content van dmermiaed by the method of Lowry r of. (1951). The wnim dinical liver testa including aspartate amvnotranefarena slseine aminotransferase, alkaline phos phatase. btlirubia. choUtterol. and triglyceride were determined by Tediaieoo sequential multiple analyzer computer (SMAC) system. Light and electron microtcopy. Small strips of liver were removed under ether anesthesia, diced into small cub**, placed immediately in ice-cold 1% osmium tetroxide (pH 7.4), and fixed for 2 hr at 4*C. Subse- i-ir- kiv-cii c.srfz't m r'-.ztpha'.e duller, dehydrated in ascending uicohol. and embedded n Epun T'ssue blocks were polymerized it aCC for 2 dais. Thin sections were cut with a diamond knife and stained with uranyl acetate and lead curate before ex amination on a Philips 300 electron microscope. For ultrastructural analysis, three rats randomly selected from controls and groups exposed for 2, 4. and 6 weeks to vinyl chloride were studied. For light microscopy, a block of tissue was tixed in buffered formalin and processed routinely for paraffin embedding. Sections 6 urn thick were stained with he matoxylin and eosin. Statistical analysis. Analysis of variance was per formed for the various groups at the different time pe riods and multiple comparisons were performed based on the results of the analysis of variance. RESULTS The protein content (mg protein/g liver) in the subcellular fractions in both control and vinyl chloride-exposed groups was the same throughout the exposure (data not shown); the enzymatic results, therefore, are expressed as micromoles of substrate con verted per minute per milligram of protein. There were no statistical differences be tween the normal and air-exposed groups in the glutathione and cytochromes P-450 con tents or in any of the enzyme activities. The nonprotein sulfhydryl content (Table 1) was significantly elevated from 26 to 54% at 2, 4, and 6 weeks in the vinyl chloride-exposed group compared to both control groups. Al though the nonprotein sulfhydryl content in creased with exposure, the increases were not statistically significant. Glutathione re ductase activity (Table 1) in the exposed group was increased by 53 to 77% at all three time periods. The increase in glutathione reductase was the same at 2 and 4 weeks of exposure but showed a further significant increase after 6 weeks of exposure. Glutathione-5-epoxide transferase (GEST, Table 1) and glucathione-5-araikyl transferase (GAST. Table 1) activities were signifi cantly higher than controls after 6 weeks of exposure, 37 and 45%, respectively. The cy tochromes /*-450 content, on the other hand. CMA 003638 - DU, TSENG, AND TAMBURRO CMA 0 0 3 6 3 9 TABLE I Sequential Chances in Hepatic Non protein Sulfhydryl, Cytochromes P-450 Content and Activities of Glutathione Reductas t and Glutathione-S-Tkansferases (Epoxide and Aralkyl) in Rats Exposed to Vinyl Chloride' Time (weeks) Nonprotein sulfhydryl (pmol/g liver) Glutathione reductase (100 X pmol/min/mg protein) GEST (100 X nmol/ min/mg protein) GAST (10 x /unol/ min/mg protein) Cytochrome P-450 (nmol/g liver) Treatment Normal control Vinyl chloride-exposed Air control Normal control Vinyl chloride-exposed Air control Normal control Vinyl chloride-exposed Air control Normal control Vinyl chloride-exposed Air control Normal control Vinyl chloride-exposed Air control 0 7.9 0.3 -- -- 5.0 0.2 -- -- 9.1 1.6 -- -- 2.4 0.4 -- -- 17.0 3.7 -- -- 2 7.8 0.4* 9.4 0.2(/ 7.1 0.3' 4.3 0.4* 6.7 0.6*' 4.5 0.2' 7.8 0.4 9.1 1.7 8.1 1.2 2.1 0.3 2.6 0.4 2.2 0.2 17.1 1.7 13.2 1.1 15.5 1.3 4 7.1 0.4* 10 2 0.6*' 6.9 0.4' 4.2 0 4* 6 3 0.5*' 3.7 0.3' 7.5 1 0 9.7 + 0.7' 6.3 0.7' 1 9 0.2 2.4 0.3 1.9 0.2 197 1.3* 15.3 1.3* 19.6 2.7 6 6.9 + 0.4* 11.4 0 6*' 7.9 0.3' 48 + 0.3* 8.9 0 7*"' 5 3 0.4' 7.7 1 0* 110 1.3*' 8.4 or 2.4 0.3* 3.2 + 0.1*' 2.1 0.2' 15.5 1.4* 10 6 + 1.0*' 14.3 1.8' ' RaU were exposed lo 28.000 ppm of vinyl chloride; normal controls and the air controls were exposed to air only. Each number represents the mean and the SEM from a group'of six rats. * Normal vs vinyl chloride-exposed, p < 0.05. ' Air control vs vinyl chloride-exposed, p < 0.05. 4 Vinyl chloride (6 weeks) exposed vs vinyl chloride (2 and 4 weeks) exposed, p < 0.05. VINYL CHLORIDE EFFECT ON ENZYMES 5 was significantly lower than controls after 6 weeks of exposure to vinyl chloride (Table 1). No differences were found in the hepatic mixed-function oxidase activity or in the serum clinical liver tests. After 2 weeks of exposure, the two control groups had gained weight but the vinyl chloride-exposed group did not (Table 2). After 4 weeks of exposure, the normal control group housed at the an imal care center had gained significantly more weight than either the air-control or vinyl chloride-exposed group. After 6 weeks of exposure, however, the vinyl chloride-ex posed group failed to gain weight; the nor mal control group gained more than the aircontrol group (Table 2). Morphological examination revealed poly hedral hepatocytes arranged in irregular plates interposed by vascular sinusoids in the livers of the control rats. This general cytoarchitecture was maintained after vinyl chloride exposure. Hepatocytes in control rats contained a prominent spherical nu cleus, numerous ovoid mitochondria, stacks of rough endoplasmic reticulum (RER), some aggregates of smooth endoplasmic re ticulum (SER), and varying amounts of ly- sosomes and glycogen particles (Fig. 2a).Few interstitial cells were scattered among the hepatocytes. These cells contained few cy toplasmic organelles and could be readily discerned at the light microscopic level by their hyperchromatic nuclei. The sinusoids were linked by fenestrated endothelium and some of the lining cells displayed phagocytic activity. After 2 to 6 weeks of vinyl chloride exposure, the principal organelle affected appeared to be the endoplasmic reticulum. Cisternac of the RER became dilated in a relatively small population of the hepato cytes in the 2-week treatment group. Four weeks after exposure to vinyl chloride, patches of dilated endoplasmic reticulum were prominently displayed in some hepa tocytes (Fig. 2b). At this stage the SER was relatively unaffected. In the 6-week treat ment group, vesiculation of SER and dis tention of RER were easily discernible in a large number of hepatocytes (Fig. 2c). How ever, other cell organelles showed no de monstrable change. These changes, though, are still beyond the resolving limit of the light microscope. The nonhepatocyte com ponents showed minimum changes which TABLE 2 Body Weights of Rats before and after Vinyl Chloride Exposure* Duration (week) Treatment Initial weight <> Final weight (*) Percentage gain 2 Normal control 400 15 433 16 VC-exposed 405 12 396 14 Air oontrol 396 17 414 18 4 Normal control 398 8 450 15 VC-exposed 410 + 16 421 15 Air control 395 10 419 5 6 Normal control 402 14 486 14 VC-expoaed 396 9 398 10 Air control 398 14 449 4 8* -2" 5` 13" 3* 6' 21" <l*r 13" * Analysis of body weight was by regression analysis followed by an analysis of variance on the residuals from the regression equation. (Residual * observed final weight -- predicted final weight front regression equation.) * Normal control vs vinyl chloride exposed, p < 0.05. * Air control vs vinyl chloride exposed, p < 0.05. * Air control vs normal control, p < 0.05. CMA 003640 W1. i ,_ . ,, 1 . r 1), No differences were found in :hs hepatic mixed-function oxidase activity or in the serum clinical liver tests. After 2 weeks of exposure, the two control groups had gained weignt out the vinyl -niwriue-cxpojea group did not (Table 2). After 4 weeks of exposure, the normal control group housed at the an imal care center had gained significantly more weight than either the air-control or vinyl chloride-exposed group. After 6 weeks of exposure, however, the vinyl chloride-exnosea arouo f-.i.rd vs i r. vc: the nor mal control group gamea more than tne aircontroi group tTuo.e 2). Morphological examination revealed poly hedral hepatocytes arranged in irregular plates interposed by vascular sinusoids in the livers of the control rats. Inis general cytoarchitecture was maintained after vinyl chloride exposure. Hepatocytes in control rats contained a prominent spherical nu cleus, numerous ovoid mitochondria, stacks of rough endoplasmic reticulum (RER), some aggregates of smooth endoplasmic re ticulum (SER), and varying amounts of ly- . ,.ere -Ca-'.-Tid ar:._.'. g '1* hepatocytes. These cells contained few cy toplasmic organelles and could be readily discerned at the light microscopic level by their hyperchromatic nuclei. The sinusoids were un^wd oy icncairuteu eaucmcmUin a.*d some of the lining cells displayed phagocytic activity. After 2 to 6 weeks of vinyl chloride exposure, the principal organelle affected appeared to be the endoplasmic reticulum. Cisternae of the RER became dilated in a relatively small population of the hepato cytes in the 2-week treatment group. Four weeks after exposure to vinyi chloride, patches of dilated endoplasmic reticulum were prominently displayed in some hepa tocytes (Fig. 2b). At this stage the SER was relatively unaffected. In the 6-week treat ment group, vesiculation of SER and dis tention of RER were easily discernible in a large number of hepatocytes (Fig. 2c). How ever, other cell organelles showed no de monstrable change. These changes, though, are still beyond the resolving limit of the light microscope. The nonhepatocyte com ponents showed minimum changes which TABLE 2 Body Weights of Rats before and after Vinyl Chloride Exposure* Duration (week) 2 4 6 Treatment Normal control VC-exposed Air control Normal control VC-exposed Air control Normal control VC-expoaed Air control Initial weight (8) 400 15 403 x 12 396 17 398 8 410 16 393 * 10 402 * 14 396 * 9 398 * 14 Final weight (8) 433 x 16 396 x 14 414 x IS 430 X 13 421 X 15 419 x 5 486 14 398 * 10 449 X 4 Percentage gain 8* -2*" 5f 13" 3* 6' 21" <i" 13" ' * Analysis of body weight wai by regression analysis followed by an analysirof variance on the residuals from the rsfrasaioo equation. (Residual " observed final weight - predicted final weight from regression equation.) * Normal control vs vinyl chloride exposed, p < 0.03. ' Air control vs vinyl chloride exposed, p < 0.03. * Air control vs normal control, p < 0.03. CMA 003641 CMA 003642 VINYL CHLORIDE EFFECT ON ENZYMES 7 were characterized by an increased accu mulation of lysosomal-like substances in some of the sinusoidal lining ceils as well as a greater tendency to accumulate lipids in the interstitial cells (Fig. 2d). DISCUSSION Glutathione conjugation is an important pathway for the metabolism of potentially harmful electrophilic metabolites of xenobiotics. Studies by Watanabe et al. (1976b,c) indicate this to be the major route for in activation of the vinyl chloride metabolites. Glutathione-S-transfcrases are a group of cytosol enzymes catalyzing the reaction of glutathione and electrophilic compounds to form less toxic and more water-soluble con jugates. Their activity during chronic ex posure to xenobiotics, like vinyl chloride, could be a key determinate in the ultimate outcome of such exposures as illustrated by the longer arrow in Fig. 1. The increase in hepatic GEST activity at 4 weeks and the later increase in GAST activity at 6 weeks suggested that in the earlier stages of ex posures most of the chlorooxirane interme diate is being adequately detoxified. During the later stages of chronic exposure, more chlorooxirane may become rearranged to yield more chloroacetaldehyde and, in turn, react with other available glutathione trans ferases or become further metabolized to chloroacetic acid. Alternatively, the exces sive chlorooxirane could rearrange sponta neously to form chloroethanol and be further oxidized to chloroacetaldehyde, which in turn may react with glutathione, or be oxi dized to monochloroacetic acid (Johnson, 1967). This would be consistent with the later increases in the aralkyl-transferases and the finding by Hefner et al. (1975) that monochloroacetic acid is found only in the urine of rats exposed for an extended time to higher levels (5000 ppm) of vinyl chloride. The increased use of alternate pathways, for chlorooxirane and chloroacetaldehyde de toxification, may reflect increased concen tration of these active metabolites allowing greater opportunity for DNA injury. A single exposure to vinyl chloride de creased hepatic nonprotein sulfhydryi com pounds in rats (Watanabe, 1976a); similar decreases of glutathione concentrations were produced in rats by other xenobiotics such as 1,1-dichloroethylene (Jaeger et al., 1974a; Reichert et al., 1978) and acetaminophen (Mitchell et al., 1973). In the present study, repeated exposure to vinyl chloride caused a significant increase of nonprotein sulfhy dryi concentrations (Table 1) analogous to the elevation of glutathione concentrations seen after the administration of carcinogens to rats (Fiaia et al., 1976). In addition, the results showed that repeated exposure to vi nyl chloride also caused an increase in he patic glutathione-5-transferase activity (Ta ble 1) similar to that seen after the administration of phenobarbital and 3-methylcholanthrene to rats (Mukhtar and Bresnick, 1976). These data suggest a mecha nism for compensatory synthesis of hepatic glutathione and glutathione-5-transferases after repeated exposure to vinyl chloride. The decreased concentration of cyto chromes F-450 found in rats after repeated exposure to vinyl chloride (Table 1) is con- Fio. 2. (a) Portion of a hepatocyte from control. Stacks or rough endoplasmic reticulum (RER) are separated by many ovoid mitochondria (M). Chromatin is finely dispersed in the nucleus (N). 9300X. (b) Hepatocyte after 2 weeks of vinyl chloride exposure. Dilation of RER appeared widespread in these two cells. Bile (B) canaliculus appeared unaltered in these rats. 5300X. (c) Four weeks after vinyl chloride exposure. Golgi complex (G) appeared unaffected while cisternal dilation continued. Distinction between SER and RER is complicated by the detachment of ribosomes. Lipid droplets (L) and glycogen (GL) often accumulated. 9500X. (d) A fat-storing interstitial cell is surrounded by several hepatocytes in a vinyl chloride-treated animal. Unlike lipid stored in hepatocytes, the shape of lipids (L) appeared irregular in these cells. 8000X. CMA 003643 \ ' 'n \ I C H L 0 A i D b T ON r. \ . '< were characterized ", nk"1;- >_2 2t,r'T'* of 'x co,;:c''*_iln some ot' the sinusoidal lining ceils as well as a greater tendency to accumulate lipids in the interstitial cells (Fig. 2d). DISCUSSION Glutathione conjugation is an important pathway for the metabolism of potentially harmful electrophilic metabolites of xeno- biotics. Studies by Watanabe et al (1976b,c) j a U*C -1 r\ > > r ^ 1o_ activation of the vine 1 chloride metabolites. Giutathione-S-transferases are a group of cytosol enzymes catalyzing the reaction of glutathione and electrophilic compounds to form less toxic and more water-soluble con jugates. Their activity during chronic ex posure to xenobiotics, like vmyl chloride, could be a key determinate in the ultimate outcome of such exposures as illustrated by the longer arrow in Fig. 1. The increase in hepatic GEST activity at 4 weeks and the later increase in GAST activity at 6 weeks suggested that in the earlier stages of ex posures most of the chlorooxirane interme diate is being adequately detoxified. During the later stages of chronic exposure, more chlorooxirane may become rearranged to yield more chloroacetaldehyde and, in turn, react with other available glutathione trans ferases or become further metabolized to chloroacetic acid. Alternatively, the exces sive chlorooxirane could rearrange sponta neously to form chloroethanol and be further oxidized to chloroacetaldehyde, which in turn may react with glutathione, or be oxi- v mof-ivhloroucJ'.': This would he :z~. ; - :h ihe later increases in the aralkyl-transterases and the finding b> Hefner et al. (1975) that monochloroacetic acid is found only in the urine of rats exposed for an extended time to higher levels ( 5000 '*Drr;) of t irv ! -h1 jr'di. The increased use of alternate pathways, for chlorooxirane and chloroacetaldehyde de toxification, may reflect increased concen tration of these active metabolites allowing greater opportunity for DNA injury. A single exposure to vinyl chloride desrecsed hepatic -.c "protein sulfh; dryl :cmpoandi in rats (vVatanabe, 1976a); similar decreases of glutathione concentrations were produced in rats by other xenobiotics such as 1,1-dichloroethylene (Jaeger era/., 1974a; Reichert et at., 1978) and acetaminophen (Mitchell er al., 1973). In the present study, repeated exposure to vinyl chloride caused a significant increase of nonprotein sulfhydryl concentrations (Table 1) analogous to the elevation of glutathione concentrations seen after the administration of carcinogens to rats (Fiala et a!., 1976). In addition, the results showed that repeated exposure to vi nyl chloride also caused an increase in he patic glutathione-5-transferase activity (Ta ble 1) similar to that seen after the administration of phenobarbital and 3-methylcholanthrene to rats (Mukhtar and Bresnick, 1976). These data suggest a mecha nism for compensatory synthesis of hepatic glutathione and glutathione-5-transferases after repeated exposure to vinyl chloride. The decreased concentration of cyto chromes P-450 found in rats after repeated exposure to vinyl chloride (Table 1) is con- FlO. 1 (a) Portion of a hepatocyte from control. Stacks of rough endoplasmic reticulum (RER) are separated by many ovoid mitochondria (M). Chromatin is finely dispersed in the nucleus (N). 9300X. (b) Hepatocyte after 2 weeks of vinyl chloride exposure. Dilation of RER appeared widespread in theae two eelle, Bile (B) canaliculus appeared unaltered in these rats. 5300X. (c) Four weeks after vinyl chloride exposure. Golgi complex (G) appeared unaffected while cisternal dilation continued. Distinction between SER end RER is complicated by the detachment of riboaomea. Lipid droplets (L) and glycogen (GL) often accumulated. 950QX. (d) A fat-storing interstitial cell is surrounded by several hepetocytes in a vinyl chloride-treated animal. Unlike lipid stored in hepetocytes. the shape of lipide (L) appeared irregular in these cells. 8000X. CMA 003644 8 DU, TSENG, AND TAMBURRO sistent with work by Reynolds et ai. (1975). This decrease in cytochromes P-450 content was also shown in vitro (Guengerich and Strickland, 1977; Ivanetich et al., 1977) sug gesting that a metabolite of vinyl chloride destroys the cytochrome. Mixed-function oxidase activity, with benzphetamine as sub strate was unaltered. With regard to the structural alterations produced by vinyl chloride, the present find ings confirmed previous observations on the selective effect of this carcinogen in the en doplasmic reticulum (Du et al., 1979). A gradual increase in the number of hepatocytes affected and the involvement of both smooth and rough GR were shown in this study. The endoplasmic reticulum is the pri mary site of protein synthesis and its dilation suggested the presence of a vinyl chloriderelated effect. The lack of a concomitant Golgi hypertropy is noteworthy since this organelle serves as the site of glycosylation and packaging of many exportable proteins. It may be inferred that the vinyl chloride effect is mainly an intracellular phenome non. The relatively late involvement of SER could reflect a further attempt at detoxifi cation by the hepatocyte. The tendency for the interstitial cells to accumulate lipid and the heightened phagocytic activity correlate with increases in collagen formation and may be evidence of low-grade cellular injury. This may prove to be the initial histological response to vinyl chloride exposure. Ultrastructural responses to chronic in halation of vinyl chloride were reported by Feron et ai. (1979). Unlike our findings, the principal effects observed were swollen mi tochondria and some proliferation of SER. The mitochondrial swelling presumably re flects the extensive vacuolization observed by light microscopy. The discrepancy prob ably resulted from differences in the dose and duration in vinyl chloride exposure since Feron et al. (1979) exposed rats to 5000 ppm vinyl chloride for 52 weeks. Such an exten sive mitochondrial lesion could result in bio chemical modifications such as a change in ATPase or cytochrome oxidase levels. The only enzyme activity measured, glucose-6phosphatase, was reduced. The authors believed that the altered glu tathione metabolism, as reflected by the in creased nonprotein sulfhydryl content, and the increased activities of glutathione-5transferases and glutathione reductase, in rat liver after repeated exposure to high doses of vinyl chloride represent an early hepatocellular adaptation to vinyl chloride exposure. ACKNOWLEDGMENTS The authors wish to express their gratitude to the staff of the B. F. Goodrich Plant in Louisville, Kentucky, for their cooperation in the exposure studies, to Dr. Richard A. Greenberg for his review, to Mr. John P. Sandoz for his help with the statistical analysis, and Mr. John Kreisle and Ms. Debra S. Eades for technical as sistance. REFERENCES Barbin, A., Bresil, H., Croisy. A., Jacquionon. P., Malaveille, C,, Montesano, R., and Bartsch, H. (I97S). Liver-microsome-mediated formation of alkylating agents from vinyl bromide and vinyl chlo ride. Biochem. Biophys. Res. Common. 67, S96-603. Bolt, H. M.. Kaffus, H., Buchter, a., and Bolt, W. (1975). Metabolism of vinyl chloride. Lancet 1, 1425. Bolt, H. M., Kaffus, H,, Kaufmann, R., Affel. K. E., Buchter, A., and Bolt. W. (1976). Metabolism of l4C-vinyl chloride in vitro and in vivo. Inserm Sym posia Ser. 52, 151 -164,1ARC Scientific Publications No. 13. Carlberg, I., AND MaNNERvik, B. (1975). Purifica tion and characterization of the flavoenzyme gluta thione reductase from rat liver. J. Blot. Chem. 250, 5475-5480. Creech, J. L., and Johnson, M. N. (1974). Angios arcoma of liver in the manufacture of polyvinyl chlo ride. J. Occup. Med. 16, 150-151. Drew, R. T,, Harter. C. Gufta. B. N., and Talley, F. A. (1975). Effects of vinyl chloride exposures to rats pretreated with phcnobarbital. Environ. Health Fersptc. 11, 235-242. Du, J. T., and Tamburro, C. H. (1976). Decreased glucose-6-phophatase activity in liver in vinyl chlo ride exposed rats. Fed. Proc 35, 329. CMA 003645 VINYL CHLORIDE EFFECT ON ENZYMES 9 Du, J. T., and Tamburro, C H. (1978). Elevated glutathione content, glutathione-S-transferase and glutathione reductase in liver of rats exposed to vinyl chloride. Fed. Proc. 37, 1545. Du, J, T. Sandoz, J. P.. Tseng, M T, and Tam burro, C H. (1979) Biochemical alterations in liv ers of rats exposed to vinyl chloride. J. Toxicol. En viron. Health 5, 1119-1132. Elmore, J. D., Wong, J. L, Laumbach, a. D,, and Streips, U, N, (1976). Vinyl chloride mutagenicity via the metabolites chlorooxirane and chloroacetaldehyde monomer hydrate. Biochim. Biophys. Acta 442, 405-419. Feron, V J., Spit, B. J., Immel, H. R., and Kroes, R, (1979). One-year time sequence inhalation toxicity study of vinyl chloride in rats. III. Morphological changes in the liver. Toxicology 13, 143-154 Fiala, S.. Mohindru, A., Kettering, W. G., Fiala. A. E , and Morris, H. P. (1976). Glutathione and gamma glutamyl transpeptidase in rat liver during chemical carcinogenesis. J. Sat. Cancer Inst. 57,591 598. Green, T., and Hathway, D. E. (1975). The biolog ical fate in rats of vinyl chloride in relation to its oncogenicity. Chem.-Biol. Interact. II, 545-562. Green, T.. and Hathway, D. E. (1977). The chem istry and biogenesis of S-containing metabolites of vinyl chloride in rats. Chem.-Biol. Interact. 17, 137150. Greim, H., Bonse, G., Radwan, Z., Reichert, D,, and Henschler, D. < 1975). Mutagenicity in vitro and potential carcinogenicity of chlorinated ethylenes as a function of metabolic oxirane formation. Biochem Pharmacol. 24, 2013-2017. Gross, H., and Freiberg, J. (1969). Alpha-halo ethers. 41. Existence of chloroethyiene oxide. J. Prakt. Chem. 311, 506-510. Guengerich, F. P,, and Strickland, T. W. (1977). Metabolism of vinyl chloride: Destruction of the heme of highly purified liver microsomal cytochrome P-450 by a metabolite. Mol. Pharmacol. 13, 993-1004. Habig, W. H,, Pabst. M. J., and Jakoby, W. B. (1974). Glutathione-5-transferases. The first enzy matic step in mercapturic acid formation, J Biol. Chem. 249, 7130-7139. Hefner, R. E., Watanabe. P. G., and Gehring. P. J. (1975). Preliminary studies of the fate of inhaled vinyl chloride monomer in rats. Ann. S. Y. Acad. Sci. 24*. 135-148. Ivanehch. K. M., Aronson, I., and Katz, 1. D. (1977). The interaction of vinyl chloride with rat he patic microsomal cytochrome P-450 in vitro. Biochem. Biophys. Res. Common. 74, 1411-1418. Jaeger, R. J,, Connolly, R. B., and Murphy, S. D. (1974a). Effect of 18 hr. fast and glutathione deple tion on 1,1-dichloroethylene-induced hcpatotoxicity and lethality in rats. Exp. Mol. Pathol. 20, 187-198. Jaeger, R. J., Reynolds. E. S.. Connolly, r. b., Moslen. M. T.. Szabo, S., and Murphy, S. D, (1974b). Acute hepatic injury by vinyl chloride in rats pretreated with phenobarbetal. Nature (London} 252, 724-726. Johnson, M K. (1967) Metabolism of chloroethanol in the rat. Biochem. Pharmacol. 16, 185-199 Kaplowitz, N., Kuhlenkamp, J., and Clifton, G. (1975). Drug induction of hepatic glutathione-Stransferascs in male and female rats. Biochem. J. 146, 351-356. Kappus, H., Bolt, H. M., Buchter, A,, and Bolt, W (1976). Liver microsomal uptake of l4C-vinyl chloride and transformation to protein alkylating metabolites in vitro. Toxicol. Appl. Pharmacol 37, 461-471. Laib, R. J.. and Bolt, H. M. (1978). Formation of 3, N'-ethcnocytidine moieties in RNA by vinyl chloride metabolites in vitro and in vivo. Arch. Toxicol. 39, 235-240. Lowry, O. H,, Rosebrough, N. J., Farr, A. L.. and Randall, R. J. (1951). Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265275. Lu, A. Y. H,, Kunztzman, R., West. S., Jacobson, M., and Conney, A. H. (1972). Reconstituted liver microsomal enzyme system that hydroxylates drugs, other foreign compounds and exogenous substrates. J. Biol. Chem. 247, 1727-1734. Malaveille, C., BarTSCH, H., Barbin, A., Camus, A. M., and MONTESANO, R. (1975). Mutagenicity of vinyl chloride, chloroethyiene oxide, chloroacctaldehyde and chloroethanol. Biochem Biophys. Res. Common. 63, 363-370. Maltoni, C., and LEFEMINE, C. (1975). Carcinoge nicity bioassays of vinyl chloride: Current results. Ann. N Y. Acad. Sci. 246, 195-218. McCann, J.. Simmon, V., Streitwieser, D., and AmES, B. N. (1975). mutagenicity of chloroacetaldehyde, a possible metabolic product of 1,2-dichloroethane, chloroethanol, vinyl chloride and cyclo phosphamide. Proc. Nat. Acad. Sci. USA 72, 31903193. Mitchell, J. R., Jollow, D. J.. Potter, W. a.. Gillete. J. R , AND BrODIE, B. B. (1973). Acetaminophen-induced hepatic necrosis. IV. Protective role of glutathione. J. Pharmacol. Exp. Ther. 187, 211217. Mukhtar, H., and Bresnick, E. (1976). Effects of phenobarbital and 3-methylcholanthrene administra tion on glutathione-5-epoxide transferase activity ip rat liver. Biochem Pharmacol. 25, 1081-1084. Omura, T., and Sato, R. (1964). The carbon mon oxide-binding pigment of liver microtomes. I. Evi dence for its hemoprotein nature. J. Biol. Chem 239, 2370-2378. Reichert, D., Werner, H. W., and Henschler, D. CMA 003646 Du, j r., and gi.,uihione content '-.j. r j,-a i- - giutatmone reduc'.Jai n n?r 01 -jta cnloride. Fed Proc 37, 1545 ; i.m., Du. J T. Sandoz. J P. Tseng. M T. and T\m8LRRO, C H (!97<?> Biochemicil altefitions in liv ers of rats exposed to vinyl chloride J Toxicoi En viron Healths. 11 19-1! 32 Elmore. J D., Wong, J. L , Lalmbalh, A. D , and Streips, U. N (19761 Vinyl chloride mutagenicity via the metabolites chlorooxirane and chloroaeetaldehvde monomer hydrate. Biochtm. Biophvs. Acta 442. 405-419 Feron, V J . Spit, B J . Immel. H R . and Kroes. R I 1979), One-year time sequence inhalation toxicity study of vmyl chloride in rats III Morphological changsa .n the ...cr ......... 13. .45 ,; 4 FlALA, S., MuhiNDRL, A., iNETTcRinai, A 0 , flAUA. A E. and Morris. H P tl9*6V Glutathione ind gamma glutamyl transpeptida.se in rat liver during chemical carcinogenesis. J "Cat Cancer Inst 57, 591 - 598. Green. T,. and Hathway, D E, (1975) The biolog ical fate in rats of vinyl chloride in relation to its oncogenicity Chem -Biol. Interact. II. 545-563. Green, T.. and Hathway, D, E 119"') The chem istry and biogenesis of S-contaimng metabolites of vmyl chloride in rats. Chem.-Biol Interact. 17. 137- 150. Greim, H.. Bonse, G.. Raowan, Z.. Reichert, D , AND HENSCHLER, D, (197$). Mutagenicity in vitro and potential carcinogenicity of chlorinated cthylenes as a function of metabolic oxirane formation. Biochtm. Pharmacol. 24, 2013-2017 Gross. H.. and Freiberg, J. (1969) Alpha-halo ethers. 41. Existence of chloroethylene oxide. J. Prakt. Chtm. 311. 506-510. Guengerich. F. P,, and Strickland, T. W. (1977). Metabolism of vinyl chloride: Destruction of the heme of highly purified liver microsomal cytochrome P-450 by a metabolite. Mol. Pharmacol 13, 993-1004. Habig, W H.. PaBST. M. J., and JaKOBY, W. B. (1974). Glutathione-S-transferases. The first enzy- - matic step in mercapturic acid formation. J Biol. Cham. 24*. 7130-7139. Hetnir. R. E, Watanare. P. G,, and Gehring, P. J. (1975), Preliminary studies of the fate of inhaled vinyl chloride monomer in rats. Amt (V. Y. Acad. Sci. 246, 135-148. Ivanbuch. K. M., aronson, I., and Katz, I. D. (1977). The interaction of vinyl chloride with rat he patic microanmal cytochrome P-450 in vitro. Biochtm. Blophys. Has. Common. 74, 1411-1418. Jaeger. R. J., Connolly, R. B,, and Murphy, S. D. (1974a). Effect of 18 hr. fast and glutathione deple tion on 1,1-dichioroethylcne-inducad hepatotoxictty and lethality in rata. Exp. Mol. Pathol. 20, 187-198. J),_ ` ` 3 A-C- - '-NO tiv'upi Aau'.u ,,cr .'jury by *in>, -n.-jr-js in ri:s prctrcited v.uh pnenobarbital. Auiurr London' 252. *:4-726, Johnson. M K (196"') Metabolism of cn! jroeih-inol in the rat. Btochem Pharmacol. 16. 135-199 K iPI OAITZ N . Ku hLENA. amp. J . tv; Cl.-- IN, 0 ( 1975) Drug induction of hepatic glutathione-5transferases in male and female rats. Btochem J 146, 551-356. Kappis. H,, Bolt, H, M . Blchter, A., and 3olt. W (1976). Liver microsomal uptake of 'C-vinyl cnloride and transformation to protein aikviaung metabolites in vitro. Toxicol Appl Pharmax I 37. 461-4*1 -x.a. '5 ; . on oui.T. h M (i9~.il. Formation cl J. N `-ethenocytidine moieties in RfuA Py vinyi chloride met ahoir.:-, on wtrr, _nd in vivo. Arch. ToXix.il. 59. 235-240 Lowry, O. H., Rosebrough, N. J.. FaRR, A. L., and Randall. R. J 11951).. Protein measurement with the Folm phenol reagent. J Biol. Chem 193. 265275. Ll. A. Y H., Klnztzman. R., West, S.. Jacobson, M,, and ConneY. A. H. (1972). Reconstituted liver microsomal enzyme system that hydroxylates drugs, other foreign compounds and exogenous substrates. J. Blot. Chtm. 247, 1727-1734. Malaveille, C., Bartsch, H., Barsin. a.. Camus. A. M., and Montesano, R. (1975). Mutagenicity of vinyl chloride, chloroethylene oxide, chioroacetaldehydc and chloroethanol. Biochtm. Biophvs Res. Common. 63, 363-370. Maltoni, c.. and LEFEMlNE. C. (1975). Carcinoge nicity bioassays of vinyl chloride: Current results. Ann. ,Y Y. Acad. Sci. 246, 195-218. McCann. J., Simmon, V,, Streitwieser, D.. and Ames. B. N. (1975). mutagenicity of chloroacetaldchyde. a possible metabolic product of 1.2-dichloroethane, chloroethanol. vinyl chloride and cyclo phosphamide. Proc Nat. Acad. Sci. USA 72, 31903193. Mitchell, J. R.. Jollow, D. J., Potter, W. a.. Gillete, J. R.. and Brooie, B. B. (1973). Acetaimnophen-induced hepatic necrosis. IV. Protective role of glutathione. J Pharmacol. Exp. Thar. 187, 211217. Mukhtar, H., and Bresnick, E (1976). Effects of phcnobarbital and 3-methylcholanthrene administra tion on glutathione-5-eposide transferase activity fn rat liver. Biochtm. Pharmacol. 23, 1081-1084. Omura. T., and SaTO, R. (1964). The carbon moooxide-binding pigment of liver tmerooomes. I. Evi dence for its hemoprotein nature. J. Biol. Chtm. 239, 2370-2378. Reichert, D.. Werner, H. W,, and Henschler. D. CMA 003647 10 DU. TSENG, AND TAMBURRO (1978). Role of liver glutathione in 1,1 -dichloroethylene metabolism and hepatotoxicity in intact rats and ,isolated perfused rat liver. Arch. Toxicol. 41 169- 178. Reynolds, E. S,, Moslem. M, T, Szabo, S., and Jaeger, R, J. (1975). Vinyl chloride-induced deac tivation of cytochrome P-450 and other components of the liver mixed function oxidase system: An in vivo study. Res. Commute Chem. Pathol. Pharmacol. 12, 685-694. Sedlak, J., and Undsay, R. H. (1968). Estimation of total protein bound and non-protein sulfhydryl groups in tissue with Ellman's reagent. Anal. Biochem. 25, 192-205. VanDuuren, B. L. (1975). Possible mechanism of car cinogenic action of vinyl chloride. Ann. (V. Y. Acad. Sci. 244, 258-267. Viola, P, L., Biootti, A., and Caputo, A. (1971). Oncogenic response of rat skin, lungs and bones to vinyl chloride. Cancer Res. 31, 516-522. Watanabe, P. G.. Hefner, R. E., Jr,, and Gehring, P. J. (1976a). Vinyl chloride-induced depression of hepatic non-protein sulfhydryl content and effects of bromosulphalcin (BSP) clearance in rats. Toxicology 6, 1-8. Watanabe, P. G., McGowan, G. R., and Gehring, P. J. (1976b). Fate of l4C-vinyl chloride after single oral administration in rats. Toxicol. Appl. Pharma col. 34, 339-352. Watanabe, P. G,, McGowan, G. R., Madrid, E. O., and Gehring. P. J. (1976c). Fate of l4C-vinyl chlo ride following inhalation exposure in rats. Toxicol. Appl Pharmacol. 37, 49-50. Watanabe, P. G., Zempel. J. A., Pegg, D. G., and Gehring, P. J, (1978). Hepatic macromolecular binding following exposure to vinyl chloride. Toxicol. ,Appl. Pharmacol. 44 571-579. CMA 003648 5 E>irtronmental Health Perspectives Vo/, U. PP- A ftm Effectiveness of Federally Required Medical Laboratory Screening in the Detection of Chemical Liver Injury by Carlo H. Tamburro* and Richard Greenberg* The increasing concern of industrialized societies over the potential health hazard of synthetic chemicals in the occupational environment has led to goverment requirements for medical laboratory screening of workers. The specific tests for such screening programs are most often selected on the basis of medical experience which utilized them in symptomatic or hospitalized populations. Required screening tests for hepatic injury including cancer in vinyl chloride workers has been systematically and prospectively studied in an industrial population working with synthetic rubber and plastics. Approximately 1300 employees were studied over a five-year period. A cohort of %9 male employees, for the purposes of analysis, were divided into a "standard" and "nonstandard" population based upon the absence or presence of significant medical disease (including liver disease). A subcohort of 130 individuals was further identified based on availabiliity of liver biopsy. Evaluation of federally required studies included alkaline phosphatase (AIM. y-glutamyl transpcptidnse ((i(iTI'), alanine nminotranserase (ALT, 801*7), aspartic aminotransferase (AST. SOOT) and bilirubin (HR). Also studied were indocyanine green clearance (ICO) and radioisotopic liver spleen scans (L-S scans). The OOTI* provided the highest positive predicted value us a screening test for identifying "nonstandard" individuals (individuals with all types of medical disease) followed by ICO, AST. ALT, L-S scan, Al*. and HR. In the identification of asymptomatic liver disease the OOTI* had the least specificity due to a high false positive rate, while the Al* provided the highest specificity. The ICG clearance however, provided the best combination of positive predictive value and sum of specificity and sensitivity. The AP provided additional increase in specificity as a follow-up study. There was no evidence that any of the other federally required tests added any additional benefit and did add significant increase in the false positive rate. These studies support the need for evaluating screening tests as to their sensitivity, specificity and positive predictive value, in asymptomatic individuals, before they are made established requirements. Introduction Industrialized societies throughout the world have become increasingly concerned over the poten tial health hazard of synthetic chemicals in the occupational environment. Governmental regula tions have increased the number and types of medical laboratory screening required for a large variety of halogenated hydrocarbons as well as other potential environmental hazards. The pri mary objective of these screening programs is to `Liver Research Center. DiviMun of Digestive Diseases, Department of Medicine, and Department of Community Health and Cancer Center, University of Louisville School of Medicine, Louisville, Kentucky 40201. October 1981 reduce disability, morbidity and mortality in work ers, especially as related to serious low-grade health hazards. In general, screening programs are instituted because of the presence in the work environment of a suspected or proven environmen tal toxin or carcinogen, which has the potential of producing low-grade injury over long periods of exposure. Most screening studies are directed toward the detection of abnormalities in certain body systems. The specific tests are frequently selected on the basis of medical experience which utilized them in symptomatic or hospitalized populations. Prior expe riences utilizing nonspecific multiphasic health sur veillance screening and maintenance have not proven to be cost effective except under certain limited 117 CMA 003649 conditions (i). The cost effectiveness of such tests, however, in the determination of medical screening requirements, has played a limited role due to the potential seriousness of these occupational agents. Little attention has been paid as to whether the effectiveness of federally required screening pro vides the best or, more importantly, a necessary benefit when applied to asymptomatic and other wise healthy worker populations. The discovery in 1974 of hepatic toxicity and cancer formation in vinyl chloride workers provided the opportunity to systematically and prospectively study the effectiveness of federally required and federally recommended medical screening proce dures for the detection of chemical liver injury, including cancer development (2). Table 1 lists the federally required medical screening procedures since 1974 for environments utilizing vinyl chloride or polyvinyl chloride. Table 2 lists the federally recommended studies for these same environments. This paper will present a preliminary assessment of the effectiveness of these federally required studies in the accurate detection and identification of chem ically induced liver injury due to halogenated hydro carbons, especially vinyl chloride. Materials and Methods The industrial population studied consisted of approximately 1200-1400 employees of a chemical plant whose two major products were synthetic rubber and plastics. The industrial plant had been in operation for over 35 years and had a predomi nance of male employees (96%), approximately 80-87% of the work force being white, 11-12% black, less than 1% of other racial origins. Turnover of the plant was approximately 10 to 15% per year with 65-70% of the work force having worked five years or more at the plant. Employee ages ranged from 18-65, with a mean of 52 years. A cohort consisting of 969 male employees who worked continually from June 1, 1976 to May 31, 1977 was, for purposes of this analysis, divided into a "standard" and a "nonstandard" population. These designations were given on the basis of a review of Tabic 1. Federally required studies for vinyl chloride workers. History and physical < 10 years as vinyl chloride worker--(annual) > 10 years as vinyl chloride worker--(semiannual) Biochemical studies SGOT (AST) SGPT (ALT) GGTP AP TB 118 Table 2. Federally recomrmnded (not required) studies. Hepatic studies LDH isoenzyme Total protein Protein electrophoresis Radioisotopic scan Kidney dysfunction (urine examination) Albumin RBC Exfoliative abnormal cells Pulmonary system FVC FEV, Chest x-ray (PA and lateral) all present standard medical data on each employ ee, including the federally required studies. Other screening studies of the medical surveillance pro grams were not utilized in the classification of overall medical status because, at that time, their clinical usefulness was unknown or controversial. All studies were performed on an annual basis; those individuals with ten years or more of employ ment were examined and screened semiannually. Compliance with medical screening studies during the five-year study period showed a continuous participation in the history and physical examina tions by over 75% of the work force, laboratory tests and chest x-rays by 86%, and liver-spleen scans by 85%. Seventeen percent failed to undergo at least one history and physical examination, 9% did not have any of the radiological studies, and only 4% failed to have laboratory studies during this period. Approximately 40-50% of these individuals who did not undergo an examination claimed to have been examined by their private physician. A subcohort of 120 individuals was further identified based on the availability of a liver biopsy performed for medical reasons, both related and not related to their work. The term "standard" is used for those individuals who, based upon the best medical opinion, demon strated no evidence of any significant medical disease, occupational or nonoccupational in origin. The "nonstandard" population included all others not included in the standard population. The subcohort population was divided into those individuals with and without histological evidence of liver injury and farther subdivided into those with and without histological features characteris tic of chemical injury. All employees had individual work histories. These consisted of a standardized job classification for all jobs within the plant since its opening and a Environmental Health Perspectives CMA 003650 rank ordering of exposure for 22 different sus pected or potentially hazardous heptatoxic chemi cals used within the work place (J-5). The agents were rank ordered on the basis of the intensity of exposure for each of the job classifications for each of the years that the plant was in operation. From this detailed work history, a cumulative exposure ran month ration (CERM) wa3 determined for each employee for each of the 22 chemicals. All histologi cal material was classified as to the presence or absence of iiver disease, and to whether the abnor malities were consistent with chemical or non chemical iryury. This classification was conducted double blindly by three experienced physicians, two pathologists, and a hepatologist (6), without knowledge of any medical data, exposure or work history. Results Although 50 or more biochemical screening tests were performed during this study period, this paper will limit itself to the evaluation of the federally required studies, the indocyanine green clearance (ICG) study at the 0.5 mg/kg dose (7, 8) and radioisotopic liver and spleen scan (9). The 100 GGTP i CG 05 SGOT A5T SGPT ALT TESTS LIVER SPLEEN size Figure l. Positive predictive values of screening tests in identification of medical disease in an asymptomatic working population (N = 969). All those screening tests with positive predictive values of greater than 70 are shown except for indirect bilirubin (due to high number of congenital indirect hyperbilirubinemia) and triglyceride determination. Above each bar in the graph are shown the sum values for sensitivity and specificity of each test. They generally follow the same ranking. October 1981 Figure 2. Frequency with which clinical biochemical tests cor rectly reflex the presence of hepatic damage in chemical workers suspected of having liver disease: (SGPT) alanine aminotransferase (ALT) (GGT) y-glutamyl transpeptidase, (SGOT) aspartic aminotransferase (AST), (Aik. Phos.) alka line phosphatase, (ICD) isocitric dehydrogenase (ICG) Indo cyanine Green clearances at 0.5 and 5.0 mg/kg dose. federally required biochemical studies include alka line phosphatase (AP), y-glutamyl transpeptidase (GGPD, alanine aminotransferase (ALT/SGPT), bilirubin (ALT/SGPT), and the aspartic amino transferase (AST/SGOT). The positive predictive values of these screening tests in identifying medical disease (including liver disease) in this asymptomatic working populatimi are shown in Figure 1. The GGTP provided highest positive predictive value as a screening tWr for "nonstandard" individuals. It also provided the highest sensitivity and specificity sum shown in brackets. The predictive value of the other tests, in decreasing positivity were ICG, AST, ALT, liver and spleen scan, AP, and bilirubin. Further evaluations were conducted on the subcohort population in whom we had both histolog ical and biochemical data concerning hepatocellular damage. If one looks at only those individuals who received liver biopsies for suspected liver disease then one would find the percent of positive tests as illustrated in Figure 2. The ALT (SGPT), GGTP, AP and AST (SGOT) demonstrate a very high degree of sensitivity in identifying individuals with hepatic disease. As shown on the right, ICG clearances at the 0.5 mg/kg level provide a similar degree of sensitivity to SGOT and AP. The higher dose ICG clearance (5 mg/kg) appears to provide the most sensitivity for latent hepatic disease. These findings are consistent with the general medical experience with hospitalized patients. Sensitivity alone however is not an adequate indicator of a test's screening value, especially when used in asymptomatic individuals. More appro priate evaluation of these tests' value as screening instruments are shown by their sensitivity, specificity 119 CMA 003651 pected or potentially lianard^us r.cp:a;.\;e ,.h cals used '.vithin the work place T'.ic ^-enta were rank ordered on the basis of the intensity of exposure for each of the job classifications for each of the years that the plant was in operation. From this detailed work history, a cumulative exposure ran month ration (CERM) was determined for each employee for each of the 22 chemicals. All histologi cal material was classified as to the presence or absence of liver disease, and to whether the abnor malities were consistent with chemical or non chemical injury. This classification was conducted double blindly by three experienced physicians, two pathologists, and a hepatologist (8). without knowledge of any medical data, exposure or work history'. Results Although 50 or more biochemical screening tests were performed during this study period, this paper will limit itself to the evaluation of the federally required studies, the indocyanine green clearance (ICG) study at the 0.5 mg/kg dose (~, 8) and radioisotopic liver and spleen scan (9). The Figure 1. Positive predictive values of screening tests in identification of medical disease in an asymptomatic working population (N 989). All those screening tests with positive predictive values of greater than 70 are shown except for indirect bilirubin (due to high number of congenital indirect hyperbilirubinemia) and triglyceride determination. Above each bar in the graph are shown the sum values for sensitivity and specificity of each test. They generally follow the same ranking. October 1981 SGPT GGT ALK SGOT ICO BILI- Q5mg50m<j PHOS RUBIN ICG Figure; 2. Frequency with which clinical biochemical tests cor rectly reflex che presence of nepatic damage in chemical workers -usoected of having liver disease (SGPT) alarine aminotransferase ALT 'JUT -'-glutamyl :rar.saeptidase. iSuuT) aspartic aminotransferase (AST). (Aik. Phos.) alka line phnsnnacase, TCD) isocitnc dehydmeerase 'ICG1 Indo cyanine Green clearances at 0.5 and 5.0 mg kg dose. federally required biochemical studies include alka line phosphatase (AP\ y-glutamyl transpeptidase (GGPT). alanine aminotransferase (ALT/SGPT). bilirubin (ALTSGPT), and the aspartic amino transferase (AST/SGOT). The positive predictive values of these screening tests in identifying medical disease (including liver disease) in this asymptomatic working population are shown in Figure 1. The GGTP provided the highest positive predictive value as a screening test for "nonstandard" individuals. It also provided the highest sensitivity and specificity sum shown in brackets. The predictive value of the other tests, in decreasing positivity were ICG, AST, ALT, liver and spleen scan, AP, and bilirubin. Further evaluations were conducted on the subcohort population in whom we had both histolog ical and biochemical data concerning hepatocellular damage. If one looks at only those individuals who received liver biopsies for suspected liver disease then one would find the percent of positive tests as illustrated in Figure 2. The ALT (SGPT), GGTP, AP and AST (SGOT) demonstrate a very high degree of sensitivity in identifying individuals with hepatic disease. As shown on the right, ICG clearances at the 0.5 mg/kg level provide a similar degree of sensitivity to SGOT and AP, The higher dose ICG clearance (5 mg/kg) appears to provide the most sensitivity for latent hepatic disease. These findings are consistent with the general medical experience with hospitalized patients. Sensitivity alone however is not an adequate indicator of a test's screening value, especially when used in asymptomatic individuals. More appro priate evaluation of these tests' value as screening instruments are shown by their sensitivity, specificity 119 CMA 003652 20 ^5 SIN5 ||H 5Pf'I A * L' A 0 E i'Sue F[gl're 3. Sensitivity and specificity of various biochemical screening tests and their sensitivity and specificity sum values (S & S) based on 78 with biopsy documentation of their hepatic status and all of the biochemical screening studies listed. All screening tests with S & S sums less than 110 (e.g. bilirubin and isocitnc dehydrogenase, are not illustrated c L I L 0 50 GROUP RATINGS Fkh re 4, Correlation between the histologic findings on liver biopsy and each individual's average total vinyl chlonde exposure based on their average CERM (Cumulative Expo sure Rank Months) ratings. Rankings: 1 = lowest possible exposure; 2 - minimal exposure, low levels; 3 - moderate exposure; 4 - worked in areas subject to occasional high excursions, or frequently high and/or had intimate contact. and sum values shown in Figure 3 in the biopsied subpopulation. Here again, 7-glutamyl transpeptidase and ICG clearance (0.5 mg dose) show the greatest sensitivity for identifying individuals with liver disease. However, GGPT had the least specificity, reflecting its high incidence of false positives. Specificity increased with the use of AST, ALT, and ICG clearance. The alkaline phosphatase pro vided the highest specificity, suggesting that mild or low grade chronic hepatic injury due to environ mental agents may be activating hepatic AP syn thesis in the absence of biliary tract obstruction or cholestasis. The ICG clearances, even at the low dose (0.5 mg/kg), clearly remains the test with the best combined sensitive and specific screening study for detection of individuals with subclinical hepatic disease. This subcohort biopsied group was further exam ined on the basis of the histological interpretation of their liver biopsies and their work exposure to vinyl chloride. All biopsied individuals were subdivided into three groups: 19 with histological evidence consistent with chemical liver injury; 30 with histo logical evidence of liver disease, nonchemical liver injury; and 29 with normal liver biopsies. Each of the histological subgroups were further subdivided based on their vinyl chloride exposure, on a scale of 1 to 4 (Fig. 4). The chemical liver injury group contained the highest percentage of individuals with the highest average rating (CERM) for vinyl chloride expo sure. In contrast, with those with liver disease, nonchemical, and those with normal livers have a 120 more even distribution of individuals relative to their degrees of vinyl chloride exposure. In our previous studies we noted that almost ail individuals with histologically specific lesion of vinyl chloride injury or angiosarcoma had a total average CERM rating of 3.5 or greater. The asterisk in Figure 3 indicates the percentage of individuals in each of the three histological groups with exposure ratings of 3.5 or greater. Again, the chemical liver injury group have the highest per centage of individuals with the high exposure ratings. This further supports previous work 4,10) identifying focal hepatocellular hyperplasia as the earliest histological characteristics of chemical injury in liver disease. A study of the frequency with which these tests are positive among those individuals with liver disease, based on their histological findings (chem ical versus nonchemical), provides additional data supporting the clinical observation that an increased AP has a greater specificity for chronic liver injury. Figure 5 shows the ratio of the proportion of positive screening tests in those with histological chemical liver disease divided by the proportion of positive tests in those whose disease is not of chemical origin. All tests, independent of their sensitivity and specificity for liver injury, were more frequently abnormal in the presence of nonchemical, subclinical liver injury, except for AP. tn contrast, AP was far more frequently abnormal in those individuals with chemical liver injury, which tended to be more chronic than acute and generally less severe. Environmental Health Perspectives CMA 003653 TESTS Ficure 5. Frequency with which biochemicaJ tests were abnor mal in those with different type of hepatic injury expresses as a ratio: (CLI) chemical liver injury, (LI)) liver disease, nonchemical. Discussion This preliminary systematic review of the posi tive predictive values and the sensitivity and specificity of federally and some non-federally required tests for chemical workers provides the first scientific and biological basis for the selection of medical screening tests for liver injury in occupa tional environments. Although these commonly used medical tests have been found by clinical experience to be effective as diagnostic tools in the symptomatically ill or hospitalized population, little clinical work has been done to determine their ability to accurately separate biological variations from early latent or underlying disease in asymp tomatic individuals. Tests which provide very high false-positive rates (decreased specificity) such as GGTP, interfere with the screening process identi fication of the high risk worker by the extra time and cost required for repeat testing, the decreased productivity for the employer, the employees' increased anxiety, and by the loss of confidence in the effectiveness of the testing program by both employees and employer. Determination of the sensitivity and specificity of screening studies for asymptomatic individuals is essential if effective recommendations are to be made a federal require ment. This evaluation process also provided the October 1981 best means of developing effective triage protocols for the screening program. For example, in this particular population of industrial workers, we have shown that the assessment of hepatic function is best accomplished by low dose ICG clearance (0.5 mg/kg). The ICG clearance is somewhat a more complicated technique (i.e., injection of substance and repeated blood sampling) but requires only 10 min to perform, and needs only one needle stick. In exchange it provides the best singular screening test for latent hepatic injury. If adequate medical facilities are not easily accessible, then ALT should be substituted. If either ICG clearance and/or ALT studies are found to be abnormal, an AP should be done and a diagnostic work-up instituted to deter mine the etiology {11). The rationale for these recommendations is based on the actual study of chronic subacute chemical injury in an asymptomatic population, not preselected because of signs or symptoms. Therefore the test's ability to correctly differentiate disease from nondisease or one type of injury from another is more accurately determined. Chemical and envi ronmental agents of low toxicity tend to produce repeated or persistent injury which accumulates over time. Tests which measure overall functional capacity quantitatively or semiquantitatively, rather than measuring acute low-grade injury over tim^^ are more likely to detect changes. For this reaso^B clearance or tolerance studies provide the be^^ means for identifying latent hepatic disease, while enzyme studies like ALT, GGPT, and SGOT usu ally reflect acute cellular injury of higher grade or degree and cannot accurately reflex accumulative damage until very late in the disease process. Tests which provide information concerning the progres sion or nonprogression of injury will be far more helpful to the practicing occupational physician. They provide him/her with a better capability to discern between nonoccupational and occupational disease, and the best available reassurance for the worker of his or her safety while allowing the greatest possibility for continued productivity and employment. Finally, these data provide a sound scientific basis upon which to modify federal requirements. The removal of specific testing requirements which, with field experience, prove not to have any significant positive predictive value, or effective sensitivity and specificity will aid in reducing overall cost and help maintain continued compliance by industry and workers. There may be theoretical reasons to maintain or continue some of the present federally required screening studies, but these reasons should be separately identified and not be confused with the purposes of the more effective test in the 121 CMA 003654 I On r- ICO ' .1 I"V3.CmL TV NON CrieM.CAL f-- 80 (u*> o -- 5 > Q 60 40 ;o - AST (SGOT) alt (SGPT) t; 11 GGTP AP TESTS ICG 05 Fi'.t rs 3. Frequency with hich bii)tr.L,TMical u*'t> uhnor- mal in '.hose with .in'ferent type hepatic injury e\|n*c.-.-i> as a ratio-. (CLI) chemical liver injury, liver (tisca.iv, nonchemical. Discussion This preliminary systematic review of the posi tive predictive values and the sensitivity and specificity of federally and some non-federally required tests for chemical workers provides the first scientific and biological basis for the selection of medical screening tests for liver injury in occupa tional environments. Although these commonly used medical tests have been found by clinical experience to be effective as diagnostic tools in the symptomatically ill or hospitalized population, little clinical work has been done to determine their ability to accurately separate biological variations from early latent or underlying disease in asymp tomatic individuals. Tests which provide very high false-positive rates (decreased specificity) such as GGTP, interfere with the screening process identi fication of the high risk worker by the extra time and coat required for repeat testing, the decreased productivity for the employer, the employees' increased anxiety, and by the loss of confidence in the effectiveness of the testing program by both employees and employer. Determination of the sensitivity and specificity of screening studies for asymptomatic individuals is essential if effective recommendations are to be made a federal require ment. This evaluation process also provided the October 1981 I - ,I1 \ j ;* t V.. '.s" * ,, * . ' . have shown that the assessment of hepatic function is best accomplished by low dose ICG clearance oj.-3 mgke). The IC'G clearance is ;omewhat a more complicated technique (i.e.. injection ofsubstar.ee and reoented blood samp",mr but remilr*- s --V \o min to perform, and needs only one needle stick. In exchange it provides the best singular screening test for latent hepatic injur.-. If adequate medical facilities are not easily accessible, then ALT >houid be substituted. If either ICG clearance and or ALT studies are found to be abnormal, an AP should be done and a diagnostic work-up instituted to deter- Tne raLioiiaie lor tnese recoinmeriiicitions ic on the actual study of chronic subacute chemic-! injury in an asymptomatic population, not preselected because of signs or symptoms. Therefore the test's ability to correctly differentiate disease from nondisease or one type of injury from another is more accurately determined. Chemical and envi ronmental agents of low toxicity tend to produce repeated or persistent injury which accumulates over time. Tests which measure overall functional capacity quantitatively or semiquantitatively, rather than measuring acute low-grade injury over time are more Likely to detect changes. For this reason, clearance or tolerance studies provide the best means for identifying latent hepatic disease, while enzyme studies like ALT, GGPT, and SGOT usu ally reflect acute cellular injury of higher grade or degree and cannot accurately reflex accumulative damage until very late in the disease process. Tests which provide information concerning the progres sion or nonprogression of injury will be far more helpful to the practicing occupational physician. They provide him/her with a better capability to discern between nonoccupational and occupational disease, and the best available reassurance for the worker of his or her safety while allowing the greatest possibility for continued productivity and employment. Finally, these data provide a sound scientific basis upon which to modify federal requirements. The removal of specific testing requiren>ents which, with field experience, prove not to have any significant positive predictive value, or effective sensitivity and specificity will aid in reducing overall cost and help maintain continued compliance by industry and workers. There may be theoretical reasons to maintain or continue some of the present federally required screening studies, but these reasons should be separately identified and not be confused with the purposes of the more effective test in the 121 003^55 detection of occupationally related liver injury. The capabilities and limitations of any new tests or old tests for new screening purposes in detection of other potential occupational hazards, should be validated before making them a required screening procedure. This would lay the foundation for sys tematic determination of effectiveness of screening procedure against a proven standard. Unvalidated federal requirements provide the worker and employer with a false sense of security and safety by what is believed to be effective monitoring. More importantly, such a situation can lead to delay in effective correction of cause and disease preven tion. Conclusions Biochemical screening for hepatic injury in asymp tomatic chemical workers can be done most effec tively by the use of liver specific clearance studies. ICG clearance provides the best combination of positive predictive value and sensitivity and specificity for functional hepatic testing at the present time. None of the present federally required studies provide any significant degree of sensitivity with out marked reduction in specificity in asymptomatic individuals. The ALT (SGOT) is the most useful among those Federal tests presently required and the alkaline phosphatase may provide additional specificity as a follow-up study in those individuals with positive ICG or ALT screening studies. There is no evidence that any of the other federal studies add any benefit and strong evidence that they significantly increase the false-positive results in well individuals. All screening studies should undergo evaluation as to their positive predictive value and sensitivity and specificity in asymptomatic individuals before becoming permanent or established requirements. Portions of this work were supported by National Cancer Institute Contract No-l-CN-55212 and Manufacturing Chem ists Association Grant. The authors wish to acknowledge the help and cooperation of Hynson, Westcott & Dunning and B. F. Goodrich Chemical Company. REFERENCES 1. Collen, M. F. Cost effectiveness of multiphasic health testing services. In: Multiphasic Health Testing Services, M. F. Collen, Ed., Wiley, New York, 1978, Chapter 17. Section F, pp. 487-530. 2. Creech, J. L., Makk, L., Whelan, J. G., Jr., and Tamburro, C. H. Hepatotoxicity among polyvinyl chloride production workers during first year of surveillance program. Gastro enterology 67: 786 (1974). 3. Greenberg, R. A., Tamburro, C. H., and Kupchella, E. C. Prospective medical surveillance program for detection and prevention of industrial related cancer. In: Prevention and Detection of Cancer; Part 1, Vol. 2: H. Nieburgs, Ed,, Marcel Dekker, New York, 1978, pp. 1921-28. 4. Tamburro, C. H., Greenberg, R. A., Newby, L. G., and Turns, D. M. Implementation and assessment of a demon stration cancer control detection and prevention program in a cohort of industrial workers. Program Contract #N01CN-55212 Final Report, Division of Cancer Control and Rehabilitation, National Cancer Institute, Bethesda, Mary land, 1978. 5. Greenberg, R. A., and Tamburro, C. H. Exposure indices for epidemiological surveillance of carcinogenic agents in an industrial chemical environment. J. Occup. Med., in press. 6. Tamburro, C. H., Makk, L., and Popper, H. Early hepatic histological alterations among chemical (vinyl monomer) workers. Gastroenterology 77: A33 (November 1979). 7. Fortwengler, P., and Tamburro, C. H. Use of dye clearance in the detection of hepatocellular injury among vinyl chlo ride workers. Clin. Res. 23: 264A (1975). 8. Tamburro, C. H., Creech, J. L, Davis, A., and Greenberg, R. A. Indocyanine green clearance as a prospective indica tor of hepatocellular chemical toxicity. Gastroenterology 75: 989 (1978). 9. Whelan, J. G., Jr., Greenberg, R., and Tamburro, C. H, Radioisotopic scans and gray scale ultrasonography in detection of liver damage. Gastroenterology 79:1129 (1980). 10. Popper, H. and Thomas, L. B. Alterations of liver and spleen among workers exposed to vinyl chloride. Ann. N. Y. Acad. Sci. 246: 172 (1976). 11. Tamburro, C. H. Chemical hepatitis, pathogenesis, detec tion and management. Med. Clin. N. Amer. 63: 545 (1979). 122 Environmental Health Perspectives CMA 003656 Reprinted from CANCER, Vol 40, No, 6, Decem ber 1977 Copyright. 1977, by the American Can cer Society, Inc J B, Lippincott Company, Printed in U S.A, URINARY GLYCOSAMINOGLYCAN PATTERNS IN ANGIOSARCOMA OF THE LIVER Kevin L. Curran, BA, MS, Charles E, Kupchella, PhD, and Carlo H. Tamburro, MD 6 Glycosaminoglycans extracted from 24-hour urine specimens from patients with hepatic angiosarcoma and from normal/controls were separated as cetylpyridinium complexes into "hyaluronic acid," "chondroitin sulfate," and "heparin" fractions, then further separated and characterized by anion-ex change chromatography and hyaluronidase susceptibility. The chromato graphic pattern of the urinary chondroitin sulfate fraction in patients with angiosarcoma of the liver differed from those of controls in that there was a relative increase in the total amount of uronic acid in a hyaluronidase-resistant fraction and a decrease in a fraction susceptible to hyaluronidase digestion. These changes appeared to become more pronounced with advancing disease. Chromatographic patterns and determinations of hyaluronidase susceptibility indicated that the resistant fraction was heparan sulfate and that the susceptible fraction was chondroitin-4-sulfate and/or chondroitin-6-sulfate. Comctr 40:3050-3053, 1977. he emergence of angiosarcoma of the urinary GAG or collagen degradative or syn Tliver and its relationship to vinyl chloride thetic products. Preliminary studies in our labo exposure*'7 has prompted a search for methodrsatory, however, demonstrated an increase in to detect this lesion. Although systematic both liver and urinary GAG in patients with screening programs are currently in opera angiosarcoma, chronic active hepatitis, and cir tion,101" there is still no single chemical in rhosis. 15 Most of the increase in urinary GAG dicator which is specific for angiosarcoma or for occurred in the chondroitin sulfate fraction and. changes which may precede this disease. in contrast to what was found for normal and The association of elevated tissue glyeosarm- other diseases, the urinary chondroitin sulfate noglycans ((IA(i) with tumors, including angio fraction was the only uronic acid positive fraction sarcoma, has been established. `,1, ,* M Clyco- found in the urine of seven of nine cases of vinyl- saminoglycans are also known to be involved in chloride-exposure-associated liver injury other normal connective tissue synthesis and collagen than angiosarcoma. This study was undertaken deposition and are elevated in connective tissue to characterize more completely the urinary disorders.1J Since angiosarcoma of the liver has "chondroitin sulfate" fraction in hepatic angio both neoplasia and fibrogenesis in its etiology17 sarcoma. GAG changes could be expected to serve to sig nal the appearance of early lesions and may he useful in evaluating advanced lesions Galambos" suggested that since the liver con Clinical Summaries Case 1 --(Hepatic Angiosarcoma--advanced) tributes very little to the overall connective tissue of the body, hepatic fibrogenesis should not be expected to result in significant increases in from the University of Louisville, Cancer Center, and the Price Institute for Surgical Research Health Sciences Cen ter. Louisville. Kentucky 40201 I hi. work was supported in part tiv an \rrteni an ('.ineer Not ictv InsMtoMooal (Irani, IN-III, a grant from the It E (ioodru h ( ompnnv and tontrat t N() 1 -(12 with the Nalional ('.inter Institute \tldress for reprints C E Kupt hello. Corner Center. I mversiiv ol I.onisville. I.otnsville, KN 40201 \t t t'pted lor [lulilit atinn April Is PC? ,\ 4(>-year-old white male worked as a chemical helper in a vinyl chloride polymerization plant for thirteen years prior to the diagnosis of angiosarcoma. Twelve years after initial employment, the patient exhibited a persistent elevation of lactic dehy drogenase and underwent angiographic studies which demonstrated multiple areas of scattered tumor stain throughout both lobes of the liver with areas of central translucenry consistent with the diagnosis of angtosart oma of the liver. Exploratory laparotomy and liver biopsy confirmed this diagnosis, and the patient was treated with adriamycin, cyclophosphamide, and methotrexate, an 3050 CMA 003657 \o 6 Glvcosaminogl'.canS is Angiosarcoma Curran et al 3051 initial response was associated with a decrease in the alkaline phosphatase activity, improvement in in docyanine green clearance and an increase in radio isotopic uptake in areas of previously defective up take \fter completion of the chemotherapy course, hepatic lunction deteriorated and the patient under went partial hepatic lobe radiation (total dose of \000 rads) over a two-month period Despite radia tion therapy, the clinical course continued to deterio rate with the development of ascites, peripheral edema, increasing jaundice, hypoalbuminemia, and marked elevations of transaminases and alkaline phosphatase activities This was followed by progres sive hepatic failure, hepatorenal svndrome and he patic coma. Autopsy findings showed extensive in volvement of the liver with angiosarcomatous tissue extending into the diaphragm and metastasis to retro peritoneal and mediastinal lymph nodes, lungs, right adrenal gland and cerebellum The right lobe of the liver demonstrated near elimination of the angiosar coma, presumably due to the radiation treatment. Urinary GAG assavs reported here were made on 24hour urine specimens collected over the two-week period before death (Fig I) Case 2 (Hepatic Angiosarcoma--moderately advanced) A 54-year-old vinyl chloride polymerization worker was first employed as a polymerization vat cleaner 28 years prior to the diagnosis of angiosarcoma. Two years prior to diagnosis, the patient had persistent biochemical liver, abnormalities although he was oth erwise asymptomatic with a normal liver-spleen scan. Angiographic studies showed peltosis hepatis, A liver biopsy revealed focal sinusoidal dilatation, mild t hrontc inflammatory reaction with portal fibrosis, KuptTer cell hyperplasia and dysplasia. Subsequent biopsies demonstrated continued sinusoidal dilitation. atypical and dvsplaslic Kupfler cells with premalignant changes. The peliosis hepatis pattern hecame more pronounced and multiple radioisotopic defects were evident on liver scan. A repeat biopsy one year after initial biochemical abnormality demon strated malignant sinusoidal cells. The patient was treated with a combination of adriamycin, cytoxan, and methotrexate with limited clinical and biochemi cal response. Death was preceded by peripheral edema, ascites, progressive hepatic failure, and coma The GAG analyses reported here were made ID and (i months before death (Fig I, |middle|) Materials and Methods Twenty-four hour urine specimens were col lected from two patients with angiosartoma ol the liver, and from two normal controls. Urine specimens were stored at -7()C! until analysis. C (etylpvridinium chloride (Sigma Chemical Company, St. Louis) was .added to the ctitire 24-hour volume to precipitate the GACs Flo I Elution Patterns of the Urinary Chondroitin Sul fate Fraction The glycovarmnoglycans (GAG) in a 24-hour urine specimen were precipitated with cetylpvndinium chlo ride (CPC) and separated as 0.4 M NaCl soluble ("hyalu ronic acid"), 1 2 M NaCl soluble ("chondroitin sulfate"), and 2 I M NaCl soluble ('`heparin") fractions Each fraction was then subjected to anion-exchange chromatographv Shown here are typical I 2 M (chondroitin sulfate) fraction elution patterns (Advanced = case 1) according to the method of DiFerrante.1 The hyaluronic acid, chondroitin sulfate, and hepa rin fractions were eluted individually according to the method of Schiller et al. '*Cetylpyridinium chloride was removed14 and the GAGs were sub jected to amon-exchange chromatography as de scribed by Schiller et al." Glycosaminoglycan fractions were applied to 1.0 X 44 cm AG1-X2 (200-400 mesh, chloride form) columns Bio Rad Laboratories, Richmond, California) and eluted stepwise with 0 0, 0.5, 1 0, 1,25, 1.50, 2.0, and ,3 0 M NaCl. At a flow rate of 1.0 ml/min, approximately sixteen 10,3 ml fractions of each molar strength of NaCl were collected and a sample of each fraction was analyzed for urontc acid by the method of Bitter and Muir.3 Stan dards of heparin (Nutritional Biochemical Com pany), chondroitin sulfate (Sigma Chemical Company), and hyaluronic acid (Nutritional Biochemical Company) were also evaluated by ion exchange chromatographv. The uronic-acid-positive fractions within cat. h individual salt fraction were pooled, dialyzed to remove salt, and concentrated. The fractions eluted bv I 25 or 1.50 M NaCl were tested for CMA 003658 3052 Cancer December 1977 Voi M) T \fl[ f. I Source Raiiu of Total L'romc \ud Kiuted in \ 25 \t/ 1 5 M Nad Normal Normal \ngiosarroma. case 2. pre-chemotherapy1 \ngiosarroma. < ase 2, post ( hcmothcrapv* Angiosarcoma. ta^r I, advanced 0^4 0 5 I (> 084} 0 ()7 \ 5000 Mmi The give osarmnoglvcans (GAG) in a 2'Miour urine specimen were precipitated with cetylpyndmium chloride (CTC) and separated as 0 4 \1 NaCI soluble ("hvaluromc acd' T i 2 St NaCI soluble (`Nhondromn sulfate"), and 2 1 M NaCI soluble ("heparin") fractiuns The CPC was removed from the I 2 M NaCl-CPCsolubtlized fraction and the GACs further purified by amon-exchange chromatography The total amount of GAG in the resulting 1 25 M and 1 50 M NaCI columneluted fractions was determined and the ratio of the two fractions was calculated ('One day prior to begin ning of chemotherapy, *Two days following chemotherapy initiation.) acid and heparin fractions revealed no qualita tive differences between controls and angiosar coma patients. The anion exchange column pat terns of the 1.2 M NaCI solubilized GAGs are shown in Fig 1. Chromatography of the urinary "chondroitin sulfate" fractions of patients with angiosarcoma yielded a comparatively large, uronic-acid positive peak in 1 25 M NaCI. The ratios of the total amount of uronic acid-positive material eluted with 1.25 VI NaCI to the total amount eluted with 1.50 M NaCI are given in Table 1. The susceptibility of the GAGs eluted with 1.25 or 1 50 VI NaCI to hyaluronidase degrada tion is given in Table 2. The GAG eluted with 1.25 M NaCI was resistant to hyaluronidase, the enzyme producing only a 43% reduction in tur bidity. The 1,50 M NaCl-eluted GAG fraction was 100% susceptible to hyaluronidase degrada tion. Discussion and Conclusion susceptibility to testicular hyaluronidase (Nutritional Biochemical Company) using a modifica tion of the cetyltrimethylammonium-bromide, turbidimetric assay described by DiFerrantc." Ri-.stius The major GAG fraction observed in all urines--both from normal controls or from pa tients with angiosarcoma--was the fraction so lubilized by 1.2 M NaCl/1% cetylpyndinium chloride (the "chondroitin sulfate" fraction). Anion exchange chromatography of hyaluronic Table 2, Hyaluronidase Susceptibility Source llepolymerizalion %' Heparin, standard Hvaluronic acid, standard Chondroitin sulfate, standard \ 25 M NaCI column-cluate. pooled frac tions from angiox.irrom.ttom patients 1 51) \| N,( :l coliimn-du'tie. pooled fractions from angioawrc umatous patients 1 5() M NaCI column-eluate. normal 50 in i <)7 <i 4^ S MX) 0 too o 1 Glvcosaminoglvcans isolated from urine were tested lor hyaluronidase susceptibility by measuring change's m turbidity developed with the addition of < ctvltnmrthviammonium bromide following incubation with hv.duronid.or Normal controls exhibited only minor amounts of I 2^ M NaCI column-eluted GAG and consequently do not ap pear in this tabic The chromatographic pattern found here for controls conforms to urinary glycosaminoglycan distributions reported by others.11,11 These pat terns suggest that there was a relative increase in urinary heparan sulfate and a decrease in chondroitin-4- and/or -6-sulfate in patients with hepatic angiosarcoma. This interpretation agrees with the Dowex I-X2 chromatographic patterns reported by Kao and Leslie11 and by others 1'1* Heparan sulfate is reported to be partially susceptible to hyaluronidase digestion,10 and this correlates well with the observed 43% diges tion of the GAG in our 1.25 M NaCI fraction. Since heparan sulfate has been shown to be associated with blood vessels,10 an increase in the urinary excretion of this GAG is not surpris ing in this vascular lesion. Also, chondroitin-4and -6 sulfates are reportedly eluted from Do wex 1-X2 columns with 1.50 M NaC^ ^and are susceptihle to hyaluronidase" suggesting that our 1 5 \| fraction is chordroitin-4- and/or i hondromn-6-sulfate. Assuming that urinary GAG patterns de scribed here are reflections of hepatic changes, it will be important to determine what processes these changes reflect, i.f., those of neoplastic growth, fibrogenesis, or cell death. In this re gard, it should be noted that 1) the ratio of heparan sulfate to chondroitin sulfate reported here for angiosarcomatous urine is similar to that reported for cirrhotic human Uver tissue by Betker,1 and 21 the shift from a hvaluronidase- CMA 003659 3052 \ ; Ntomal c^'jC 2. prc-< hemmhr r.ipv1 \t i .f 2 ;m>>.t ncmoUAcr.ipv * \nms.irroma, c,i<r I .idv.mcrd \uti- I'hr ^l\c uv.iminngivf in-* (. \( >n ,i Jt.'-.mic urine ^pectmcn were precipitated with i efvlpvndinmm 1 hiurtde COC, md -separated a-i `i4 M N,t( ! voluble ' hvaluronic and I 2 M \ i( I '"!uble "c-hnndroitin suifate" and 2 ! M NaC I iotob!e heparin " fraenunv ,-pr i0% -- ~ h* r ^ : '[ \jt 1j>( . oi'jPMzed irac'ion am -nr 1 r \<,- panned bv tmon-eychange '-hromatoiiraphv The total amount of (- \C in the resulting 1 23 M and i M ViC.l coiumn- eluted fractions was determined and the ratio of the two fractions was calculated 'One dav prior to begin ning of chemotherapy 2Two dnvs following chemotherapy initiation terns of the ! 2 M NaCI solubilized G-\Gs are shown m Fig I, Chromatographs' of the urinarv "chondroitin sulfate" fractions of patients with angiosarcoma yielded a comparatively large, ironic-.0 id positive peak .c. i 25 M Vj( ! The ratios ot the total amount of urontc acid-positive material eluted with 1 25 \| NaCI to the total amount eluted with 1 50 M NaC! are ji-.en n Table I 1 he susceptibility of the GAGs eluted with I 2s or I 50 M NaCI to hvaluronidase degrada tion is given in Table 2. The GAG eiuted with I 2' \ I N',i C] 'Vi- nft';f[nr -o L, '' ' er/vne producing sr.iy a -13% reduction ;.n tur- biditv The 1,50 \[ NaCI-eluted GAG fraction was 100% susceptible to hyaluromdase degrada tion Disc.rssioN \so Conclusion susceptibility to testicular hvaluronidase (Nutri tional Biochemical Company) using a modifica tion of the cetyltrimethylammonium-bromide, turbidimetric assay described by DiFerrantc.* Rt-.suus The major GAG fraction observed in all urines--both from normal controls or from pa tients with angiosarcoma--was the Fraction so lubilized by 1 2 M N'aCl/1% cetylpyridinium chloride (the `chondroitin sulfate'' fraction). Anion exchange chromatography of hyaluronic Tal* 2 Hyaluromdase Susceptibility Source Depolvmenzauon V Heparin, standard HvaJuromc acid, standard Ohondrottin sulfate, standard 1 25 M NaCI column-eluate, pooled fractions from singtonarromatom patients 1 30 M NaCI column-eluatc. pooled fractions from singiosarromatnus patterns 1 30 VI NaCI column-eluate, normal 30 m 07 6 4y 3 1000 100 0 1 (ilycosaminoglveani isolated from urine were icsied tor hyaluromdase susceptibility by measuring changes in turbidity developed with the addition of i-eivUrimcihvlammomum bromide following incubation with h.hun-nid.iMNormal controls exhibited onlv minor aniounis of I 2S M NaCI cnlumn-cluted CAC and rim.sequenilv ito noi ap pear in this table The chromatographic pattern found here for controls conforms to urinary glvcosaminoglycan distributions reported by others.1121 These pat terns suggest that there was a relative increase tn urinary heparan sulfate and a decrease tn rhonrlroitin-4- and/or -6-sulfate in patients with hepatic angiosarcoma. This interpretation agrees with the Dowex I-X2 chromatographic patterns reported by Kao and Leslie12 and bv others 21* Heparan sulfate is reported to be partially susceptible to hvaluronidase digestion,20 and this correlates well with the observed 43% diges tion of the GAG in our 1.25 M XaCi fraction. Since heparan sulfate has been shown to be associated with blood vessels,20 an increase in the urinary excretion-of this GAG is not surpris ing in this vascular lesion. Also, chondroiun-4and -6 sulfates are reportedly eluted from Do wex 1-X2 columns with 1.50.VI NaCI21'and are susrepuhir to hvaluronidase20 suggesting that our 1,5 M fraction is t hordroitin-4- and/or rhondroitin-b-sulfate. Assuming that urinary GAG patterns de scribed here are reflections of hepatic changes, it will be important to determine what processes these changes reflect, i.e., those of neoplastic growth, fibrogenesis, or ceil death. In this re gard. it should be noted that 1) the ratio of heparan- sulfate to chondroitin sulfate reported here for angtosarcomatous urine is similar to that reported for cirrhotic human liver tissue by Becker.2 and 2) the shift from a hvaluronidasr- CMA 003660 No 6 Glycosaminoglycans in Angiosarcoma Curran et al. 3053 susceptible to a hvaluronidase-reststant G AG is consistent with the suggestion by Hutterer and Rubin" that the stabilization of collagen de pends on a shift to a hvaluronidase-resistant (iAG envelope surrounding the collagen bundle Although Hutterer and Rubin attribute this to an augmentation of dermatan sulfate, Becker2 reported that the GAG pattern in human <ir- rhosis was characterized by the augmentation of dermatan sulfate and heparan sulfate If the ob served changes in urinary GAG are reflective of vinvl chloride-exposure-associated fibrosis, the fact that fibrosis is a precursor of angiosarcoma " inditaios that the observations reported here ( onstttute a promising lead in early detection of vinyl-chloride-induced liver disease. Ki ll i<l M l.s 1 \nghilrn, I j Metabolism of at id muc opolvsac charides in hepatoma and norm.il liver Onc/'ioq\ 30 3()4-3l" 1')'4 2 Becker, K \nd mucopolvsarcharides m experimental .ind human cirrhosis In Collagen Metabolism in the I,nor M Popper and K Becker F.ds New York, Stratton Inter continental Medical Book Corporation, lc'72, pp 4S s2 3 Bitter, i , and Muir, H A modified uronic a< id c arha/ole reac I ion Anal Bwthmt 4 A AO- A14, |%2 4 ( reec h, | I , and Johnson, M N \ngiosarcoma or die Iiut in the rn,inuta< tore of poiwmvl c hlonde 7 (inupfb hntut! Mill If) 1 si t | 3 | , | l)'f4 ^ Mil'tinihlr \ I lie niGiMici'iiii'ni ol mio.iiv iiinin| mi) V s.h t toil ides \nul lUixhrm 21 9K 106. I067 fi t )i) c'jTAMie N I urbidinietnc mcMMiirmt'iu ol <ic ul muc opolw.u c harides and hv.iluromd.ise activnv 7 ( hmn 22*1 303- 306, |63o, la Ik, 11 , ("reet h, | 1, , I le.nh, I) VV , Johnson, M N , and Key. M M Hepatic disease among workers at a vinvl chloride polymerization plant 7-bW,-l 730 36-63, l(D4 8 Calambos, J T Connective tissue metabolism and cirrhosis In Collagen Metabolism in the I-iver, II Popper and K Becker, Eds New N'ork, Stratton Intercontinental Medic a! Book Corporation. 1973, pp 37 6! <> tiasic, C , and Caste T Removal of static aeid from the i ell coat in tumor < ells and vas< ular endoihelium and ns cIIcmn m metastasis /Vr* Sail. \md ,Sn / S 1 1H i P2 11 "7"\ f%2 10 Crccnbeni R \, I amburro. (' 11 , and Kupchella, C f', A prospective medical surveillance program for the deteciion and prevention of occ upai urnallv-related catuet hi Prevention and Detection of Cancer, II K Nioburgv Editor, Part I, Volume 2. Marcel Dekker. Inc . NV (In prrss) 1 1 Hutierer f*` , and Rubin, E, Mueopolysacc handes in reversible and irreversible experimental hepatic fibrosis In ( ollatjrn Metabolism tn ihe I aver. II Popper and K Bet kcr l.ds New N ork. Stratton Intercontinental Medical Book Corporation !O-'A, pp aU 12 Kao K 3 T . and Leslie. J C Micro fractionation ,md determination of urinary glvrosaminoglvcans Buxhem \l*f 'Mp A20, l<T4 1 3 Koizumi, ( Nakamura, N , and \br, H Changes m and muc opolvsac eharide in the liver in hepatic fibrosis link him IhuphM ,-h/u. 148^49-736, 1067 14 Korn, K 1) Isolation of heparin from mouse mast ' ell minor / Hml hhm. 234 1323-1329 1939 I ^ Kupc Bella, C K . and I amburro, C H Crtnary and iiNvnr u;K< ON.immoglvc an patterns in hepatic angiosarcoma hi Pirvention and Detection of Canter, II l\, Nu-burgs, laluof, Part 1, Volume 1, Marcel Dekker, Inc, \Y (In press) U Makk, 1 , Creech, J I, Whelan, j (I , and Johnson, \l N laver damage and anginsareoma in vinyl chloride winkers \ systematic detec i ion program, JA XI 4 2 A<) 64-68, 1974 r Popper, II, and Thomas, L B., Alterations of liver and spleen among workers exposed to vinyl chloride Ann. W had S'(( 246 172-194, I973 18 Rich. C , and Mvcrs. W P E Excretion of acid muc npolysaccharidrs m the urine of patients with malignant neoplastic diseases J. hah and (dm ited, 34 223-228 1939 16 Schiller, S, Slover, (. \. and Dorfman. \ A method for the separation of acid mucopolysaccharides Its applic anon to the isolation of heparin from the skin of rats J Hud (hrtn 2Vi 983-9H7 I%| 2** Sharon, N, Complex Carbohydrates, I heir Ohemistiv Biosynthesis. and Functions Reading. Massachusetts, \dclison*Wes|ev Publishing ( ompanv. |973 21 \ aradi D P , Cifonelli, [ A . and Dorfman, A Ihe a< id muc opolysaci handes m normal urine Biochim Bwphys, 1 ,ta |4| 103-1 17 1667 22 Yamamoto, K , and Terayama, H Comparison of cell coat acid mucopolysaccharides of normal liver and vari ous ascites hepatoma cells ('.amer Hti 33 2237-2264 1973 CMA 003661 Prevention and Detection of Cancer PARTI. PREVENTION Volume 1. Etiology Edited by Herbert E. Nieburgs Mount Sinai School of Medicine of The City University of New York New York, New York COPYRIGHT 1977 by MARCEL DEKKER, INC. MARCEL DEKKER, INC. New York and Basel CMA 003662 URINARY AND 1 ISSUE GLYCOSAMINOGLYCAN PATTERNS IN HEPATIC ANGIOSARCOMA Charles E. Kupchella and Carlo H. Tamburro Cancer Center and Department of Medicine University of Louisville School of Medicine Louisville, Kentucky 40201 I. INTRODUCTION The recent discovery of a relationship between vinyl chloride and angiosarcoma of the liver has received much attention (1-3). Although there are now systematic detection programs for vinyl chloride workers (3,4), there is as yet no specific chemical abnormality that serves as a good indicator of early, vinyl-chloride-induced liver injury and angio sarcoma. Alpha feto-protein has been a relatively valuable serological marker for hepatocellular carcinoma (5), but is has not as yet proven use ful in the detection of angiosarcoma (6). New leads are needed if more specific tests are to be developed for angiosarcoma. The literature suggests that the glycosaminoglycans in the urine and/or blood should be evaluated as a possible aid in early detection. The production of sulfated glycosaminoglycans is characteristic of malig nant vascular tumors of the skin and some pathologists use this feature as a diagnostic aid (7). Barr and Bonin (8) Observed a strong positive alcian-blue, glycosaminoglycan staining reaction in human angiosarcoma tissue and suggested than an attempt be made to qualitate and quantitate the production of glycosaminoglycans in the neoplasms, serum, and urine of those at risk. They pointed out that the urinary glycosaminoglycans may have diagnostic significance in angiosarcoma and, if so, a glycosami noglycan spot test might easily be employed as a gross screening test of vinyl chloride production workers. A number of other observations place the glycosaminoglycans in a relevant position with regard to angiosarcoma. Angiosarcoma is accom panied by connective tissue abnormalities (2,9) and changes in tissue, urinary, and blood glycosaminoglycans have been found to occur in many connective-tissue disorders -- including connective tissue disorders of the liver (10-14) -- as well as in hepatic cancer (15-17). Supported in part by grants from the B. F. Goodrich Company and the American Cancer Society (IN-111) and a contract with the National Cancer Institute (NOl-CN-55212). 9 IS CMA 003663 The purpose of this study was to make a preliminary determination of the glycosaminoglycan oatterns in tissue and urine associated with angiosarcoma of the liver and with vinyl-chloride-induced liver injury other than angiosarcoma and to compare these patterns with those in normal controls and those associated with other liver disease. Our goal was to evaluate the use of glycosaminoglycan patterns in the early detec tion of vinyl-chloride-induced liver injury and angiosarcoma and to explore the role of the glycosaminoglycans in the etiology of vinyl chloride injury. II. PROCEDURES AND MATERIALS USED Urine specimens were collected as occasional samples from: 9 normal controls; 9 individuals with histories of occupational exposure to vinyl chloride and having abnormal, liver, biochemical studies; 6 with "other" cancers prior to surgery; 3 with angiosarcoma; 8 with active viral hepatitis; 6 with cirrhosis; 2 with lung-liver metastases; and 4 with metabolic disorders of the liver (congenital and indirect hyperbiliru binemia) . In one case of angiosarcoma. 24-hr urines were collected on alter nate days beginning 2 weeks prior to death. Urine samples were collected without preservative and frozen at -76 until analysis. Specimens were divided into two 25 ml samples and one 5 ml sample. Urinary creatinine was measured on the 5 ml sample using a Technicon Autoanalyzer. The degree of urinary glycosaminoglycan polymerization was estimated by dialyzing one 25 ml sample for 24 hours in tap water; the sample was then treated identically to an undialyzed sample by the method of DiFerrante (18) using cetylpyridiniurn chloride as a precipitant. After resolubilization of the glycosaminoglycans in water, duplicate samples were assayed for total uronic acid by the modi fied carbozole reaction of Bitter and Muir (19). The remaining glycosaminoglycans were reprecipitated with cetylpyridinium chloride and separated into the wash, hyaluronic acid, chondroitin sulfate, and heparin fractions as described by Schiller et. al.(20). Each of the fractions was assayed for uronic acid ( ug per mg of creatinine). Autopsy tissue was obtained in 2 cases of hepatic angiosarcoma (tumor tissue and non-tumor tissue adjacent to tumor), 2 cases of cirrhosis, and in 3 control cases (gun-shot wound victims without liver pathology) and analyzed for glycosaminoglycans by a previously reported modification (21) of the method of Schiller et. al. (20). Uronic acid was determined in each of the hyaluronic acid, chondroitin sulfate, and heparin fractions. Pieces of tissue were subjected to alcian-blue-periodic-acidSchiff staining with and without hyaluronidase and diastase pretreatment. These procedures were carried out according to the methods described by Mowry (22). Ascitic fluid was also obtained at autopsy in one case of angio sarcoma and analyzed for glycosaminoglycans. The fluid was centrifuged 916 CMA 003664 Tne purpose or - s ssu'Jy was to mane a preliminary determination of tne glycosaminoglycan patterns in fssue and urine associated with angiosarcoma or tne i<er and with vinyl-cnloriae-induced liver injury otner tnan ang1 os artpma and to compare these patterns with those in "s' u! -r. v^ - -Ssj2 associated with os oar 1 . r disease, uai1 goal was to evaluate tne use of glycosaminoglycan patterns in the early detec tion of vinyl-chloride*!nduced liver injury and angiosarcoma and to explore the role ;f the glycosaminoglycans in the etiology of vinyl cnloride injury. I! PROCEDURES AND MATERIALS USED urine specimens were collected as occasional samples from: 9 no-mal controls; } '-c1. - duals with 'mstor'es of occupational exposure to vinyl chloride and saving aonorma), liver, biochemical studies; 6 with "other1 cancers prior to surgery; 3 with angiosarcoma; 3 with active viral hepatitis; 6 with cirrhosis; 2 with lung-liver metastases; and 4 with metabolic disorders of the liver (congenital and indirect hyperbiliru binemia) . In one case of angiosarcoma, 24-hr urines were collected on alter nate days beginning 2 weeks prior to death. Urine samples were collected without preservative and frozen at -76 until analysis. Specimens were divided into two 25 ml samples and one 5 ml sample. Urinary creatinine was measured on the 5 ml sample using a Technicon Autoanalyzer. The degree of urinary glycosaminoglycan polymerization was estimated by dialyzing one 25 ml sample for 24 hours in tap water, the sample was then treated identically to an undialyzed sample by the method of DiFerrante (18) using cetylpyridinium chloride as a precipitant. After resolubilization of the glycosaminoglycans in water, duplicate samples were assayed for total uronic acid by the modi fied carbozole reaction of Bitter and Muir (19). The remaining glyco saminoglycans were reprecipitated with cetylpyridinium chloride and separated into the wash, hyaluronic acid, chondroitin sulfate, and heparin fractions as described by Schiller et. al.(20). Each of the fractions was assayed for uronic acid ( ug per mg of creatinine). Autopsy tissue was obtained in 2 cases of hepatic angiosarcoma (tumor tissue and non-tumor tissue adjacent to tumor), 2 cases of cirrhosis, and in 3 control cases (gun-shot wound victims without liver pathology) and analyzed for glycosaminoglycans by a previously reported modification (21) of the method of Schiller et. al. (20). Uronic acid was determined in each of the hyaluronic acid, chondroitin sulfate, and heparin fractions. Pieces of tissue were subjected to alcian-blue-periodic-acidSchiff staining with and without hyaluronidase and diastase pretreatment. These procedures were carried out according to the methods described by Mowry (22). Ascitic fluid was also obtained at autopsy in one case of angio sarcoma and analyzed for glycosaminoglycans- The fluid was centrifuged 916 CMA 003665 and the sediment analyzed as tissue above. The supernatant was treated by the method for urine described above. III. RESULTS A sunmary of the urinary glycosaminoglycan measurement is given in Table I. Normal controls had the least urinary glycosaminoglycans (measured as uronic acid) of all groups. All other groups showed some elevation. The levels in angiosarcoma, hepatitis, cirrhosis, and liver metastases were significantly elevated (P < .05) over normal controls. The cirrhotic group exhibited the greatest variance in urinary glycosami noglycans. No significant differences were found in urinary creatinine levels between groups. There were no significant differences between groups in either the percentage of the total glycosaminoglycans that was dialyzable (Table I) or in the percentage of the unfractionated total that appeared in the hyaluronic acid, chondroitin sulfate, and heparin fractions. Seven of 9 vinyl-chloride-exposed individuals other than those with angiosarcoma had positive chondroitin sulfate fractions with nega tive hyaluronic acid and heparin fractions. This was true in only 3 of 32 other urines evaluated in this same manner. The pattern of daily glycosaminoglycan excretion prior to death due to angiosarcoma in one individual is given in Figure 1. Total tissue glycosaminoglycan levels for angiosarcoma tumors, fibrotic tissue adjacent to tumors, cirrhotic liver tissue and normal liver tissue are shown in Figure 2. Fractional hyaluronic acid, chon droitin sulfate, and heparin levels are given in Figure 3. Histochemically, angiosarcomatous tissue exhibited a strong alcianblue positive staining reaction. Alcian-blue staining was only slightly less in "non-tumor" tissue adjacent to tumor masses. The staining reaction in tissue from normal liver was very weak and only slightly stronger in cirrhotic liver tissue. The strong alcian-blue reaction in angiosarcomatous tissue did not occur if sections were pretreated with hyaluronidase. Ascitic fluid sediment was uronic-acid-positive in only the hyaluronic acid fraction -- 112 ug uronic acid per gram of dry, defatted sediment; ascitic fluid supernatant contained 1.7, 1.2, and 0.2 ,ug uronic acid per ml in the hyaluronic acid, chondroitin sulfate, and heparin fractions, respectively.IV. IV. DISCUSSION The literature indicates that normal male creatinine excretion is 1.5 g per 24 hrs (23). Thus, our normal mean (Table I) of 3.2 t ,4 ug cetylpyridinium chloride-precipitable uronic acid per mg creatinine falls in the middle of the normal ranges reported by Varma et. al. (24), 917 CMA 003666 2.6 - 4.7 ^g/mg; OiFerrante and Rich (25), 2.9 - 4.8 yg/mg; and Kao and Leslie (26), 1.8 - 4,9 ug/mg. Although our study was not controlled for age, Goldberg and Cotlier (27) have shown that urinary glycosaminoglycan excretion is constant from ages 20-70. Manley et. al. (28) have shown that the proportion of urinary glycosaminoglycans in the chondroitin sulfate fraction is constant from ages 20-70. Manley et. al. also reported that the chondroitm sulfate fraction is highest at birth and that it gradually drops until age 20, suggesting that urinary chondroitin sulfate reflects tissue growth. The fact that we found no differences between groups in the creatinine concentration is significant in that it indicates that occa sional samples do reflect 24-hour excretion when normalized to creatinine. Precedent for expressing glycosaminoglycan measurements as a function of creatinine content in occasional urine samples has been established by DiFerrante and Rich (25) and Pennock (29). Manley et. al. (28) have shown that the creatinine/uronic acid ratio is steady from ages 20-70. Our results indicate that the liver diseases evaluated are accom panied by elevated urinary glycosaminoglycan excretion. Our tissue data suggests that this reflects liver-tissue glycosaminoglycan elevation and conforms to the reports by others that both hepatic connective tissue disorders (10-14) and hepatic cancer (15) result in increased hepatic glycosaminoglycan levels. It may be significant that the angiosarcoma patients had half the urinary glycosaminoglycan excretion of patients with liver metastases and that our analysis of angiosarcomatous tumor tissue exhibited half the glycosaminoglycan content reported by Kojima et. al. (15) for hepatocellular carcinoma. While our data suggest that liver disease results in a decrease in the proportion of highly polymerized glycosaminoglycans, variance was large within each group and none of the differences between groups were statistically significant. Although we have not completed the characterization of isolated glycosaminoglycan fractions, our data indicate: 1) that the chondroitin sulfates are the primary urinary glycosaminoglycans in both normal controls and in disease states; 2) that the chondroitin sulfates and heparin are the dominant glycosaminoglycans in normal and cirrhotic livers (Figure 3). Chondroitin sulfate is elevated in the fibrotic, non tumor, portions of angiosarcomatous livers while heparin is the predomi nant glycosaminoglycan in tumor tissue. Hyaluronic acid is also apparent ly elevated relative to chondroitin sulfate in angiosarcomatous tumors (Figure 3); and 3) that hyaluronic acid is the sole glycosaminoglycan in ascites fluid sediment. These gualitative data are in general agreement with those reported by others. Goldberg and Cotlier (27), Douglas et. al. (30), and Varma et. al. (24) have reported that the chondroitin sulfates are the predominant urinary glycosaminoglycans. Varma et. al. reported that 2/3 of urinary glycosaminoglycans are chondroitin-4-and chondroitin-6-sulfate and this agrees with our data on normal controls and on those with liver disease. Kojima et. al. (15) reported that in hepatocellular carcinoma 918 CMA 003667 -.c'cteerg and _otl ier on 15 constant from r-pcrt zn te~ jrinarv glycosaminoglycans m '.ie :r:ror9itin i,'r'j:s -`r-jction is constant from ige: 21- 1 '-'Sr'?/ et j'; p -^o *1 fraction is highest at oirth ana tna gracually drops until age 20 suggesting that ur-nary cnondroitin sulfate refects tissue growth. The fact "rat we *te-no "o ai ffe-erces between groups in toe creatinine concentration is significant ;n t.nat it indicates mat occa sional samples to reject 24-rour excretion -when normal iced to creatinine Precedent for expressing j1ycosaminog1vcan measurements as a function of creatinine content in occasional :jnir? sa'des nas teen estate1'snea by 'Cr r3 Z 'r**3 Cur results mo'cate tnat me 11 (sr diseases evaluated are acccm- panied Sy e'evated urinary g;/cos am'nog'/can excretion, Our tissue data suggests that this refects liver.tissue giycosaminoglycan elevation and conforms to the reports by others that teeth nepatic connective tissue disorders (10-id) jrp "iji;-; ttr.ce- IS' 'au' i in Tcea'iO '"epatic giycosaminoglycan levels it tee significant that the ang'osarconia patients had half tie urinary I'ycossminog i/can er.'et'on of patients with liver metastases ahd tnat our aralysis of angiosarcomatous tumor tissue exniteited naif the giycosaminoglycan content reported by Xojima et. al. (15) for hepatocellular carcinoma. While our data suggest that liver disease results in a decrease in the proportion of highly polymerized glycosaminoglyeans, variance was large within each group and none of the differences between groups were statistically significant Although we have not completed the character!zation of isolated giycosaminoglycan fractions, our data indicate: 1) that the cnondroitin sulfates are the primary urinary glycosaminoglycans in both normal controls and in disease states; 2) that the chondroitin sulfates and heparin are the dominant glycosaminoglycans in normal and cirrhotic livers (Figure 3). Chondroitin sulfate-is elevated in the fibrotic, nontumor, portions of angiosarcomatous livers while heparin is the predomi nant giycosaminoglycan in tumor tissue. Hyaluronic acid is also apparent ly elevated relative to chondroitin sulfate in angiosarcomatous tumors (Figure 3); and 3) that hyaluronic acid is the sole giycosaminoglycan in ascites fluid sediment. These qualitative data are in general agreement with those reported by others. Goldberg and Cotlier (27), Qouglas et. al. (30), and Varma et. al. (24) have reported that the chondroitin sulfates are the predominant urinary glycosaminoglycans. Varma et. al. reported that 2/3 of urinary glycosaminoglycans are chondroitin-4-and chondroitin-6-sulfate and this agrees with our data on normal controls and on those with liver disease. Kojima et. al. (15) reported that in hepatocellular carcinoma 018 CMA 003668 tissue, chondroitin sulfates and hyaluronic acid were increased 33 and 10 times, respectively, over amounts found in healthy livers; the heparin and heparan sulfate proportions dropped. This contrasts with our data on angiosarcoma tissue, i.e. heparin and hyaluronic acid increased 5 and 10 times, respectively, over normal tissue; chondroitin sulfate levels rose but fell in proportion to other glycosaminoglycans. Galambos and Shapira (10) reported that the chondroitin sulfates are dominant in normal livers and in hepatic fibrosis, but Kojima et. al. (15) report that chondroitinase-resistant and hyaluronidase-resistant glycosaminogly cans are dominant. Kuroda et. al. (31) also reported that heparan sulfate is the dominant glycosaminoglycan in the normal liver. Our histochemical observation that nearly all of the increased alcian-blue positive material in angiosarcomatous livers was susceptible to hyaluronidase digestion suggests that the observed chondroitin sulfate elevation is due to-chondroi tin-4- and/or chondroi tin-6-sul fate. The increases in liver and urinary glycosaminoglycans may well reflect an important role of these substances in the process of fibrogenesis and in tumor growth, Galambos and Shapira (10) reported that hyaluronic acid was elevated during hepatic fibrogenesis. If a similar fibrotic process is operative in angiosarcoma, it may be that the ob served tumor-tissue heparin increase is reflective of tumor growth. We did observe a four- to six-fold greater heparin level in tumor tissue than in adjacent, non-tumor tissue. The observation that the chondroitin sulfates tend to be the exclusive uronic-acid-positive constituents in the urine of individuals is paradoxical in that those glycosaminoglycan fractions that are most elevated in angiosarcomatous tissue are those that are absent from the urine of individuals who may well have early, vinyl-chloride-induced liver injury. This pattern may be due to the selective action of lyso somal, glycolytic enzymes in the liver and/or may reflect the role of the chondroitin sulfates in early fibrotic changes in the liver. Certainly the potential usefulness of this pattern in early detection warrants the more complete evaluation now ongoing in our laboratory.V. V. SUMMARY Glycosaminoglycans were measured in urine and tissue of patients with hepatic fibrosis and hepatic cancer including vinyl-chlorideexposure-associated liver injury and angiosarcoma. Angiosarcoma, hepa titis, cirrhosis, and liver metastatic patients exhibited significantly elevated glycosaminoglycan excretion. Angiosarcoma tissue exhibited elevated glycosaminoglycan levels with the greatest increases in the heparih fraction. Histochemically, angiosarcomatous tissue gave a strong alcian-blue staining reaction which could be prevented by pretreatment with hyaluronidase. Although vinyl-chloride-exposure-associated liver injury other than angiosarcoma was not accompanied by a significantly elevated glycosaminoglycan excretion, this condition tended to be associated with a urinary glycosaminoglycan excretion pattern in which the chondroitin sulfate fraction was the only uronic-acid-positive fraction. 9 u> CMA 003669 TABLE I, Urinary Glycosaminoglycan Levels in ug Uronic Acid per mg Creatinine by Liver Diseases Category Patient qroup normal control vinyl chloride exposed Cases 9 9 ug uronic acid per mg creatinine 1 1 S.E.) 3.2 * .4 4.1 .4 % uronic acid not dialyzable (* 1 S.E.) 65 5 39 6 other cancer 6 4.5 1.5 39 +- 6 other liver disease 4 5.1 0.8 37 T 13 angiosarcoma hepatitis 3 8 7.6 1.6 8.5 1.8 41 t 22 52 t 14 cirrhosis 6 12. 7 i 3 53 9 1iver metastasis 2 13. 8 - .9 42 920 CMA 003670 r IC - T ^ j' ; `_c. a's Creatinine 3y Live1- C'ses-es Ca 'arpx / ,,a '^rzr.-c n C 1 J p 0 f' .1 9 Patient qrouD normal control vnyl chloride exposed Cases 9 9 og uromc acid per mg creatinine ___ (11 S.E.) 3.2 * A 4.1 i .4 4 uronic acid not dialytable ( 1 S.E.) 55 t 5 39 * 6 other cancer Other liver disease 6 4 4.5 - 1.5 5.1 t 0.8 39 6 37 13 angiosarcoma hepatitis cirrhosis 3 8 6 7.6 * 1.6 8.5 i 1.8 12.7 3 41 t 22 52 i 14 53 t 9 1iver metastasis 2 13.8 - .9 42 920 CMA 0036^^ DAYS PRIOR TO LIVER DEATH FIG. 1. Urinary glycosaainglycan output in one angiosarcoma patient during the 16-day period prior to death. 921 CMA 003672 TOTAL GIYCOSAMINOGIYCANS in MICROGRAMS URONi C laoo r ijlOOO - i o o l>> or -J U> Tumor Case I Non- rumor Adjacent Cose I Tumor Cose 2 Non-iumor Adjocent Cose 2 Cose 3 Cose 4 Case 5 Cose 6 FIG. 2. Glycosaminoglycan concentration in angiosarcomatous, cirrhotic and normal human liver tissue. Cose 7 1200 r u O 10 tNg Q Tumor Coil h^on-lornor A^jocertl Ciji* I Tumor Cult 2 Nonlomor Adjocont Cast 2 Co 3 C.n* 4 Cai `j Cuse 6 FIG. 2. Glycosaninoglycan concentration in anglosarcoraatous, iirrhotic and noraal human liver tissue. Con 7 003674 <NW> Q o o <-r*i Cose l Adjacent Cote I Case 2 Adjacent Coe* 2 Cot* 3 Cate 4 Cote 5 Cote 6 FIG. 3. Fractional concentrations of g1ycosaminog1/cans in angiosarcomatous, cirrhotic and normal human liver tissue. VI. REFERENCES 1. Creech, J. L. and Johnson, M. N. Angiosarcoma of the Liver in the Manufacture of Polyvinyl Chloride. J. Qccup. Med. 16: 150-151. 1974. -- 2. Falk, H., Creech, J. L., Heath, D. W., Johnson, M. N., and Key, M. M. Hepatic Disease Among Workers at a Vinyl Chloride Polymerization Plant. JAMA 230: 59-63, 1974. 3. Makk, L., Creech, J. L., Whelan, J. G., and Johnson, M, N. Liver Damage and Angiosarcoma in Vinyl Chloride Workers: A Systematic Detection Program, JAMA 230: 64-68, 1974. 4. Creech, J. L., Makk, L., Whelan, J. and Tamburro, C. H. Hepatotoxicity Among Polyvinyl Chloride Production Workers During First Year of Surveillance Program. Gastroenterology 67,: 786, 1974. 5. Kohn, J. and Weaver, P. C. Serum Alpha Fetoprotein in Hepatocellular Carcinoma. Lancet 2: 334-336, 1974. 6. Tamburro, C. H., Makk, L. and Creech, J. L. Unpublished observation. 7. Girard, 0., Johnston, W. C., and Grahm, J. H. Cutaneous Angiosar coma. Cancer 25: 868-883, 1970. 8. Barr, R. and Bonin, M. "Letters." JAMA 231(9): 914, 1975. 9. Popper, H., and Thomas, L. B. Alterations of Liver and Spleen Among Workers Exposed to Vinyl Chloride. Ann. NY Acad. Sci. 246: 172-194, 1975. 10. Galambos, J. T., and Shapira, R. Natural History of Hepatitis: IV Glycosaminoglycuronans and Collagen in the Hepatic Connective Tissue. J. Clin. Invest. 52(11): 2952-2962, 1973. 11. Koizumi, T., Nakamura, N., and Abe, H. Changes in Acid Mucopoly saccharide In the Liver in Hepatic Fibrosis. Biochim. Biophys, Acta. 148: 749-756, 1967. 12. Kojima, J. Studies on the Metabolism of Hepatic Connective Tissue in Fibrosis of the Liver. Med. J, Osaka Univ. 16: 419-429, 1964. 13. Rubin, E. Autoradiographic Characterization of Sulfated Acid Mucopolysaccharides in Experimental Cirrhosis. J. Histochem. Cytochem. 14: 688-689, 1966. 14. Patrick. R. S. and Kennedy, J. S. The Synthesis of Sulfated Mucopolysaccharide at Sites of Hepatic Fibrosis Induced by Carbon Tetrachloride, Amyloidosis, and the Implantation of Catgut. J. Pathol. Bacteriol. 88: 549-555, 1964. 15. Kojima, J., Nakamura, N., Kanatani, M. and Ohmori, K. The Glycosaminoglycans in Human Hepatic Cancer. Cancer Res. 35(3): 542-547, 1975. 924 CMA 003676 1. Creech, J. L. and Johnson, M. Angiosarcoma of the Liver in the Manufacture of Polyvinyl Ch'orJde. J. Occud. Med. 15: 150-151 1974. -- 2- H,, Cfijecn, v. l., neatn, j. Jonnson, M. N., and Key, M. M. Hepatic Disease .Among Workers at a Vinyl Chloride Polymerization Plant. JAMA 230: 59-63, 1974. 3. MakJt, L. , Creech, _ J. L., nhe'an, J. G. , and Jonnson, M. V. Liver Damage and Angiosarcoma in Vinyl Chloride Workers; A Systematic Detection Program. JAMA 230: 64-68, 1974. 4. Creech, J. 1., ''akk, L., 'a-b.j'--o, C. - -sratctcx:- C'ty -r.org Polyvinyl Cn'sr-de Production Wooers Dcri"g rirst rear of Surveillance Program. Gast-oenterology 57: 756, 1974. 5. Konn. J, and Weaver, P. C. Serum Alpha Fetoprotein in Hepatocellular Carcincma. Lancet 2: 334-336, 1974. 6. Tamburro, C. H., Makk, L. ar.d Creech, J. L. Unpublished observation. 7. Girard, D., Johnston, W. C., and Grahm, J. H. Cutaneous Angiosar coma. Cancer 25; 353-233, 1970. 8. Barr, R. and Bonin, M. "Letters." JAMA 231(9): 914, 1975. 9. Popper, H., and Thomas, L. B. Alterations of Liver and Spleen Among Workers Exposed to Vinyl Chloride. Ann. NY Acad. Sci. 246: 172-194, 1975. 10. Galambos, J. T., and Shapira, R. Natural History of Hepatitis: IV Glycosaminoglycuronans and Collagen in the Hepatic Connective Tissue. J. Clin. Invest. 52( 11): 2952-2962, 1973. 11. Koizumi, T., Nakamura, N., and Abe, H. Changes in Acid Mucopoly saccharide in the Liver in Hepatic Fibrosis. Biochim. Biophys. Acta. 148: 749-756, 1967. 12. Kojima, J. Studies on the Metabolism of Hepatic Connective Tissue in Fibrosis of the Liver. Med. J. Osaka Univ. 16/. 419-429, 1964. 13. Rubin, E. Autoradiographic Characterization of Sulfated Acid Mucopolysaccharides in Experimental Cirrhosis. J, Histochem. Cytochem. 14: 688-689, 1966. 14. Patrick. R. S. and Kennedy, J. S. The Synthesis of Sulfated Mucopolysaccharide at Sites of Hepatic Fibrosis Induced by Carbon Tetrachloride, Amyloidosis, and the Implantation of Catgut. J. Pathol. Bacteriol. 88; 549-555, 1964. 15. Kojima, J., Nakamura, N., Kanatani, M. and Ohmo.ri, K. The Glycosaminoglyeans in Human Hepatic Cancer. Cancer Res. 35(3): 542-547, 1975. 924 CMA 003677 16. Anghileri, L. J. Metabolism of Acid Mucopolysaccharides in Hepatoma and in Normal Liver. Oncology 30: 304-317, 1974. 17. Yamamoto, K., and Teryama, H, Comparison of Cell Coat Acid Muco polysaccharides of Normal Liver and Various Ascites Hepatoma Cells. Cancer Res. 33: 2257-2264, 1973. 18. DiFerrante, N. M. The Measurement of Urinary Mucopolysaccharides. Anal. Biochem. 2l_: 98-106, 1967. 19. Bitter, T., and Muir, H. A Modified Uronic Acid Carbazole Reaction. Anal. Biochem. 4: 330-334, 1962. 20. Schiller, S., Slover, G. A., and Qorfman, A. A Method for the Separation of Acid Mucopolysaccharides: Its Application to the Isolation of Heparin from the Skin of Rats. J. Biol. Chem. 236(4): 983-987, 1961. 21. Kupchella, C., and Steggerda, F. The Distribution of Acid Muco polysaccharides in the Canine Gastrointestinal Mucosa. Trans. NY Acad. Sci. 34: 351-360, 1971. 22. Mowry, R. W. Alcian Blue Techniques for the Histochemical Study of Acidic Carbohydrates. J. Histochem. and Cytochem. 4: 407, 1956. 23. Sunderman, F. W. and Boerner, F. Normal Values in Clinical Medicine. W. B. Saunders. Philadelphia, p. 353, 1949. 24. Varma, R. S., Varma, R., Allen, W. S., and Wardi, A, H. Urinary Excretion of Acid Mucopolysaccharides in Schizophrenia. Biochem. Med. 11(4): 358-369, 1974. 25. DiFerrante, N. and Rich, C. The Determination of Acid Aminopolysaccharide in Urine. J. Lab. Clin. Med. 48: 491-494, 1956. 26. Kao, K. and Leslie J. Micro Fractionation and Determination of Urinary Glycosaminoglycans. Biochem, Med. 9(4): 317-326, 1974. 27. Goldberg, J. and Cotlier, E. Specific Isolation and Analysis of Mucopolysaccharides (Glycosaminoglycans) from Human Urine. Clin. Chim. Acta. 41: 19-27, 1972. 28. Manley, G., Severn, M. and Hawksworth, J. Excretion Patterns of Glycosaminoglycans and Glycoproteins in Normal Human Urine. 0. Clin. Pathol. 21: 339-345, 1968. 29. Pennock, C. A. A Modified Screening Test for Glycosaminoglycan Excretion. J. Clin. Path. 22: 310, 1969. 92S CMA 003678 30. Douglas, C. , Nowak, J. and Danes, B. Mucopolysaccharides in Urine During Normal Human Development. Pediatr. Res 7: 724-727, 1973. 31. Kuroda, 0., Saito, S. , Seno, N., Nagase, S. , and Anno, K. Isolation and Chemical Characterization of Mucopolysaccharides from Rat Tumors. Cancer Res. 34(2): 308-312, 1974. 926 CMA 003679 C , J arc Ii-es. 3 y,Jco;;Qi/saccharides ir --no 2unng Nor-a: human 3e/e'osirent. ?ediatr. Res 7: 724-727, -'3. 31. :<urcda, J,, Sana, 3., Sene, N., Nagase, S, , and Anno, X. Isolation and Che^'ci1 Oaracter'::f :n or "ucotolysaccharides from Rat '-tor; ujncar Res, . JUo-212, 1374, 926 CMA 003680 C. E. Kupchelia et al. Table 1 Characteristics ot tha transplantable hepatomas studied Tumor Ho* 7777 Growth rat designation Fast Time (days) until tumor reached 3* cnvlong axis ia Histology Poorly differen tiated 5123tc Intermediate 35 Moderately dif ferentiated 96 ISA Slow 89-90 Well differen tiated * According to tn* data of Hruban et al. (14). Metastatic potential* ++ 0 General metastatic char acteristic observed here Scattered lung micrometastasas but no gross metastases ev ident at sacrifice Multiple large metasta ses to lungs grossly avidani in lungs of all animals at sacrifice No lung motaatasss ev ident by sampling at sacrifice Collagen* 4- 0 0 Anlon-Exchang* Chromatography. Following uronic acid measurement, the uronic acid-positive material was pooled by tumor line with tumor and liver tissue material pooled sepa rately. The cetylpyridinium chloride was removed as described by Korn (17), and then each fraction was dialyzed and sub jected to Dowex 1-X2 (200 to 400 mesh) anion-exchange chromatography as described by Schiller et al. (28). The 0.0, 0.5,1.0.1.25, t .5, and 2.0 m NaCt etuate fractions were eluted stepwise in 15 to 30 fractions (10 ml) each. One-mi samples were assayed for uronic acid to produce an elution profile. Chromatographic fractions were pooled across all tissue groups, dialyzed, and concentrated, yielding 8 pooled fractions which were then subjected to enzymatic characterization. Enzymatic Characterization. Each of the fractions were subjected to digestion by Streptomycea hyaluronidase (digests only hyaluronic acid), bovine testicular hyaluronidase (digests hyaluronic acid, chondroitin, and chondroitin sulfate but not heparin, dermatan sulfate, or heparan sulfate), and chondroitinase ABC (digests hyaluronic acid, chondroitin, chondroitin sulfate, and derma sulfate but not heparin or heparan sulfate) as described by Kojima et al. (16). Histochemistry. Small pieces of liver and tumor as well as lung and intestine in selected animals were fixed in Zenker's fluid, embedded in paraffin, cut at 6 pm, and subjected to hematoxylin and eosin, Alcian blue-periodic actd-Sehiff (21), and Masson's trichrome (21) staining. Alcian blue-periodic acid-Schiff staining was also carried out with and without prior digestion in chondroitinaae ABC and bovine testicular hyalu ronidase (9). For the condroiHnase ABC study, hydrated tissue sections were incubated with enzyme (16) for 2 hr at 37s; control sections were incubated in buffer only. Statistical Analysis. Statistical analysis of tissue GAG data was carried out in a sequential format starting with an analysis of variance. Significant differences among the means were further evaluated using the Newman-Keuis procedure de scribed by Snedecor and Cochran (29). Uronic acid levels in the 0.04 m NaCI fraction isolated from the 3 tumors were similar and significantly higher (p < 0.05) than the levels found in normal liver. In the 1.2 m NaCI fraction, tumors exhibited 5- to 6-fold greater (p < 0.05) uronic acid levels than normal liver. In the 2.1 m NaCI fraction, uronic acid levels were simitar for alt tissues except that Tumor 9618A had significantly higher (p < 0.05) levels than did Tumor 5123tc. There was a statistically significantly greater (p < 0.05) level of uronic acid in the 0.03 m NaCI fraction for the livers of animals bearing Tumor 5123tc compared to normal liver and the livers of animals bearing the other 2 lines. Except for this, differences among livers were unremarkable. RESULTS The amounts of GAG in Tumors 7777, 5123tc, and 9618A and in normal liver are presented by fractions in Chart 1. "Uronic acid" levels in the 0.03 m NaCI fraction were similar in Tumors 7777 and 5l23tc but were significantly (p < 0 05) lower than the levels found in tumor 9618A and in normal liver. 420 me avers of non-twwor-toeerWQ <M) animals tor each ot 4 sequentially coaected sail traction*. Sara. geometric maana. OtWarancaa between P/t and S/N m the 0.03 m NaCI traction, between tumor Hama and normal War in bom tba 0 * m and 1.2 at NaCI traction, and between f and 5 m ma 2.1 m NaCI traction ara statistically tlgnHIcant <p < 0.05). CANCER RESEARCH VOL. 41 CMA 003681 Chromatographic data for the 0.4 m NaCI and 1.2 m NaCl GAG fractions, those fractions which were appreciably larger in tumor tissue versus normal liver, are presented in Chart 2 together with the patterns obtained for these same fractions isolated from normal liver and the liver of tumor-bearing ani mals These data indicate that, even though the uronic levels in the initial fractions were similar from tumor line to tumor line, there were some qualitative GAG differences in the fractions isolated from different tumor lines. This is even more apparent in composite Chart 3, which was derived by multiplying the mean of the individual uronic acid levels shown in Chart 1 by the percentage of distribution shown in Chart 2 and then summing within each chromatographic fraction. GAG s in Morris Hepatomas Enzymatic Characterization of Tumor GAG's The results of the enzymatic characterization of GAG'S iso lated from, tumor tissue are summarized in Table 2. On the basis of the elution patterns reported by Kao and Leslie 0 5) and corroborated in our own study of authentic GAG's and on the enzyme susceptibilities for each fraction given in Table 2, we arrived at the identities of the predominant GAG in each chromatographic fraction given in Table 2, Column 7. Histochemlcal Observations Tumors. Each of the 3 tumor types exhibited significantly more intense Alclan blue staining than did normal liver or host liver; GAG's were generally distributed throughout the tumor tissue. Tumor sections subjected to hyaluronldase or to chon- 0.4M NaCIlolgbl* matarial SO- 1 2M NaCIsoluble material A 8 C HjO OSM 10M 12$M ISM J.OM Anion exchenge column NeCI tractions Chart 3. Anlon-exchango chromatographic patterns obtained tor the QAO'a isolated initially aa 0.4 m N*Cl-*otobto and 1.2 m NxO-Mfubto eefytpyridinlum chloride complexes. A, Tumor 7777: S, Tumor 5123tc; C. Tumor 9S18A: 0. liver* ot tumor-bearing animats pooled acroea tumor types: E. normal liver. This chart I* a composite ot Charts i and 2 and la based on the percentages (given In Chart 2) of the arithmetic mean of the indMdual uronic acid values Chart i) for each tissue type by traction. droitinase ABC exhibited reduced levels of Aldan blue staining consistent with biochemical measurements and the enzymatic characterization of chemically isolated fractions. Host Livers. The livers of animals bearing Tumors 7777 and 9618A were histologically indistinguishable from normal liver. Liver tissue from animals bearing Tumor 5l23te consistently exhibited a slightly greater vacuolar appearance than did nor mal liver (Fig. 18). Urinary GAG Excretion 0 H^OOSM^M HjO 0.5M 1.0M 1.25M 1.5M 20M Anion exchange column NaCI fractions Chart 2. Anion-exchange chromatographic profiles lor the GAG-cetyipyndmium chloride comptexee eotubMzed Initially In 0.4 m and 1.2 m NaCI. A. Tumor 7777; a. Tumor St23lc; C. Tumor 9S18A; 0, pooled liver tissue from tumor- beanng animals. E. normal liver. Recoveries ranged (torn 65 to 107% Over the entire study, mean total uronic acid excreted per pair of animals per 24 hr for control animals and animats bearing Tumors 7777, Sl23tc, and 9618A were 118 7 (S.E.), 118 3.203 12, and 201 7 /ig, respectively. GAG excretion by animals bearing Tumors 5123tc and 9618A were statistically significantly (p < 0.05) elevated oyer controls and animals bearing Tumor 7777. There were 22, 8, 8, and 27 twenty-four-hr collections assayed, respectively. The smalt number of 7777 and 5123tc samples was due to the rapid growth of these tumors and time in transit after inoculation. Urinary excretion profiles for animals bearing Tumor 5123tc over time and in relation to tumor size are illustrated in Chart 4. A similar pattern over a longer time span was observed in animals bearing Tumor 9618A. in both 5123tc and 96ISAbearing animals, uronic acid excretion appeared to be greater after the tumors reached larger sizes, but no regression with tumor size was apparent in either case. FEBRUARY 1981 CMA 003682 421 Chromatographic data 'or tha 0 4 m N'aC: arc ` 2 m 'iaCl GAG `ract o.~3, Vaatioos which vsre zppracabiy arger m tumor tissue versus normal liver are presented m Chart 2 taged'er w ;,n the patterns ootameu 'or meae same tract.c.^s sc itT-d 'sr-J1 .;er and the .,-i' a, .-mcr-cear.'rg a-,i- -- a = ''sseda'a rS cate that, eve" ''cuco - = =.-:s --- 1 I CiA G S >P A*Crf*$ ^ l-'Z Z' 2*" Z 2 A tdere were some qualitative GAG differences in the fractions isolated from different tumor lines. This is even more apparent m composite Chart 3, which was derived by multiplying the mean of the individual uronic aetd levels shown in Chart 1 by the percentage of distribution shewn in Chart 2 and men summing within each chromatographic fraction. Enzymatic Characterization of Tumor GAG's The results of the enzymatic characterization of GAG s iso lated from tumor tissue are summarized in Table 2. On the basis of the elution patterns reported by Kao and Leslie (15) and corroborated in our own study of authentic GAG'S on tne enzyme susceptibilities for each fraction given in Table 2, we arrived at the identities of the predominant GAG in each chromatographic fraction given in Table 2, Column 7. Histochemicai Observations Tumors. Each of the 3 tumor types exhibited significantly more intense Alcian blue staining than did normal liver or host liver, GAG s were generally distributed throughout the tumor tissue. Tumor sections subjected to hyaluronidase or to chon- 0.M tubloiuot* malarial 1.2MN*CIsotubto mannai a 9 C Chart 3 Anion-aicnanga ctiromatograoMc oattama ootamad tor tha QAQ i laolaiad initially aao.a u NaCI-*oiubto and 1.2 u NaO-aoiuW* catyWynOimum eWorld* eomoMaaa. A, Tumor 7777; B. Tumor S123te; C, Tumor 381SA, 0. Ilvars of tumor-baanng ammala pooM acroaa tumor tygoa; S. normal Inar. Tha chart it a composite of Charta t and 2 and e baaed on tha percentage* (gtran in Chart 2) ot tha arimmadc maan of tha mdhndual uranic acid value* (given in Chart t) tor aaeh Uaaua tyo* by fraction. droitinase ABC exhibited reduced levels of Alcian blue staining consistent with biochemical measurements and the enzymatic characterization of chemically isolated fractions. Host Livers. The livers ot animals bearing Tumors 7777 and 9618A were histologically indistinguishable from normal liver. Liver tissue from animals bearing Tumor 5123tc consistently exhibited a slightly greater vacuolar appearanca than did nor mal l)var(F1g. IS). Urinary GAG Excretion 0 E Anton xcha ' column NaCI fraction* Chart 2. Amon-aaehanga ttuumalucrapMc prolltaa lor ttto OAG-catytoynOirv. um cMonda complaaao aakiMIxod Wnte*y m 0 * u am i 2 u Nad A, Tumor 7777: 0. Tumor 912310-, C. Tumor MtSA; D. pooled War tlaaua from tumor* Soaring animat*; g, normal Kvar. Raeovanaa rangaO tram SS to 107%. Over the entire study, mean total uronic add excreted per pair of animals per 24 hr for control animals and animals bearing Tumors 7777, 5l23tc, and 9618A were 118 7 (S.E.), 116 3,203 * 12. and 201 7pg, respectively. GAG excretion by animals bearing Tumors 3123tc and 9618A ware statistically significantly (p < 0.05) elevated over controls and animals bearing Tumor 7777. There were 22, 8, 8, and 27 twenty-four-hr collections assayed, respectively. The smalt number of 7777 and 5123tc samples was due to the rapid growth of these tumors and time in transit after inoculation. Urinary excretion profiles tor animals bearing Tumor 5123te over time and in relation to tumor size are illustrated in Chart 4. A similar pattern over a longer time span was observed in animals bearing Tumor 9618A. in both 5i23tc and 9618Abearing animals, uronic acid excretion appeared to be greater after the tumors reached larger sizes, but no regression with tumor size was apparent in either case. FEBRUARY 1981 CMA 003683 421 C. e Kupchella et al Table 2 (Oenhhcation of tha predominant GAG >n each of ouranion-exchange chromatographic tractions The identification (Column 7) of the predominant GAG or GAG's present <n each of our anton-exchange chromatographic fractions (Column 1) was deduced from (a) solubilities of cetytpyndimum chloride complexes. (b) anion*exchange chromatographic patterns compared to those that W* obtained for authentic GAG a and those reported by Kao and Leslie (15) (Columns 2 and 3). and (c) the susceptibility ol each fraction to muoopolysacchandases (Columns 4. 5. and S) Dowex 1-X2 faction (m NaCI) GAG's reported Oy Kao and Leslie (15) to 0e primarily eluted in this trection Otner GAG s reported to be partially eluted in this fraction digested by Streptomyces hyalurondase % digested by hov*ne testic ular hyaluroni- dase % digested by chondroiti- nase A8C Predominant GAG >n our fraction 00 05 10 125 15 20 Hyaluronic acid (84)J Heparan sulfate (78) Chondroitin 4-sulfate (66) Chondroitin 6-suifate (63) Heparin (72) None Hyaluronic acid (12) Heparan sulfate (14) Heparin (18) Dermatan sulfate (19) Chondroitin 6-sulfate (19) Dermatan sulfate (67) Chondroitan 4-suH fate (13) 70-100 35-61 0 0 30 0 80-100 61 0 <2 25 100 Hyaluronic acid 0 Hyaluronic acid 0 Heparan sulfate 0 Heparan sulfate 22 Heparan sulfate and/or Itoparln Hepann 0 * Numbers m parentheses, percentage eluted per fraction. <n Chari 4. Urinary GAG excretion in relation to tumor growth for animals bearing Tumor 5123tc and In control animals. , tumor-boring animals: O, control animals. We had sufficient urinary GAG'S for anion-exchange chro matography only in the case of control animals and animals bearing Tumor 9618A. A chromatographic comparison of these 2 profiles indicated that the elevation in urinary GAG s occurred across all column fractions to about the same degree. DISCUSSION The results depicted in Charts i to 3 conform to the gener alization (7. 16, 18) that tumors, including hepatic tumors, exhibit high levels of GAG's relative to the tissue of origin. Charts 2 and 3 suggest that differences between tumors may be related to the behavioral properties of the tumors. Chart 3 reveals a gradation from normal liver, through the liver of tumorbearing animals, well-differentiated, slowly growing tumor tis sue to faster-growing, metastatic tumor lines. The gradation shown in Chart 3 is even more striking if the patterns for tumor lines 7777 and 5l23tc are inverted. Since tumor line 5l23tc was more highly metastic than was line 7777 (Table 1), such an inversion would arrange the tissues according to metastatic potential and raises the possibility that the patterns, perticularly Fraction 1.0 m. are related in some way to metastatic potential. It is postulated that the 1.0m NaCI chromatographic fraction contains an undersulfated form of heparan sulfate such as that described by Kuroda et al. (18) in AH109 hepatomas. Both Kuroda et al. (18) and Saito (25) reported that heparan sulfate is the major GAG constituted in AH109A hepatic tumors. Kuroda et al. also reported that most of this heparan sulfate is eluted in 1.0 m NaCI in anion-exchange chromatography and that heparan sulfate is also the predominant GAG in normal liver and suggested that this may mean that the tumor heparan sulfate comes from tumor cells and not from connective tissue elements within the tumors. The predominance of heparan sulfate in Morris hepatomas and in AH109A hepatomas do not conform with the fact that hyaluronic acid and chondroitin sulfate have generally been identified as the predominant GAG'S in animal tumors (4, 7). These findings likewise do not conform with the report by Kojima et al. (16) that chondroitin sulfate and hyaluronic acid are the predominant GAG's in human hepatocellular cancer. The possibility that tumor heparan sulfate is related in some direct way to tumor behavior has been raised by others. A role for sulfated GAG's in cell recognition and adhesion has been proposed by Dietrich et al. (8). and Chiarugi and Vannucchi (3) have proposed that cell surface heparan sulfate regulates both cell division and transport. The histological appearance of the tumors studied here and the histological normalcy of the livers of tumor-bearing animals agree with the findings reported by Hruban et al. (14). Although the vacuolar apipearance of the livers of animals bearing Tumor 5123tc is not abnormal in rodents, this characteristic was uniformly present in all Tumor 5123tc-bearing animals and was found in no others. The possibility exists that this structural feature is related to the high concentrations of the uncharac- 422 CANCER RESEARCH VOL 41 CMA 003684 GAG s in Morris Hepatomas terized, 0.03 m NaCI-soluble, uronic acid-positive-matenal found in these livers. The fact that urinary GAG excretion was elevated in animals bearing Tumors 5123tc and 9618A but not in animals bearing Tumor 7777 suggests that urinary GAG excretion may reflect properties of certain tumors and is not simply an indirect result of me presence of tumor Overall levels of excretion of GAG s m our control and experimental animals agree with levels reported for rats by Lehtonen ef a/. (19). It remains to be determined what the source(s) of the elevated urinary GAG is (are). There have been reports of striking increases in GAG synthesis in tumor cells (see Ref. 20), but Kojima ef a/, (16) have suggested that decreased degradation may account for increased GAG in some tumors. Our investi gations provide no evidence that the source of the abnormal GAG increases in tumor tissue and urine is anything other than the tumor cells themselves; however, a possibility that these results stem from an impaired ability of the liver to degrade circulating GAG'S cannot be ruled out. CONCLUSION Our data clearly show that hepatomas 7777, 5123tc. and 9618A have GAG compositions that are appreciably different than that of normal liver. These data show that there are also qualitative differences in GAG between tumor lines and that heparan sulfate patterns parallel growth rate and possible degree of malignancy. Heparan sulfate is the predominant GAG in the hepatomas studied. The increased urinary excretion of GAG's for 2 of the 3 tumor lines examined suggests that urinary GAG analysis may prove useful in the detection and diagnosis of some hepatic tumors. REFERENCES 1 Bitter. T . and Muir, H. A modified uronic acid carbazote reaction. Anal Bioehem , 4: 330-334. 1902. 2. Cameron. and Pauling. L. Ascorbic acid ana me olycosammoglycans. Oncology (BaaaO 27. 181-192, 1973. 3. Chiarugi. V. P , and Vannucchi. S. Surface heparan sulfate as a control element in eukariotic cede: a working modal. J. Thaor. Biol.. St - 459-475, 1978. 4. Choi. H V., Meyer. K.. and Swarm, R. Mucopolysaccharide and proteinpoiysaccharide of a transplantable rat chondrosarcoma. Proc. Natl. Acad. Scl. U. S. A., 88. 877-879. 1971. 5. Cudkowicz, 0. The potyiacchandas of a human carcinoma. Br. J. Cancer. 10: 759-762. 1956. 6. Curran. K. L., KupchaKa. C. E.. and Tamburro. C. H. Urinary glycoaammoglycan patterns in angiosarcoma of the liver. Cancer (Phila.), 40. 30503053. 1977. 7. Oamshelsky, I.. Oppenheimer. E. T . HeriNer-Watkms. O , and Wilhite, m Mucopolysaccharides In animal tumors. Cancer Res.. 28. 229-232. 1966. 8 Dietrich. C.P. Sampaio. L O . Toledo. O.M S. and Cassaro. C M F Cell recognition and adhesiveness: a possible biological role tor the suttated mucopolysaccharides. Bioehem. Biophys. Res. Commun.. 75 329-336. 1977 9 Drury, R A B . and Wallington, E. Carleton s Histological Techntq 210-219 New Vork. Oxford University Press. 1967 10 Friman C , and Juvani. M Urinary excretion and gtycosammogiycans in malignant diseases oi ihe haemopoietic and lymphatic tissues Acta Med Scard `OH 1ib-122 1975 i i Gordin, A , Edgren J , Friman. C and Holmstrom. T A case ol disseminated hemagiomatosis with cutaneous, hepatic and skeletal mamtestations and increased urinary excretion of glycosaminoglycans. Acta Med. Scand., 198 525-530, 1975 12 Hatae. Y . Yoda. V . and Makita. A. Glycosaminoglycans in small ctH carcinoma ot human lung- histologically characteristic pattern. Gann. 70 389-390. 1979 13 Hruban, Z . Mochizuki. Y , Slesers. A., and Moms. H. p A comparative study ol cellular organelles of Morris hepatomas. Cancar Res.. 32 853867, 1972 14. Hruban. Z.. Morns. H. P., Mochuuki. V . Meranze. 0 R . and Slesers, A. Light microscopic observations ot Morris hepatomas. Cancar Rea. .31. 752762, 1971. 15. Kao. K. Y. T . and Leake, J- G Microlractlonation and determination ot urinary glycosaminoglycans. Bioehem Med.. 9. 317-326, 1974. 16. Kojima, J . Nakamura, N.. Kanatani. M . and Ohmon. K. The glycosammoglycans in human hepatic cancer. Cancer Res., 35. 542-547. 1975. 17. Korn. E. 0. The isolation of heparin from mouse mast ceH tumor. J. Biol. Chem , 234 1325-1329. 1959. 18. Kuroda. J.. Saito, S.. Sane. N., Nagaae, S,, and Anno. K. Isolation and chemical charactarizatton of mucopolysaccharides from rat tumors. Cancar Res . 34 308-312.1974. 19. Lehtonen, A.. Nanto, V.. and Kasanen. A. The effects of some antiinflam matory agents on me urinary excretion of mucopolysaccharides m the rat. Ann. Med. exp.; Fenn . 45. 32-34. 1967. 20. Manley. G.. Bower. L-. and Anaon. A. Urinary excretion of gtycoaaminoglycans in disseminated neoplasm. J. Clm. Pathol., 3f: 447-453. 1975. 21. McManus. J. F. A., and Mowry. R. W. Staining Methods: Histological and Histochemical. New York: Paul B. Hoeber. Inc. (Harper and Brothers). 1960. 22 Nakamura. N,, Mun, Y.. Tanigaki. Y.. and Koima. J. Changes in the cellular glycosaminoglycans of cultured mastocytoma celts induced by sodium bu tyrate. Biochim. Biophys. Acta 627. 60-70, 1980. 23. Ozzello. L . and Speer. F. 0. The mucopolysaccharides in the normal and diseased breast: their distribution and significance. Am. J Pathol, 34 993- 1005. 1958. 24 Rich. C . and Meyers. W. P L. Excretion of ack) mucopoiyseccha^^B. the urine of patients with malignant neoplesUc diseases, j. Lao. CIh^Bo 54 223-228. 1959. 25. Saito. S. Mucopolysaccharides of rat ascites hepatoma cells. Gann. 64 247-255, 1973. 25. Sakaki. T . Tauruim. N., Maeda, J., and Matsuda, H. Studies on the influ ences ot acid mucopolysaccharides on the growth of Tawa sarcoma. J. Osaka Dental Univ . 4:113-122. 1970. 27 Sanders. F K.. and Smith. J. Effect of collagen and sod polysacchandes on the growth ot BHK/21 ces Si sami-sokd media Nature (Lond ). 227 513- SIS. 1970. 28 Schifler. S.. Siover. G. A., and Oorfman. A. A method for the separation of acid mucopolysaccharides: its application to the isolation ot heparin from the skm of ran. J. Biot. Chem.. 230:963-987. 1961 29. Snedacor, G. w , and Cochran, w. G. Statistical Methods, EO. 6. p. 273. Ames, lows: Iowa Slate University Press, 1967. 30. Takeuchi. J. Growth promoting effect of acid mucopolysaccharides on Ehrlich ascites tumor. Cancar Res., 26' 797-602.1966. 31. Winterbourne. D. J., and Mora. P. T Distribution ot giycocontugatos m mouse fibroblasts with varying degrees of tumongemcity. J. Supramof. Struct.. 7. 91-100. 1977. FEBRUARY 1981 CMA 003685 3 T C3 '.t NsCi-soiubie. urcric aco-ce: r - - -t -i - ' -IfC "^e 'ac! tr.a; urinary GAG exers'i.^n v ;. ~c ;r ~ ="i--! =;.s - ,, wi* _,,." *' . a.".a i * ^c" r.' "~ 'ai an^a, 3 a^v. 1:a a . a a 'izzroz ; ':.:s by Ub:c-i -,f a/ 0 3) It remains to be determined wnat the source(s) of the elevated urinary GAG is (are). There have been reports sr striking ncreases m GAG sytVhers n tumor ce'*s (see Pe* 20. tut Ko;ima er a/ (to) nave suggested ;na; decreased degradation may account for increased GAG in some tumors. Our investi gations provide no evidence that the source of the abnormal GAG increases in tumor tissue and urine is anything other than the tumor cells themselves, however, a possibility that these results stem from an impaired ability of the liver to degrade circulating GAG s cannot be ruled out CC.'.CLwSiC.'J Our data ciearly show that hepatomas 7777. 5i23tc, and 961 SA have GAG compositions that are appreciably different than that of normal liver. These data show that there are also Qualitative differences in GAG between tumor lines and that heparan sulfate patterns parallel growth rate and possible degree of malignancy. Heparan suifate is the predominant GAG in the nepatomas studied. The increased urinary excretion of GAG s for 2 of the 3 tumor lines examined suggests that urinary GAG analysis may prove useful in the detection and diagnosis of some hepatic tumors. REFERENCES 1 Sitter. T . and Muir. H A modified uronic acid barcarole reaction Anal Siocnam . 4 330-304. 1963 2. Cameron . and Pauling, L Aicorbtc acid and me qlycoaamtnoglycana. Oncology (Basel) 2 7 181-192. 1370. 3. Chiarugi, v P . and Vannucctn. S, Surface necaran aulfate aa a control element in euXanotic cells: a working model. J. Theor. Biol.. Sr 459-475. 1978 4. Cnot, H. V . Meyer. K.. and Swarm, R. Mucaoolyseccriande and oroteinDOlysacchande ol a tranaolantaole rat chondrosarcoma. Proc. Natl. Acad. So U S. A.. 6* 877-879. 1971. 5. Cudkowiej, 0. The pofyseecnandes Of a human carcinoma Br. J. Cancer, 10 759-762. 1956. 6 Curran. K. I... KuDCheHe. C. . and Tamhurro. C. H Urinary glycoiaminogiycan Datlema m angioaarcoma of the ever. Cancer (Phila). 40 30503053. 1977 7. damahefsliy, I., Opeenhenner. E. T., Hentier-Watkini. 0 . and Wilhite. M Mucopotyiecchendee *1 anknat tumors. Cancer Res.. 26 229-232. 1966 5 .vun cutarocus necA'ic and ^nai^iai 'i'-'jiSu -''s2rv-:^cr-:,*iG'r> Vs $g,, - 2 . 4 3 ^ 525-53C 1975 1 2 Hafae Y y, and Makita, A Glycosanuro'giycans .p small cp`i care.noma of ruman lung histologically characteristic pattern. Gann 70 389-*l90 1979 '-,h 2 ijtjRi v 5-*S'r5 A in v-Kf-id, m p j. ^ study of cellular organelles of Morns hepatomas Cancer Res 32 353- 367. 1 972 \ x HruDan, Z , Morns, h, P , Mocnuuki. y Meranz*. O R . and GJeser*. A i_i^rt microscopic ooservations of Moms hepatomas Cancer 31 752* '52,1971 15 Kao. K Y T . and Leslie. J G Microfractionation and determination of urinary gfycosammogfycans Biochem Med , 9 317-326,1374 16. Koiima. J,. Nakamura. N., Kanatani, M , and Ohmon, K The giycosammo- giycans in human heoafic career Cancer Res . 35 542-547 1375, 1 7 <C'n D The isolation c* "eoarin from /^cuse mast zz'l j ' 0 Cr*m 234 1325-1329 '959 i"vufodd, J , Sa.to. S-. Seno, N,# Negate, 3 , and Anno. K Isolation and onemicai characterization of muccociysaccrandes from rat tumors Cancer Res,34 3C3-312. 1974, 19 Lehfonen, a , Naoto. V . and Kasanen. A The effects of some antiinflam matory agents on the urinary excretion of mucopolysaccharides in me rat. Ann Med exp : Fenn,, 45. 32-34, 1967 20 Manley. G . Bower. L., and Anson. A Urinary excretion of gtycosammogty- -a->s ,n disseminated necciasm J C:.n Pathol,. 3f 447-453 1973 21 McManus. J, F a., and Mowry. R W Staining Methods* Histological and HfstccnemtcaL New York: Paul 8 hoeber, ;ng. {Harper and Brothers). 1960 22 Nakamura. N . Mun, Y., Tantgaki. Y . and Koiima. J Changes m the cellular giycosamtnogtycans of cultured mastocytoma certs induced by sodium bu tyrate. Biochim Siophys. Acta 627 60-70. i960. 23 Ozzefto, L,. and Speer, F, D. The mucopolysaccharides in the normal and diseased breast: their distribution and significance. Am J Pathol 34 9931005, 1953. 24 Rich, C . and Meyers. W P L. Excretion of acid mucopolysaccharides in the urme of patients with malignant neoplastic diseases. J Lab. Cim Med. 54 223-228. 1959. 25 Saito. S Mucopolysaccharides of rat ascites hepatoma ecus Gann. 64 247-255, 1973, 26 Sakaki, T , Tsurumi. N.. Maeda. J . and Matsuda, H. Studies on the influ ences of acid mucopolysaccharides on the growth of Tawa sarcoma, J. Osaka Dental Urn* . 4. it3-122. 1970. 27 Sanders. F K.. and Snath, j. Effect of eoltagen and sad polysaccharides on th* growth of BHK/21 cent ei sem^sortd medn. Nature (Lond ). 227 513- Si 5, 1970. 28. Schiller. S-. Stoner. G. A., and Oorfman, A. A method for the separation of acid mucopofysacchandes: its application to the isolation of hepann from the skin of rats. J. Biol. Cham.. 238. 963-987, 1961. 29. Snedeeor. G. W , and Cochran, w. G- Statistical Methods, Ed 6, p. 273. Ames. Iowa: Iowa State University Press, 1967. 30 Takeuchi, J. Growth promoting effect of acid mucopolysaccharides on Ehrlich ascites tumor. Cancer Res., 26 797-802. 1966. 31. Winterbourne, 0, J., and Mors. P. T Dtstnbutton of glycoconrugates in mouse fibrobiests with varying degrees of tumongemcity. J Supremo*. Struct., 7. 91-100, 1977 FEBRUARY 1961 CMA 003686 423 C. E. Kupchella et at. Fig. 1. A. typical lung matastasas In animals baaring Tumor 7777. Thasa wars found mors traquantty and saniar and wars largar in animals baaring Tumor Si 23te. No lung matastaaas wars found In any ol ttia animats baanng Tumor 961SA xSO. S, kvar ol animals baanng Tumor 51231c showing dear cytoplasm-vacuolar appearance characterlatlc ol such Hvars. xSO. 424 CANCER RESEARCH VOL. 41 CMA 003687 EPA 560/6-81-002 9 EPIDEMIOLOGY STUDIES SCREENING FOR THE EARLY DETECTION OF DISEASE IN INDIVIDUALS EXPOSED TO VINYL CHLORIDE JANUARY 1981 FINAL REPORT U.S. ENVIRONMENTAL PROTECTION AGENCY OFFICE OF PESTICIDES AND TOXIC SUBSTANCES WASHINGTON, D.C. CMA 003688 EPA 560/6-81-002 January 1981 SCREENING FOR THE EARLY DETECTION OF DISEASE IN INDIVIDUALS EXPOSED TO VINYL CHLORIDE Carlos H. Tamburro Charles Kupchella 3 Kenneth Taylor Emanuel Landau^ by Richard Greenberg 2 Hi 1 degarde Maricq Joseph Whelan, Jr. c Joseph Seifter 1. University of Louisville Louisville, Kentucky 2. University of South Carolina Charleston, South Carolina 3. Yale University New Haven, Connecticut 4. American Public Health Association Washington, D. C. 5. U.S. Environmental Protection Agency Washington, D. C. Project Officer Jane Keller Office of Pesticides and Toxic Substances Washington, D.C. U.S. ENVIRONMENTAL PROTECTION AGENCY WASHINGTON, D.C. 20460 CMA 003689 DISCLAIMER This project has been funded with Federal Funds from the Environmental Protection Agency under contract number 68-01-3859. The content of this publication does not necessarilyref1ect the views or policies of the U.S. Environmental Protection Agency, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government. CMA 003690 DISCLAIMER This project has been funded with Federal Funds from the Environmental Protection Agency under contract number 68-01-3859. The content of this publication does not necessarilyreflect the views or policies of the U.S. Environmental Protection Agency, nor does mention of trade names, commercial products, or organi2ations imply - .wv ar. * ,,i3 u ^ . oovern^ent, CMA 003691 ABSTRACT A prospective collaborative study was conducted to compare the effectiveness of four clinical techniques in the detection of liver damage due to vinyl chloride monomer exposure. A chemically exposed and medically monitored worker population was identified by histopathological and biochemical documentation. Three techniques were non-invasive: a) grey scale ultrasonography of the liver, b) microvascular skin capillary assessment, and c) urinary analysis of glycosam.inoglycan excretion. The fourth technique was the standard 99mTc sulfur colloid radionucleotide liver spleen scan. The screening studies were performed on a randomly selected single cohort of chemical workers; some of whom were known to have disease. All four techniques were analyzed for their sensitivity and specificity as compared to results of the liver biopsy and biochemical blood test classification. Although all four screening techniques had a sensitivity and specificity sum greater than one, none were significantly better than could be explained by chance or the use of a biased coin. Reclassification of the population into those with more severe biochemical abnormalities improved the sensitivity of all screening tests, but only the sensitivity and specificity sum for the GAG test were statistically significant at the 0.05 level. There was no significant correlation between any pair of screening tests. None of the four screening tests agreed with the biopsy results better than might be obtained by ' 'i biased coin or chance. These screening studies as presently constituted, do not provide sufficient sensitivity and specificity to warrant their use in community screening for subclinlcal asymptomatic hepatic injury due to chemical exposure. CMA 003692 X INTRODUCTION The initial reports of primary liver cancers (angiosarcoma) in rats exposed to vinyl chloride by Maltoni, et al (1) and the discovery of similar liver tumors y in vinyl chloride polymerization workers by Johnson and Creech (2) has lead to considerable environmental concern regarding corimunities surrounding chemical industries which utilize potentially carcinogenic agents such as vinyl chloride. Several investigators have reported on various screening techniques as effective indicators of vinyl chloride chemical injury. Four such techniques - ultrasono graphy (3), radionucleotide scanning (4), na'ilbed capillary visualization (5), and glycosaminoglycan (GAG) (6) excretion - were reported to have some possible usefulness in detecting early chemical injury to the liver. In order to determine the useability of these techniques for community screening, the American Public Health Association (APHA) and Environmental Protection Agency (EPA) funded a multi-center collaborative study designed to determine the comparative sensitivity and specificity of these various techniques in detecting and identifying chemical related hepatic injury in asymptomatic individuals. Materials and Methods Population Selection The chemical worker population consisted of 1,178 active (Group B, Figure 1) employees as of September 1, 1977, and 70 employees who had had liver biopsies regardless of current employment status (Group A, Figure 1); they were undergoing annual medical screening for the identification of work-related disorders. The medical screening consisted of an annual or semi-annual (for those employees with 10 or more years of employment) comprehensive history and physical examinations, laboratory screening studies consisting of 35 biochemical tests, chest and abdomi nal X-rays, and radionucleotide liver-spleen scan. CMA 003693 The initial reports of primary liver cancers (angiosarcoma) in rats exposed to vinyl chloride by Maltoni, et al (1) and the discovery of similar liver tumors in vinyl chloride polymerization workers by Johnson and Creech (2) has lead to considerable environmental concern regarding communities surrounding chemical industries which utilize potentially carcinogenic agents such as vinyl chloride. Several investigators have reported on various screening techniques as effective indicators of vinyl chloride chemical injury. Four such techniques - ultrasono graphy (3), radionucleotide scanning (4), na'ilbed capillary visualization (5), and glycosamincglycan (GAG) (6) excretion - were reported to have seme possible usefulness in detecting early chemical injury to the liver. In order to determine the useability of these techniques for community screening, the American Public Health Association (APHA) and Environmental Protection Agency (EPA) funded a multi-center collaborative study designed to determine the comparative sensitivity and specificity of these various techniques in detecting and identifying chemical related hepatic injury in asymptomatic individuals. Materials and Methods Population Selection The chemical worker population consisted of 1,178 active (Group B, Figure 1) employees as of September 1, 1977, and 70 employees who had had liver biopsies regardless of current employment status (Group A, Figure 1); they were undergoing annual medical screening for the idertification of work-related disorders. The medical screening consisted of an annual or semi-annual (for those employees with 10 or more years of employment) comprehensive history and physical examinations, laboratory screening studies consisting of 35 biochemical tests, chest and abdomi nal X-rays, and radionucleotide liver-spleen scan. CMA 003694 2 This population was selected for the collaborative study to determine the comparative effectiveness of four screening techniques in detecting liver damage as indicated by 1) past histopathological documentation of liver injury of 2) current hepatic dysfunction identified biochemically. Three of the four techniques were included as potential non-invasive pro cedures suitable for determining the effects of vinyl chloride in a comnunity population. These included 1) grey scale uTtrasorography of the liver as devel oped by Taylor and colleagues (7, 13, 14), 2) a nailbed skin capillary evaluation of the middle and distal phalanges of the fingers as developed by Maricq and associates (8), and 3) urinary analysis of glycosaminoglycan excretions (GAG) as published by Kupchella and associates (9). The fourth method included for comparison purposes was the standard radionucleotide liver-spleen scan utilizying 99mTc colloid and interpreted by Whelan and associates (10). The non-invasive screening studies were performed during a single week on a group randomly selected from all chemical company employees. The workers were selected on the basis of complete medical and work data for 1976-1977, and all those employees who had investigative liver biopsies performed during the screening program (1974-1977). The selection process for these workers is illustrated in Figure 1. The biochemical data and radioisotopic scans were part of the routine medical surveillance system for the employees. The pathological data was based on the last or most recent liver biopsy (s) which were performed for medical reasons, both related and not related to their work. Positive and negative results were determined as defined in Table 1 which list the technique, the evaluation or evaluators, and the criteria used. The employees targeted for examination were selected by simple random sampling from all available employees. CMA 003695 3 One hundred and twenty one of the targeted 170 employees (71 percent) particip^^|. Twenty six declined to participate or could not be scheduled; 13 did not keep their scheduled appointment. After participation, nine employees were discovered to have been misclassified as biochemically abnormal. They had some biochemical ,/ abnormalities but not all (classified intermediate) and have been eliminated from the final analysis. Four biochemically abnormal individuals had abnormally low test values and they were included in the analysis. Liver 8iopsies Liver biopsies were performed by the transjugular technique (11) and provided two to five biopsies from various areas of the liver. In addition, some individuals had second biopsies performed by the percutaneous or wedge biopsy via mini-laparotomy procedures. Pathological data was recorded in a computerized format identifying all histological abnormalities in a semi-quantitative fashion. Biopsies were read without knowledge of the individual's medical history or chemical exposure by two pathologists and a hepatologist with extensive experier in hepatic chemical injury. All biopsies were classified as 1) normal, 2) abnormal, a) chemical injury, and b) non-chemical injury. The non-biopsy groups were drawn from those currently employed, and based on biochemical liver "function tests" individuals were sorted into positive, negative and indeterminate for hepatic disease. Only the positive and negative are included in this study. The ultrasonic evaluation and its relative ability to identify hepatic damage due to vinyl chloride has been published elsewhere (3). Microvascular techniques and the method of evaluation by Dr. Maricq and co-workers are also published in part (8). The experimental work on GAG excretion in vinyl chloride workers and the techniques for differentiating the electrophoretic patterns in patients with 003696 One 'uncred end tv.enty one c- ca-cezed 170 a-clnyees ''71 oercent) ca-v'cl^ated. ` si.v : :o : :a :c~adj:ad; 13 did not keep their scheduled appointment. After participation, nine employees were discovered to have been misclassified as biochemically abnormal. They had some biochemical abnormalities but not all (classified intermediate) and have been eliminated from the final analysis. Four biochemically abnormal individuals had abnormally low test values and they were included in the analysis. Liver Biopsies Liver biopsies were performed by the transjugular technique (11) and pro vided two to five biopsies from various areas of the liver. In addition, some individuals had second biopsies performed by the percutaneous or wedge biopsy via mini-laparotomy procedures. Pathological data was recorded in a computerized format identifying all histological abnormalities in a semi-quantitative fashion. Biopsies were read without knowledge of the individual's medical history or chemical exposure by two pathologists and a hepatologist with extensive experience in hepatic chemical injury. All biopsies were classified as 1) normal, 2) abnormal, a) chemical injury, and b) non-chemical injury. The non-biopsy groups were drawn from those currently employed, and based on biochemical liver "function tests" individuals were sorted into positive, negative and indeterminate for hepatic disease. Only the positive and negative are included in this study. The ultrasonic evaluation and its relative ability to identify hepatic damage due to vinyl chloride has been published elsewhere (3). Microvascular techniques and the method of evaluation by Dr. Maricq and co-workers are also published in part (8). The experimental work on GAG excretion in vinyl chloride workers and the techniques for differentiating the electrophoretic patterns in patients with CMA 003697 4 angiosarcoma and connective tissue damage of the liver has been published else where (9). The effectiveness of radioisotopic {radionucleotide) scanning as a technique for identifying anatomical lesions in vinyl chloride workers is in preparation (12). Method of Analysis The biopsied group and the non-biopsied (biochemical) group were analyzed independently. For the biopsied group sensitivity and specificity were estimated for each screening test by assuming that the biopsy was correct. For the bio chemical group the biochemical classification was assumed to be correct. For each analysis the data consisted of a simple cross classification. In a perfect screening test, the sum of sensitivity and specificity would be two. In a screening test which provided results no better than could be obtained by using a biased coin, the sum of sensitivity and specificity would be equal to one. We, therefore, estimated 95% confidence limits for the sum of sensitivity and spec ificity and observed whether or not one is included within these limits. As a 2 test of statistical significance this is equivalent to the usual X test for independent proportions. Finally, in Table 4 we looked at the association (as measured by the phi coefficient) between each pair of screening tests. For these comparisons we used all employees who received both screening tests regardless of their biopsy status. In this case, we assumed both tests were subject to error and estimated the phi coefficient (**) between them. Finally, for completeness, we give the biochemical classification for the 51 employees included in the biopsy group. Results Table 2 compares the histological and biochemical results for the 51 biopsied employees included in the study. Twenty-two of these employees had biochemical abnormalities as defined in Table 1. There was no significant CMA 003698 5 correlation between the biochemical and biopsy classification for the 29 employees with positive or negative biochemical classifications (r$ = 0.21; X? * 0.20). The biochemical studies used in this analysis were those determined at the time of this and not at the time the biopsy was performed. In all instances of disagreement, the biopsy was positive and the biochemical results negative (P< 0.001). Figure 2 provides the sensitivity and specificity for each of the screening tests when compared to biopsy results. In no case is the sum significantly greater than one, indicating that the results are not statistically significantly better than could be obtained using the biased coin. With the exception of the GAG studies, similar results are obtained when comparing the sum of sensitivity and specificity in the biochemical group (Figure 3). The sum of sensitivity and specificity for the GAG studies are just statistically significant with 95 percer^| confidence limits of 1.06 to 1.52. Table 3 gives the frequency distribution of the results of the GAG test for employees with normal and abnormal biochemical results. The distributions differ in their spread (variance) and not in their location (means). Finally, the correlation matrix for the four tests are given in Table 4. There is no significant correlation, as measured by rf, between any pair of the screening tests. This Is also true when they are sorted by biopsy status. After reviewing the results, a reclassification of the 51 employees who had biopsies was assessed in regard to whether there was chemically induced liver damage. Ultrasonographic evaluation was reclassified by Dr. Taylor and the liver biopsies by Drs. Tamburro and Popper. Table 5 gives the results of this additional analysis which demonstrates that there was no agreement that could not be explained by chance (P * 0.60). CMA 003699 ,'i.iK.Qan " a ^ ^ ^ " `i t " 1 ' * ` ' / * '; " I * ` "i " 1J '3 " 1 '" Z Z _ ,; tcs'1 r i v""? j' ' c " a--, i' s' " 1 ' 2 s i f i i j,"," = ., 1* : ,.; = Q.ZO). The biochemical studies used in this analysis were those determined at the time of this and not at the time the biopsy was performed. In all instances of disagreement, the biopsy was positive and the biochemical results negative (P< 0.001). Figure 2 provides the sensitivity and specificity for each of the screening tests when compared to biopsy results. In no case is the sum significantly greater than one, indicating that the results are not statistically significantly better than could be obtained using the biased coin. With the exception of the GAG studies, similar results are obtained when comparing the sum of sensitivity and specificity in the biochemical group (Figure 3). The sum of sensitivity and specificity for the GAG studies are just statistically significant with 95 percent confidence limits of 1.06 to 1.52. Table 3 gives the frequency distribution of the results of the GAG test for employees with normal and abnormal biochemical results. The distributions differ in their spread (variance) and not in their location (means). Finally, the correlation matrix for the four tests are given in Table 4. There Is no significant correlation, as measured by r$, between any pair of the screening tests. This is also true when they are sorted by biopsy status. After reviewing the results, a reclassification of the 51 employees who had biopsies was assessed in regard to whether there was chemically induced liver damage. Ultrasonographic evaluation was reclassified by Or. Taylor and the liver biopsies by Drs. Tamhurro and Popper. Table 5 gives the results of this additional analysis which demonstrates that there was no agreement that could not be explained by chance (P = 0.60). CMA 003700 6 Discussion The increasing industrialization in highly developed Western countries, such as the United States, continues to provide concern, not only for the health and safety of the industrial workers, but also for the surrounding communities in the areas of these industries. It is highly desirable to identify and validate the reliability of screening and diagnostic techniques which will identify the early development of Injury due to the exposure of a variety of chemicals such as vinyl chloride. The assessment of newly developing techniques on a high-risk, exposed worker population, who have been carefully screened and prospectively followed, provide the most reliable method for determing both the sensitivity and specificity of these technical procedures in the asymptomatic subclinical high- risk exposed community population. The failure of such techniques to provide sufficient sensitivity in the presence of a required specificity Is of critical clinical importance. This is especially so where the Incidence of disease is relatively low and the population exposed large. Tests which provide a high sensitivity but of low specificity can and do medically stigmatize the population under surveillance leading to unnecessary anxiety and socioeconomic disturbances which can far outweigh the benefit of early detection of even serious disease in a smaller population. Far too often screening techniques which have been developed in a highly diseased, clinically overt, hospitalized population are applied to an asymptomatic, clinically well-working populations without adequate determination of the sensi tivity and specificity at this earlier stage of disease development. In this study, all four techniques had, in the highly diseased hospitalized population, demonstrated either a sensitivity or specificity suggestive for the Identification of underlying chemically related liver disease. Some techniques (Maricq and Kupchella) appeared ideal for community studies since they were non- invasive, relatively inexpensive, and provided a means of screening which would be highly accepted by a community. CMA 003701 7 This prospectively designed study has allowed us to estimate the ability of ultrasonography, nailbed capillary assessments, radioisotopic scanning, and glycosaminoglycan excretions to correctly predict the presence and absence of hepatic disease as documented by an exposed population. These studies clearly show that none of the four screening techniques sufficiently agree with either the biopsy, the biochemical results, or each other in a well defined population; they do not provide sufficient sensitivity or specificity to be useful as early indicators of chemical exposure injury. The inclusion of nine individuals with intermediate biochemical results, who were originally misclassified as abnormal, would decrease the sum of sensi tivity and specificity in all three screening tests. With their exclusion only the GAG tests provided results better than might be expected by chance (P< 0.05). Even the GAGs, from a pragmatic point of view, provide too high a false positive rate to be useful in its present stage. Possibly, with increased refinement and further study, this might provide a simple non-invasive technique for the identification of increased collagen changes related to chemical injury. More i[mediately, it should be duplicated to rule out chance. Had we eliminated the four employees with an abnormally low biochemical test values, the sum of sensitivity and specificity for the GAGs and nailbed skin capillary screening tests would have been slightly reduced and that for the scan slightly Increased. The GAG would still be of bordering significant (X^ = 3.73), and the scan not significant (x| * 1.46). Finally, the tests not only disagree with each other, but within the biopsy group there was no agreement between the biopsies and biochemical results. It should be noted, however, that the biochemical studies used in analysis were those chronologically closest to September, 1977, and not to the date of biopsy. CMA. 003702 7 i ;:tively cas'j.ied study has allcwad '.a so essimaoa a.is sj;''Sv of r-ccrspry, nail bed cacillary assessments, raaic i sotopic scanning, and glycasaminoglycan excre ;ons to correctly predict the presence and absence of hepatic disease as documented by an exposed population. These studies clearly snow that none of the four screening techniques sufficiently agree with either the biopsy, the biochemical results, or each other in a well defined population; they do not provide sufficient sensitivity or specificity to be useful as early indicators of chemical exposure injury. The inclusion of nine individuals with intermediate biochemical results, who were originally misclassified as abnormal, would decrease the sum of sensi tivity and specificity in all three screening tests. With their exclusion only the GAG tests provided results better than might be expected by chance (P< 0.05). Even the GAGs, from a pragmatic point of view, provide too high a false positive rate to be useful in its present stage. Possibly, with increased refinement and further study, this might provide a simple non-invasive technique for the identification of increased collagen changes related to chemical injury. More iirmediately, it should be duplicated to rule out chance. Had we eliminated the four employees with an abnormally low biochemical test values, the sum of sensitivity and specificity for the GAGs and nailbed skin capillary screening tests would have been slightly reduced and that for the scan slightly increased. The GAG would still be of bordering significant (X? * 3.73), and the scan not significant (x| * 1.46). Finally, the tests not only disagree with each other, but within the biopsy group there was no agreement between the biopsies and biochemical results. It should be noted, however, that the biochemical studies used in analysis were those chronologically closest to September, 1977, and not to the date of biopsy. CMA 003703 8 The test determinations and the biopsies may have been as long as three years apart. An analysis of the biochemical tests value done at the time of the biopsy would have more accurately reflected the liver status, as shown by histology (15). It has been shown, in previously published studies, that biochemical and histological findings each correlate with chemical injury and chemical exposure (16, 17). CMA 003704 9 References 1. Maltoni, C. and Lefemine, G. Carcinogenicity to bioassays of vinyl chloride I. Research plan and early results. Environmental Perpectives, 7:387-405, 1974. 2. Creech, J. L. and Johnson, M. N. Angiosarcoma of the liver in the manufac ture of polyvinyl chloride. Journal of Occupational Medicine, 16:150-151, 1974. " 3. Taylor, K. J. W., Williams, D. M. J., Smith, P. M. and Dach, B. W. Grey scale ultrasonography for monitoring industrial exposure to hepatotoxic agents. Lancet, i_: 1222-1224, 1975. 4. Whelan, J. G. Jr., Creech, J. L. and Tamburro, C. H. Angiographic and iso topic characteristics of hepatic angiosarcoma found in vinyl chloride workers. Radiology, 118:549-557, March 1976. 5. Maricq, H. R., Johnson, M. N., Whitstone, C. L. and LeRoy, E. C. Capillary abnormalities in polyvinyl chloride production workers. JAMA, 236, 1368- 1371, 1976. ----- 6. Kupchella, C. E. and Tamburro, C. H. Urinary and tissue glycosaminoglycan patterns in angiosarcoma and other vinyl chloride exposure - associated livej^ injury. In Prevention and Detection of Cancer, Part I, 1:915-926, ed. Niebergs, H., Marcel Dekker, Inc., 1977. 7. Taylor, K. J. W., Carpenter, D. A., Hill, C. R. and McCready, V. R. Grey scale ultrasound imaging the anatomy and pathology of the liver. Radiology, 119: 415-423, 1976. 8. Maricq, H. R. and LeRoy, E. C. Patterns of finger capillary abnormalities in connective tissue disease by wide field microscopy arthritis pheum, 16: 619-629, 1973. -- 9. Curran, K. L., Kupchella, C. E. and Tanfcurro, C. H. Urinary glycosaminoglycan patterns in angiosarcoma of the liver. Cancer, 40:3050-3053, 1977. 10. Whelan, J. G. Jr., Greenberg, R. and Tamburro, C. H. The effectiveness of radioisotopic scans and grey scale ultrasonography in. the detection of liver damage. Gastroenterology, 79:1129, 1980. 11. Rosch, J., Anconovic, R. and Dotter, C.T. Transjugular approach to the liver, biliary system and portal circulation. American Journal of Roentgenology, 125:602-608, 1975. 12. Whelan, J. G. Jr., Creech, J. L. and Tamburro, C. H. Primary liver cancer detection in vinyl chloride workers by radioisotopic scanning (In Preparation). 13. Taylor, K. J. W. Glees,J. P., Smith, T. A. and Carpenter, D. A. Ultrasonic examination of the liver. In Ultrasound in Medicine, Vol, 2j pp. 173-174 (Eds.) White, D. N. and Barnes, R., Plenum Press, flew York, 1976. CMA 003705 1. Mai torn, C. and Lefemine, G. Carcinogenicity to bioassays of vinyl chloride I, Research plan and early results. Environmental Perpectives, 7:387-405, 1974. 2. Creech, J. L. and Johnson, M. N. Angiosarcoma of the liver in the manufac ture of polyvinyl chloride. Journal of Occupational Medicine, 16:150-151, 1974. -- 3. Taylor, K, J. W., Williams, 0. M. J., Smith, P. M. and Oach, B. W. Grey scale ultrasonography for monitoring industrial exposure to hepatotoxic agents. Lancet, i_:1222-1224, 1975. 4. Whelan, J. G. Jr., Creech, J. L. and Tamburro, C. H. Angiographic and iso topic characteristics of hepatic angiosarcoma found in vinyl chloride workers. Radiology, 118:549-557, March 1976. 5. Maricq, H. R., Johnson, M. N., Whitstone, C. L. and LeRoy, E. C. Capillary abnormalities in polyvinyl chloride production workers. JAMA, 236, 1368- 1371, 1976. ----- 6. Kupchella, C. E. and Tamburro, C. H. Urinary and tissue glycosaminog!yean patterns in angiosarcoma and other vinyl chloride exposure - associated liver injury. In Prevention and Detection of Cancer, Part I, 1:915-926, ed. Niebergs, H., Marcel Dekker, Inc., 1977. 7. Taylor, K. J. W., Carpenter, D. A., Hill, C. R. and McCready, V. R. Grey scale ultrasound imaging the anatomy and pathology of the liver. Radiology, 119: 415-423, 1976. 8. Maricq, H. R. and LeRoy, E. C. Patterns of finger capillary abnormalities in connective tissue disease by wide field microscopy arthritis pheum, 16: 619-629, 1973. ~ 9. Curran, K. L., Kupchella, C. E. and Tanturro, C. H. Urinary glycosaminoglycan patterns in angiosarcoma of the liver. Cancer, 40:3050-3053, 1977. 10. Whelan, J. G. Jr., Greenberg, R. and Tanturro, C. H. The effectiveness of radioisotopic scans and grey scale ultrasonography in. the detection of liver damage. Gastroenterology, 79:1129, 1980. 11. Rosch, J., Anconovic, R. and Dotter, C.T. Transjugular approach to the liver, biliary system and portal circulation. American Journal of Roentgenology, 125:602-608, 1975. 12. Whelan, J. G. Jr., Creech, J. L. and Tamburro, C. H. Primary liver cancer detection in vinyl chloride workers by radioisotopic scanning (In Preparation). 13. Taylor, K. J. W,, Glees,J. P., Smith, T. A. and Carpenter, 0. A. Ultrasonic examination of the liver. In Ultrasound in Medicine. Vol. pp. 173-174 (Eds.) White, D. N. and Barnes, R., Plenum Press, Hew York, 1976. CMA 003706 10 14. Taylor, K. J. W. and Carpenter, D. A. Comparison of radioisotopic and ultra sound examination in the investigation of hepatobiliary disease. In Ultrasound in Medicine, Vol. 1, pp. 159-167 (Ed.) White, D. N., Plenum Press, New York, 1976. 15. Clarmont, R.J. and Chalmers, T.C. The transaminase tests in liver disease. Medicine, 46:197-207, 1967. 16. Tamburro, C.H. and Greenberg, R. Effectiveness of federally-required medical laboratory screening in the detection of chemical liver injury. Environmental Perspectives, 1980. In Press. 17. Tamburro, C.H. and Greenberg, R.A. Identification of human toxicity and carcinogenicity by ethylene derivatives in mechanisms of toxicology and hazard evaluation (eds. Holmstedt, B., et al) Elsevier/North-Holland Biomedical Press, New York, New York, pp. 319-334, 1980. 18. Tamburro, C.H., Makk, L. and Popper, H. Early hepatic histological alterations among chemical (vinyl monomer) workers. Gastroenterology, 77:A43, 1979. CMA 003707 Table 1 CLASSIFICATION Biopsy Biochemical Ultrasound Microvascular DETERMINED BY CRITERIA 1) Dr. Popper, Pathologist a) Positive by consensus agreement if medically significant pathology 2) 3) Dr. Makk, Pathologist Dr. Tamburro, Hepatologist is present. Negative otherwise. b) Pathology is of chemical or non chemical origin. (18) Liver "Function" Tests Group 1 Group 2 SGPT GGTP ICG BILIRUBIN SGOT ALK. PHOSPHATASE Positive if two or more Group tests were abnormal or if one Group 1 and both Group 2 tests were abnormal. Negative if all six tests were normal. Intermediate otherwise. Dr. Taylor Defined as negative if the over all impression was normal. Posi tive otherwise. Dr. Maricq Defined as negative if there were no microvascular abnormalities. Positive otherwise. Glycosaminoglycans uromc _ UG Uronic Acid acit* MG Creatinine Defined as positive for values less than 2.0 or greater than 4.8. Negative otherwise. Liver Scan Dr. Whelan Defined as positive if any patho logical defect was detected. Negative Otherwise. CMA 0 0 3 7 0 8 TaDie i 003709 CLASSIFICATION Biopsy Biochemical Ultrasound Microvascular DETERMINED BY CRITERIA 1) 2) 3) Dr. Dr. Dr. Popper, Pathologist Makk, Pathologist Tamburro, Hepatologist a) Positive by consensus agreement if medically significant pathology 'is present. Negative otherwise. b) Pathology is of chemical or non chemical origin. (18) Liver "Function" Tests Group 1 Group 2 SGPT GGTP ICG BILIRUBIN SGOT ALK. PHOSPHATASE Positive if two or more Group tests were abnormal or if one Group 1 and both Group 2 tests were abnormal. Negative if all six tests were normal. Intermediate otherwise. Dr. Taylor Defined as negative if the over all impression was normal. Posi tive otherwise. Dr. Maricq Defined as negative if there were no microvascular abnormalities. Positive otherwise. olycosnminoglycans Uronic _ UG Uronlc Acid acld MG ^;reatinine Defined as positive for values less than 2.0 or greater than 4.8. Negative otherwise. Liver Scan Dr. Whelan Defined as positive if any patho logical defect was detected. Negative Otherwise. TABLE 2 COMPARISON OF BIOPSY AND BIOCHEMICAL DETERMINATIONS OF THE PRESENCE OF LIVER DISEASE 12 BIOPSY BIOCHEMICAL POSITIVE ' NEGATIVE SUM B IOCHEMICAIlY TOTAL INDETERMINATE POSITIVE NEGATIVE SUM 3 0 3 18 21 88 26 29 r = 0.20966 X? 0.200 N.S. 2 Matched xj - (18 " 1) /l8 -16.1 P < 0.001 15 7 22 36 15 51 CMA 003710 TABLE 3 13 FREQUENCY (F) AND RELATIVE FREQUENCY (R.F.) OF GAGS FOR NORMAL AND ABNORMAL BIOCHEMICAL RESULTS GAG Normal ------ 1 R.F. <2 2<3 3<4 4<5 5+ 4 16 13 1 2 0.111 0.444 0,361 0.028 0.056 Sum Mean Variance 36 1.000 2.886 0.746 Abnormal f R.F. 6 0.250 7 0.292 6 0.250 1 0.042 4 0.167 24 1.000 3.160 1.776 t = 0.968 N.S. S8 F = 2.381 P < 23,35 T Test Independent Means 0.05 F Test Independent Variances ( Two Taile CMA 003711 IA3LS 3 FREQUENCY (F) AND RELATIVE ERECL'ENCY (?..?.) OF GAGS FOR NORMAL AMD A3NOP01AL ^ j,~\ n* 1X in *_> S O Jj j- GAG <2 2<3 3<4 4<S 5+ Sum Mean Variance Normal a. R.F. 4 0.111 16 0.444 13 0.361 1 0.028 2 0.056 36 1.000 2.886 0.746 Abnormal c1_ R . F . 6 0.250 7 0.292 6 0.250 1 0.042 4 0.167 24 1.000 3.160 1.776 t * 0.968 N.S. T Test Independent Means $ F = 2.381 P < 0.05 F Test Independ nt 23/35 Variances { Two Taile CMA 003712 14 TABLE 4 CORRELATION MATRIX (Ro) BETWEEN POUR SCREENING TESTS USED TO PREDICT THE PRESENCE OR ABSENCE OIF LIVER DISEASE GAG CAPILLARY ANALYSIS ASSESSMENT GAG ANALYSIS 0.08 CAPILLARY ASSESSMENT (113) * ULTRASOUND STUDY (87) (84) RADIOISOTOPIC SCAN (120) (114) ULTRASOUND STUDY -0.03 0.03 (88) RADIOISOTOPIC SCAN 0.005 0.07 0.06 None of the correlations are statistically significant (a = 0.05). The correlations are given above the diagonal. The sairple size is given in parenthesis below the diagonal. CMA 003713 TABLE 5 KECLASSIFICATTCN CF BICPSIED EMPLOYEES FOR CHEMICALLY INDUCED ABNORMALITIES Biopsy Ultra Sound Chemical Abnormality Other Chemical Abnormality Other 7 26 5 13 SUM 33 18 SUM 12 39 51 Specificity = 13/39 = 0.333 Sensitivity 35 7/12 = 0.583 Sum - 0.916 CMA 003714 TABLE 5 KECLftSSIFICATICN CF BICPSHD EMPLOYEES FOR CHEMICALLY INDUCED ABNOH-ffiLITIES Biopsy Ultra Sound Chemical Abnormality Other Chemical Abnormality Other 7 26 5 13 SUM 33 18 SUM 12 39 51 Specificity * 13/39 = 0.333 Sensitivity = 7/12 = 0.583 Sum = 0.916 CMA 003715 GROUP A FIGURE 1 GROUP B 16 *Not seen by Doctor Taylor CMA 003716 luuaii ^ >> +j O (OU a co xJ c ci SO CO \ 1 H c0) CO 0 Nurrfoer Employees Positive Biopsies Negative Biopsies SENSITIVITY AND SPECIFICITY FOR FOUR SCREENING TESTS FOR THE PREDICTION OF LIVER ABNORMALITIES {AS DETERMINED BY BIOPSY) .02 22 1.13 -j" - Sum: | plUS I'*,.,., Specificity Sensitivity T ^95% Limits / 1.04 .53 .75 80 .60 .40 .64 [Sensitivity [Specificity 40 GAG (51) 36 15 Microvsscular Ultrasound (49) 34 15 (51) 36 15 Scan (51) 36 15 CMA 0 0 3 7 1 7 '+H>j, U 'H oa> tao "D cr;l ;> <} r.-t (HJ ri os u 10 ] 0 Nidi her fV.ployees Posit, hr* Biopsies Nrpat.i'v Biopsies SENSITIVITY AND SPECIFICITY FOR FOUR SCREENING TESTS FOR THE PREDICTION OF LIVER ABNORMALITIES (AS DETERMINED BY BIOPSY) 36 34 36 36 15 15 15 15 CMA 0 3 7 1 8 CMA 0 0 3 7 1 9 TM +>Hj> O <HH 0o) wa a co +-> > 4HJ CO c CO Figure 3 SENSITIVITY AND SPECIFICITY FOR FOUR SCREENING TESTS FOR THE PREDICTION OF LIVER ABNORMALITIES (AS DETERMINED BY REVIEWED BIOCHEMICAL TESTS) Sum: Specificity Nurriber Bnployees Positive Biochemical Negative Biochemical (60) 24 36 (57) 22 35 (37) (61) 24 37 STUDIES ON THE MUTAGENICITY OF VINYL CHLORIDE METABOLITES AND RELATED CHEMICALS A. D. Laumbach, S. Lee, J. Wong, and U. N. Streips Department of Microbiology and Immunology University of Louisville School of Medicine Department of Chemistry Louisville, Kentucky 40201 I. INTRODUCTION The studies by Viola et al (26) and Maltoni et al (15) established the carcinogenic potential of vinyl chloride monomer. The detection of angiosarcoma in industrial workers exposed to polyvinyl chloride suggested a causal relationship between this chemical and the development of hepatic abnormalities (4,12). Hefner et al (7) have delineated the metabolic fate of inhaled vinyl chloride in rats and proposed that the epoxide, chlorooxirane, and chloroacetaldehyde were the carcinogenic intermediates. Their hypothesis has been supported by the work of several laboratories (3,14,16,19, Elmore, Wong, Laurabach and Streips, submitted for publication) using bacterial strains as mutagenic indicators. In this communication we present additional data concerning the mutagenicity and the potential mechanisms of action of several vinyl chloride metabolites, including the previously unreported chloroacetadehyde monomer hydrate, chloroacetaldehyde dimer hydrate, and chloroacetaldehyde trimer. Epichlorohydrin, a mutagenic/carcinogenic (21,25) methylene homolog of chlorooxirane was also examined.II. II. PROCEDURES AND MATERIALS USED A. Bacterial Strains The bacterial strains utilized in these studies are presented in Table I. The Bacillus subtilis strains were all maintained on AK agar (BBL). Salmonella typhimurium cultures were obtained from B. N. Ames (l) and were stored on Nutrient Agar (Difco) plus 5g NaCl per liter. CMA 003720 Laumbach, A.D., Lee, S., Wong, J., and Streips, U.N. B. Mutagenicity Assays The indirect assay utilized repair deficient strains of B. subtilis. The procedure was a modification of the "rec-assay" described by Kada et al (9). Cells were grown overnight in Nutrient Broth (Difco) at 37C in a rotary incubator shaker, then diluted tenfold in phosphate buffer (pH 7.0). The suspended cultures were streaked onto Nutrient Agar plates (Difco). Filter paper discs (6 mm) were saturated with the chemical solutions to be examined, then were placed onto the agar plates next to the streaked bacterial cultures. Following incubation at 37C overnight the plates were examined and the lethality and mutagenic potential of the test chemicals were assessed by comparing inhibition zones between the B. subtilis 168 wild type, a repair-capable strain and the various DNA repair-deficient strains. In all these studies 4-nitroquinoline-l-oxide (4NQQ) was used as the positive mutagenic control. Direct mutagenicity assays utilized the S. typhimurium tester strains described by Ames (l). The chemicals were examined by the methods of McCann et al (16). The cultures were grown in Nutrient Broth plus 0.5/5 NaCl overnight in a rotary incubator shaker at 37C. A mixture of the test chemical (0.1 ml) in dimethyl sulfoxide (DMSO) and 2 ml of soft agar (0,6? agar, 0.6? NaCl, 0.5 mM biotin, and 0.5 mM histidine) was added to 0.1 ml of the bacterial culture. The solutions were mixed thoroughly and overlaid onto minimal plates [Vogel-Bonner E medium (27), 1.5? agar, and 2? glucose]. Control samples were prepared by omitting the test chemicals. For the positive mutagenesis control, 4NQ0 was added to the mixtures in place of the test chemicals. All plates were incubated for 48 hr at 37C prior to the enumeration of revertant colonies. C. Chemical Compounds The chemical compounds utilized in these studies were prepared, purified, and analyzed by previously reported techniques (Elmore, Wong, Laumbach, and Streips, submitted for publication). D. Preparation of DNA Transforming DNA was isolated from B. subtills cultures by the method described by Young and Wilson (297. In some of the experiments the cultures were pretreated for 15 min either with chloroacetaldehyde (16 mM) or epichlorohydrin (16 mM) prior to the extraction procedure. In alternate experiments S-9 liver homogenate mix was added to the compounds prior to addition to bacteria. The S-9 liver homogenate contains per ml, 0.3 ml of the S-9 fraction, 8 mMMgCl2> 33 mM KD1, 5 mM glucose-6-P, 4 mM NADP, and 100 mM sodium phosphate (pH 7.4). The DNA concentration in all lysates were assayed by the method of Richards (20). 003721 E. Treatment of DNA In vitro with Chemicals A sample of B. subtilis transforming DNA (0.9 ml) in standard saline citrate (SSC) (0.15 M MaCl-0.015 M trisodium citrate, pH 7.0) was combined with 0.1 ml chloroacetaldehyde (1.0 M in DMSO) or 0.1 ml U STUDIES ON THE MUTAGENICITY OF VINYL CHLORIDE METABOLITES epichlorohydrin (1.0 M in DMSO). The mixture was allowed to react for 1 hr v.ith occasional shaking. Following this treatment the treated DNA was dialyzed at OC against three 500 ml changes of SSC for 24 hrs. In alternate experiments the DNA-chemical mixtures were placed in a dialysis bag and immersed in the S-9 liver homogenate mix. These samples were dialyzed in SSC as above. F. Competent Cultures for Transformation Assays The procedures for the development of competence were similar to those described (23). B. subtilis cells were grown in a modified Spizizen's minimal medium (GMI) (29) for 90 min at 37C after cessation of logarithmic growth in a rotary incubator shaker. The cells were then diluted tenfold into GMII medium (29) and incubated for an additional 60 min at 370 in the shaker. At this time the culture has attained maximum competence. G. Transformation Procedures A sample (0.1 ml) of extracted, treated or untreated DNA was added to 0.8 ml of the competent cultures and incubated at 37C for 30 min in the shaker. The reaction was terminated by the addition of 0.1 ml of deoxyribonuclease (500 yg/ml, Worthington Biochem. Corp.) for 15 min at 37C. The cells were plated on appropriate selective minimal media and incubated at 37C for 48 hrs. III. RESULTS A summary of preliminary mutagenesis screening experiments with potential vinyl chloride monomer metabolites and related compounds is presented in Table II. It is evident that chlorooxirane and chloroacetaldehyde are the ultimate mutagens in this system. These results agree with the published data (3,16). In addition, this table describes the mutagenicity of the other chemical forms of chloroacetaldehyde, not ably a monomer hydrate, a dimer hydrate, and a trimer. The hydrate and dimer hydrate forms have been shown to form an equilibrium mixture by the spontaneous rearrangement of chloroacetaldehyde under physiological conditions (Elmore, Wong, Laumbach, and Streips, submitted for publication),and these hydrate forms must be regarded as potential metabolites of consequence. Purified dimer hydrate and trimer were synthesized under laboratory conditions. Neither acetaldehyde, chloroacetic acid, nor chloroethanol showed a significant level of mutagenicity in these assays. Other investigators have reported the mutagenicity of chloroethanol, however, either high concentrations or activation with microsomal enzymes was required for activity (3,16). Our results agree with those of McCann et al (16). These experiments suggest a molecular relationship involving the proximity of the chloride group to the aldehyde moiety for mutagenic activity. In this regard we are currently examining structurally analogous ketones, substituted with various halogens. Epichlorohydrin (l-chloro-2,3 epoxypropane) was also mutagenic in screens using the Salmonella tester strain TA100. o ea i or* 3 * in 'ere placed in a dialysis bag and immersed in the S-9 liver homogenate mix. These samples v/ere as above F n ompetent Cultures for Transformation Assays The procedures for the development of competence were similar to those described (23). B. subtilis cells were grown in a modified Spizicen's minimal medium (GM) (29) for 90 min at 37C after cessation of logarithmic growth in a rotary incubator shaker. The cells were then (29) and incubated for an additional 60 min at 37C in the shaker. At this time the culture has attained maximum e G. Transformation Procedures A sample (0.1 ml) of extracted, treated or untreated DNA was added to 0.8 ml of the competent cultures and incubated at C for 30 min in the shaker. The reaction was terminated by the addition of 0.1 ml of deoxyribonuclease (500 yg/ml, Worthington Biochem. Corp.) for 15 min at 37C. The cells were plated on appropriate selective minimal media and incubated at 37C for 48 hrs. III. RESULTS A summary of preliminary mutagenesis screening experiments with potential vinyl chloride monomer metabolites and related compounds is presented in Table II. It is evident that chlorooxirane and chloroacetaldehyde are the ultimate mutagens in this system. These results agree with the published data (3,16). In addition, this table describes the mutagenicity of the other chemical forms of chloroacetaldehyde, not ably a monomer hydrate, a dimer hydrate, and a trimer. The hydrate and dimer hydrate forms have been shown to form an equilibrium mixture by the spontaneous rearrangement of chloroacetaldehyde under physiological conditions (Elmore, Wong, Laumbach, and Streips, submitted for publication),and these hydrate forms must be regarded as potential metabolites of consequence. Purified dimer hydrate and trimer were synthesized under laboratory conditions. Neither acetaldehyde, chloroacetic acid, nor chloroethanol showed a significant level of mutagenicity in these assays. Other investigators have reported the mutagenicity of chloroethanol, however, either high concentrations or activation with microsomal enzymes was required for activity (3,16). Our results agree with those of McCann et al (16). These experiments suggest a molecular relationship involving the proximity of the chloride group to the aldehyde moiety for mutagenic activity. In this regard we are currently examining structurally analogous ketones, substituted with various halogens. Epichlorohydrin (l-chloro-2,3 epoxypropane) was also mutagenic in screens using the Salmonella tester strain TA100. CMA 003723 Laumbach, A. D.,Lee, S., Wong, J., and Streips, U. N. Further experiments examined the effect of proposed metabolites on several different DNA repair deficient strains of B. subtilis. Chlorooxirane and the different forms of chloroacetaldehyde were all found to specifically inhibit the growth of strain MC-1, which lacks recombination repair (17) (Table III). Epichlorohydrin was capable of moderate reactivity only in the presence of the S-9 fraction. Quantitative mutagenesis assays with Salmonella strain TA100, an indicator for base-pair substitution mutations, revealed that chloro acetaldehyde monomer had the highest mutagenic capacity of all the reactive metabolites (Table IV). The monomer-dimer hydrates, dimer hydrate, and trimer show progressively decreasing mutagenic efficiency as evidenced by the higher chemical concentration required for eliciting maximum reversion. All forms of chloroacetaldehyde were very toxic, thus the mutagenic response of each compound was limited to a narrow range of concentrations. However, epichlorohydrin, a weak mutagen by comparison, has a broad mutagenic spectrum and a corresponding low toxicity. Since the mutagenic activity of the compounds constituted strong evidence that DNA was a primary target of attack, we examined the inter action of chloroacetaldehyde and epichlorohydrin with transforming DNA. It is known that the biological activity of transforming DNA can be altered by exposure to physical and chemical agents (3,22). Previous studies have shown that chloroacetaldehyde can bind to DNA in vitro (ll). Accordingly, transforming DNA isolated from B. subtilis 168W was treated with either chloroacetaldehyde or epichlorohydrin as described in Materials and Methods. The treated DNA was examined in transformation assays utilizing several different auxotrophic strains of B. subtilis as the recipients. Data presented in Table V reveals that in vitro treatment of DNA with either compound has little or no apparent efTect on the biological activity of this DNA in transformation. Since both chloroacetaldehyde and epichlorohydrin demonstrated mutagenic activity in the Salmonella TA100 strain, we examined the effect of these two compounds on B. subtilis DNA in vivo. Transforming DNA was isolated from B. subtilis following a 15 min exposure to the mutagenic chemicals. The" DNA concentration was calculated from these samples, and levels equivalent to those used in the in vitro assays were added to competent cultures. The results of these transformation assays are shown in Table VI. Two major effects are evident with chloroacetaldehyde in vivo treated DNA. First, there was a major depression of the biological activity in the transforming DNA. Secondly, the depression showed genetic marker specificity. Moreover, the DNA segments containing genetic markers which have previously been shown to be associated to macromolecular structures such as the cell membrane (6,24,28) or the cell wall (Streips, Doyle, Sueoka, Brown, and Fan, submitted for publication) were selectively protected from attack by chloroacetaldehyde and epichlorohydrin. The activity of epichlorohydrin was less in these experiments, however, the patterns of specific marker inactivation are quite similar. The addition of the S-9 mix to the chemicals prior to addition to the cells, did not cause significant alteration in trans formation efficiency (results not shown). In some samples there was an effect on the transforming DNA by DMSO, therefore all transformation values were corrected to account for this parameter. CMA 003724 STUDIES ON THE MUTAGENICITY OF VINYL CHLORIDE METABOLITES IV. DISCUSSION The major findings reported in this manuscript can be summarized: 1) We have confirmed the mutagenicity of chioroacetaldehyde and chlorooxirane, and extended it to include the additional potential metabolites, chioroacetaldehyde monomer hydrate, dimer hydrate and trimer, as well as the previously unreported chlorooxirane homolog, epichlorohydrin, 2) We have shown that recombination repair appears to be the mechanism for the correction of vinyl chloride metabolite elicited damage. 3) Chioro acetaldehyde causes a decrease in the biological activity of transforming DNA only if the cells are treated with the mutagen prior to the extraction of the DNA. In vitro studies showed no effect. 4) Epichlorohydrin apparently differs markedly from the vinyl chloride metabolites in mutagenic activity. To understand the mutagenic potential of sin environmental carcinogen, such as vinyl chloride and related chemicals,it is necessary to determine both its metabolic fate and probable mechanism of action for alteration of cellular processes. This report, as well as others, (3,14,16) has identified the potential active metabolites in vinyl chloride monomer mediated carcinogenesis. Furthermore, on the basis of a series of studies in microbial systems, we can postulate probable mechanisms of action of the vinyl chloride monomer metabolites and related chemicals. Under standing of these mechanisms is necessary for the development of possible blocking agents to the carcinogenic activity. Recombination repair appears to be induced to correct DNA lesions caused by vinyl chloride monomer metabolites and epichlorohydrin. Salmonella strain TA10Q which lacks excision repair (uvr-), yet retains the capacity for recombination repair is capable of recovery and can express mutation following exposure. Furthermore, experiments with several repair-deficient B. subtilis mutants demonstrate that only the recombination repair mutant is specifically sensitive to the active metabolites, whereas the excision repair mutants and the wild type strain are relatively unaffected. The nature of the lesions may specifically evoke the recombination repair mechanism (10), or, alternatively, the chemical reactivity of the metabolites may directly suppress other repair. It is known that recombination repair is inducible, while other types of repair are mostly constitutive (5). Since chioroacetaldehyde has been shown to specifically interact with proteins containing -SH groups (J. Hoffman, personal communication), it is possible that the chemical could inactivate the constitutive repair enzymes leaving the repair to an inducible system. The requirement for recombination repair of damage induced by these chemicals suggests the potential route of mutagenesis in bacteria. Recombination repair has been shown to be error prone (16). In this sense it resembles postreplication repair in mammalian cells (13)- Thus, we can postulate that the analogous error prone repair pathway, post replication repair, may function in mammalian cells in response to vinyl chloride metabolite elicited damage. A relationship between post replication repair caused errors and somatic mutation and carcinogenesis a has been suggested in patients with the skin disease, xeroderma pig mentosum (13). 003725 .- zzr. ce : Ve have confirms.:: one mutagenici' ,y c:` chlcroacetaldehyde ar.d chlcrooxi~xr.e, si.r-^d 3X'26r.i"sa'2'1'. "i *" ~ ^ir.2J.udcs ids 3-ddiiior*2_l 'coisr^i-'`a. chloroacetaldehyde monomer hydrate, diner hydrate and trimer, as well as * v*r --- have shctm that recombination repair appears to be the mechanism for the correction of vinyl chloride metabolite elicited damage. 3) Chloro- acetaldehyde causes a tease m ;iological activity of transforming DMA only if the cells are treated with the mutagen prior to the extraction cf the DMA. In vitro studies showed no effect. 4) Epichlcrohydrin apparently differs r.arhedly from, the vinyl chloride metabolites in To 'understand she mucigenic pocencial of an environmental carcinogen, such as vinyl chloride and related chemicals,it is necessary to determine both its metabolic fate and probable mechanism of action for alteration of cellular processes. This report, as well as others, (3,14,16) has identified the potential active metabolites in vinyl chloride monomer mediated carcinogenesis. Furthermore, on the basis of a series of studies in microbial systems, we car. postulate probable mechanisms of action of the vinyl chloride monomer metabolites and related chemicals. Under standing of these mechanisms is necessary for the development of possible blocking agents to-the carcinogenic activity. Recombination repair appears to be induced to correct DNA lesions caused by vinyl chloride monomer metabolites and epichlorohydrin. Salmonella strain TA100 which lacks excision repair (uvr-), yet retains the capacity for recombination repair is capable of recovery and can express mutation following exposure. Furthermore, experiments with several repair-deficient B. subtilis mutants demonstrate that only the recombination repair mutant is specifically sensitive to the active metabolites, whereas the excision repair mutants and the wild type strain are relatively unaffected. The nature of the lesions may specifically evoke the recombination repair mechanism (10), or, alternatively, the chemical reactivity of the metabolites may directly suppress other repair. It is known that recombination repair is inducible, while other types of repair are mostly constitutive (5). Since chloroacetaldehyde has been shown to specifically interact with proteins containing -SH groups (J. Hoffman, personal communication), it is possible that the chemical could inactivate the constitutive repair enzymes leaving the repair to an inducible system. The requirement for recombination repair of damage induced by these chemicals suggests the potential route of mutagenesis in bacteria. Recombination repair has been shown to be error prone (16). In this sense it resembles postreplication repair in mammalian cells (13)- Thus, we can postulate that the analogous error prone repair pathway, postreplication repair, may function in mammalian cells in response to vinyl chloride metabolite elicited damage. A relationship between post replication repair caused errors and somatic mutation and carcinogenesis has been suggested in patients with the skin disease, xeroderma pig mentosum (13). CMA 003726 Laumbach, A. D., Lee, S., Wong, J., and Streips, U. N. The increased inhibitory activity of the chloroacetaldehyde dimer and triiner forms for the other repair-deficient B. subtilis strains (Table III) may have been nonspecific killing of the cells, since all the strains other than MC-1 showed identical levels of inhibition. The necessity for metabolic activation of epichlorohydrin could reflect either a lack of permeability of the nonactivated compound or the requirement of a metabolite of this compound as the true mutagenic species. Neither chloroacetaldehyde nor epichlorohydrin seemed to affect transforming DNA in vitro (Table V). Although several investigators have reported that CAA specifically modifies bases and causes mismatched base pairs (2,11), this reaction in vitro does not seem to affect the biological activity of the D^T. In contrast, DNA which was isolated from cells treated with either chloroacetaldehyde or epichlorohydrin (in vivo, Table VI) was severely affected. The overall biological activity of the transforming DNA is depressed,- and it appears that the regions of the genome which are not protected by either the cell membrane or cell wall are most susceptible to attack and inactivation. It has also been postulated both in Escherichia coli and B. subtilis that the replication origin, terminus, and replication fork are all outer surface bound (18,24). Thus, these would be protected regions from chloroacetaldehyde attack and the nonreplicating DNA in the cytoplasm would be most susceptible. In this connection, recent experiments in our laboratory (Laumbach, Lee, Wong, and Streips, manuscript in preparation) have shown that chloroacetaldehyde causes enhanced mutation levels in cultures with nonreplicating genomes. This may imply that chloroacetaldehyde could be active in mammalian cells during growth stages where little DNA synthesis occurs. The mode of action of epichlorohydrin, a known carcinogen (25), differs from that of the vinyl chloride monomer metabolites. Although epichlorohydrin causes similar base substitution mutations in Salmonella tester strain TA 100, it is a comparatively weaker alkylating agent based on quantitative assay. Epichlorohydrin also exhibits a lower toxicity level than vinyl chloride monomer metabolites, thus epichloro hydrin can demonstrate mutagenic activity through a wider range of concentrations. In addition, our laboratory has preliminary evidence that epichlorohydrin produces higher levels of mutation in Salmonella cultures which are actively replicating DNA than in cultures which have been arrested in DNA replication (Laumbach, Lee, Wong, and Streips, manuscript in preparation). The different activity spectra between chlorooxirane and its homolog epichlorohydrin points out the necessity for a multifaceted study of carcinogens. V. SUMMARY Our laboratories have utilized strains of B. subtilis and Salmonella typhimurium to investigate the mutagenicity of vinyl chloride* i tabolites and related compounds. The major findings reported in this mscript are: 1) Confirmation of mutagenicity of chloroacetaldehyde i chlorooxirane. 2) Description of mutagenicity of additional potential metabolites of vinyl chloride, chloroacetaldehyde monomer hydrate, dimer hydrate, and trimer, as well as the mutagenic CMA 003727 STUDIES ON THE MUTAGENICITY OF VINYL CHLORIDE METABOLITES carcinogenic chlorooxirane homolog, epichlorohydrin. 3) Recombination repair is postulated to be the mechanism for correcting vinyl chloride metabolite elicited damage. 4) Chloroacetaldehyde affects the transformation activity of DNA only if cells are treated with the mutagen prior to the extraction of the DNA. In vitro the chemical had no effect. 5) Epichlorohydrin differs from vinyl chloride metabolites in mode of action. VI. ACKNOWLEDGEMENTS We wish to thank Mary A. Kinnaman for her extremely able technical assistance. We are grateful to Dr. Jerald Hoffman for making available preliminary results and to Dr. B. N. Ames for providing the Salmonella tester strains. This work was supported by a grant from the B. F. Goodrich Company to the Cancer Center at the University of Louisville, School of Medicine. VII. LITERATURE CITED 1. Ames, B.N., Lee, F.D., and Durston, W.E. An Improved Bacterial Test System For Detection And Classification Of Mutagens And Carcinogens. Proc. Nat. Acad. Sci. U.S.A., 70: 782-786, 1973. 2. Barrio, J.R., Secrist, J.A., and Leonard, N.J. Fluorescent Adenosine And Cytidlne Derivatives. Biochem. Biophys. Res. Comm., 46: 597-604, 1972. 3. Bartsch, H., Malaveille, C., and Montesano, R.M. Human, Rat, And Mouse Liver-Mediated Mutagenicity Of Vinyl Chloride In S. typhimurium Strains. Int. J. Cancer, 15: 429-437, 1975. 4. Creech, J.L., and Johnson, M.N. Angiosarcoma Of The Liver In The Manufacture Of Polyvinyl Chloride. J. Occup. Med., 16: 150-151, 1974. 5. Ganesan, A.K., and Smith, K.C. Recovery Of Recombination Deficient Mutants Of Escherichia coli K-12 From Ultraviolet Irradiation. Cold Spring Harbor Symp. Quant. Biol., 33: 235-242, 1968. 6. Ganesan, A.T., and Lederberg, J.. A Cell-Membrane Bound Fraction Of Bacterial DNA. Biochem. Biophys. Res. Comm,, 18: 824-835, 1965. 7. Hefner, R.E., Watanabe, P.G., and Gehring, P.G. Preliminary Studies Of The Fate Of Inhaled Vinyl Chloride Monomer (VCM) In Rats. Ann. N. Y. Acad. Sci., 246: 135-148, 1975. 8. Jensen, R.A., and Haass, F.L. Analysis of Ultraviolet Light-Induced Mutagenesis By DNA Transformation In Bacillus subtilis. Proc. Nat. Acad. Sci. U.S.A., 50_: 1109-1116, 1963. 9. Kada, T., Tutikawa, K., and Sadaie, Y. In Vitro And Host-Mediated "Rec-Assay" Procedures For Screening Chemical Mutagens; And Phloxine, A Mutagenic Red Dye Detected. Mutation Res., 16: 165-174, 1972. CMA 003728 STUDIES ON l ) * A'j /"i-* v'---**i V> Vl a / ' " ^ ^ in *3 transformation activity of DNA only if cells are treated with che mutagen prior to the extraction of the DMA. In vitro the chemical had no effect. 5) Epichlorohydrin differs from vinyl"-chloride metabolites in mode of action. ACKNOWLEDGEMENTS Y/e wish to thank Mary A. Kinnaman for her extremely able technical assistance. We are grateful to Dr. Jerald Hoffman for making available preliminary results and to Dr. 3. N. Ames for providing the Salmonella J. a ^ - -- w * MJ C*.* - * - -rt' -..r- , ^ ,, j. ^ V,,* <3 If W * -- H ^*2 .J ^ w W ->*.<, <w ^ J* DX* -- * ,, - " Q U * *. y, .*4 * Goodrich Company to the Cancer Center at the University of Louisville, School of Medicine. VII. LITERATURE CITED 1. Ames, B.N., Lee, F.D., and Curston, W.E. An Improved Bacterial Test System For Detection And Classification Of Mutagens And Carcinogens. Proc. Nat. Acad. Sci. U.S.A., 70: 732-786, 1973. 2. Barrio, J.R., Secrlst, J.A., and Leonard, N.J. Fluorescent Adenosine And Cytidine Derivatives. Biochem. Biophys. Res. Comm., 46: 597-604, 1972. 3. Bartsch, H., Malaveille, C., and Montesano, R.M. Human, Rat, And Mouse Liver-Mediated Mutagenicity Of Vinyl Chloride In S. typhimurium Strains. Int. J. Cancer, 15: 429-437, 19757 4. Creech, J.L., and Johnson, M.N. Angiosarcoma Of The Liver In The Manufacture Of Polyvinyl Chloride. J. Occup. Med,, 16: 150-151, 1974. 5. Ganesan, A.K., and Smith, K.C. Recovery Of Recombination Deficient Mutants Of Escherichia coll K-12 From Ultraviolet Irradiation. Cold Spring Harbor Syrup. Quant. Biol., 33} 235-242, 1968. 6. Ganesan, A.T., and Lederberg, J. A Cell-Membrane Bound Fraction Of Bacterial DNA. Biochem. Biophys. Res. Comm., 18: 824-835, 1965. 7. Hefner, R.E., Watanabe, P.G., and Gehring, P.G. Preliminary Studies Of The Fate Of Inhaled Vinyl Chloride Monomer (VCM) In Rats. Ann. N. Y. Acad. Sci., 246: 135-143, 1975. 8. Jensen, R.A., and Haass, F.L. Analysis of Ultraviolet Light-Induced Mutagenesis By DNA Transformation In Bacillus subtilis. Proc. Nat. Acad. Sci. U.S.A., 50: 1109-1116, 1963. 9. Kada, T., Tutikawa, K., and Sadaie, Y. In Vitro And Host-Mediated "Rec-Assay" Procedures For Screening Chemical Mutagens; And Phloxine, A Mutagenic Red Dye Detected. Mutation Res., 16_: 165-174, 1972. CMA 003729 Laumbach, A. D., Lee, S., Wong, J., and Streips, U. N. 10. Laumbach, A. D., and Felkner, I. C. Formation Of A 4-Nitroquinoline-l-Oxide Complex With DNA In Normal And Repair-Deficient Strains Of Bacillus subtilis. Mutation Res., 15_: 233-245, 1972. 11. Lee, C. H., and Wetmur, J. G. Physical Studies Of Chloroacetaldehyde Labeled Fluorescent DNA. Biochem. Biophys. Res. Commun., 50: 879-885, 1973. 12. Lee, F. I., and Harry, D. S. Angiosarcoma Of The Liver In A Vinyl Chloride Worker. Lancet, 1: 1316-1318, 1974. 13* Lehmann, A. R. Postreplication Repair Of DNA In Mammalian Cells. Life Sci., 15: 2005-2016, 1974. 14. Malaveille, C. H., Bartsch, H., Montesano, R., Barbin, A., Camus, A. M., Croizy, A., and Jacquignon, P. Mutagenicity Of Vinyl Chloride, Chloroethylene Oxide, Chloroacetaldehyde And Chloroethanol. Biochem. Biophys. Res. Commun., 63: 363-370, 1975. 15. Maltoni, C., and Lefemine, G. Carcinogenicity Bioassays Of Vinyl Chloride. Environm. Res., 7_: 387-405, 1974. 16. McCann, J., Simmon, V., Streitwieser, D., and Ames, B. N. Mutagenicity Of Chloroacetaldehyde, A Possible Metabolic Product Of 1,2-Dichloroethane (Ethylene Dichloride), Chloroethanol (Ethylene Chlorohydrin), Vinyl Chloride, And Cyclophosphamide. Proc. Nat. Acad. Sci., U.S.A., 72: 3190-3193, 1975. 17. Okubo, S., and Romig, W. R. Impaired Transformability Of Bacillus subtilis Mutant Sensitive To Mitomycin C And Ultraviolet Radiation. J. Mol. Biol., 15: 440-454, 1966. 18. Olsen, W. L., Heidrich, H. G., Hannig, K., and Hofshneider, P. H. Deoxyribonucleic Acid-Envelope Complexes Isolated From Escherichia coli By Free-Flow Electrophoresis: Biochemical And Electron Microscope Characterization. J. Bacteriol., 118: 646-653, 1974. 19. Rannug, U., Johansson, A., Ramel, C., and Wachtmeister, C. A. The Mutagenicity Of Vinyl Chloride After Metabolic Activation. "ffilO, 3: 194-197, 1974. 20. Richards, G. Modifications Of The Diphenylamine Reaction Giving Increased Sensitivity And Simplicity In The Estimation Of DNA. Anal. Biochem., 57: 369-376, 1974. 21. Strauss, B., and Okubo, S. Protein Synthesis And The Induction Of Mutations In Escherichia coli By Alkylating Agents. J. Bacteriol., 79: 464-473, 1955: 22. Strauss, B., Reiter, H., and Searashi, T. Recovery From Ultra violet And Alkylating Agent-Induced Damage In Bacillus subtilis. Rad. Res. Supp., 6: 201-211, 1966. 23. Streips, U. N., and Young, F. E. Transformation In Bacillus subtills Using Excreted DNA. Molec. Gen. Genetics, 133: 4*7-55, 1974. CMA 003730 STUDIES ON THE MUTAGENICITY OF VINYL CHLORIDE METABOLITES 24. Sueoka, N., and Quinn, Y/. Membrane Attachment Of The Chromosome Replication Origin In Bacillus subtilis. Cold Spring Harbor Symp. Quant. Biol., 33: 695-V05, 1968. 25. Van Duuren, B. L. On The Possible Mechanism Of Carcinogenic Action Of Vinyl Chloride. Ann. N. Y. Acad. Sci., 246: 258-267, 1975. 26. Viola, P. L., Bigotti, A., and Caputo, A. Oncogenic Response Of Rat Skin, Lungs And Bones To Vinyl Chloride. Cancer Res., 31: 516522, 1971. 27. Vogel, H. J., and Bonner, D. M. Acetylomithinase Of Escherichia coll: Partial Purification And Some Properties. J. Biol. Chem., 2137 97-106, 1956. 28. Yamagudln, K., and Yoshlkawa, H. Association Of The Replication Terminus Of The Bacillus subtilis Chromosome To The Cell Membrane. J. Bacteriol., 124: 1030-1033, 1975. 29. Young, F. E., and Wilson, G. A. Bacillus subtilis. In: Handbook Of Genetics. Ed.: Robert C. King, Plenum Press, New York, 1: 69114, 1974. CMA 003731 --io n Cri: --* 7 25. Van Duuren, B. L. On The Possible Mechanism Of Carcinogenic Action Of Vinyl Chloride. Ann. N. Y. Acad. Sci., 226: 258-267, 1975. * -wa. f *. ^ * * vuh>4vU j *1 Rat Shin, Lungs And Bones To Vinyl Chloride 522, 1971. 'w* !!3*sa'" n ^ ~ =. c & - Cancer Res 31: 516- 27. Vogel, H. J., and Bonner, D. M. Acetylomithinase Of Escherichia coli: Partial Purification And Some Properties. J. Biol". "EJnen., 2lf? 97-106, 1956. 28. Yaraagudin, K., and Yoshikawa, H. Association Of The Replication Terninus Of The 3acillus subtilis Chromosome To The Cell Membrane. J. Bacteriol., 124: 1030-1033, 1975. 29. Young, F. E., and Wilson, G. A. Bacillus subtilis. In: Handbook Of Genetics. Ed.: Robert C. King, Plenum Press, New York, 1: 69114, 1974. caza 03 732 Laumbach, A. D., Lee, S., Wong, J., and Streips, U. N. Table I Bacterial Strains Bacillus subtilis Genotype Origin and Comments RUB 783 BR 151 BUL 709 BUL 714 Hcr-9 (JB01-200) MC-1 FB-13 168WT purB6, leu-8, hisAl, metBIO trpC2, lys-3, metBIO ura-1, hisAl, leu-8, metBlO cysA, hisAl, leu-8, metBlQ trpC2 trpC2, recB2 trpC2 prototroph U. Streips B. Reilly This laboratory This laboratory S. Okubo and W. Romig, her" S. Okubo and W. Romig, recC. Hadden, uvr" A. Laumbach and I. Felkner Salmonella typhimurium Mutations in Strains His" LPS DNA Repair R Factor Mutation Detected TA1535 TA100 TA1537 TA1538 TA98 hisB46 rfa hisB46 rfa hisC3076 rfa hisD3052 rfa hisD3052 rfa uvrB uvrB uvrB uvrB uvrB base-pair substitution pKMIOl base-pair substitution frameshift frameshift pKMIOl frameshift CMA 003733 STUDIES ON THE MUTAGENICITY OF VINYL CHLORIDE METABOLITES Table II Mutagenic Activity Assayed by Bacterial Test Systems Compounds Indirect Screen Direct Testa B. subtilis S. typhinrurium "Repair-Assay" Strain TA100 Acetaldehyde Chloroacetic Acid Chloroethanol Vinylidene Chloride Vinyl Chloride Chlorooxirane Chloroacetaldehyde (monomer) Chloroacetaldehyde (monomer-dimer hydrates) Chloroacetaldehyde (dimer hydrate) Chloroacetaldehyde (trimer) Epichlorohydrin NRb NR NR NR NR +c +++ ++ + NR NR NR NR NR NR +++ ++ + + + Experiments performed in absence of liver homogenatemediated activation. Er no reaction detected c + Reactive d++ Moderately reactive e+++ Very reactive CMA 003734 STUDIES ON THE I.TITA' CHLORIDE :s e II 55ayec ;y oacrerial Te;si oystems Compounds B. subtilis "Repair-Assay'' Direct Test3, S. typhimurium Strain iAIGO Acetaldehyde Chlcroacetio Acid Chloroethanol Vinylidene Chloride Vinyl Chloride Chlorooxirane Chloroacetaldehyde (monomer) Chloroacetaldehyde (monomer-dimer hydrates) Chloroacetaldehyde (dimer hydrate) Chloroacetaldehyde (trimer) Epichlorohydrin NRb ?!?. NR NR +c +++6 ++ + + NR NR tt.L NR NR ++d +++ ++ + + "Experiments performed in absence of liver homogenatemediated activation. ^NR no reaction detected c Reactive ^++ Moderately reactive e+-t-. very reactive CMA 003735 Laumbach, A. D., Lee, S ., Wong, J . , and S treips, 003736 Table III "Repair-Assay" with Bacillus subtilis Strains o Molar Growth Inhibition in Millimeters5 Compounds Concentration 168WT MC-1 Hcr-9 FB-13 (hcr+, rec+) (hcr+, rec~) (her-, rec+) (uvr+, rec+) z! Chloroacetaldehyde (monomer) 0.10 2.0 28.0 4.0 3.0 Chloroacetaldehyde (monomer-dimer hydrate) 0.115 NIb 23.0 NI NI Chloroacetaldehyde 0.097 2.0 10.0 2.0 2.0 (dimer hydrate) Chloroacetaldehyde 0.096 7.0 15.0 6.0 7.0 (trimer) Chlorooxirane Epichlorohydrin 0.26 HI 10.0 NI NI 0.997 NI NI NI NI Epichlorohydrin (plus liver homogenate )c 0.997 HI Average inhibition calculated from multiple experiments. ^No inhibition detected. <=9,000 x g supernatant (S-9) + NADPH generating system. 3.0 NI NI Table IV Quantitative Mutagenicity Assay by Salmonella TAIOO Reversion STUDIES ON THE MUTAGENICITY OF VINYL CHLORIDE METABOLITES CMA 0 0 3 7 3 7 Compound Concentration in Soft Agar Layer mM/Platea Chloroacetaldehyde (monomer) Chloroacetaldehyde (monomer-dimer hydrate) Chloroacetaldehyde (dimer hydrate) Chloroacetaldehyde (trimer) Epichlorohydrin 0.0004 0.054 0.490 0.240 4.746 highest effective non-toxic concentration for reversion. ^Spontaneous background revertants subtracted. Average Number Revert ants/PIate*5 265 977 311 159 2856 Table IV Quantitative Mutagenicity Assay by Salmonella TA100 Reversion CMA 0 0 3 7 3 8 Concentration in to<1 Compound Soft Agar Layer mM/Platea Average Number Revertants/Plate^ fl ) l* Chioroacetaldehyde 0.0004 265 ( monomer} waMCO <* >I Chloroacetaldehyde (monomer-dimer hydrate) 0.054 977 r.3 h7 Chioroac et aldehyde (dimer hydrate) 0.490 311 ti Chloroacetaldehyde (trimer) 0.240 159 Epichlorohydrin 4.746 2856 F -f 1f "1> highest effective non-toxic concentration for reversion. I *- T) ^Spontaneous background revertants subtracted. [;) IJ O rII T0-43l i<0 Laumbach, A. D ., Lee, S., Wong, J ., and Streips 003739 Table V Effect of Chloroacetaldehyde and Epichlorohydrin of Transforming DMA In vitro Recipient Strains Relative Transformation Efficiency8 metBlO leu-8 cysA hisAl ura-1 trpC2 lys-8 purB6 u z Epichlorohydrin treated !=> DNA BUL 714 .92 .97 .97 1.16 RUB 783 .91 .60 .98 .77 BUL 709 .99 .95 1.02 .85 BR 151 1.48 1.07 .62 Chloroacetaldehyde treated DNA BUL 714 1.43 .92 .55 .91 RUB 783 .93 1.45 .75 .89 BUL 709 .86 1.33 .90 .75 BR 151 NDb .77 Relative transformation efficiency calculated: number of transformants with treated DNA b^ot determined number of transformants with untreated DNA Conditiasns for competence and transformation as described in Materials and Methods. Table VI EFFECT OF CHLOROACETALDEHYDE AND EPICHLOROHYDRIN ON TRANSFORMING DNA IN VIVO STUDIES ON THE MUTAGENICITY OF VINYL CHLORIDE METABOLITES 003740 Recipient Strains*1 Relative Transformation Efficiency metBlO leu-8 cysA hisAl ura-1 trpC2 lys-3 purBl6 Chloroacetaldehyde in vivo treated DNA BUL 714 RUB 783 BUL 709 BR 151 .54 .09 .08 .36 .33 .10 .35 .53 .07 .32 .34 .37 .13 .11 .10 Epichlorohydrin in vivo treated DNA BUL 714 RUB 783 BUL 709 BR 151 .55 .17 .25 .46 .52 .15 .38 .59 .15 .50 .36 .37 .11 .19 NDC Relative transformation efficiency calculated: number transformants with treated DNA number transformants with untreated DNA ^Conditions for competence and transformation as described in Materials and Methods. cNot determined. Table VI EFFECT OF CHLOROACETALDEHYDE AND EPICHLOROHYDRIN ON TRANSFORMING DNA IN VIVO STUDIES ON THE MUTAGENICITY G? VINYL CHLORIDE 003741 Recipient Strains*3 Relative Transformation Efficiency8 metBlO leu-8 cysA hisAl ura-1 trpC2 lys-3 purBl6 !F H1I C1 Chloroacetaldehyde CJ in vivo treated DMA t '< Id u BUL 714 .54 .09 .08 .36 RUB 783 .33 .10 .35 .10 BUL 709 .53 .07 .32 .34 BR 151 .37 .13 .11 Epichlorohydrin in vivo treated DNA BUL 714 RUB 783 BUL 709 BR 151 .55 .17 .25 .46 .52 .15 .38 .59 .15 .50 .36 .37 .11 .19 NDC Relative transformation efficiency calculated: number transformants with treated DNA number transformants with untreated DNA ^Conditions for competence and transformation as described in Materials and Methods. cNot determined. VINYL CHLORIDE MUTAGENICITY AND CARCINOGENICITY VIA THE METABOLITES CHLOROOXIRANE AND CHLOROACETALDEHYDE MONOMER HYDRATE. Uim D. Elmore and John L. Wong* Department of Chemistry, University of Louisville Andrew D. Laumbach and Uldis N. Streips* Department of Microbiology, University of Louisville Louisville, Kentucky, USA 40208 . SUMMARY Mutagenicity tester strains of Bacillus and Salmonella were used to assay vinyl chloride in nutrient broth at a practical concentration level. Also screened without exogenous activation were seven potential metabolites of vinyl chloride in their pure forms as well as the related epichlorohydrin. Chlorooxirane, chloroacetaldehyde, chlcroacetaldehyde monomer hydrate, chloroacetaldehyde dimer hydrate, chloroacetaldehyde trimer, and epichlorohydrin produced significant mutagenic activity in Salmonella typhimurium strains sensitive to base-pair mutation. A recombination repair deficient strain of Bacillus subtil is was inhibited in growth by these compounds, whereas excision repair deficient and wild type strains of Bacillus subtil is were relatively unaffected. On the basis of these assays a working hypothesis for the vinyl chloride carcinogenesis mechanism is proposed which involves chlorooxirane and chloroacetaldehyde monomer hydrate as the ultimate carcino genic metabolites of vinyl chloride. CMA 003742 L INTRODUCTION The carcinogenic potential of vinyl chloride monomer 1 was initially 'll established by Viola et al_. [1J and Mai torn' et al. [2] with inhalation experi ments using laboratory animals. Detection of angiosarcoma in polyvinyl chloride workers suggested a causal relationship between industrial exposure to vinyl chloride and the development of pathological conditions in humans. This contention was supported by epidemiological data which revealed an association between exposure to 1 and the onset of hepatic abnormalities including angio- -v sarcoma [3,4], A report by Hefner et al_. [ 5] on the metabolic fate of 1 in rats indicated that 67% of the vinyl chloride inhaled by the rats was metabolized and excreted in the urine. The metabolic products identified were N-acetyl-S(2-hydroxyethylJcysteine and thiodiglycolic acid [5,6] which are sulfhydryl conjugates of chloroethanol f2\f and chloroacetic acid i3\, respectively. Chloro-'* " ,, ,* oxirane 4^ and chloroacetaldehyde O5i were speculated to be the carcinogenic forms. We report herein the mutagenicity and carcinogenic potential of these i compounds in their pure forms. Several reports [7,8,9] have appeared recently using the Salmonella tester strains to test vinyl chloride and several of its supposed metabolic derivatives. Vinyl chloride and chloroethanol were found to be mutagenic after activation by liver homogenates [9]. Direct mutagenicity of vinyl chloride was also reported by McCann et aK [7] and Bartsch eal_. [10] at 20% v/v in air (200,000 ppm). Since the solubility of vinyl chloride in water at 25C and 1 atm has been determined to be 7.79 x 10"^ mole fraction [11] or 2,900 ppm, we have conducted further testing at this concentration level to secure a practical CMA 003743 0 dose-response comparison with its proximate metabolites. Regarding the latter, the exact chemical forms of the proximate metabolites previously tested are often questionable. Chloroacetaldehyde, like formaldehyde [12], dichloroacetaldehyde Cl31 and chloral [14], can exist in combinations of four forms depending on the history of sample preparations: the monomer 5, the monomer hydrate r6\j, the dimer hydrate i7\j, and the trimer ^8. McCann -e---t----a--l- * [ 7] used vacuum distilled chloroacetaldehyde without a follow-up analysis of its content. This distillate may have consisted of chloroacetaldehyde monomer 5 and its cyclic trimer 8 if water was totally absent, or it may have been a mixture of chloroacetaldehyde hydrates 6 and 7 in an aqueous medium. Bartsch t al_. tlOl tested a commercial aqueous chloroacetaldehyde solution which, according to our analysis reported herein, had an acidic pH and approximately equal concentrations of the two hydrates 6 and 7. This solution was also contaminated by ethanol to the extent of 10%. We have therefore conducted individual assays of pure compounds, or assays of a known mixture of the specific forms of chloroacetaldehyde. Furthermore, the mutagenicity observed for chlorooxirane 4 [9] may be attributed to a chloroacetaldehyde hydrate rather than the chlorooxirane integrity. Under the 37C aqueous testing conditions reported, chlorooxirane decomposed with a half life of 1.6 min [10] to chloroacetaldehyde. For this reason we have also screened epichlorohydrin 9 which is a stable chloro-epoxide homolog of 4 as a comparative assay to interpret the observations of the activity of chlorooxirane. This investigation used the above-mentioned compounds 1^ - 9 in mutagen assays without exogenous enzyme activation. A preliminary screen was performed using DNA repair-deficient mutants of Bacillus subtilis. This was followed by quantitative testing of the compounds for mutagenicity CMA 003744 I JC J, t JWI t >< ' 1 * w . I------.J . tested ar -V. W i.M IJ J I w. " *- j -- -i *- r cuasr zr?.:'. z. Chi orze.zi:: '. dichlcroacetaldahyda [13], and chloral [14], can exist in combinations of four forms depending on the history of sample preparations: the monomer 5, the monomer hydrate 6, the dimer hydrate 7, and the trimer 8, McCann et al. f 7] used vacuum distilled chloroacetaldehyde without a follow-up analysis of its content. This distillate may have consisted of chloroacetaldehyde monomer 5 s4 i wj.l It msrLi . 3 if water was totally absent, or it may have been a mixture of chloroacetaldehyde hydrates 6 and 7 in an aqueous medium. Bartsch r\j et aK [10] tested a commercial aqueous chloroacetaldehyde solution which, according to our analysis reported herein, had an acidic pH and approximately equal concentrations of the two hydrates 6 and 7. This solution was also contaminated by ethanol to the extent of 10%. We have therefore conducted individual assays of pure compounds, or assays of a known mixture of the specific forms of chloroacetaldehyde. Furthermore, the mutagenicity observed for chlorooxirane 4 [9] may be attributed to a chloroacetaldehyde hydrate rather than the chlorooxirane integrity. Under the 37C aqueous testing conditions reported, chlorooxirane decomposed with a half life of 1.6 min [10] to chloroacetaldehyde. For this reason we have also screened epichlorohydrin 9 which is a stable chloro-epoxide homolog of 4 as a comparative assay to interpret the observations of the activity of chlorooxirane. This investigation used the above-mentioned compounds 1^ - 9 in mutagen assays without exogenous enzyme activation. A preliminary screen was performed using DNA repair-deficient mutants of Bacillus subtil is. This was followed by quantitative testing of the compounds for mutagenicity CMA 003745 4 with Salmonella typhimurium LT-2 strains [15]. Thus, a combination of these two screening procedures have led to information on the mutagenicity and carcinogenic potential of these compounds as well as their chemical mode of action. MATERIALS AND METHODS The bacterial strains used in the bioassays are shown in Table I. The Salmonella tester strains were designed to detect chemical carcinogens as mutagens [15]. The recombination and DNA repair-deficient Bacillus subtilis strains were used in the repair assays as an indirect test for mutagenicity [16,17]. Nutrient broth (Difco) and nutrient broth plus 0.5% NaCl was used for growth of stock cultures of Bacillus and Salmonella strains respectively. Nutrient agar (Difco) served as a solid medium for the growth of Bacillus strains in "repair-assays". The pour plates used with Salmonella strains consisted of molten (45C) soft agar which contained 0.6% agar, 0.6% NaCl, 0.5mM biotin, and 0.5 mM histidine. The minimal agar plate was composed of Vogel-Bonner E medium [18], 1.5% agar, and 2% glucose. Vinyl chloride gas was obtained from Matheson Scientific; aqueous chloroacetaldehyde (45% by wt.) from ICN Pharmaceuticals; epichlorohydrin from Matheson Coleman and Bell; and other chemicals from Aldrich Chemical Co. Mutagenicity Assays Salmonella - Vinyl chloride 1 was tested by the method of Ames [7], A mixture of 0.1 ml of 1 in broth and 2 ml of top agar was added to 0.1 ml of cell culture. The solutions were then mixed and poured immediately onto the surface of a minimal medium plate. After incubation for 48 hrs at 2J?C, colonies were counted and recorded. For compounds 2 - 9,sample solutions of known concentrations CMA 003746 were prepared in dimethyl sulfoxide (DMSO). A 0.1 ml aliquot of the sample solution was admixed with 0.9 ml of the tester strain culture. Then, a 0.1 ml sample of this mixture was added to 2 ml of molten soft agar and applied to the surface of a minimal agar plate. Control plates for detection of the spontaneous reversion rates were prepared for each tester strain by omitting only the compounds tested. Pour plates were incubated for 48 hr at 37"C before revertant colonies to prototrophy were counted. For positive mutagenesis control, plates containing the mutagen 4-nitroquinoline-N-oxide ./nre used. Bacillus - The "repair-assay" procedure was a modification of the "recassay" procedure of Kada et al_. [19]. They were grown overnight in nutrient broth then diluted 10 fold in phosphate buffer (pH 7.0). Strains were streaked with pipettes onto nutrient agar plates. Filter paper discs (6 mm) saturated with test solution were placed upon the bacteria streaks. Following incubation for 24 hr at 37C, growing bacteria were visible except in the inhibition zone The lethality and mutagenic potential of compounds were assessed by comparison of inhibition zones between the 168 wild type strain and the DNA repair-deficient strain and the DNA repair-deficient strains. The control used was 4-nitroquinolineN-oxide. Survival assays for 6 and 7 (cf.. Fig. 3.) were made in MY-1 broth [17] solutions. Cultures were grown in tryptose blood agar base for 16 hrs, inoculated into MY-1 broth, and viable cell counts were done on TBAB agar plates. Compound Synthesis and Purification Chlorooxirane 4 prepared by the method of Walling and Frederick [20J was in % higher purity (95% pure) than that by molecular chlorination [21] (50% pure). Thus, t-butyl hypochlorite and ethylene oxide at -10C with 200 watt tungsten lamp irradiation yielded chlorooxirane 4i glpc (gas liquid phase chromatography) 003747 u tR * 1.2 min at 508C, infrared absorption ( vcm"*) 900, 1250, 1320, and 1710 as described previously [21], and PMR as shown in Table II. Derivatization of 4 with an excess of acidic 2,4-dinitrophenylhydrazine solution gave glyoxal-bis-dinitrophenylhydrazone, mp 317-318 C (317C reported [21]). Chloroacetaldehyde monomer 5 was obtained in the purest form by cracking the chloroactaldehyde trimer 8 at 95C and distilling it into dry 0MS0; % glpc = 2.25 min at 100C and PMR as shown in Table II. Chloroacetaldehyde dimer hydrate 7 - A solution containinq 49.5 ml of 38% hydrochloric acid, 30 ml of the 45% aqueous chloroacetaldehyde solution, and 55.5 ml of water was distilled over a 9 inch Vigreaux fractionating column. The fraction (1/5 of the initial volume) collected from 87-100C was redistilled. This second distillate at 83-92C crystallized after 3 days at -15c. Upon sublimation of 60C and 1 atm, white crystals of 7 were obtained; mp 55C and PMR as shown in Table II. --Chloroacetaldehyde trimer 'b8 - Concentrated sulphuric acid (7.5 ml) was added to the 45% aqueous chloroacetaldehyde solution (5 ml) with vigorous stirring and external cooling (-5*C). The crystalline precipitate was filtered after standing overnight at -15C, washed with 5 ml of cold 20% aqueous methanol, and recrystallized 5 times from cold methanol; mp 87-88C, corresponding to that reported by Natterer [13], and PMR as shown in Table II. Quantitation of vinyl chloride in nutrient broth - The concentration of vinyl chloride 1 in the broth solution was determined by an extraction method in conjunction with glpc. This method involved (1) establishing the CMA 003748 7 linearity of the response of 1 on glpc, (2) constructing a standard curve % of vinyl chloride weight vs_. peak area, and (3) extracting the broth with methylene chloride followed by glpc determination. (1) Linearity of response - A standard solution of 1 was prepared by condensing it (bp -13.4C) at -78C onto a known weight of methylene chloride in a 1 ml volumetric flask fitted with a serum cap. The condensed vinyl chloride was determined by weighing. A typical solution thus prepared was 0.2877 M and was subjected to glpc analysis by varying injection sizes from 1-9 ul. The correlation of vinyl chloride weight and peak area was made by a least square computer routine: slope = 0.877, with an index of correlation of 0.972 (ideal 1.00) up to 6 yl (0.1078 mg) of injection. (2) A standard linear curve using the above technique was established for 0.01-0.10 mg of vinyl chloride. (3) Extraction of broth Vinyl chloride was allowed to bubble through the nutrient broth for 30 min at 25C. A 1.0 ml broth sample was then extracted with 4 x 2 ml of methylene chloride, the extracts were combined and then made up to 10 ml in a volumetric flask with methylene chloride. Glpc analysis of this solution and application of the standard curve showed that there was 7.0 mg of 1 in the 10 ml solution, or the concentration of 1 in the broth was 0.0107 M. Repeated determination showed it to be 0.0105 M. High Pressure Liquid Chromatographic (HPLC) Analysis of the Commercial 45% Aqueous Chloroacetaldehyde - Reverse phase HPLC on the commercial 45% chioroacetaldehyde solution (7 M, pH 2.6) resolved it into two components: tRl =6.3 min and tR2 = 8.8 min in a ratio of 40:60. The ratios of the two peaks on the chromatogram changed as the solution pH was varied by addition of 1 N NaOH at room temperature: 43:57 (pH 3.7), 44:56 (pH 5.0), 48:53 (pH 7.6), CMA- 003T49 ! -3 ^ J -r'-' ni kj W -J i i. ^ solution was admixed witn 0.9 ml of :ne tester strain culture. Then, a 'N -J *, O - H * ** - -A. of the spontaneous reversion rates were prepared for each tester strain by omitting only the compounds tested. Pour plates were incubated for 48 hr at 37aC before revertant colonies to prototrophy were counted. For positive mutagenesis control, plates containing the mutagen 4-nitroquinoline-N-oxide .''re used. Bacillus - The "repair-assay" procedure was a modification of the "recassay" procedure of Kada et al_. [19]. They were grown overnight in nutrient broth then diluted 10 fold in phosphate buffer (pH 7.0). Strains were streaked with pipettes onto nutrient agar plates. Filter paper discs (6 mm) saturated with test solution were placed upon the bacteria streaks. Following incubation for 24 hr at 37C, growing bacteria were visible except in the inhibition zone. The lethality and mutagenic potential of compounds were assessed by comparison of inhibition zones between the 168 wild type strain and the DNA repair-deficient strain and the DNA repair-deficient strains. The control used was 4-nitroquinoline N-oxide. Survival assays for 6 and 7 (cf.. Fig. 3.) were made in MY-1 broth [17] solutions. Cultures were grown in tryptose blood agar base for 16 hrs, inoculated into MY-1 broth, and viable cell counts were done on T8AB agar plates. Compound Synthesis and Purification Chlorooxirane 4 prepared by the method of Walling and Frederick [20] was in 'V higher purity (95% pure) than that by molecular chlorination [21] (50% pure). Thus, t-butyl hypochlorite and ethylene oxide at -10C with 200 watt tungsten lamp irradiation yielded chlorooxirane 4; glpc (gas liquid phase chromatography) CMA 003750 (CANCER RESEARCH 41, 419-424, Febfuwy 198t] 0000-5472/81/0041-0000*02.00 Tissue and Urinary Glycosaminoglycan Patt rns Associated with a Fast, an Intermediate, and a Slow-growing Morris Hepatoma*1 8 Charles E. Kupchella,1 E. Elaine Drake, Jeffrey Kennedy, Kevin L. Curran, Raya Warick, and H. P. Morris Cancer Center, university ot Louisville, Louisville. KentucHy 40201 fC K.E E D.J K,K L C R w l. nnd the Department of Biochemistry, Cancer Research Unit College at Medicine. Howard University. Washington, D. C [H P M. I ABSTRACT MATERIALS AND METHODS The purpose of this investigation was to evaluate the glycosaminoglycans (GAG's) in different behavioral-histological types of i.m.-transplanted hepatomas and in the liver and urine of animals bearing these tumors. Groups of 10 Buffalo rats carrying fast-growing (7777), intermediate (5123tc), and slowgrowing (9618A) Morris hepatomas were studied as the tumors reached 3 cm. Urinary and tissue GAG's were isolated by proteolysis, separated as cetylpyridinium complexes, and measured as uronic acid. The GAG's were further purified using anion-exchange chromatography and characterized with mucopolysaccharidases. Tissue GAG's were also evaluated histochemically using Alcian blue staining and mucopolysac charidases. Tissue from fast-growing, intermediate, and slowgrowing tumors exhibited greater GAG levels than did normal liver in the hyaluronic acid (0.4 m NaCI-soiuble) fraction and in the chondroitin sulfate-heparan sulfate (1.2 m NaCI-soluble) fraction. The livers of tumor-bearing animals exhibited GAG levels similar to those of normal liver. Increased urinary GAG excretion was appreciated in animals bearing Tumors 5123tc and 9618A but not in those bearing Tumor 7777. INTRODUCTION An increasing number of reports cite the presence of com paratively high levels of GAG's3 in both animal tumors (4, 7, 1 a, 22) and human tumors (5, 12, 16). There also have been reports citing qualitatively and quantitatively abnormal urinary GAG excretion in association, with malignant tumors (6,10.11, 23-25). Although there have been attempts to establish any functional relationship between tumor GAG's and tumor cell properties (23, 25-27, 30, 31), the significance of elevated GAG's in malignant tumors remains obscure. In view of the possibility that GAG's play an important role in the expression of one or more malignant cell properties, our purpose here was fb evaluate the GAG patterns associated with transplantable hepatomas exhibiting different growth rates and metastatic properties. Because it has been suggested that tumor GAG may be contributed by normal host tissue in re sponse to the presence of hepatic tumor (7), a secondary purpose was to evaluate the influence, if any, of "remote" hepatomas on the GAG's of the host liver; a third purpose was to evaluate the urinary GAG patterns in hepatoma-bearing animals. ' Supported by American Cancer Society Grant IN-111B, by a grant trom me Manufacturing Cbamiata Aaaociation. and in part by USPHS Qranta CA 10729 and CA 24S201. 1 Present address: Biotogtcal Sciences, Murray State University, Murray, Ky. 42071. To whom requests for reprints should be addressed. 1 The abbreviation used M: GAG, glycosammogtycsn Received Aprs 11. i960; accepted October 24. i960. Materials. Hyaluronic acid (umbilical cord) was purchased from Nutritional Biochemical Corp. (Cleveland. Ohio); chon droitin sulfate (whale and shark cartilage) and sodium heparin were purchased from Sigma Chemical Co. (St. Louts, Mo.). Authentic samples of heparin, chondroitin 4-sulfate, heparan sulfate, and hyaluronate were also kindly supplied by Dr. M. B. Matthews, University of Chicago. Bovine testicular hyaluronidase was purchased from ICN Pharmaceuticals (Cleveland, Ohio), Streptomyces hyaluronidase was obtained from Calbiochem (La Jolla, Calif.), and Proteus vulgaris chondroitinase ABC was purchased from Sigma. Experimental Design. Forty male Buffalo rats were shipped from Lab Supply Company (Indianapolis, Irtd.) to Washington, 0. C., where 10 were inoculated bilaterally (thigh) with Tumor 7777. 10 were inoculated with Tumor 5123tc. and 10 were inoculated with Tumor 9618A. These were shipped to Louisville with 10 controls. Throughout the study, animals were provided free access to food and water even when placed in metabolic cages for urine collections. Urine collections were made twice each week from animals bearing Tumor 7777 and once each week from animals bearing Tumors 5123tc and 9618A. Collections were made alterna tively on one-half of the animals in each group, with 5 control animal collections made each time a collection was made from tumor-bearing animals. Tumors. Line 5123tc is a tissue culture variant of a moder ately differentiated trabecular hepatocellular carcinoma in duced with dietary administration of A/-2-fluorenylphthalamic acid. Tumors were received and studied here in the 165th generation. Tumor line 7777 is a poorly differentiated hepatocellular carcinoma induced by dietary administration of N-2fluorenylphthaJamic acid. This line was studied in the 159th transplant generation. Tumor line 9618A is a well-differentiated hepatocellular car cinoma induced by 2-(4`-methy!)benzoylaminoffuorene. This tumor was studied in the 13th generation. Characteristics of these 3 tumors observed in our laboratory and selected characteristics reported by Hruban at al. (13,14) are summarized in Table 1 (see also Fig. 1). Extraction and Purification of GAG's from Liver and Tumor Tissue. Dry defatted tissue was subjected to proteolysis, tri chloroacetic acid precipitation, and dialysis, and the GAG's were separated as cetylpyridinium chloride complexes into 0.03 m NaCI-soluble, 0.4 m NaCI-soiuble. i .2 m NaCI-soluble. and 2.1m NaCI-soluble fractions as described by Schiller etal. (28) and measured as uronic acid (see Schiller et at.) by the carbazole method of Bitter and Muir (1). FEBRUARY 1981 CMA 003751 419 7 linearity of tine response cf 1 on glpc, (2) constructing a standard v*k / I l ij1 I i`i ! i < w v j w Cl fs Ct ' - U ) '-h -\ N, J / t.ethylene chloride followed oy glpc determir.ati; v `; Lmcaricv of response - A standard solution of 1 was prepared by condensinc it (bo -13.4C) at -78C onto a known weight of methylene chloride in a 1 ml volumetric flask fitted with a serum cap. The condensed vinyl chloride was determined by weighing. A typical solution thus prepared was 0.2377 M and was subjected to glpc analysis by varying injection sizes from 1-9 ul. The correlation of vinyl chloride weight and peak area was made by a least square computer routine: slope = 0.877, with an index of correlation of 0.972 (ideal 1.00) up to 6 U1 (0.1078 mg) of injection. (2) A standard linear curve using the above technique was established for 0.01-0.10 mg of vinyl chloride. (3) Extraction of broth - Vinyl chloride was allowed to bubble through the nutrient broth for 30 min at 25*C. A 1.0 ml broth sample was then extracted with 4 x 2 ml of methylene chloride, the extracts were combined and then made up to 10 ml in a volumetric flask with methylene chloride. Glpc analysis of this solution and application of the standard curve showed that there was 7.0 mg of 1 in the 10 ml solution, or the concentration of 1 in the broth was 0.0107 M. Repeated determination showed it to be 0.0105 M. High Pressure Liquid Chromatographic (HPLC) Analysis of the Commercial 45% Aqueous Chloroacetaldehyde - Reverse phase HPLC on the commercial 45% chloroacetaldehyde solution (7 M, pH 2.6) resolved it into two components: tRl = 6.3 min and tR2 = 8.8 min in a ratio of 40:60. The ratios of the two peaks on the chromatogram changed as the solution pH was varied by addition of 1 N NaOH at room temperature: 43:57 (pH 3.7), 44:56 (pH 5.0), 48:53 (pH 7.6), CMA 003752 and 56:44 (pH 8.2). The first eluted component at t^l = 6.3 min increased while the second at t^2 8.8 min decreased at raised pH's. This indicated a retrograde aldol condensation type hydrolysis. The first peak was assigned to the monomer hydrate 6 and the second to the dimer hydrate 7. When the 45% solution was diluted to 0.44 M at pH 3.6, refluxed for 1 hr, cooled, and chromatographed, the ratio of t^l:t^2 became 33:67. This suggests an acidcatalyzed condensation of 6 to form the dimer hydrate 7. When these two f\t components were analyzed by glpc, both emerged at the same retention time (t_ = 2.8 min at 60C) as that of the monomer 5. Dehydration of 6 and 7 K^ must have occurred under the glpc conditions thereby reverting them to the monomer form. As shown in Table II, the PMR spectra of 6 and 7 are resolved from one another. Thus, the PMR spectrum of the 45% commercial solution also revealed the presence of both hydrates 6 and 7 with approximately the same integrals. Instrumentation High Pressure Liquid Chromatography (HPLC) - A Waters Associates model 600 pump combination (dual) with a model 440 differential 254 nm ultraviolet detector were used for the analysis of aqueous chloroacetaldehyde solutions. A Porasil Bondapak yC-jg 4 mm x 30 cm column was used with a 80:20 v/v 0.1 N NH4H2P04-Me0H isocratic eluant (pH 4.9) at 1 ml/min. These two components were also resolved on a Reeve Angel Parti si1 10 0DS 4.6 mm x 25 cm column using the same eluant. cm 003753 Gas Liquid Phase Chromatography (GLPC) - Analysis of vinyl chloride 1, chlorooxirane 4, and chloroacetaldehyde 5 were performed on a Carle model 9500 flame ionization gas chromatograph. A 10% SE-30 on Anakron 60/70 packed column was used at 30C for vinyl chloride determination. A 20% Carbowax on chromosorb W column at 50"C and 40 ml/min of Helium was used for analysis of 4 and 5 unless specified otherwise. Both were 0,125 inch x 5 feet stainless steel columns. Proton Magnetic Resonance (PMR) - Spectra were obtained using a Varian A-6Q A and a Perkin Elmer R-12 spectrometer. Solutions of 020 and DMSO-dg were used with 3-(trimethylsilyl)propanesulfonic acid sodium salt as internal reference. Tetramethylsilane was used as a reference in CDC13 and CCl^, Probe temperature was 38 t. RESULTS AND DISCUSSION Mutagenicity assays with Bacillus and Salmonella for compounds 1-9 are 'V 'X, summarized in Table III. They are grouped into three categories in subsequent discussion. Further testing data are included in Tables IV-VI and Figures 2-4. Nonmutagenicity of vinyl chloride, chloroethanol, and chloroacetic acid Only high concentrations of vinyl chloride (20% v/v in air) have produced mutagenic action in previous assays with the Salmonella tester strains [9,10]. We have found that tests with both the Salmonella and the Bacillus cultures were negative within the practical solubility range of vinyl chloride in the nutrient broth under ambient conditions. Figure l shows the stability of a CMA 003754 Gas Liquid Phase Chrc~ / . - ' rnaiysis or v'ryi ce : orooxi rer.e ^ . arc c . _ J L -- i U . mob e t ^joO i 1 a.ue i o n i z ^i o < i ^ o ^ > o i ii ^ii l\ ^ v c *. ^^^ ^ packed column was used at 309C for vinyl chloride determination. A 20% Carbo.'/ex on chromosorb eolcn at 50 C and 40 ml/min of Helium '-'as usee for analysis of 4 and 5 unless specified otherwise. Both were 0,125 inch 'V 'V x 5 feet stainless steel columns. Proton Magnetic P,eso"onc3 fp'-'p) - Spectra were obtained using a Varian A-50 A and a Perkin Elmer R-12 spectrometer. Solutions of 0^0 and CMSO-dg were used with 3-(trimethylsilyl)propanesulfonic acid sodium salt as internal reference. Tetramethylsilane was used as a reference in CCClj and CCl^. Probe temperature was 38 11. RESULTS ANO DISCUSSION Mutagenicity assays with Bacillus and Salmonella for compounds 1-9 are 'V n, summarized in Table III. They are grouped into three categories in subsequent discussion. Further testing data are included in Tables IV-VI and Figures 2-4. Nonmutaqenicity of vinyl chloride, chloroethanol, and chloroacetic acid Only high concentrations of vinyl chloride (20% v/v in air) have produced mutagenic action in previous assays with the Salmonella tester strains [9,10]. We have found that tests with both the Salmonella and the Bacillus cultures were negative within the practical solubility range of vinyl chloride in the nutrient broth under ambient conditions. Figure ] shows the stability of a 37ss presaturated vinyl chloride broth solution at 258C and 1 atm. The initial concentration of 0.022 M of vinyl chloride 1 rapidly decayed by 50% in 15 hr. Thereafter the escape of 1 from the broth slowed considerably. In the next 45 hr there was a further decline of only 18%. Thus the bacteria strains (8. subtil is MC-1 and Salmonella IA 100) were exposed to the stabilized solution of 0,0106 M (723 ppm) of vinyl chloride in the nutrient broth. The negative activity observed was not unexpected since vinyl chloride lacks the electrophilic character common to many mutagens (23], Although the direct mutagenicity of vinyl chloride at 200,000 ppm (20% v/v in air) observed previously [10] may be real, the chronic human exposure problem most likely requires metabolic activation of vinyl chloride to an electrophilic reactive form [8]. Also tested were chloroethanol 2 and chloroacetic acid 3 which probably are metabolic intermediates as shov/n in Scheme I. Both S-(2-hydroxyethyl)cysteine and SCHEME I: Vinyl Chloride Metabolites CH2 = CHC1 C1CH--CH?0H 2 ------------- --2 2 * SH 02C_CH^NH3 ^H2 I ho-ch2-ch2-s cellular ci-ch2-co2h 3--H * s(ch2-co2h)2 thiodiglycolic acid in Scheme I have been identified [6] as the urinary metabolites of vinyl chloride. Neither chloroethanol 2 nor chloroacetic acid 3 exhibited any mutagenic effects at 1 mM concentration in our mutagenicity assays (cf. Fig. 2 ). Although Bartsch et al,P0] found considerable mutagenic CMA 003756 activity with chloroethanol 2 for TA 1530 strain in the absence of microsomal activation, our observations corroborate with those of McCann et al^. [7] who showed that 2 was weakly mutagenic directly even at high concentrations for the sensitive TA 100 and that it showed only trace activity for TA 1535. The enhanced activity of 2 after microsomal activation observed by these groups suggests that chloroacetaldehyde derivatives were being formed from chloroethanol. We have found potent mutagenicity and lethality with the various forms of chloroacetaldehyde as shown in Figures 2-4 and Tables IV-V. Mutagenesis of chloroacetaldehyde, the monomer hydrate, dimer hydrate, and the trimer - When chloroacetaldehyde 5 was distilled into distilled water, a mixture of the monomer hydrate 6 and the dimer hydrate 7 was formed instantly as determined by HPLC and PMR spectroscopy. Analysis of the commercial 4555 aqueous chloroacetaldehyde solution with the same techniques showed a 50:50 mixture of the two hydrates. Upon standing under dry conditions, the monomer 5 cyclized to form trimer 8. The trimer was sparingly soluble in water, but was disproportionated upon heating in water to form the hydrates 6 and 7. Purified samples of chloroacetaldehyde 5, the commercial 4555 chloro acetaldehyde solution containing a 50:50 mixture of the hydrates 6 and 7, the dimer hydrate 7, and the trimer 8, in DMSO solutions were tested for mutagenic potential. The data in Table IV summarizes the results of the repair assays with B. subtil is. Chloroacetaldehyde 5 and the monomer hydrate 6 specifically inhibited the growth of B. subtil is MC-1, a mutant lacking recombination repair of DNA. These unrepaired ONA lesions then led to cell death. However, compounds 5 and 6 did not inhibit the wild type B_. subtil is or those mutant CMA 003757 activity wicn criiOroc^ridnoi 2 tor activation, Cjr observation son 6 do di shewed that 2 was wea>> " -. jo ic directly even at hi oh coooennoiors -'or* the sensitive TA ICO and that it showed only trace activity for TA 1535. The enhanced activity of 2 after microsomal activation observed by these groups suggests that chloroacetaldehyde derivatives were being formed from chloro- ethanol. We have found potent mutagenicity and. lethality with the various forms of chloroacetaldehyde as shown in Figures 2-4 and Tables IY-V. Mutagenesis of chloroacetaldehyde, the monomer hydrate, dimer hydrate, and the trimer - When chloroacetaldehyde '5V was distilled into distilled water. a mixture of the monomer hydrate 6 and the dimer hydrate 7 was formed instantly as determined by HPLC and PM.T spectroscopy. Analysis of the commercial 45* aqueous chloroacetaldehyde solution with the same techniques showed a 50:50 mixture of the two hydrates. Upon standing under dry conditions, the monomer 5 cyclized to form trimer 8, The trimer was sparingly soluble in water, but was disproportionated upon heating in water to form the hydrates 6 and 7. Purified samples of chloroacetaldehyde 5, the commercial 45X chloroacetaldehyde solution containing a 50:50 mixture of the hydrates 6 and 7, the dimer hydrate 7, and the trimer 8, in DMSO solutions were tested for mutagenic potential. The data in Table IV summarizes the results of the repair assays with B. subtilis. Chloroacetaldehyde 5 and the monomer hydrate 6 specifically inhibited the growth of B. subtilis MC-1, a mutant lacking recombination repair of DNA. These unrepaired DNA lesions then led to cell death. However, compounds 5 and 6 did not inhibit the wild type B. subtilis or those mutant CMA 003758 strains (Hcr-9, FB-13) having the capacity for recombination repair. In contrast, purified samples of the dimer hydrate 7 and trimer 8 inhibited all of the mutants as well as the wild type, although MC-1 again demonstrated the most sensitivity. It is unlikely that 7 and 8 caused DNA lesions that 'V are different from those by 5 and 6 because the excision repair mutants and 'V 'V the wild type were inhibited to an equivalent extent. Inhibition of these strains by 7 and 8 may result from a metabolic poisoning or cellular damage similar to that observed in Escherichia coli after exposure to vinyl chloride waste [24]. Viability assays on B. subtil is strains during short term exposure to the chloroacetaldehyde 5 revealed that recombination repair was essential for recovery (Figure 2). The B_. subtil is strains with recombination repair cap abilities displayed repair kinetics as evidenced by the shoulders in curves in Figure 3. The sensitive MC-1 strain lacking recombination repair yet having excision repair was rapidly killed following the exposure. We surmise that these DNA lesions caused by the chloroacetaldehydes were repaired solely by the recombination repair mechanism. The direct mutagenic potential of the samples 4-9 was determined by the mutations expressed in Salmonella strain TA 100. The dose response relation ships of these compounds (cf. Table V), and the dose-response curves (cf. Figure 4 ) were determined with strain TA 100. At high concentrations the dose response curves for all compounds became nonlinear because the toxicity of the chemicals reduced the number of potential revertants on the plates,. The reduction of the cell population by high concentrations was confirmed by viable counts and observations of decreased background lawn on the test plates. CMA 003759 Data from the dose response curves (Figure 4 ) showed that chloroacetaldehyde 5 and the monomer hydrate 6 were more active than 7 and 8. "V 'N. <\, i\. The monomer 5 was the most mutagenic as predicted because its electrophilic carbonyl group remains intact. The restoration of the carbonyl character to the monomer hydrate 6 can be projected in hydrophobic environment of the cell membrane by loss of water, hence its mutagenicity is also explicable. The substantial activity of the dimer hydrate 7 can be attributed to its ready equilibrium with the monomer hydrate 6 in aqueous medium. The relationship of the activity of the cyclic trimer 8 to the chloroacetaldehyde action cannot be deduced on the basis of the curves in Figure 4. Nevertheless, these mutations were the base-substitution type and were expressed in Salmonella strains TA 100 and TA 1535. The mutagenic response in TA 1535 was very weak as compared to TA 100. It should be noted that the latter contains an "R" factor that enhances mutation by an error-prone recombination repair mechanism following DNA damage [7]. Comparison of mutagenic response of chlorooxirane with its methylene homolog epichloroh.ydrin - The prevailing opinion [s] is that chlorooxirane 4 is the primary metabolite of vinyl chloride and is derived from the action of the microsomal mixed function oxidase. This compound, prepared independently from chlorination of ethylene oxide, was found to rearrange readily to chloro acetaldehyde in aqueous or DMSO solution at ambient temperatures. A kinetic study of a 0.15 M solution of 4 in a D20-DMS0-d6 (80:20) mixture at pD 7.1 and 4C by PMR technique showed that the rearrangement followed first order kinetics, k (sec"1) 2.5 x 10"4. This translatesto a half life of 46.2 min CMA 003760 9?ta from the dose resoc~z= carves CFire 4) showed that chloro- aeeta'dehyde 5 i'-.c ^ ;v/:.raze C ware more active than 7 and 3. `u "V 'X, The monomer 5 was the most mutagenic as predicted because its electrophilic 'V carbonyl group remains intact. The restoration of the carbonyl character to the monomer hydrate 5 can be projected in hydrophobic environment of the cell membrane by loss of water, hence its mutagenicity is also explicable. The substantial activity of the dimer hydrate 7 can be attributed to its A# ready equilibrium with the monomer hydrate 5 in aqueous medium. The relation- ship of the activity of the cyclic trimer 8 to the chloroacetaldehyde action cannot be deduced on the basis of the curves in Figure 4. Nevertheless, these mutations were the base-substitution type and were expressed in Salmonella strains TA 100 and TA 1535. The mutagenic response in TA 1535 was very weak as compared to TA 100. It should be noted that the latter contains an "R" factor that enhances mutation by an error-prone recombination repair mechanism following DNA damage [7]. Comparison of mutagenic response of chlorooxirane with its methylene homoloq epichlorohydrin - The prevailing opinion rsl is that chlorooxirane 4 is the primary metabolite of vinyl chloride and is derived from the action of the microsomal mixed function oxidase. This compound, prepared independently from chlorination of ethylene oxide, was found to rearrange readily to chloro acetaldehyde in aqueous or DMSO solution at ambient temperatures. A kinetic study of a 0.15 M solution of 4 in a D20-DMS0-d5 (80:20) mixture at pD 7.1 and 4*C by PMR technique showed that the rearrangement followed first order kinetics, k (sec"1) = 2.5 x 10"4. This translates to a half life of 46,2 min CMA- 003781 at 4C as compared to 1,6 min at 37 C[9]. Such instability allows only limited testing in the cold as well as interpretation of the results. On the other hand, epichlorohydrin 9, which can be considered the epoxidation metabolite of allyl chloride, is a methylene homolog of 4. Both 4 and 9 are structurally bis-alkylating agents, hence comparing their mutagenic activities will lend insight into the mode of action of chlorooxirane 25,26], Table IV tabulates the mutagenic response of these two compounds in the Bacillus repair assay and Table VI the Salmonella TA 100 reversion. Due to the instability of chlorooxirane at 37C, the assays shown in Table VI were preincubated at 3C in solution with the Salmonella TA 100 strain before plating to insure that chlorooxirane could reach the cells intact. Mutant strains of B. subtil is were not inhibited when exposed to high concentrations of epichlorohydrin 9. v In contrast, chlorooxirane 4 selectively inhibited the rec- strain MC-1 in a manner similar to chloroacetaldehyde hydrate 6. Tests with Salmonella showed that strain TA 100 was very susceptible to the mutagenic action of both epoxides 4 and 9. However 9 was less toxic to the tester strains than 4. The low toxic effects of 9 indicate a different type of DNA lesion compared to that 'V caused by chlorooxirane 4, It is possible that 9 may react with DNA by a 'V, ^ mechanism which does not cause potential lethal strand-scissions. On the other hand, chlorooxirane 4 may act on the bacteria via a NIH shift [27,28] to form chloroacetaldehyde 5 or 6. Conceivably, chlorooxirane can also behave as a diradical intermediate rather than a conventional S^l or 5^2 type alkyl ating agent in its reaction with DNA. Vinyl chloride carcinogenesis mechanism hypothesis - Among the compounds tested in the metabolic Scheme II, chloroacetaldehyde 5 and chlorooxirane 4 003762 i0 SCHEME II: The metabolic pathways of vinyl chloride [5], I. C1-CH=CH, 1- Cl-CtL-QL-OH 2- SCHEME I urine alcohol Cl-CHrCH0 5 dehydrogenase C1-CH2-CH0 5- C1-CH2-C02H 3 SCHEME I urine II. ci-ch2-ch2-oh H22 ^ ase -ci-ch2-ch2-ooh C1-CHc5-CH0 <5v III. oxidase1 C1-CH=CH2 1------------------'V Cl-CH-CHo0 4 I____ h * C1-CH--CH0 5 & 'V/ were the most mutagenic with the lowest toxic side effects. Hence, they may * qualify to be the active carcinogenic derivatives of vinyl chloride. Consider ing the aqueous milieu of the metabolic environment, however, the chloroacetaldehyde monomer hydrate 6 is a more realistic choice as an ultimate carcinogen than the monomer 5 which reacts immediately with water. Viability assays followina exDOSure of the B. subtil is mutants to 6 indicated that the compound --- *\f induced recombination repair. The strain with the rec phenotype (MC-1) was immediately inactivated. All the rec+ strains showed survival ability and repair kinetics of similar nature. It has been shown [27] that mammalian cells have postreplication repair of DNA which has many similar features to the recombination repair in bacteria. In addition, a relationship between this mammalian postreplication repair process and mutation as well as carcinogenesis 003163 SCHEME II: The catabolic pathways of vinyl chi erica [ 5 ]. Cl-CH=CH, C1-CH2-CH0 5- - '-'It C1-CH2-C02H 3 [5lcohol dehydrogenase jr'.re C1-CH2-CH0 5 SCHEME I urine II. C'l -Cn^-CH^-OH h202 cataIase :i-CH,,-CH--CCH Cl-CHo-CH3 5 III. oxidase CI-CH=CH2 1 ------------------ C1-CH-CH-0 4 I____ h C1-CH.-CK0 5 were the most mutagenic with the lowest toxic side effects. Hence, they nay qualify to be the active carcinogenic derivatives of vinyl chloride. Consider ing the aqueous milieu of the metabolic environment, however, the chloroacetal dehyde monomer hydrate 6 is a more realistic choice as an ultimate carcinogen than the monomer 5 which reacts immediately with water. Viability assays following exposure of the B_. subtil is mutants to 6 indicated that the compound induced recombination repair. The strain with the rec phenotype (MC-1) was immediately inactivated. All the rec+ strains showed survival ability and repair kinetics of similar nature. It has been shown [27]that mammalian cells have postreplication repair of DNA which has many similar features to the recombination repair in bacteria. In addition, a relationship between this mammalian postreplication repair process and mutation as well as carcinogenesis CMA 003764 has been suggested E4J. Cells from patients with the skin disease xeroderma pigmentosum lack the ability to excise pyrimidine dimers and must rely on postreplication repair to remove these lesions [24]. It is believed that this error prone process of postreplication repair is responsible for the production of somatic mutations and cancer in patients with this disease. Since chloroacetaldehyde monomer hydrate 6 induces recombination repair in bacteria 'V responding to lesions, it may also be capable of activating the error prone postreplication repair in exposed mammalian cells. At the molecular level, chloroacetaldehyde is known to react with N^ and N nitrogensof adenosine and N 3 and N4 nitrogensof cytidine in single-stranded DNA [28,29]. It therefore appears that chloroacetaldehyde monomer hydrate 6 should merit our consideration as an ultimate carcinogenic metabolite of vinyl chloride. The lower mutagenic activity of chloroxirane 4 compared to 6 may reflect the unstable nature of chlorooxirane as an a-chloroether. While the carcino genic chloromethyl methyl ether is a bifunctional alkylating agent [3CS, the mutagenic activity of chlorooxirane cannot be so categorized, especially when it is compared with epichlorohydrin 9. One mode of action of chlorooxirane is a rearrangement to chloroacetaldehyde via the NIH shift [27,28], Another is a homolytic ring cleavage to yield a stabilized diradical inter mediate CICH-CHgO. Both are capable of reacting with DNA, thereby accounting for the mutagenicity of 4. In mammalian cells, chlorooxirane,being a reactive epoxide, could be trapped by a glutathione epoxide transferase [31] at a faster rate than the detoxification of chloroacetaldehyde via the less active aldehyde dehydrogenase [32] . Such detoxification of chlorooxirane could CMA 003765 explain the reported decrease in sulfhydryl level in liver cells during metabolism of short chain halo hydrocarbons including vinyl chloride [31]. Perhaps the lower mutagenicity of chlorooxirane *4\t in these bacterial assays, compared to the chloroacetaldehydes, is also attributable to its being detoxified faster. We therefore consider both chlorooxirane 4 and the chloroacetaldehyde monomer hydrate 6 to be the ultimate carcinogenic metabolites of vinyl chloride. Their reactions with DNA causing mutations in Bacillus and Salmonella suggest a causal relationship with vinyl chloride carcinogenesis in human and laboratory animals. Acknowledgements This work was supported by grants from the B. F. Goodrich Co. This grant program was initiated and administered by the Cancer Center of the University of Louisville. The Salmonella tester strains were kindly provided by Dr. B. N. Ames of the University of California, Berkeley. We also thank George D. Stratton, Jr. and S. E. Yen for their able assistance. CMA 003766 explain ire reported decrease in Ai ulli fhydryl level in liver cells coring metabolism of short chain halo hycrccuroons including vinyl c.nl price l31 j . Perhaps the lower mutagenicity of chlorooxirane 4 in these bacterial assays, compared to the chloroacetaldehydes, is also attributable to its being detoxified faster. We therefore consider both chlorooxirane 4 and the cnloroacetaldehyde monomer hydrate 6 to be the ultimate carcinogenic metabolites of vinyl chloride. Their reactions with DNA causing mutations in Bacillus and Salmonella suggest a causal relationship with vinyl chloride carcinogenesis in human and laboratory animals. Ac know!edgements This work was supported by grants from the B. F. Goodrich Co. This grant program was initiated and administered by the Cancer Center of the University of Louisville. The Salmonella tester strains were kindly provided by Dr. B. N. Ames of the University of California, Berkeley. We also thank George D. Stratton, Jr. and S. E. Yen for their able assistance. CMA 003767 REFERENCES AND FOOTNOTES Iu *To whom correspondence should be addressed. 1 Viola, P. L., Bigotti, A. and Caputo A. (1971) Cancer Research 31, 516-522 2 Maltoni, C. and Lefemine, G. (1974) Environmental Research 7, 387-405 3 Creech, J. L. and Johnson, M. N (1974) Journal of Occupational Medicine 16, 150-151 4 Lee, F. I. and Harry, 0. S. (1974) Lancet 1, 1316-1318 5 Hefner Jr., R. E., Watanabe, P. G. and Gehring, P. J. (1975) Annals of New York Academy of Science 246, 135-148 6 Watanabe, P. G., McGowan, G. R. and Gehring, P. J., Annals of New York Academy of Science, in the press 7 McCann, J., Simmon, V., Streitwieser, D. and Ames, B. N. (1975) Proceedings of National Academy of Science U.S.A. 72, 3190-3193 8 Rannug, U.,Johnsson, A., Ramel, C. and Wachtmeister, C. A. (1974) Ambio 3, 194-197 9 Malaveille, C., Bartsch, H., Barbin, A., Camus, A. M. and Montesano, R. (1975) Biochemistry Biophysics Research Comnunication 63, 363-370 10 Bartsch, H., Malaveille, C. and Montesano, R. (1975) Internationl Journal of Cancer 15, 429-437 11 Hayduk, U. and Landle, H. (1974) Journal of Chemical Engineering Data 19, 253-257 12 Walker, G. (1953) Formaldehyde, p. 86, Reinhold, New York 13 Natterer, K. (1882) Monatsheffe fur Chcmie 3, 443-464 CMA 003768 19 14 Harris, G., ed (1965) Dictionary of Organic Chemistry, Vol. 2, pp. 589 956 15 McCann, J., Spingarn, N. E., Kobari, J. and Ames, B. N. (1975) Proceedings of National Academy of Science U.S.A. 72, 979-983 16 Okudo, S. and Romig, W. R. (1965) Journal of Molecular Biology 14, 130-142 17 laumbach, A. D. and Felkner, I. C. (1972) Mutation Research 15, 233-245 18 Vogel, H. J. and Bonner, D. M. (1956) Journal of Biological Chemistry 218, 97-106 19 Kada, T. Tutikawa, K. and Sadaie, Y. (1972) Mutation Research 16, 165-174 20 Walling, C. and Fredericks, P. S. (1962) Journal of American Chemical Society 84, 3326-3331 21 Gross, V. H. and Freiberg, J. (1969) Journal fur Praktische Chemie 311, 506-510 22 Zief, M. and Schramm, C. H. (1964) Chemistry and Industry April 18, pp. 660-661 23 Kappus, H. Bolt, H. M., Buchter, A. and Bolt, W. (1975) Nature 257, 134- 135 24 Hagstrom, A. (1974) Ambio 3, 77-79 25 Voogd, C. E. (1973) Mutation Research 21, 52-53 26 Fjellstedt, T. A. Allen, R. H., Duncan, B. K. and Jakoby, W. B. (1973) Journal of Biological Chemistry 248, 3702-3707 27 Lehmann, A. (1974) Life Science 15, 2005-2016 28 Barbin, A., Bresil, H., Croisy, A., Jacquignon, P. Malaveille, C., Montesano, R. and Bartsch, H. (1975) Biochemical and Biophysical Research Communications 67, 596-603 o03 14 Harris, (1 955) Dictiona-y of Organic Cr.enistry, Vol. 2, pp. 559 15 .-.Cuanr., J., Spingarn, .i. c., ;.csari, J. and Kms s, 3. J. iizj .-roccco :rgs of National Academy of Science U.S.A. 72, 979-983 16 Okudo, S. and Romig, W. R. (1965) Journal of Molecular Biology 14, 120-142 17 Laumbach, A. D. and Felkner, I. C. (1972) Mutation Research 15, 233-245 18 Vogel, H. J. and Bonner, D. M. (1956) Journal of Biological Chemistry 213, 97-106 19 Kada, T. Tutikawa, K. and Sadaie, V. (1972) Mutation Research 15, 155-174 20 Walling, C. and Fredericks, P. S. (1962) Journal of American Chemical Society 84, 3326-3331 21 Gross, V. H. and Freiberg, J. (1969) Journal fur Praktische Chemie 311, 506-510 22 Zief, M. and Schramm, C. H. (1964) Chemistry and Industry April 18, pp. 660-661 23 Kappus, H. Bolt, H. M., Buchter, A. and Bolt, W. (1975) Nature 257, 134- 135 24 Hagstrom, A. (1974) Ambio 3, 77-79 25 Voogd, C. E. (1973) Mutation Research 21, 52-53 26 Fjellstedt, T. A. Allen, R. H., Duncan, B. K. and Jakoby, W. B. (1973) Journal of Biological Chemistry 248, 3702-3707 27 Lehmann, A. (1974) Life Science 15, 2005-2016 28 Barbin, A., Bresil, H., Croisy, A., Jacquignon, P. Malaveille, C.,' Montesano R. and Bartsch, H. (1975) Biochemical and Biophysical Research Communication: 67, 596-603 CMA 003770 20 29 Kochetkov, N. K. Shibaev, V. N., and Kost, A. A. (1971) Tetrahedron Letters 22, 1993-1996 30 Leong, B. K., Macfarland, H. N. and Reese, W. H. (1971) Archives of Environmental Health 22, 663-666 31 Johnson, M. K. (1965) Biochemical Pharmacology 14, 1383-1385 32 Weiner, H., King, P., Hu, J. H. J. and Bensch, W. R. (1974) Alcohol and Aldehyde Metabolizing Systems (Thurman, R, 6., ed.), pp. 101-113, Academic Press, New York CtAA. 003'7'71 LEGENDS TO FIGURES fig. 1. Stability of vinyl chloride in nutrient broth at 25C, 1 atm. Fig. 2. Survival of Bacillus subtil is MC-1 after incubation with: 1.0 nil chloroethanol 2 ( '5*--^); 1.0.nM cHToroacetic acid ,3 (O -- o); 5.76 mM (concentra tion based on CIH2CCHO) chloroacetaldehyde (45% aqueous solution) 6 and 7 ( 0-t> 0.1 mM 4-ni troquinol ine-N-oxide (0-0); and untreated control cells (0--0), Mid-logarithmic cultures grown in MV-1 broth were incubated with compounds at 37C. Samples were plated at the times indicated from which viable cell counts were recorded. The mutagen 4-nitroquinoline-N-oxide served as a positive mutagenicity control. Fig. 3. Survival of Bacillus subtil is strains in the presence of 5.0 mM (concentration based on ClHgCCHO) chloroacetaldehyde (45% aqueous solution) and 7. Cultures were grown to mid-logarithmic phase in MY-1 broth and then incubated with the chloroacetaldehyde solution at 37C. Samples were plated at the times indicated from which the percent survival was determined. FB-13 uvr", rec+ (! -- +);Hcr-9 her", rec* (0--0); 168M wild type (0 -- 0); and MC-1 uvr+, rec- (e> -- o). Fig. 4. Dose response curves with Salmonella typhimurium TA100. Sample solutiot of known concentrations prepared in UMSO were mixed with tester strain culture and soft agar. Plates were poured, incubated at 37C for 48 hrs, and then^^pn for revertant colonies to prototrOphy. Chloroacetaldehyde monomer 5 ( chloroacetaldehyde (45% aqueous solution-concentration based on GH2CCHO) 6 and 7 (0-0); chloroacetaldehyde dimer hydrate .{ i>-- >); chloroacetaldehyde trime jg ( x--x); and epichlorohydrin ^ a). CMA 003772 :s TO FTC Fig. 1. Stabi'i ty of .,ry I V -j ci . r, ,, --^ a . - 0 u < rig. 2. Survival of Bacillus subti 11 s .VC-1 after incubation with: 1.0 r!' chioroethanol ( ^S* -- 'iTo nM ct1 oroacetic acid 3 (o -- a); 5.75 m" (cancer,tra tion based on Cll^CCHO) chloroaceta idehyde (45,'S aqueous solution) 6 and 7 ( >-- 0.1 mM 4-nitrocuinoline-N-oxide (0-0); and untreated control cel^fs (o --o). Mid-logarithmic cultures grown in MY-1 broth were incubated with compounds at 37aC. Samples were plated at the times indicated from which viable cell counts were recorded. The mutagen 4-nitroquinoline-N-oxide served as a positive mutagenicity control. 3. Survival of Bacillus subti1 is strains in the presence of 5.0 (concentration based on C1 r^CCHG) cn1crcacetaIdehyde (45* aqueous solution) .5 and 7. Cultures were grown to mid-logarithmic phase in MY-1 broth and then incuBated with the chlorcacetaldehyde solution at 37C. Samples were plated at the times indicated from which the percent survival was determined. F3-13 uvr", rec+ (!-- +);Hcr-9 her", rec* (0--0); 163M wild type (o --o); and MC-1 uvr+, rec" (c --1>). Fig. 4. Dose response curves with Salmonella typhimurium TA100. Sample solutic of known concentrations prepared in LJMSQ were mixed with tester strain culture and soft agar. Plates were poured, incubated at 37C for 48 hrs, and then scot for revertant colonies to prototrophy. Chloroacetaldehyde monomer 5 (o-o); chloroacetaldehyde (45% aqueous solution-concentration based on CIH2CCHQ) 5 anc 7 (0-0); chloroacetaldehyde dimar hydrate ^.(4>-o); chloroacetaldehyde trims % ( x--k); and epichlorohydrin (u-*0). 03 773 COHCEHTRATIOK M ILL! KOLAS Figure Xoo Fiyui e L 10 C E ILS /m l i i CMA 003776 ___ J___ 10 is GO m iuius Fic 4 / ,0 / TABLE I: Bacteria Tester Strains A. Salmonella typhimurium LT-2 Tester Strains3 aAll tester strains contain uvrB repair mutations which eliminate the excision repair system; mutations in the histidine operon; and rfa mutations which alter the cell wall by increasing permeability and eliminating pathogenicity. bThe resistance transfer factor,"R" factor,enhances the error-prone recombination repair system thus making the strains more susceptible to mutation [15]. cThe strains susceptible to base-pair substitution contain mutations in the histidine G46 operon and those susceptible to frameshift mutation contain mutations in the histidine operon C 3076 (TA 1537) or D3052 (TA 1538, TA 98). Strains "R" factor6 Mutation detected0 TA 1535 TA 100 TA 1537 TA 1538 TA 98 _ base-pair substitution + base-pair substitution - frameshi ft - frameshift + frameshift B. Bacillus subtil is Tester Strains aTrp" denotes a requirement for tryptophan; Mit-S denotes sensitivity to mitomycin C. bhcr+ denotes a host-cell reactivation DNA repair capacity; her" lacks a host-cell reactivation DNA repair capacity; rec+ denotes a recombin ation DNA repair capacity; rec" lacks a recombination DNA repair capacity; uvr- is sensitive to ultraviolet-induced DNA damage. _________ Strains Phenotype3 DNA Repair' 168 M llcr-9 FB-13 MC-1 Prototroph (wild type) Trp" Trp" Trp",Mit-S her , rec her", rec uvr", rec her , rec CMA 003780 TABLE II. PMR Spectra of Vinyl Chloride Derivatives Compounds C1HC-CH20 4 C1H2C-CH0 5 0. C1H2C-CH(0H)2 6 'V C1H,C-CH-0H `1 0 1 cih2c-ch-oh 7 *\/ CHoCl A 1 (T^n`0'^>ch9ci 2L 8 Solvent CC14 CC14 DMSO-dg CD-jOD-DpO 1:8 3(pD 6.1) Spectra,, 6TMS*0 (JH- ' 2.75 (q, CH, J-1.5) 2.85 (q, CH, 0=2.4) 4.90 (q. CH, J-2.4,1.5) 4.00 (d, CH, 3=2.2) 9.57 (t. CH?, 0=2.2) 3.50 (d, CM) 9.60 (t, CH2) 3.60 (d, CHo, 0=5) 4.60 (t, CH, 0*5) DMSO-Dg CDoOD-D 0 1:8 3(pD 20.1) 3.55 (d, CH,, 0=4.9) 5.05 (t. ch; J=4.9) 3.60 (d, CH,, 0=4.5) 4.83 (t. CH, 0*4.5) cci4 DMSO-dg 3.52 (d, CHo, 0=4.7) 5.08 (t. CH7 0=4.7) 3.75 (d. CHo, 0=4.1) 5.45 (t, CH, 0=4.1) CMA 0378l TABLE II. ?:*3 Spectra cf Vinyl eWorld' Osriv a tives Cc.~p0ur.d5 C1KC1 -CM2.0, *4 C1H2C-CH0 5 C1H2C-CH(0H)2 6 Sol vent cci4 CC14 DMSO-dg CD,CD-D?0 1:8 3(pD 0.1) Stectra,, - i. - = u I ^ . 2.75 (q. CH. >1.5) 2.S5 (q, CH, >2.4) 4.90 (q, CH, >2.4,1.5) 4.00 (d, CH, >2.2) 9.57 (t. CH?1 0=2.2) 3.50 (d, CH) 9.60 (t, CH2) 3.60 (d, CH2, 0=5) 4.60 (t, CHT >5) C1H-C-CH-0H cl 0 ` C1H2C-CH-0H 7 ' DMSO-Dg 3.55 (d, CH-, 0=4.9) 5.05 (t, CHf >4.9) CD.OD-O.O 1:8 3.60 (d, CH-, 0=4.5) J(pO z0.1) t 4.83 (t, CHT >4.5) CH2C1 A H2<r^0^vCH2Cl 5 cci4 DMSO-dg 3.52 (d. CH2, 0=4.7) 5.08 (t, CHT >4.7) 3.75 (d, CH-, >4.1) 5.45 (t. CHT 0-4.1) CMA 003782 TABLE III: Summary of Mutagen Activity in Microbial Systems NI - No inhibition of growth detected in Bacillus subtil is MC-1; NR = No increase of revertants in Salmonella t.yphimurium iA 100 compared to control;+ =active; ++ = very active. Acetaldehyde, a potential metabolite of and allyl chloride, the parent olefin of were negative in these two systems. Compounds Tested Control: 4-nitroquinoline-N-oxide 1v H,C=CHC1 c 2 C1H2C-CH20H 3 C1H9C-C00H c. 4 ClHC-CH,,-0 * 1 2_i 5 C1H-C-CH0 \ t- 6 C1H2C-CH(0H)2 7 ClH,C-CH0H-0-CH0H-CH9Cl 8 (ClH2C-CH0-)3 9 C1H,,C-CH-CH,,0 * 2 1____ l Bacillus subtil is Repair Assay ++ NI NI NI + ++ ++ ++ ++ NI Salmonella typnimurium Reversion Assay ++ NR NR NR ++ ++ ++ ++ + 44 CMA 003783 TABLE IV: Growth Inhibition of Bacillus subtil is Strains . Inhibition was measured in mm after 24 hr at 37C as described in text; NI denotes no inhibition Mutaqen Molarity C1Hc.C-CH0 ^5 0.100 Chloroacetaldehyde & (45% aqueous solution) 0.115 C1H0C-CH-QH 2i 7 0 n, cih2c-ch-oh 0.097 168M 2.0 NI 1.5 MC-1 27.7 22.5 9.5 Hcr-9 3.7 FB-13 2.7 NI NI 1.5 1.5 CIH2Cv/\^CK2CI 1 CII2CI % C1HC-CH70 1___ "J 4 * C1H92C-Ci2Hj-CH,0 9n. 0.096 0.260 0.113 4-Nitroquinoline-N-oxide 0.001 (control) 7.0 NI NI 10.0 14.5 10.0 NI 18.0 6.0 NI NI 15.0 .0 NI NI 15.C CMA 003784 k n r* i j i u i o n wd s r a d ^ ^ r=o in ::n after 24 nr at 37C as descr d anotes no inh'bi tion '** Mutacen cih2c-cho 5 Molarity 0.100 Ch'onscataldshyde 5^i 7 (vO-j SC t ^ w t vt* ^ G* l *- CliLC-CH-CH 21 7 0 -C C'H2C-CH-0H 0.097 1 53" 2.0 111 1.5 MC-1 27.7 22.5 9.5 Hcr-9 3.7 F3-1 2.7 l\ i 11 * 1.5 l.E cn:2e^\ -ck2ci Ci!2Cl 8 0.096 CIHC-CH-O i___ h 4 'V 0.260 C1H2,C-Ci_H_- CH2-01 9~ 0.113 4-Nitroquinoline-N-oxide 0.001 (control) 7.0 NI NI 10.0 14.5 10.0 NI 18.0 6.0 NI NI 15.0 7.( NI NI 15. CMA 003785 30 TABLE V. Relative Mutagenicity of the Four Forms of Chloroacetaldehyde with s.typhimurium TA 100 45% Aqueous Soln 6 : 7 50:50 Monomer 5 Dimer Hydrate 7 Trimer 8 Molarity 5.3xl0-5 2.7xl0'5 1.4xl0'5 5.9x1 O'6 3.4xl0"6 1.7x1 O'6 8.6xl0"7 4.3x1 O'7 Revertants Molarity 977 1.3xl0'5 723 6.7xl0"6 512 3.3xl0"6 194 1.7x10`6 120 8.4xl0"7 61 4.2xl0-7 36 2.1xl0`7 10 l.lxl0-7 Revertants 18 68 88 361 404 238 185 131 Molarity 4.8xl0`4 2.4x1 O'4 1.2xl0"4 6.OxlO'5 3.0xl0'5 1.5x1 O'5 7.5x1 O'6 3.8x1 O'6 Revertants 311 259 193 107 83 30 23 11 Molarity 4.8x1 O'4 2.4xl0"4 1.2xl0**4 6.0xl0-5 3.OxlO'5 1.5xl0`5 7.4xl0"6 3.7x1 O'6 Revertants 144 159 .101 39 27 18 12 -0 CMA 003786 TABLE VI: Reversion of S.typhimurium TA1Q0 by Chlorooxirane 4 and Epichlorohydrin-!} 'V. ^ Broth solutions of or ^ with TA100 were preincubated at 3C before plating. Duplicate plates were evaluated after 48 hrs at 37C.'` The average number of revertant colonies per plate minus the number of spontaneous reversions were recorded. Preincubation Time 0 hr 1 hr 2 hr 4 hr 6 hr Epichlorohydrin 9 (1.0 mM) ^ 186 204 297 202 154 Chlorooxirane 4 (0.26 mM) ^ 31 4 6 114 44 CMA 003787 Broth solutions of or 9 v/ith TAT CO 'were preincubated at 3C before plating. Cupiicar.e pioles ,<ere evaluated after 43 hrs at 378C.J The average number of revertant colonies per plate minus the number of spontaneous reversions were recorded. Preincubation Tire_________ 0 hr 1 hr 2 hr 4 hr 6 hr Epichlorohydrin 9 n ,o -"i______ ^ 135 204 297 202 154 Chlorooxirane 4 (0.26 -') ^ 31 4 6 114 44 CMA 003788 microbiology series ri-Jt volume 5 - iswv ;v:,v . \\'&~"}. - : '-' -- .. ~"'v'"' -v. "" _ : .\. ' ,'tsS?. ^edited by- vn-l . v$?vv ** - -' V'#* . W -1 . LCecil Felkner . * >-- >jv. '*?*.* ^ w' . -- :*: * CMA 003789 BACTERIAL MUTATION MONITORS FOR ACTIVE METABOLITES OF CHEMICAL CARCINOGENS: B, SUBTILIS ASSAYS FOR MUTATION AND DNA REPAIR ULDIS N. STREIPS University of Louisville, Louisville, Kentucky ANDREW D. LAUMBACH Food and Drug Administration, Washington, D. C, RONALD E. YASBIN The Pennsylvania State University, University Park, Pennsylvania I. Introduction II. Microbial Assays A. Salmonella Reversion Test B. Bacillus Repair Assay C. Bacillus Forward Mutation Analysis D. Bacillus Comptest III. Discussion References 13X 134 134 135 137 137 141 143I. I. INTRODUCTION Chem'ical monomers, such as vinyl chloride, styrene, and acrylylnltrile, are of great importance in many industrial synthesis reactions. They are the fundamental components of polymers used extensively in the economy of the industrial nations. The detection of tumors (e.g., angio sarcoma) In workers exposed to vinyl chloride in the Industries of several nations suggested that there was a relationship between this industrial monomer and the development of hepatic disorders (13,40). As a result, both industry and government have become Increasingly active In research 131 CMA 003790 132 U. N. Streips, A. D. I.aumbach, K. E. Yasbin concerning the potential carcinogemcity of extensively used chemical com pounds. Unfortunately, the assays for tumor production arc time consuming (70) and the number of existing and newly produced chemical formulations huge. Therefore, it is imperative that rapid assays for mutagenicity and potential carcinogenicity be performed to narrow the spectrum of chemicals which should be tested for tumor production in animal systems. Epoxides are generally thought to be intermediates in microsomal, enzyme-mediated oxidation of olefinic compounds (29). As such, chlorooxlrane and its rearrangement product chloroacetaldehyde (CAA) (20) were suspected to be the carcinogenic metabolites of vinyl chloride. Several laboratories have confirmed the mutagenicity of these compounds (4,20,39, 42,45,59). Our laboratory also tested the various multimer forms of chloroacetaldehyde, as well as chlorooxlrane, and found them to be muta genic (20). It is reasonable to assume that other industrial monomers may also form reactive metabolites upon oxidation. In this chapter we describe several rapid assays, primarily utilizing DNA repair functions as the screening methodology, for the detection of biological activity in known mutagens [ethylmethanesulfonate (EMS) and methylmethanesulfonate (MMS)] as well as in the oxidation products of th chemical monomers vinyl chloride and styrene. Repair of mutagen-elicited lesions in DNA is Important for the survival of affected cells. However, t DNA repair has also become Increasingly implicated in pathways leading to additional mutagenesis and possibly carcinogenesis (49,50,73). The discovery that certain UV-sensitive mutants of Escherichia coli are not mutated following UV radiation'led to the hypothesis of error-prone and error-free types of DNA repair (73,75), There are three distinct pathways for the repair of damaged DNA following radiation or exposure to chemical agents (photoreactivation, excision repair, postrepltcation repair). The process of photoreactlvation involves the binding of the pyrimidine dimer by the photoreactivating enzyme and the "splitting" of the dimer following the exposure of the enzyme-dimer complex to photoreactlvating light (37,61,66), Significantly, this mechanism of DNA repair does not result in the production of mutations and is therefore error-free (5,74,75). Damaged DNA can also be repaired in the dark via the excision mecha nism (10,69). Excision repair (uvr genes) Involves the enzymatic recogni tion and removal of the damaged DNA followed by repair replication (56). In this process, the single-strand gap produced following the removal of the damaged base or bases is filled in by a DNA polymerase which uses the undamaged strand as a template. In E, coli. the most efficient and accu rate excision repair system requires, among other gene products, a func tional DNA polymerase I (12,24,36). However, alternate pathways of ex cision repair have been demonstrated in cells lacking DNA polymerase I (44, 82, 84, 85). Excision repair is also virtually error-free, although there is a low frequency of mutation which has been ascribed to errors in the alternate pathways of this type of DNA repair (8,53,74). One of our rapid tests, the DNA repair assay, examines the ability of colls lacking this typo of repair to survive chemical damage (Tabic 3). CMA 003791 t u. i 7'; ...:J Lw_ number of existing and newly produced chemical focmalatioas huge. Therefore, it is imperative* that rapid assays tor mutagenicity and potential carcinogenicity be performed to narrow the spectrum of chemicals which should be tested for tumor production in animal systems. Epoxides are generally thought to be intermediates in microsomal, enzyme-mediated oxidation of olefinic compounds (29). As such, ehtorooxivane and its rearrangement product chloroacetaldehyde (CAA) (20) were suspected to be the carcinogenic metabolites of vinyl chloride. Several laboratories have confirmed the mutagenicity of these compounds (4,20,39, 42,45, 59). Our laboratory also tested the various multimer forms of chioroncataldenyde, as well as chlorooxirane, and found them to be muta genic (20). It is reasonable to assume that other industrial monomers may also form reactive metabolites upon oxidation. In this chapter we describe several rapid assays, primarily utilizing DNA repair functions as the screening methodology, for the detection of biological activity in known mutagens [ethylmethanesulfonate (EMS) and methylmethanesulfonate (MMS)] as well as In the oxidation products of the chemical monomers vinyl chloride and styrene. Repair of mutagen-ellclted lesions in DMA is Important for the survival of affected cells. However, DNA repair has also become Increasingly Implicated In pathways leading to additional mutagenesis and possibly carcinogenesis (49,50,73). The discovery that certain UV-sensltiv* mutants of Escherichia coli are not mutated following UV radiation'led to the hypothesis of error-prone and error-free types of DNA repair (73,75). There are three distinct pathways for the repair of damaged DNA following radiation or exposure to ch mical agents (photoreactivation, excision repair, postrepltcatlon repair). The process of photoreactivation involves the binding of the pyrimidine dimer by the photoreactivatlng enzyme and the "splitting" of the dimer following the exposure of the enzyme-dimer complex to photoreactivatlng light (37, 61,66), Significantly, this mechanism of DNA repair does not result in the production of mutations and Is therefore error-free (5,74,75). Damaged DNA can also be repaired in the dark via the excision mecha nism (10,69). Excision repair (uvr genes) involves the enzymatic recogni tion and removal of the damaged DNA followed by repair replication (56), In thts process, the single-strand "gap produced following the removal of ' the damaged base or bases is filled in by a DNA polymerase which uses the undamaged strand as a template. In E, colt, the most efficient and accu rate excision repair system requires, among other gene products, a func tional DNA polymerase I (12,24,36). However, alternate pathways of ex cision repair have been demonstrated in cells lacking DNA polymerase I (44, 82, S4, 85). Excision repair is also virtually error-free, although there Is a low frequency of mutation which has been ascribed to errors in the alternate pathways of this type of DNA repair (8, 53,74). One of our rapid tests, the DNA repair assay, examines the ability of cells lacking this typo of repair to survive chemical damage (Table 3). CMA 003792 6 / tinctorial Mutation Monitors for Ac tive Metabolites 133 Pholoreaclivation and excision repair are processes which restore the integrity of the damaged chromosome prior to the replication of the DNA. Following the replication of the DNA, integrity can still be restored by the proecss(cs) of postreplication repair. Postrepliealion repair is believed to involve the filling of gaps left in daughter strand DNA following the repli cation of damaged DNA (32, 62). The closing of these single-strand gaps is accomplished by at least two types of mechanisms in E. coli (83). First, a constitutive type of postrcplication repair has been shown to involve recom bination (22,23,63). Second, one or more types of postreplicatlon repair are inducible and their mechanisms of action are unknown (64,65). Postreplication repair has an error-prone element (6,35,65,73,74), which is an inducible, independent minor pathway. Specifically, mutants of E. colt lacking the products of the recA+ and/or lexA+ genes are unable to Induce this error-prone repair system (48,75). All postreplication repair in E. coli is dependent on a functional recA+ gene product (68) while Inducible postreplication repair is also dependent on a functional lexA+ (exrA+) gene product (64). In addition to error-prone, inducible postreplicatlon repair, recA and/or lexA mutations prevent the Induction of prophage (9,18), Welgle (72) or UV reactivation, and W mutagenesis (15,48,54), the induction of recA+ gene product (21, 26, 27,47,64), the inhibition of exonuclease V (43), as well as other physiological changes following the inhibition of DNA repli cation (69,74). The pleiotroplc effects of these mutations led Radman (58) to propose the SOS hypothesis. This theory contends that damage to DNA and/or the inhibition of DNA replication results in the release of a signal which simultaneously activates various functions which aid the cells and/or prophage In survival (74). Therefore, error-prone repair Is the result of an efficient but Inaccurate repair mechanism which is induced in an eff rt to prevent cell death. Based on the observations of certain inducible phenomena, the presence of an SOS system can be assumed in nonenteric bacteria (28,30,67,76,86), In lower eucaryotes (51,57), as well as in mammalian cells in culture (7, 14,34,41). It is important to note that the components of SOS systems appear to differ from species to species. Specifically, an Induced DNA modification system has been found in Bacillus subtilis (76), whereas In Haemophilus Influenzae. W reactivation has been demonstrated although error-prone repair Is not present (28, 33,67). Thus, there appears to be significant divergence in the composition of SOS systems. In mammalian systems the relationship between the mechanisms of DNA repair and mutagenesis Is not as readily obtainable as in the prokaryotes. However, there do appear to be analogous mechanisms of repair in the two systems. Specifically, chemical damage (GO) as well as photochemical damage (11) of DNA can be repaired by mammalian cells. Recently, interest has begun to focus oa the possible effect of inducible SOS-likc systems on carcinogenesis (49,74). In the tester systems devel oped by Ames and his colleagues (1,2,46), most of tho carcinogenic agents tested have been shown to require a functional error-prone repair system CMA 003793 134 Lf. N. Strcips, A. L). I.aumbnch, R. E. Yashin in order to generate mutations in the hacteria (40). A phage induction func tion has been proposed for their system (71). In addition, aflatoxin and `1-nilt ociuinoline-1 -oxide (known carcinogenic agents) both induce prophage X and therefore activate the bacterial SOS system (25, ;tl). Moreau et at, (49, 50) developed a tester system for carcinogenic agents which utilizes the induction of prophage X. Based on their results, they have suggested that an SOS-likc system, including induction of proteases, exists in eucaryotes, and that the activation of this system can lead to cancer. If this theory is correct, a model system in which precocious activation of SOS functions occurs should be a sensitive tester for potential carcinogens. The bacterium B. subtilis possesses many of the desired prerequisites for this type of sen sitive tester system. It has been demonstrated that when B. subtilis differentiates into its com petent state, SOS functions are precociously activated (76,77). In addition, competent B. subtilis cells have enhanced sensitivity to UV irradiation (55, 77, 80). This enhanced sensitivity is the result of the Induction of defective prophages in the competent bacteria (Refs. 77 and 77a). Removal of the defective prophages and/or the inhibition of SOS induction eliminated the enhanced sensitivity of the competent cells to UV energy. We have included this review to substantiate our rationale in utilizing several types of repair assays, each measuring a different step in the re pair process, as a part of our battery of rapid assay systems. It is evident that bacteria have several methods for repairing both UV-induced and chem ical damage to DNA. Therefore, assays which measure for the induction of error-prone repair (SOS) or for the necessity of repair (our series of re pair assays) represent valuable information when considering the biological activity of potential mutagens and/or carcinogens.II. II. MICROBIAL ASSAYS A. Salmonella Reversion Test The mutagenic capability of the postulated active derivatives of the chemi cal monomers vinyl chloride and styrene, as well as ethylmethanesulfonate (EMS) and methylmethanesulfonate (MMS), was examined using the Salmo nella typhimurtum tester strains described by Ames (1) using methodology already published by McCann et al. (45) and by our laboratory (20,39) (Table 1). Chloroacetaldehyde (CAA) was prepared, purified,, and analyzed by previ ously reported techniques (20). Styrene oxide was prepared by treating styrene in methylene chloride with 1.2 equivalents of M-chloroperbenzolc acid at room temperature for several days. The solution was washed successively with aqueous NaSO^lI and aqueous NallCOj, then dried and evaporated. Both EMS and MMS were obtained from the Aldrich Chemical Company (Milwaukee, Wis.). CMA 003794 i;M C'u: in ordvr to gencrriLC mut-lions ia t'.rj itufin (-III). A phngc ir.-i'Jc:io:i func tion ha?- 1 ecu f:n `v.:- - -l . ; In a;! 'ititu, " a. .; . - ; J i-i*ii s,` j4.,,z-1 *,,c-^.%.- ^ " - * c sk * *liO^t,iiio n^c .. -( . , . .s.'c ~, k ,'. * ,,* X m.:l lav! ofoi c acr. l'.uv.p '. ic iaitcnr.! SC.5 syntcn.: iJMoreau el a!, (4D, 5u) developed a tester system for carcinogenic agents which utilize^ the induction of prophage X. Based on their results, they have suggested that an SOS-likc system, including induction of proteases, exists in eucaryotes, and that the activation of this system can lead to cancer. If this theory is correct, a model system in which precocious activation of SOS functions occurs should be a sensitive tester for potential carcinogens. The bacterium B. subtilis possesses many of the desired prerequisites for this type of sen sitive tester system. It has been demonstrated that when B. subtilis differentiates into its com petent state, SOS functions are precociously activated (7G,77). In addition, competent B. subtilis cells have enhanced sensitivity to LIV irradiation (55, 77, 80). This enhanced sensitivity is the result of the induction of defective prophages In the competent bacteria (Refs. 77 and 77a). Removal of the defective prophages and/or the inhibition of SOS induction eliminated the enhanced sensitivity of the competent cells to UV energy. We have included this review to substantiate our rationale in utilizing several types of repair assays, each measuring a different step in the re pair process, as a part of our battery of rapid assay systems. It is evident that bacteria have several methods for repairing both UV-induced and chem ical damage to DNA. Therefore, assays which measure for the Induction of error-prone repair (SOS) or for the necessity of repair (our series of re pair assays) represent valuable information when considering the biological activity of potential mutagens and/or carcinogens.II. II. MICROBIAX ASSAYS A. Salmonella Reversion Test The mutagenic capability of the postulated active derivatives of the chemi cal monomers vinyl chloride and styrene, as well as ethylmethanesulfonate (EMS) and metbylmethanesulfonate (MMS), was examined using the Salmo nella typhimurium tester strains described by Ames (1) using methodology already published by McCann et al. (45) and by our laboratory (20,39) (Table 1). Chloroacetaldehydo (CAA) was prepared, purified* and analyzed by previ ously reported techniques (20). Styrene oxide was prepared by treating styrene in methylene chloride with 1.2 equivalents of M-chloroperbenzolc acid at room temperature for several days. The solution was washed successively with aqueous NaSOgH and aqueous NallCOg, then dried and evaporated. Both EMS and MMS were obtained from the Aldrich Chemical Company (Milwaukee, Wis.). CMA 003795 Table I. Bacterial Strains Used in These Studies Strain Relevant repair genotype Source S. typlumurluin TA100 B. subtilis GSV1025 B. subtilts MC-1 B. subtilis GSY1627 B. subtilis VUB214 B. subtilis BD224 B. subtilis HA10G B. subtilis VQB133 B. subtilis RUB827 B. subtilis FB13 B. subtilis HC-9 B. subtills GSY1641 B. subtilis 168W B. subtilis RUB818 uvrH (it factor) recAl recB2 recD27 recD27 rec K4 recF7 recH342 polAS uvr her mtc-41 wild type wild type B. Ames C. Iladden C. Hadden c. Hadden c. Hadden c. Hadden c. Hadden c. Hadden R. Yasbin C. Hadden C. Hadden c. Hadden A. Laumbach R. Yasbin The assays with the S. typhimurtum strain TA100, an indicator for base pair substitution mutations (45), revealed that all four chemicals had con siderable mutagenic activity (Table 2). Our laboratory has shown previously that the chloroacetaldehyde consists of several different forms, all muta genic in this assay (20). None of these chemicals caused significant rever sion with strain TA98, a framesht(t indicator. B. Bacillus Repair Assay The repair assay, using various repair-deficient strains (Table 1) of B. subtil is. has been described by our laboratory (20,39). This is an indirect mutation assay, modified from the procedures described by Kada et al. (33). As discussed in the introduction to this chapter, when a mutagenic agent Table 2. Mutagenicity of Chemical Monomers Assayed by Salmonella TA100 Reversion___________________________ ________________________ Concentration in soft agar Average number Compound laver. mM/plate revertants/platea Chloroacetaldehyde 0.10 456 Styrene oxide 0.10 620 Methylmethanesulfonate 0.05 956 Ethylmethanesulfonate 0.10 295 Spontaneous revertants have been subtracted. CMA 003796 130 U. N. Streips, A. I). I.aumbaeh, R. E. Yasbin affects DNA, the cell is required to repair the damage. If the cell is unable to express the repair capability, the majority of cells will not survive. Thus, a measure of the biological reactivity of a chemical is to test it against several strains, one being wild type for repair, the others being deficient in one of the repair systems. If the wild tjpc strain survives but any of the repair mutants do not, the chemical is assumed to be reactive (see also Chapter 4). In Table 3 we show the reactivity of CAA, MMS, EMS, and styrene oxide against a scries of repair mutants of B. subtilis. First of all, styrene oxide does not appear to be reactive in any of these assays. This suggests that over a wide spectrum of concentrations tested, styrene oxide does not cause damage to DNA, which must be repaired by the available repair sys tems. Even when these tests were performed with cold incubation (see Chap. 4; Refs. 33 and 63), a procedure which increased sensitivity to CAA threefold, there was no discernible reaction with styrene oxide. On the other hand, all forms of CAA (20), MMS, and EMS were quite reactive in this system. However, these compounds affected only strains which Table 3. Effect of Chemical Monomers on Repair-Assay with B. subtilis Strain GSY102S Relevant repair genotype recAl _______________Growth inhibition. mma Chloroacetaldehyde, Styrene oxide, 0.1 M 0.1 M 13. S NI MMS, EMS, 0.1 M 0.1 M 14.5 8.1 MC-1 recB2 11.3 NI 13.0 6.4 GSY1627 recD27 11.5 NI 11.2 4.S VUB133 recD27 HI NI NI NI BD224 recE4 8.8 NI 16.4 6.1 HA 106 recF7 6.1 NI 11.5 7.2 VUB133 recH342 NI NI 3.0 1.5 Hcr-9 ; FB-13 GSY1641 her uvr mtc-41 NI NI NI NI NI NI NI NI NI. NI NI XT 168W wt NI NI NI NI aGrowth inhibition was measured In millimeters from the edge of the disc to the first evidence of growtlv (radius). Values are an average of seven experiments. Inhibition due to toxicity, as measured on wt cells (168W), has been subtracted from these values. Note: N1 = no growth inhibition. CMA 003797 13G Lr. ch, R. E. Yashin r.'.'.'-'w1.j Ci"'.. . . - -:o v.vn:-.'-i `.'.v t. l ^ is unable .vjr,i\c. T'y, r w" r.:im . , '-! - " 1' '>f a ; b.-nI is to tes; it against several strains, one being wile! type tor repair, the others being deficient in one of the repair systems. If the wild type strain survives hut any of the repair mutants do not, the chemical is assumed to he reactive (see also Chapter 4). In Table 3 we show the reactivity of CAA, AIMS, EMS, and styrene oxide against a scries of repair mutants of B. suhtilis. First of all, styrene oxide does not appear to be reactive in any of these assays. This suggests that over a wide spectrum of concentrations tested, styrene oxide does not cause damage to DNA, which must be repaired by the available repair sys tems. Even when these tests were performed with cold incubation (see Chap. 4; Piafs. 33 and 03), a procedure which increased sensitivity to CAA threefold, there was no discernible reaction with styrene oxide. On the other hand, all forms of CAA (20), MMS, and EMS were quite reactive in this system. However, these compounds affected only strains which Table 3. Effect of Chemical Monomers on Repair-Assay with B. subtllls Strain GSY1025 Relevant repair genotype recAl Growth inhibition. mma Chloroacetald ehyd e, Styrene oxide, 0.1 M 0.1 M 13.5 NI MMS, EMS. 0.1 M 0.1 M 14. 5 8.1 MC-1 recB2 11.3 NI 13.0 6.4 GSY1627 recD27 11.5 NI 11.2 4.S VUB133 recD27 Nl NI Nl NI BD224 recE4 8.8 NI 16.4 6.1 HA106 recF7 6.1 NI 11.5 7.2 VUB133 recH342 NI NI 3.0 1.5 Hcr-9 her NI NI NI NI FB-13 uvr NI NI NI NI. GSY1641 mtc-41 NI NI NI XT 168W wt NI NI NI NI "Growth inhibition was measured in millimeters from the edge of the disc to the first evidence of growth (radius). Values are an average of seven experiments. Inhibition due to toxicity, as measured on \vt cells (168\V), has been subtracted from these values. Note: NI " no growth inhibition. CMA 003798 were deficient in the type of postrepiication repair known as recombination repair. Moreover, only specific recombination genotypes recA, recB, rccK, recF. and one strain of recD were affected by CAA, MMS, and EMS. One strain of recD, recH, as well as the excision repair-deficient mutants (her, uvr. polA) were inhibited slightly (EMS and MMS) or not at all (CAA). All strains were confirmed to possess the recombination deficiencies by transformation and transduction assays (19). These results suggest that CAA, MMS, and EMS may cause similar lesions, all of which require for repair certain components of the recombination repair pathway. In this regard, 4-nitroquinoline-l-oxide (4NQO) reacts identically in this system (Laumbach and Streips, unpublished data). C. Bacillus Forward Mutation Analysis An additional procedure for determining the mutagenic capability of a chem ical compound is to examine its ability to cause forward mutations. This test serves the same purpose as the reversion assay described in Section II. A. Both are direct tests for mutation capability. However, the response is less limited for the forward mutation systems. Cultures of B. subtills 168 wild type (Table 1) were subjected to a 15-mln exposure to CAA, styrene oxide, MMS, and EMS at various concentrations. The cultures were washed and allowed to incubate for 16 hr to express mu tation to streptomycin resistance. The cells (0.1 ml) were plated on tryp- tose blood agar base (TBAB) plates for total viable count. Similar plates were overlaid with soft agar (0.8% agar) containing 1 mg/ml dihydrostreptomycin sulfate to determine the number of mutants. The results from this ' experiment are shown in Table 4. It is evident that CAA, MMS, and EMS I I ' Table 4. Induction of Mutants Following Exposure to Chemical Monomers i Chemical Treatment Total number bacteria*1 Total Mutations mutant per 10 colonies cells Relative mutation frequency**. Chloro| acetaldehyde 0 5 mM 6.0 X108 5.5 x 10 2.4 70.8 0.4 12.9 32.3 |i Styre*ne oxide 0 5 mM 7.0 x 10 6.6 x10 3.1 10.2 0.4 1.5 3.8 Methylmethanesulfonate 0 5 mM 6.0 x10 5.4 X10 2.4 85.8 0.4 15.9 39.8 EthyIraethanesulfonate 0 5 mM 7.0 X108 5.2 Xl08 3.1 68.3 0.4 13.1 32.8 aB. subtilts 168 wild type cells were used in all these experiments. h # mutations per 10 cells no treatm nt Relative mutation frequency = # mutatlona pCr 108 cells with treatment CMA 003799 K,133 U. N. Ktreips, A. L>. I.aumbach, II. Yasbin are strong mutagens in this assay. L'pichlorohydrin, a methylene homologue of chlorooxirane, was also reactive. Styrene oxide showed very weak re activity in initial tests and is being examined further in this system. I D. Bacillus Complest !b, subtilis strain RUB827 (Table 1) was grown to competence using the pro cedure previously described (79), then exposed to DXA which had been previously isolated from strain RUB818 as described by Yasbin (78). After 25 min of incubation, followed by 5 min additional incubation with 100 ug/ml of deoxyribonuclease I, the competent culture was centrifuged (15,000g for 1 min) and resuspended in minimal salts for UV irradiation (76,77) or in GM2 (79) for treatment with chemical agents. The cells were exposed for 30 min to the chemical agents before they were plated on appropriate media. When needed, microsomal activation mixture was added to the competent cells and chemical agents according to the procedures of Ames. McCann, and Yamasaki (2). The total number of viable cells was determined by colony growth on minimal media with glucose, supplemented with all the amino acids {20 Mg/ral) for which the bacteria are auxotrophic (79). The number of transformants were determined by colony growth on the same minimal media which was lacking either tryptophan or methionine. i. Significantly, only 20% of a B. subtilis culture achieves the competent state (81). Therefore, In a "competent" culture of B. subtilis, 80% of the bacteria are noncompetent. The competent and noncompetent members of the culture can be separated by their growth on selective media following transformation. Thus, the number of noncompetent bacteria of strain RUB827 can be determined on minimal media supplemented with both tryptophan and methionine. The competent cells can be calculated from the number f transformants (the number of colonies formed on minimal medium lacking either tryptophan or methionine) (52). The precocious activation of the SOS system in competent B. subtilis makes thes'e bacteria more sensitive to SOS-lnducing factors than non competent bacteria (Refs. 77 and 77a). This enhanced sensitivity is due to the fact that the resident prophages are Induced at lower doses in the competent cells. In Figure 1 the data are presented for an experiment involving the UV Irradiation of a competent culture of RUBS27. The en hanced sensitivity of the competent cells was readily visible. This enhanced sensitivity can be expressed as the percent of transformation obtained after each flucnce of radiation. If the sensitivity of the competent cells and the noncompetent cells to UV were identical, then the percentage of cells trans formed should remain constant, independent of UV irradiation. However, if the competent cells are more UV-sensitivc, then as the fluence increases, CMA 003800 i3s \ A\ A. i). H. A. 'i i are strong mutagens in t;u> assay. V,<I* ',,-Ml.i,.' v ___ _ -I. f. , -.-.--- -V >. ....-.....-.-.-.-.-.-...-.-......... nctivitv >. In it in L U-s1.^ ' / '.^ ' :', ; ipk'iucrolr dri.i, a. ctayleue humotogue D. Bacillus Comptest is. subtilis strain IVJB327 (Table 1) was grown to compcisacc; using the pro cedure previously described (79), then exposed to DXA which had been 'previously isolated from strain RUBS1S as described by Vasbin (78). After 25 min of incubation, followed by 5 min additional incubation with 100 ug/ml I of deoxyribonuclease I, the competent culture was centrifuged (I5,000g for 1 min) and resuspended in minimal salts for UV irradiation (76,77) or in GM2 (79) for treatment with chemical agents. The cells were exposed for 30 min to the chemical agents before they were plated on appropriate media, j When needed, microsomal activation mixture was added to the competent j cells and chemical agents according to the procedures of Ames. McCann, and Yamasaki (2). The total number of viable cells was determined by colony j growth on minimal media with glucose, supplemented with ail the amino acids (20 Mg/ml) for which the bacteria are auxotrophic (79). The number of transformants were determined by colony growth on the same minimal media which was lacking either tryptophan or methionine. i Significantly, only 20% of a B. subtills culture achieves the competent state (81). Therefore, in a "competent" culture of B. subtilis. 80% of the bacteria are noncompetent. The competent and noncompetent members of the culture can be separated by their growth on selective media following transformation. Thus, the number of noncompetent bacteria of strain RUB- 827 can be determined on minimal media supplemented with both tryptophan and methionine. The competent cells can be calculated from the number of transformant! (the number of colonies formed on minimal medium lacking either tryptophan or methionine) (52). The precocious activation of the SOS system In competent B. subtilis makes thes'e bacteria more sensitive to SOS-Inducing factors than non competent bacteria (Refs. 77 and 77a). This enhanced sensitivity Is due to the fact thei the resident prophages are induced at lower doses in the competent cells. In Figure 1 the data are presented for an experlm nt involving the UV irradiation of a competent culture of RUB827. The en hanced sensitivity of the competent cells was readily visible. This enhanced sensitivity can be expressed as the percent of transformation obtained after l each fluence of radiation. If the sensitivity of the competent cells and the noncompetent cells to UV were identical, then the percentage of cells trans formed should remain constant, independent of UV irradiation. However, if the competent cells are more UV-sensitivc, then as the fluence incr ases. CMA 003801 6 / Bacterial Mutation Monitors for Active Metaholites 130 Figure 1. Survival of the total population of strain RUB827 () and of the competent portion of the culture (o) following UV irradiation. the competent cells (as represented by the transformants) should be killed at a greater rate than the noncompetent cells. In Figure 2 the relative transformation efficiency is shown at the various UV fluences. As the fluence increased, the relative transformation efficiency decreased^ The same types of results were obtained when the carcinogenic compound MMS was added to the competent cells (Figure 2). However, the mutagenic, but not carcinogenic, compound, EMS, did not produce a decreasing transfor mation efficiency with increasing dose (Figure 2). Thus, this assay appears to be highly specific for only those components which activate the SOS sys tem. Chloroacetaldehyde, styrene oxide, and several other potential car- . cinogenlc chemicals are being examined tn this system at the present tim . Bacterial strain RUB827 was chosen for this assay because previous results (77) had shown that the poIA5 mutation enhanced the sensitivity of this bacterial strain to SOS-lnducing substances. Compounds are considered potentially carcinogenic when the relative transformation efficiency (the CMA 003802 Figure 2. Relative transformation efficiency (RT) of strain RUB827 follow ing exposure to various concentrations of MMS () and EMS (a) as well as following exposure to various fluences of UV Irradiation (o). The RT is calculated by dividing the percent transformation at a given fluence or con centration by the percent transformation at 0 fluence or concentration. percent transformation of the treated sample divided by the percent trans formation of the untreated sample) is 0.05 or less. This novel B. snbtilts tester system measures the toxicity of agents as well as the ability of these same agents to induce the SOS system. Also, the Cbmptest identifies the concentrations at which the compounds have toxic and SOS-induclng capabilities. The main advantage of this assay is that it apparently distinguishes between substances which are mutagenic and those which activate SOS functions (potential carcinogenic compounds). This property of the Comptest can be noted readily when the relative activities of EMS and MMS are compared. CMA 003803 Figure 2. Relative transformation efficiency (RT) of strain RUB827 follow ing exposure to various concentrations of MMS (a) and EMS (B) as well as following exposure to various fluences of UV irradiation (o). The RT is calculated by dividing the percent transformation at a given flucnce or con centration by the percent transformation at 0 fluence or concentration. percent transformation of the treated sample divided by the percent trans formation of the untreated sample) is 0.05 or less. This novel B. subtllis tester system measures the toxicity of agents as well as the ability of these same agents to induce the SOS system. Also, the Cbmptest Identifies the concentrations at which the compounds have toxic and SOS-induclng capabilities. The main advantage of this assay is that it apparently distinguishes between substances which are mutagenic and those which activate SOS functions (potential carcinogenic compounds). This property of the Comptest can be noted readily when the relative activities of EMS and MMS are compared. CMA 003804 G / Bacterial Mutation Monitors for Active Metabolites 141 III. DISCUSSION In this chapter we have described a batter}- of microbial (particularly B. subtil is) rapid tests for the assay of mutagenicity. The known mutagens EMS and MMS were used as standards in all these tests. In addition, we have also initiated a progressive study of industrial chemicals and their oxidation products using the described test systems. The initial results for chloroacetaldehyde and styrene oxide, the oxidative products of vinyl chloride and styrene, respectively, have been presented. In all of the assays, except the Comptest, both EMS and MMS were highly reactive. It is Interesting to note that EMS was not active In the Comptest, although MMS showed a high level of activity. This finding becomes highly significant when the carcinogenic properties of EMS and MMS are examined. EMS is not carcinogenic, whereas MMS Is a potent carcinogen (17,52). Thus, on the basis of these preliminary results, it seems promising that the Comptest may distinguish between carcinogens and noncarcinogenlc mutagens. We are presently expanding this test to examine a whole battery of chemicals, including chloroacetaldehyde and styrene oxide, to verify this prediction. Since the Comptest specifically examines events associated with SOS induction, it may discriminate for chemicals which induce the error-prone repair systems. Such chemicals could Include all the carcinogens. In this connection, our laboratories have also initiated the interferon induction inhibition (III) test (Ref. 3a). Interferon Is an antiviral agent produced by mammalian cells and whole animals in response to viral infections. Preliminary studies in other laboratories have suggested that Newcastle disease virus (NDV) induced interferon production could be Inhibited by pretreatment of cell cultures with known carcinogens such as benzo&y}pyrene (16). We have confirmed and extended these observations to Include EMS, MMS, styrene oxide, and other known carcinogens. Our study (double-blind) has demonstrated that styrene oxide and MMS are active in this assay, but EMS shows no activity. This mammalian (including human cell) rapid assay may mea sure for SOS-like events in eucaryotlc cells. Among these events could be the Induction of protease, which we postulate could inhibit .Interferon induction in virus and polynucleotide-stimulated cells. Furthermore, several known carcinogens are active in this system, whereas noncarcino genlc analogs are inactive (Refs. 3, 3a, and 16). An overall comparison of the various microbial tests performed In ur laboratories Is presented In Table 5. It is evident that a composite evalua tion of many assays must be made when considering the biological activity of a chemical. This is shown by the variability in reactions of styrene oxide. It is interesting to note that this chemical is essentially nonreactive in the Bacillus tests performed so far. yet It is mutagenic in the Salmonella test and very active in the in assay. If this chemical Is also inactive in the CMA 003805 112 U. N. Streips, A. O. Laumbach, II. E. Yasbin Table !>, Composite Mutagenicity Spectrum of Chemical Monomers Chcm icnls Forward Salmonella Comptest Repair assay mutation Chloroacetaldehyde Styrene oxide Methylmethanesulfonate + + + ND NTD + T + + + Ethylmethanesulfonate + - ++ aResults are expressed as (+) positive in the assay performed, (-) nega tive in the assay performed, and (ND) not determined. Styrene oxide was weakly reactive (Table 4) and must be considered to have borderline activity in the forward mutation assay (t). Comptest assays, then it must be considered that styrene oxide may not enter Bacillus cells. Alternatively, styrene oxide may not elicit the types of repair we are examining in the B. subtilis system. An additional point must be made in regard to the repair assay. Two mutants with recD27 genotype gave dissimilar results when tested with reactive chemicals. Since they both were found to be Rec-, we must assume that one or the other of these mutants does not have the recD27 genotype. Chloroacetaldehyde has been highly reactive in all assays done to date and must be considered as the probable metabolite of vinyl chloride, re sponsible for the biological effects elicited by this industrial chemical. This conclusion has also been made by other laboratories testing vinyl chloride metabolites (4, 20). In conclusion, we have described one rapid assay (Comptest) which may have the capability of detecting carcinogens. In addition, we are refining a rapid mammalian test (111), which also may identify carcinogenic chemicals. These two assays, along with the other tests described in this chapter, pro vide a battery of rapid procedures for screening a large number of chemi cals for biological activity. As such, they are potentially valuable in our overall research programs dealing with environmental pollutants and indus trial monomers. ACKNOWLEDGEMENTS We are grateful to Dr. John L. Wong for valuable discussions and for pro viding the chemicals used in this study. We appreciate the critical appraisal of this contribution by Dr. R. J. Doyle, We wish to thank Camille Jones, Mary L. Kinnaman, Carole Kolb, Rosemarie Michl, and Donna M. Crabb for excellent technical assistance and Drs, C. Iladdcn and B. N. Ames for bacterial strains. This work was supported in part by grants from the CMA 003806 M2 U. N. .Streips, A. . Laumbach, 11. E. Yashin TaMo rlwMMl'-''- _ c.. .M ;;:c::nejrsa Forward mutatic.-. C hi o roac etald ehycl e SO f + Styrene oxide + NTD Methylmethanesulfonate + ++ Ethylmethanesulfonate + - ++ aResults are expressed as (+) positive in the assay performed, (-) nega tive in the assay performed, and (ND) not determined. Styrene oxide was weakly reactive (Table 4) and must be considered to have borderline activity in the forward mutation assay (1). Comptest assays, then it must be considered that styrene oxide may not enter Bacillus cells. Alternatively, styrene oxide may not elicit the types of repair we are examining in the B. subtilis system. An additional point must be made in regard to the repair assay. Two mutants with recD27 genotype gave dissimilar results when tested with reactive chemicals. Since they both were found to be Rec", we must assume that one or the other of these mutants does not have the recD27 genotype. Chloroacetaldehyde has been highly reactive in all assays done to date and must be considered as the probable metabolite of vinyl chloride, re sponsible for the biological effects'elicited by this industrial chemical. This conclusion has also been made by other laboratories testing vinyl chloride metabolites (4,20). In conclusion, we have described one rapid assay (Comptest) which may have the capability of detecting carcinogens. In addition, we are refining a rapid mammalian test (III), which also may identify carcinogenic chemicals. These two assays, along with the other tests described In this chapter, pro vide a battery of rapid procedures for screening a lnrge number of chemi cals for btologtcal activity. As such, they are potentially valuable In our overall research programs dealing with environmental pollutants and indus trial monomers. ACKNOWLEDGEMENTS We are grateful to Dr. John L. Wong for valuable discussions and for pro viding the chemicals used in this study. We appreciate the critical appraisal of this contribution by Dr. R. J. Doyle. We wish to thank Camille Jones, Mary L. Kinnaman, Carole Kolb, Rosemarie Michl, and Donna M. Crabb for excellent technical assistance and Drs. C. Hadden and B. N. Ames for bacterial strains. This work was supported in part by grants from the CMA 003807 G / 13acteri.il Mutation Monitors for Active Metabolites 143 Kentucky Tobacco Health Research Institute, the Manufacturing Chemists Association, and B. F. Goodrich Company to U. N. Streips and the Uni versity of Louisville Cancer Center, and a grant from the Department of Knergy (EP-78-S-02-49G4) to H. E. Yasbin. REFERENCES 1. Ames, B. N., Durstan, W. E., Yamasaki, E. (1973) Proc. Nat. Acad. Sci. U. S. 70:2281. 2. Ames, B. N., McCann, J., Yamasaki, E. (1975) Mutat. Res. 31:347. 3. Barnes, M. C., Sonnenfeld, G. (1979) Abst, Ann. Mtg. ASM, p. 253. 3a. Barnes, M. C., Sonnenfeld, G. (1980) Oncology (In press). 4. Bartsch, H., Malaveille, C., Montesano, R. M. (1975) Int, J. Cane r. 15:429. 5. Bridges, B. A. (19G9) Ann. Rev. Nucl. Set. 19:139. 6. Bridges, B. A., Dennis, R. E., Munson, R. J. (19G7) Genetics 57: 897. 7. Bridges, B. A., Huckle, J. (1970) Mutat. Res. 10:141. 8. Bridges, B. A., Mottershead, R. (1971) Mutat. Res. 13:1. 9. Brooks, K., Clark, A. J. (19G7) J. Virol. 1:283. 10. Carrier, W. L., Setlow, R. B. (1971) In Methods In Enzymology. L. Grossman (ed.), Vol. 21, Part D, p. 230, Academic, New York. 11. Cleaver, J. E. (1974) In Advances In Radiation Biology. J. T. Lett, H. Adler, and M. R. Selle(eds.), Vol. 4, p. 1, Academic, New York. 12. Cooper, P. K., Hanawalt, P. C, (1972) Proc. Nat. Acad. Set, U. S. 69:1156. 13. Creech, J. L., Johnson, M. N. (1974) Environ, Res. 1:387. 14. D'Ambrosio, S. M., Setlow, R. B. (1967) Proc. Nat. Acad. Set, U, S. 73:2396. 15. Defais, M., Fauquet, P., Radman, M., Errera, M. (1971) Virology 43:495. 16. DeMaqyer, E., DeMaeyer - Gulganard, J. (1965) JXCI 34:265. 17. deSerree, F. J. (1976) Mutat. Res. 38:165. 18. Donch, J., Greenberg, J., Green, M. H. L. (1970) Genet. Res. 15: 87. 19. Dubnau, D., Ctrlglano, C. (1974) J. Bactertol. 117:488. 20. Elmore, J. D., Wong, J. L., Laurobach, A. D., Streips, U. N. (1976) Biochlm. Blophvs. Acta 442:405. 21. Emmerson, P. T., West, S. C. (1977) Mol. Gen. Genet. 155:77. 22. Ganesan, A. K. (1974) J. Mol. Biol. 87:103. 23. Ganesan, A. K., Seawill, P. C. (1975) Mol. Gen. Genet. 141:189. 24. Glickman, B. (1974) Blochem. Blophvs. Acta 335:115. 25. Goze, A., Sarasin, A., Moulc, Y., Devoret, R. (1975) Mutat. Res. 28:1. CMA 003808 Ml U. N. Streips, A. D. Laumbnch, R. E. Yashin 26. Caul as, 1.. C. , Mount, D. \V. (1077) I'roc. Nat. Acad. Set. U. S. 74: 5280. 27. Giulas, I.. J., l'ardce, A. B, (1075) Proc. Nat. Acad. Set. U. S. 74: 2300. 2S. Harm, W. , Rupert, C. S. (1903) ?Verenhungsiehre 04:330. 29. Heidelbcrger, C. (1975). Ann, Itev. Biochem. 44:79. 30. Hill, T., Prakash, L., Strauss, B. (1972) J. Bacteriol. 110:47. 31. Ikenaga, M., Ichikawa-Ryo, H,, Hondo, S. (1975) J. Mol. Btol. 92: 341. 32. Iyer, V. N., Ilupp, W. D. (1971) Biochem. Biophys. Acta 228:117. 33. Kada, T., Tutikawa, K., Sadaie, Y. (1972) Mutat. Res. 1G:1C5. 34. Kao, F.-T., Puck, T. T. (1969) J. Cell Physiol. 74:245. 35. Kato, T., Rothman, R. H., Clark, A. J. (1977) Genetics 87:1. 36. Kelly, R. B., Atkinson, M. R., Huberman, J. A., Kornberg, A. (1969) Nature 224:495. 37. Kelner, A. (1949) J. Bacteriol. 58:511. 38. Kimball, R. F., Setlow, J. K., Liu, M. (1971) Mutat. Res. 12:21. 39. Laumbach, A. D., Lee, S., Wong, J., Streips, U. N. (1977) In Proceedings ot the Third International Symposium on Detection and Prevention of Cancer, H. Neiburgs (ed.), Vol. I, p. 155, Marcel Dekker, New York. ) 40. Lee, F. I., Harry, D. S. (1974) Lancet 1:1316. 41. Lytle, C. D., Benane, S. G., Blockstahler, L. E. (1974) Photochem. Photobtoi. 20:91. 42. Malaveille, C. H., Bartsch, H., Montesano, R., Barbin, A., Camus, A. M., Crolzy, A., Jacqulgnon, P. (1975) Btochim. Biophys. Res. Commun. 63:363. 43. Marsden, H. S., Pollard, E. C., GInoza, W., Randall, E. P. (1974) J. Bacteriol. 118:465. 44. Masker, W., Hanawalt, P. C., Shlzuya, H. (1973) Nature, New Biol. 244:242. 45. McCann, J., Simmon, V., Streitwelser, D., Ames, B. N. (1975) Proc.tNat. Acad. Set. U. S. 72:3190. 46. McCann, J., Ames, B. N. (1977) In Origins of Human Cancer, H. H. Hiatt, J. D. Watson, and J. A. Wtnstin (eds.), p. 1431, Cold Spring Harbor Laboratory, N. Y. 47. McEntee, K. (1977) Proc. Nat. Acad, Sci. U. S. 74:5275. 48. Mlura, A., Tomtzawa, J. (1968) Mol, Gen. Genet. 103:1. 49. Moreau, P,, Bailone, A., Devoret, R. (1976) Proc, _Nat. Acad. Set. U. S. 73:3700. 50. Moreau, P., Devoret, R. (1977) In Origins of Human Cancer, H. H. Hiatt, J. D. Watson, and J. A. Winstin (eds.), p. 1451, Cold Spring Harbor Laboratory, N. Y. 51. Nasim, A. (1968) Genetics 59:327. 52. Ncster, E. W. , Schafer, M., Lederberg. J. (1963) Genetics 43:529. 53. Nishioka, H. , Doudney, C. O. (1970) Mutat. Res_, 9:3-19. CMA 003809 Ml U. N. Slreips, A. D. Laumbrich, ft. C. Yashin 32?l. i , O w:. us ( i.. a'*, 1'.i: . - u, ^ i . i - - i >, . k'OC. 1 a *. 'v, k. * - 2c:io. 28. Harm, W. , Rupert, C. S. (19G3) 7. Verenbungslehrc 94:336. 29. Heidelberger, C. (1975). Ann, Hcv. Biochem. 44:79. 30. Hill, T., Prakash, L. , Strauss, B. (1972) J. Bacterial. 110:47. 31. Ikenaga, M., Ichikawa-Ryo, H., Hondo, S. (1975) J. .Mol. Biol. 92: 341. 32. Iyer, V. N., Rupp, W. D. (1971) Biochem. Biophys. Acta 223:117. 33. Kada, T,, Tutikawa, K., Sadale, Y. (1972) Mutat. Res. 1G:1G5. 34. Kao, F.-T., Puck, T. T. (1969) J. Cell Physiol. 74:245. 35. Kato, T., Rothman, R. H., Clark, A. J. (1977) Genetics 37:1. 36. Kelly, R. B., Atkinson, M. R., Huberman, J. A., Kornberg, A. (1969) Nature 224:495. 37. Kelner, A. (1949) J. Bacteriol. 58:511. 38. Kimball, R. F., Setlow, J. K., Liu, M. (1971) Mutat. Res. 12:21. 39. Laumbach, A, D., Lee, S., Wong, J., Strelps, U. N. (1977) In Proceedings of the Third International Symposium on Detection and Prevention of Cancer. H. Neiburgs (ed.), Vol. I, p. 155, Marcel Dekker, New York. 40. Lee, F. I,, Harry. D. S. (1974) Lancet 1:1310. 41. Lytle, C. D., Benane, S. G., Blockstahler, L. E. (1974) Photochem. Photobiol. 20:91. 42. Malaveille, C. H., Bartsch, H., Montesano, R., Barbin, A., Camus, A. M., Croizy, A., Jacquignon, P. (1975) Btochtm. Biophys. Res. Commun. 63; 363. 43. Marsden, H. 8., Pollard, E. C., Glnoza, W., Randall, E. P. (1974) J. Bacteriol. 118:465. 44. Masker, W., Hanawalt, P. C., Shizuya. H. (1973) Nature, New Biol. 244:242. 45. McCann, J., Slnunon, V., Streltweiser, D., Ames, B. N. (1973) Proc.Nat. Acad. Scl. U. S. 72:3190. 46. McCann, J., Amea, B. N. (1977) In Origins of Human Cancer. H. H. Hiatt, J. D. Watson, arai J. A. Winstln (ads.), p. 1431, Cold Spring Harbor Laboratory, N. Y. 47. MoEntee, K. (1977) Proc. Nat. Acad. Set. U. S. 74:5275. 48* Mlura, A., Tomizawa, J. (1968) Mol, Gen, Genet. 103:1. 49. Moreau, P., Bailone, A., Devoret, R. (1976) Proc. Nat. Acad. Scl. U. S. 73:3700. 50. Moreau, F., Devoret, R. (1977) In Origins of Human Cancer, H. H. Hiatt, J. D. Watson, and J. A. Winstin (eds.), p. 1451, Cold Spring Harbor Laboratory, N. Y. 51. Nasim, A. (1968) Genetics 59:327. 52. Ncster, E. W., Schafer, M., Lederberg, J. (1963) Genetics 43:529. 53. Nlshioka, H., Doudney, C. O. (1970) Mutat. Res. 9:349. CMA 003810 (t / Bacterial Mutation Monitors foi Active; Metabolites Mj -l. Ognv. n, II., Shiinada, K. , Tomi/awa, J. I. (1J103) Mol. Con. Genet. J_0J_;227. 53. Okubo, S., noting', W. It. (1955) J. Mol. Hiol. 56. I'ottijolm, A. R. , llaaavvalt, P. C. (1964) J. Mol. Biol. 0:395. 57. l'rukash, I.. (1974) Genetics 78:1101. 53. Had man, M. (1975) la Molecular Mechanisms for Repair of DXA. P. Hunawalt and R. B. Setlow (cds.). Part A, p. 355, Plenum, New Y rk. 59. Raunug, U., Johansson, A., Hamel, C., Wachtmeister, C. A. (1974) AM BIO 3:194. 60. Regan, J. D., Setlow, It. B. (1974) Cancer Res. 34:3318. 61. Rupert, C. S. (1975) In Molecular Mechanisms for Repair of DNA, P. Hanawalt and R. B. Setlow (eds.). Part A, p. 73, Plenum, New York. 62. Rupp, \Y. D., Howard-Flanders, P. (1968) J, Mol. Biol. 31:291. 63. Rupp, W. D., Wilde, C. E., III, Reno, D. L., Howard-Flanders, P. (1971) J. Mol. Biol. 61:25. 64. Sedgwick, S. G. (1975) Proc. Nat. Acad. Sci. U. S. 72:2753, 65. Sedgwick, S. G. (1975) J, Bacteriol. 123:154. 66. Setlow, J. K. (1966) Radtat, Res, (suppl.) 6:141. 67. Setlow, J. K., Boling, M. (1972) J. Mol. Biol, 63:349. 68. Smith, K. C., Meun, D. II. C. (1970) J. Mol. Btol. 51:459. 69. Swenson, P. A. (1976) In Photochemical and Photobiologtcal Reviews. K, C. Smith (ed.), Vol. 1, p. 269, Plenum, New York. 70. Viola, P. L., Bigottt, A., Caputo, A. (1971) Cancer Res. 31:516. 71. Wheeler, L., Halula, M., Demeo, M. (1979) Abst. Ann. Mtg, ASM. p. 119. 72. Weigle, J. J. (1953) Proc. Nat. Acad. Sci. (Wash.) 39:628. 73. Within, E. M. (1967) Brookhaven Symp, Biol. 20:17. 74. Witkin, E. M. (1969) Mutat. Res. 8:9. 75. Witkin, E. M. (1976) Bacteriol. Rev. 40:869. 76. Yasbin, R. E., Wilson, G. A., Young, F. E. (1975) J. Bacteriol. 121:296. 77. Yasbin, R. E., Wilson, G. A., Young, F. E. (1975) J, Bacteriol. 121:305. 77a. Yasbin, R. E., Wilson, G. A. Young, F. E. (1980) Appl. Environ. Microbiol. (In press). 78. Yasbin, R. E. (1977) Mol. Gen. Genet. 153:211. 79. Yasbin, R. E. (1977) Mol. Gen. Genet. 153:219. 80. Yasbin, R. E., Fernwalt, J. D., Fields, P. I. (1979) J. Bactertol. 137:391. 81. Young, F. E. (1967) Nature 213:773. 82. Youngs, D. A., Smith, K. C. (1973) J. Bacteriol. 116:175. 83. Youngs, D. A., Smith, K. C. (1973) Nature. New Biol. 244:240. 84. Youngs, D. A., Smith, K. C. (1976) J. Bacteriol. 125:102. 85. Youngs, D. A., van dor Schwren, E., Smith, K. C. (1974) J. Bacteriol. 117:717. 86. Zhcvner, V. D., Shestakov, S. V. (1972) Arch, Mikrohiql. 56:349. CMA 003811 Journal or Bacteriology, June 1979, p 915-922 0021-9193/79/06-0915/09302,00/0 Vol. 138, No 3 Selective Association of the Chromosome with Membrane in a Stable L-Form of Bacillus subtilis SARAH HOROWITZ,1 RONALD J. DOYLE,1 FRANK E. YOUNG,2 and ULD1S N. STREIPS'' Department of Microbiology and Immunology, Schools of Medicine and Dentistry, University of Louisville, Louisville, Kentucky 40232, and Department of Microbiology, Schools of Medicine and Dentistry, University of Rochester, Rochester, Neui York 146421 Received for publication 11 January 1979 A stable L-furm (Sal-1) of Bacillus subtilis was found to have retained a markedly modified chromosome-membrane association when compared to intact cells. The membrane-deoxyribonucleic acid complex of the L-form was similar to that of its parental strain in quantity and stability. Genetic analysis of the L-form membrane-deoxyribonucleic acid complex revealed enrichment for markers close to the replication origin, but not for internal markers, indicating preferential attachment of the origin of chromosomal replication to the membrane. These results are in close agreement with those found for the parental bacterial form. In contrast, the replication terminus region was not preferentially attached to the membrane, of the L-form, even though it is enriched in the bacterial form. The association of the chromosome with the membrane at the replication terminus does not appear to be necessary for cell growth and separation, but because the L-form divides aberrantly, it may be one of the factors required for normal deoxyribonucleic acid segregation and septation. Interaction between the chromosome and the cell surface has been well documented in pro caryotes. A stable attachment of the chromo some to the cytoplasmic membrane has been demonstrated in gram-negative as well as gram positive bacteria (IS, 22). The DNA in the membrane-DNA complex was found to be specifically enriched for the replication point and for genetic markers close to the origin and terminus of chromosomal replication (10. 12, 19, 24, 27, 28, 34, .35). This attachment was proposed to aid in the segregation of the newly replicated chro mosomes and in cell division (14). Further work with Escherichia coli (9). and in our laboratories with Bacillus subtilis (3, 26), has demonstrated a stable attachment of the chromosome to the (rigid) cell wall as well as to the cytoplasmic membrane. The DNA attached to the cell wall in B. subtilis was found to be enriched for ge netic loci near the origin and terminus of chro mosomal replication (26). The chromosomemembrane-cell wall complex has been proposed to be held together by specific binding proteins (5, 9, 11. 29, 30). This complex may create an apparatus functional in cell division processes. Research with procaryotes lacking cell wall has demonstrated that the genome of Myco plasma galtisepticum is also associated with the membrane, showing specific enrichment for the replication point and possibly the origin of chro mosomal replication (20, 21). Klectron micro graphs have revealed the presence of chromo some-membrane association in protoplasts and unstable L-forms of B. subtilis (23). However, to date, genetic characterization of the chromo some-membrane complex in procaryotes with out cell walls has been lacking. In this report, we describe studies on the DNA-membrane association in a stable L-form of R subtilis. Sal-1 (30). This organism requires 1.2 M NaCl for stabilizstion, does not have cell wall polymers, and does not revert to the bac terial form (6). Moreover, this L-form appears to have lost cell division control and undergoes aberrant cell diviaion (R, W. Gilpin, personal communication), as do other stable L-forms (7, 33). Therefore, the L-form becomes a good model to test our hypothesis that loss of control in cell division may be due to the loss of one or more of the components in the cell division complex, such as the cell wall and proteins as sociated with the peptidoglycan. The present study offers the first evidence that a stable Lform retains attachment of the chromosome to the membrane. Furthermore, this attachment is specific only for the origin of chromosomal rep lication. The region for termination of genome replication appears to have no specific affinity for the surface of this L-form. (This work i a part of the dissertation to he submitted by S. Horowitz to the Graduate Fac ulty of the University of Louisville in partial 915 CMA 003812 916 HOROWITZ KT Al, J Bactf.riol. fulfillment of the requirements for the PhD. decree.) MATKKIM.S \M> VH-THIHiw Bncteriul strum*. 'I he L ^uhfilis -ii.un,- u.-oh tn i hi, -tudy are lr-inl m Tahir I Wt thunk H Yi.-hik. iu.i for strain TLS13? N Sueok.i for strain Mnhu iul6. J, Copeland for strain BCiN Harford for strain* BD170. BD'20'2, VDB15. and VUB41, S Phillips for strain QB9J2. ami ). Kane for strain BDH2, All strains were maintained on trsptose hloud agar base tTBABl (Difco Laboratories, Detroit. Mich j '1 tie stable L-form, Sal-1, kindlv provided by R. W. Gilpin. is a derivative of B, subtihB R151 (Jti) and retains the metBIO marker (II. The L-form was main tained by passage in stabilizing liquid medium (below). The parental strain, B. unbuilt BR151 (vs ? trpC2 metBIO. was transformed to protntrnphv for (vs-.? and trpC2 with 8 subtilts wild-tvpe 168 DNA to maintain the same auxotrophic background as the L-form, and was designated as strain HUMtkt. Media.and growth conditions. The L-form, Soll, was grown m IS ml of Sal-l medium, a modified glucose minimal medium Hcscnbed by Spizizen (251 with 0.016T (wt/voli Mg<.T-6H:0 tinstead of 00`2'i MgSOi-7HiO-1.2 M Nad) for stabilization, oO.Vr ca sein hydrolysate, and 50 pg of (.-methiomne per ml. The culture was shaken in a rotary water bath (New Brunswick Scientific Co., New Brunswick. N.J.) at 37'C and 150 rpm. B. subtilts BUL404 was propagated in Sal-1 medium without NaCl, under the same growth conditions. Both cultures were grown for two or more generations in the exponential phase in the presence of deoxyadenosine (200 pg/ml; Sigma Chemical Co., St. Louis, Mo.) and (2-"Cjth>midine (0.1 /iCi/ml, 51 mCi/mmol; Schwarz/Mann, Orangeburg. N.Y.). Cell growth was monitored on a Klett-Summrrson colorimeter with filter no. 34. Lysis. The L-forma were harvested by centrifuga tion at 5,000 rpm for 5 min and then washed twice in cold NCI' buffer it) 01 M sodium citrate. 0,od M K lli'O.-IH-O. and 0.04 M KH.I'Oi. admitted to pH 7 4 | l?|i whirh contained I `2 M N.iCI The cells were then ni-penhed in a -mall mluinc nl NCI1 pin- 1 M NaO bulier ,lntl tw-ll by osmo'.u -hot s uue to :n adtution of NCI' Poller without NaCl. 4 he resulting lysate was a four to eightfold concentration t1 original culture, and contained a final concentration of 0.1 M NaCl, The 8. tubtilts BUI-404 cells were har vested and w ashed as Sal-1. but in NCI' buffer without NaCl, then concentrated four- to eightfold in NCI', and Ivsed by the addition of Ivsozvmo (0.5 mg/ml; Worthington Biochemical Corp., Freehold, N.J.) for 5 min at 37C. The lysate was then adjusted to tu M NaCl. Both lysates were either left untreated, or weio sheared for .10 a on a Vortex mixer (Fisher Scientific Co.. Springfield, Mass) set at position no. 6. Renografin gradient*. Linear (0 to 38%) Rcntvgrafin density gradients (RanograAn-76, from Scubb and Sons, Inc., Princeton, NJ., in NCP plus 0.1 M NaCl buffer) were prepared by the method of Ivarie and Pent (12). Sample* (1,0 to 1.5 ml each) of the Lforro or the parental attain lyaataa were layered onto .30-ml cold Renografin gradients, and th* gradient* were centrifuged in a Beckman L2S6B ultracentrtfug* at 25,000 rpm, 2*0 for 5.5 h, using an SW2S.1 rotor. After centrifugation, 1.2-ml fractions war* collected into sterile tube*. Samples of these fractions (0.23 to 0.50 ml each) wn counted in Multiaol scintillation fluid (Isolab Inc., Akron, Ohio) in a liquid scintillation countar. The remaining volume of aach of th* peak fraction* was maintained at 4*C for transformation experiments. Transformation experiments. Th* procedure* for the development of competence and transforma tion ware those of Boylan et al. (2). Auxotrophic strains ot B. subtili> (Table 1} were grown in GMl (2) at 37C in a New Brunswick gyratory incubator shaker at 250 rpnp Cell growth wa* monitored on a KletiSummcraon colorimeter (filter no. 54) until the ce*aanon of logarithmic growth. After 90 min of further Tahle 1. List of stroma used DaaignatMn Genotype Derivation or nirc* Sal-1 (stable L-form) BR7 BR151 RUB783 BUL201 BUL320 BUL401 BUL404 BUL406 BUL5Q2 BUL705 BUL700 BUL714 BUL7I7 BLTL720 BUL726 BUL728 metBIO leuAB lrpC2 trpC2 lye-3 metBIO leuAB metBIO hisAI purBS leuAB purAtS hitAl leuAB metBIO thr-5 hisAI leuAB gltA292 metBIO metBIO leuAB thyA thyB leuAB metBIO ly* 3 leuAB metBIO hisAI pyrAI leuAB metB10 cyaA 14 htsA t leuAB purA16 metC3 leuAB metBIO trpC2purBS leuAB eitKS leuAB trpC2 lys-3 R. W. Gilpin B. E. Reilly B. E. Reilly Thie laboratory This laboratory BUL714 x BD170 thr-5 trpC2 QB922-g(tA292 trpC2 X TL8I32-W8 thyA thyB U8132 BR151 x W168* BR151 x Sal-1" 168TT-tipC2 thyA thyB x MuAuSuiA-purA IS leuAB metBS BR15I x RUB783* RUB783 x VCJB4S-Ai*A/ argC4pyrAI` RUB783 X BD82-eysA/-T BUI-20: x WTUl-trpC2 meter RUB783 x BD202-^7>rr RC 122-.`mC2 ci'ffS x MuguSulS* BK151 xTI-8lo2` * Obtained by congresaion (17); general notation; recipient celt * honor DNA. CMA 0038X3 Vol. 138,1979 L-FORM MEMBRANF.-DNA COMPLEX 917 incubation, the cells were diluted tenfold into GMII (2) and incubated an additional 60 min, at which tune the cultures were maximallv competent Peak frac tious from the Kcnugrufin (tmiinmts served donor DMA. These fractions were diluted in Spizizen s min imal salts solution to avoid saturation and to reduce the concentration of the Renografin (8). The compe tent cells were incubated with the DNA for 30 min at 37*C in the incubator shaker Samples 10 1 ml each) were plated in triplicate on selective media and incu bated at 37*C. Transformants were enumerated after 48 h of incubation. Each genetic marker was analysed for enrichment from 3 to 16 times to attain statiatical validity, tn general, selection for transformants was done on glucose-minimal agar 125) supplemented with the appropriate amino acids and bases. Selection for the citKS transformants was on minimal agar plates which contained 0.19- sodium lactate (J. T. Baker Chemical Co., Phillipsburg, N.-J.) instead of glucose. Selection for thyA transformants was by incubation at 46C on plates without thymine (18). The thyB trans formants were sensitive to growth at 46*C without thymina and were evaluated by subtracting the num ber of transformants ithy`) at 46'C from those at 37*C (18). Selection for gltA292 transformants was on glu cose-minimal agar in the absence of 400 *ig of glutamic acid per ml (4). RESULTS Distribution of labeled DNA on Reno grafin gradients. To determine the presence of a membrane-DNA (m-DNA) complex in the L-form. a culture of Sal-1 was grown in Sal-1 medium, and its DNA was uniformly labeled with [2-'`C(thymidine. The doubling time of the L-forms, as determined by Klett absorbance units, was approximately 2 h under these growth conditions. The L-forms were washed and then lysed by osmotic shock. For comparison, the parental strain BUL404 was also grown and labeled under similar conditions, having a dou bling time of 40 min. The parental strain was also washed and then lysed by exposure to ly sozyme (0.5 mg/ml) in hypotonic solution. A sample (1 ml) of each lysate was immediately layered onto a 0 to 38% linear Renografin density gradient. The remainder of each lysate was sheared by blending in a Vortex mixer for 30 s, and then 1.0- to 1.5-ml samples were applied onto Renografin gradients. The gradients were centrifuged and collected, and samples were as sayed for radioactivity (Fig. L). In the absence of shearing, the L-form DNA formed a distinct peak at the lower part of the gradient (Fig. 1A). The peak corresponds to the membrane-at tached DNA (m-DNA) peak that was found in the unsheared lysate of the parental strain BUL404 (Fig. la). When sheared, part of the L- form DNA formed a slower-sedimenting peak (Fig. IB) which corresponda to the free DNA (f- Fto. 1. Distribution of the lab*ltd DNA from the L-form. Sai l, on Renografin gradients. The L-forms ui*r* grown, labeled, washed, and tyeed as described. Unsheared (A) or sheared (B) lysates were sedimented on 0 to 38% Renografin gradients. Fractions were collected and counted for recovered radioactivity. In all samples, 90 to 100% of the input radioactivity was recovered For comparison, the parental strain. B. subtihs BIT404 was grown, labeled, and washed under similar conditions, lysed by exposure to lysotyme (0.5 mg/ ml) in hypotonic solution, and the unsheared (a) or sheared (b) lysates were sedimented on Renografin gradients as abovr Shcored lysates demonstrate membraneattaehed (m-DKA) and free (f-DNA) nucleic acid Vou US. 1979 L. 7zny-', mmmskanf-dna comply x t? -ubation, the Jeiis r; ,,-u ' . > .J' and incubated an additional **'> -i - i *r " - ` he luiturcj aero m.iximallv .'i.-wiii:! I'-ik t.il i'r'in ihc itcnugriiiln (ic t, * ; - s.twi i. i; nor DNA. Thu?* fractions were diluted in Spi/ucn 3 rum- imsl salts solution to avoid saturation and to reduce the concentration of the Kenografm (8) The compe tent ceil# were incubated with the DNA for JO inin at 37*C in the incubator shaker Samples (0 1 ml each) were plated in triplicate on seleitr. e media and .titu bated at 37*C. Transformants were enumerated after 48 h of incubation. Each genetic marker waa analyzed for enrichment from 3 to IS times to sttain statistical validity. In general, selection for transformants was done on glucose-minimal agar 125> supplemented with the appropriate amino acids and bases. Selection for the cirAfj transformants was on minimal agar plates which contained 0 IT sodium lactate lJ. T Baker Chemical Co. Phillipsburg. N 1 ) tpstead of glucose Selection for thyA transformants was by incubation at 46*C on platea without thymine < 18). The thyB trans formants were sensitive to growth at 4b'C without thymine and were evaluated by subjecting the num ber of transformants (thy *) at 4fi"C from thus* at 37'C 118). Selection for gltA292 transformants was on glu cose-minimal agar in the ahsence of 400 ug of glutamic acid per ml (4). RESULTS Distribution of labeled DNA on Renogrmfln gradients. To determine the presence - embranu-DN.A 'mA) rort.pl.". [, f-.ru a iibuie of S 1 i -1 was grown S,.| I -y* 111 ''*> MV A *; ** ! ' '**'!' ' w ltd L J- `C ] h u mu lint- I he douiiung tunc of iln- [.-forms, as determined by Klett absorbance units, was approximately 2 h under these growth conditions. The L-forms were washed and then lysed by osmotic shock. For comparison, the parental strain BUL404 was also grown yn-J labeled under similar conditions, having a dou bling time of 40 min. The parental strain was also washed and then lysed by exposure to ly sozyme (0.5 mg/ml) in hypotonic solution. A sample (1 ml) of each lysate was immediately layered onto a 0 to 38T linear Kenografin density gradient. The remainder of each lysate was sheared by blending in a Vortex mixer for 30 s, and then 1.0- to l,5*ml samples were applied onto Renografin gradients. The gradients were centrifuged and collected, and samples were as sayed for radioactivity (Fig. 1). In the absence of shearing, the L-form DNA formed a distinct peak at the lower part of the gradient (Fig. 1A>. The peak corresponds to the memhrane-attached DNA (m-DNA) peak that was found in the unsheared lysate of the parental strain BUL404 (Fig. la). When sheared, part of the L- form DNA formed e slower-sedimenting peak (Fig. IB) which corresponds to the free DNA (f- Fm. 1. Dimnbudon of the labeled DNA from the l-form. Soi l, on Rtnogrofin gradient* The L-forme toon groan, labeled, leoahed, and lyeed as deaenbed. Unehaond (A) or sheared (S) lymstee wart tedimented on 0 to 38% Rtnogrofin gradient*. Fractions man collected and counted for recovered radioactivity. In all tampion. 30 to 108% of the input radioactivity wan recovered. For comparison, the parental strain. 8. subtili* BIT.404, mat groom, labeled, and washed under tuntlar conditions, lysed by exposure to lymityme (0.S mg/ ml) in hypotonic solution, and the uneheand (a) or sheared (b) lysates men sedimented on Rtnogrofin gradients as above. Sheared lysates demonstrate membrane-attached (m-DNA) and fret (f DNA) nucleic acid CMA 003S15 918 HOROWITZ ET Al. J Bactf.kioi DNA) peak of the paienul atiain (Fig. lb). However, a significant propoiuon of the labeled DNA of the L-fumi was retained in the m-DNA peak as firmK attached DNA : !'iy l It) a-, in the ft;1: ental h 'rti"* ' h * 'T i, -- > rn vf UNA distribution I'm the L-lonn, as weli a.' the parental hactena! fnim, in unsheaied and sheared lysates is in close agreement with that described by Ivarie and Pene for B. subtilis 168T", using Renografin gradients (12). To com pare the distribution of the L-form labeled DNA in the Renografin gradients to that of its parental strain, the results of several experiments were pooled, and the average percentage of DNA in each peak ( the standard error of the mean) was calculated. Approximately 80% of the la beled DNA in the unsheared L-form lysate is attached to the membrane (m-DNA) (Table 2). Upon shearing for 30 s, approximately 30% of the DNA remained firmly attached to the mem brane (m-DNA), whereas about 55% of the DNA appears as free DNA (f-DNA). These results are statistically not different from those found in the unsheared and sheared lysates of the parental strain BUL404 as determined by an independent t test (Table 2). This observation suggests that the L-form has retained an m-DNA complex physically similar to that of the bacterial form. To compare the stability of the m-DNA com plex of the L-form to that of its parental strain, both lysates were subjected to different shearing times, and the resulting DNA distribution on Renografin gradients was observed. There was no significant decrease in the percentage of m- DNA for both the L-form and the parental strain when sheared for times ranging from 15 s up to 60 s (S. Horowitz, Ph.D. thesis. University of Louisville, Louisville, Ky., 1979), We, therefore, routinely used a shearing time of 30 s for the experiments described below. Genetic analysis of membrane-attached DNA isolated from the L-form and its pa rental strain. The DNA bound to the mem brane was used in transformation assays to de termine the genetic composition of the m-DNA Taslx 2. Effect of shearing on m-DNA complete from the L-form, Sal-1, and Us parental strain, B. subtilis BUUOT Stain Sei-t BUL404 Treatment of lysate No sheerinf Sheared* Noshaarin# Shaarad* m-DNA IS) 8133 a 2.78 31.93 1.01 33.00 1.28 34.08 X 2.88 f-DNA <%) 12.88 * 2.77 58.09 1.08 12.41 0.83 501 2.08 " Tb pncshina for tin Mpsriuon of m-DNA and f-DNA ind th* assays for radioactivity ar* ss daambad in tht ltfand to Fif. 1. Th* lyiatw wet* sheared for 30 on a Vortat miiar set at poaition S complex from the L-form and compare it to the piofile found in membrane-associated DNA fiom the parental strain BLT.404. Different i hioiRo.ioinal markers, chosen to provide an an tl\sis ot tht entire chinininump, aero exam ined. and (he membrane enrichment indexes (MEl) ere calculated by the method of Sueuka and Quinn (27). The marker leuAS was used as a standard in these calculations. The average ME1 values for each marker tested ( the stan dard error of the mean) are presented in Table 3. For statistical analysis, a two-factor analysis of variance with repeated measures on one factor was run. The between factor was the 13 genes we had examined. The within factor was the parental (BUL404) versus L-form (Sal-1) con dition. The Duncan range test was used for post hoc comparisons to find markers among which a significant difference occurred (31). The ge netic and statistical analyses revealed that the DNA attached to the membrane in the L-form was enriched for the markers purAlS and cysA14, which are close to the origin of chro mosomal replication (16, 37), but not for tha internal genes hisAl. purB6, thr-S, leuAS (stan dard), metC3, pyrAl, lys-3, and thyA (Table 3). These results showed preferential attachment of the origin of chromosomal replication to tha membrane in the L-form. In this regard, the In form m-DNA complex was not statistically dif- Tasi.f 3. Genetic analysis of m -DNA notated from the L-form. Sal l. and its parental strain, B. subtilis Bl'f.404 (j*n*uc muiker examined" purAW cysAN hisAl purBS thr-5 leuAS metC3 pyrAl lys-3 thyA IrpCS gltA2S2 thyB citKS Sai l MEI* 1.74 0.11 1.78 0.17 0.98 0.13 1.10 0.12 1.17 0.11 1.00 0.81 010 1.06 0.10 1.17 0.19 1.07 0.02 0.99 0.05' 1.16 0.09' 0.82 0.09 0.95 0.07 BLL404 MEI' 1 55 0.13 1.57 0.23 0.90 + 0.12 1.06 0.14 1.11 0.13 1.00 0.92 0.U 1.05 0.14 1.10 0.21 1.40 0.17 157 0.11 1 80 0.18 1.90 0.48 2 20 0.24 " Recipient strain* used in these experiment* were BUL201, 320, 401. 502, 706, 709. 7t4.717. 720, 726, 728, RUB783. end BR7. * Calculated by the method of Sueoke end Quinn (27). MEI - (U/leuAS) in m-DNAl/f(x//euA8) in fDNA1, where x is the number of transformants for any marker tested, end leuAS i* the number of transform ant* for the standard marker leuAS. ' P < 0.01 when compared to RUL404. by analysis of variance. CMA 003816 VOL. l.'W, 1979 L-FORM MEMBRANE-DNA COMPLEX 919 ferent from the m-DNA complex of the parental strain BUL404 (Table 3). In contrast, the repli cation terminus region of the L-form was not found to be preferentially attached to the mem brane. The terminus markers trpC2, gltA292, thyB, and citK5 were enriched in the m-DNA complex of the bacterial form, BUL4Q4, but were not enriched in the L-form (Table 3). The MEI of the genetic markers tested in the m-DNA complex of the L-form and its parental strain are presented in relation to their relative posi tions in the bidirectional replication map of B. subtilis in Fig. 2. As a control, incubation of the parental lysate in the preaence of 1.2 M NaCi did not bring about losa of the specific membrane attachment of genetic markers close to either the replication origin or the terminus. Genetic analysis of m-DNA isolated from the hybrid BUL405. Since the L-form ap peared to have lost the specific attachment of the terminu.-, to the membrane, it wus unpin lant to evaminv whether a chionKi-oiiiiul (raj,'tient from the terminus region of the L-form would be able to regain the specific attachment when placed in the parental, bacterial form. Therefore we constructed a hybrid, BL L405. which is a derivative of the parental strain BR151. The hybrid contains the internal marker lys-3 and the terminus marker trpC2 from the L-form (Table 1). Strain BUL405 was grown, labeled, washed, lysed, and sheared as described above for the parental strain BUL404, Both m-DNA and f-DNA were isolated on Renografin gra dients, and genetic analysis was performed. The internal marker, lys-3, of the hybrid, BUL405, was not enriched in the membrane and had an MEI value not statistically different from that of the parental strain. BUL404, and the L-form, Sal-1 (Table 4). The terminus marker, trpC2, was enriched in the membrane of the hybrid, BUL405, aa it waa in the parental strain. This enrichment, indicated by the MEI values, was significantly different from that found for the Lform trpC2 gene which was not enriched in the membrane. DISCUSSION This study demonstrates that Sai-1. a stable L-form of Bacillus subtilis BR1S1, has retained an attachment of the chromosome to the mem brane. In Renografin gradients, this attachment was similar in sedimentation, quantity, and la bility to that of the parental bacterial form in both sheared and unsheared lysates. Further more. these results were in close agreement with the- attachment found in another strain of It. submit (12). Our findings represent the first description of sn m-DNA complex in a stabie L* form, and it appears that the permanent loss of cell wall polymers did not prevent chromosomemembrane association. Such an attachment may be a universal feature of all procaryotes. Several laboratories have described both specific and Fig. 2. MEI for tht genetic markers examined from m-DNA of tht L-form, Sal-1, and its parental strain, B. subtitle BUL404, correlated to the map poetiion of them markers tn B subtilis 168. The map positions for the different gents were calculated from the bidirectional replication map of B. subtilis 168 of Leptsant Kejzlarova et aL (Kg and Young and Wil son (37), The MEI values ( standard error of the mean) are taken from Table 3. (A) m-DNA from tht L-form, Sal-1; (B) m-DNA from the parental strain, B VL404. Tht dotted tine represents the areas of expected membrane enrichment of genetic mailers, based on our results and those in the literature (10, 35). Tabus 4. Genetic analysis of m-DNA isolated from B. subtitle 8U1406 Genetic mariar tanuaad* lys-3 trpd BUL40S MEI 1.11 0.00 1.44 % 0.15' BUL40* MEI* 1.10*0.21 1.570.U' Sal-I MEI* 1.17 0.19 0.90 0.05 " Recipient strains used in these experiments were BUL706, BUL728, and BR7. * The MEI value* were taken from Tsbl* 3 far rompsnson. 1 p < 0.01 when compered to the MKI of the trpC2 marker in Sal-1, using an independent l test. \ 01.. : '"!{ 'N r. 2 12 = 'n'i BL'L404 Tab1 - - v- : i-aum ro us pref-rs".;..jtiuct.ec ;w '!= " > ornr.e The terminus markers trpC2, gUAL`92. thy-3. and citKo were enriched in the m-DNA complex of the bacterial form, 8UL404. but were not enriched in the L-form (Table .'ll The MEI if the jer.eti: markers tested tn the m-DNA complex of the L-foim end its parental strain are presented in relation to their relative posi tions in the bidirectional replication map of B, mbtili in Fig. 2. As a control, incubation of the parental lysate in the presence of 1.2 M NaCl did not bnng about loss of the specific membrane attachment of genetic markers close to either the replication origin or the terminus. Genetic analysis of m-DNA isolated from il J........ !( .r T * r 3 .`HI 11 1* * l: -U- io an a* 04f o on < * oiiriot( :11; 1i; jiII: r - J 14 f" U LJ (*, Lj- VJ ') l > - '1 : `o h ; 1 iv > r [[ _ , c.'.. L- <- -v tmiiu .d t hr>fit. 'W m ^'t:., i um tne terminus legion of me L mrm wmid I'f able to regain the specific attachment when placed in the parental, bacterial form. Therefore e constructed a hybrid, BL L4Q5. which -s n ae.v.ati.e of the parenal -iram BRIM Toe hvbrid contains the internal marker /vs ,? and the terminus marker trpC2 from 'he L-form -Table 1), Strain BLL4G5 was grown, labeled, wished, lysed, and sheared as described move for the parental strain BUL4Q4. Both m-I)N,\ and f-DNA were isolated on Kenografm gra dients, and genetic analysis was performed 1 he imernal marker, 'ye-?, of the hybrid. BIT -o')1, was not enriched in the membrane and had an MEI value not statistically different from that of tne parental strain, BLL4G4. and the L-form. Sal-1 (Table 4). The terminus marker. trpC2, was enriched in the membrane of the hybrid. BUL405, as it was in the parental strain. This enrichment, indicated by the MEI values, was significantly different from thst found for the L- form trpC2 gene which was not enriched in the membrane. DISCUSSION This study demonstrates that Sal-1, a stable L-form of Bacillus .eubttlis BR151, has retained an attachment of the chromosome to the mem brane. In Renografin gradients, this attachment was similar in sedimentation, quantity, anti sta bility to that of the parental bacterial form in both sheared and unsheared tvxates. Further more. these results were in close agreement with the attachment found in another strain of />. xubtUin (12). Our findings represent the first description of an m-DNA complex in a stable Lform, and it appears that the permanent kiss of cell wall polymers did not prevent chromosomemembrane association. Such an attachment may be a universal feature of all procaryotes. Several laboratories have described both specific and FlO. 2. MEI far (** genetic markers nomine^ than m-DNA of dm L-form. Sal-1, and its parenta! drain. B. suhtilis BUL404. correlated to tht map pomlion ofdmtt markon in B. subtihs 168, Tht map potititmt far dm difftrtnl fna won ealeulattd from dm bedirwetional replication map of B. subtihs 168 of Ltpttanl Kejsiarova tt aL (IQ and Young and Wil son (37). T7m MSI valors (% standard error of dm moan) art toMtn from Tablt 1 (A) m-DNA from dm L-form, Sai l; (B) m-DNA from dm partntal strain. BUL404. Tht datttd lint rtprtstnts (hr areas of rxpectfd membrane enrichment of gtnette markers, dosed on our remits and thorn in tht literature (10, 35). Tails 4. Genelu.- analysis of m-DNA isolated from B. subtitle BVL406 matter auL4tMH 8UL404 MET SeM MET nifflinaT lyt-3 trpCS l.U a 0.08 U0 0J1 1.17 0.19 i.44 % air 1J70.I1' 0.90 0.06 ' Recipient strains uaed in iheae experiment* were BUL706, BUL728, and BR7. 'The MEI values were taken from Table 3 for romiMinKHi. ' P < 0.01 when ramparad to the MKI of the trpC2 marker in SeM. using an independent t teat. CMA 003818 920 HOROWITZ KT AL. J Bactfhiol. nonspecific attachment of the chromosome to the membrane (10. 12, 13, 24, 27. 28, 34, 35) To determine whether the m-DNA complex from the L-fuim contained 'pecific regions of the chromosome, we used samples of peak fractions from the Renografin gradients in transformation assa>s and calculated memhrane enrichment in dexes (MEI) (Table 3). Such genetic analysis revealed that the origin of chromosomal repli cation was preferentially attached to the mem brane in the L-form, as in the parental bacteria] form, tn contrast, the specificity for the attach ment of the terminus region had been lost in the L-form. This finding indicates that the two sites of attachment may be chemically and physically independent and may involve different binding proteins, as has been suggested previously (29, 34). The persistent binding of the origin region also suggests that this specific attachment to the membrane is of critical importance. The propor tion of m-DNA (32%) found in the L-form, de spite the loss of terminus attachment, could be explained by more extensive association of the origin, when compared to the parental form. Alternatively, radioactive counts due to nonspe cific attachments (13) may mask the loss of radioactivity from terminus detachment. Since the L-form, Sal-1, is known to divide aberrantly (R. W. Gilpin, personal communica tion), the loss of specific attachment of the ter minus region to the membrane may be a con tributing factor to the loss of division control in this organism. The integrity of the DNA-membrane-protein-cell wall complex was proposed to be crucial in maintaining normal cell division (26). Moreover, dynamic terminus attachment to the membrane in correlation with the cell cycle has been proposed to play an important rote in cell division control in B. subtihs (32). All these (acton lead to the speculation that loss of terminus attachment may be critical in the loss of normal DNA segregation and ceil division processes. Several possible factors could have been re sponsible for the loss of specific terminus attach ment to the membrane in the L-form. The most prohable are: a change in the affinity of the DNA in the replication terminus region of the chromosome for the membrane or for specific binding proteins; the permanent loss of cell wait along with a disappearance of an outer surface site for terminus attachment (26). a loss or al teration of binding proteins which may aid in the specific attachment of the terminus (o the membrane or cell wall (5. 9, 11, 29, 30); and, finally, the presence of a high concentration ot sail (I 2 M NaCI) over many generations in the growth medium of the L-form, which might af fect the terminus-membrane attachment, but not the origin-membrane association. Results from genetic analysis of the h>bnd BL'l4U5 I Table 4) appeared to negate the first possibility and indicated that no change in the affinity of the L-form terminus region DNA had occurred, because the trpC2 locus from the Lform was membrane-enriched when integrated into the parental strain. Nevertheless, this ar gument cannot be fully ruled out, since it is possible that only a very small fragment of the L-form trpC2region entered the BRI51 strain-- enough to make it tryptophan-independent, but not enough to make a significant change in the affinity of the terminus for the surface. It is also possible that the specificity of terminus attach ment is determined by discrete sequences within the terminus region of the chromosome rather than by the whole region itself (35). The trpC2 fragment, which originated from the L-form, although found in membrane preparations, may not be one of such specific binding sequences. Partial loss of the specificity of DNA-mem brane attachment in the L-form could also be due to the inactivation of binding proteins from the membrane and/or the cell wall in the chro mosome-surface complex. Such proteins could have been lost or altered as a consequence of the removal of the cell wall, or during continued growth in the presence of high salt. Alterna tively, it is possible that a mutation occurred in one or more of these proteins, because the Lform has undergone silent and apparent muta tions since it was isolated (1). Furthermore, the profile of the membrane-protein composition (6) and the membrane-lipid composition in the L* form (S. Horowitz, Ph.D, thesis) was signifi cantly different from that of the parental bac terial form when both were cultured under the same growth conditions. In addition to the foregoing studies on the DNA-membrane complex from the L-form, we also have generated information which better defines the extent of the specific membrane at tachment of the chromosome in B. subtihs. The specific attachment of the origin, as examined in transformation experiments by use of the markers purAlS and cysA 14, statistically con firmed that the cy*Al4 gene was part of the origin-membrane complex (28), as is the purAlS marker. In the terminus region, the thyA locus had an MEI value of 1.40 which appeared to indicate that it was membrane enriched, as noted by Yamagucht and Yoshikawa (35), How ever, it did not pass statistical analysis, and therefore must be considered as an internal marker. Thus, it appears that the thyA locus may be the most proximal marker to the termi- CMA 003819 Vul. 1J8. 1979 I- KORM M K Ml1UA N K-1)N A COMI'I.KX 921 mis region on the "ngh< ' -ide of the ehroniosoMi*' 'ee ref 16 and 37 and Kot 1 for a map of tho H 'fi hnnio.soinri 'l he ttpCI nt-n., oiioi oosU LOiiMUered 'o 'n1 an internal in.i.' k : i3., Jm, was significant!,' "nnched in the mcnitnnne. and may oe `he beginning of toe terminus region on the other side of the bidirectionally replicated chromo"otne. The lys-3 marker, which is near to trpC'2. was not membrane-enriched in any ex pel iments. The observation of Neuhard et al. i ia>, describing the temperature sensitivity of thvmidylate synthetase R in B. subtiUs, enabled us to screen separately for thyA and thyB markers, thus adding the thyB gene to the list of terminus loci enriched in the membrane. The ME! of the different terminus markers werincreased as the distance towards the terminus of replication was decreased, suggesting stronger chromosomal attachment near the terminus point. The hisAl and purBS loci which have appeared to he somewhat enriched in some of the previous reports (10, 24. 27, 28) were not found to be membrane enriched in our study, and belong to the internal region of chromo somal replication. The attachment of the chromosome to the membrane, as demonstrated in the present work with strain BUL404, seems to be less extensive and less specific (lower enrichment values) than the attachment of the genome to the bacterial cell wall as reported by Streips et al. (26). These findings could indicate basic differences between the in vivo chromosome-surface complex 'in cluding cell wall) in bacteria and the complex isolated after cell wall removal. ACKNOWl^iKjMENTS We lhank R, W. Gilpin for pru\ tding the L-form. Snl-1. and for helpful Healfoexpiew* our gratitude ^ Hart for helping with the aMtu-ond analyst*. Thia work was aupported by National Science Foundation gram PCM78-0B903 to R.i.D and () N.S., by a grant from th* Manufacturing Chenuat* A**ocwuton to U N S.. and by ln*u* tutional American Canter Numv grant IN-lllR from the Univanity of Loumvillr t > S H and U N N UTERATI RK CITED 1. BatHngtr.riiK iintll T 5 nung 1?^ TnuuUormcth of Baallw* Mihfi!.- iritr.H . 'mini; abilitv id d< i*\ i'm* nucleic arid in Nw,- * L-fomt* or protM Becterivt. 2. Boytaa. K. N. H Mrndalwn, D. Brook, ud k\ H Young* 1972- Regulation ol the baitenel ted wall analyma of a mutant uf liaaltu* nuhtilw del* i me m biceynthari* of inthoit and. I Bactanol 110:2 J90 X Brown, W. C ft. J Doyle, and U. N. Straipi. I*7ti Companion of varwu* procedure far reuioxtoa proUtna and nude* acida from call walla of BacMua *ubtiii*. Prap. Biochrm 6:47: -I&* 4. Dedondor, ft A, J.-A. Lrpruuit, J- Lapeaant>Keicla* rove. A. BUlault, M. Sieinmeta, and F. Kunn. i;"~ Conatnittion of a kit of reference airama for r.ipid genetic mapping in Bacillus subtths 166 Appl Environ. Microbiol J.V969-99J I >n> le, K, ./ , t' N, ''imp**, V. S (.' Fan, V t H:(, v n, H. Moblt'v, .ind M. Mnn-FicUl I1'" ( t-i proit-m in /<<! u/ti' suMili* ) K' 'mu. \\iU. . .11 f.dpin, K- W., F K-Young, and A N. thaUcrjcv i'i'i < 'h.if.Tt leriz.ii ton of h >rat>le [. |orm of lit,, -//oh v<(h/,/|S I**-1' I M,it leriol l tiilpin, K W . and S. S. Nagy. 197o Tmu-o, ,,h...... rvtph\ ol llcudhm >itbtdi'* l. form** ri-f)li* annjt m ikimkI medium .J H,irnrj<d 127' * Hadden, C-, and E. W. Neater. 1%* Fiinfuji.nn ol t ompeiem celU in the Boctlfu' 'uhtih - > ,*wni I Rjiteriol 95;rtTh->W4.> formation Heidrich, H. G-, and W. L. Olaen. (h'oxvniKtnu ' leu at id emelupe ^umpli-Ke^ fn*ni A o hvn hi a t <>h A coniplenospvt ific protein -inti iL I'vviMr tunuion tor ihv stability of the cnmplea. I, Cell Hiol 67:444-4M), 1 Hye, R J., M. A. O'Sullivan, K. Howard, and N. Sueoka. Membrane iswuiaimn nf ^igin terroi* nua, and repluation fork in BquIIua *ub(iU*. p .H.i.'-t*) In D Schlessinger led ), Microbiology-- 197H Amern ail Society for Microbiology, Washington. U.C i 1 Imada, S., L E. Carroll, and N. Sueoka, 1976 DNA* membrane complex in Bacillu* p lib-122 tn O. Schleaamger (ed., Microbiology --197H Ameruan Society for Microbiology, Washington. DC. 12. Ivarie. R. D-. and J, J. Pena. 1970 Aaaociatton of the Bnctflu* whtths thromoeome with the cell meinbr uir reaolution of free and bound deoxyribonucleic acid on Renografm gradienLi J Hactenoi. 1(M:8J9-A.V> H Ivarie, fL D., and J. J. Pene. 19?'t Annociation of many regions of the RactltuH mhtUui chromoeon:* with the cell membrane J Bactenol 114:57U.S78 H. Jacob, S* Bretinae, and F. Curin 196-1 On the regulation of DNA replication in bacteria. Cold Spring Harbor Symp Quant Biol. 19:329-346. 15 Leihowitz. P. J., and M, Schacchtar. 1975 The attichmem uf ihe xu terml (hromnnome to the cell num- brum- lnt Hun ('ytoi 41;l-2f*. >" Lapaeant*Kcjibirova, J., d.*A. Lepeitant, J. Walk*, A. Hillaulk and ft, Dedonder. 1975 Revision ot the linkage map of llnulfu* -<ubtih* lhA induction.* for circularity of the chmmnHoine *1 HMtenol- 12I "2U M4 Neater. L W,, M. Schafer, and J. Lederberf. I9M flene link.igo in DNA transfer a riuMter of genen <un- 11`rm-d wnh aromatic biouyntheata m Bacillus* suhttlr* \ * i* i mw 46:59-9<51. 1' Neuhard, J., A H Price. L Sc hack, and L Thomao- vn. 197H Two thymidvlate nynihetaaev in Bacillus* -ohnhti Prut Natl. Acad Sci. USA 75:1194-11^ 11 Parker, 11 U. and D. A- Glaaar. L97i rhromo^omal ^Me** of DNA-memhrane attachment in E^hcruhta ro/i. J Mol Rio! 87:15.1-166. J Uuinlan, D. C,, and *1. ManilofY. 197*2. Membrane auw^ (tation ot the deoxyrihonucleK* attd gruwing*pomt re gion in Mycttplasma ealli'wttUcum 1 Bactenol 112: i 5-I.I7**' Niuinlan. D. C., and J. Mamloff. I97;t IkHOkvnlatoiM I?" a' id syntheiM!- in <vn brommalv growing A/vcfyi/<runn aalli.yrpticum I Rscti-riol llS:ll7-lJU Kyter, A. 196H AaMKiation of the nurieua and the mem* brane of ba<iena: a murphologMal tudv. Bactertul Rev 3S:;-44 i Kyter. A** and (>. E. Landnaa I9HA Morphologu-nl Hiudv of the attachment of nucletnd to metnhrane in 04-.1II1. protupbi-^l.'s and reverting: prutoplan. <*t Ituuthis* -uh/Uis, p Uo-12-t In L. R 1*u/e te<l t, Mu rohioi protopi,i>.i*. v|ihcrof|iL'vi> and 1. lorm* !h<* Wi'I.mr* mi Wilkiiu* c > . Hultim*)ri- -. Snyder, H. W,, and F. fc. Young, IW* Ajeanmtam \ uL. 1J3, `.373 Mr MilitAMvU.N \ u 'Ml'I.rA :rji f ' ' l O* ^ ~ '1 1 I ' '' : .r- : M s.isM. , 2?1, w,is .liiinficantl' "'ir'c'i-'l in the n*i:>r:in*. -."M nid. Oc `he beiirtnir,' A t:>e tetminu.- il-l'Io ; 'ii the other side of n* hiilire- tionnlh replicated .-hromo*.oi;ie The .;_w .7 marber. which is near to 'n'C2, i*as not membrane-enriched in any py- peiimehi= *1 tie obsc-naijiin ot Neahaid et .1. 118h descriOing the temperature sensitivity o: tnvmidyiate synthetase B tn 8 suhtih*. enabled us to -.creen sepi-.i'-., for v.A and tu.u maikurs. tnus adding th> `n\B gene to the h.it ot terminus loci enriched in the membrane The MEI of the different terminus markers wer" increased as the distance towards the terminus if replication was dec*'1 -ncgest'.ne rur cer chromosomal attachment near the terrmnus point The HisAI and purBS loci which have appeared to be somewhat enriched in some of the previous reports (10, 24. 27, 23) were not found to be membrane enriched in our study, and belong to the internal region of chromo somal replication. The attachment of the chromosome to the membrane, as demonstrated m the present work with strain BUL404, seems to be less extensive and less specific (lower enrichment values) than the attachment of the genome to the bacterial cell wall as reported by Stmps et al, (26). These findings could indicate basic differences between the in vivo chromosome-surface complex (in cluding cell wall) in bacteria and the complex isolated after cell wall removal. ACKNOWI-EixJMEfTTS W thank R. W, Gilpin fnr |iru.Ktv\f iha L-furm. Sii-1. .tie for hatpfii) dxumnuwi W r alw> nprivour griulurii- in T Hart fur httping with th.- ..Inrun .1 rnalrm. Thia work waa auppuncd by National Sewnci- Ho..iH;uion gram PCM7S-QWW to R i D ami U N 3 . by a (rant (run bt Manufacturing Cbemiaia Aanuriauun to tf N 8: and bv lru*ututional American Canc.r Suo-tv cram 1N-1UB front tinUruvwnaty of Loumvdle 11 S H -M'l U N S LITERATI KK CfTED 1. Bettingvr.G. Euand k. 11- 4 eung. IS7S Tranalomu in >t of nrr//e* wibrif.. irti, . -nimg ahiliix ni d<- *-1 e-,,. nudnc ai-af m I,-.'.* p L-forma ur proit Battenei. ta:*kt:-S>x< 1 laylaa IL 4., N. >1. Mukrlna D. Brook., and s' Yeuag. IV72. RuU(iimi ot th* bwionai m 1 wall, atmfyiop of a mutant of Bonita* Hrl.i uv m bionymh*** of m-hJ. 4 Bactamd 11*;, ,7ki 3. lion W, C* Re J. Doyl*. aid U. N. SCrvtp*. CoApnnof^i 9f vanouit procadMfaa fin* rviitutm* ofu- tmna and nocWtr acid* from cod wJb of dol\Lu* *yb- tUU Pr#p. BtOCftro^ 0;4?C-lfw. 4. Dadondor. IL A* J,*A. LtfiMut. J. L*ji*anntwK*i*l*. rova. A* BUUttlt, Ms SkttfMti, tad F. Kunwt. l j7' Conauiiertaon of i kit of itfowm.1 strain* for n-.piri iwioin.K \i f- K > oun^, and A, N ( hriitt-rit-v ''1 * r ' ' *'r -I' ">'* (I 1 s' ll'if L ft/YTl .* .'in, - m. -'f `La I Hv te-ujl U'I;4o^-494 i -tlpm, M, W , ;md S S Ni*|fy, ~ - ,r rs|ii*it ui /JiTi 'Uu* s<(iV(/f> (, fy/rmw f*'nn Rt'DC .n 1 hv'rru'1 127 ''Him ' Hdtid"n, C. ind fe! W Nts'.*r .n ji ' t.t?lU in the Buciltu* /><i/t, rrdn^'ornui.yn s\-:*ni -I Hji'mnfK US-HTfk-Jwa,, Heidnth, H. ( . snd W L. Olson. !'*'*. iVxvriifunu*" -ii Ui -T ...... . yonipirTl** fftmt h w h\ ht'l' m \ (.-iiiinlct sn*i itlc prot-'in iol n* tim.'UMi :or iho >iabiluy >( th* cnmpltx. I Colt Hul 57:444-4M) ' Kvnf, K. JM M. A. O'Sullivan. K. Ho^jrd, snd Sueoka. l1*^ Msmoranr a?*t>f*unvn iius, and rtpijLjuon lorn in HauJu* p <,-*) VO Sn;fsini{pr *<r i Mtr`')iiin;.,sv -- ^.'mh >"<- s'tv -o- M<.ri>niolo<y 'Vasnir^'on OC 11 ImadA. S., L E. Carroll, anti N. Suo^ka. (9*6 D'^'A- mppinranr '`nnpirt in Qacutv* nibrUt* p II'j-IJ'J !n 0 Scnlrs^meer ). Microniuln^) -- I97t Amaru an Scotty Tor Micmluulofy, DC, 12. Wan*. R. O,. and J. J- Han*. 1970 Association of th* dnct/lun suMifn ihruTKaw* wuh ih# call nrtnbriAr rintoluuon tt( Irtc and bound dtotynbunudfic a*l on ^?noraf'n ^ajicnu ) Bacicnoi l04:dJ^-AV [ I lvaria. R. D., and J. J. Pan*. I'tT'j AMKiunticn of mn*y 'rrona of :ha utMUi* khromnaon:* wjih iha trll mvmlM'anr ) Bat-icnol. LU:57US7k, H Jacob, F-. 9, Brawwr, and F. Cuxia. 1963. On th* rtfuUtion of DNA laplkshon in bactona. Cold Spnnf Harbor Symp. Quant. BioL 29:329-346. M Lihewitz. H. J,, and M. Schahur \H7*> TH# r(l.tco- m*m uf 'ht* -MH'irnNl t hfwnwom* it* (hr call rrMfi* 1 orwni ini ji.-k C'>iol 41:1-26. ' Lapaannt'Ki'j/lHrnvo, J., I.*A. Lapa^iant. J. Walk*. A. Hdlouit. and R, (>^dnnd*r. 197^ Rcvt^mn tt Hl- inW^ija mop '>f tihnhs lh6 mdu-itun> fur iiriL.)^nt\ .f :ht hrunrtwifflr, / Rai'tannl. 121 o-i4 Naactr, E, W,, M. Schafor, and J. L#d+rb*i*f. iwtl Vrna unk.iiif* in DNA rrantftr riuatvr of wm* run. i>rn< d with ammatM hiuayn(hu in facxlUt* Nittfili* ' N^unard, J.. A, R. Pricr. I- S*Ju*ck- and E. TVxhm* a* a. 19?N Two rhvmtdvlat* ymhataM in Bftn/Iua *"hnh*. Prv Natl. Arad. Set. USA 1' Harare. D. l,, and D. A- Gla**r. 1971 ('hromcaaimai 'ht-- of DNA'ma<ohrat itlat-hmcnt in EM-h+mhut run J. Mol Had. ' tAimlaift* U. C.. and 4. MialMf. 19T2. Mambrprw .*f miHin ol (ho douNvnhnnucWic attd jnowuiC'pnmi ra- rfion m Mycitp1o.'*ma jailii+fMirum J Boctonul. 112: , ;j7b Duinlan, D. C,, and 4. Mantlnfr 771. UntvnixHiiH Iru id symbol!* m vmhmAnt*4v sroanni .1/vit taih**pticttm 1 R*fi*rvil llI:M7.|ji. ,, Kyt*f. A* I9MM AnuL-iatua of th* imhIow and tho mom- hrano uf hat ion*: mov^bokiyicaf Mudv. Sactonul. Rrv 32:.-Vl < Kytar, a*d O. K> r igm. Monbi(*nral study of (ho aitarhcnom of nurlo***H to momhranp tn naeilii. prmotikuu* add rovertunc prutootaat-* **f IUn tilu* >uM(Um, p 110-12:1. in 1- H. Huso <l 1. Mnottowl ppMopiuHis plM'ftitdaMa and (. tortv* Th^ ttnl Wilkut^ i. it Haltmauv .. >nydr. R. and Y. K. Yittsnf. I9*ft AMateiuiNtn CMA 003821 922 HOKQWITZ ET AL J Bactfkjol between the chromosome and the cytnpU^mic mem brane m 8ac(Uu* ubnlts Binihem Bivphys Res O'mrmm 3SM64-\>' , * **pi7j/,en. J 19A8 MV information <>f ^i*n h^tnicjIK <lui t lent trains of Bat tlfus *u fifths b\ deox\ nhonuiIc.ue Pryc Nad Acad Sci USA 44.1072-10?? 2b Streips, U. N , ft. J. Doyle, W, D Crsbb, M. A, Court* nev, and V. 5. C. Fart. 1976 Significance of f)NA attachment to the cell surfeit? of Bnullo- p 6,1-68 tn A Portqles, R Lopei. and M Espinosa ted f. Modem trends tn bacterial transformation and trana fection Elsevier Press, Amsterdam 27 Sueoka* S,, and W. G. Quinn, l%8 Membrane attach ment of the chromosome replication ongin in Bacillus sidbtiit* Cold Spring Harbor Symp Quant Riol 33: 695-706, 28 Sueoka* N.. R. J. Bishop, N. Harford, K H. Kenneu, A, O'Sullivan, and W. G. Quinn. 197J Chromosome replication and cell metabolism m Racillu'' '-ubiiUs, p 73-87 tn Z Vanek. Z Hosmlek, and ) Cudlin led }, Genetics of industrial microorganism*. Academia, Pra gue. 29 Sueoka* N-, and J, M. Hammers. 1974 Isolation of DNA-memhrane complex in Raciffu* <nb(t(ii Pr<x Natl. Acad. Sci, U S A 71:4787-4791 30. Toyoda, H., K. Yamaguchi, and H. Yoshikawe* 1978 Chromosome-membrane oasocuition in Bacillus >ub tilt# IV Further purification of PNA-menibran# uw- plex by using a combination uf centrifugation and elec trophoresi* J Bactenoi 136:799-802 (l Wjner, B. J I9?l >mele factor experiments having re peated me^ures on the same element-,. p 261- I1'? In 'M.in'iudl prim in experiment,il deMitn 2nd e<l M.<.,faw Hill, Roy*, Uu New. \ork 12 Winston, and T, Matsushita. 1976 Protein symht'io. and the release of the replication terminus from the cell membrane in Rnn/fu.s'.ubft/i'., p 123-127 In O Nchlev singvr ted ), Microbiology--1976 American Solely for Mit robwlogy, Washington. D C 13 Wyrick, I\ B,, and H. J, Rogers. 1973 Isolation and characterisation of cell wall-defective variants of Bacil lus nubUh* and Bactltu* hcheiuformu. J Bactenoi 116-456-465 34 Yamaguchi. K., and H. Yoihikawa. 1971 Topography of chromosome mcmbranejunction n Hanitut* >ubfdis Nature ILnndonl New Biol, 244:204-206 15 Yamafuchi, K., and H. Yoshikawa. 1975 Association of the replication terminus of the Bactllus \ubute* chromosome to the cell membrane / Bactenoi. 124: 1030-1033 36 Young* F. * P. Haywood, and M. Pollock. 1970. Isolation of L-forma of BaciUus subttlis which grow in liquid medium J. Bactenoi. 102:867-870. ,17. Young* F. E** and G. A. Wilson, 1976. Revision of the linkage map of Bacillus subtiUs, p. 686-703. fn G. O. Fuman led.), Handbook of biochemiatiy and molecular biology, 3rd ed , section B, vol, 11. CRC Prvm. Cleveland. Ohio. CMA 003822 Reprinted frdfO Microbiology 11W0 c 19*0 American Society for Microbiology Genetic Analysis of DNA-Surface Interactions in Bacillus .subtilis ULDIS N STRF.IPS. SARAH HOROWITZ. *sd RONALD J DOYLE Deportment oj Mtcrohmlogs and Immunology School of Mediant. University of Lmiisalle Louisville. Kentucky 40232 In a paper which has provided the foundation for many productive experiments in molecular biology and bacterial cell division. Jacob. Bren ner, and Cuzin (II) proposed the "replicon" model to explain some of the mechanisms of DNA initiation and replication. One major con cept of their proposal was that portions of the cellular genome were attached to the cytoplasmic membrane. That assumption made it possible to account for proper segregation of the genome into daughter cells concomitant with cell growth. Subsequently, research in several laboratories demonstrated that the chromosome of various bacteria was indeed attached to the surface. Moreover, the attachment was at the origin and terminus of replication and at the replication point, as well as at several nonspecific points along the chromosome (4, 5, 9, 13, 17, 18). The specific association of the chromosome at sites involved in DNA replication reinforces the con cept of a causal relationship among DNA-surface attachment, cell division, and chromosome segregation. To study this relationship further, we have investigated DNA-surface attachments in Bacillus subtilis under conditions in which normal cell division and DNA segregation are disturbed. MEMBRANE-DNA COMPLEX IN A STABLE L-FORM A stable L-fonn from B. subtilis, sab I, has been propagated in liquid media since 1969 (19). The L-form grpws in the absence of any cell wall and is stabilized by 1.2 M NaCl. sab I and its salt-independent derivative, sig-l, have been shown to divide aberrantly, producing progeny with altered cytoplasm and DNA contents (6. 7). In this respect, these two L-forms are similar to other L-forms isolated from B. subtilis (14). We. have used sal-1 as a model for an organism which has lost proper cell division and DNA segregation control. To determine whether the DNA-membrane at tachment remained intact in these aberrantly dividing cells, we isolated DNA-membrane com plexes by use of Renografin gradients (9). The complexes were then assayed for DNA content by transformation, and membrane enrichment indices were calculated (9). As shown in Fig. 1. when compared to spheroplasts (panel A), the L-form (panel B) retained enrichment for genetic markers close to the origin of replication. purA 16 and cysAI4. Several internal markers were not enriched in either sample. However, selective enrichment for genes close to the terminus of replication, such as trpC2, gltA292, citKS, and thyB, was found to be lost in the L-form but not in spheroplasts. It is attractive to speculate that this loss of replication terminus attachment con tributes to the loss of division control in the L-form. There are several factors which could have contributed to the loss of replication terminus binding (9). Among the most probable alterna tives are the absence of a cell wall or the pres ence of high salt (1.2 M NaCl) in the growth medium. If the loss of cell wall contributed to the dissociation of the terminus of replication from the surface, then it follows that the cell wall contributes to the makeup of the normal DNA-surface complex. CELL WALL-DNA COMPLEXES To examine the possibility that the cell wall is involved in attaching the chromosome to the surface of B. subtilis, we isolated cell walls at various stages of growth and examined the preparations for transforming activity (1, 15: R. J. Doyle et al., submitted for publication). Not only did isolated cell walls possess trans forming activity, but the transformation was specifically enhanced for genetic markers around the replication origin and terminus (Fig. 1C). The extent of enrichment in the cell wall differs from that observed with membrane-DNA com plexes. Thus, hisAI, pyrAI, and sacA were en riched in the cell wall samples, whereas the same markers were not enriched in membrane-DNA complexes (9. 15. 18). In contrast, cysA!4, a marker which was shown to be membraneenriched. was not found to be specifically attached to the cell wall (9, 15. 16). The signif icance of these differences is not obvious at the present time. There is an overall symmetry ap- 1*4 CMA 03824 STREIPS HOROWITZ. AND DOYLE 285 AB Flo. I. Genetic analysis of surface-DNA complexes in Bacillus jubtilu. Genetic markers on the chromosome of B. jublilis and its L-form, sal-1. were examined for enrichment in membrane and wall preparations (9. IS). Represented are the enrichment maps for spheroplast membrane-DNA (A). L-form fsal-l) membrane-DNA (B). cell wall-associated DNA (C). and membrane-DNA from 1.2 M NaCI-treated cells (D). The genetic mop of B. subtilis is that presented by Young and Wilson 120). Loch marker is designated as enriched f ). nonennched (O). or no) examined!'). Replication origin 10). replication terminus IT), parent in the cell wall-DNA profile (Fig. 1C), which is not present in the membrane-DNA preparations (Fig. IA and B). This loss of sym metry could be due to changes in the topography of the in vivo DNA-surface complex after re moval of the cell wall. Along with these altera tions it is possible that prolonged growth in the absence of a cell wall could also result in CMA 003825 286 DNA-SURFACE inter \l! IONS the dic-ociation of the terminui from the mem bone EFFECT OF SALT ON MEMBRANEDNA COMPLEXES An alternate possibility for the loss of attach ment of the replication terminus by the L-form is the presence of 12 M NuCI in the growth medium of this organism. A high concentration of salt may destroy the integrity of the surface-DNA complex. The effect of salt on morphology has been documented <3, 12). We grew B. subiilis BUL 404 imetBlO) in 1.2 M NaCl and isolated membrane-DNA com plexes at various intervals. In Fig. ID. we show the membrane-DNA profile from cells grown in 1.2 M NaCl for 120 min. It is obvious that growth in salt dissociated the replication termi nus from the membrane in the bacterial strain. Recent experiments suggest that replication terminus dissociation occurs within the first 30 min after addition of 1.2 M NaCl to the culture growth medium. Moreover, after 120 min of growth in 1.2 M NaCl, the entire population of cells has assumed abnormal morphological char acteristics. suggestive of changes in cell division control. Thus, we have been able to emulate the L-form state of replication terminus detach ment and induce abnormal cell division by grow ing B. subiilis cells in high salt. Presumably, piasmolysis has occurred in these cells. There fore, it is attractive to postulate that the replica tion terminus site may be labile to plasmolytic detachment of the membrane from the wall. Such removal could destroy the integrity of this suriface-DNA site. In contrast, the replication origin-surface complex appears to be inviolate to the effects of high sail concentrations. Origin of replication binding has also been maintained in the L-form. This suggests that the replication origin-surface complex may be of primary and vital importance for cell survival. DISCUSSION On the basts of the foregoing series of ex periments, several conclusions can be drawn concerning the surface-DNA complex in B. subiilis. First, the in vivo complex contains not only the chromosome and membrane, but also pcptidoglycan. Second, of all the binding sites, the attachment of the origin of replication is of primary importance. This attachment is main tained in all samples assayed to date. Third, the replication terminus attachment appears to be labile to high salt and perhaps prolonged growth in the absence of the ceil wall. Thus, the L-form and cells grown in 1.2 M NaCl have k"i pieteiemial attachment at the terminiis uf icplie.tutm Iciminui attachment and notmal morphology cun be restored after removal ot the 1.2 M NaCl (S. Horowitz el al.. submitted for publication). Finally, once the attachment in the terminus region is lost, cells are observed to have altered morphology and division patterns. These data suggest that the DNA-surface com plex is a vital structure for the procaryotic cell and has a dynamic role in the maintenance of the cell cycle. The complex may regulate not only DNA replication but also DNA segre gation and cell division events. If the complex is disturbed, then aberrancies in cell division processes occur. It is possible that information for the proper maintenance of the complex and its functions resides in the proteins found in both the cell wall and cell membrane (2. 8. 10). Such interactions, however, remain uncharacterized at the present time. The ability to manipulate the constitution of the DNA-surface complex and the morphology of cells leads to the pos sibility that the individual functional com ponents may be isolated and studied. ACKNOW LDGMEATS Wt thank Joey OoMim for lechmcai awstanc* in a pan of this research. Tarry White provided valuable discussion Junb|the preparation of tha manuscript This research was supported by National Setanet Founda tion pant PCM 74-00903 to R J.D. and U.N.S.. by grant from the Manufrctunng Chemists Association to U.N S., and by Institutional American Cancer Society grant IN-IIIR from the University of Louisville lo S.H and U N.S. LITERATURE CIRS I two, W. C.. a. J. Dark, and L. N. Strvipa. 197*. Companion of vanous procedures far removing pro tein and nucleic acids from call walls of Xmi/fw luMUt. Prep giochem. 4*479-410. 2. OeySs. X. j.. U. N. Ssiaipa, V. S. C. Fan. W. O. Bro-m H. MoMty. and J. M. MlldlM 1977. CeH wall protein in Aanffnr uhii. J Bncteriol. 129:547- 549 ). UMK T., . Ward, and H. apes. 1975. Formation of call wail polymers by reverting protoplasts of 6w ilhii tiihruifiirmn.J, Bacterid. 124:423-432. 4. rVsbsin. W. 1972. The DNA/membrane fraction of Pittmmarncriis contains a DNA repheation cumpiea. i. Mol. Biol. 79:343-397. 3. Csnmn. A. T.. and J. Ladarharg. 1963. A cell membrane bound fraction of bacterial DNA. Biochem. Biophyv 6 Res. Common. 14:424-43). . GBpta, It. W,, and S. S. Nagy. 1976 Time-lapse pbotography of bullsr m/in/n L-form-> rtpiicutmg in liquid medium. J. Bnctenol. 127:1014-1021 7 GRpin, R. W,. and S. K. PMsarsan. 1974 Adaptation of a liable L form of Ruutlmi wrbn/n to minimal -allmedium without osmotic siabih/ers 1 ttactcnid 125: K4J-H49 4 Harman. J. M., and H. W. Taber. |V77 Altered ac cumulation of a membrane protein unique to a membrane-deuaynbonuclcic acid cumptes m a tints ifutiaiion mutant of ffiir *f/u: iubrrl/\ 1 Racicnol 139:1224 -12" CMA 003826 286 D.\ A-S>1 RF u_`E I NT', < \. ' Itb the iIi".>ciatton ,'f tr.c i-.1- ' ~v Tit"' effect of salt on membrane- DNA COMPLEXES An alternate possibility for the loss of attach ment of the replication terminus b> the L-form is the presence of 1.2 M NaCI m the growth medium of this organism. A high concentration of salt may destroy the integrity of the surface-DNA complex. The effect of salt on morphology has been documented 13, 12). We grew B, subnlis BUL 404 (mttBlQ) in 1.2 M NaCI and isolated membrane-DNA com plexes at various intervals. In Fig. ID. we show the membrane-DNA profile from ceils grown in 1.2 M NaCI for 120 min. U is obvious that growth in sail dissociated the replication termi nus from the membrane in the bacterial strain. Recent experiments suggest that replication terminus dissociation occurs within the first 30 min after addition of 1.2 M NaCI to the culture growth medium. Moreover, after 120 min of growth in 1.2 M NaCI, the entire population of cells has assumed abnormal morphological char acteristics, suggestive of changes in cell division control. Thus, we have been able to emulate the L-form suue of replication terminus detach ment and induce abnormal cell division by grow ing B. iubtUii cells in high salt. Presumably, plasmolysis has occurred in these ceils. There fore, it is attractive to postulate that the replica tion terminus site may be labile to plasmolytic detachment of the membrane from the wall. Such removal could destroy the integrity of this surface-DNA site. In contrast, the replication origin-surface complex appears to be inviolate to the effects of high salt concentrations. Origin of replication binding has also been maintained in the L-form. This suggests that the replication origin-surface complex may be of primary and vital importance for cell survival. DISCUSSION On the basis of the foregoing series of ex periments, several conclusions can be drawn concerning the surface-DNA complex in B. smJbuMt. First, the in vivo complex contains not only the chromosome and membrane, but also pcptidoglycan. Second, of all the binding sites, the attachment of the origin of replication is of primary importance. This attachment is main tained in all samples assayed to date. Third, the replication terminus attachment appears to be labile to high salt and perhaps prolonged growth in the absence of the cell wail. Thus, the L-form and ceils grown in 1.2 M NaCI have i". r,-T_'.'"u, .t`__ -- -nrtt 'h-j k'T'i,, itl.lkb'T'CTt and v ' , morphology .an be restored alter removal ot ;ne 1.2 M NaCI iS Horowitz e: al.. submitted lot publication! Finally once the attachment in the terminus region is lost, cells are observed to have altered morphology and division patterns These data suggest that the DN A-surface .am ple x is a vuul structure for the procaryotic cell and has a dynamic role in the maintenance of the cell cycle. The complex may regulate not only DNA replication but also UNA segre gation and cell division events. If the complex ts disturbed, then abcrrancies tn cell division processes occur. It Is possible that information for the proper maintenance of the complex and us functions resides m the proteins found in both the cell wall and cell membrane (2. 8. 10). Such interactions, however, remain uncharactenzed at the present time. The ability to manipulate the constitution of the DNA-surfacc complex and the morphology of cells leads to the pos sibility that the individual functional com ponents may be isolated and studied. ACKNOWLEDGMENTS We thuh Jo*i Dobbun tor technical aursrancs in t part o( this research. Terry Whitt provided mhuMi di.tuvu.in Junitf tht preparation af Lht manuscript. Thu research was wiepuntil by Nationml Semite* Founda tion pant KM 7I43S90J to ft.J.D. and U.N S . by a gram troth tha ManuDetunn* Chemists Association to U N S . and by lasdtuuoiial American Cancer Society (rant IN-lliH front the University of Louisville to S.H and U N.3. UTTXATUeX CITED 1. Inn. W. C.. ft. J. Day*. ^ L. ,N. Stretp. 1974. Comparison of -snout ptocedwos toe lathovute pro tons and nucleic acids front cell walls of ftnillet rubttda. Prep. Biodtem. 4t479-AM. 2. DaySa. ft. J- U. N. Stamps, V. S. C. tut w. D. bev-w. H. Maairy, aed J. M. MaaaMd. im. Cttt watt protein in ftentfru teprifu l Sactenot. I24i*a7-5S4 3. DM. T,, ft. Wmd. aad M. lapn. 1*7*. Formation of eatt wall polymers try reverms* proioptnis of luv Jint MrWomen. I. Sanenot 1241*23-432. a. FtnMa. W. t**2. The ON A;membrane fraction uf PnmasuriKTU contatas a ONA reptiewou cumptex. J. Mol. 3mt. 7li3i3-397. 3 Ganeaao. a. T., and J. Ltd- Sir*. 194*. A cett membrane- bound fraction of bactcnpl DNA. Iliocbem. Bmpbyv Res. Common IM24- 43J. 4. GttpM ft. W,, nad S. S. Napy. 1474. Tinw-lapse pbuiur rapby of iMrittui <uhlii L-form. repbeattup m liotnd medium. J. Bacicnot 127:1011-IU2I 7 GftpM ft. W,, sod S. K. IV74. Adaptation uf u viable L-furm uf thuillv tunn/it to nuetmai -ail. medium without aunM viabtb/ers I duuemu. I25t S4S-MU n Hermes. J. M.. and N. W. Timer. 1977 Altered ac cumetatton of a mendwane pruem unique to a mem. brans-deotynbonucleiv och! cumtdva m a dnu tiuiuimo ntutaesofttntiflm mhtitii J Bacicmd IJBtl224-I2H CMA 003327 STREIPS, HOROWITZ. AND DOYLE 287 * Horowitz, S , R. J. Itotle. t K3nuH|:. and t, N Stretpv p let n\ f in'-ov.i ,:i"n >'t ;nc , hr. `irin'.o nt with t hL Tii.''nhr.ifli,* m i ^i.ih|, I term .<1 H,u 'mi ; bjNfertv.i uh-yi* c in Imada. S , L. Carroll, and > Sucnka 197h Membrane OS a rumple* in finish' uifrttfts, p IJ6-I22 /> 0 SthlcAsinger ted ) Mitrohiolog) -- 1976 \mcrn.an Solely for Mu.fobu*U'&> Washington D C I I Jacob, f\, S. Brenner, and F. Cuzin. ISK>3 On the regulanon of ON A replication m bacteria, Cold Spnng Harbor Simp Quant Biyl 2a,)2y- UK I' Kusaka, I. 1971 Electron microscopic observations on growing and dividing protoplasts of Hu<illn> wt`v` tenum. J Cen. Microbiol 63:199-202 13 UiN>*itx, P.. and M. Schaectcr. 1975 The attachment of (he bacterial chromosome to the cell membrane. Int. Rev Cytol. 41:1-2*. f4 Ryttr. A. end O. E. Landman. 1964. Electron micro scope study of the relationship between mesosome loss and the stable Lsutc lor protoplast state! m Bautins subulh. i. Bactenol. 8St4?6-467 15 Streipe, U. N,, R. J. Deyle. W. D. Crabb. M. A. Courtney, cad V. S, C. Fan, 1976 Significance of ONA attachment to the cefl surface of Buullm subulis, p 63-68 In A. Ponolea, R. Lopes, and M Espinosa ted,). Modern (fend" r. -'.h'.p.j transformation and ira*s!e*ni *n t- i-cvi^r N -r " HwHand Publishing Co \mMcrd.tm in vieoka. N . K, J Hi,hup, S Harford, K. H. Keiini-tt. \ O'Sullivan, and VS 0. Quinn, 197] (_nromosornv renh wjtmn and ^cIt metabolism m Bu<itin\ \nhnU\ p 'V- 87 hi Z Nahik Z HoStalek And J Cudlin ied i Oeneocs of industrial microorganisms Academia Prague 17 St^oka. V, and VV. Qgmn. 1968 Membrane Attachment of the chromosome replication origin n Buctilm v6- ri/i.v Cold Spnng Harbor Symp Quant Biol 33:69?- 70? )8 Yamagucht, \. and M. Yoshikawa. 1977 Chromosome membrane association in Bactllu* subnh\ HI Isolation and characterization of a DNA-protem complex carry <n| replication ongm markers, J. Mol. Biol. 110: 219-253 19 Young, F. E.. P. Haywood, and M. Pollock, 1970. Isolation of L-forms of Bacillus suhttlts which grow in liquid medium, j Bacteno). 102:867-870, 20. Young, F and G. Wtbon, 1976. Revision of the linkage map of Baal/as subt/tis, p 686-703. MG D FasmaiHed.i. Handbook of biochemistry and molecular biology. 3rd ed,, sect. B. vol. 2. CRC Press. Cleveland. Ohio, CMA 003828 15 TRANSFORMATION - 1978 Proc. 4th Internat. Mtg. Bacterial Transformation 245 RESTRICTED CHROMOSOME-MEMBRANE ASSOCIATION IN A STABLE L-FORM OF BACILLUS SUBTILIS. S. Horowitz. R.J. Doyle, and U.N. Streips, Department of Microbiology and Immunology, University of Louisville, Schools of Medicine and Dentistry, Health Sciences Center, Louisville, Kentucky, 40232 (USA) ABSTRACT A stable L-form (sal-1) of Bacillus subtills BR151 was found to retain a chromosome-membrane association. The attachment of the chromosome to the membrane in the L-form is specific for the replication origin but not for the replication terminus. These results are in contrast to those obtained with the parental, cell-walled strain in which both the origin and terminus of chromosomal replication are preferentially attached to the membrane. INTRODUCTION Attachment of the chromosome to the cytoplasmic membrane and to the cell wall has been well demonstrated in cell-walled prokaryotes and is postulated to be instrumental in DNA segregation and division control [1,2,3,4]. This attachment is specific for the origin and terminus of chromosomal replication and for the replication point [1,4,5]. In Mycoplasma the chromosome was also found to be attached to the membrane [1,6]. Studies of unstable L-forms, utilizing electron microscopy, have also revealed chromosome-membrane association [7]. Yet, the genetic specificity of the chromo somal attachment in cell wall-deficient prokaryotes has not been studied. In this report, we demonstrate that a stable Lform (sal-1) of B. subtilis BR151, which has no cell wall polymers and divides aberrantly [8,9, R.W. Gilpin personalcommunication], retains a chromosome-membrane association. The attachment is preferential for the region at the origin of chromosomal replication, but the specific attachment of the region close to the replication terminus has been lost. MATERIALS AND METHODS Bacterial strains. The stable L-form (sal-1) of B^. subtills BR151 [8,9] was kindly provided by R. W. Gilpin and retains the metBlQ marker. The parental strain, jS. subtilis BR151 lys-3. trpC2, metBlQ, was transformed to prototrophy for CMA 003829 246 lys-3 and trpC2 with B^, subtills W168 DNA to maintain the same auxotrophic background as the L-form. All strains used as recipients in the transformation experiments are listed elsewhere [10]. Media and growth conditions. The L-form (sal-1) was grown in sal-1 medium [10] which contains 1.2M NaCl for stabilization, according to a modification of the technique described by Gilpin t al [9], in the presence of deoxyadenoslne (200 yg/ml) and l^C-thymidlne (0.1 iiCi/ml, 53 mCl/mmole) for about two generations. The culture was washed in NCP buffer [11] plus 1.2M NaCl and lysed in NCP buffer by osmotic shock. The parental strain, subtllis BR151 metBIQ. was grown and labeled under the same conditions as the L-form but without the NaCl, washed In NCP buffer, and lysed by addition of lysozyme (0.5 mg/ml). Renografin gradients. Lysates of the L-form and its parental strain were applied onto 0-38Z linear Renografin density gradients [11] and were centrifuged, fractionated and counted according to a modification of the technique described by Ivarie and Pene [11]. Transformation procedures. The procedures for the development of competence and transformation were according to ehose described by Boylan et al [12]. Dilutions of the peak fractions from the Renografin gradients served as donor DNA. RESULTS Attachment of the chromosome to the membrane. Uniformly labeled DMA from an unsheared lysate of the L-form sal-1 (Materials and Methods) forms one distinct band at the lower part of the Renografin gradient. This band corresponds to the membrane-DNA (mDNA) peak which is found in the untreated lysate of the parental strain _B, subtills BR151 metBIQ (results not shown). The DNA attached to the membrane of both the Lform lysate and the lysate from the parental sCrain contains about 80Z of the label incorporated. Shearing the L-form lysate by vortexing for 30 seconds at setting no. 6 causes a decrease in the amount of labeled DNA which appears in the m-DNA peak and generates a slower sedimenting peak which bands at the upper part of the Renografin gradient. This new peak corres ponds to the free DNA (f-DNA) peak of the parental strain (results not shown). The distribution of the uniformly labeled DNA from the L-form sheared lysate consists of about 55Z label which appears as free DNA and about 30Z which appears as firmly attached membrane associated DNA. This distribution is similar to that of the uniformly labeled DNA from the sheared CMA 003830 CMA 003831 :4? lysate of the parental strain [10], Genetic analysis of the specificity of chromosome attachment. Since the DNA isolated in the Renografin gradients can be used in transformation assays, genetic analysis for the specificity of DNA attached to the membrane was performed, using peak fractions from the sheared lysate of the L-form and the parental strain as donor DNA. Genetic markers were tested throughout the chromosome and included markers at the origin and terminus regions of chromosomal replication as well as Internal loci. The specificity of the attachment was determined by the membrane enrichment index (KEI), calculated according to Sueoka and Quinn (2) using leuAS as a standard. Several transformation experiments were performed for each marker with both the L-form and the parental strain DNA peak samples,and the average MEI was calculated. An MEI value which equals or exceeds 1.40 was taken to Indicate that the marker is enriched in the DNA attached to the membrane, hence preferentially bound. An ME1 value close to 1.00 was accepted to show no specificity for the attachment to the membrane. The results from the transformation experiments are shown in Figure 1. None of the internal markers that we have examined has any specificity for association with the membrane in both the L-form and the parental strain. The origin region of chromosomal replication (purA16 and cysA14) in the L-form is preferentially attached to the membrane as is the case in the parental strain. However, the region close to the replication terminus (trpC2, thyB, citK5, gltA292, and thyAl does not appear to be attached to the membrane of the L-form although it is preferentially attached to the membrane In the parental strain. DISCUSSION This study demonstrates that the L-form, sal-1, retains a membrane-chromosome association in both sheared and unsheared lysates. This attachment is quantitatively similar to that of the parental strain and may indicate the importance of maintaining such an association for normal cell growth. Genetic anelysis for the specificity of the membranechromosome association in the L-form shows that the preferential attachment of the origin of chromosomal replication Is retained. However, specificity for the attachment of the terminus region has been lost, when compared to the parental strain. This finding suggests that the two sites of attachment are different, and Chnt the attachment of the origin is of primary Importance. The loss of association of the terminus to the membrane in the [.-form could be the lysate of the parental strain [10]. 247 Genetic analysis of the specificity of chromosome attachment. Since the DNA isolated in the Renografin gradients can be used in transformation assays, genetic analysis for the specificity of DNA attached to the membrane was performed, using peak fractions from the sheared lysate of the L-form and the parental strain as donor DNA. Genetic markers were tested throughout the chromosome and included markers at the origin and terminus regions of chromosomal replication as well as internal loci. The specificity of tha attachment was determined by the membrane enrichment index (MEI), calculated according to Sueoka and Quinn [2] using leuAS as a standard. Several transformation experiments were performed for each marker with both the L-form and the parental strain DNA peak samples,and the average MEI was calculsted. An MEI value which equals or exceeds 1.40 was taken to indicate that the marker la enriched in the DNA attached to the membrane, hence preferentially bound. An MEI value close to I.00 was accepted to show no specificity for the attachment to the mombrsne. The results from the transformation experiments are shown in Figure 1. None of the internal markers that we have examined has any specificity for association with the membrane In both the L-form and the parental strain. The origin region of chromosomal replication (purA!6 and cysA14) in the L-form is preferentially attached to the membrane aa la the case In the parental strain. However, the region close to the replication terminus (trpC2. thyB. citK5, gltA29Z, and thyA) does not sppesr to be attached to the membrane of the L-form although It Is preferentially attached to the membrane In the parental strain. DISCUSSION This study demonstrates that the L-form, sal-1, retains e membrane-chromosome association In both sheared and unshearad lyaatee. This attachment la quantitatively similar to that of the parental strain and may indicate the importance of maintaining such sn association for normal csll growth. Genetic analysis for the specificity of the membranechromosome association in the L-form shows that ths preferential attachment of the origin of chromosomal replication la retained. However, specificity for the attachment of the terminus region has been lost, when compared to the parental strain. This finding suggests that the two sites of attachment arc different, and chat the attachment of the origin is of primary Importance. The loss of assocLatLon, of the terminus to the membrane In the t.-form 'could lie the CMA 003832 248 5 SO ? Figure 1. Genetic analysis of m-DNA from the L-form and ita parental strain. Genetic markers along the entire chromosome of both the L-form and Its parental strain were examined for specific enrichment in membrane samples (ME1) by transformation (Materials and Methods). Represented is the genetic map of the L-form sal-1 (A) and of the parental strain BR151 metBlO (B). The genetic map of Bacillus subtills was constructed from those of Lepesant-Kejzlarova et_ al [13] and Young and Wilson [14]. Each marker is designated as enriched (e) or non-enriched (o) in m-DNA samples. Replication origin (0); Replication terminus (T). CMA 003833 249 result of one or more of the following physiological and/or genetic alterations. First of all, the loss of the cell wall may have removed an outer surface site for terminus attachment [15,16]. Secondly, a loss or alteration of a binding protein(s) which aids in the specific attachment of the chromosome to the membrane (or cell wall) could be reflected In the loss of association of the terminus region. Also,a temperate bacteriophage which inserts in this area of the chromosome [17] could participate In the binding of the genome. The L-form could have lost such a bacteriophage. Finally, the presence of a high concentration of salt (1.2M NaCI) throughout the growth cycle of the L-form could be responsible for the dissociation of specific, bound regions of the chromosome [3,9]. We are presently investigating all of these possibilities. The chromosome of B. subtills has been found to bind to the cell wall polymer [4,15]. It is attractive to speculate that any alteration of the integrity of the cell wall-proteinmembrane-DNA complex would result in aberrant division and DNA segregation patterns. Our observation that the L-form, sal-1, of subtills is altered in a part of the replication complex, the attachment of the terminus of chromosome replication, provides initial support for such a speculation. It also provides opportunity for further studies on the role of the bacterial outer surface in DNA replication and segregation, and cell division. ACKNOWLEDGMENTS We wish to thank F. E. Young and R. W. Gilpin for many fruitful discussions. This work was supported by an NSF grant (?CM 78-08903) to R.J.D. and U.N.S., a grant from the Manufacturing Chemists Association to U.N.S., and an Institutional American Cancer Society grant from the University of Louisville to S.H. and U.N.S. t CMA 003834 249 result of one or more of the following physiological and/or genetic alterations. First of all, the loss of the cell wall may have removed an outer surface site for terminus attachment [15,16}. Secondly, a loss or alteration of a binding protein(s) which aids in the specific attachment of the chromosome to the membrane (or cell wall) could be reflected in the loss of association of the terminus region. Also, a temperate bacteriophage which inserts in this area of the chromosome [17] could participate in the binding of the genome. The L-fonn could have lost such a bacteriophage. Finally, the presence of e high concentration of salt (1.2M NaCl) throughout the growth cycle of the 1-form could be responsible for the dissociation of specific, bound regions of the chromosome [3,9]. We are presently investigating all of these possibilities. The chromosome of B. subtills has been found to bind to the cell well polymer [4,15]. It is sttractlve to speculate that any alteration of the Integrity of the cell wsll-proeeinmembrane-ONA complex would result in aberrant division and DMA segregation patterns. Our observation that the L-form, sal-1, of i. subtills Is altsred In a part of the replication complex, the attachment of the terminus of chromosome replication, provides initlsl support for such s speculation. It also provides opportunity for further studies on the role of the beeterlei outer surface in DNA replication and sagregatlon, and cell division. ACKNOWLEDGMENTS We wish to thank 7. E. Young end R. W. Gilpin for many fruitful discussions. This work was supported by an NS7 grant (7CM 78-08903) to R.J.D. and U.N.S., a grant from the Manufacturing Chemists Association to U.N.S., and an Institutional American Cancer Society grant from the University of Louisville to S.H. and U.N.S. CMA 00383s 250 REFERENCES 1 D.W. Smith and P.C. Hanawalt, Biochin. Biophys. Acta 1967, 149,519-531. 2 N. Sueoka and W.G. Quinn, Cold Spring Harbor Symp. Quant. Biol. 1968,33,695-705. 3 H.C. Heidrich and W.L. Olsen, J. Cell Biol. 1975,67,444460. 4 U.N. Strelps, R.J. Doyle, W.D, Crabb, M.A. Courtney and V.S.C. Fan in "Modern Trends In Bacterial Transformation and Transfection", edited by A. Portoles, R. Lopez and M. Espinosa, p. 63-68, Elsevier Press, Amsterdam, 1976. 5 R.J. Hye, M.A. O'Sullivan, K. Howard and N. Sueoka in "Microbiology-1976", edited by D. Schlessinger, p. 83-90, American Society for Microbiology, Washington, D.C., 1976. 6 D.C. Quinlan and J. Maniloff, J. Bacteriol. 1972,112,13751379. 7 A. Ryter and O.E. Landman in "Microbial Protoplasts, Spheroplasts and L-forms", edited by L.B. Cuze, p. 110-123, The Williams and Wilkins Company, Baltimore, 1968. 8 F.E. Young, P. Haywood and M. Pollock, J. Bacteriol. 1970, 102,867-870. 9 R.W. Gilpin, F.E. Young and A.N. Chatterjee, J. Bacteriol. 1973.113.486-499. 10 S. Horowitz, R.J. Doyle, F.E. Young and U.N. Streips, submitted for publication. 11 R.D. Xvarie and J.J. Pene, J. Bacteriol. 1970.104.839-850. 12 R.J. Boylan, N.H. Mendelson, D. Brooks and F.E. Yoimg, J. Bacteriol. 1972.110.281-290. 13 J. Lepesant-Kejzlarova, J.A. Lepeaant, J. Walla, A. Blllault and R. Dedonder, J. Bacteriol. 1975,121,823-834. 14 F.E. Young and G.A. Wilson in "Nucleic Acids IX", edited by G.D. Fasman, p. 686-703, CRC Press, Cleveland, Ohio, 1976. CMA 003836 251 15 W.C. Brown, R.J. Doyle and U.N. Strelps, Prep. Biochem. 1976,6,479-482. 16 R.J. Doyle, U.N. Strelps, V.S.C. Fan, W.C. Brown, H. Mobley and J.M. Mansfield, J. Bacterlol. 1977,129,547-549. 17 S.A. Zahler, R.Z. Korman, R. Rosenthal and H.E. Hemphill, J. Bacterlol. 1977,129,556-558. CMA 003837 251 15 W.C. Brown, R.J. Doyle and U.N, Streips, Prep. Biochem. 1976,6,479-482. 16 R.J. Doyle, U.N, Streips, V.S.C. Fan, W.C. Brown, K. Mobley and J.M. Mansfield, J. Bacteriol. 1977,129,547-549. 17 S.A. Zahler, R.Z. Konnan, R. Rosenthal and H.E. Hemphill, J. Bacteriol. 1977,129,556-558. CMA 003838 INTERFERON: PROPERTIES AND CLINICAL USES A. Khan, N.O. Hill, and G.L. Dorn, cds. pp. 589-598, Wadley Institutes of Molecular Medicine, Dallas, TX, 1980. ^ INHIBITION OF INTERFERON INDUCTION AS A SCREEN FOR THE CARCINOGENIC POTENTIAL OF CHEMICALS'* Gerald Sonnenfeld, Mary Carol Barnes, Julia Schooler and Uldis N. Streips Department of Microbiology and Immunology University of Louisville School of Medicine Louisville, Kentucky, USA ABSTRACT The induction of murine interferon by Newcastle disease virus has been shewn to be inhibited by pre-treatment of fibroblasts with Various carcinogens. The present study has extended the range of carcinogens to include chloroacfetaldehyde. Chloroethacol and chloroacetic acid, rarely carcinogen ic analogs of chloroacetaldehyde-, had no significant effect on interferon induction by Newcastle disease virus. When polyriboinosinic-polyribocytidylic acid was used as an interferon inducer, induction of interferon was also inhibited by pretreatment of the cells with chloroacetaldehyde. Addition of reduced glutathione, which can trap active metabolites of carcinogens, abrogated the effect of a carcinogen on Inter feron induction, suggesting that fibroblasts can activate carcinogens in the Inhibition of Interferon induction system. Following further verification of the ability of the inhibit ion of interferon induction system to discriminate among chemicals on the basis of carcinogenic potential, this test could become part of a comprehensive battery of t-_';ts for chemical-carcinogenicity. *This work was supported by grants from the American Cancer Society No. IN111C and the Manufacturing Chemists' Associat ion, and the National Aeronautics and Space Administration under Interchange NCA2-ORA00-901. M.C.B. was the recipient of a Graduate Student Research Grant from the University of Louisville Graduate School, and J.S. was a Summer Research Scholar of the University of Louisville School of Dentistry. CMA 003839 590 INTERFERON: PROPERTIES AND CLINICAL USES INTRODUCTION Many different: tests have been developed over the past several years to determine the carcinogenic potential of chemicals. Several of the early tests, such as the Ames Salmonella test (1), were able to determine the mutagenic potential of chemicals, but could not consistently discrim inate between mutagens and carcinogens. Several recently de veloped assays, which Involve determination of the activat ion of "SOS'* repair in bacteria, have apparently been able to differentiate between carcinogens and mutagens (2,3). How ever, these tests suffer from the drawback that they are carried out in prokaryotic cell systems which may not be analagous to the systems operative following exposure of a mammalian cell to a chemical. Mammalian tests to determine the carcinogenic potential of chemicals have been developed, but also suffer from sev eral drawbacks. Transformation of cell cultures or Induct ion of tumors in animal models require extended periods of time for the carcinogenic potential of a chemical to be ex pressed. In addition. If a chemical is carcinogenic only at a very low rate, massive quantities of animals would be re quired to observe the carcinogenic event (A). DeMaeyer and DeKaeyer-Guignard have shown in early studies that pre-treatment of rat fibroblasts with several carcinogens including benzo- (a) -pyrene and 3-methylcholan- threne resulted in reduced interferon production^ when the cultures were challenged with vaccinia virus (5,6) . Benzo(c)-pyrene, a rarely carcinogenic mutagen and an analog of benzo-(a)-pyrene, had no effect .on interferon Induction. He have recently extended these studies to a murine system and have shown that several other suspected carcinogens, includ ing Aflatoxin-B., 2-amlnofluorene, styrene oxide, and methyl methanesulfonate, all inhibited Interferon induction by New castle disease virus (NDV) in mouse embryo fibroblasts (7). Ethyl methanesulfonate, a rarely carcinogenic analog of methyl methanesulfonate, also had no effect on interferon induction. The present study represents an extension of the exam ination of the effects of carcinogens on the induction of interferon. Pro--treatment of mouse embryo fibroblasts with the postulated carcinogen chloroacetaldchyde resulted in CMA 003840 SECTION ft*: IN VITRO TESTING 591 inhibi.ci.on of interferon induction when the cultures were chnllanged with NDV. Treatment of the cultures with chloroacetic acid and chloroethanol, rarely carcinogenic analogs of chloracetaldehyde, did not significantly affect interferon induction. These data suggest that the inhibition of inter feron induction system may be useful as an assay for the car cinogenic potential of chemicals, following more extensive studies and confirmation of results. In addition, chloroacetaldehyde treatment inhibited interferon induction by poly-- riboinoslnic-polyribocytidylic acid (poly I:C), suggesting that carcinogen treatment does not disrupt the interaction of virus with the cell membrane resulting in inhibited inter feron production. Finally, application of reduced gluta thione to the cell cultures in conjunction with the carcino gen benzo-(a)-pyrene resulted in abrogation of the effects of the carcinogen on Interferon induction. These data suggest that a system exists in fibroblasts which can activate carcinogens in the interferon inhibition system. MATERIALS AND METHODS Mouse Embryo Fibroblast Cultures: C57B1/6 derived mice were bred and maintained in our laboratory. Fifteen to eighteen day old embryos were surgically removed from pregnant mice, minced, trypslnized in 0.25Z trypsin 1-300 (ICN Pharaceutl-- cals, Cleveland, OH) and then suspended in minimum essential medium (Grand Island Biological, Grand Island, NY) supple mented with 10Z fetal calf scrum, penicillin and streptomycin and glutamine. Second or third passage^culutres were used in all experiments and were plated in 25cm6 tissue culture flasks (Falcon Plastics, Oxnard CA). Chemicals: Chloroacetaldehyde, chloroethanol and chloroacetlc acid were generous gifts of Dr. John Wong, Department of Chemistry, University of Louisville. Polyriboinosinic and polyrlbocytldyllc acids were obtained from F-L Biochemicals, Milwaukee, WI. Benzo-(e) -pyrene was obtained from Aldrich Chemical Company, Milwaukee, WI. Reduced Clutathione was obtained from Calbiochem, La Jolla, CA. Interferon Induction: Mouse Type I interferon was' produced in fibroblasts with the Herts strain of Newcastle disease virus, as described elsewhere (8). After inactivation of residual Inducing virus by pH 2 treatment at 4 C for four days, the tissue culture supernatants were assayed for anti viral (interferon) activity. Polyriboinosinic anti polyribo- CMA 003841 inhibition of interferon induction when the cultures were chnllanged with NDV. Treatment of the cultures with chloroacetic acid and chloroethanol, rarely carcinogenic analogs of chloracetaldehyde, did not significantly affect interferon induction. These data suggest that the inhibition of inter feron induction system may be useful as an assay for the car cinogenic potential of chemicals, following more extensive studies and confirmation of results. In addition, c&Ioroacetaldehyde treatment inhibited interferon induction by polyriboinosiaic-polyribocytidylie acid (poly I:C), suggesting' that carcinogen treatment doss not disrupt the interaction of virus with the cell membrane resulting in inhibited inter feron production. Finally, application of reduced gluta-- thione to the cell cultures in conjunction with the careino-^ gen benzo-(o)-pyrene resulted in abrogation of the effects of the carcinogen on Interferon induction. These data suggest that a system exists in fibroblasts which can activate carcinogens in the interferon inhibition system. MATERIALS AND METHODS Mouse Embryo Fibroblast Cultures: C57B1/6 derived mice were bred and maintained in our laboratory. Fifteen to eighteen day old embryos were surgically removed from pregnant mice, minced, trypsinized in 0.25Z trypsin 1-300 (ICN Pharaceuticals, Cleveland, OH) and then suspended in minimum essential medium (Grand Island Biological, Grand Island, NT) supple mented with 10Z fetal calf scrum, penicillin and streptomycin and glutamine. Second or third passage2eulutres were used in sll experiments and were plated in 25cm tissue culture flasks (Falcon Plastics, Oxnard CA). Chemicals i Chloroacetaldehyde, chloroethanol and ehloroacetic add were generous gifts of Dr. John Vong, Department of Chemistry, University of Louisville. Polyriboinosinic and polyribocytidylie acids were obtained from P-L Biochomicals, Milwaukee, WI. Benzo-( e)-pyrene was obtained from Aldrich Chemical Company, Milwaukee, WI. Reduced Glutathione was obtained from Calbiochera, La Jolla, CA. Interferon Induction: Mouse Type I interferon was produced in fibroblasts with the Herts strain of Newcastle disease virus, as described elsewhere (8). After inactivation of residual inducing virus by pH 2 treatment at 4 C for four days, the tissue culture supernatants were assayed for anti viral (interferon) activity. Polyriholnosinic and polyribo- CMA 003842 592 INTERFERON: PROPERTIES AND CLINICAL USES cytidylic acid were completed to poly I:C by heating at A5C fox one hour. Type 1 interferon was induced with poly I:C by adding 50 yg of poly I:C to tissue cultures for 90 min., and' adding additional fresh medium to the mouse embryo fibroblast cultures. DEAE-Dextran was included with' the poly I:C to insure maximum of induction of interferon (9), After twenty-four hours of incubation, the culture supernat ants were harvested and assayed for antiviral activity. Interferon Assay: Interferon titers-were determined by plaque reduction on mouse L-929 cells using the Indiana strain of bovine vesicular stomatitis virus (10). The inter feron titer corresponded to the reciprocal of the highest dilution of test sample that reduced virus plaques by 502. One interferon unit in this assay equals 0.88 NIB-C-002-904511 reference units. RESULTS Chloroacetaldehyde, chloroethanol, and chloroacetic acid were solublized in dimethyl sulfoxide (DMSO) and diluted to appropriate concentrations in tissue culture medium. The chemicals were then added to different confluent monolayers of mouse embryo fibroblasts. Following twenty-four hours of incubation at 37C, the culture supernatants were removed and Interferon was Induced with NDV. The results shown In Table 1 indicate that only treatment with chloroacetaldehyde resulted in decreased Interferon production of 502 or greater. TABLE 1 EFFECT OF PRETREATMENT OF CELL CULTURES WITH CHLORO ACETALDEHYDE AND ITS ANALOGS ON INTERFERON INDUCTION BY NDV Treatment*_______________ Interferon Titer ________X Decrease NDV Only DMSO + NDV Chloroacetaldehyde + NDV Chloroacetic acid + NDV Chloroethanol + NDV 300 500 69 225 209 -- -- 772 252 302 *A11 chemicals were applied at a concentration of 0.005 pM CMA 003843 SECTION IV: IN VITRO TESTING 593 Since the interferon assay has an innate two-fold vari ability due to the biological nature of the assay (11), only differences of 50% or greater were considered to demonstrate an effect of a carcinogen on interferon induction. When chloroacetaldehyde, chloracetic acid and chloroethanol were applied to confluent monolayers and interferon was induced with poly IsC, inhibition of interferon induction of 50% or greater was again only observed in- chloroacetaldehyde treated cultures (Table 2). Chloroacetic acid and chloroethanol had much less of an effect on the induction of Interferon. TABLE 2 EFFECT OF PRETREATMENT OF CELL CULTURES WITH CIIL0R0ACETALDEHYDE AND ITS. ANALOGS ON INDUCTION OF INTERFERON BY POLY IsC Treatment* Poly IsC only DMSO + poly IsC Chloroacetaldehyde + poly IsC Chloroacetic add + poly IsC Chloroethanol 4- poly IsC Interferon Titer 430 385 187 268 304 Z Decrease -- 10Z 57Z 37Z 29Z *A11 chemicals were applied at a concentration of 0.005 pm. Benzo-(a)-pyrene was solublized in DMSO and applied to confluent monolayers of mouse embryo fibroblasts at a con centration of 0.5 pm. As has been previously reported for induction of Interferon by NDV (7), this treatment resulted in a significant drop in the titer of interferon induced by poly IsC (Table 3). Addition of reduced glutathione with the benzo-( ci)--pyrene resulted in abrogation, at least in part, of the inhibitory effects of bcnzo-(cd-pyrene on interferon induction by poly IsC (Table 3). Glutathione itself had minimal If any effect on the induction of inter feron (Table 3). 003844 CMA SECTION IV: IN VITRO TESTING 593 Since Che interferon essay has an innate two-fold vari ability doe to the biological nature of the assay (11), only differences of 50% or greater were considered to demonstrate an effect of a carcinogen on interferon induction. When chloroacetaldehyde, chloracetic acid and chloroethanol were applied to confluent monolayers and interferon was induced with poly I:Ct inhibition of interferon induction of 50% or greater was again only observed in'chloroacetaldehyde treated cultures (Table 2). Chloroacetic acid and chloroethanol had much less of an effect on the induction of interferon. TABLE 2 EFFECT OF PRETREAIMENT OF CELL CULTURES WITH CIIL0R0ACETALDEHYDE AND ITS. ANALOGS ON INDUCTION OF INTERFERON BY POLY I:C Treatment* Poly I:C only DMSO + poly I:C Chloroacetaldehyde + poly I:C Chloroacetic add + poly I:C Chloroethanol + poly I:C Interferon Titer 430 385 187 268 304 Z Decrease -- 10Z 57Z 37Z 29Z *A11 chemicals were applied at a concentration of 0.005 pm. Benzo-(o)-pyrene was solublized in DMSO and applied to confluent monolayers of mouse embryo fibroblasts at a con centration of 0.5 pm. As has been previously reported for induction of Interferon by NDV (7), this treatment resulted in a significant drop in the titer of interferon induced by poly I:C (Table 3). Addition of reduced glutathione with the benzo-( ci)-pyrene resulted in abrogation, at least in part, of the Inhibitory effects of benzo-(cd-pyrene on interferon induction by poly I:C (Table 3). Glutathione itself had minimal if any effect on the induction of inter feron (Tabic 3). CMA 003845 594 INTERFERON: PROPERTIES AND CUNIC/U. USES TABLE 3 EFFECT OF CONCOMITANT ADDITION OF GLUTATHIONE AND BENZO-(a)-PYRENE ON INTERFERON INDUCTION BY POLY I;C Treatment Poly I:C only Benzo-(a)-pyrene + poly I:C. Benzo-(a) -pyrene + 0.1 pm glutathione + .poly I:C Behzo-(cj)-pyrene + 0.01 pm glutathione + poly X:C 0.1 pm glutathione + poly I:C 0.01 |im glutathione + poly I:C Interferon Titer 266 76 181 150 347 196 * X Decrease -- 71Z * 31Z 43Z 26Z Addition of glutathione to cultures ot mouse embryo fibro blasts did not result in the induction of detectable level of Interferon. DISCUSSION The induction of Type I interferon has been shown tp be inhibited by several carcinogenic chemicals (5-7). Of par ticular interest is the observation that when several pairs of highly carcinogenic chemicals and their rarely or noncarcinogenlc analogs, e.g. benxo-(o)-pyrene and benzo-(e)pyrene, ethyl methanesulfonate and methyl' methanesulfonate were tested, only application of the proven carcinogen resulted in the inhibition of interferon induction (5-7). We have now extended this observation to include the presumed carcinogen chloroacetaldehyde, and its rarely carcinogenic analogs chloro acetic acid and chloroethanol. After many additional pairs of carcinogens and analogs arc tested in the future, the inhibition of interferon induction by chemicals may prove useful in the screening of chemicals for carcino genic potential. i CMA 003846 SECTION IV: IN VITRO TESTING 595 Dimethyl sulfoxide has previously been shown to have a non-statislically significant minimal effect on the induction of interferon by NDV (7). In the present study, no effect of DMSO on interferon induction was observed. The mechanism of the inhibition of interferon induction was also examined in the present study. Previous work has shown that pre-treatment of rat fibroblasts with carcinogens did not inhibit the plaquing efficiency of vaccinia virus (5). Since poly I:C induction of Interferon is also 'inhibited by pre-treatment of fibroblasts with carcinogens. It is not likely that the carcinogen treatment is affecting the binding of viruses to the cell membrane or the budding, of viruses. Many carcinogens must be activated by microsomal oxidases in order to form final active products before an effect in bacterial or mammalian systems can be observed (12). Benzo(o)-pyrene is included among these chemicals. Reduced glutathione can trap these active products and prevent the occurence of a carcinogenic event (12). Application of reduced glutathione to the inhibition of interferon induction | system resulted in at least partial abrogation of the effects of benzo-(a)-pyrene on interferon Induction. These data suggest that fibroblasts must contain the type of activation system for carcinogens that renders them potent for the inhibition of interferon induction. REFERENCES 1. Ames, B.N., Durstan, W.E., and Yamasaki, E. Carcinogensare mutagens: A simple test system combining liver homogenates for activation and bacteria for assay. Proc. Natl. Acad. Set., 70:2281-2285, 1973. 2. Moreau, D., Balione, A., and Devoret R. Prophage lambda induction in Escherichia coli KI2 envA envB: A highly sensitive test for potential carcinogens. Proc. Natl. Acad. Scl., 73:3700-3704, 1976. 3. Streips, U.N., Laumbach, A.D., and Yasbin, R.B. Bacillus subtilus assays for mutation and DNA repair. In, Bacterial Mutation Monitors for Active Metabolites of Chemical Carcinogens; cd. Felkner, I.C. Marcel Dekkcr, New York, New York. In Press, 1979. ^ 003847 SECTION IV: If) VITHO TcSTlr.'G C25 Dimethyl sulfoxide has previously been shown to havt* a r.ca-sleiisiically si^ai fioarc niai-al nr feet on the iw-'oe t 1 on of interferon by NDV (7). In the present study, no effect of DMSO on interferon induetion was observed. The mechanism of the inhibition of interferon induction was also examined in the present study. Previous work has shown that pre-treatment of rat fibroblasts with carcinogens did not inhibit the plaquing efficiency of vaccinia virus (5). Since poly I:C induction of interferon is also 'inhibited by pre-treatment of fibroblasts with carcinogens, it is not likely that the carcinogen treatment is affecting the binding of viruses to the cell membrane or the budding of viruses. Many carcinogens must be activated by microsomal oxidases in order to form final active products before an effect in bacterial or mammalian systems can be observed (12). Benzo l-pyrene is included among these chemicals. Reduced glutathione can trap these active products and prevent the occurence of a carcinogenic event (12). Application of reduced glutathione to the inhibition of interferon induction system resulted in at least partial abrogation of the effects of benzo-(a)-pyrene on interferon induction. These data suggest that fibroblasts must contain the type of activation system for carcinogens that renders them potent for the inhibition of Interferon induction. REFERENCES1 1. Ames, B.N., Durstan, W.E., and Yamasaki, E. Carcinogens are mutagens: A simple test system combining liver homogenates for activation and bacteria for assay. Proc. Natl. Acad. Scl.. 70:2281-2285, 1973. 2. Moreau, D., Balione, A., and Devoret R. Prophage lambda induction In Escherichia coli K12 envA envB: A higlily sensitive test for potential carcinogens. Proc. Natl. Acad. Scl.. 73:3700-3704, 1976. 3. Streips, U.N., Laurabach, A.D., and Yashin, R.E. Bacillus subtilus assays for mutation and DNA repair. In, Bacterial Mutation Monitors for Active Metabolites of Chemical Carcinogens; cd. Felkner, l.^C. Marcel Dekkcr, New York, New York. In Press, 1979. CMA 003848 596 INTERFERON: PROPERTIES AND CLINICAL USES t* A. Ames, B.N. Identifying environmental chemicals causing mutations and cancer. Science, 204:587-593. 1979. 5. DeMaeyer-Guignard, J., and DeMaeyer, E. Effect of car cinogenic and noncarcinogenic hydrocarbons on interferon synthesis and virus plaque development. J. Natl. Cancer Inst., 34:265-276, 1965. | 6. DeMaeyer, E., and De-Maeyer-Guignard, J. ' Inhibition by 3-methylcholanthrene of interferon formation in rat embryo cells infected with Sindbis virus. J. Natl. Cancer Inst,. 32:1313-1321, 1964. 7. Barnes, M.C., Streips, D.N./and Sonnenfeld, G. Inhibit ion of interferon induction: Differentiates among chemi cals on the basis of carcinogenic potential. Submitted for publication, 1979. 8. Sonnenfeld, 6., Handel, A.D., and Merlgan, T.C. The immunosuppressive effect of Type II mouse Interferon on antibody production. Cell. Immunol.. 34:193-206. 1977.* 9. Platinini, F., Caritagalll, P., Gagnoni, S., and Rita, G. Effect of DEAE-dextran on production of interferon by . synthetic double-stranded RNA in L cell cultures. Prbc. Soc. Exp. Biol. Med.. 128:708-711, 1968. 10. Brodeur, B.R., Weinstein, Y., Melmon, K.L., and Merlgan, T.C. Reciprocal changes in interferon production and lnmnme responses of mouse spleen cells fractionated over columns of lnsolublized conjugates of histamine. Cell. Imwmol.. 29:363-372.. 1977.11 11. Lockhart, R.Z. Criteria for acceptance of a viral inhibitor as an interferon and a general description of the biological properties of known Interferons. In, Interferon and Interferon Inducers; ed., Pinter, N.B. American Elsevier, New York, New York, pp 11-28, 1973. 12. Chasseaud, L.P. Glutathione (reduced) and glutathione S-txansferase blocks the carcinogenic event by trapping the ultimate metabolites. Adv. Cancer Res.. 29:176-255, 1979. CMA 003849 - SECTION IV: IN VITRO TESTING 597 DISCUSSION UNKNOWN: Do you think that this effect is restricted only to these groups of carcinogens and works also with aflatexins? SONNENFELD: It does work with aflatoxin* We have shown it with aflatoxin previously as well and there has just been a report published using aflatoxins and the B2 aflatoxin. UNKNOWN: And what about tumor promotors? SONNENFELD: There have been reports using coal dust and also asbestos fibers as well, that 'there was reduced induc tion of interferon. The alfatoxin story is very interesting because these investigators at Western Virginia University used four different types of aflatoxin. One type being more potent carcinogen than the other and they found that the de gree of inhibition of induction was related to the potency of the type of aflatoxin that they used. DEGRE: Do you have any idea whether other cell types could be used for the same tests? SONNENFELD: We have used fibroblast. DeMeyer has used fibroblasts. There has been one study in vitro. I cannot remember the chemical right off hand, but if mice were treated with this chemical they could not produce in vivo serum Inter feron. I do not know about other things* Of course, we would like to get into studies with Type II interferon as well in lymphoid cells but to this point this type of work has only been done using fibroblast. CROB: You might have mentioned it but could you please repeat what the pretreatment schedule was before interferon induction. SONNENFELD: What we would do was treat those cells with the carcinogens for twenty-four hours and then wash and make sure that there was no toxicity of the carcinogen. UNKNOWN: Jay, just to support you, the emulsifiers that we used severely reduced the ability of cells to pro duce interferon. If you use them in very small concentra tions in leukocyte culture, they cause chromosome breaks to a substantial degree higher than the normal controls. CMA 003850 SECTION IV: IN VITRO TESTING 597 UNKNOWN: Do you think that this efleet is restricted only to these groups of carcinogens and works also with aflatexins? SONNENFELD: It does work with aflatoxin. We have shown it with aflatoxin previously as well and there has just been a report published using aflatoxins and the El aflatoxin. UNKNOWN: And what about tumor promotors? SONNENFELD: There have been reports using coal dust and also asbestos fibers as well, that 'there was reduced induc tion of interferon. The alfatoxin story is very interesting because these investigators at Western Virginia University used four different types of aflatoxin. One type being more potent carcinogen than the other and they found that the de gree of inhibition of induction was related to the potency of the type of aflatoxin that they used. DEGKE: Do you have any idea whether other cell types could be used for the same tests? SONNENFELD: We have used fibroblast. DeMsyer has used fibroblasts. There has been one study in vitro, I cannot remember the chemical right off hand, hut if mice were treated with this chemical they could not produce in vivo serum inters feron. I do not know about other things. Of course, ve would like to get into studies with Type II interferon as well in lymphoid cells but to this point this type of work has only been done using fibroblast. GRQB: You might have mentioned it but could you please repeat what the pretreatment schedule was before interferon induction. SONNENFELD: What we would do was treat those cells with the carcinogens for twenty-four hours and then wash and make sure that there wes no toxicity of the carcinogen. UNKNOWN: Jay,, just to support you, the emulsifiers that we used severely reduced the ability of cells to pro duce interferon. If you use them in very small concentra tions in leukocyte culture, they cause chromosome breaks to a substantial degree higher than the normal controls. CMA 003851 598 INTERFERON: PROPERTIES AND CLINICAL USES * SONNENFELD: Which might suggest some carcinogenic potential of the material. UNKNOWN: Yes, I couldn't agree more on the need of shifting from a prokaryotic system to an eukaryotic system and 1 think that your approach is particularly'exciting and promising, especially considering that you can use a human cell. SONNENFELD: Yes, we have begun studies along those lines- CMA 003852 Oncology 38: VS-lOl (1981) Effect of Carcinogens and Analogs on Interferon Induction Mary Carol Barnes, Uldis N. Streips and Gerald Sonnenfeld Department of Microbiology and Immunology, University of Louisville, School of Medicine, Louisville, Ky. Key Words. Interferon Carcinogen inhibition I Abstract. Pretreatment of mouse embryo fibroblasts with several chemicals, including 7,12-dimethylbenz-(a)anthracene, 2-aminofluorene, afiatoxin Blt benzo-(a)-pyrene, styrene oxide, and the No. 4 fraction of tobacco smoke condensate, resulted in severely reduced production of interferon when the cells were challenged with Newcastle disease virus. All of the above chemicals are proven or strongly suggested carcinogens. When the analogs methyl methanesulfonate, a potent carcinogen, and ethyl methanesulfonate, a weak carcinogen, were applied to cells, interferon induction was only inhibited by the methyl methanesulfonate. Therefore, carcinogens may directly inhibit the induction of interferon by Newcastle disease virus. Introduction Interferon is now being used in several clinical trials for the treatment of various types of tumors, including non-HodgkinVtype lymphomas [1], Therefore, it is of interest to determine the type of interactions that exist between interferon and tumor-inducing substances, such as carcinogens. Early experiments by DeMaeyer and DeMaeyerGuignard [2, 3} suggested that application of known chemical carcinogens, such as benzo-(a)-pyrene, to rat embryo fibroblast cultures prior to challenge with vac cinia virus resulted in inhibition of the induction of interferon by the virus. An additional finding of sig nificance was that benzo-(e)-pyrene, a weak carcino gen, had no effect on the induction of interferon. These data suggested that carcinogens might have a specific inhibitory effect on the induction of interferon. The results of the present study confirm and extend these observations. Several known and suspected po tent carcinogens, including methyl methanesulfonate (MMS), markedly inhibited the induction of interferon by Newcastle disease virus (NDV) in mouse embryo fibroblasts. The rarely or weakly carcinogenic analog of MMS, ethyl methanesulfonate (EMS) [4, 5] had little or no effect on the induction of interferon by NDV. Therefore, the new data suggest that interferon induc tion is inhibited by pretreatment of cell cultures with an extended group of potent carcinogens. Materials and Methods Mouse Embryo Fibroblast Cultures. B6 H-2* Iyt2.1 mice, a gift of Thomas Huff and Samuel Wellhausen, Department of Microbiology and Immunology, University of Louisville, Ky., were bred and main tained in our laboratory. IS- to 18-day-old embryos were surgically removed from mothers, minced, trypsinized in 0.25% trypsin 1-300 (Pharmaceuticals, Cleveland. Ohio) and then suspended in Gtbco minimal essential medium (MEM) with 10% fetal calf serum. Second or third passage cultures were used in all experiments and were plated in Falcon 25 cm' tissue culture flask*, and were used immediately upon reaching eonfluency. Chemicals. Afiatoxin B,, 2'aminofluorene,benzo-(a)-pyrene, EMS and MMS were obtained from Aldrich Chemical Co., Milwaukee. Wise. 7,12-Dimethylbenz-(f< (-anthracene and No. 4 fraction of to bacco smoke condensate were received from the Kentucky Tobacco Health and Research Institute. Lexington, Ky. Dimcthylsulfoxide (DMSO) was obtained from J.T. Baker Chemical Co.. IMulhpshurg. N.J. Styrene oxide was kindly provided by Dr. John Wong, Depart ment of Chemistry. University of Louisville, Ky. Interferon Production. Mouse type I interferon was produced in fibroblasts with the Herts strain of NDV, as described elsewhere (6J. CMA 003853 Carcinogen* anil Interferon 99 After inactivation of residual inducing virus by ptl 2 treatment at 4C for 4 days, the tissue culture supernatant fluid was assayed for interferon activity. Interferon Assay Interferon titers were determined by plaque reduction on mouse L-929 cells using Indiana strain bovine vesicular stomatitis virus. The interferon titer corresponded to the reciprocal of the highest dilution of test sample that reduced virus plaques by 50%. One interferon unit equals 0,88 NIH HG-002-904-511 reference units [6|. Results Table I. Effect of carcinogens on interferon production Treatment of fibroblasts Antiviral titer % decrease from control Newcastle disease virus 7.12-Dimethylbenz-(a)anthracene + NDV 39.0ji Af 3.9//M 2-Aminofluorene + NDV 0.050/iM 0.005 1.000 100 86 21 247 90 91 98 81 Viability of Mouse Embryo Fibroblast Cultures Mouse embryo fibroblasts were allowed to grow into a confluent monolayer. After reaching confluency, dif ferent cultures were treated for 24 h with several con centrations of the chemicals used in this study. These chemicals included 7,12-dimethylbenz-(a)-anthracene, 2-aminofluorene, aflatoxin B, No. 4 fraction of tobac co smoke condensate, benzo-(a)-pyrene, MMS, EMS, DMSO and ethanol. Immediately after removal of the chemicals, the fibroblast cultures were stained with 2 x neutral red, a vital stain. All of the cultures that were treated with the dosages of chemicals that were used in the study were >95% viable by staining criteria. Addi tional cultures that had chemical-containing medium replaced with fresh MEM remained viable for at least another 24 h. Effect of Known Carcinogens on Interferon Induction The known carcinogens 2-aminofluorenc and 7,12dimethyibenz-(a)-anthracene were diluted in MEM and added to confluent fibroblast cultures. After 24 h of incubation, the culture supernatants were removed. The cultures were then challenged with NDV for 24 h and the supernatants were harvested, pH 2 treated, and then assayed for interferon activity. Reductions in inter feron titers 90% were obtained when compared to noncarcinogen treated controls (table I). Several different dosages of chemicals were initially used to determine the dosage for optimum effect without affecting via bility of the cultures, and the data in table I represent the optimal dosages. Effects of Several Chemicals on Interferon Induction Mouse embryo fibroblasts were treated with several different dosages of various chemicals before attempted induction of interferon with NDV. Aflatoxin Blt No. 4 fraction of tobacco smoke condensate, benzo-(a)-pyrene, and styrene oxide treatments all inhibited the induction of interferon by 90% or greater (table II). Table II. Blind study of the effect of various chemicals on inter feron induction Chemical treatment1 Interferon titer SE % decrease Pnov* from control None (NDV control) Aflatoxin B, 0.05 nM No. 4 fraction, tobacco smoke1 condensate. lOx Benzo-a-pyrene, 0.05 pM Styrene oxide, 0.05 /iM 1,456 455 105 82 89 20 122 74 149 60 _ 93 95 92 90 --. <0.05 <0.05 <0.05 <0.05 1 Mouse embryo fibroblasts were pretreated with the appropriate chemical for 24 h. Supernatant* were removed, cells washed and NOV applied for 24 h. Cell supernatants were then harvested and assayed for antiviral activity. 1 Statistical analysis was performed by means of a Student's T analysis. 1 Arbitrary dilutions of the No. 4 fraction of tobacco smoke con densate were used. Table 111. Blind study of the effect of a weak and a potent car cinogen on interferon production Chemical treatment* Interferon titer SE % decrease1 P from control None (NDV control) Ethyl methancsulfonate 0.05 nM Methyl methanesulfonate 0.05/sAf 1,456 455 1,680 294 137 82 -- 91 >0.5 <0.05 1 Experimental protocol as in table II. 1 EMS and MMS were dissolved in ethanol only. Table 11 represents the effects of optimum dosages of the chemicals used. DMSO was used as a solvent for all of the above chemicals. Pretreatment of fibroblast cultures with DMSO resulted in a smalt (0--40%), not 003854 100 Barnes/Streips/Sonnenfeld statistically significant decrease in the titers of inter feron induced by NDV. EMS and MMS were solubilized in ethanol. Pre treatment of mouse fibroblasts with ethanol had no effect on the titers of interferon induced by NDV. Pre treatment of the cell cultures with EMS had a minimal effect on the induction of interferon by NDV, while MMS pretreatment severely decreased the titer of inter feron induced by NDV (table 111), Discussion The results of the present study indicate that several well-known carcinogens, including 7,12-dimethylbenz(a)-anthracene, benzo-(a)-pyrene, 2-aminofluorene, aflatoxin B,, and the No. 4 fraction of tobacco smoke condensate, all inhibited the induction of interferon by NDV. Styrene oxide, an important industrial chemical, which was positive in the Ames assay, but negative in all Bacillus assays done to date, also inhibited the in duction of interferon by NDV. Further studies of the carcinogenic potential of styrene oxide may be required. DMSO has been used in clinical trials as a drug for the treatment of arthritis. However, these trials have been suspended due to the teratogenic potential of the chemical (7, 8]. It is of interest to note that pretreat ment of cells with DMSO did not result in the signifi cant inhibition of interferon induction by NDV. EMS and MMS are closely related analogs which differ with respect to carcinogenic potential as deter mined by tumor induction [4, S). MMS is a highly car cinogenic mutagen, while EMS is a rare or weakly car cinogenic mutagen. In the present study, MMS strongly inhibited the induction of interferon by NDV, while EMS had no significant effect. In a recent study, fibro blasts were pretreated with four different forms of aflatoxin [9]. The degree of inhibition of interferon in duction correlated with the proven carcinogenic poten tial of the aflatoxin forms, i.e. the more carcinogenic the form of aflatoxin, the greater the inhibition of inter feron induction. In view of these findings and the pre viously reported discrimination between bcnzo-(a)pyrene and benzo-(f)-pyrene (3] and the presently observed discrimination between EMS and MMS, car cinogens may directly inhibit the induction of inter feron. The mechanism by which carcinogens inhibit the in duction of interferon is not known. A toxic effect of the carcinogens to the cell cultures does not seem likely. since the results of this study indicate that the target cells are alive when they fail to produce interfemR However, a general effect of the carcinogens on RNA, or protein synthesis,and differential dosage ef fects of potent and weak carcinogens cannot be ruled out and are the subject of future studies. Other possible mechanisms include alteration of the cell membrane by the carcinogen resulting in reduced interaction of virus with the cells. Recent studies show ing that pretreatment of cells with carcinogens inhibited the induction of interferon by poly-inosinic-poly-cytidylic acid, a nonviral inducer of interferon [10]. In view of those studies and the earlier finding that carcinogen pretreatment does not inhibit the replication of the viruses used to induce interferon [2], it is unlikely that virus-membrane interactions are severely disrupted by carcinogen treatment of cells. Another possible mechanism could involve induction of a protease by the carcinogen in a manner analogous to bacterial `SOS' repair systems [11], This protease could degrade induced interferon, or interfere with the control mechanisms for interferon induction. The true nature of the proposed effect of carcinogens on inter feron induction is the subject of extended studies. The implications of these studies toward an under standing of the mechanisms of carcinogenesis are still obscure. It is possible to speculate that there man^a balance between interferon levels and oncogenea^pd that balance may be upset by the action of a carcino gen. In addition, if further studies continue to indicate a differential effect on interferon induction by strong and weak carcinogens, the effect of a chemical on inter feron induction may have some use in determination of the carcinogenic potential of that chemical. Acknowledgements We thank Dr. R.E. Yashin of the Pennsylvania State University for providing unpublished data and helpful suggestions. This work was supported by grams from the Manufacturing Chemists' Associa tion, the American Cancer Society No. 1N-U1C. and the National Aeronautics and Space Administration under Interchange NCA2OR400-90I. M.C. B. was the recipient of a Graduate Student Re search Grant from the Graduate School of the University of Louis ville. References 1 Merigan, T.C.; Sikora, K.; Breeden. J.H.; Levy, R.. and Rosen berg, S. A,; Preliminary observations on the effect of human leuko CMA 003855 KO 9 True' 'i:u:"!ucally significant decrease in the titers of inter;_r:n nduccd by NDV. EMS and MMS were solubilized in ethanol. Prermcnt of mouse fibroblasts with ethanol had no effect on the tii rs of interferon induced by NDV. Prctreatment of th cell cultures with EMS had a minimal effect on the induction of interferon by NDV, while MMS pretreatment severely decreased the titer of inter feron induced by NDV (table III). Discussion The results of the present study indicate that several well-known carcinogens, including 7,12-dimethylbenz(ct)-anthracene, benzo-(a)-pyrene, 2-aminotluorene, aflatoxin B,, and the No. 4 fraction of tobacco smoke condensate, all inhibited the induction of interferon by NDV. Styrene oxide, an important industrial chemical, which was positive in the Ames assay, but negative in all Bacillus assays done to date, also inhibited the in duction of interferon by NDV. Further studies of the carcinogenic potential of styrene oxide may be required. DMSO has been used in clinical trials as a drug for the treatment of arthritis. However, these trials have been suspended due to the teratogenic potential of the chemical (7, 8]. It is of interest to note that pretreat ment of cells with DMSO did not result in the signifi cant inhibition of interferon induction by NDV. EMS and MMS are closely related analogs which differ with respect to carcinogenic potential as deter mined by tumor induction (4, 5). MMS is a highly car cinogenic mutagen, while EMS is a rare or weakly car cinogenic mutagen. In the present study, MMS strongly inhibited the induction of interferon by NDV, while EMS had no significant effect. In a recent study, fibro blasts were pretreated with four different forms of aflatoxin [9J. The degree of inhibition of interferon in duction correlated with the proven carcinogenic poten tial of the aflatoxin forms, i.e. the more carcinogenic the form of aflatoxin, the greater the inhibition of inter feron induction. In view of these findings and the pre viously reported discrimination between benzo-(a)pyrene and benzo-(f)-pyrene [3| and the presently observed discrimination between EMS and MMS, car cinogens may directly inhibit the induction of inter feron. The mechanism by which carcinogens inhibit the in duction of interferon is not known. A toxic effect of the carcinogens to the cell cultures does not seem likely. since the results of this study indicate that the ir,e: ceils ire .lihe when they fail to produce mtertcren. How user, a general affect of the carcmcgens *>.t DN.\. RNA. or protein synthesis and ditferential dosage ef fects of potent and weak carcinogens cannot ue ruled out and are the subject of future studies. Other possible mechanisms include alteration of the cell membrane by the carcinogen resulting in reduced interaction of virus with the cells. Recent studies show ing that pretreatment of cells with carcinogens inhibited the induction of interferon by poly-inosinic-poly-cytidylic acid, a nonviral inducer of interferon [10], In view of those studies and the earlier finding that carcinogen pretreatment does not inhibit the replication of the viruses used to induce interferon [2], it is unlikely that virus-membrane interactions are severely disrupted by carcinogen treatment of cells. Another possible mechanism could involve induction of a protease by the carcinogen in a manner analogous to bacterial `SOS' repair systems [11]. This protease could degrade induced interferon, or interfere with the control mechanisms for interferon induction. The true nature of the proposed effect of carcinogens on inter feron induction is the subject of extended studies. The implications of these studies toward an under standing of the mechanisms of carcinogenesis are still obscure. It is possible to speculate that there may be a balance between interferon levels and oncogenesis and that balance may be upset by the action of a carcino gen. In addition, if further studies continue to indicate a differential effect on interferon induction by strong and weak carcinogens, the effect of a chemical on inter feron induction may have some use in determination of the carcinogenic potential of that chemical. Acknowledgements We thank Dr. ft. . Yatbm of the Pennsylvania State University for providing unpublished data and helpful suggestions. This work was supported by grams from the Manufacturing Chemists' Associa tion, the American Cancer Society No. lN-ltlC. and (he National Aeronautics and Space Administration under Interchange NCA2OR400-ycl. M.C.B. was the recipient of a Graduate Student Re search Grant from the Graduate School of the Umtcrviy of Louis ville. References 1 Mengan. T.C.; Sikora. K.; Breeden. J.H.; Levy, R., and Rosen berg. S. A.: Preliminary observations on the effect of human leuko- CMA 003856 Carcinogens anti Interferon 101 eyre interferon in non-Hodgkin's lymphoma. New F.ngl J. Med 299: 1449-1453(1979). 2 DeMneyer. E. and DeMacyer-Guignard. J.: Inhibition by 3-methylehi'lanthrcne of interferon formation in rat embryo cells infected with Sindbis virus, i. natn. C.mcer Inst. 12 1317-1321 (1964), t DcM.icvcr-Guignard. J. and De.Mjever b . Utect ol carcinogenic and noncarcinogeme hydrocarbons on interferon synthesis and \irus plaque development J nain Cancer Inst 34, 265-276 (1065), 4 Roe. F. L C.; Mitchley, B C. V,, and Walters. M.. Tests for car cinogenesis using newborn mice: 1,2-benzanthracene. 2-naphthylamine. 2-naphthylhydroxylamine. and ethyl methanesulfonate. Br J. Cancer/7, 255-260 (1963). 5 Casto. B.C.; Janosko, N.. and DiPaolo, J A.: Development of a focus assay for transformation of hamster cells in vitro by chemical carcinogens. Cancer Res. 37: 3508--3515 (1977). 6 Sonnenfeld. G.; Mandel, A.D., and Mengan. T.C.. The immuno suppressive effect of type II mouse interferon on antibody pro duction Cell. Immunol. 34; 193-206 (1977). 7 Caujolle, F M.E.;Caujolle.D,H.;Cros,S.B.,andCalvet,M.-M.J.. Limits of toxic and teratogenic tolerance of dimethylsulfoxide Ann N. Y. Acad. Sci./4/.- 110-126(1967). 8 Rubin, L F : Toxicity of dimethylsulfoxide, alone and in combina tion. Ann N.Y. Acad. Sci 24J. 98-103 (1975). 9 Hahon. N.; Booth. LA., and Stewart. J.D.: Aflatoxin inhibition ot viral interferon induction, Antimicrob. AgentsChemother, 16: 277(1979) ill 'uinncnicld. G , Barnes. M.C., Schooler, J., and Strops. U N.: Inhibition of interferon induction as a screen for the carcinogenic potential of chemicals, in Khan. Hill, Dorn, Interferon- properties and clinical uses. pp. 589-598 (Wadlcy Inst. Molec. Med.. Dallas 1980). 11 Wiikin, E. M.: Ultraviolet mutagenesis and inducible DNA repair in Escherichia colt. Bact. Rev. 40: 869-907 (1976). Dr. Gerald Sonnenfeld, Department of Microbiology and Immunology, University of Louisville School of Medicine, Health Sciences Center, Louisville, KY 40292 (USA) CMA 003857 CMA 003858 13 IMMUNOPATHOLOGIC OBSERVATIONS IN LIVER ANGIOSARCOMA Enrique Espinosa, M.D. Department of Pathology University of Louisville School of Medicine Louisville, Kentucky I. INTRODUCTION In hepatic fibrosis and angiosarcoma associated with vinyl chloride exposure of industrial workers, manifestations of the disease could not be detected in most cases until the process was far advanced (1). Normal values of liver function tests were reported in a case with significant vinyl chloride hepa tic fibrosis (2), and only a small percentage of workers of a plant unit where seven cases of liver angiosarcoma were diag nosed had abnormal blood screening tests (3). Thus, conven tional liver function tests do not appear to be sensitive in dicators of vinyl chloride liver disease. Development of more sensitive methods for detecting the disease in early stages would be of great importance. An approach to this may be pro vided by immunologic studies. In such a study the question arises whether the fibrotic and angiosarcomatous livers contain antigens that are different from those present in normal tis sue and whether such changes could stimulate an immunologic response. The purpose of this study was to search for anti genic changes in the angiosarcomatous tissue and to test for possible presence of an antibody response in the host. II. PROCEDURES AND MATERIALS USED A. Patients' Sera and Tissue Specimens Serum samples from B.F. Goodrich Co. workers with histor ies of vinyl chloride exposure of several years included sam ples from two individuals with liver angiosarcoma, ten with liver dysfunction with fibrosis and ten with normal liver function tests. The patients with angiosarcoma died and the diagnosis was confirmed at autopsy and portions of tumor and neighboring liver tissues were obtained at autopsy. Patients with liver dysfunction with fibrosis included individuals with 927 CMA 003859 abnormalities in liver function tests and fibrosis detected at biopsy. Tissues were also obtained from" coroner's autopsies of healthy individuals a few hours after death by gunshot wounds. B. Tissue Extracts and Antisera Liver angiosarcoma and adjacent liver tissue and post-mor tem tissues considered to be normal were frozen and stored at -70aC until used. Portions were cut, thawed and homogenized in 2-3 volumes of distilled water in a Potter-Elvehjem grinder in an ice bath until a smooth suspension was obtained. After centrifugation at 20,000 x G for 30 min, the supernatant fluid containing the aqueous extract was lyophilized. Albino rab bits were immunized with the tumor or normal liver extracts in Freund's complete adjuvant and sera collected and stored fol lowing procedures detailed elsewhere (4). Reaction of these immune sera with human serum or plasma was eliminated by ab sorption with 100 mg of lyophilized, pooled normal human serum /ml antiserum. Antisera were routinely absorbed in this man ner prior to use. Additional absorption with tissue extracts was carried out with 100 mg lyophilized extract/ml antiserum. Absorptions followed a procedure described previously (5). C. Treatment of Tissue Extracts Enzymatic treatment of tissue extracts was carried out with Pronase and trypsin as previously described (6). Periodate oxidation was done according to Rajam et al. (7). Ammonium sulfate and cold ethanol fractionations were carried out as detailed (8). D. Immunodiffusion and Immunofluorescence Double immunodiffusion was carried out in 0.8% agarose in phosphate-buffered saline pH 7.2 (PBS) containing 0.1% sodium azide. Circular wells, 2 mm in diameter, 3 mm apart were used. Immunoelectrophoresis was performed according to Schei- . .degger (9) using 0.8% agarose in 0.02S M Veronal buffer at pH 82 In immunofluorescent studies cryostat sections of rat kid ney, stomach or intestine and liver (4 microns) were used as substrate for antimitochondrial, antismooth muscle and anti nuclear antibodies. Liver sections from rats exposed 4, 8 and 14 days to 1-2% vinyl chloride for 4 hr/day were also used. The sections were covered with dilutions of patients' sera for 45 min at room temperature, washed twice in PBS for 10 min and then covered with fluorescein conjugated IgG fraction of rab bit anti-human immunoglobulins serum (Cappel Laboratories, Inc.) for 45 min and washed as before prior to examination. Cryostat sections of liver angiosarcoma and liver tissue con sidered to be normal (4 microns) were washed twice in PBS for 10 min to wash off nonfixed immunoglobulins. After drying. 928 CMA 003860 sections were stained with fluorescein conjugated IgG frac tions of rabbit anti-human IgG serum and goat anti-human IgM serum (Cappel Laboratories, Inc.). Sections were washed as a-- bove and examined under the fluorescent microscope. Represen tative frozen sections were stained with hematoxylin and eosin to allow correlation between immunofluorescence and the histo logical findings. The antigenic preservation of the tissues was indicated by the demonstration of their staining by anti nuclear factor according to the indirect immunofluorescent procedure. E. Elution of Tumor-bound IgG The tumor and liver tissues were extracted five times with PBS to wash off nonfixed immunoglobulins and then extracted at pH 2.5 to release bound IgG according to a procedure applied in the elution of renal-bound antibody (10). F. Circulating Tissue Antigens Liver-specific antigen lsa (8), tissue antigens of wide organ distribution (4) and bile antigens (11) were tested in the patients' sera by immunodiffusion as described previously. III. RESULTS A. Angiosarcoma-related Antigen To test for presence of new antigens appearing in liver angiosarcoma, antiangiosarcoma serum was absorbed with human serum and liver extract and tested by immunodiffusion with ex tracts of both normal liver and angiosarcoma tumor at varying concentrations. This absorption eliminated all reactivity with liver extracts prepared from five normal individuals but not with the angiosarcoma extracts where one line of precipi tation remained (Fig. 1). This line of precipitation could still be seen after additional absorption of the antiserum with kidney extract. In contrast, absorption with the tumor extracts eliminated completely the angiosarcoma-related line of precipitation. Absorption with spleen and lung extracts also eliminated this line of precipitation. Thus, the antigen appeared to be of restricted tissue distribution and not an giosarcoma specific. The antigen was inactivated by trypsin and Pronase and thus appeared to be a protein or closely asso ciated to protein. Incubation of the tumor extracts for 1 hr at 4*C in citrate buffer, pH 2.5, resulted in inactivation of the antigen whereas incubation in phosphate buffer, pH 5.0, neutral or alkaline pH up to pH 10.0, did not affect it. The antigen was shown to be relatively thermolabile. - Incubation of the tumor extracts for 30 min in PBS at 25 and 56"C did not affect the antigen whereas incubation at 70C and higher 929 CMA 003861 serum (Cappel Laooratorlgs, Inc.). Sections were washed as abova and examined under the fluorescent microscope. Reoresentative frozen sections were stained with hematoxylin and eosin to allow correlation between immunofluorescence and the histo logical findings. The antigenic preservation of the tissues was indicated by the demonstration of their staining by anti nuclear factor according to the indirect immunofluorescent procedure. E. Elution of Tumor-bound IgG The tumor and liver tissues were extracted five times with PBS to wash off nonfixed immunoglobulins and then extracted at pH 2.5 to release oound IgG according to a procedure applied in the elution of renal-bound antibody (10). F. Circulating Tissue Antigens Liver-specific antigen LSA (8), tissue antigens of wide organ distribution (4) and bile antigens (11) were tested in the patients' sera by immunodiffusion as described previously. III. RESULTS A. Angiosarcoma-related Antigen To test for presence of new antigens appearing in liver angiosarcoma, antiangiosarcoma serum was absorbed with human serum and liver extract and tested by immunodiffusion with ex tracts of both normal liver and angiosarcoma tumor at varying concentrations. This absorption eliminated all reactivity with liver extracts prepared from five normal individuals but not with the angiosarcoma extracts where one line of precipi tation remained (Fig. 1). This line of precipitation could still be seen after additional absorption of the antiserum with kidney extract. In contrast, absorption with the tumor extracts eliminated completely the angiosarcoma-related line of precipitation. Absorption with spleen and lung extracts also eliminated this line of precipitation. Thus, the antigen appeared to be of restricted tissue distribution and not an giosarcoma specific. The antigen was inactivated by trypsin and Pronase and thus appeared to be a protein or closely asso ciated to protein. Incubation of the tumor extracts for 1 hr at 4*C in citrate buffer, pH 2.5, resulted in inactivation of the antigen whereas incubation in phosphate buffer, pH 5.0, neutral or alkaline pH up to pH 10.0, did not affect it. The antigen was shorn to be relatively thermolabile. Incubation of tha tumor extracts for 30 min in PBS at 25s and 56*0 did not affact the antigen whereas incubation at 70*C and higher 929 CMA 003862 completely inactivated it. The antigen precipitated mainly at 20-30% saturated ammonium sulfate and at ethanol concentra tions of 30-70% (Table I). B. Absence of a Normal Tissue Antigen in Angiosarcoma Antiliver serum absorbed with human serum gave several arcs of precipitation in immunoelectrophoresis with extracts of normal liver and angiosarcoma tissue (Fig. 2a). These lines could not be seen following additional absorption of the anti serum with normal liver. In contrast, absorption with liver angiosarcoma extract failed to eliminate one of the arcs of precipitation (Fig. 2b). Thus, the tissue antigen related to this arc of precipitation appeared to be absent in the angiosarcomatous tissue whereas the antigens corresponding to the other arcs were present. The absent antigen in angiosarcoma was shown to be present in kidney and lung extracts in addi tion to liver by absorption and direct immunodiffusion tests. Physicochemical characterization studies indicated this anti gen to be unaffected by Pronase and trypsin and inactivated by periodate treatment. The antigen was relatively thermostable withstanding incubation at 70C for 30 min in PBS. The anti gen was destroyed following incubation of the liver extract in citrate buffer at pH 2.S or lower for 1 hr at 4*C; incubation at pH 5.0 or higher (up to pH 10.0) did not affect it. This antigen precipitated over a wide range of ammonium sulfate and ethanol concentrations (Table I). C. Angiosarcoma-bound IgG IgG fluorescent staining appeared in a linear pattern in the peripheral portion of the tumor cells suggesting in vivo binding by the tumor of the immunoglobulin. This staining is illustrated in Fig. 3. Fig. 4 shows the angiosarcomatous cells surrounding irregular vascular spaces. Relatively coarse, linear fluorescence was also present along some hepa tic cords and strands of connective tissue. There was no ev idence of IgM. Control post-mortem liver tissue did not show any significant fluorescence of bound IgG. Staining for IgG of the tumor sections did not change after several washings at pH 7.2 indicating that the IgG was firmly bound to' the tumor. In contrast, sections showed marked diminution of staining after washing at pH 2.5. Elution of bound IgG from salineextracted tumor homogenates was thus attempted at acid pH. with the five successive saline extractions the amount of sa line soluble IgG gradually diminished to nondetectable levels; and at acid .pH bound IgG was released from the homogenate. Similar treatment of liver homogenates did not demonstrate presence of bound IgG (Table II). D. Circulating Autoantibodies and Tissue Antigens Serum autoantibodies to nuclei, mitochondria and smooth muscle were negative in all patients examined. In addition, 930 CMA 003863 serum from these patients did not show reactivity with liver from rats exposed to vinyl chloride. Liver-specific antigen LSA C8), bile antigens (11) and other tissue antigens (4) as sociated with liver damage were not detected in these pa tients . IV. DISCUSSION The immunologic characteristics of cancer have been under intense investigation during recent years, and antigenic dif ferences between normal and malignant tissue are considered to be fundamental factors in the immunologic approach to cancer therapy and diagnosis. Liver angiosarcomatous tissue was thus analyzed in this work for presence of neoantigens, normal tis sue antigens and tumor-bound immunoglobulins. Several normal tissue antigens were found by immunodiffusion to be present in the tumor, but one antigen of rather wide organ distribution was not detected. These findings are in agreement with obser vations in other tumors indicating that tumor cells contain many of the antigens of their original hosts and lack some normal tissue antigens. For example, immunohistochemical studies have shown the loss of kidney antigens in stilbestroland x-ray-induced kidney tumors (12), of skin antigen in 3methylcholanthrene-induced mouse squamous cell carcinoma (13) and of certain muscle antigens in 20-methylcholanthrene-induced rat rabdomyosarcoma (14). By far the most extensively studied class of tumors are the chemically induced hepatomata where deletion of liver antigens have been shown in tumors in duced with 4-dimethylaminoazobenzene (15, 16), diethylnitrosamine (15) and 2-acetomidofluorene in the rat (15, 17) and oaminoazotoluene in the mouse (18). In human carcinoma, loss of. antigens have been reported in squamous cell carcinoma (12, 19), loss o| tfie ABH blood group isoantigens in some ' solid tumors (20V 21) and of HL-A isoantigen in lymphoma (22). In addition, it has- been well documented that as cells trans form from a normal state to malignancy they may gain new anti genic specificities. Tumor-specific transplantation antigens have beeq/d^sjnonstrated in a number of experimentally induced tumors (23-26) as well as in spontaneous tumors in man (27-' l 29). In the present report immunodiffusion analyses of liver angiosarcoma and other human tissues with rabbit antiangiosar coma serum did not indicate the presence of a tumor-specific antigen but rather of an antigen found in lung and spleen but not in liver and kidney. This antigen is being further char acterized in our laboratory. Of particular interest is the demonstration of tumor-bound IgG by immunofluorescence and elution experiments. This find ing must however be interpreted with caution and should be confirmed in biopsy specimens. The tumor-bound IgG may repre sent specific antitumor antibody, antibody fixed by the tumor tissue "nonspecifically" or part of both. Further speculation is premature until it has been shown that the staihing pattern 931 CMA 003864 ser-^n from these pattents did not show reactivity with liver from rats exposed to vir.yl rhltride. Liver-soecific antigen L5A (3), btle antigens (11) and other tissue antigens (4) as sociated with liver damage were not detected in these pa tients . IV. DISCUSSION The immunologic characteristics of cancer have been under intense investigation during recent years, and antigenic dif ferences between normal and malignant tissue are considered to be fundamental factors in the immunologic approach to cancer therapy and diagnosis. Liver angiosarcomatous tissue was thus analyzed in this work for presence of neoantigens, normal tis sue antigens and tumor-bound immunoglobulins. Several normal tissue antigens were found by immunodiffusion to be present in the tumor, but one antigen of rather wide organ distribution was not detected. These findings are in agreement with obser vations in other tumors indicating that tumor cells contain many of the antigens of their original hosts and lack some normal tissue antigens. For example, immunohistochemical studies have shown the loss of kidn^^antigens in stilbestroland x-ray-induced kidney tumors (12*of skin antigen in 3methylcholanthrene-induced mouse squamous cell carcinoma (13) and of certain muscle antigens in 20-methyleholanthrene-indueed rat rabdomyosarcoma (14). By far the most extensively studied class of tumors are the chemically induced hepatomata where deletion of liver antigens have been shown in tumors in duced with 4-dimethylaminoazobenzene (IS, 16), diethylnitrosamine (15) and 2-acetomidofluorene in the rat (15, 17) and oaminoazotoluene in the mouse (18). In human carcinoma, loss of antigens have been reported in squamous cell carcinoma (12, 19), loss of t(ie ABH blood group isoantigens in some t solid tumors (20^ 21) and of HL-A isoantigen in lymphoma (22). In addition, it has been well documented that as cells trans form from a normal state to malignancy they may gain new anti genic specificities. Tumor-specific transplantation antigens have beer)/.demonstrated in a number of experimentally induced tumors (23-26) as well as in spontaneous tumors in man (27' 29). In the present report isnunodiffusion analyses of liver angiosarcoma and other human tissues with rabbit antiangiosar coma serum did not indicate the presence of a tumor-specific antigen but rather of an antigen found in lung and spleen but not in liver and kidney. This antigen is being further char acterized in our laboratory. Of particular interest is the demonstration of tumor-bound IgC by immunofluorescence and elution experiments. This find ing must however be interpreted with caution and should be confirmed in biopsy specimens. The tumor-bound IgG may repre sent specific antitumor antibody, antibody fixed by the tumor tissue "nonspecifically" or part of both. Further speculation is premature until it has been shown that the staining pattarn 931 CMA 003865 is due to the deposition of a specific antibody, that the elu ted antibody is specific or until the relevant antigen has been identified. Work is in progress to determine the precise significance of the finding of IgG in the tumor. V. SUMMARY Immunodiffusion analyses of human liver angiosarcoma asso ciated with vinyl chloride exposure indicated presence in the tumor of an antigen not detected in normal liver and kidney but found to be present in lung and spleen. This antigen was shown to be a protein, inactivated by Pronase and trypsin, relatively susceptible to heating and to acid pH and precipi tated mainly at 20-301 saturated ammonium sulfate and at 30701 ethanol concentrations. The tumor was shown to contain several antigenic constituents of normal tissue but one normal tissue antigen was not detected. This antigen was character ized as a substance unaffected by Pronase and trypsin and in activated by periodate. It was relatively thermostable, af fected by acid pH and precipitated over a wide range of ammon ium sulfate and ethanol concentrations. Tumor specimens ob tained at autopsy contained bound IgG as shown by immunofluor escence and elution experiments suggesting possible in vivo binding of IgG to the tumor. Acknowledgements: The author wishes to thank Drs. W. M. Christopherson, G. R. Schrodt and P. H. Carstens for fruitful discussion and advice about the histologic sections and Drs. C. Tamburro and L. Maak for providing serum samples and au topsy material. I also wish to thank Miss M. VanBraun and Mrs. V. Petrey for skilled technical assistance. This work was supported in part by a grant from B. F. Goodrich, Co. 932 CMA 003866 TABLE I. Angiosarcoma-related Antigen and Antigen Absent from the Tumor in Ammonium Sulfate and Ethanol Fractions A Presence of Fraction tested Angiosarcoma-related b antigen Antigen absent from c Angiosarcoma Ammonium sulfate: 0-20% saturation 20-30% saturation + ++ - 30-50% saturation - +++ 50-70% saturation - ++ Ethanol: 0-20% - ++ 20-30% - +++ 30-50% ++ +++ 50-70% d SN ++ " ++ + a +++, ++, + indicate strength of double diffusion reaction in dilution assay, b Detected in angiosarcoma fractions, c Detected in liver fractions, d SN * supernate of the 70% ethanol precipitation, dialyzed and lyophilized. CMA 003867 933 TABLE I. Angiosarcoma-related Antigen and Antigen Absent from the Tumor in Ammonium Sulfate and Ethanol Fractions a Presence of Fraction tested Angiosarcoma-related b antigen Antigen absent from c Angiosarcoma Ammonium sulfate: 0-20% saturation 20-30% saturation 30-50% saturation 50-70% saturation Ethanol: 0-201 20-30% 30-50% 50-70% d SN + ++ - ++ ++ + + ++ ++ +++ +++ + +++, ++, + indicate strength of double diffusion reaction in dilution assay, b Detected in angiosarcoma fractions, c Detected in liver fractions. d SN supernate of the 70% ethanol precipitation, dialyzed and lyophilized. 933 CMA 003868 TABLE II. igG in Saline and Acid Extracts of Angiosarcoma and Liver Tissues Height solid extracted from a Preparation tested 1 gm (wet weight) tissue Presence of IgG (mg) Angiosarcoma: Saline extract 1 Saline extract 2 29.8 8.6 +++ ++ Saline extract 3 Saline extract 4 Saline extract 5 Acid extract 6.2 5.6 6.1 6.1 + M- Liver s Saline extract 1 Saline extract 2 Saline extract 3 Saline extract 4 Saline extract 5 Acid extract 47.4 14.5 8.6 7.4 6.4 9.0 +++ *-++ + - a Tested by immunodiffusion at concentrations of the eluates ranging up to 2%. Present at concentrations 0.05-0.1* (+++); 0.2-0.5* (++); 1-2% (+); negative at 2* {-). 934 CMA 003869 FIG. 1. Demonstration of angiosarcoma-related antigen. Peripheral wells have 2-fold serial dilutions of liver angiosarcoma extract (a) and normal liver extract (b). Dilutions are clockwise and start at 100 mg/ml in the upper right well. Central wells in each plate contain rabbit antiangio sarcoma serum absorbed with normal human serum and liver extract. FIG. 2. Demonstration of tissue antigen absent in angiosarcoma, (a) Trough contains rabbit antihuman liver serum absorbed with normal human serum, (b) Trough contains the antiliver serum additionally absorbed with angiosarcoma extract. In both plates top wells have 10% solution of liver extract and lower wells angiosarcoma extract. Anode is to the right. 93S CMA 03870 FIG. 1. Demonstration of angiosarcoma-related antigen. Peripheral wells have 2-fold serial dilutions of liver angiosarcoma extract (a) and normal liver extract (b). Dilutions are clockwise and start at 100 mg/ml in the upper right well. Central wells in each plate contain rabbit antiangio sarcoma serum absorbed with normal human serum and liver extract. FIG. 2. Demonstration of tissue antigen absent in angiosarcoma, (a) Trough contains rabbit antihuman liver serum absorbed with normal human aorta, (b) Trough contains the antiliver serum additionally absorbed with angiosarcoma extract. In both plates top wells have lot solution of livbr extract and lower wells angiosarcoma extract. Anode is to the right. 93S CMA 003871 414 l.^p:.'.o Antigen BT-1 BT-2 BT-J BA Temper[50 mm in PUS) 2S'C . -U 56 c -- -- ;70"C 100JC _ --_ r -i-, Unaffected; , partial inactivation; --, complete inactivation. tested. This distribution of BT-1 and BA -.vas confirmed by direct immunodiffusion rests of bod) fuidsand tissue extracts with the antibile serum. Occurrence of BT-2 and BT-3 could not be studied by direct testing since the antiliver serum reacted wtth other tissue antigens not detected in bile. Species specificity studies In order to study the species distribution of the antigens, the antisera were absorbed with biles of various mammalian species and tested against two-fold dilutions of human bile. Absorption of the anti sera with human bile abolished reaction with all antigens (with the exception of BA if biles not containing BA were used), thus confirming their presence in humans. In contrast, absorption with pig, dog, cat and guinea-pig biles did not significantly affect any of the lines of precipitation. Absorption w ith Rhesus monkey bile abolished only reaction with BA, thus indicating the presence of an immunologically related antigen in this species. These findings were confirmed by direct double immunodiffusion tests of the animal biles: none of the antigens could be shown, except BA which was demonstrated in monkey bile. However, immunodiffusion patterns comparing the reaction of human and monkey biles showed a line of fusion with a spur consistent with partial identity or a cross-reactive relationship of the BA from the two species. Physicochemical properties Treatment of bile with Pronase resulted in loss of the lines of precipitation given with the antisera. In the controls with no enzyme and with enzyme but without incubation, the antigens remained reactive. The control with enzyme alone was unreactive. This suggests that the bile antigens are proteins or closely associated to protein. In order to study the effect of heating on the bile antigens, a 10% solution of bile in PBS was incubated at various temperatures for 30 min and tested with the antisera. As indicated in Table 2, BT-1 and BT-3 were relatively susceptible to heat treatment; in contrast, BT-2 and BA withstood boiling temperature. All antigens were unaffected by treatment with neutral or alkaline buffers up to pH 10 0. Buffer at pH 5-0 partially inactivated BT-1 but did not affect the other antigens. Treatment at pH 2-5 and 3-5 resulted in complete inactivation of BT-1 and BT-3 and partial inactivation of BT-2. BA was unaffected even at pH 2-5. As shown in Table 3, BT-1 was precipitated at ammonium sulphate concentrations between 30 and 70% of saturation, BT-2 at 70-100% and BT-3 at 30-50%t BA precipitated over a wider range of ammonium sulphate concentrations. Treatment with ethanol (Table 4) resulted in inactivation of BT-1; the other antigens precipitated mainly at ethanol concentrations ranging from 30 to 70%. BT-1 and BT-5 in patients' sera Scrum samples from thirty-three patients and from forty-eight normal blood donors were examined CMA 003879 Antigens in bile Table 3, Antigens in (N'HjjSOa frictions (NHiJj.SO* fractions (% saturated) Antigen 0-30 30-50 50-70 70-100 SN* BT-l BT-2 BT-3 BA _ - -i- -- -- - -r-f -- +- 4- + + + -f +* + -f | K + +, -t- Indicate strength of double diffusion reaction in dilution assay, * SN =* supemate of the 100% saturated solution. Table 4. Antigens in ethanol fractions Ethanol fractions (v/v%) Antigen 0-30 30-70 SN* BT-If BT-2 BT-3 BA _ ++ - ++ -- +++ + -M- +, + +,+, indicate strength of double diffusion reaction in dilution assay. SN, supemate of the 70% ethanol precipitation, dialysed and lyophilized. t Ethanol susceptible. Table 5, BT-2 and BT-3 in patients' sera Group Number of cases Number of caves with circulating with circulating Number of cases BT-2 BT-3 Acute hepatitis: infectious 5 44 senim 7 66 alcoholic 5 44 carbon tetrachloride 1 1 1 Liver rirrhosii 4 00 Obstructive jaundice 2 0 0 Liver ingiosarcotna 2 0 0 Pneumonia 5 00 Pancreatitis 2 00 Normal subjects 48 0 0 415 for the presence of the bile antigens by double immunodiffusion. BT-1 and BA could not be detected in any of the scrum samples. As indicated in Table 5, BT-2 and BT-3 were detected in sera of most cases with acute hepatitis but not in other liver diseases, pneumonia, pancreatitis and normal subjects. Identity of antigens BT-2 and BT-3 detected in bile and patients' seta was shown by the fusion of their lines of precipitation and by their elimination following absorption of the antiliver scrum with bile. CMA 003880 416 E. Espinosa DISCUSSION The present data have demonstrated presence in bile of antigens of limited and of wide body distribution. Of limited body distribution was antigen HA, detected in bile and saliva of certain individuals, and antigen BT-1 found to be present in bile, urine and kidney. A wide body distribution was shown by antigens BT-2 and BT-3 which were detected in bile, liver, kidney, spleen and lung. Antigens previously demonstrated in bile include both normal serum proteins and apparently bile- specific proteins (Rawson, 1962; Hardwicke ttal., 1964; Clausen ft at'., 1965; Yoon ft a!., 1966; Wales tt al., 1969; Englcrt ct al., 1970). The reported number of bile proteins grew with increasing sensitivity of the techniques employed from four, by paper electrophoresis (Yerschure, 1956; Wales tt al., 1969), to up to sixteen total proteins and four specific bile proteins, by disc electrophoresis combined with immunodiffusion and immunoclectrophorcsis (Englert tt al., 1970). Three `bile-specific protein com ponents' were identified in human bile by Yoon tt al. (1966) and up to four by Englert tt al. (1970) by immunoclectrophorcsis using rabbit antihuman bile scrum absorbed with normal human scrum. Since other body fluids and tissues were not tested, these studies did not rule out the possibility that such antigens are present in other body fluids and tissues as well as bile. Of the components identified by Yoon tt al. (1966), one (alpha-1 biliprotein) moved slower and another (biliprealbumin) faster than serum albumin in immunoelectrophoresis; a third component (biliproalbumin) moved as serum albumin. One of the antigens described by Englert tt al. (1970) moved as biliprealbumin and the three others in the post albumin zone which includes the serum globulins. The antigens described in the present work had immunoelectrophoretic mobilities ranging between that of alpha-and beta-serum globulins and thus may relate to the antigens described previously in the post-albumin zone. Antigens with albumin and prealbumin electrophoretic mobilities were not detected in this work. Neither was evidence obtained of antigens with strict specificity for bile. However, one bile antigen (BA) was found to be shared with saliva only. This antigen was not searched for in saliva in a preliminary study and was, thus, considered to be present in bile only (Espinosa, Shelton & Shaw, 1975). s The antigens described in this work separated from each other in Sephadex G-200 gel filtration ammonium sulphate fractionation and, thus, appear to be distinct and different molecular entities. methods are being employed for their purification and isolation. AH antigens were inactivated by Pronasc, suggesting that they are proteins or closely associated to proteins. In addition, they were precipitated at 70% ethanol concentration, excepting BT-1 which was inactivated by ethanol and was generally rather labile. In contrast, BT-2 and BA were relatively resistant to boiling temperature and to acid pH. Of special interest to liver disease is the finding of BT-2 and BT-3 in the blood of patients with acute hepatitis. These antigens are part of a group of circulating tissue antigens of wide organ distribution which have been found to be present in patients with liver injury (Espinosa, 1974). Two more of these antigens, referred to as CTA-2 and CTA-3, have been recently characterized (Espinosa, 1976). CTA-2 is a protein with a relative electrophoretic mobility of a beta-serum globulin and molecular weight in the range of 67,000-80,000; CTA-3 moves in immunoclectrophorcsis like an alpha-l-serum globulin, is resistant to Pronase and behaves like a protein of molecular weight of about 300,000. Other circulating tissue antigens detected in liver disease include the following liver-specific antigens: (a) F-antigen of wide interspecies cross-reactive properties (Bodmer, 1969; Rosenmund, 1971; Smith St Iverson, 1973); (b) a protein located within the cytoplasm of hepatocytes and of molecular weight of about 190,000 (Meyer zum Buschenfelde St Miescher, 1972); and (c) LSA, a protein of electrophoretic mobility of serum gammaglobulins and molecular weight in the range of 82,000-93,000 (Espinosa, 1973). Signifi cance of these findings in the pathogenesis and clinical assessment of hepatitis is worth investigating. These studies should include correlations of the levels of these antigens in diseases of the liver and of other organs and investigation of their autoimmunogenic properties. The possibility that bile antigens are related to autoimmune liver disease is suggested by the demonstration of cell-mediated immune response to a protein fraction of human bile in patients with primary biliary cirrhosis and active chronic hepatitis (Eddleston tt al., 1973). 416 E. Espinosa The present diti '-r c .ierv r";rr! TT '.tTr.f'rens .T :nd -it'", rie body distribution. Oflimitcd body distribution ujs antigen BA, detected in bile and sails a ot certain indisiduals, and inticcn BT-1 found to be present in bile, urine and kidney. A wide body distribution was shosvn by antigens BT-2 and BT-3 which were detected in bile, liscr, kidney, spleen and lumr. Antigens previously demonstrated in bile include both normal serum proteins and apparently bile- specific proteins (Rawson, 1962, Hardwicke d .iL, 1964, Clausen et si., i965, Yoon ct ai., 1966; Wales et ai., 1969; Englcn et ai., 1970). The reported number of bile proteins grew with increasing sensitivity of the techniques employed from four, by paper electrophoresis (Yerschure, 1956; Wales et ai., 1969), to up to sixteen total proteins and four specific bile proteins, by disc electrophoresis combined with immunodiffusion and immunoclectrophorcsis (Englert et ai., 1970). Three `bile-specific protein com ponents' were identified in human bile by Yoon et ai, (1966) and up to four by Englert et ai, (1970) by Immunoelectrophoresis using rabbit antihuman bile scrum absorbed with normal human scrum. Since other body fluids and tissues were not tested, these studies did net rule out the possibility that such antigens are present in other body fluids and tissues as well as bile. Of the components identified by Yoon rf ai, (1966), one (alpha-1 biliprotein) moved slower and another (biliprealbumin) faster than serum albumin in immunoelectrophoresis; a third component (biliproalbumin) moved as serum albumin. One of the antigens described by Englert et ai. (1970) moved as biliprealbumin and the three others in the post albumin zone which includes the serum globulins. The antigens described in the present work had immunoeiectrophoretic mobilities ranging between that of alpha- and beta-serum globulins and thus may relate to the antigens described previously in the post-albumin zone. Antigens with albumin and prealbumin electrophoretic mobilities were not detected in this work. Neither was evidence obtained of antigens with strict specificity for bile. However, one bile antigen (BA) was found to be shared with saliva only. This antigen was not searched for in saliva in a preliminary study and was, thus, considered to be present in bile only (Espinosa, Shelton & Shaw, 1975). The antigens described in this work separated from each other in Sephadex G-200 gel filtration and ammonium sulphate fractionation and, thus, appear to be distinct and different molecular entities. Both methods are being employed for their purification and isolation. All antigens were inactivated by Pronase, suggesting that they are proreins or closely associated to proteins. In addition, they were precipitated at 70% ethanol concentration, excepting BT-l which was inactivated by ethanol and was generally rather labile. In contrast, BT-2 and BA were relatively resistant to boiling temperature and to acid pH. Of special interest to liver disease is the finding of BT-2 and BT-3 in the blood of patients with acute hepatitis. These antigens are part of a group of circulating tissue antigens of wide organ distribution which have been found to be present in patients with liver injury (Espinosa, 1974). Two more of these antigens, referred to as CTA-2 and CTA-3, have been recently characterized (Espinosa, 1976). CTA-2 is a protein with a relative electrophoretic mobility of a beta-serum globulin and molecular weight in the range of 67,000-30,000; CTA-3 moves in immunoelectrophoresis like an alpha-1-serum globulin, is resistant to Pronase and behaves like a protein of molecular weight of about 300,000. Other circulating tissue antigens detected in liver disease include the following liver-specific antigens; (a) F-antigen of wide inicispecies eross-reactive properties (Bodmer, 1969; Roeenmund, 1971; Smith St Iverson, 1973); (b) a protein located within the cytoplasm of hepatocytes and of molecular weight of about 190,000 (Meyer zum Biischcnfeldc Sc Mieschcr, 1972); and (c) LSA, a protein of electrophoretic mobility of serum gammaglobulins and molecular weight in the range of 82,000-93,000 (Espinosa, 1973). Signifi cance of these findings in the pathogenesis and clinical assessment of hepatitis is worth investigating. These studies should include correlations of the levels of these antigens in diseases of the liver and of other organs and investigation of their autoimmunogenk properties. The possibility that bile antigens are related to autoimmune liver disease is suggested by the demonstration of cell-mediated immune response to a protein fraction of human bile in patients with primary biliary cirrhosis and active chronic hepatitis (Eddleston et ai., 1973). CMA 003882 Antigens in bile 417 The author wishes to thank Miss Margaret VenBraun, Mrs Virginia F l'ctrcy anti Mr Gordon I., Shaw for assistance during this ini estimation. This work was supported in part by grants from Eli Lilly and Company and B. F, Goodrich and Company REFERENCES Boomer, A. (1969) Nachweis ernes hcpatozcllularen Shadens mit dcr Gelprazipitations methode niit llilfc ernes Maus* Auto-Antikorpers. Path. Mtcrobwi. (Basil), 33, 357, Clausen, J , Kruse, 1. Sc Dam, H (1965) Fractionation and characterization of proteins and lipids in bile. Scand. J. din. Lab. Invest. 17, 325. Eddleston, A.LAV F., McFarlane, I.G., Mitchell, C.G., Reed, VV.D. Sc Williams, R.B. (1973) Cell-mediated immune response in primary biliary cirrhosis to a protein fraction from human bile, Brit. mid. J. iv, 274. Englert, E., Jr., Wales, E.E., Jr. Sc Straight, R C, (1970) The proteins of human gallbladder bile with and without gallstones. Clin. ehim, Acta, 29, 319. Espinosa, E. (1973) Grculating tissue antigens. II. Studies on an organ-specific antigen of human liter. Lab. Invest. 29, 556. Espinosa, E. (1974) Grculating tissue antigens. L Tissue antigens in serum of patients with diseases invoicing injury of the liver and of other organs. Clin, cip. Immunol. 16, 153. Espinosa, E. (1976) On two tissue antigens detected in pathologic sera. Ijib. Invest. 34. 314. Espinosa, E. Sc Kaplan, M.H. (1968) Antigenic analysis of human heart tissue. Identification of antigens with specificity restricted to heart and skeletal muscle in acid extracts of myocardium. J, Immunol. 100, 1020. Espinosa, E. Sc Kaplan, M.H. (1970) Antigenic analysis of human heart tissue. Antigens with restricted organ distribution in acid extracts of human myocardium. J. Immunol, 105,416. Espinosa, E , Shelton, R.V A Shaw, G L. (1975) Tissue specificity and properties of two bile antigens. Fed Proc 34, 1033. Hardwicke, J., Rankin, J.G., Biker, K.J, & Preisig, R. (1964) The loss of protein m human and canine hepatic bile. Clin. Sci. 26, 509. Meyer zum Bischeneelde, K.H. 8c Miesgeier, P.A. (1972) Liter specific antigens. Purification and characterization. Clin, e.\p. Immunol. 10, 89. RawsOn, A.J. (1962) Human bile proteins. I. Proteins identified by antibody to human scrum. Clin. Cbem. 8, 310. Rosenmu.nu, A. (1971) Ein let erspezifisches Antigen im Serum ton Lctcrkrankcn. Sihu'iiz. mid. Wschr. 101,1023. Scheidegger, J.J. (1955) Une micro-methode de I'immunoelectrophorese. Int. Arch. Allergy, 7, 103. Smith, J B. 8c Iverson, G-M. (1973) Occurrence of liverspecific antigen in adult human serum. Chn. tip. Immunol. 13, 209. Verschure, J.C.M. (1956) Electro-chromograms of human bile. Clm. chim. Aeta, 1, 38. Wales, E.E., Jr., Englert, E., Jr., Wingaro, R.T., Maxwell, J.G. 8c Stevens, L.E. (1969) Disc electro phoresis-immunodiffusion of serum proteins in normal human gall-bladder bile, Proc. Sot. tip. Biol. (.V. Y), 132, 146. Voon, D., Shim, B. 8c Kil, T. (1966) Bile-specific protein components in human hepatic bile. J. Lab. din. Mid. 67, 640. CMA 003883 ABSTRACTS CMA 003884 ABSTRACT MUST B RECEIVED AT SOCIETY OFFICE BY THURSDAY, DECEMBER 1, 1977 DO NOT FOLD THIS FORM 1 Indicate below the numbers and titles of sessions in which your abstract might be programed (see Topic Category List); Unified Mail to: CAREFUL SELECTION IS CRITICALLY IMPORTANT. 1978 FASEB Dr. Kenneth M. Endioott, Executive Officer American Association of Pathologists Ist #...12.8.............. Title ..iRBRimoBathplogy.............. 2nd #..0.2.6............... Title ..X.uM.r..Immunology............ 3rd #...1217................ Title ..Immunobiology................... (See paragraph below for the number of photocopies to be submitted with abstracts.) Abstract Form 9650 Rockville Pike Bethesda, Maryland 20014 PRESENTATION PREFERENCE Preferred choice (CHECK ONE ONLY) jBc Poster presentation Slide presentation Indifferent 16 mm. films (silent or optical sound) are permitted if essential to 10-minute slide session presentation. MOVIE YES........... NO........... Submit justification by letter to Soci ety Office, with abstract. IMPORTANT: See sample abstracts, typing and mail ing instructions on reverse side; use enclosed Check List for preparation of abstract. PATHOLOGY LYMPHOCYTE TRANSFORMATION TESTS IN VINYL CHLORIDE (VC) WORKERS. H. Philip Fortwengler*, Michael E. Dever*, Carlo H. Tamburro**, and Enrique Espinosa. Univ. of Louisville School of Medicine , Louisville, KY 40201 Previous reports have shown circulating immune complexes in VC workers and a tumor-associated antigen in VC-related liver angiosarcoma. This suggests immune stimulation by a tissue or plasma antigen induced by or conjugated with VC or a metabo lite. In this work, the in vitro lymphocyte reactivity for such antigens was tested in 79 VC workers including 25 with liver abnormalities, and 20 normal individuals having no ex posure to VC. The liver angiosarcoma antigen preparation in cluded the tumor-associated antigen and other tissue antigens as shown by immunodiffusion. Stimulation Indices (SI) were calculated from cellular incorporation of trltiated thymidine. Mean SI in the normal individuals for antigens of angiosarcoma and normal liver tissues were 4.7 (SE1.7) and 3.8 (-0.9) and in VC workers, 1.8 (0.2) and 2.5 (0.3) respectively. Mean SI for PHA and Con A in the normal individuals were 235 (35) and 209 (30) and in VC workers 201 (23) and 180 (19) re spectively. Thus, these results suggest that VC workers have a decreased lymphocyte response to antigens of liver angiosar coma and normal liver tissues rather than the hypothesized in creased reactivity. This appears to be due to a lower overall lymphocyte responsiveness in these chemical workers. (Supported in part by the Manufacturing Chemists Association) The original typed copy of this abstract )rm (for reproduction by photo-offset in EDERATION PROCEEDINGS) must be rbmitted together with 9 photocopies, one ;t of author index cards, three abstract' lentification cards, and return postaL All laterial must reach the Society office NO ATER THAN THURSDAY, DECEMBER , 1977. All compounds that are designated by code or initial letters must be identified adequately in the abstract, e.g., MJ-1999: 4-(2-isopropylamino-l-hydroxyethyl) methamesulfonanilide hydrochloride. Each Abstract Form submitted MUST BE SIGNED by a member of the AMERICAN ASSOCIATION OF PATHOLOGISTS.* [AILING ADDRESS OF FIRST AUTHOR (Please Print or Type) H. Philip Fortwengler............... Univ. of Louisville Sch. Med Enrique Espinosa. M.D. ^ (Member'* Neme:Bra flrlnt or Type) / (Membrai Signature) ...Louisville,...Ke.TLtHCky....... Zip ...4Q2Q1...... Member's telephone no.: Area Code ....5.Q2... # ..588 5525. Telephone no.: Area Code ...50? # 5.8.8-5.251 *See the enclosed unified rules for eligibility of papers CMA 003885 ABSTRACT MUST BE RECEIVED AT SOCIETY OFFICE BY TUESDAY, DECEMBER 19 Please consider this abstract for inclusion in the tentatively listed minisymposium, M____;---------- Indicate below the numbers and titles of sessions in which your abstract mieh be_prormed (see Topic Category List); , u 410 Title Liver Pathophysiology 21 a 451u4o _ Title Title ParhpM.nl ngy .of KnHnrhel -him. ant carcinogenesis; Chemical. Vira. 1979 FASEB Abstract Form DO NOT FOLD THIS FORM Mail to: Dr. Kenneth M. Endicott, Executive Offit American Association of Pathologists 9650 Rockville Pike Bethesda, Maryland 20014 PATHOLOG FACTOR VIII CONTENT AS EVIDENCE FOR ENDOTHELIAL ORIGIN OF VINYL CHLORIDE ASSOCIATED LIVER ANGIOSARCOMA (VCA). H. Philip Fortwengler*, Douglas Jones*, Carlo H. Tamburro* and Enrique Espinosa (SPON: G. Randolph Schrodt). Univ. of Louisville School of Medicine, Louisville, JCY. 40232 To ascertain the endothelial cell origin of VCA we have looked for Factor VIII in the tumor since this factor ap pears to be specific for such cells (Hoyer et al., 1973). Frozen sections of three VCA and one idiopathic case were examined for presence of Factor VIII by indirect immuno fluorescence. Sections of angiosarcoma demonstrated a . strikingly increased specific fluorescence which lined ^ enlarged sinusoids. This intense fluorescence was easily' seen on low power(lOOx) as an irregular or splotchy pat tern. Occasional striations of linear fluorescence which did not follow hepatic cords were also present. A similai pattern of staining was also given by the Idiopathic angle sarcoma but was never seen in sinusoids of normal liver. These observations were further supported by the fin^ktg that absorption of Factor VIII antiserum with exper^Jntalangiosarcoma tissue resulted in complete inhibition of immunofluorescence. These findings demonstrate that VCA and idiopathic angiosarcoma include proliferating cells containing Factor VIII and therefore strongly suppo: an endothelial cell origin of this tumor. (Supported in part by the Manufacturing Chemists Association.) All compounds that art designated by codt or initial Utters must be identified adequate in the abstract, e g., MJ-1999: 4-(2-isopropylamino-l-hydroxyethyl) methanesulfona tide hydrochloride. MAILING ADDRESS OF FIRST AUTHOR (Please Print or Type. Provide full name rather than initals.) .. .Phil.lp. <?r.tvsngler.,. .MS.......................... 538 MDR Building 511 South Floyd Street............................ University of Louisville Med. School Louisville, KY zip ... *?.2.9?....... Telephone No.; Area Code. 5.02#.. .58.8t5.251........... Each Abstract Form submitted MUST BE SIGNED by a membe of the AMERICAN ASSOCIATION OF PATHOLOGISTS. G. Randolph Schrodt ________________________ ft a (Members Name Please Print or TypapProvtde full namaj ^(Member's Signature) Member's telephone no,; Area Code....... 502...........#-----588-5341. Signing member, are you willing to chair a session? ( ) yes, category #---------- ( ) no If yes, are you also willing to give an overview or introductory I ( ) yes { ) no CMA 003886 ABSTRACT MUST 3E RECEIVED at SOCIETY of F ICE SY TUESDAY, DECEMBER 19 P'.ttsc consider this ibjinei fcr inclusion in the 'eniauvely b':ed miniiytnposium, M____; ___ ___________________________________________________ Indicate below the numbers and titles of sessions m which your abstract mih be protramed (see Topic Caiegory List); ir -? Till- Liver P-athoohys^ln-r_______ , tl 406 ~ Tnie uardinoganesTs; Lr.&.uicii, ''TTrai '1979 FASfcB Abstract Vw a d i i DO NOT FOLD TM!S F^RM Mail to: Dr, Kenneth M. Endicort, Executive Off American Association of Pathologists 9650 Rockville Pike liW-yis, 20014 PATHOLOC FACTOR VIII CONTENT AS EVIDENCE FOR ENDOTHELIAL ORIGIN OF VINYL CHLORIDE ASSOCIATED LIVER ANGIOSARCOMA (VCA). H. Philip Fortwengler*, Douglas Jones*, Carlo H. Tamburro* and Enrique Espinosa (SPON: G. Randolph Schrodt). Univ. of Louisville School of Medicine, Louisville, ICY. 40232 To ascertain the endothelial cell origin of VCA we hav looked for Factor VIII in the tumor since this factor ap pears to be specific for such cells (Hoyer et al., 1973). Frozen sections of three VCA and one idiopathic case were examined for presence of Factor VIII by Indirect immuno fluorescence. Sections of angiosarcoma demonstrated * strikingly increased specific fluorescence which lined enlarged sinusoids. This intense fluorescence was easily seen on low power(lQOx) as an irregular or splotchy pat tern. Occasional striations of linear fluorescence which did not follow hepatic cords were also present. A simila pattern of staining was also given by the idiopathic angl sarcoma but was never seen in sinusoids of normal liver. These observations were further supported by the finding that absorption of Factor VIII antiserum with experiment talangiosarcoma tissue resulted in complete Inhibition of immunofluorescence. These findings demonstrate that VCA and idiopathic angiosarcoma include proliferating cells containing Factor VIII and therefore strongly suppo an endothelial cell origin of this tumor. (Supported in part by the Manufacturing Chemists Association.) AU compound! that art designated by code or initial letters must be identified adequate in the abstract, e.g., MJ-1999: 4-(2-tsopropylamino-l-hydroxyethyI) methanesulfons lide hydrochloride. MAILING ADDRESS OF FIRST AUTHOR (Please Prim or Type. Provide full name rather than iniuls.) .. .Philip. .Eor.tvngler.t. M.S>........................ 538 MDR Building 511 South Floyd Street........................... University of Louisville Med. School Louisville, KY %p 40202 Telephone No.; Area Code. 5.02#.. .58.8r5.251........ Each Abstract Form submitted MUST BE SIGNED by a memb< of the AMERICAN ASSOCIATION OF PATHOLOGISTS. G. Randolph Schrodt __________________________ . V7T/-\ ntmaJilyWrnbw't Nm. PIMM PVmi at TypafProMe* tu yfMwnfcer'x 5>gnanitel Member's telephone no.; Area Code....... 502........... #-- 588-5341 Signing member, are you willing to chair a session? ( ) yes. category #---------- ( ) no If yes, are you also wilting to give an overview or introductory lecture? ( ) yes ( ) no CMA 003887 CLINICAL RESEARCH Abstract Reproduction Form TYPE name, address, and telephone number o author who should receive correspondence. .... 502/582-2211 ex. 366 onto: 502/895-2955 home Name --_--------------- ----------------- --------------------------------------------------------------Address_____C_a__rl_o H. Tamburro, M.D. University of Louisville Scliool of Merlioinft Department of Medicine Louisville, Kentucky 40201 THIS FORM MUST BE SIGNED BY A MEMBER FOR OFFICE LSL ONLY Dale Payment (S 10.00) CHECK Preferred Sub-Specialty Classification: ____ Cardiovascular Clinical ____ Epidemiology Clinical ____ Pharmacology ____ Dermatology ___ Endocrinology* ___ Gastroenterology ____ Genetics _X _ Health Care Research Hematology ____ Immunology & ___ Conn. Tissue ___ Infectious Disease ____Metabolism* ___ Oncology ____ Pulmonary ___ Renal & Electrolyte traditionally, Endocrinology ha* included papers dealing with the thyroid, adrenal and pituitary glands, and gonads, while ab stracts dealing with the purathy* r<tids, calcium and phosphorus metabolism, bones. thyrocukilorun, diabetes, insulin, glucagon, and growth hormone have been considered under Mewboitsm. -------- ---- TYPE ABSTRACT 11 ERF:/ BE SURE TO STAY WITHIN BORDER................... USE OF DYE CLEARANCE IN DETECTION OF HEPATOCELLULAR INJURY AMONG VINYL CHLORIDE WORKERS. P. Fortwengler* and C. H. Tamburro. Department of Medicine, Digestive Diseases & Nutrition Section, University of Louisville, School of Medicine, Louisville, Kentucky. Serious hepatic injury associated with exposure to potentially harmful indus trial chemicals and the need for a sensitive method of early detection has been previously noted among vinyl chloride workers. (Creech, J. L., et al, Gastroent., 67:786, 1974). Forty-three of the 1,183 employees of this poly vinyl chloride production plant with biochemical, radioisotopic, and histologi cal evidence of liver dysfunction were evaluated using indocyanine green (ICG) dye clearance, at both the 0.5 mg/Kg and 5.0 mg/Kg dose. Percentage dis appearance rates (PDR) were determined by both blood and ear densitometry with automatic computer calculation of PDR. Liver histology was normal in 9, minor changes in 10, fat and minimal fibrosis in 14, portal fibrosis in 4, fibrosis and lobular distortion in 2, and angiosarcoma in 4. Individual bio chemical tests correctly indicated hepatic status in 80-83% of the cases with 8% false positives and 10-15% false negatives. Radioisotopic studies alone correctly indicated liver status in only 55%, with,45% either false positive or false negative. The 0.5 mg/Kg dose of ICG correctly indicated liver status in 80% with no false positives, but failed to indicate liver dysfunction in 20%. However, at the 5.0 mg/Kg dose, all cases of liver dysfunction were detected. Ear densitometry values were essentially the same as blood values. Ear densitometry with electronic calculator allows a dye clearance determination over 10 minutes without drawing blood. This data demonstrates an effective fheans for early detection of industrial chemical hepatotoxicity. PLEASE CHECK ABSTRACT CAREFULLY FOR APPEARANCE BEFORE MAILING IMPORTANT The instructions accompanying this form must be followed COMPLETELY for all abstracts which are to appear in CLINICAL RF-SEARCH. Ab stracts which do not conform either will be re typed by the publisher at a cost of $15.00 to the author, or rejected. Revised June 1974 "The sponsoring member affirms that the material herein will not have been previously published or presented at any national meeting, that any animal stud ies conform with the "Guiding Principles in the Care and Use of Animals" ot the American Physiological Society and that any human experimentation has been conducted according to a protocol approved by the institutional committee or. ethics of human investigation or -if no such committee exists-that it conforms with the principles of ihc Declaration of Helsinki of the World Medical Associa tion (CLINICAL RESEARCH 14:195, 1966). MEMBER'S SIGNATURE . .C.^LC.(,/ Li.l.ft-J.- CMA 003888 ' " X<f& RECEIVE0 AT j-C.c^ r Jrriu or in UrtiJA 7 ( r^jiTwArtf i ( i^/o : __ --:-------L /lh umber* nd ml** ( sessions in which your abstract DO NOT FOLD THIS FORM is* <** Top.* Otciory List); i", nta Biochemical Pharmacology ------------ Ti||e Enzymes - General 1978 AS3C rule ----------------------------------------------------------------------------- UmuI Technique* Abstract Form rtiiul typed copy of this sbsuact form mmt be submitted together photocopies. VSA K<r \^> 4 RTANT: jnple abstracts, typing and mailistructions on reverse side; use ad Check List for preparation of ct. e original typed copy of this ct form (for reproduction by -offset in FEDERATION PROHNGS) must be submitted to with 7 photocopies. streets submitted for the "Edulal Techniques" poster session do revent a member from submitting lonsoring an abstract for the r sessions. ard projectors for 2" x 2" and t 4" slides will be available in all sessions. Other audio-visual aids e provided at cost if the request ustification accompany this ab. Authors will be billed following eeting. DECREASED GLUC0SE-6-PH0SPHATASE ACTIVITY IN LIVER IN VINYL CHLORIDE EXPOSED RATS. J.T. Du* and C.H. Tamburro* (SPON: M. Fonda) Dig. Dis. & Nutr. Sect., Dept. Med., Cancer Center, Unlv. of Louisville Med. Sch., Lou., Ky. 40201. Increases in key glycolytic enzymes paralleling hepatoma tumor growth (Heinrich, et al., FEBS Letters, 42:145, 1974) and decreases in key gluconeogenic enzymes prior to and with the development of hepatomas (Isok, et al., Voprosy. Med. Khim 19:568, 1973) have been shown. We exposed adult SpragueDawley rats to 10,000-20,000 ppm of vinyl chloride (VC), 4-S hrs./day, 5 days/wk. for 3-4 wks. (40-140 hrs. exposure) to Induce liver injury and angiosarcoma formation. Glucase-6phosphatase, a key gluconeogenic enzyme in the liver micro somal fraction, decreased 25% over control (P<0.05 from com bined data). Other microsomal proteins and enzymes related to VC metabolism, i.e., P450, NADPH-cytochrome c reductase and mixed function oxidase were unchanged in the same microsomal fraction with no differences in either mitochondrial cyto chrome oxidase or blood transaminases. The decrease in glucose-6-phosphatase is similar to the lower gluconeogenic enzyme findings in hepatomas. This may reflect an increase^ glycolysis and ribose-5-phosphate production associated with de novo purine biosynthesis prior to tumor development, inhi bition in enzyme synthesis, or increased breakdown due to VC exposure. Work is under way to determine the mechanism. The decrease in glucose-6-phosphatase may be an early biochemical lesion usable as an indicator of liver injury associated with the subsequent development of angiosarcoma. (B.F.Goodrich Grant) Supported by NCI Contract #N01-CN-55212 All compounds that are designated by code or initial letters must be identified adequately in the abstract, e.g., MJ-1999:4-(2-isopropylamino-1 -hydroxyet hy!) methane>ultbnanilide hydrochloride. Each Abstract Form submitted MUST BE SIGNED by a member of the American Society of Biological Cltemists. NG ADDRESS OF FIRST AUTHOR r. Du., Ph.p. rersity of Louisville, MDR Bldg., Rm.535 S... .Floyd. .StTee.bjLpu.., .Ky... Zip AQ2P.1.. one no.: Area Code 39?.......... #. ??.~3???......... CMA 003889 Member's Signature Telephone No.: Area Code 502 588-5228 Mail to: Hobart A. Harte, Executive Officer American Society of Biblogical Chemists 9650 Rockville Pike Bcthesda, Maryland 20014 ** / ne^ciUE6D AT luCiti Y UhHI-t 8T inUM^JAI, -c^nLAnr U, 13/0 umb<r, ,nd Mile* of <n "h:ch your abstract t^r.mm^twTop.cCttcjory List,; li m/T . rm. JochgsicAl Pharmacology '-Mt Ea=ygea - General-------------- .Till* L*>iui Tc^^J4JC* i(ul typed py Of pMtocoptes- *>''** form :T,U3 be ttifcmittttl tejsthsr DO NOT FOLD THIS FORM icnc Acnn iwiu ngju Ahcfrppr Pnrrn ,-ljv.i CtOt 1^1 iii \ \A<?<-. N'O, ;<\ 4 \\\ RTANT: jnple abstracts, typing and mailistructions on reverse side; use ed Check List for preparation of ct. e original typed copy of this ct form (for reproduction by -offset in FEDERATION PRO>INGS) must be submitted to with 7 photocopies. streets submitted for the "Edulal Techniques" poster session do event a member from submitting >onsoring an abstract for the r sessions. ard projectors for 2" x 2** and t 4" slides will be available in all sessions. Other audio-visual aids e provided at cost if the request ustification accompany this ab. Authors will be billed following ecting. DECREASED GLUCOSE-6-PHOSPHATASE ACTIVITY IN LIVER IN VINYL CHLORIDE EXPOSED RAIS. J.T. Du* and C.H. Tamburro* (SIGN: M. Fonda) Dig. Dis. & Nutr. Sect., Dept. Med., Cancer Center, Unlv. of Louisville Med. Sch., Lou., Ky. 40201. Increases in key glycolytic enzymes paralleling hepatoma tumor growth (Heinrich, et al., FEBS Letters, 42:145, 1974) and decreases in key gluconeogenic enzymes prior to and with the development of hepatomas (Isok, et al., Voprosy, Med. Xhim 19:563, 1973) have been shown. We exposed adult SpragueDawley rats to 10,000-20,000 ppm of vinyl chloride (VC), 4-3 hrs./day, 5 days/wk. for 3-4 wks. (40-140 hrs. exposure) to , induce liver injury and angiosarcoma formation. Glucose-6phosphatase, a key gluconeogenic enzyme in the liver micro somal fraction, decreased 25Z over control (P<0.05 from com bined data). Other microsomal proteins and enzymes related to VC metabolism, i.e,, P450, NADPH-cytochrome c reductase and mixed function oxidase were unchanged in the same microsomal fraction with no differences in either mitochondrial cyto chrome oxidase or blood transaminases. The decrease in glucose-6-phosphatase is similar to the lower gluconeogenic enzyme findings in hepatomas. This may reflect an increase in glycolysis and ribose-5-phosphate production associated with de novo purine biosynthesis prior to tumor development, inhi bition in enzyme synthesis, or increased breakdown due to VC exposure. Work is under way to determine the mechanism. The decrease in glucose-6-phosphatase may be an early biochemical lesion usable as an indicator of liver injury associated with Che subsequent development of angiosarcoma. (B.F.Goodrich Grant) Supported by NCI Contract ffNOl-CN-55212 All compounds that are designated by code or initial letters must be identified adequately in the abstract, e.g., MJ-1999: 4^Msopropylamino-1-hydroxyethy I) methaneiulfonaniiide hydrochloride. ____________________ Each Abstract Form submitted MUST BE SIGNED by a member of the American Society of Biological Chemisis. SG ADDRESS OF FIRST AUTHOR C... Du.,Ph.D..................................................... rersity of Louisville, MDR Bldg., Rm.535 5... .Floyd. .strpetnLou...* .Ky.-. Zip AQ2P.I. an* no.: Area Code A9?........... #. ......... Member's Signature Telephone No.: Area Code $d...= .588-5228........ Mail to; Robert A. Harte, Executive Officer American Society of Biological Chemnti 9650 Rockville Pike Bethesda, Maryland 20014 CMA 003890 winMti IVIOil oc nc^sivcu Al ;>uutl Y UI-MUt UY UVbUNESDAY, 11 JANUARY 1978 Pleax consider this abstract for inclusion in the tentatively listed minisymposium, M 27 i Chpmfeal Carcinnffenpgi a Indicate below the numbers and titles of sessions in which your abstract might be programed (see Topic Category List); 1st a M27Title Chemical Carcinogenesis 2nd #--L6.1Title Biochemical Pharmacnlnyv 3rd #__Title Educational Techniques poster session The original typed copy of this abstract form must be submitted together with 8 photocopies. DO NOT FOLD THIS FORM 1978 ASBC/AAI Abstract Form If abstract is scheduled for regular ses sion I prefer; O Poster presentation CXSlide presentation Final decision will be made by Program Committee IMPORTANT; See sample abstracts, typing and mail ing instructions on reverse side; use enclosed Check List for preparation of abstract. The original typed copy of this abstract form (for reproduction by photo-offset in FEDERATION PRO CEEDINGS) must be submitted to gether with 8 photocopies. Abstracts submitted for the "Edu cational Techniques" poster session do not prevent a member from submitting or sponsoring an abstract for the regular sessions. Standard projectors for 2" x 2" and 314" x 4" slides will be available in all slide sessions. Other audio-visual aids can be provided at cost if the request and justification accompany this ab stract. Authors will be billed following the meeting. ELEVATED GLUTATHIONE CONTENT, GLUTATHIONE-S-TRANSFERASE AND GLUTATHIONE REDUCTASE IN LIVER OF RATS EXPOSED TO VINYL CHLORIDE Du, J.T.* and Tamburro. C.H.* (SPON: McCeachin, R. L.) Dig. Dis. & Nutr. Sect., Dept. Med., Cancer Center, Univ. Lou. Med. Sch., Loy., Ky.t 40232 Vinyl chloride (VC) is believed to be metabolized to chloroethylene oxide (CEO) and chloroacetaldehyde, and detoxified by way of glutathione. Rats were exposed to 28,000 ppm VC, 7hrs/ day, 5days/wk for 4 and 6 weeks, the activity of glutathione epoxide-S-transferase (GEST) was elevated 30 to 54% over normal control and air control (9.71+ 0.68 vs 7.52+ 0.97 and 6.30+ 0.68) respectively. However, the activity of glutathione aralkyl-S-transferase (GAST) was not significantly elevated until 6 weeks of exposure to VC. The content of reduced glutathione was also elevated 45% in the VC treated group and the activity of the glutathione reductase, the enzyme to re generate glutathione from the oxidized form was elevated 50%. These results demonstrate that VC exposed rats have the capa city to maintain glutathione reductase activity and glutathione concentration for detoxification. Further, it suggests that the primary route of VC metabolism is initial oxidation to CEO and then detoxification by GEST directly. With longer exposure and probable saturation of the direct route, there is greater rearrangement of CEO to chloroacetaldehyde, and de toxification with glutathione as supported by the delayed in duction of GAST (Supported by a grant from Manufacturing Chemists Association). Alt compounds that are designated by code or initial letters must be identified adequately in the abstract, e.g., MJ-1999: 4-(2-isopropyhmino-l-hydroxyethyl) methanesulfonanilide hydrochloride. __________________________________________ Each Abstract Form submitted MUST BE SIGNED by a member of the American Society of Biological Chemists. McGeachin,_R..L. MAILING ADDRESS OF FIRST AUTHOR (Please print or type ) Du^_ Julie T. 535 MDR Building, P. 0. Box 35260 . . ,U. of. LMed.. .Sch..,. Lciu.,. ,Ky... zip. .40232 Telephone no.; Area Code .. .5P.2... . 588f-.525.1......... (Member's Signature) Telephone No.: Area Code. .$Q? . it. .588.-5.290.... Mail to: Russell J. Hilmoe, Executive Officer American Society of Biological Chemists 9650 RocKville Pika Bethesda, Maryland 20014 CMA 003891 ASPET/SOT 1982 Louisville, Kentucky Deadline for Receipt: May 7, INS Abstract Processing Fee; $20.00 Select Category Numbers and Titles (See list over) 1- 2A. Presentation Preference _ Poster X_ Oral (slides) _ Indifferent If your choice is not available X_ Willingly accept the alternative __Grudgingly accept the alternative Withdraw the abstract Poster boards are S' wide x 4' high. Serum Bile Acids (SBA) Screening For Chemical Hepacotoxicity* Gary Llss* and Carlo H. Tamburro* (SPON: William Waddell). NIOSH Robert Taft Laboratory, Cincin., OH. & Liver Research Center, Div. Occupational Health, Depts. of Medicine & Com munity Health, Univ. of Louisville, Lou., Ky. 40292. Standard liver tests have limited ability to detect lj^Mnt liver disease (Environ. Health Perspect. 1981; 41:117-11!^^ Serum bile acids are more sensitive indicators of liver in jury. During the medical surveillance of 1000 chemical work ers 67 liver biopsies were investigated for histological evi dence of chemical injury (Gastroenterology 1979; 77:A33); 15 had chemical liver disease (CLD), 27 nonchemical liver disease (NCLD) & 25 had normal biopsies (NBX). Fasting SBA-cholylglycine (CG) & conjugates of cholic acid (CCA) were studied by ^^I-radioimmunoassay in these groups and 416 "normal" work ers. Mean + S.E.M. for CG in the CLD, NCLD, NBX & normal groups were 95.17 + 28.75, 27.25 + 4.43, 34.60 + 7.13 & 14.9 +0.88 ug/ml, respectively. Analysis of variance showed sig nificant differences (p<0.035) in the 3 biopsy groups; p<0.001 for all 4 groups. The CLD values were significantly differ ent in CG levels of the NBX (p<0.02). No significant differ ence in CG levels in NCLD and NBS groups; CCA' showed similar results. Fasting SBA showed promise in detecting latent chemical liver injury. Prepare your abstract carefully. It will be printed by photo-offset exactly as received. See the example (ever) for lay-out and style. For elite typewriters (12 pitch), set the margins at 36 and 98 (62 spaces). Use 25 lines. For pice typewriters (10 pitch) set margins at 30 and 81 (51 spaces). Use 25 lines. The first typed letter should nearly or just touch the top and left type box (blue margin lines). Title. Leave no margins. Use a short, descriptive title. Use upper and lower case letters. Authorship. Underline names. Place an asterisk* after the name of each author not a member of ASPET or SOT. If no author is a member, type the following after the name of the last author: (SPON: sponsor's name). Continue on the same line with the authors' institutional affiliation(s), city, state and zip code. Text. Start the text on the next line, indenting 3 spaces. Subsequent lines should extend the entire width of the type box. The text should have all of the elements of a report: introduction, method, result(s) and conclusion. It is improper to substitute "The results will be discussed" for the results and conclusion. Adequately identify all chemical compounds used. Do not fold abstract. Be certain abstract is prepared securely for mailing. All subsequent correspondence mtmt reference the first author to enable us to identify the abstract. The Program Committee selects chairmen and overview ers from volunteers. See below. First author phone #: (502) 588-5251 Membership of author or sponsors. ASPET^j SOT 1. Member signature 2. Name (Type or print) William Waddell, M.D. 3. Will you chair a session* __ ysj ^no In what category Will you give a 25 minute overview? __ yes; X no; In what category Mail the original, 3 copies, a self-adcheaeod post card for program confirmation and 1 check to Kay A. Croker, Acting Executive Officer, ASPET, 9650 Rockville Pike, Bethesda, MD 20814, USA. Reprints cost $15 per 100, lots of 100 only. Order below and fill in the mailing label, below left Invoice: Abstract Processing Fee Reprints: Quantity ordered__cost: TOTAL $20.00 I Mailing label for reprints (Name and Adthess): Mailii* label to the first author (Name and Address): CMA 003892 ASPET/SOT 1932 Loutsviilt, Xcitucity Deadline (or Receipts May T, 1982 *.hstrct Processing Fee: 120.00 Select Category Numbers and Titles lies ust over) 1- lfi_ 2.7i Presentation Preference Poster X_ Oral (slides) _ Indifferent If your choice is not tvelleble Willingly accept the alternative _Grudgingly accept the alternative Withdraw the abstract Poster boards are 6' wide x 4' high. Seruta Bile Acids (S3A) Screening Fnr Chemical "ernr; *3-:i :iV/ . Gary Ll33* and Carlo H. Tivburro* (3?CN: willian '.'Jiiiai l'1 . NIOSH Unbare Taf; Laboratory, Cincin., DM. 3 Liter Center, Div. Occupational Health, Debts, or .".eciciae j Cc".- raunity Health, L'niv. of Louisville, Lou,, Ky. 10292. S 2 2 n - 2 r "i Liver tests h 3 v s li'nited 3biiit'f to Pettit ^ ' i'' ` t J -- -d ^ j ^ c ^( rc^ 'i a ^ ^ ' 1 ^1 * 1 Serun bile acids are core sensitive indicators of liver in jury. Daring the radical surveillance of 1CC0 ah ami cal vo til ers 67 liver biopsies were investigated for histological evi dence of chemical injury (Gastroenterology 1979; 77:A33); 15 had chemical liver disease (CLD), 27 nonchemical liver disease (NCLD) & 25 had normal biopsies (NBX). Fasting SBA-cholylgly- cine (CG) 4 conjugates of cholic acid (CCA) were studied by ^Sf.radioinnnunoassay in these groups and 416 "normal" work ers. Mean + S.E.M. for CG in the CLD, NCLD, NBX & normal groups were 95.17 + 28,75, 27.25 + 4.43, 34.60 + 7.13 4 14.9 +0.88 ug/ml, respectively. Analysis of variance showed sig nificant differences (p<0.035) in the 3 biopsy groups; p<0.001 for all 4 groups. The CLD values were significantly differ ent in CG levels of the NBX (p<0.02). No significant differ ence in CG levels in NCLD and NBS groups; CCA showed similar results. Fasting SBA showed promise in detecting latent chemical liver injury. Prepare your abstract carefully. It will be printed by photo-offset exactly as received. See the example (over) for lay-out and style. For alite typewriters (12 pitch), set the margins at 38 and 98 (62 spaces). Use 25 lines. For pics typewriters (10 pitch) set mergins et 30 end 81 (31 spaces). Use 25 lines. The first typed letter should nearly or just touch the top end left type box (blue margin lines). Title. Leeve no margins. Use a short, descriptive title. Use upper and lower case tatters. Authorship. Underline names. Piece an asterisk * after the name of each author not a member of ASPET or SOT. If no author is a member, type the following after the name of the last authors (SPONt sponsor's name). Continue on the seme line with the authors' institutional effiUationis), city, state and zip code. Text, Start the text on the next line, indenting 3 spaces. Subsequent lines mould extand the entire width of the type box. The text mould here all of the elements of a reports introduction, method, resuitU) and conclusion. It is improper to subetitute "The results will be discussed" for the results and conoluBlan. Adequately identify all chemical compounds used. Do not fold abstract. Be certain abstract is prepared securely for mailing. All subsequent correspondence mimt reference the tint mitbor to enable us to identify the abstract. The Program Committee selects chairmen and overview ers from volunteers. See below. First author phone f: (502) 588-5251 Membership of author or sponsorK /JPEtX; SOT 1. Member signature 2. Name (Type or print) VVNSVv Will:Lm Waddell, M.D. 3. Will you chair a session? __ ya i^no In what category Will you give a 25 minute overview? yea: X no: In what category Mail the orighial, 3 eopise, e aelf-ettemad pest end for program confirmation and 1 check to Kay A. Croker, Acting Exaeutiva Officer, ASPET, 9650 Rockville Pike, Bethmda, MD 20814, USA. Reprints cost $15 per 100, lots of 100 only. Order below and fill in the mailiaf label, below left tavoiem Abstract Proemsing Pee Reprintst Quantity ordered__eoeti TOTAL $ 20.00 L=a= Mailing label for taps tots (Name and Adtheaah Mailing label to the first author (Name and Ackhessh CMA 003893 FORM FOR ABSTRACTS TO BE PUBLISHED IN GASTROENTEROLOGY Type Abstract in Space Below 7 Check the most appropriate cate gory below: Absorption Secretion Motility Liver/Biliary /Bile Salt a G.L Hormones Morphology Clinical Immunology/Microbiology I prefer presentation at: Poster Session (3 Regular Forum (3 Please publish this abstract in GASTROENTEROLOGY at a cost of S25.00 ($30.00 if bill ing required) (3 My check payable to the Amer ican Gastroenterological Asso ciation is enclosed. Please bill me-instructions en closed. Please do not publish this ab stract in GASTROENTER OLOGY. TISSUE AND URINARY GLYCOSAMINOGLYCANS (GAG) CHANGES IN HEPATIC FIBROSIS C. E. Kupchella. J, 0. Jarvis, K. L. Curran, R. A. Greenberg, and C. H. Tamburro Cancer Center and Digestive Diseases and Nutrition Section, University of Louisville School of Medicine, Louisville, Kentucky. GAG's are felt to play an important role in the collagen fibril formation and collagen bundle stabilization. It has been assumed that hepatic fibrogenesis contributes little to the total body connective tissue and would not be reflected in urinary excretion of GAG degradation or syn thetic products. We have previously found, however, that increased urinary GAG's occur in humans with cirrhosis, hepatitis, and hepatic angiosarcoma (Kupchella, C.E., Tamburro, C. H., Clin. Res. 25:329, 1977; Curran, K. L., Kupchella, C. E., Tamburro, C. H., Cancer (in press), 1977). In the present study, changes in urinary and hepatic GAG's were measured in rats with CCL4-induced hepatic damage. Highly significant increases (P <.001) in total GAG's and in the hyaluronic acid (HA), chrondrotin sulfate (CS), and the heparin (H) fractions were found in hepatic tissue after three weeks and persisted through nine weeks of ex posure. Histological examination after three weeks showed hepatic necrosis with extensive fatty metamorphosis without histochemical (Trichrome-Alcian Blue-PAS) evidence of fibrosis. Control hepatic tissue GAG levels were 31+5 pg (of uronic acid) per gram of dry defatted liver. CCL4 treated livers showed a 2 to 4 fold increase in GAG levels; 81.4 ug at 3; 98 pg at 6; and 113 pg/g at nine weeks, re spectively. Increases occurred mainly in the CS and H frac tions. The urinary CS fraction was significantly (P < .05) elevated over the nine weeks of exposure; the HA fraction was elevated during the first two weeks only, and the H fraction excretion appeared to be directly related to CCL4 injections. These data support our previous observations in humans and demonstrate that urinary GAG excretion pattern changes occur before histochemical evidence of collagen formation and suggest that urinary GAG patterns may be a useful indicator of early hepatic fibrogenesis. MAILING ADDRESS OF PRINCIPAL AUTHOR Carlo H. Tamburro, M.D. 511 South Floyd Street Room 535 MDR Building Louisville, Kentucky 40201 ______________________________ zip ------------ If, in the conduct of these studies, human subjects were exposed to risks not required by their medical needs, the author affirms that the study was approved by an appro priate committee, or, if no such committee was available and informed consent was needed, it was obtained in accordance with the principles set forth in "The Institu tional Guide to DHEW Policy on Protection of Human Subjects". MA 003894 CLINICAL RESEARCH Abstract Reproduction Form TYPE name, address, and telephone number of author who should receive correspondence in Box A and complete Box B. Telephone 502-588-5245 (Area code) office 502-239-1818 (Area code) home 8 Name Address Charles E. Kupchella, Ph. D. Associate Director, Cancer Center 213 MDR Building Health Sciences Center Louisville. Kentucky 40201 Date Payment (II 0.00) Check number Purchase order October 1. 1 CHECK Prtftmd Sub-Sptcialry damflcation: ____ Cardiovascular Clinical ____ Epidemiology Clinical ____ Pharmacology ____ Dermatology ____ Endocrinology* ____Gastroenterology ____Genetics ____ Health Care Research ____ Hematology Immunology & ____ Conn. Tissue ____ Infectious Disease -- Metabolism* ____ Oncology ____ Pulmonary ____ Renal & Electrolyte `Traditionally. Endocrinology tin included pepeti dealing with th* thyroid, adnnal and pituitary |Urnib, and gonads, while ahstracts dealing with th* parathy roid*. calcium and phosphorus metabolism, bones, thyrocalcitonia, diabetes, insulin, glucagon, and growth hormon* have been considered under SferehoHm. URINARY CH0NDRQITIN SULFATE FRACTION PATTERNS IN HEPATIC ANGIOSARCC C. E. Kupchella, K. L. Curran, and C, H, Tamburro. Cancer Center, University of Louisville, Louisville, Kentucky. This study was undertaken to evaluate the usefulness of urinary glycosaminoglycan patterns in the detection of hepatic angiosarcoma and in monitoring the course of this disease. Glycosaminoglycans e tracted from 24 hr urines from 2 patients with angiosarcoma of the liver and from normal controls were separated as cetylpyridintum co plexes into "hyaluronic acid," "chondroltln sulfate," and "heparin" fractions. These fractions were further purified by anion-exchange chromatography. The "chondroltln sulfate" fraction isolated from angiosarcoma patients consistently demonstrated an increase io_the amount of glycosaminoglycan eluted in 1.25 M NaCl and a concJ^ant decrease in the 1.5 M NaCl elution peak. This "shift" was affluent related to the course of the disease and was unaffected by chemo therapy. As indicated by the following results, the ratio of 1.25 to 1.5 M elution peaks may be of value In monitoring the course of angiosarcoma: 2 normal controls, .364 and .316; angiosarcoma (prechemotherapy), .843; angiosarcoma (post-chemotherapy), .971; angio sarcoma (advanced), 5.00. Characterization of the 1.5 M and 1.25 F NaCl elution peaks by hyaluronldase susceptibility and comparison w elution patterns reported in the literature suggest that the observ shift was from chondroitin 4 and/or chondroitin 6 sulfate to hepara sulfate. Preliminary data indicate that the excretion pattern in vinyl-chloride-associated liver injury other than angiosarcoma is characteristically different from patterns associated with hepatiti cirrhosis, or metastases to the liver. These observations may be related to progressive connective tissue proliferation in angiosarct IMPORTANT The initractions accompanying this form must be followed COMPLETELY for all abstracts which am to appear in CLINICAL RESEARCH. Ab stracts which do not conform either will be re typed by the publisher at a cost of $15.00 to the author, or rejected. Revised June 1976 CMA 003895 IMS FORM AS WELL AS THE FORM LETTER OF TRANSMITTAL MUST BE SIGNED BY A MEMBER MEMBER'S SIGNATURE > clinical research Abstract Reproduction Form 3 TYPE name, address, and telephone number of author who should recede correspondence in A and complete 3ox B. ..`.;phcns . " '------------------ (Axes wode) office i Are* code) home Name Charles S. Kupchella, Ph. D.______ Associate Director, Cancer Center Address------------------------------------------------------------------- 213 MPR Building_________________ Health Sciences..Center___________ _________ Louisville. Kentucky 40201______ B ou* Payment (S10.00) Cheek niunber Purdue* onler October 1. : . CHECK Prtferrtd Sub-Sptcialy Ckmflcanon: ____ Cardiovascular Clinical ___ Epidemiology Clinical ____ Pharmacology ____ Dermatology ____ Endocrinology* ____ Gastroenterology ____Genetics ____Healthcare Research ____ Hematology Immunology k ____ Conn. Tissue ____ Infectious Disease ____ Metabolism* ____Oncology ____ Pulmonary ____ Renal k Electrolyte TiBdMoMUy. mtocrinoiowy Da* tadaded p*pn> destine wilt, tke thyroid, Mhonai aid pituitwy linm, ttd fdnad*, white hitract* dwtlne with da tauky* told*. eddim aid phmyhote* boMit nd powtfc ItaraoM kiTSii io--idwyil --dee Mtteheetem URINARY CHONDROITIN SULFATE FRACTION PATTERNS IN HEPATIC ANGIOSARCC C. E. Kupchella, K. L. Curran, and C. H. Tamburro. Cancer Center, University of Louisville, Louisville, Kentucky. This study was undertaken to evaluate the usefulness of urinary glycosaminog1yean patterns In the detection of hepatic angiosarcom* and in monitoring the course of this disease. Glycosamlnoglycans tracted from 24 hr urines from 2 patients with angiosarcoma of the liver and from normal controls were separated as cetylpyridlnfurn cc plexes Into "hyaluronic acid," "chondroltln sulfate," and "heparin' fractions. These fractions were further purified by anlon-exchangt chromatography. The "chondroltin sulfate" fraction isolated from angiosarcoma patients consistently demonstrated an increase In the amount of glycosaminoglycan eluted in 1.25 M NaCl and a concomitant decrease in the 1.5 M NaCl eluticn peak. This "shift" was apparent related to the course of the disease and was unaffected by chemo therapy. As Indicated by the following results, the ratio of 1.25 to 1.5 M elution peaks may be of value In monitoring the course of angiosarcoma: 2 normal controls, .364 and .316; angiosarcoma (pre chemotherapy) , .843; angiosarcoma (post-chemotherapy), .971; angio sarcoma (advanced), 5.00. Characterization of the 1.5 M and 1.25 fNaCl elution peaks by hyaluronldase susceptibility and comparison v elution patterns reported In the literature suggest that the observ shift was from chondroitln 4 and/or chondroitin 6 sulfate to hepara sulfate. Preliminary data indicate that the excretion pattern in vinyl-chloride-associated liver injury other than angiosarcoma is characteristically different from patterns associated with hepatiti cirrhosis, or metastases to the liver. These observations may be related to progressive connective tissue proliferation in anglosarc IMPORTANT The inatructiGn* accompanying tfaii fonn muat ba followed COMPLETELY for iQ abctracts which ait to appear in CLINICAL RESEARCH. Ab stract! which do not conform either will be re typed by the publisher at a coat of S15.00 to the author, or rejected. Reviled June 1976 THIS FORM AS WELL AS THE FORM LETTER OF TRANSMITTAL MUST BE SIGNED BY A MEMBER MEMBER'S SIGNATURE V CMA 003896 CLINICAL RESEARCH Abstract Reproduction Form TYPE name, address, and telephone number of author who should receive correspondence in Box A and complete Box B. Telephone (502) 588-5253---------------- (Area code) office (502) 895-2955 (Area code) home A Name Address Carlo H. Tamburro. M.D.___________________ University of Louisville Cancer Center 511 South Floyd Street Room 535 MDR Building______________________ Louisville, Kentucky 40201 ____ B Date Piymwtt ($10.00) Check munbar Purchase older 1/14/77----------17-1185_______ CHECK Preferred Sub-Specialty Classification: ____ Cardiovascular Clinical ____ Epidemiology Clinkml ____ Pharmacology ____ Dermatology ____ Endocrinology* ____ Gastroenterology ____ Genetics v Health Care Research ____ Hematology Immunology & ____ Conn. Tissue ____ Infectious Disease ____ Metabolism* ____Oncology ___ Pulmonary -- Renal A Electrolyte 'Traditionally, Endocrinology has indudsd paptts dsiiini with tha thyroid, adnnal and pituitary Standi, and fonadi, whila abjtracts dealing with tha parathy roids, calcium and phosphorus metabolism, bona, thyrocatclto nin, diabetas, insulin, decagon, and growth hormooa hm boa* considerad undar MnsMUa. URINARY GLYCOSAMINOGLYCAN EXCRETION PATTERNS IN CHEMICALLY INDUCED LIVER INJURY AND CANCER C. E. Kupchella*and C. H. Tamburro**, Cancer Center, University of Louisville, Louisville, Kentucky, Glycosamlnoglycans (GAG) are essential compounds of connective tissue (CT) matrix and are Increased with CT proliferation. Urinary GAG pat terns were studied in 9 individuals with vinyl chloride (VC) induced chemical Injury, 2 VC Induced angiosarcomas, 8 viral hepatitis, 6 alco holic cirrhosis, 7 non-hepatic cancers, and 9 controls. Hepatic histo logical and electron microscopic (EM) studies were obtained in all but normal controls which were studied biochemically, radioisotopically, and physically. Urines were analyzed for creatinine, total GAG, and uronic acid content. GAG levels (ug uronlc acid/mg creatinine) In con trols were 3,2 + .4, in VC exposed 4.1 +4, and in alcoholic liver dis ease 5.1 + 0,8. In contrast, in angiosarcoma they were 7.6 + 1.6 and hepatic metastasis, 13.8 + .9. Total GAG's were further separated into hyaluronic acid (HA), chondroitln sulfate (CS) and heparin (H) frac tions. Although no significant differences were found in total GAG's 78Z (7/9) of VC exposed had positive CS and negative HA and H fractions in contrast to 9Z (3/32) of the other cases. Further characterization by anion-exchange chromatography showed a shift in the CS fraction composition. Total GAG's eluted at 1.25 M NaCl, were increased and at 1.50 M NaCl decreased. These changes became more pronounced as the disease developed. Enzyme digestion indicates that 1.25 M fraction is heparan sulfate (HS). EM studies showed increased sinusoidal collagen. Since HS is associated with blood vessels, its increased urinary excre tion in early VC injury and angiosarcoma (vascular Injury and sinusoids cell cancer) may be used as an early indicator of chemical Injury and liver cancer formation. IMPORTANT The instructions accompanying this form must be followed COMPLETELY for all abstracts which are to appear in CLINICAL RESEARCH. Ab stracts which do not conform either will be re typed by the publisher at a cost of SI5.00 to the author, or rejected. Revised June 1976 THIS FORM AS WELL AS THE FORM LETTER OF TRANSMITTAL MUST BE SIGNED BY A MEMBER .(%jdbkk MEMBER'S SIGNATURE CMA 003897 Check the most appropriate category below. Absorption O Secretion Q Motility :Z Liver/'Biliary/Bile Salt G.I. Hormones O Morphology Clinical O Immunology/Microbiology 13 Publish this abstract in GASTRO ENTEROLOGY at a cost of $25.00 ($30.00 if billing required) S3 Check payable to the American Gastroenterological Association enclosed. Bill me--instructions enclosed. Do not publish this abstract in GASTROENTEROLOGY. 1C Type Abstract in Space Below TISSUE AND URINARY GLYCOSAMINGLYCANS IN TRANSPLANTABtt HEPATOMAS. C. E. Kupchella, K. L. Curran, E. Drake^^P Kennedy, and C. H. Tamburro. Cancer Center, and Divraion of Digestive Diseases and Nutrition, University of Louisville, School of Medicine, Louisville, Kentucky. The purpose of this investigation was to evaluate: a) the glycosaminoglycans (GAGs) in different behavioral/ histological types of intermuscularly transplanted hepa tomas, b) GAG patterns in tumor tissue in relationship to degrees of fibrosis and necrosis, c) the GAG changes in t'. livers of tumor-bearing animals, and d) urinary GAG ex cretion as a function of tumor growth. Three types of Morris hepatomas, 7777, 5l23tc, and 9618A, which differ ii jrates of growth, metastatic potential, fibrosis, and necrosis, were studied. Urinary and tissue GAGs were ex tracted as cetylpyridinium complexes and measured as uron: (acid. Tissue GAGs were also evaluated histochemically .using alcian blue staining with and without enzyme pretreatment. Tumor tissue exhibited four- to six-fold 'greater GAG levels in the hyaluronic acid (90 +10, 91 + [12, and 111 + 9 vs/ 25 + 3 ig uronic acid/g dry liver, re spectively) and chondroitin sulfate (261 + 29, 217 + 51, [and 208 + 22 vs. 47 + 7 yg uronic acid/g, respectively) ifractions than normal liver; the heparin fractions did not differ significantly (31 +7, 17+4 and 67+11 vs. 39 + 10). The livers of tumor-bearing animals exhibited slightly greater hyaluronic acid levels than normal dowers Moreover, increased urinary GAG excretion was evided^Hlf te jtwo weeks in animals bearing fast-growing tumors. The GAG tissue levels in fast vs. slow-growing tumors were not sig jnificantly different. This further supports our previousl ,reported studies of urinary GAG excretion in human hepatic [angiosarcoma (Curran, K. L., e al., Cancer 40 (6): 30503053) in suggesting that urinary GAG analyses may be usefu in the detection, screening and diagnosis of hepatic cance TYPE name, address, and telephone number of author who should receive correspondence: Namc Charles E. Kupchella, Ph.P., Associate Director, Cancer Center_ . .. 213 MDR Building. University of Louisville, P. 0. Box 35260, Louisville, Ky. Address ---------------------------- Zl------------------- ---------------------------------------------- ----------------------------------- 4023T' Telephone ___ (502) 588-5245-------------------------------------- -------- ----------------- ----------------------- -- IMPORTANT The principal author affirms that the material herein will not have been previously published or presented at any meeting of a national society and that if in the conduct of these studies, human subjects were exposed to risks not^> quired by their medical needs, that the study was approved by an appropriate committee or, if no such committee available and informed consent was needed, it was obtained in accordance with the principles set forth in "The InstW {Avail Guide to DHEW Policy on Protection of Human Subjects." CMA 003898 Check the most appropriate category beio'.v: Q Absorption 3 Secretion O Motility Z Liver/Biliary,'Bile Salt G.I. Hormones Morphology Clinical Immunology/Microbiology 13 Publish this abstract in GASTRO ENTEROLOGY at a cost of S25.00 (S30.00 if billing required) B Check payable to the American Gastroenterological Association enclosed. Bill me--instructions enclosed. Do not publish this abstract in GASTROENTEROLOGY. 1C Type Abstract in Space Below TISSUE AND URINARY GLYCOSAMINGLYCANS IN TRANSPLANTABLE HEPAiONAS. C. S. Kupchella, K. L. Curran, E. Drake, J. Kanr.aiv, and C. h. Taaburro. Cancer Cancer, and Division of Digestive Diseases and Nutrition, University of Louisville, School of Medicine, Louisville, Kentucky. j The purpose of this investigation was to evaluate: a) the glycosaminoglycans (GAGs) in different behavioral/ histological types of interauscularly transplanted hepa tomas, b) GAG patterns in tumor tissue in relationship to degrees of fibrosis and necrosis, c) the GAG changes in tl livers of tumor-bearing animals, and d) urinary GAG ex cretion as a function of tumor growth. Three types of Morris hepatomas, 7777, 5123tc, and 96I8A, which differ ir rates of growth, metastatic potential, fibrosis, and necrosis, were studied. Urinary and tissue GAGs were ex tracted as cetylpyridinium complexes and measured as uroni jacid. Tissue GAGs were also evaluated histochemically using alcian blue staining with and without enzyme pre!treatment. Tumor tissue exhibited four- to six-fold greater GAG levels in the hyaluronic acid (90 +10, 91 + T2, and 111 + 9 vs. 25 + 3 Mg uronic acid/g dry liver, re spectively) and chondroitin sulfate (261 + 29, 217 + 51, and 208 + 22 vs.. 47 + 7 ug uronic acid/g, respectively) fractions than normal liver; the heparin fractions did not differ significantly (31 +7, 17+4 and 67 + 11 vs. 39 + ;10). The livers of tumor-bearing animals exhibited ~~ slightly greater hyaluronic acid levels than normal livers Moreover, increased urinary GAG excretion was evident afte |two weeks in animals bearing fast-growing tumors. The GAG 'tissue levels in fast vs. slow-growing tumors were not slg Inificantly different. This further supports our previousl ^reported studies of urinary GAG excretion in human hepatic angiosarcoma (Curran, K. L., et al., Cancer 40 (6): 3050- 3053) in suggesting that urinary GAG analyses may be usefu in the detection, screening and diagnosis of hepatic cance TYPE name, address, and telephone number of author who should receive correspondence: Namc Charles E, Kupchella. Ph.D., Associate Director, Cancer Center ... 213 MDR Building, University of Louisville, P. 0. Box 35260, Louisville, Ky. Address -- -- -- 1 ~4'0232"` Telephone___(502) 588^245------------------------------------------------------------ --------- ---------- ------------------- IMPORTANT The principal author affirms that the material herein will not have been previously published or presented at any meeting of a national society and that if in the conduct of these studies, human subjects were exposed to risks not re quired by their medical needs, that the study was approved by an appropriate committee or, if no such committee was available and informed consent was needed, it was obtained in accordance with tits principles set forth in "The lnstituinkuG Guide to DHEW Policy on Protecti u of Human Subjects." CMA 003899 CLINICAL RESEARCH Abstract Reproduction Form TYPE name, address, and telephone number or' author who should receive correspondence in Box a and complete Boxes B, C and D. Telephone ^02------ 588-5245------------ (Area code) office 502 (Area code) A Name _ Address Charles E. Kupchella Cancer Center University of Louisville Louisville, Ky. 40232 N0 This abstract is submitted to: t; American Federation for Clinical Research (name of organuation, selected from list on form letter of transmntai) 239-1818 home B (See Rule 15) Date -- Payment (Si3.00) Check number__ Purchase order number. Issued by. (name of institution) A copy of this abstract must be attach to original purchase order to aid in idt tmcacion. i^ I CHECK OSE i tor national meetings only) J Prefer poster session presentation Q I Consider for poster session if not | selected for oral presentation AJ j Do not consider for poster I session under any circumstances C3 D | CHECK S1SGLE SCBSPECULTY CLASSIFICA TION: cod* Cardiovascular* __ no.; (___) Clinical Epidemiology......... ___ Clinical Nutrition................. __ Clinical Pharmacology___ _ Dermatology.......................... __ Endocrinology (see Rule 5) Gastroenterology................. -- -- ! Genetics.................................. -- Health Care Research........ ........ Hematology .......................... ........ Hypertension ....................... .-- Immunology & Rheumatology __ Infectious Disease............... __ Metabolism (see Rule 5).., - Oncology................................ X- Pulmonarv ............................ -- Renal Sl Electrolyte.............__ 'For abstracts submitted to Cardio vascular only, select single subcategory and enter code no. (l-6i inspace above: (1) Clinical: (2) Basic Science; O) Electrophysiology-Dysrhythmias; (4) Echo cardiography. (5) Radiology-Radio nuclides: (6) Other. Subclassification is designed to aid in reviewing process onlv and is independent of program selection. TYPE ABSTRACT HERE/BE SURE TO STAY WITHIN BORDER GLYCOSAMINOGLYCAN CHANGES ASSOCIATED WITH HEPATIC TUMORS: THE CONTRIBUTIONS OF REGENERATION AND NECROSIS. C. E. Kupchella, E. M. Secskas,* J. S, Kennedy,* and E. Espinosa*. Cancer Center and Department of Pathology, University of Louisville, School of Medicine, Louisville, Kentucky. Although glycosaminoglycans (GAGs) have been shown to be elevated in many types of animal and human tumors Including hepatic tumors, the cause and significance of these changes in neoplasia are still open questions. Regeneration and necrosis are operative in hepatic cancer and the purpose of this investigation was to evaluate the GAG changes associated with hepatic regeneration and hepatic necrosis. Regeneration was induced in male Sprague Dawley rats by partial hepatectomy and hepatic GAGs were evaluated at 4, 8 and 12 days post operatively. Necrosis was induced by: a) ligating the medium lobe, b) by resecting and placing median lobes in the peritoneal cavity and c) by resecting median lobes and incubating them in vitro in sterile saline. Analyses were carried out after 5 days of treatment. While regenerating livers exhibited GAG levels chat were not statistically different from sham operated controls or non-operated controls, in vivo necrosis was accompanied by 3-4 fold increases in tissue GAGs. These data suggest that necrosis may make a substantial contribution to the elevated GAG levels found in some tumors. CIin.Ru. 27 PLEASE CHECK ABSTRACT CAREFULLY FOR APPEARANCE BEFORE MAILING BOTH THIS FORM AND THE FORM LETTER OF TRANSMITTAL MUST BE SIGNED BY A MEMBER (RULE 2) Charles E. Kupchella, Ph.D, (please type name) Revised May 1978. MEMBER'S SIGNATURE' ... . . ______ r. CMA 003901 . . r /.7/ Check the most appropriate category below; Absorption Secretion Motility O Liver/ Biliary/Bile Salt G.I. Hormones Morphology Clinical Immunology/Microbiology S Publish this abstract in GASTRO ENTEROLOGY at a cost of S25.00 ($30.00 if billing required) IS Check payable to the American Gastroenterological Association enclosed. Bill me--instructions enclosed. Do not publish this abstract in GASTROENTEROLOGY. Type Abstract in Space Below 11 iURINARY GLYCOSAMINOGLYCAN PATTERNS IN HUMAN HEPATIC ANGIO- SARCOMA, HEPATOMA, AND IN WORKERS AT RISK POR ANGIOSARMtt. K. L. Curran, C. E. Kupchella, J. Sandoz, and C. H. ' ,Tamburrol Cancer Center and'Division' o^Digestive Diseases and Nutrition, University of Louisville, Scho 1 of Medicine, ILouisville, Kentucky. l i i J A previous study reported an abberation in glyccsamlno- glycans (GAGs) eluted from anion-exchange columns with 1.25 land 1.5 M NaCl in the urine of patients with hepatic anglo~! sarcoma. A controlled pilot study examined urinary GAG I patterns in workers at risk for angiosarcoma. Six indi- j jviduals with a history of high vinyl-chloride exposure and ' 'documented liver disease were paired with individuals with ! |a high exposure index to vinyl-chloride but no clinical j jliver disease. Similarly six persons with low exposure I {but abnormal liver function were paired with low exposure/ I {normal liver function individuals. A 24-hour urine was , collected from each Individual and the GAGs analyzed by ' [anion-exchange chromatography. Individuals with clinically lactive liver disease at the time of this study were found { !to have urinary GAG excretion patterns which were similar 1 !to those described in angiosarcoma. No significant dif- | Jferences were found between high vs. low vinyl-chloride 'exposure or between those with inactive liver disease and , {those with normal liver function. GAG excretion was als , 'studied in an additional hepatic angiosarcoma and human ' hepatoma and confirm the reported urinary changes. Th^u {findings support the concept that urinary GAGs are ln-^B creased only in active hepatic disease and may be usefux in, evaluating the degree of activity at the various stages f liver disease in humans* 1 I TYPE name, address, and telephone number of author who should receive correspondence: Charles e. Kuoche.iia. Ph.p.. Associate Director. Cancer Center--.---------------------- . ., 213 MDR Building, University of Louisville, P. 0.-.Box 35260/ Louisville, Ky. Address ---- 4QZ3Z Telephone (502) 588-52.45---------- ----------------------------------------------------------------------------------------------- IMPORTANT The principal author affirms that the material herein will not have-been previously published or Pr**ent^ * mretinp of a nation^1 society and hf*t if ;n <*v ' - .'w studies, human sub- tposed to risks not re- CMA 003900 CLIMCAL research Abs:r:ic: P^eoroduction Form .>? name, address, ind telephone number of iOih'jr #ho shnoid receive oorresponder.ee in Sox s-.d -- Doxe: 3. C and 0. 7e;epn.ne J0238-5245 i vea -:c4*> oiucs 302 [ -\:zl > A .>arrte Charles . Kupchella Cancer Center University of Louisville Louisville, Ky. 40232 ai This ibsirac: is ;ub; an Federation :cr Cli si am .-i f "h 239-1313 j B (See Hule 15) ! Da:: Psymem (513 00)__ Checlt number PurcSuie orfi*: number,, Issued by - (name of ;ns;iiuuon) A copy of thu abstract must be atttc to ongitui purchase order to aid us u. tukauon. TYPE ABSTRACT HERE/BE SURE TO STAY WITHIN BORDER i <. ; a:i'cxosc 1 ' _Pr.a:oterr npaotsiotenralsemsseieontinpz"sresoennlyu)uon __ I Consider for poster wtssion if not ' I ielected for oral presentation 21 ! rvs not consider tor poster i session under any circumstances >D j CHECK SINGLE SCBSFECIALTY CLASSIFICATION; ! cod* ! Cardiovascular*_____ no.; ( t ; C.imcai Epidemiology......... __ I Clinicai Nutrition................. | Clinicai Pharmacology .... __ j Dermatology.......................... __ J1 Endocrinology (see Rule 5) __ Gastroenterology................. __ : Genetics.................................. I Health Care Research......... ! Hematology ......................... - Hypertension ........................ -- Immunology A Rheumatology __ Infectious Disease............... -- j Metabolism (sec Rule 5)... ___ j Pulmonary ............................ ___ Renal St Electrolyte............. __ `For abstracts ubimurt to Cardiovascular only, select single subcategory and enter code no. (l-bi in space above: (1) Clinicai; Id) Basic Soescs (3) Elec:rophysiology*Dysrhythmtasi (a) Echo* cardiography: (5) Radiology-Radionudsdes; (4) Other. Subelaasification is designed to aid in reviewing process onlv and is independent of program teicction. GLYC0SAMIN0GLYCAN CHANGES ASSOCIATED WITH HEPATIC TUMORS: THE CONTRIBUTIONS OF REGENERATION AND NECROSIS. C. E. Kunchella. E. M. Secskas,* J. S. KannedV,* and E. Esolnosa*. 'Cancer Cancel and Department of Pathology, University of Louisville, School of Medicine, Louisville, Kentucky. Although glycosaainoglycans (GAGs) have been shown- to be elevated in many types of animal and human tumors including hepatic tumors, the cause and significance of ehese changes in neoplasia are still open questions. Regeneration and necrosis are operative in hepatic cancer and the purpose of rh-ts investigation was to evaluate the GAG changes associated with hepatic regeneration and hepatic necrosis. Regeneration was induced in mala Sprague Dawley rats by partial hepatectomy and hepatic GAGs were evaluated at 4, 8 and 12 days post operatively. Necrosis was induced by: a) ligating the medium lobe, b) by resecting and placing median lobes in the peritonea, cavity an** c) by resecting median lobes and incubating them in vitro in sterile saline. Analyses were carried out after 5 days of treatment. While regenerating livers exhibited GAG levels chat were not statistically different from sham operated controls or non-operaced controls, in vivo necrosis was accompanied by 3--4 fold increases in tissue GAGs. These data suggest that necrosis may make a substantial contribution to the elevated GAG levels found in some tumors. CtiM.Re/. 270)''3S? please check abstract carefully for appearance before mailing BOTH THIS FORM AND THE FORM LETTER OF TRANSMITTAL MUST BE SIGNED BY A MEMBER I RULE 2) Charles E. Kupchella, Ph.D. <pt3M typ* turn*) Revised May 1973 MEMBER'S SIGNATURE........... ____ r. CMA 003902 (\rt\ea\QM pw3nt HAcm ^jract/vnorJ oecemoe^ ^78 S/|n Oisgo 3097 Early Detection of Disease in Individuals Ex posed to Vinyl Chloride. Richard Greenberg. KID. and Carlo Tainburro. I/O. University of Louisville. Lonisi die. Kentucky This is 3 collaborative comparison study of the effectiveness of ultrasound, naii bed capillary study and urinary glycos.imincglycan excretions in vinyl chloride workers with biochemical dysfunction or histapathologtcally documented injury. It represents an effort to detect early mor bidity in vinyl chloride workers. Included is Dr. Maricq's method of investigating capil laries of the middle and distal phalanges of the fingers, including the nailfold, as a screening test for early changes from vinyl chloride exposure. Another new method ts Dr. Taylor's "grey scale" ultrasonographic method which has not yet been widely adopted for diagnosis of liver disease. The University of Louisville group reported that the first indicator of liver changes occurred in the vascular sinusoids and was reflected by the appearance of mucopolysaccharides in the urine. The results of the massive clinical study in Louisville should soon be able to indicate whether more sensitive criteria of early dis ease may be forthcoming, especially from the liver function testing. The special value of the capillary and ultrasound method re main doubtful at this time for early detection of the effects of exposure to vinyl chloride. 13 193 CMA 003903 Pretinlalion choice: J tatlorin jj, or Poster Session [J u SCIENTIHC PAPER Title: Computer Assisted Morphologic Quantitation of Collagen in Human Liver Biopsies. N'ame(s) of authors, institutional affiliations and addresses: G. H. Barrows, M.D., M.J. Joyce, G.R. Schrodt, M.D., R. Greenberg, Ph.D., C.H. Tamburro, M.D. Departments of Pathology, Epidemiology and Biostatistics, and Medicine, University of Louisville, Louisville, Kentucky 40202. Underline name of author to notify and give -Treliephuone (iAt rea\)_5_02-588-5341 E--xt., AddrMtr Department of Pathology, University of Louisville School of Medicine, P.0. Box 40232, Louisville, Kentucky 40232 -----------------------------------------------------------------------------------------------------------------------------------------------------------Zip------------------- Abstract -- Double space. Because unfixed liver is rarely available for analytical studies, appraisal of the extent of collagen must be made from appropriately stained sections of lives biopsy material leading to judgemental as well as sampling error. We have de veloped a computer assisted morphologic technique to evaluate the distribution of collagen in liver biopsy material. We examined multiple biopsy samples from patients dying suddenly with no history of liver disease to establish normal values. A Hewlett-Packard 9864-A digitizer and 9815-A micro computer were m to quantitate areas of trichrome stainable collagen within the biopsy samples. Random area selection and statistical analysis proved to be important consider ations in determining the experimental model and these factors were readily in corporated into the calculator program. Stainable collagen estimates in normal liver varied from 0 to 6.IX (mean = 1.25X) in patients with no evidence of liver disease and showed considerable variation even in different biopsies from the same patient. As anticipated, subcapsular biopsies had more collagen than deep biopsy; however, the specimens from several deep biopsies often had significant variation in collagen content. Differences in central, mid-zonal and portal collagen could be obtained using this method. The computer assisted morpholog ical technique for evaluating hepatic collagen appears to be readily adaptable to clinical studies. This study of normal liver implies considerable care must hp takpn hpfnr* the rlinical diagnosis of significant flbEgS^S can ** made-- (Limit abstract to 250 words r less) CMA 003904 (OVER) PreM'tit.it choice; 1 l.itlorin M ur .''c^muii cj SCIENTIf-IC PAPER T.`.I?: Computer Assisted Morphologic Quantitation 0/ Collagen in Human Liver 3icpsies. Name(s) of authors, instihiHonaJ affiliations and addresses: G. H. Barrows, M.D., M.J. Joyce, G.R. Schrodt, M.D., R. Greenberg, Ph.D., C.H. Tamburro, M.D. Departments of Pathology, Epidemiology and Biostatistics, and Medicine, University of Louisville, Louisville, Kentucky 40202. Underline name of author to notify and give Telephone (Area) 502-588-5341 ^________ Department of Pathology, University of Louisville School of Medicine, P.0. Box 40232, Louisville, Kentucky ,, 40232 --------------------------------------------------------------------------------------------------------------------- -Zip--------------------- Abstract -- Double space. 1 Because unfixed liver is rarely available for analytical studies, appraisal ot the extent of collagen must be made from appropriately stained sections of live biopsy material leading to judgemental as well as sampling error. We have de veloped a computer assisted morphologic technique to evaluate the distribution of collagen in liver biopsy material. We examined multiple biopsy samples from patients dying suddenly with no history of liver disease to establish normal values. A Hewlett-Packard 9864-A digitizer and 9815-A micro computer were used to quantitate areas of trichrome stainable collagen within the biopsy samples. Random area selection and statistical analysis proved to be important consider ations in determining the experimental model and these factors were readily in corporated Into the calculator program. Stainable collagen estimates in normal liver varied from 0 to 6.1* (mean * 1.25*) in patients with no evidence of liter disease and showed considerable variation even in different biopsies from the same patient. As anticipated, subcapsular biopsies had more collagen than deep biopsy; however, the specimens from several deep biopsies often had significant variation in collagen content. Differences in central, mid-zonal and portal collagen could be obtained using this method. The computer assisted morpholog ical technique for evaluating hepatic collagen appears to be readily adaptable to clinical studies. This study of normal liver implies considerable care must hm fakpn h#fnr the cliniral diagnosis of sionificant fibCPSlS can be.made.-----(Limit abstract to 250 words or loss) (OVER) CMA 3905 \;nciican Association For'1 lie Study of Fixer Pisceses ABSTRACT FORM _15_ Type name, address, and telephone number of author vdw should rcceixc correspondence in Box C. and complete Boxes A and B. iForoffice > TYPE ABSTRACT BELOW. BE SURE TO STAY WITHIN BORDER Cost of Publication Payment enclosed ($25.00) Check no_____ *2Sl_______ (Payable to Elsevier Publishing Co.) or Payment delayed ($30.00) Purchase order no.. Issued by________ lln'liturion Name) A copy of this abstract must be at tached to original purchase order to aid in identification. B IMPORTANT The principal author affirms that the material herein: 1) will not have been previously published or presented at any national meeting of a national so ciety; 2) that if human subjects were exposed to risks not required by their medical needs, the study was ap proved by an appropriate committee or. if no such committee was available and informed consent was needed, it was obtained in accordance with the principles enunciated in "The Institu tional Guide to DHFW Policy on Pro tection of Human Subjects": or 3) any animal studies conform with the "Guiding Principles in the Care and Use of Animals" of the American Physiological Soyiejs (/'X'Jj . \ ; (Jv-l&r&'T) ISijin.tlu.f t`f Pnns.ip.tl Author) EARLY HEPATIC HISTOLOGICAL ALTERATIONS AMOK CHEMICAL (VINYL MONOMER) WORKERS. C.H. Tamburro, L. Makk, and IS. Popper, Division of Digestive Diseases and Nutrition and St. Anthony Hospital, University of Louisville, Louisville, Ky., and Stratton Laboratory for Study of Liver Disease, , Mt. Sinai School of Medicine, New York, N.T. Industrial vinyl chloride exposure is associated with hepatic subcapsular, portal, and perisinuscmdal fibrosis land hyperplasia of both sinusoidal cells and. hepatocytes. jEarlier histological studies, mainly on autopsy material, 1 lindicated that focal mixed hyperplasia (of hepatocytes and jsinusoidal cells) is the early histologic alteration indi cative of exposure. To substantiate this observation and .its potential use in screening workers, liver biopsies from ,55 persons were investigated in double blind duplicative fashion; 34 were workers exposed to chemicals with hepatic ibiochemical abnormalities, and 21 were a comparison group composed of 8 non-exposed persons and 13 e^osed workers without hepatic abnormalities who had biopsies for noniliver related reasons. Of the exposed workers with abnor malities, 12 (35%) had a hepatic lesion consistent with [exposure; 6 (18%) had focal hepatocytic h.ypplasia; 6 (18%) had focal mixed hyperplasia. In contrast, eoly 5 of the jcomparison group (2 exposed and 3 non-exposed persons) had : similar findings; 1 exposed and 2 non-exposed had only jfocal hepatocytic hyperplasia. The other exposed worker j had focal mixed hyperplasia as did the other non-exposed j (Individual who, subsequent to the biopsy, vis found to have i 'angiosarcoma. Thus, only 19% of the comparison group demonstrated hepatic lesions with chemical oposure, half j of whom had worked with chemicals. Of the 55 biopsies, 22 j were re-read double blindly 2 to 4 times ov a 2-year Jperiod. In addition, 9 of these Individuals had second or j third biopsies also read double blindly. Delicate readings were identical in 86%, 9% had 1 positive and 1 questionable | reading, and only 5% were read differently. Focal hepato- j cytic hyperplasia, in addition to the previoasly described ! pixed hyperplasia, appears to be the earliest identifiable j changes consistent with chemica.1 exposure and seem to be , precursors of angiosarcoma. They are useful in the screening of chemical workers. C NL.m- Carlo H- Tamburro, M.D.. Professor of Medicine A,, d,,dresx_5_1_1___S_.__F__l_oyi_d,l__R_o_o_m___5__3_5_ Louisville, KY. 40202 Telephone. Office: 588-5252 (502) uome: JM-JSSS (502) due.i lode) code) CMA 003906 16 Chloroacetaldehyde-induced Damage to Bacillus subtilis. A.D. Laumbach, U.N. Streips* and J.L. Wong. Bureau Foods, FDA and U. Louisville, Louis ville, Ky. Chloroacetaldehyde (CAA), a proposed metabolite of vinyl chloride meta bolism, has been shown to be mutagenic in the Salmonella assay system, and to specifically inhibit the growth of a recombination-deficient mutant of Bacillus subtilis (Ellmore, et_ a^L BBA, 442: 405, 1976). We have examined further the biological activity of this compound. First, CAA caused a significant increase in induced mutations to streptomy cin resistance. Using i). subtilis 168, the relative mutation frequency (treated 5mM CAA/control) was 13.6, while using an her mutant of B. subtilis the frequency was 14.2. Secondly, a survey of the available repair-deficient mutants of B_. subtilis revealed, that the only strains sensitive to CAA were of the rec-genotype. Furthermore, the several rec-mutants, shew three types of response to CAA. Strong killing by CAA (inhibition of growth of 10 mm or more) was exhibited on the recA, recB, rec-4 (GSY1616), and recD27 (GSY1627) strains. Intermediate killing (3mm-8mm) was shown with strains bearing the recE, recF, and rec-13 (BD246) loci. All other strains examined (recC, recH, mtc-41, uvr, her, polA, as well as several repair-proficient cultures) grew in the presence of 1.1M CAA. These studies are part of an ongoing program sponsored by the MCA to determine the molecular basis for vinyl-chloride induced carcinogenesis. CMa 0^3907 16 Chloroacetaidenyde-mduced Damage Co Bacillus subtilis. A.D. Laumbueh, U.N. Streips* and J.L. Wong, Bureau Foods, FDA and L'. Louisville, Louis ville, ;;v. dhlorja^ccalueiiyaa (.CAA), a proposed metabolite of vinyl cnloride meta bolism, has been shown to be mutagenic in the Salmonella assay system, and to specifically inhibit the growth of a recombination-deficient mutant of 3acillus subtilis (Ellmore, et_ ai_ BBA, 442: 405, 1976). We have examined further Che biological activity of this compound. First, CAA caused a significant increase in induced mutations to streptomy cin resistance. Using B_, subtilis 163, the relative mutation frequency (created 5mM CAA/control) was 13.6, while using an her mutant of IJ. subtilis the frequency was 14.2. Secondly, a survey of the available repair-deficient mutants of B_. subtilis revealed, that the only strains sensitive to CAA were of the rec-genotype Furthermore, the several rec-mutants, show three types of response to CAA. Strong killing by CAA (inhibition of growth of 10 mm or more) was exhibited on the recA, recB, rec-4 (GSY1616), and recD27 (GSY1627) strains. Intermediate killing (3mm-8mm) was shown with strains bearing the recE, recF, and rec-13 (BD246) loci. All other strains examined (recC, recH, mtc-41, uvr, her, polA, as well as several repair-proficient cultures) grew in the presence of 1.1M CAA. These studies are part of an ongoing program sponsored by the MCA to determine the molecular basis for vinyl-chloride induced carcinogenesis. CMA 003908 A13oTT iO Tiiley:. 2 or 5 18 INHIBITION OF INTERFERON INDUCTION AS A SCREEN FOR THE CARCINOGENIC ' POTENTIAL OF CHEMICALS. Gerald Sonnenfeld," Mary Carol BarneS, and Uldis N. Streips, Department of Microbiology and Immunology, Univ. of Louisville School of Medicine, Louisville, Kentucky 40232. Type I Interferon was produced after challenge of mouse embryo fibroblasts with Newcastle disease virus. This production of interferon could be sharply reduced by pretreatment of the cells with the known car- ; cinogens benz-a-pyrene, aflatoxin, 2-aminofluorine, and the #4 fraction of tobacco smoke condensate. This finding suggested that inhibition of interferon induction <1--1--I) may serve as an indicator for the carcino genic potential of chemicals. We tested the screening potential of this assay by testing two chemical analogs, methylmethanesulfonate (MMS) a car cinogen and mutagen and ethylmethanesulfonate (EMS) a non-carclnogenlc mutagen. Both of these chemicals are active in the Ames Salmonella assay but can be separated in "SOS" induction assays such as the Comptest. In ' the I-I-I assay, MMS was highly positive while EMS showed no inhbltlon f Interferon induction. Thus, the I-I-I assay appears to have th potential to screen for carcinogens. Preliminary results using the carcinogen chloracetaldehyde and its non-carcinogenic analogs chloroethanol and chloroacetic acid support this idea. Following further verification of the dis crimination potential of the I-I-l assay, this test could become an integ ral part of a comprehensive battery of tests for chemical carcinogenicity - Ahj-t This A [lie Ah original Indica V* ' i in .1 .i. Ki.v, J, )u70. A.ilimissi'V.i aT. early dttte in Lo appreciated LV-M-ly y: , ;!' u';i`o.i'iai within the Line lines. Your afj>uttct i!l appear ii isua -`l' lh Ii.iu'it ! of C 1 ..v.d Ileiiiatol- sv unci Oue.olo^v exactly as sunnailced. VUr.se sene : clw:- ,, AhMiirr; 1;. .! n.jruina ,uv 'a'n.tccl on the r.-veiae side of this page. 11* ...v.-nate category for your abstract. vI I : .>d i -..I l t.l'.tl : u rit;.) 'L .-'inc f . ' . * f 11r ! . i * ' (! .led Tri\ : '/uc 10!i no- < Gerald Sonnenfeld, Ph.D. Department of .Microbiology and Immunology, University of Louisville School of Medicine, health Sciences Center Louisville, KY 40232. . 0 : JiU.VI ; , ;- s.A. CMA 003910 17 Chloroacetaldehyde-induced Damage to Bacillus subtilis. A.D. Laumbach, U.N. Streips*, and J.L* Wong. Bureau Foods, FDA, and University of Louisville, School of Medicine, Health Sciences Center, Louisville, Ky. 40232 (USA). Chloroacetaldehyde (CAA), a proposed metabolite of vinyl chloride metabo lism, has been shown to be mutagenic in the Salmonella assay system, and to specifically inhibit the growth of a recombination-deficient mutant of Bacillus subtilis (Ellmore, e al., BBA, 442:405. 1976). We have examined further the biological activity of this compound. First, CAA caused a significant increase in induced mutations to streptolycin resistance. Using ]}. subtilis 168, the relative mutation frequency (treated 5mM CAA/control) was 13.6, while using an her mutant of 11. subtilis the frequency was 14.2. Secondly, a survey of the available repair-deficient mutants of B^. subtilis revealed, that the only strains sensitive to CAA were of the rec-genotype. Furthermore, the several rec-mutants, show three types of response to CAA. Strong killing by CAA (inhibition of growth of 10 mm or more) was exhibited on the recA, recB, rec-4 (GSY1616), and recD27 (GSY1627) strains. Intermediate killing (3mm-8mm) was shown with strains bearing the recE, recF, and rec-13 (BD246) loci. All other strains examined (recC, recH, mct-41, uvr, her, polA, as well as several repair-proficient cultures) grew in the presence of 1-.1M CAA. The effects of CAA on the biological activity of DNA molecules in vitro are being studied. These studies are part of an ongoing program sponsored by the Manufacturing Chemists Association to determine the molecular basis for vinyl-chloride induced carcinogenesis. CMA 003909 i 2 or 5 13 INHIBITION Or INTERFERON INDUCTION AS A SCREEN FOR THE CARCINOGENIC POTENTIAL OF CHEMICALS. Gerald Sonnenfeld," Mary Caxol BarneS, and Uldis N. Streips, Department of Microbiology and Immunology, Univ. of Louisville School of Medicine, Louisville, Kentucky 40232. Type I interferon was produced after challenge of mouse embryo fibroblasts with Newcastle disease virus. This production of interferon could be sharply reduced by pretreatment of the cells with the known car cinogens benz-o-pyrene, aflatoxin, 2-aminofluorine, and the #4 fraction of tobacco smoke condensate. This finding suggested that inhibition of interferon induction (I-I-I) may serve as an indicator for the carcino genic potential of chemicals. We tested the screening potential of this assay by testing two chemical analogs, methylmethanesulfonate (MMS) a car cinogen and mutagen and ethylmethanesulfonate (EMS) a non-carcinogenic mutagen. Both of these chemicals are active in the Ames Salmonella assay but can be separated in "SOS" induction assays such as the Comptest. In the I-I-I assay, MMS was highly positive while EMS showed no lnhbieion of interferon induction. Thus, the I-I-I assay appears to have the potential to screen for carcinogens. Preliminary results using the carcinogen chloracetaldehyda and its non-carcinogenic analogs chloroethanol and chloro- , acetic acid support this idea. Following further verification of the dis crimination potential of the I-I-I assay, this test could become an integ ral part of a comprehensive battery of tests for chemical carcinogenicity.. , h'J !Y` i - ' U' ` Itllo l u fuin, J, lyTfl is.ilunissi-u; ii' This h . ' ;v-;i/. i 1* *r ' , i It'Jl * ill '.rtl s' ; zlie Ah -y iSUS >;' Ml , \>\\i i'i.! .-f C l ..I' .it ICt'lZiitttjl' ; '"flprimi: -. . cU1;-. . , it' .ii >.t iiiv [niiich'.-.- 1 - t...-.\ i r *' } 1 h it j t i 1 i ,1 l i ! u.f (L ; :u rit,,-) I iu^ r, . ' ill.; 1 .1 i (: .Iim' Tri\ : : early date iti l.c appr^-iatei! y.-,,. ( . , t, ;ri,,Gerald Sonnenfeld, Ph.P. Department of Microbiology _ and Immunology, University of Louisville School of Medicine, Health Sciences Genter Louisville, KY 40232. G-Mvl ,|. .it |.'... ...t .;V ' I7 S.A. CMA 0039H 19 Reproduction Copy for PROGRAM AND ABSTRACTS of the 30TH ANNUAL MEETING OP THETISSUE CULTURE ASSOCIATION, INC. Properties of 14 Week Maintenance Cultures of PLC/PRF/5 Cells. P.B. JOHNSTON5-, E. Espinosa, S. Chia and S. Caple. Univ. of Louisville, Louisville, Ky 40232__________________________________ ____________________ _______ __________ This established cell line was originally described as being short lived with in a given passage, with considerable cell loss by day 6 to 8. However.after 80 passages, we found that culturing in glass roller bottles at 0.25 RPM allowed maintenance for at least 34 weeks with cells exhibiting excellent morphology without sloughing. After treating monolayers with Hoechst 33258 DNA stain, intense cytoplasmic fluorescence was seen in a proportion of the cells, possibly a function of the subviral hepatitis B genome. HBsAg of culture supernates was monitored and solid phase RIA assay revealed 31 to 47 ratio-units/0.2 ml after 2 to 14 weeks in medium with 10% fetal calf serum. The 14 week old cultures were extremely well adapted to maintenance in serum free medium for 14 additional days when supplementing with 300 ng/ral of insulin, in contrast to 1 week cultures which survived well for only 5 days without serum.. The HBsAg titers of supernates from these 1 week or 14 week cultures, were very similar to the previous harvests in 30% serum. In addition, both young and 14 week cultures were similar in their rate of acid production; and in that both elaborated scrum albumin, fibrinogen, transferrii and a]pha-2 macroglobulin, the proteins tcsLcd to date. CMA 003912 ABSTRACT MUST BE RECEIVED AT SOCIETY OFFICL \j Y IUE5DAY, DECEfiBER 19 Please consider this abstract for inclusion in (he tentatively lifted mimsymposium, m43 Chemical Carcinogenesis Indicile below (he numbers and (ides of sessions in which your absiract migh be programed (see Topic Category List); It 4Q7------ Title Tumor. Biology^--------------------------------- M ( A iTil le Tnmrtr Mr di* g*n.c la/a FASEB Abstract Form 00 NOT FOLD THIS FORM Mail to: American Association of Immunologist: 9650 Rockville Pike Bethcsda, Maryland 20014 IMMUNOLOt LIVER-SPECIFIC F ANTIGEN IN TRANSPLANTABLE HEPATOMAS HAVINT DIFFERENT GROWTH RATES. Enrique Espinosa, Sue Chia*, Sam Caple* and Charles Kupchella*. Dept. Pathology and Cancer Center, Univ. of Louisville School of Medicine, Louisville, KY 40232 In the course of studies of antigenic changes associatec with liver neoplasia we measured the relative concentratiot of liver-specific F antigen in fast (7777), slow (9618A) ar medium (5123tc) growing Morris hepatomas in comparison wit! its concentration in normal adult rat liver. -F antigen was solubilized from each tissue by aqueous homogenization. T1 supernatant containing the antigen was then tested by a double diffusion dilution assay capable of detecting a min: mum of 1-2% the concentration of antigen present in normal rat liver. The F antigen antiserum used was prepared ac cording to Fravi and Lindenmann by immunizing CBA mice witf BALB/c mouse liver extract. F antigen was undetectable in the fast growing hepatoma and ranged from less than 22^o 10% of the normal liver concentration in the slow gri^Bg one. The medium growing hepatoma showed similar or h^^er concentration than normal rat liver. Thus, the level of F antigen' in these tumors does not appear to correlate wit their rates of growth. (Supported tji iya-at J>y the Manu facturing Chemists Association) All compounds that are designated by code or initial letters must be identified adequa in the abstract, eg., MJ-I999: 4-(2-isopropyIamino-l-hydroxyethyl) meihanesulfon lide hydrochloride. MAILING ADDRESS OF FIRST AUTHOR (Please Print or Type. Provide full name rather than initals.) Enrique Espinosa, M-D. .... 511. .South .Floyd .S.tre.et ___Louisy.ill#........KY___ zip .40232 -- Telephone No.: Area CodP.?.. # 588-5525.............. CMA 003913 Each Abstract Form submitted MUST BE SIGNED by a member of the AMERICAN ASSOCIATION OF IMMUNOLOGISTS. Enriqye Espinosa IMeitiSw's kitin' Pleas* Pnnt or bide tuU name > /OtAMVAM ^7; 7 (Member's Signature) Member's telephone no.: Area Code.. 502 . #.588.-55.25... ABSTRACT MUST BE RECEIVED AT SOCIETY OFFICE uY lots DAY. DECEf BER 19 P'.case consider ihis abstract for inclusion in ihe icmauvely Tinted mims> mposium, 3 . ihemic^LXatcinogejiesis. Indicate below >.he numbers and min of sessions m which >nut abstract r-vjh be programed (see Topic Category Lis<); if ?n . Tide Liver-Paehophys-i-oLogy---------------:/ +Q2-------Tuie TtfaoE-Biok&gy----------------------------------3/ Tide Tumor. Specifia- Antigens------------- 1979 FASEB Abstract Form n DO NOT FOLD TM!S i-'PRM Mail to: American Association of ! .-nmunologis' 9550 Rockville Pike Bethcsda, maryland C0C14 presentation preference Preferred choice (CHECK ONE ONLY) 3 Poster presentation Slide presentation O Indifferent If your first choice is unavailable, you wilt SI Accept the alternative Withdraw the abstract 16 mm. films (silent or optical sound) are permitted if essential to 10-minute slide session presentation. MOVIE YES......... NO.,.3.... Submit justification by letter to Society Office, with abstract. IMPORTANT Before you complete this form: See the enclosed rules for eligibility of papers. Set sample abstracts and typing instruc tions on reverse tide. Use enclosed Check List for preparation of abstract. IMMUNOLO LIVER-SPECIFIC F ANTIGEN IN TRANSPLANTABLE HEPATOMAS HAVIN DIFFERENT GROWTH RATES. Enrique Espinosa, Sue Chla*, Saro Caple* and Charles Kupchella*. Dept. Pathology and Cancer Center, Univ. of Louisville School of Medicine, Louisville KY 40232 In the course of studies of antigenic changes associate with liver neoplasia we measured the relative concentratio of liver-specific F antigen in fast (7777), slow (9618A) a medium (5123tc) growing Morris hepatomas in comparison wit its concentration in normal adult rat liver. `F antigen wa solubilized from each tissue by aqueous homogenization. T supernatant containing the antigen was then tested by a double diffusion dilution assay capable of detecting a min mum of 1-2% the concentration of antigen present in normal rat liver. The F antigen antiserum used was prepared ac cording to Fravi and Lindenmann by immunizing CBA mice wit BALB/c mouse liver extract. F antigen was undetectable in the fast growing hepatoma and ranged from less than 22 to 10% of the normal liver concentration in the slow growing one. The medium growing hepatoma showed similar or higher concentration than normal rat liver. Thus, the level of F antigen in these tumors does not appear to correlate wi their rates of growth. (Supported in part by the Manu facturing Chemists Association) The qri*inl typed copy of this atmraa form (for reproduction by photo-offset) with - 8 photocopies, - one tel of author index card^ three ibstnct idciuificMioa cards, - and return Program Confirmation Postal All compounds that are designated by code or initial letters must be identified edequa in the abstract, e.f., MJ-1999: 4-(2-isopropylamino-l -hydroxyethyl) melhanesulfon lide hydrochloride. must reach the Society office NO LATEX THAN TUESDAY. DECEMBER 19. Each Abstract Form submitted MUST BE SIGNED by a member of the AMERICAN ASSOCIATION OF IMMUNOLOGISTS. MAILING ADDRESS OF FIRST AUTHOR (Pi Print or Type. Provide fun name rather than initals.) Enrique Espinosa, M.D. .... 511 .South .Floyd .Sxreet. _________ Enriqye Espinosa IM(irider s Name. PPleas* Pont or tufl vm. I / U^7tJL T ViMemtMf s Signature! Member's telephone no.: Area Code.. 50%.............. #. 588-55.25... ....Louisville....... KY.... Zip .49.23.2............. Telephone No.: Area Cod^.P?.. #.5.88--5525............... CMA 003914 sVillCl I i ^ i .L International Association For The Study Of [he Liver 2: Abstract Submission Dates: IASL - June 15, 1980 AASLD- July 15,1980 ABSTRACT FORM Abstract submitted to: AASLD ^ IASL Type name, address, and telephone number of author who should receive correspondence in Box D, and complete Boxes A and B and C. (For office use) CATEGORIES - Indicate category of choice 1. Immunology 2. Collagen Fibrosis 3. Morphology 4. Viral Hepatitis 5. Chronic Hepatitis (5) Metabolic Liver Disease 7. Complications of Liver Disease 8. Alcohol and the Liver 9. Bile Acids 10. Bilirubin Cholestasis 11. Pediatric Liver Disease 12. Miscellaneous Cost of Publication Payment enclosed ($25.00) Check no.____ ^62______ (Payable to Elsevier Publishing Co.) or Payment delayed ($30.00) Purchase order no-------------------------- Issued by (Institution Name) A copy of this abstract must be at tached to original purchase order to aid in identification. Payment must be made in U.S. Currency. B IMPORTANT The principal author affirms that the material herein: I) will not have been previously published or presented at any national meeting of a national so ciety: 2) that if human subjects were exposed to risks not required by their medical needs, the study was ap proved by an appropriate committee or, if no such committee was available and informed consent was needed, it was obtained in accordance with the principles enunciated in "The Institu tional Guide to DHEW Policy on Pro tection of Human Subjects"; or 3) any animal studies conform with the "Guiding Principles in the Care and Use of Animals " of the American Physiological Society. I Signature of Principal Author) TYPE ABSTRACT BELOW. BE SURE TO STAY WITHIN BORDE OXIDATIVE AND DETOXIFYING ABILITY OF LIVER MESENCHYME PARENCHYMAL CELLS IN THE METABOLISM OF XEN0BI0TICS Du, J. and Tamburro, C.H., Liver Research Laboratory Dept, of Medicine, University of Louisville, Louisvi. The metabolism of xenobiotics, by the liver most oftt volves oxidation via mixed function oxidase (MFO) am hydryl detoxification with glutathione. Hepatic tox: and carcinogenicity are modified by these two pathwa} Vinyl monomer (e.g. vinyl chloride) exposure is assc with liver parenchymal cell (hepatocytic cell, HC) tc and mesenchymal cell (sinusoidal cell, SC) carcinoger even though the hepatocyte is the main cell for xenofc oxidation. The metabolic ability of these two cell g was assessed by isolation of SC using pronase digest! and HC using collagenase profusion. Viability assess 902 by trypan blue exclusion and cross contamination mated (< 12) by pryuvate kinase activity. MFO activi (demethylation benzphetamine), glutathione transferas (E) using (1,2, epoxy-(P-nitro nitrophenoxy) propane substrate, glutathione transferase A,C and E (A) usit nitrobenzyl chloride) as a substrate and glutathione tase (GR), were determined in subcellular isolated HC SC fractions. Results based on protein content (N mo mln/mg protein) were: E A GR MFO SC 42,9 + 17 18.5 + 6 47.4 + 2 7.49 + HC 82.5 + 11 215.0 + 61 65.0 + 7 16.0 + and activity in mole/10^ cell were: E SC 1.16 HC 65.6 A 0.5 220. GR 1.28 52. MFO 0.026 4.7 These results demonstrate both the greater average, a as total, cellular ability of HC to oxidize xenobioti More importantly, it demonstrates an increased abilit detoxify the xenobiotic's toxic metabolites. This gr ability may account for less severe HC injury and pla important role in preventing HC malignant, transformat In contrast, SC ability to oxidize but its lesser abi to detoxity xenobiotic. may account for SC's potential --ai-irgTvap-t--t-rana formation.- -- D Nam* Carlo H. Tamburro. M.D. Address Division of Digestive ill""South Floya - MuR Diseases & blag. Km. Nutrltl 535 Louisville, Kentucky 40292------------------------ March 1980 CMA 003916 Telephone, Office: 502-588-5252 .Home: 502-895-29 (area code) (area code) ABSTRACT MUST BE RECEIVED AT SOCIETY OFFICE BY TUESDAY, DECEMBER 19 Pirate consider ihis abstract Tor inclusion in the tentatively titled minitympotium. M Ai : Chemical Carcinogenesis Indicate below the numbers and titles of sessions in which your abstract migh be programed (see Topic Category List): 1/ 430 Title Liver Pathophysiology 21 AQ3------ Title Tumor Biology--------------------------------- u tT09------TMle Tumoir~Speclf Hr~Anr Igens------------- 1979 FASEB Abstract Form 21 DO NOT FOLD THIS FORM Mail to: Dr. Kenneth M. Endicott, Executive Office American Association of Patholo 9650 Rockville Pike BethxHa, Maryland 20014 . --------- PATHOLOGY TWO LIVER ANTIGENS UNDETECTABLE IN A FAST GROWING niNE OF TRANSPLANTED HEPATOMA (MORRIS HEPATOMA >777). Enrique Espinosa, Sam Caple*, Charles KupcheH^^ and Sue Chia*. Dept. Pathology and Cancer Center, Univ. >f Louisville School of Medicine, Louisville, KY 0232 During investigations of antigenic changes in liver tumor we examined fast (7777), medium (5123tc) and slow (9618A) growing Morris hepatomas for presence of liver antigens. Analyses of normal liver and hepatoma tissue saline extracts were performed by immunodiffusion methods using.rabbit anti serum to rat liver extract absorbed with pooled normal rat plasma. Hepatoma 5123tc and 9618A gave immunoelectrophoreti patterns similar to normal liver. In contrast, hepatoma 777 showed 2 arcs of precipitation missing. This was confirmed by the finding that after absorption of antiserum with hepa toma 7777 two lines of precipitation given by normal liver extract persisted whereas absorption with extract of normal liver, hepatomas 5123tc or 9618A abolished all lines of_pre- cipitatlon. The liver antigens found to be absent in toraa 7777 were characterized as proteins relatively unliKle to heating and acid pH and unrelated to liver-specific F antigen. In Sephadex G-200 and Bio-Gel A-5m columns these antigens eluted as proteins of approximately 51,000 and 240,000 daltons respectively. The fact that 7777 is the fastest growing and least differentiated of the tumors studi suggests a possible functional relationship between the absent antigen and these properties^ (Supported in part by the Manufacturing Chemists Association) All compounds that are designated by code or initial letters must be identified adequately in the abstract, e.g., MJ-1999: 4-(2-isopropyIamino-I-hydroxyethyi) methanesulfontni tide hydrochloride. MAILING ADDRESS OF FIRST AUTHOR (Please Print or Type. Provide full name rather than initals.) Enrique Espinosa, M.D. 511 South Floyd Street Louisville KY 502 Telephone No.: Area Code 40232 Zip............... 588-5525 Each Abstract Form submitted MUST BE SIGNED by a member of ihe AMERICAN ASSOCIATION OF PATHOLOGISTS. Enrique Espinosa ^ (MiMcrotitr't Nm Pleas* Print ov^Tvp* fovtdt M name.) _ IMatnbat s Siganatuure) >?502 588-5525 Member's telephone no.: Area Code. ......... #................. Signing member, are you willingto chair a session? ( ) yes. category #----------- C ) no If yes, are you also willing to give an overview or introductory ( )yes ( )no CMA 003915 , U1II.J i International Association I I'lte Stuu y Ot I ho La ?- Aor~c: Sabir,lisioa Dates. IA5L - June i 5, I960 ABSTRACT FORM AASLD July 15, 1SS0 Abstract submitted to: AASLD "B IASL Type name, address, and telephone number of aath ;r v be sitould receive ce ;n gcx n( -~,j :c^-;eca Boxes A . 3 ; C. 2- CATEGORIES - Indicate category of choice . !;n:nt,no!va> 2. Collagen Fibrosis 3. Morphology 4. Viral Hepatitis 5. Chronic Hepatitis (0 Metabolic Liver Disease 7. Complications of Liver Disease 8. Alcohol and the Liver 9. Bile Acids 10. Bilirubin Cholestasis 11. Pediatric Liver Disease 12. Miscellaneous A Cost of Publication Payment enclosed ($25.00) Check no, ^2 (Payable to Elsevier Publishing Co.) or Payment delayed ($30.00) Purchase order no Issued by I Institution Name) A copy of this abstract must be at tached to original purchase order to aid in identification. Payment must be made in U.S. Currency. B IMPORTANT The principal author affirms that the material herein: I) will not have been previously published or presented at any national meeting of a national so ciety; 2) that if human subjects were exposed to risks not required by their medical needs, the study was ap proved by an appropriate committee or. if no such committee was available and informed consent was needed, it was obtained in accordance with the principles enunciated in "The Institu tional Guide to DHEW Policy on Pro tection of Human Subjects": or 3) any animal studies conform with the "Guiding Principles in the Care and Use of Animals" of the American nosological Society. (Signature of Principal Author) TYPE ABSTRACT 3ELOW. 3E SURE TO STAY WITHIN 30RDI OXIDATIVE AND DETOXIFYING ABILITY OF LIVER MESENCHYM PARENCHYMAL CELLS IN THE METABOLISM OF XENOBIOTICS Du, J. and Tamburro. C.H.. Liver Research Laboratori Dept, of Medicine, University of Louisville, Louisvi The metabolism of xenobiotics, by the liver most oft volves oxidation via mixed function oxidase (MFO) an hydryl detoxification with glutathione. Hepatic tox and carcinogenicity are modified by these two pachwa Vinyl monomer (e.g. vinyl chloride) exposure is assi with liver parenchymal cell (hepatocycic cell, HC) ti and mesenchymal cell (sinusoidal cell, SC) carcinoge; even chough the hepatocyte is the main cell for xenoi oxidation. The metabolic ability of these two cell was assessed by isolation of SC using pronase digest: and HC using collagenase profusion. Viability asses: 90Z by trypan blue exclusion and cross contamination mated (< IS) by pryuvate kinase activity. MFO activ: (demethylation benzphetamine), glutathione transfers: (E) using (1,2, epoxy-(P-nitro nitrophenoxy) propane substrate, glutathione transferase A,C and E (A) us it nitrobenzyi chloride) as a substrate and glutathione taee (GR), were determined in subcellular Isolated H( SC fractions. Results based on protein content (N me min/mg protein) were: E A GR MFO SC 42.9 + 17 18.5 + 6 47.4 + 2 7.49 + HC 82.5+11 215,0+61 65.0+7 16.0 + and activity in mole/10^ cell were: E SC 1.16 HC 65.6 A 0.5 220. GR 1.28 52. MFO 0.026 4.7 These results demonstrate both the greater average, a as total, cellular ability of HC to oxidize xenobioti More importantly, it demonstrates an increased abillt detoxify the xenoblotlc's toxic metabolites. This gt ability may account for lesa severe HC injury and pla important role in preventing HC malignant transformat In contrast, SC ability to oxldlza but its lesser abi to detoxity xenobiocic may account for SC'a potantial -mnlAgiien6--tanfrBafc4otVf -- - --------------------- D Nam* Carlo H. Tamburro. M.D.----------------------------- AAA Division of Digestive Diseases & Nutrltl Address bU _ MDK Bldg.--VSTTJ5------- Louisville Kanteehy 40292-- -- C Indicate if the Presentor is: Fellow Student Mann 19SO Tetephoiw, lares code)(area code) CMA 003917 ASPET/SOT 1982 Louisville, Kentucky Deadline for Receipts May 7, 1981 Abstract Processing Pee: $20.00 Select Category Numbers and Titles (See list over) 1. 79. 2. 15 Presentation Preference _ Poster Oral (slides) __ Indifferent If your choice is not available _ W illingly accept the alternative Grudgingly accept the alternative Withdraw the abstract Poster boards are 6* wide x 4' high. 23 Toxic, Carcinogenic & Safe Exposure Levels in Vinyl Chloride Induced Hepatic Angiosarcoma. Carlo H. Tamburro* and Richard A. Greenberg* (SPON: William Waddell). Depts. of Medicine & Community Health, Univ. of Louisville, Lou., Ky. 40292. Vinyl chloride (VC) induced hepatic angiosarcoma (HA) has rapidly decreased since 1974. The relationship of latency & duration of exposure was studied in 21 North American (NA) cases and 62 worldwide (WW) reported cases; 92% of NA & WW cases had long exposure (LE) (10-33 yrs) ; 7 cases (8%) had short exposures (SE) (3.5-7 yrs). The mean duration in both LE & SE groups was 18.5 yrs. with 2 peaks at 15 & 25 yrs. Deaths from HA began in 1955, peaked in 1976 & rapidly fell thereafter. HA peaked in Canada (CA) in 1973 and USA 1974, France (FR) 1976, & Germany (GE) 1978. Operations began in CA in 1941 & USA 1939-44; FR 1941-57, & GE 1952-60. Exposure levels reported in 1940's of ^2000 ppm; in 1950's ^200-500 ppm, in early 1960's ^50-500 ppm, late '60's & early '70's ^50-250 ppm, & 1-10 ppm after '74. No relationship was found in total average exposures (TAE) range of 50 ppm; 200-1000 ppm range showed a direct relation of lower TAE to shorter latency, while higher TAE (1000-2000 ppm) had longer laten cies. The data suggests VC has a biological threshold level, an optimum carcinogenic level and subcarcinogenic toxic level for occurrences of HA. Prepare your abstract carefully, it will be printed by photo-offset exactly as received. See the example (ever) for lay-out and style. For elite typewriters (12 pitch), set the margins at 36 and 98 (62 spaces). Use 25 lines. For pica typewriters (10 pitch) set margins at 30 and 81 (51 spaces). Use 25 lines. The first typed letter should nearly or just touch the top and left type box (blue margin lines). Title. Leave no margins. Use a short, descriptive title. Use upper and lower case letters. Authorship. Underline names. Place an asterisk* after the name of each author not a member of ASPET or SOT. If no author is a member, type the following after the name of the last author: (SPON: sponsor's name). Continue on the same line with the authors' institutional affiliation(s), city, state and zip code. Text. Start tht text on the next line, indenting 3 spaces. Subsequent lines should extend the entire width of the type box. The text MwuM have all of the elements of a report: introduction, method, result(s) and conclusion. It is improper to substitute "The results will be discussed" for the results and conclusion. Adequately identify all chemical compounds used. Do not fold abstract. Be certain abstract is prepared securely for mailing. All subsequent correspondence must reference the first author to enable us to identify the abstract. The Program Committee selects chairmen and overviewera from volunteers. See below. First author phone #: (502) 588-5251 Membership of author or ponsor^ASP.T~Xj SOT _X_ 1. Member signature 2. Name (Type or print) Will 3. Will you chair a session? In what category Will you give a 25 minute overview? __ yes; _)no; In what category Mall the original, 3 copies, a aelf-ad*emad post card for program confirmation and 1 chock to Kay A. Croker, Acting Executive Officer, ASPET, 9650 Rockville Pike, Bethesda, MD 20814, USA. Reprints cost $15 per 100, lots of 100 only. Order below and fill in the mailing label, below left. Invoice: Abetract Processing Pee Reprints: Quantity ordered__coat: TOTAL $ 20.00 _______ $_______ Mailing label for reprints (Name end Address): Mailing label to the first author (Name end Address): CMA 003918 PREPRINTS CMA 003919 The reproducibility in identifying these histological lesions is quite high. Under double blind conditions there was consistent histological identification of chemical injury in both duplicate readings of single biopsies as well as single readings of multiple biopsies in the same individual. Inconsistency appears to occur more frequently among the duplicate readings of wedge biopsies and between wedge and needle biopsies than between duplicate readings of needle biopsies or readings of multiple needle or wedge biopsies in the same individual. This may be due to the increased time and attention needed for wedge section review in order to determine the absence or presence of these lesions anywhere in the specimen. There are, however, some pitfalls that require caution. The quality of the histological preparation is of vital importance. Technical artifact, excessive thickness, torn or fragmented spec imens will greatly reduce the ability to correctly identify these lesions. The use of silver impregnation for assessment of the reticular structure is very useful in verifying focal areas of hepatocellular hyperplasia. The presence of fatty infiltration and/or chronic disease (hepatitis, granuloma, etc.) may prevent the identification of these lesions. We were, however, able to correctly identify the presence or absence of these early lesions in the six individuals who also had a concomitant granulomatous process (four individuals with granuloma were repeatedly read as negative and two individuals with gran uloma were repeatedly read as positive for chemical liver injury). The more frequent elevation of alkaline phosphatase in those individuals with histological evidence of chemical injury may provide some guidance in sequential uses of laboratory screening tests. During the first phase, tests with the greater sensitivity for identifying liver injury, such as the ICG and ALT can be used. Positive results can be followed up by using tests of greater specificity, such as the alkaline phosphatase, to further identify those at high risk of chemical injury. CMA 003920 10 of acute or chronic injury, appear to be the earliest change associated with chemical exposure. Subsequently, there occurs perisinusoidal fibrosis associated with a proliferation of fibroblasts and/or its precursor cell, the Ito cell. In later stages sinusoidal cell activation occurs involving both the macrophagic and endothelial lining cells. These changes are associated with focal sinusoidal enlargement. In the late stage mixed hyperplasia (hepatocyte and sinusoidal cells) becomes prominent, followed by sinusoidal cell dysplasia with palasading of cells, and finally malignant transformation. Our data demonstrates that focal hepatocytic hyperplasia is the earliest consistently identifiable histological finding in industrial workers exposed to vinyl monomer chemicals. The very limited occurrence of these findings in non-chemical workers and the fact that focal mixed hyperplasia, a more advanced lesion, was only found in chemically exposed workers and in a non chemical worker with vinyl chloride exposure (via hair spray), adds further confirmation that these lesions are related to prolonged or repeated periods of chemical exposure. Whether these lesions should be viewed as precursors to the development of cancer or only as an early reflection of the degree and duration of chemical exposure cannot be determined at this point in time. The unchanging and stable status of screening studies over a seven year follow-up period and the histo logical similarities among the serial biopsies of these workers suggest that these early lesions are non-progressive when the affected individuals are removed from exposure (7). The presence of sinusoidal cell dysplasia, however, has been associated with the later development of angiosarcoma as illustrated in one of our cases and may represent a true precancerous or a non-reversible lesion. CMA 003921 this worker cohort had shown that all the cases of angiosarcomas had an average vinyl cnloride exposure ranking (AER) of 3.5 or greater. AER's of 3.5 or greater occurred in 45 of the CLI group. In contrast, those with liver disease (LD) alone, only 22% had a vinyl chloride exposure rating of 3.5 or greater and in the chemically exposed workers with no evidence of liver disease, 32" had a rating of 3.5 or greater. Biochemical Studies A study of the biochemical abnormalities among the two histological groups of chemical workers with evidence of liver disease is illustrated in Figure 7. The five screening tests with the best degree of sensitivity and specificity in identifying latent liver injury were reviewed. Indocyanine green clearances, at the 0.5 mg/kg dose, provides the most sensitive and the most specific of the laboratory screening study, followed by the alanine amiontransferase and aspartase aminotransferase, gamma glutamyl transpeptidase and alkaline phosphatase (6). However, specificity for chemical injury appeared to be associated more with abnormalities in the alkaline phosphatase, which was far more frequently abnormal in workers with chemical liver disease than in those with non-chemical liver injury. In contrast, the ALT, AST, GGT, and ICG were more frequently abnormal in those workers with liver disease of a non-chemical origin. DISCUSSION Figure 8 provides the most likely progression in the development of vinyl chloride and other vinyl monomers hepatic injury and cancer development. These findings confirm those previously reported by Thomas and Popper (2) in their original pathological studies of vinyl chloride associated hepatic angiosarcoma and hepatic injury. Focal hepatocytic megalocytosis associated with focal increases in reticulum structure, in the absence of other evidence CMA 003922 this worker cohort had shown that all the cases of angiosarcomas had an average vinyl chloride exposure ranking (AER) of 3.5 or greater. AER's of 3.5 or greater occurred in 45% of the CLI group. In contrast, those with liver disease (LD) alone, only 22% had a vinyl chloride exposure rating of 3.5 or greater and in the chemically exposed workers with no evidence of liver disease, 32% had a rating of 3.5 or greater. Biochemical Studies A study of the biochemical abnormalities among the two histological groups of chemical workers with evidence of liver disease is illustrated in Figure 7. The five screening tests with the best degree of sensitivity and specificity in identifying latent liver injury were reviewed. Indocyanine green clearances, at the 0.5 mg/kg dose, provides the most sensitive and the most specific of the laboratory screening study, followed by the alanine amiontransferase and aspartase aminotransferase, gamma glutamyl transpeptidase and alkaline phosphatase (6). However, specificity for chemical injury appeared to be associated more with abnormalities in the alkaline phosphatase, which was far more frequently abnormal in workers with chemical liver disease than in those with non-chemical liver injury. In contrast, the ALT, AST, GGT, and ICG were more frequently abnormal in those workers with liver disease of a non-chemical origin. DISCUSSION Figure 8 provides the most likely progression in the development of vinyl chloride and other vinyl monomers hepatic injury and cancer development. These findings confirm those previously reported by Thomas and Popper (2) in their original pathological studies of vinyl chloride associated hepatic angiosarcoma and hepatic injury. Focal hepatocytic megalocytosis associated with focal increases in reticulum structure, in the absence of other evidence CMA 003923 A comparison of the reproducibility of the biopsy readings is shown in Tables 3, 4 and 5. Table 3 illustrates the high degree of consistent readings between multiple biopsies in the same individual. In only one individual was one of the four biopsies read differently--the wedge was read differently from the three needle biopsies. Table 4 makes a comparison between readings of multiple needle biopsies or multiple wedge biopsies from a single individual. In these cases there was consistent agreement in all readings. Table 5 illustrates the consistent readings of the same biopsies over the three-year period. All the needle biopsy duplicate readings and 12 of the 14 wedge duplicate readings were the same. At least 96% (43/45) of all readings, whether duplicate or from multiple biopsies, provided reproducible and consistent findings. Correlation of Histological Lesions With Exposure Biopsies from all the chemical workers were divided into three categories on the basis of final diagnosis: 1) individuals with no evidence of histological liver disease (N) 2) individuals who had liver disease but no evidence of chemical injury (LD) 3) individuals who had chemical liver injury identified by the histolog ical criteria specified (CLI). A comparison was made between the average vinyl chloride exposure ranking of each worker and his/her final histological diagnosis. The exposure rankings were not known to those making the final histological determination at any time during the three year period. As illustrated in Figure 6 the group of workers with evidences of chemical liver injury had the highest percentage of individuals with the highest ranking of exposure to vinyl chloride. Previous studies of 0039^4 for each job for each year based on the best available information and chemicals ,vas based on cne various job classifications he/sne held during their employment. A detailed explanation of this system and an actual in-the-field demonstration of the ability 0* this system to identify work-related liver injury (angiosarcoma) has been published elsewhere (4,5). RESULTS Thirty-seven percent (13/35) of the exposed workers with screening test abnormalities had nepacic lesions consistent with chemical exposure. Among the exposed workers without biochemical abnormalities 23% (3/13) had hepatic.lesions consistent with chemical exposure. In the non-worker comparison group none of those with normal biochemical screening tests and 17/19 (89%) with abnormal biochemical screening tests had hepatic lesions consistent with chemical expo sure. Of the remaining two, one had focal hepatocellular hyperplasia; the other had focal mixed hyperplasia and early peliosis hepatis. This latter individual was later found to have angiosarcoma possibly from vinyl chloride hair spray exposure. The former individual had no history of chemical exposure and/or alcohol consumption, but did have hepatobiliary tract disease in the form of biliary stones. Table 1 lists the hepatic lesions found in those with and without biochemical abnormalities for the entire group. Reproducibility An evaluation of the reproducibility of these biopsy readings was carried out in 17 individuals who had 32 biopsies. Seven individuals had only one biopsy (2 needle, 5 wedged) and 10 individuals had the-ir biopsies read in dupli cate. Four had needle biopsies with 9 readings and 6 had wedge biopsies with 14 readings (Table 2). CMA 003925 for each job for each year based on the best available information and industrial expertise. Each worker's cumnulative exposure to each of the 22 chemicals was based on the various job classifications he/she held during their employment. A detailed explanation of this system and an actual in-the-field demonstration of the ability of this system to identify work-related liver injury (angiosarcoma) has been published elsewhere (4,5), RESULTS Thirty-seven percent (13/35) of the exposed workers with screening test abnormalities had hepatic lesions consistent with chemical exposure. Among the exposed workers without biochemical abnormalities 23% (3/13) had hepatic lesions consistent with chemical exposure. In the non-worker comparison group none of those with normal biochemical screening tests and 17/19 (89%) with abnormal biochemical screening tests had hepatic lesions consistent with chemical expo sure. Of the remaining two, one had focal hepatocellular hyperplasia; the other had focal mixed hyperplasia and early peliosis hepatis. This latter individual was later found to have angiosarcoma possibly from vinyl chloride hair spray exposure. The former individual had no history of chemical exposure and/or alcohol consumption, but did have hepatobiliary tract disease in the form of biliary stones. Table 1 lists the hepatic lesions found in those with and without biochemical abnormalities for the entire group. Reproducibil ity An evaluation of the reproducibility of these biopsy readings was carried out in 17 individuals who had 32 biopsies. Seven individuals had only one biopsy (2 needle, 5 wedged) and 10 individuals had their biopsies read in dupli cate. Four had needle biopsies with 9 readings and 6 had wedge biopsies with 14 readings (Table 2). CMA 003926 hepatocellular hyperplasia; b) focal increased reticulum associated with the hepatocytic hyperplasia; c) perisinusoidal fibrosis; d) focal hyperplasia; e) focal mixed hyperplasia associated with increased reticulum deposition; f) sinusoidal cell activation; g) sinusoidal dilitation; h) portal and capsular fibrosis; i) sinusoidal dysplasia; and, j) hepatic angiosarcoma. The presence of focal hepatic cellular hyperplasia with focal increases in reticulum, in the absence of other evidences of hepatocellular disease or steatosis, were considered the minimum presumptive evidence of chemical exposure. The presence of focal mixed hyperplasia with increased reticulum and/or peri sinusoidal fibrosis was considered evidence of chemical injury of a more advanced stage. Sinusoidal cell activation and dilitation of the sinusoids were con sidered evidences of chemical injury of an even further stage of advancement. The presence of subcapsular fibrosis, (on wedge biopsies only) and/or increases in both portal and sinusoidal fibrosis, with sinusoidal cell activation and sinusoidal dilitation was considered the most advanced stage of chemical injury characteristic of vinyl chloride or vinyl monomer exposure. Finally, sinu soidal cell dysplasia and/or evidence of malignant transformation, constituted definitive evidence of chronic vinyl chloride exposure and the terminal stages of injury. Work and exposure histories were available for all the chemical workers. The work histories provide a total and average exposure data for 22 chemicals based on rank order exposure estimates. These exposure work histories consisted of an estimate of exposure to each of 22 different chemicals for all of the 350 job classifications used since the start of the chemical plant. Each of the chemicals exposure was ranked on a scale of zero to six for each of the 22 chemicals. These exposure ranking estimates were done separately CMA 003927 nucleoli with a moderate amount of cytoplasm which reacted on PAS staining and persisted after diastase reaction. These cells are interpreted as activated fibroblasts, are associated with an increase in fat storing sinusoidal cells and augmented peri sinusoidal fibrosis tissue (Figure ). 3. Fibrotic changes consisting of an excell of irregular connective tissue in the periportal zone, frequently arranged around proliferating bile ductules, and focal thickened subcapsular accumulation of corrective tissue into the parencyma (Figure 7). These histological lesions are believed to be the three earliest hepatic findings associated with vinyl monomer chemical injury. The hepatocytic changes are referred to as focal hepatocytic hyperplasia or FHH. The sinu soidal cell changes (sinusoidal cell hyperplasia) associated with the hepatocytic changes are referred to as focal mixed hyperplasia or FMH. For purposes of this study, early fibrotic changes were not classified since the majority of liver biopsies were of needle type, which could not be evaluated for subcapsular changes. The liver biopsies were coded and read blindly to identify (1) histologi cal evidence of hepatic disease; (2) any characteristics of non-chemical injury excluding steatosis; (3) steatosis with or without fibrosis and, (4) evidence of chemical injury identified by the following histological findings: a) focal CMA 003928 A group of 30 individuals who were not chemical workers but had undergone abdominal surgery with liver biopsies, were used as a comparison population. All of the comparison population's liver biopsies were performed during the same three-year period, at the same hospital, by mentiers of the same medical- surgical team, with the informed consent of the individuals. This group had similar hepatic biochemical studies with the exception of the GGT and ICG clearance. None had a history of extensive exposure of working with halo- genated hydrocarbons, although one patient had a history of household contact with vinyl chloride via commercial home products. Pathology Earlier experiences with autopsy and animal material from vinyl chloride associated angiosarcoma have recognized certain hepatic lesions in the non- tumorous area of the liver parenchyma ( ). They include: 1. Hepatocytic changes consisting of foci of enlarged hepatocytes with increased amount of cytoplasm, and large hyperchromatic nuclei (Figure 1). These cells are intermixed with hepatocytes of normal and smaller sizes and create focal, indistinct nodular areas (Figure 2a). These areas are often associated with very slight increased sinusoidal dilitation and focal increase in the reticulin framework illustrated best by silver inpregnation (Figure 2b). 2. Sinusoidal lining cell changes consisting of an increase in cell number associated with nuclear changes, particularly conspicuous in areas of sinusoidal widening. This proliferation of sinusoidal cells involves (a) Kupffer cells (macrophages) with PAS-positive granules; (b) fibroblasts with elongated, almost rectangular nuclei with loose vesicular chromatin patterns and inconspicuous 003929 CtAft. I 'iTP.OO'JCT I ON Industrial vinyl chlor'de and ot'ier v'nyl tcmerer exposure is associated with various hepatic histological abnormalities. These include subcaosular, portal, and perisinusoidal fibrosis as well as hyperplasia of both hepatocytes and sinusoidal cells. Early histological studies, mainly on autopsy material, had shown focal mixed hyperplasia (hyperplasia of hepatocytes and sinusoidal cells) to be an early histological alteration associated with vinyl chloride exposure (1,2,3). To substantiate this observation and determine its potential use in medical surveillance screening of the exposed workers, liver bioDsies from 73 individuals were studies in duplicate blind fashion to determine 1) if these early histological findings were associated with extensive vinyl chloride exposure, and 2) if these histological findings occur in non-exposed popula tions or were associated with other diseases. MATERIALS AND METHOOS Liver biopsies from 48 vinyl monomer chemical workers with and without hepatic biochemical abnormalities were investigated. All 48 chemical workers had had hepatic biochemical studies performed on an annual or semi-annual basis. These included aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (AP), total bilirubin (TB), gamma glutamyl transpeptidase (GGT), prothronfcin time (PT), and indocyanine green clearance (ICG). In addi tion, history and physical examinations, liver-spleen scans, and abdominal plain fllnB had been performed annually. Thirteen of the chemical workers had hepatic biopsies for non-liver related reasons while 35 had biopsies performed because of screening biochemical and/or liver-spleen radioisotopic scan abnor malities. CMA 0393o INTRODUCTION Industrial vinyl chloride and other vinyl monomer exposure is associated with various hepatic histological abnormalities. These include subcapsular, portal, and perisinusoidal fibrosis as well as hyperplasia of both hepatocytes and sinusoidal cells. Early histological studies, mainly on autopsy material, had shown focal mixed hyperplasia (hyperplasia of hepatocytes and sinusoidal cells) to be an early histological alteration associated with vinyl chloride exposure (1,2,3). To substantiate this observation and determine its potential use in medical surveillance screening of the exposed workers, liver biopsies from 78 individuals were studies in duplicate blind fashion to determine 1) if these early histological findings were associated with extensive vinyl chloride exposure, and 2) if these histological findings occur in non-exposed popula tions or were associated with other diseases. MATERIALS AND METHODS Liver biopsies from 48 vinyl monomer chemical workers with and without hepatic biochemical abnormalities were investigated. All 48 chemical workers had had hepatic biochemical studies performed on an annual or semi-annual basis. These included aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (AP), total bilirubin (TB), gamma glutamyl transpeptidase (GGT), prothrontin time (PT), and indocyanine green clearance (ICG). In addi tion, history and physical examinations, liver-spleen scans, and abdominal plain films had been performed annually. Thirteen of the chemical workers had hepatic biopsies for non-liver related reasons while 35 had biopsies performed because of screening biochemical and/or liver-spleen radioisotopic scan abnor malities. CMA 003931 ABSTRACT Earlier histological studies of industrial vinyl chloride exposure, obtained mainly on autopsy material, indicated that focal mixed (hepatocytes and sinu soidal cells) hyperplasia is the earliest histological alteration indicative of exposure. To substantiate this observation and its potential use in screening workers, 93 liver biopsies from 78 persons were investigated in double blind duplicative fashion; 35 were exposed chemical workers with hepatic screening tests abnormalities, and 13 were exposed workers without hepatic abnormalities who had liver biopsies for non-liver related reasons. A comparison group consisted of 30 individuals who were not chemical workers, but had liver biop sies for non-hepatic related reasons during the same time period. Of the exposed workers, 23 (48%) had a hepatic lesion consistent with exposure; 17 (35%) had only focal hepatocytic hyperplasia; six (13%) had focal mixed hyper plasia or more advanced lesions. In contrast, only five of the comparison group had similar findings; four (13%) had only focal hepatocytic hyperplasia and one (3%) had focal mixed hyperplasia and sinusoidal dilitation. On a sibsequent biopsy, this individual was found to have angiosarcoma and a history of using hair spray containing vinyl chloride as propellant. Ten individuals had 28 multiple biopsies also read doifcle blindly, and 10 individuals had 23 readings of the same biopsy; 21/23 (91%) duplicate readings and 27/28 (96%) multiple biopsy readings in the same individuals were identical. Only 18% of either duplicate and/or multiple biopsy readings had disagreements in their biopsy assessment. Focal hepatocytic hyperplasia, in addition to the previously described mixed hyperplasia, appears to be the earliest identifiable change consistent with chemical exposure and both lesions are present prior to the development of angiosarcoma. These lesions can be consistently identified and are useful in the screening of chemical workers for evidence of exposure. CMA 003932 EARLY HEPATIC HISTOLOGICAL ALTERATIONS AMONG CHEMICAL (VINYL MONOMER) WORKERS 1 23 Carlo H. Tamburro, Laszlo Makk, and Hans Popper l Liver Research Center, Department of Medicine 21 and St. Anthony Hospital, University of Louisville 3 Louisville, Ky., Stratton Laboratory for Study of Liver Diseases, Mt. Sinai Hospital fo Medicine, New York, New York 1 Portions of this work supported by NCI Contract No-l-CN-55212 and the Manufacturing Chemists Association Grant VC7.0 CMA 003933 The reproducibility in identifying these histological lesions is quue high. Under double bl'^i conditions t^ere -as consistent hi;toloc;ccl identification of chemical injury in both duplicate readings of single biopsies as well as single readings of multiple biopsies in the same individual. Inconsistency appears to occur more frequently among the duplicate readings of wedge biopsies and between wedge and needle biopsies than between duplicate readings of needle biopsies or readings of multiple needle or wedge biopsies in the same individual. This may be due to the increased time and attention needed for wedge section review in order to determine the absence or presence of these lesions anywhere in the specimen. There are, however, some pitfalls that require caution. The quality of the histological preparation is of vital importance. Technical artifact, excessive thickness, torn or fragmented spec imens will greatly reduce the ability to correctly identify these lesions. The use of silver impregnation for assessment of the reticular structure is very useful in verifying focal areas of hepatocellular hyperplasia. The presence of fatty infiltration and/or chronic disease (hepatitis, granuloma, etc.) may prevent the identification of these lesions. We were, however, able to correctly identify the presence or absence of these early lesions in the six individuals who also had a concomitant granulomatous process (four individuals with granuloma were repeatedly read as negative and two individuals with gran uloma were repeatedly read as positive for chemical liver injury). The more frequent elevation of alkaline phosphatase in those individuals with histological evidence of chemical injury may provide some guidance in sequential uses of laboratory screening tests. During the first phase, tests with the greater sensitivity for identifying liver injury, such as the ICG and ALT can be used. Positive results can be followed up by using tests of greater specificity, such as the alkaline phosphatase, to further identify those at high risk of chemical injury. CMA 003934 12 CONCLUSIONS Focal hepatocellular hyperplasia, in addition to the previously described focal mixed hyperplasia, appear to be the earliest identifiable changes consistent with chemical exposure. These histological lesions are useful in identifying high risk exposed chemical workers as part of a medical screening surveillance program. Biochemical studies may help identify which individuals warrant a liver biopsy for further identification of the cause of the liver abnormalities. CMA 003935 REFERENCES 1. Popper H, and Thomas LB. Alterations of liver and spleen among workers exposed to vinyl chloride. Annals of the New York Academy of Sciences 1975; 245:172-194. 2. Thomas LB and Popper H. Pathology of angiosarcoma of the liver among vinyl chloride-polyvinyl chloride workers. Annals of the New York Academy of Sciences 1975; 246:268-277. 3. Gedigk P, Muller R and Bechtelsheimer H, Morphology of liver damage among polyvinyl chloride production workers. A report on 51 cases. Annals of the New York Academy of Sciences 1975; 246:278-285. 4. Greenberg RA and Tamburro CH. Exposure indices for epidemiological sur veillance of carcinogenic agents in an industrial chemical environment. Journal of Occupational Medicine 1981; 23:No. 5, 353-358. 5. Greenberg RA and Tamburro CH. Monitoring exposure to hazardous chemicals in an industrial setting: a method of demonstrated utility. Journal of Occupational Medicine ,198 1 6. Tamburro CH and Greenberg RA. Effectiveness of federally-required medical laboratory screening in the detection of chemical liver injury. Environmental Pfg^pectives, 1981 , 41:117-122 7. Tamburro CH, Davidson CS, Fisher MM, et al. Medical surveillance for chemical hepatotoxicity. Guidelines for Detection of Hepatotoxicity Due to Drugs and Chemicals. Davidson CS, Leevy CM and Chamberlain EC (ed.) U.S. Department of HEW, NIH Pub. No. 79-313, 1979, Chapter 5; 60-80. CMA 003936 :rrro>re*:ices 1 Pepper H and Thoras lS. 4|"erations Oi vsr dno scl99n i.Tcm wor\a,'3 exposed to vinyl chloride. Annals of the i'lew York Academy of Sciences 1 975; 245:172-194. 2. Thomas LB and Popper H. Pathology of angiosarcoma of the liver among vinyl chloride-polyvinyl chloride workers. Annals of the New York Academy of Sciences 1975; 246:268-277. 3. Gedigk P, Muller R and Bechtelsheimer H. Morphology of liver damage among polyvinyl chloride production workers. A report on 51 cases. Annals of the New York Academy of Sciences 1975; 246:278-285. 4. Greenberg RA and Tamburro CH. Exposure indices for epfcemiological sur veillance of carcinogenic agents in an industrial chemical environment. Journal of Occupational Medicine 1981; 23:No. 5, 353-358. 5. Greenberg RA and Tamburro CH. Monitoring exposure to hazardous chemicals in an industrial setting: a method of demonstrated utility. Journal of Occupational Medicine ,198 6. Tamburro CH and Greenberg RA. Effectiveness of federally-required medical laboratory screening in the detection of chemical liver injury. Environmental Prg^pectives, 198J , 41:117-122 7. Tamburro CH, Davidson CS, Fisher W, et al. Medical surveillance for chemical hepatotoxicity. Guidelines for Detection of Hepatotoxicity cue to Drugs and Chemicals. Davidson CS, Leevy CM and Chamberlain EC (ed.) U.S. Department of HEW, NIH Pub. No. 79-313, 1979, Chapter 5; 60-80. CMA 003937 OXIDATIVE AND GLUTATHIONE-RELATED DETOXIFYING ENZYME CAPABILITIES IN HEPATOCYTES AND NONHEPATOCYTES OF RAT LIVER Julie T. Du, D.S. Eades, and Carlo H. Tamburro Liver Research Center, Department of Medicine and Division of Occupational Health, Department of Community Health and Cancer Center University of Louisville, School of Medicine Louisville, Kentucky 40292 CMa 00393q 2 ABSTRACT Liver is the main organ in the body with the capacity to metabolize xenobiotics. Mixed function oxidase (MFO) and glutathione transferase(s) are the major enzymes involved. The activities of MFO (demethylation of benzphetamine), glutathione transferase(s) (glutathione ^-epoxide transferase, GEST, 1,2-epoxy-[-nitrophenoxy] propane as substrate; and glutathione S-aralkyl transferase, GAST, -nitrobenzyl chloride as substrate), and glutathione reductase (GR) were determined in the subcellular fractions of isolated nonhepatocytes and hepatocytes. The nonhepatocytes were isolated by pronase digestion and the hepatocytes by collagenase perfusion. The viability of the cells were assessed by trypan blue exclusion and shown to be over 90% viable in the hepatocyte and nonhepatocyte fractions. Cross-contamination of isolated cells was estimated by the relative distribution and specific activities of the L- and M2-type pyruvate kinase isoenzymes and was found to be insignificant. The specific activity based on protein content showed that the nonhepatocytes had about 7% GAST activity, 50% GEST and MFO activities, and 73% GR activity compared to hepatocytes. Based on activities per million cells, the nonhepatocytes had approximately 1/450 GAST activity, 1/100 MFO activity, 1/60 GEST, and 1/40 GR activity as compared to hepatocytes. Based on distribution of enzyme activity in hepatic cells per gram liver, the nonhepatocytes had less than or equal to one hundredth the activities of MFO, GAST, GEST, and GR as compared to hepatocytes. This study demonstrates that the nonhepatocytes have the capacity to metabolize and detoxify xenobiotics but that they are significantly less capable than the hepatocytes in oxidation and glutathione-related detoxification. CMA 003939 3 INTRODUCTION Liver is the main organ in the body in regard to the metabolism of xenobiotics including drugs and carcinogens. Hepatocytes are thought to be the cells that are mainly responsible for this metabolism, but little is known about the role of nonhepatocytes. These cells, mainly endothelial and Kupffer cells, because of their anatomical location may interact with xenobiotics before they even enter hepatocytes. Thus, it is important to compare nonhepatocytes to hepatocytes as to their enzymatic capability to oxidize and detoxify xenobiotics. In this study, we concentrated on the enzymes for oxidation (mixed function oxidase, MFO) and glutathione-related detoxifying enzymes (glutathione S-epoxide transferase, GEST; glutathione S-aralkyl transferase, GAST; and glutathione reductase, GR) in the microsomal and soluble fractions of isolated nonhepatocytes vs. hepatocytes as well as in whole 1iver tissue. MATERIALS Oxidized and Reduced Glutathione (GSSG and GSH respectively), Metrizamide, Bovine Serum Albumin (BSA Cohn fraction V), NADPH, ADP, NADH, Phosphoenol pyruvate (PEP), DNAse, Fructose^ ,6-diphosphate and Lactate dehydrogenase (pyruvate kinase free) were obtained from Sigma Chemical Co., St. Louis, MO. Gey's Balanced Salt solution was purchased from GIBCO, Grand Island, NY. 1,2-Epoxy-3-(p-nitrophenoxy) propane from Eastman Kodak Co., Rochester, NY., p-nitrobenzyl chloride from Matheson, Coleman and Bell, Norwood, OH.. Benzaphetamine Hydrochloride was a gift from The Upjohn Co., Kalamazoo, Mil Collagenase Type II was purchased from Worthington Biochemical Crop., Freehold, NJ. Pronase E was purchased from E. Merck, Darmstadt, Germany. All other chemicals were reagent grade and were obtained from commercial sources. Double distilled water was used throughout the experiment. Tests for Cell Purity and Integrity: The purity and integrity of cells isolated were tested by light microscopy (phase contrast), by trypan blue exclusion, and by the relative distribution and specific activities of the Land Mg-type pyruvate kinase isoenzymes. The M2-type pyruvate kinase (in the nonhepatocytes) is not activated by the addition of fructose-1,6-diphosphate but the L-type pyruvate kinase (solely in hepatocytes) is activated 11-fold (5). Cell Homogenization and Fractionation: A Polytron homogenizer was used at setting 5 for 60 sec to obtain ceTT homogenates. The homogenate was then centrifuged at 8000 g for 10 min in a refrigerated Sorvall centrifuge. The supernatant was centrifuged at 100,000 g for 60 min at 4C in a Beckman ultracentrifuge with a swinging bucket rotor. The rotors were equipped with microadaptors to enable the centrifugation of samples of 1 ml or less. Enzyme Assays: The rate of GSH conjugate formation was determined for glutathione S-transferases. P-Nitro benzyl chloride was the substrate for GAST with E=l900 M"Tcm"T at llO nm; 1,2-epoxy-(p-nitrophenoxy) propane was the substrate for GEST with E=510 M^cnr1 at J60 nm. Glutathione reduc CMA 003940 i I'iTn \Juui, I t ON uiver is the main organ in the body in regard to the metabolism of xenobiotics including drugs and carcinogens. Hepatocytes are thought to be the cells that are mainly responsible for this metabolism, but little is known about the role of ncnhepatocytes. These cells, mainly endothelial and Kupffer cells, because of their anatomical location may interact with xenobiotics before they even enter nepatocytes. Thus, it is important to compare nonhepatocytes to hepatocytes as to their enzymatic capability to oxidize and detoxify xenobiotics. In this study, we concentrated on the enzymes for oxidation (mixed function oxidase, MFO) and glutathione-related detoxifying enzymes (glutathione ^-epoxide transferase, GEST; glutathione ^-aralkyl transferase, GAST; and glutathione reductase, GR) in the microsomal and soluble fractions of isolated nonhepatocytes vs. hepatocytes as well as in whole 1iver tissue. MATERIALS Oxidized and Reduced Glutathione (GSSG and GSH respectively), Metrizamide, Bovine Serum Albumin (BSA Cohn Fraction V), NAOPH, ADP, NADH, Phosphoenol pyruvate (PEP), DNAse, FructoseJ,6-diphosphate and Lactate dehydrogenase (pyruvate kinase free) were obtained from Sigma Chemical Co., St. Louis, MO. Gey's Balanced Salt solution was purchased from GIBCO, Grand Island, NY. 1,2-Epoxy-3-(p-nitrophenoxy) propane from Eastman Kodak Co., Rochester, NY., p-nitrobenzyl chloride from Matheson, Coleman and Bell, Norwood, OH. Benzaphetamine Hydrochloride was a gift from The Upjohn Co., Kalamazoo, MI. Collagenase Type II was purchased from Worthington Biochemical Crop., Freehold, NJ. Pronase E was purchased from E. Merck, Oarmstadt, Germany. All other chemicals were reagent grade and were obtained from commercial sources. Oouble distilled water was used throughout the experiment. Tests fof Cell Purity and Integrity: The purity and integrity of cells isolated were tested by light microscopy (phase contrast), by trypan blue exclusion, and by the relative distribution and specific activities of the Land M2-type pyruvate kinase isoenzymes. The M2-type pyruvate kinase (in the nonhepatocytes) is not activated by the addition of fructose-1,6-diphosphate but the L-type pyruvate kinase (solely in hepatocytes) is activated 11-fold (5). Cell Homogenization and Fractionation: A Polytron homogenizer was used at setting 5 for 60 sec to obtain cell homogenates. Thehomogenate was then centrifuged at 8000 g for 10 min in a refrigerated Sorvall centrifuge. The supernatant was centrifuged at 100,000 g for 60 min at 4C in a Beckman ultracentrifuge with a swinging bucket rotor. The rotors were equipped with microadaptors to enable the centrifugation of samples of 1 ml or less. Enzyme Assays: The rate of GSH conjugate formation was determined for glutathione S-transferases. P-Nitro benzyl chloride was the substrate for GAST with E*l900 M"^cm'' at "310 nm; 1,2-epoxy-(p-nitrophenoxy) propane was the substrate for GEST with E510 M^cnr' at T60 nm. Glutathione reduc CMA 003941 4 tase was determined by the reduction of oxidized glutathione, GSSG, as measured by the disappearance of NADPH (6). Mixed function oxidase was estimated by measuring the disappearance of NADPH in the NADPH dependent demethylation of benzphetamine (7). All asays were done using the appropriate soluble or microsomal fractions of freshly isolated cells. The activities were linear functions of protein concentration and of time. Protein Determination: Before homogenization, the cells were washed with 6eyTs balanced salt solution without BSA to completely remove the BSA contained in the perfusion medium. The protein content in the soluble, microsomal and homogenate fractions were then determined by Lowry's method (8). METHODS Animals: Sprague-Dawley male rats (300-400 g) were used. Cell Isolations: Hepatocytes were isolated by a collagenase perfusion method (1). The liver was perfused in situ in a perfusion chamber, first with Krebs Hensleit bicarbonate buffer [KHBB) without calcium ion (pH 7.4) followed by KHBB with the addition of 0.02% (w/v) collagenase, 4 mM calcium and 2% bovine serum albumin for 40 min. The flow rate was 14 ml/min. at 37C, and constant aeration was applied with 95% 02-5% CO2. After perfusion the liver was blanched completely, excised, rinsed in KHBB, blotted dry, and weighed. Glisson's capsule was removed, the cells were shaken loose by gentle agitation in KHBB with 2% BSA, and collected by centrifugation at 50 for 5 min. The nonhepatocytes were isolated essentially by the method of Mills and Zucker-Franklin (2) modified by Knook and Sleyster (3). The liver was first perfused with Gey's balanced salt solution (3 min), followed by the addition of pronase (0.2% w/v, 3 min) and then digested with pronase (0.2%, 60 min). In order to minimize DNA induced cell aggregation, 0.5 mg DNAse was added to the pronase digestion step. The digested tissue was filtered, and the cells collected by centrifugation at 300 g (3 min), washed and differentially centrifuged with concentrated Metrizamiae (1400 g, 15 min) to remove red blood cells.(4) The cells were counted on a hemocytometer. RESULTS AND DISCUSSION Cell Viability and Purity: The cells used for the determination of enzyme activities in this study were shown to be greater than 90% viable, and there was essentially no hepatocyte contamination in the nonhepatocyte preparation as judged by the pyruvate kinase activation method (Table 1). In pure hepatocytes, pyruvate kinase is activated eleven-fold by the addition of fructose-1,6-diphosphate (1 n#i) In nonhepatocytes, the activity is only slightly activated showing essentially no hepatocyte contamination (5). Enzyme Activities: The GEST, CAST, GR and MFO activities from the soluble and microsomal fractions of isolated cells and from liver were determined and the specific activities are shown in Table 2. The specific activities in the enzyme- CMA 003942 5 isolated hepatocytes are higher than those obtained from the fractions of whole liver homogenate (Table 2) suggesting that the collagenase treatment does not appreciably affect the enzymes studied. The nonhepatocytes had about 7% GAST activity, 50% GEST and MFO activities and 73% GR activity as compared to hepatocytes; these results are based on protein content.. There have been no reports about the glutathione-related enzyme activities in nonhepatocytes. Cantrel and Bresnick (9) have studied benzpyrene hydroxylase in various hepatic cells by determining the fluorescence of the reaction product, 8-hydroxy benzpyrene, and found that the specific activity of benzpyrene hydroxylase in the nonparenchymal liver cells was only 7.5% of the activity of that in parenchymal cells. This is much lower than the ratio of MFO activities in nonparenchymal to parenchymal cells obtained here (50%). We measured the disappearance of NADPH in the NAOPH-dependent demethylation of benzphetamine (7). The difference in these studies may be due to the use of different assay methods. Nevertheless both studies demonstrate that the nonparenchymal cells have less MFO activity than hepatocytes. The subcellular fractions from nonhepatocytes were determined to contain 0.027 mg soluble protein per million cells, and 0.0063 mg microsomal protein per million cells; the subcellular fractions from hepatocytes contained 0.8 mg soluble protein per million cells and 0.29 mg microsomal protein per 10b cells. Thus, the enzyme activity expressed per million cells as shown in Table 3 was calculated from the specific activity previously shown in Table 2. The nonhepatocytes had approximately 1/450 GAST activity, 1/100 MFO based on cell number activity, 1/60 GEST activity, and 1/40 GR activity as compar^ to hepatocytes. ^ The total protein contents in nonhepatocytes and hepatocytes were determined to be 0.07 and 2 mq/per million cells respectively. The diameter of a nonhepatocyte is 8-1 lu (3) and that of a hepatocyte is 18-30u ,(10). Assuming the cells are spheres, the volume ratio of a hepatocyte to a nonhepatocyte is the ratio of the diameters cubed, or close to 27. Thus, the ratio of the protein content in the hepatocytes to nonhepatocytes is within the range of the volume ratio of the two types of cells. The distribution of enzyme activity liver (Table 4) can be calculated from the activity expressed per million cells (Table 3) and the cell number per gram tissue. In the rat, there are approximately 100x10 hepatocytes and 54X10 nonhepatocytes per gram liver (3,11). Thus, the nonhepatocytes had less than or equal to one hundredth the activities of MFO, GAST, GEST and GR as compared to hepatocytes per gram liver. From this study, it is concluded that the nonhepatocytes have significantly less capability in the oxidation and glutathione-related detoxification of xenobiotics as compared to hepatocytes. CMA 003943 0 isolated hepatocytes are higher than those obtained from the fractions of wnoie liver homogenate (Table 2) suggesting that the collagenase treatment does not appreciably affect the enzymes studied. The nonhepatocytes had about 1% GAST acti 'ty, 50% GEST and MFO activities and 73% GR activity as compared to hepatocytes; these results are based on protein content.. There have been no reports about the glutathione-related enzyme activities in nonhepatocytes. Cantrel and Bresnick (9) have studied benzpyrene hydroxylase in various hepatic cells by determining the fluorescence of the reaction product, 8-hydroxy benzpyrene, and found that the specific activity of benzpyrene hydroxylase in the nonparenchymal liver cells was only 7.5% of the activity of that in parenchymal cells. This is much lower than the ratio of MFO activities in nonparenchymal to parenchymal cells obtained here (50%). We measured the disappearance of NADPH in the NADPH-dependent demethylation of benzphetamine (7). The difference in these studies may be due to the use of different assay methods. Nevertheless both studies demonstrate that the nonparenchymal cells have less MFO activity than hepatocytes. The subcellular fractions from nonhepatocytes were determined to contain 0.027 mg soluble protein per million cells, and 0.0063 mg microsomal protein per million cells; the subcellular fractions from hepatocytes contained 0.8 mq soluble protein per million cells and 0.29 mg microsomal protein per 10& cells. Thus, the enzyme activity expressed per million cells as shown in Table 3 was calculated from the specific activity previously shown in Table 2. The nonhepatocytes had approximately 1/450 GAST activity, 1/100 MFO based on cell number activity, 1/60 GEST activity, and 1/40 GR activity as compared to hepatocytes. The total protein contents in nonhepatocytes and hepatocytes were determined to be 0.07 and 2 mg/per million cells respectively. The diameter of a nonhepatocyte is 8-llu (3) and that of a hepatocyte is 18-30u (10). Assuming the cells are spheres, the volume ratio of a hepatocyte to a nonhepatocyte is the ratio of the diameters cubed, or close to 27. Thus, the ratio of the protein content in the hepatocytes to nonhepatocytes is within the range of the volume ratio of the two types of cells.. The distribution of enzyme activity liver (Table 4) can be calculated from the activity expressed per million cells (Table 3) and the cell number per gram tissue. In the rat, there are approximately 100x10 hepatocytes and 54X10 nonhepatocytes per gram liver (3,11). Thus, the nonhepatocytes had less than or equal to one hundredth the activities of MFO, GAST, GEST and GR as cohered to hepatocytes per gram liver. From this study, it is concluded that the nonhepatocytes have significantly less capability in the oxidation and glutathione-related detoxification of xenoblotlcs as compared to hepatocytes. CMA 003944 6 ACKNOWLEDGMENTS This work was supported by a grant from the Chemical Manufacturers Association. REFERENCES 1. M.N. Berry and D.S. Friend, Or. Cell Biol. 43, (1969). 2. D.M. Mills and D. Zucker-Franklin. Am. Jr. Pathol. 54, 147 (1969), 3. D.L. Knook and E.C. Sleyster, Exp. Cell. Res. 99, 445 (1976). 4. R.N. Zahlten, H.K. Hagler, M.E. Nejtek, and C.J. Day, Gastroenterol. 75, 80 (1978). ------ -------------- 5. T.J.C. Van Berkel, J.K. Knight, R.G. Slee and J.F. Koster, Achives. Biochem. Biophys. 179, 1 (1977). 6. I. Carlberg and B. Mannervick, J. Biol. Chem. 250, 5475 (1975). 7. A.Y.H. Lu, R. Kuntzman, S. West, M. Jacobson and A.H. Conney, J. Biol. Chem. 247, 1729 (1972). " 8. O.H. Lowry, N.J. Rosenbrough, A.L. Farr and R.J. Randall, J. Biol. Chem. 193, 265 (1951). 9. E. Cantrell and E. Bresnick, J. Cell Biol. 52, 316 (1972) 10. Evaluation of Liver Function (Eds. S.O. Waife,; E.L. Platcow and C.E. Hamnond) 2nd Ed. P. 11, Lilly Res. Lab, Indianapolis, IN (1974). 11. D.K. Knook, E.C. Sleyster and M.J. Van Noord, hi Cell Impairment in Aging and Development (Eds. V.J. Christofalo and E. Holekova). p. 155 Plenum Publishing Corp., New York (1975). CMA 003945 7 Table 1. Determination of hepatocyte contamination in nonparenchymal cells of liver (pyruvate kinase activation method) Tissue or Cell Type Nonparenchymal Cell Hepatocyte Whole Rat Liver Activation of pyruvate kinase by addition of F-l 6-P2 Th is Lab Van Berkel and Koster 1.41 0.18 (3) 1.008 (18X 9.4 11.8 (2) 11.77 (19) 6.79 0.70 (4) 6.52 (22) Table 2. Enzyme activities of various types of hepatic cells and tissues Enzyme GEST GAST GR MFO Fraction SOL. SOL. SOL. MC Specific activity (nmol/min/mg protein) Nonhepatocyte Hepatocyte While Liver Tissue 42.9 18.5 47.4 7.49 17 (7) 6 (6) 2 (4) 2.3 (5) 82 11 (8) 275 61 (8) 65 7 (10) 16 5 (6) 74.1 (23) 213 (18) 50.3 (34) 12.6 (18) CMA 003946 7 Table 1. Determination of hepatocyte contamination in nonparenchymal cells of liver (pyruvate kinase activation method) Tissue or Cel 1 Type Nonparenchymal Cell Hepatocyte Whole Rat Liver Activation of pyruvate kinase by addition of F-l 6-Pj Th i s Tab--------------- Van Serkel and Koster 1.41 0.18 (3) 1.008 (18') 9.4 11.8 (2) 11.77 (19) 6.79 0.70 (4) 6.52 (22) Table 2. Enzyme activities of various types of hepatic cells and tissues Enzyme GEST GAST GR MFO Fraction SOL. SOL. SOL. MC Specific activity (nmol/min/mg protein) Nonhepatocyte Hepatocyte While Liver Tissue 42.9 18.5 47.4 7.49 17 (7) 6 (6) 2 (4) 2.3 (5) 82 n (8) 275 61 (8) 65 7 OO) 16 5 (6) 74.1 (23) 213 (18) 50.3 (34) 12.6 (18) CMA 003947 e Table 3. Enzyme activities of hepatocytes and nonhepatocytes Enzyme GEST GAST GR MFO Fraction SOL. SOL. SOL. MC Activity (nmol/mi n/10 cells) Nonhepatocytes Hepatocytes tfepatocytes/nonhepatocytes 1.16 65.6 56.6 0.5 250 440 1.28 52 40.6 0.048 4.4 96.7 Table 4. Distribution of enzymatic activity in various hepatic cells Enzyme GEST GAST GR MFO Fraction SOL. SOL. SOL. MC Activity (nmol/min/g liver) Nonhepatocytes Hepatocytes 63 (1X) 6,560 (99*)a 27 (0.1%) 22,000 (99.9*) 69 (1.3*) 5,200 (98.7*) 2.6 (0.6*) 464 (99.4) a: % activity in each cel 1 type per amount of liver CMA 003948 PREPRINT PRELIMINARY ASSESSMENT OF THE USEFULNESS OF URINARY TOTAL GLYCOSAMINOGLYCAN EXCRETION IN THE DETECTION OF SUBCLINICAL LIVER DISEASE 1 Charles E. Kupchella Richard A. Greenberg R.V. Warick Carlo H. Tamburro Fran the Cancer Center and the Departments of Medicine and Canmunity Health School of Medicine, University of Louisville Louisville, KY 40292 CMA 003 949 1 Present Address: Department of Biological Sciences Murray State University, Murray, Kentucky 42071 page 2 Summary Urinary glvcosaminoglycan excretion measured as uronic acid was studied in 60 asymptomatic vinyl monomer chemical workers. Thirty-six had normally and 24 had abnormally functioning livers as judged by a standard battery of biochemical liver tests. Urine samples were assayed for both creatinine and total cetylpyridinium chloride-precipitable uronic acid. Individual values were declared to be normal or abnormal based on the range 2.0 - 4.8 ^g uronic acid/ mg creatinine established as the normal range at the outset. The frequency distribution of glycosaminoglycans for the groups of normal and abnormal biochemistries were statistically significantly different. The distributions did not differ in their means (t58=0.968), but did differ in their variability (F23 35=2-381; P<0.05). The observation that liver disease increases the range of value for total urinary glycosaminoglycan excretion, but not the mean, suggests that this parameter is not a useful general screening test in asymptomatic individuals with hepatic dysfunction. The increased variance in urinary glycosaminoglycan excretion in individuals with liver disease may reflect or be related to a stage of the disease process. At certain stages of disease, glycosaminoglycan incorporation (into scar tissue for example) may predominate while in other stages there may be a predominance of glycosaminoglycan-releasing, tissue destruction and/or an impairment of the liver's ability to.clear glycosaminoglycans from the blood. These possibilities warrant further study with greater differentiation of disease categories and stages and with resolution of specific glycosaminoglycans patterns. cMA 003950 INTRODUCTION page 3 Changes in tissue, urinary and blood glycosaminoglycans (GAGs) have been found to occur in many connective tissue disorders including those of the liver (Galambos and Shapira, 1973; Koizumi, et a_l. 1967; Kojima, 1964; Rubin, 1966; and Patrick and Keenedy, 1964) and in hepatic cancer (Kojima, et_ a_l. , 1975; Anghileri, 1974; Kupchella, e_t ad 1981;- and Yamamoto and Teryama, 1973). In a previous report, we described elevated urinary excretion of uronic acid in a number of liver disease states (Kupchella and Tamburro, 1976) and showed that angiosarcoma patients exhibit a urinary GAG excretion pattern- distingui from that of normal urine by relative increases in a hyaluronidase resistant fraction and a decrease in a hyaluronidase susceptible fracti (Curran et aJL. , 1977) . The purpose of this study was to assess the potential usefulness of total urinary GAG determinations in the identification of liver disease among chemical (iruylmonomer) worked other than hepatic angiosarcoma. MATERIALS AND METHODS One hundred non-biopsied workers (50 with and 50 without hepatic biochemical abnormalities) were selected at random from among more than 1200 vinyl chloride production workers at the Louisville plant of the B.F. Goodrich Company. Participation was voluntary and 36 workers without biochemical abnormalities and 24 workers with biochemic abnormalities elected to participate. Criteria used to determine the biochemical status of these workers' livers are summarized in Table 1. CMA 003951 INTRODUCTION -acre i Changes in tissue, urinary and blood glycosaminoglycans (GAGs) have been found to occur in many connective tissue disorders including those of the liver (Galambos and Shapira, 1973; Koizumi, et al. 1967; Kojima, 1964; Rubin, 1966; and Patrick and Keenedy, 1964) and in hepatic cancer (Kojima, et al. , 1975; Anghileri, 1974; Kupchella, el; al_, 1981; and Yamamoto and Teryama, 1973). In a previous report, we described elevated urinary excretion of uronic acid in a number of liver disease states (Kupchella and Tamburro, 1976) and showed that angiosarcoma patients exhibit a urinary GAG excretion pattern distinguii from that of normal urine by relative increases in a hyaluronidase resistant fraction and a decrease in a hyaluronidase susceptible fractic (Curran et al., 1977). The purpose of this study was to assess the potential usefulness of total urinary GAG determinations in the identification of liver disease among chemical (iruylmonomer) workers other than hepatic angiosarcoma. MATERIALS AND METHODS One hundred non-biopsied workers (50 with and 50 without hepatic biochemical abnormalities) were selected at random from among more than 1200 vinyl chloride production workers at the Louisville plant of the B.F. Goodrich Company. Participation was voluntary and 36 workers without biochemical abnormalities and 24 workers with biochemic abnormalities elected to participate. Criteria used to determine the biochemical status of these workers' livers are summarized in Table 1. CMA 003952 pas^e 5 RESULTS Sixty GAG determinations were made, 6 of 36 (16.7%) workers witb normal hepatic biochemistries had GAG excretions of less than 2 or equal to or more wiian 4.3 ug uronic acid/ma creatinine. This was true for ten of the groups of 24 (41.7?,) workers whose bio chemistries were abnormal. in Table 2. The frequency distributions are given DISCUSSION We attempted here to assess the diagnostic, and other aspects of the minimally expensive, minimally time-consuming method of determining urinary GAGs. Because we found that the effect of liver disease on urinary GAG was to increase variance without appreciably altering the mean, we have clearly not identified a new diagnostic test- These results suggest however that liver disease may impact on urinary GAG excretion in multiple ways. Subclinical hepatis dys function may favor uptake or incorporation of GAGs (fibrotic tissue for example) while other diseases or disease stages are dominated by GAG-releasing tissue destruction and/or the impairment of the liver's ability to clear the blood of circulatory GAGs. 3954 page 4 Urine specimens were collected during routine screening and frozen until analysis. Twenty-five ml. amounts were subjected to urinary GAG analysis in triplicate as described by DiFerrante (1967) and 5 ml. samples assayed for creatinine using a Technicon Autoanalyzer. Uronic acid determinations of cetylpyridiniumchloride complexed GAGs were made according to the modified carbozole method of Bitter and Muir (1962) and expressed as micrograms of uronic acid per mg of creatinine. Based on our work with normal controls (Kupchella and Tamburro, 1976) and reports by others (Varma, et al., 1974; DiFerrante and Rich, 1956) we defined the normal range of urinary GAG excretion as 2.0 to 4.8 jug of uronic acid per mg of creatinine, assuming that the normal adult male excretes 1.5 grams of creatinine every 24 hours (Sunderman and Boerner, 1940). Our definition of normal range was established in advance of the study from previous data on a "normal" cohort (640 adult males examined annually for 3-5 yrs). The data includes physical examination, liver spleen scan, SMAC 20, chest x-ray, 0.5 mg/kg dose ICG clearance, negative HBgAg serology, cbc and urine analysis. The GAG analysis was done blindly, i.e., clinical diagnosis of the patients from whom the urine was collected was not known. CMA 003953 3 RESULTS Sixty SAG determinations were made, 6 of 36 (16.7%) workers with normal hepatic biochemistries had GAG excretions of less than 2 or equal to or more cnan 4-8 ug uronic acid/ma creatinine. This was true for ten of the groups of 24 (41,7s.) workers whose bio chemistries were abnormal. in Table 2. The frequency distributions are riven DISCUSSION We attempted here to assess the diagnostic, and other aspects of the minimally expensive, minimally time-consuming method of determining urinary GAGs. Because we found that the effect of liver disease on urinary GAG was to increase variance without appreciably altering the mean, we have clearly not identified a new diagnostic test. These results suggest however that liver disease may impact on urinary GAG excretion in multiple ways. Subclinical hepatis dys function may favor uptake or incorporation of GAGs (fibrotic tissue for example) while other diseases or disease stages are dominated by GAG-releasing tissue destruction and/or the impairment of the liver's ability to clear the blood of circulatory GAGs. CMA 003955 page 7 TABLE 2 Frequency (f) and Relative Frequency (R.F.) of glycosaminoglycan excretion levels for workers with normal and abnormal biochemical results. A T-test of independent means (t58=0*968) revealed no significant difference, however an F-test of the independence of variance (two-tailed) revealed significance (^23,35*2.381) at the P<0.Q5 level. GAGa <2 2< 3 3< 4 h <5 Normal . f R.F. U 0.111 16 0.1* 1*1* 13 0.361 1 0.028 Abnormal f R.F. . 6 0.250 7 0.292 6 0.250 1* 0.167 Sum Mean Variance 36 2.886 0.7^6 1.000 2U 3.160 1.776 1.000 a ^ig uronic acid/mg creatinine 39S7 page 6 TABLE 1 Liver function tests used to determine "biochemical abnormality". Biochemical abnormality was established if two or more tests in group 1 were abnormal or if one from group 1 and two from group 2 were abnormal (negative otherwise). GROUP 1 Alanine Amino Transferase (ALT) GROUP 2 Gammaglutamyl Transpeptidase (GTP) Indocyanine Green Clearance (ICA) Bilirubin (total) (TB) Alkaline Phosphatase (AP) Aspartate amino transferase (AST) CM* 3956 page 7 TAB13 2 Frequency (f) and Relative Frequency (R.F.) of glycosarainoglycan excretion levels for workers with normal and abnormal biochemical results, A T-test of independent means (^3=0.968) revealed no significant difference, however an F-test of the independence of variance (two-tailed) revealed significance (^23,35"2.381) at the P<0.05 level. GAGa <2 2< 3 3< 4 1* < 5 Normal . f R.F. h 0.111 16 0.444 13 0.361 1 0.028 Sum Mean Variance 36 2.866 0.7U6 1.000 Abnormal f R.F. 6 0.250 7 __ 0.292 6 0.250 k 0.167 2k 3.160 1.776 1.000 a jag uronic acid/mg creatinine CMA 003958 page 8 REFERENCES 1. Anghileri, L. J. (197*0 Metabolism of Acid Mucopolysaccharides in Hepatoma and in Normal Liver. Oncology ,22.: 30**-317. 2. Bitter, T., and Muir, H. (1962) A modified Uronic Acid Carbatole Reaction. Anal. Biochem. *: 330-33**. 3. Curran, K. L., Kupchella, C. E. and Tamburro, C. H. (1977) Urinary Glycosaminoglycan patterns in Angiosarcoma of the Liver, Cancer, U0(6): 3050-3053. 4. DiFerrante, N., and Rich, C. (1956) The Determination of Acid Aminopolysaccharide in Urine. J. Lab. Clin. Med. 1*8: **91-1*9U. 5. DiFerrante, N. M. (1967) The Measurement of Urinary Mucopoly saccharides. Anal. Biochem. 21: 98-106. 6. Galambos, J. T. and Shapira, R. (1973) Natural History of Hepatitis: IV Olycosaolnoglycuronans and Collagen in the Hepatic Connective Tissue. J. Clin. Invest. 52(11): 2952-2962. 7. Koizumi, T., Nakamura, N., and Abe, H. (1967) Changes in Acid Muco polysaccharide in the Liver in Hepatic Fibrosis. Bioeham. Biophya. Acta. 1**8: 7*9-756. fUKoJima, J, (196**) Studies on the Metabolism of Hepatic Connective Tissue in Fibrosis of the Liver, Med. J, Osaka Univ, 16: **19-*29. l.KoJima, J., Nakamura, N., Kanatanl, M, and Obmori, K. (1975) The Glycosamlnoglycans in Human Hepatic Cancer. Cancer Res. 35(3): 5*2-5**7. 10 . Kupchella, C. E. and Tamburro, C. H. (1976) Urinary and Tissue Glycosaminoglycan Patterns in Hepatic Angiosarcoma. In: Detection and Prevention of Cancer. H. E. Nieburgs, Ed., Part 1, Volume 1, Marcel-Dekker, Inc., New York. CMA 003959 P*e 9 11. Kupchella, C. E., Drake, E. E., Kennedy, J., Curran, K. L. and Warick, R. (i960) Tissue and Urinary Glycosaminoglycan Patterns Associated vith a Past, an Intermediate and a Slcnr-grovlng Morris Hepatoma, 1981. Cancer Res. 4l:419-424. 12/ Patrick, R. 3. and Kennedy, J. S. (1964) The Synthesis of Sulfated Mucopolysaccharide at Sites of Hepatic Fibrosis Induced by Carbon Tetrachloride, Amyloidosis, and Implantation of Catgut. J. Pathol. Bacterid. 88: 549-555. 13, Rubin, E. (1966) Autoradiographic Characterization of Sulfated Acid Mucopolysaccharide in Experimental Cirrhosis. J. Histochem. Cytochem. l4> 688-689* l4t Sunderman, P. W. and Boerner, P. (1949) Normal Values in Clinical Medicine. W. B. Saunders, Philadelphia, p. 353* ^15 Varma, R. S., Varna, R., Allen, W. S. and Ward!, A. H. (1974) Urinary Excretion of Acid Mucopolysaccharides in Schizophrenia. Biochem. Med. 11(4): 358-369. 16. Yamomoto, K., and Teryama, H. (1973) Comparison of Cell Coat Acid Mucopolysaccharides of Normal Liver and Various Ascites Hepatoma Cells. Cancer Res. 33: 2257-2264. 0396q t Xupchella, G. 3., Urake, 3. 3., Kennedy, J., Curran, X. L. and Varic. (i960) Tissue and Urinary Glycosaminoglycan Patterns Associated vith a Fast, an Intermediate and a Slov-groving Morris Hepatoma, 1981. Cancer Res. 4l:419-424. ^2, Patriclc, R. S. and Kennedy, J. S. (1964) The Synthesis of Sulfated Mucopolysaccharide at Sites of Hepatic Fibrosis Induced by Carbon Tetrachloride, Amyloidosis, and Implantation of Catgut. J. Pathol. Bacteriol. 88: 549-555. 23, Rubin, E. (1966) Autoradiographic Characterization of Sulfated Acid Mucopolysaccharide in Experimental Cirrhosis. J. Histochem. Cytochea. l4.i 688-689. 14 Sunderman, F. W. and Boerner, F. (1949) Normal Values in Clinical Medicine. W. B. Saunders, Philadelphia, p. 353. 15# Varma, R. S., Varma, R., Allen, W, S. and Wardi, A. H. (1974) Urinary Excretion of Acid Mucopolysaccharides in Schizophrenia. Biochem. Med. 11(4): 358-369. 16. Yamomoto, K., and Teryama, H. (1973) Comparison of Cell Coat Acid Mucopolysaccharides of Normal Liver and Various Ascites Hepatoma * Cells. Cancer Res. 33' 2257-2264. CMA 003961