Document 0qqLOZarDqo05vbjDrVeeQwqV

Correlated Flow Cytometric Analysis of H-ras p21 and Nuclear in Multiple Myeloma DNA By Hirouki Tsuchiya, Joshua Epstein, Peter Selvanayagam, John R. Dedman, Raymond Alexanian, and Bart Barlogie Gary Gallick, Correlated analysis of the H-ras oncogene product (p21) and of nuclear DNA content was performed by flow cytom- etry (FCM) in patients with DNA-aneuploid multiple myeloma (MM). Bone marrow cells from normal donors and MM patients in remission served as controls. Seventy-four percent of 23 patients with active MM had higher p21 fluorescence in aneuploid tumor cells than were observed in normal donor or myeloma remission bone marrows; 39% of the 23 patients also showed high H-ras p21 expression R ECENT ADVANCES in molecular biology have revealed the involvement of cellular oncogenes in neo- plasia.'3 In hematologic tumors they are frequently acti- vated by chromosomal translocations with breakpoints at or near their sites.' In contrast to extensive cytogenetic and molecular studies in lymphomas and leukemias, such investi- gations are limited in multiple myeloma (MM). This slowly proliferating B-cell tumor poses major difficulties to obtain- ing metaphase spreads for chromosomal analysis. In the absence, so far, of MM-specific cytogenetic anomalies,7 molecular studies could not be guided by morphological aberrations. Review of the available cytogenetic literature and the authors' own studies suggested a frequent involve- ment of chromosomes I and 1 1 , which are the sites of the N-ras and H-ras I oncogenes, respectively. This observation and the role of the ras family oncogenes in inducing the malignant phenotype in DNA transfection experiments'3 led to the current systematic evaluation of ras oncogene expres- sion in MM cells. The commonest mechanism of ras-gene activation in human neoplasia is gene mutation, although DNA amplification has also been reported.8 The ras-gene product is a 21-Kd protein (p21) that is located at the inner plasma membrane.9 Its role in normal growth and differen- tiation is still being investigated. From the University of Texas System Cancer Center. M.D. Anderson Hospital and Tumor !nstitute. Departments of Hema- tology (Hiroyuki Tsuchiya, Joshua Epstein, Peter Selvanayagam. Raymond Alexanian. Bart Barlogie) and Tumor Biology (Gary Gallick), and The University of Texas Health Science Center. Department of Physiology and Cell Biology (John R. Dedman), Houston. Submitted December 16, 1987; accepted April 26. 1988. Supported in part by Grants CA37161, CA28771, CA16672 from the National Cancer Institute and The Cullen Trust Foundation (Bart Barlogie) and GM29323 from National !nstitutes of Health (John R. Dedman). Address reprint requests to Bart Barlogie. MD. M.D. Anderson Hospital and Tumor !nstitute, 1515 Holcombe Blvd. Box 30, Houston. TX 77030. The publication costs ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. section /734 solely to indicate this fact. I 988 by Grune & Stratton, !nc. 0006-4971/88/7202-0068$3.00/0 in diploid cells. There was an inverse relationship between p21 levels and the presence of trisomy 1 1 ; especially high p21 levels were noted in patients without trisomy 1 1 . The frequent elevation of p21 protein in aneuploid plasma cells suggests the involvement of the H-ras oncogene in the pathophysiology of MM. which is further supported by a shorter survival among patients with high p21 levels. e 1988 by Grune & Stratton, Inc. Given the heterogeneity of human neoplasia, much effort has been expanded to in situ examination of oncogene expression, using either autoradiographic or more recently flow cytometry (FCM) studies.'#{176}" Thus Andreeff et al'#{176} have recently demonstrated the feasibility of quantitating by FCM ras-protein content in single cells. Since MM exhibits an abnormal nuclear DNA content in at least 80% of cases,'2 the authors elected to examine ras p2 1 expression in relation- ship to DNA aneuploidy using FCM of doubly stained bone marrow cells from 23 patients with MM. While preliminary studies with a pan-ras antibody (Y 1 3-259) had revealed high levels of ras p21 in MM cells, recent studies in malignant lymphoma showing high messenger RNA (mRNA) levels of H-ras prompted the authors to focus on H-ras p21 in MM as well.'3 Compared to normal bone marrow cells, high levels of H-ras p21 were present in aneuploid plasma cells and seemed to confer a poor prognosis. MATERIALS AND METHODS Preparation of anti-H-ras p21 antibody. The anti-H-ras p21 antibody was prepared using synthetic peptide corresponding to a unique sequence of H-ras protein (amino acid no. I 60- 1 79),14 which was purchased from Peninsula Laboratories (San Carlos, CA). The method followed the previous report,'5 except for using a sheep rather than a rabbit as the recipient animal. Briefly, the synthetic peptide was coupled to keyhole limpet hemocyanin and injected subcutaneously with Freund's adjuvant into a sheep. The selected amino acid sequence is common to both c- and v-H-ras p21.'4 The reactivity of the generated antiserum with the synthetic peptide was determined by enzyme-linked immunosorbent assay (ELISA), with a titer of greater than I :10,000. Antiserum was purified by CN- Br-activated Sepharose column (Pharmacia, Uppsala, Sweden) and was termed HAS2. Using Kirsten or Harvey murine sarcoma virus-transformed normal rat kidney cell lines (Ki- or Ha-NRK), the specificity of HAS2 was confirmed by immunoblotting with the biotin-avidin rabbit antisheep peroxidase system (Vector Laborato- ries, Burlingame, CA) and DNA/p21 FCM described below. The reactivity and localization of HAS 2 to H-ras p21 was also deter- mined by fluorescence microscopy using NRK, Ki-NRK, and Ha- NRK cell lines as well as a human myeloma cell line (ARH-77'6). Cells were stained according to a three-step indirect immunofluo- rescence assay described below. The antibody HAS2 is especially suitable for DNA-correlated FCM, since ethanol can be used as a fixative that provides higher resolution of DNA measurements than methanol. In contrast, methanol fixation is required when using the Yl3-259 monoclonal antibody (MoAb), which reacts with all p21 species (ie, N-, H-, and K-ras p21). 796 Blood. Vol 72. No 2 (August), 1988: pp 796-800 CORRELATED FLOW CYTOMETRY ANALYSIS 797 Study samples. Following informed consent, heparinized bone marrow cells were obtained from 23 patients with DNA-aneuploid MM (untreated, seven; primary resistant, five; relapse, I 1 ; Table 1) as well as from eight MM patients in remission and eight healthy donors. Aneuploid MM samples were selected on the basis of abnormalities of DNA/RNA (acridine orange) or DNA-clg pat- terns (propidium iodide/fluoresceini5othiocyanate).hl$ In ques- tionable cases normal blood lymphocytes were directly mixed with the patient sample. Remission marrow was defined by the presence ofless than 0.5% DNA-aneuploid more than 75% cytoreduction.'9 cells in patients who had achieved Donors for allogeneic bone marrow transplantation (BMT) were used as normal controls. The human T-cell--acute lymphocytic leukemia cell line (Molt-4) expresses high H-ras p21 protein level and was used as a positive control in each experiment. DNA/p21 flow cytometry. Marrow aspirates were subjected to Ficoll-Hypaque (FH) gradient separation (SG 1.077), and inter- phase cells were collected and washed twice in phosphate-buffered saline (PBS). A small amount of cells was used for confirmation of DNA aneuploidy. The remaining sample was fixed in 70% ethanol at 4#{176f}oCr at least 24 hours. Cells were then processed for 2-parameter FCM analysis by a three-step indirect immunofluorescence method with subsequent propidium iodide reaction for DNA counterstain- ing. Fifty microliters of HAS2 antibody at a concentration of 500 zg/mL were added to 106 cells in 50 iL of incubation buffer; incubation buffer consisted of 4% inactivated normal human serum with 0.5% Tween 20 (Sigma, St. Louis) in PBS. As nonspecific control, 50 zL of purified normal sheep !gG at a concentration of 500 g/mL was used (Organon Teknika-Cappel, several experiments HAS2 was preincubated Malvern, PA). In with the synthetic peptide coupled to sepharose and the supernatant used as an additional control (blocked HAS2). Cells were then incubated successively with 100 tL of biotinylated rabbit antisheep IgG (Vector Laboratories) at a dilution of 1:50 and with 100 L of fluorescein isothiocyanate (FITC)-labeled streptavidin (BRL, Gai- thersburg, MD) at a dilution of 1:100. Each incubation step was carried out at 4#{176}fCor 30 minutes. Following each step, cells were washed twice with 0.5% Tween 20 in PBS. Finally, cells were incubated with 70 tg of ribonuclease A (Cooper-biomedical, Mal- vern, PA) in 1 mL of PBS at 37#{176}fCor ten minutes and DNA Table 1 . Characteristics of Patients With DNA Aneuploid Myeloma High v Low H-ras p21 Levels Disease Statust Ig Types Serum B2M Chromosome H-ras p21 #{149ia} rel res K L G A D High Low Yes 1 1 Trisomy No 9 354 84731 <9 46 1 83911 55 1 10 1 5 5 0 Abbreviations: t. untreated; rel. relapsed; res, primary resistance; K. kappa; L, lambda. 4 H-ras p2 1 fluorescence was calculated by the methods described in . Materials and Methods. Value 9 corresponds to the median H-ras p21 level in aneuploid cells from 23 patients with aneuploid myeloma (Fig 4). tDisease status at time of study. ImmunogIobulin typing by immunoelectrophoresis and cytoplasmic Ig analysis using FCM. One patient did not produce the heavy chain. Serum beta-2-microglobulin levels were measured by radioimmu- noassay.2e Data were available for 2 1 patients. High denotes 6mg/L andlow <6mg/L. P< .1. Pa'ients with chromosome 1 1 trisomy. Data were available for 1 1 patients. #{182}Onpeatient with 1 1 tetrasomy and another patient with chromo- some 1 1 trisomy plus del ( 1 1 ) (q22) were included. counterstained with 40 jL of propidium iodide (Sigma, 500 g/mL in PBS) for five minutes at room temperature. Five thousand cells were analyzed with a FACS !I flow cytometer (Becton Dickinson, Mountain View, CA). Data were further analyzed with a Terak 8600 computer (Terak Co, Scotsdale, AZ) using the GRAF3D and CUBE3D programs (Coulter Electronics, Hialeah, FL). Fluorescence intensities were analyzed independently for aneu- ploid and diploid G0/G, cell compartments and were expressed as mean channel numbers. The mean channel number for H-ras p21 was obtained from gated G,10 areas shown in Figs I A through C, thus eliminating any influence of cycle stage-related differences in ras expression. The H-ras p2 1-specific fluorescence was derived by subtracting nonspecific fluorescence (obtained from normal sheep !gG staining) from total fluorescence (obtained from HAS2 stain- ing). To avoid interexperimental variation, H-ras p21-specific fluo- rescence was always standardized using the fluorescence intensity of Molt-4 cells. Statistical analysis. The relationship between H-ras p2 1 fluo- rescence intensity and other markers was examined by Student's and chi-square tests, and Kaplan-Meier plots were used for survival analysis. RESULTS Specificity of HAS2 antibody. !mmunoblotting experi- ments with HAS2 at a concentration of 1 : 100 revealed strong reactivity in Ha-NRK cells. In contrast, Ki-NRK cells, which showed high levels of p21 when developed with the pan-ras MoAb Yl 3-253 (results not shown) did not react with HAS-2. Ha-NRK revealed two discrete bands corre- sponding to H-ras p2! and its phosphorylated form with a MW 92K 67K - 45K - 30K 24K - 20K -b 18K -- Fig 1 . Specificity of anti-H-ras p21 antibody HAS2 was firmed by immunoblotting using NRK, Ki-NRK. and Ha-NRK Immunoblotting was performed with the biotin-avidin rabbit sheep peroxidase system. concells. anti- 798 slower migration, as described previously20 (Fig 1 ). Specific- ity of HAS2 was also confirmed by DNA/p21 FCM; the H-ras p2 1 specific fluorescence level of Ha-NRK was 2.7 times as high as that of NRK. Microscopic examination showed strong cytoplasmic staining of Ha-NRK cells (origi- nally infected with v-H-ras) and strong membrane staining of ARH-77 myeloma cells, but weak reactivity with Ki-NRK and NRK cells (data not shown; Fig 2). In additional control TSUCHIYA ET AL experiments, the fluorescence level of cells reacted with blocked HAS-2 was indistinguishable from background fluo- rescence. DNA/p21 FCM analysis. Representative examples of DNA/p21 FCM distributions for an aneuploid MM and morphologically normal bone marrow specimens from a healthy donor and a MM patient in remission are depicted in Fig 3. Markedly higher H-ras p21 fluorescence is observed in hyperdiploid plasma cells of the myeloma patient compared to seemingly normal diploid hematopoietic cells (Fig 3 A) and by comparison with morphologically uninvolved marrow (Figs 3 B and C). Among the 23 patients bone with aneuploid MM, high H-ras p2 1 fluorescence levels greater than 5 (upper limit of H-ras p2! level in normal bone marrow) were present in 74% of aneuploid and in 39% of diploid subpopulations (Fig 4). With one exception, the latter had lower p21 levels than the corresponding aneuploid tumor cells. Lowest p2 1 levels were noted in the eight patients with MM in remission (all 2), whereas four of the eight normal donors had values between 3 and 5. Clinical relevance of p21 expression. No relationship was noted between H-ras p21 levels and immunoglobulin isotypes (Table 1) or plasma cell phenotype analyzed by FCM (surface beta 2 microglobulin, B2, iS, Rl-3 and B4).2' Serum beta-2-microglobulin levels weakly correlated with H-ras p21 levels (Table 1; P < .1). Of I I patients with available cytogenetic data, six had trisomy I 1 with or without additional numeric or structural aberrations (Table 1). Such patients had significantly lower p21 levels than the remaining five without chromosome 1 1 involvement ( 1 2. 1 v 5.5; P < .01). Although serial studies are p21 expression changes during lacking to determine the disease course, whether the clinical relevance of high p2 1 expression at any time was examined. Patients were divided into two groups according to the level of p21 expression; patients with lower than the median level of p21 (n = 1 1) and those with p21 levels median (me- dian 9, Fig 4). There was no difference in the median times from diagnosis to the date of p21 analysis between these two groups. None of the 1 1 patients with low p2! expression has died compared to only six survivors =with high levels, with a projected months (Fig 5; P .03). among the I 2 patients median survival of 46 Fig 2. Microphotographs of ARH-77 human myeloma cells. Phase contrast (A) and immunofluorescence (B) of the same area stained for H-ras p21 with HAS2 antibody and background control stained with normal sheep lgG (C). Cells were stained with HAS2 or normal sheep lgG as a primary antibody. then incubated with biotinylated rabbit antisheep lgG and fluorescein isothiocyanate- labeled streptavidin. DISCUSSION FCM has already emerged as a valuable tool in studying the cellular heterogeneity in MM. Thus drug resistance was closely associated with the presence of DNA hypodiploidy and low plasma cell RNA content.22 A novel common acute lymphoblastic leukemia antigen (CALLA)/c!g pheno- type has been defined in aneuploid myeloma, generally conferring a favorable prognosis.2' Similarly, specific thera- py-related measurements of anthracycline drug uptake23 and glucocorticoid receptor expression have been conducted.24 The authors have now demonstrated the feasibility of using FCM for quantitation of H-ras p21 expression using the HAS2 antibody. Counterstaining with propidium iodide for nuclear DNA content permitted correlated analysis sepa- CORRELATED FLOW CYTOMETRY ANALYSIS 799 w 0z w0 `U 0 1 I'. 0d STEM LINE RESIDUAL DIPLOD CELLS MVELOMA N REMISSION (8) NORMAL BONE MARROW (8) Fig 4. H-ras p21 expression in aneuploid and diploid cells of patients with active myeloma, myeloma in remission, and normal bone marrow. Solid lines connect data derived from the same patients. Numbers in parentheses are the numbers of subjects studied. H-ras p21 fluorescence was calculated as described in "Materials and Methods." rately for aneuploid tumor and diploid cells. Since the amino acids involved most commonly in ras mutations (nos. 1 2, 13, or 61) and those recognized by the HAS2 antibody (nos. 160-179) do not coincide, the mechanism underlying high H-ras p21 expression in MM remains yet to be elucidated. High p21 expression by aneuploid tumor cells in almost 75% of MM patients suggests an involvement of H-ras in MM. Elevated p2l levels in presumably normal DNA-diploid cells 1.c #{149}" 11 S, ,, H-ru p21 Tote 12 11 oi. p 6 0J3 0 Fig 3. Correlated DNA/H-ras p21 FCM analysis of cells from aneuploid myeloma (A). normal bone marrow (B), and myeloma in remission (C). Cells were processed for 2-parameter FCM analysis by a three-step indirect immunofluorescence and propidium iodide counterstaining method. Five thousand cells were analyzed for each sample and displayed in CUBE3D mode. The hyperdiploid nature of MM cells (A) was ascertained from DNA-clg FCM revealing monoclonal clg confined to hyperdiploid cells (not shown).171 In each experiment MOLT-4 cells served as a positive standard. 0.2 0o.x$ A ES 0 30 60 90 120 Months Fig 5. high (9) cells. Tick patients. Kaplan-Meier survival analysis in myeloma according to or low (<9) H-ras p21 expression in aneuploid tumor marks indicate living patients; closed circles deceased 800 TSUCHIYA ET AL in nine patients raises the question of disease involvement of these cells, analogous to previous observations with CALLA and cIg.#{176T}he more favorable clinical course of patients with low p2! expression is presently unexplained but is reminis- cent of observations in gastric cancer.25 Although weak, the correlation between cellular p21 expression and serum beta- 2-microglobulin might suggest that high p21 expression is associated with a more aggressive clinical course with higher tumor stage.26 Of the six patients with trisomy 1 1, one had high p21 expression, while five had low levels of H-ras p21. Molecular studies are underway to define the mechanism of H-ras expression in MM and its possible relationship to the number of chromosomes 1 1 . High ras expression in MM has not been reported previously. Shen et a127 reported high p21 levels in 75% of patients with acute leukemias. Point muta- tions,27 rare allelic restriction fragments,28 and elevated mRNA levels'3'29 of H-ras gene have been demonstrated in human lymphocytic leukemia and lymphoma. Studies in cell lines and clinical samples from MM patients failed to reveal a relationship between p21 expres- sion and cellular proliferation as studied by bromodeoxy- uridine incorporation and Ki-67 MoAb reactivity. Multipa- rameter FCM analysis is ideally suited to examine the phenotypic and cytokinetic characteristics of tumor cells with high oncogene expression, thus opening entirely new avenues of cell biological investigation of human neoplasia. ACKNOWLEDGMENT We thank Drs Tetsuo Fukumoto, Jun Minowada, and Kiyoshi Takatsu for useful suggestions; Dr Nguyen Van and Nam Hoang for FCM analysis; Leslie Smallwood for clinical data analysis, and Matti J. Scott-Thomas for preparation of the manuscript. REFERENCES I . Varmus HE: The molecular genetics of cellular oncogenes. AnnRevGenet 18:553, 1984 2. Cooper GM, Lane M-A: Cellular transforming genes and oncogenesis. Biochim Biophys Acta 738:9, 1984 3. Barbacid M: Human oncogenes, in DeVita VT Jr. Hellman S, Rosenberg SA (eds): Important Advances in Oncology. Philadel- phia, Lippincott, 1986, p 1 4. Lewis JP, MacKenzie MR: Non-random chromosomal aberra- tions associated with multiple myeloma. Hematol Oncol 2:307, 1984 5. Dewald GW, Kyle RA, Hicks GA, Greipp PR: The clinical significance of cytogenetic studies in 100 patients with multiple myeloma, plasma cell leukemia, or amyloidosis. Blood 66:380, 1985 6. Chen K-C, Bevan PC, Matthews JG: Analysis of G banded karyotypes in myeloma cells. J Clin Pathol 39:260, 1986 7. Gould J, Alexanian R, Goodacre A, Pathak 5, Barlogie B: Plasma cell karyotype in multiple myeloma. Blood 7 1:453, 1988 8. Balmain A: Transforming ras oncogenes and multistage carci- nogenesis. Br J Cancer 51:1, 1985 9. Willingham MC, Pastan I, Shih TY, Scolnick EM: Localiza- tion of the src gene product of the Harvey strain of MSV to plasma membrane of transformed cells by electron microscopic immunocy- tochemistry. Cell 19:1005, 1980 10. Andreeff M, Slater DE, Bressler J, Furth ME: Cellular ras oncogene expression and cell cycle measured by flow cytometry in hematopoietic cell lines. Blood 67:676, 1986 1 1 . Watson JV: Oncogenes, cancer and analytical cytology. Cytometry 7:400, 1986 12. Barlogie B, Alexanian R: Cellular aspects of myeloma: Bio- logic and clinical implications, in Multiple Myeloma and Other Paraproteinemias. Edinburgh, England, Churchill Livingstone 1986, p 154 1 3. Slamon Di, deKernion JB, Verma IM, Cline Mi: Expression of cellular oncogenes in human malignancies. Science 224:256, 1984 14. Tanaka T, Slamon Di, Battifora H, Cline Mi: Expression of p21 ras oncoproteins in human cancers. Cancer Res 46:1465, 1986 15. Tanaka T, Slamon Di, Cline Mi: Efficient generation of antibodies to oncoproteins by using synthetic peptide antigens. Proc NatI Acad Sci USA 82:3400, 1985 16. Drewinko B, Mars W, Minowada J, Burk KH, Trujillo JM: ARH-77, an established human IgG-producing myeloma cell line. I. Morphology, B-cell phenotypic marker profile, and expression of Epstein-Barr virus. Cancer 54:1 883, 1984 17. Barlogie B, Latreille J, Freireich EJ, Fu C-T, Mellard D, Meistrich M, Andreeff M: Characterization of hematologic malig- nancies by flow cytometry. Blood Cells 6:719, 1980 18. Barlogie B, Alexanian R, Pershouse M, Smallwood L, Smith L: Cytoplasmic immunoglobulin content in multiple myeloma. J Clin Invest 76:765, 1985 19. McLaughlin P, Alexanian R: Myeloma protein kinetics fol- lowing chemotherapy. Blood 60:851, 1982 20. Shih TY, Weeks MO, Gruss P, Dhar R, Orszlan 5, Scolnick EM: Identification of a precursor in the biosynthesis of the p21 transforming protein of Harvey murine sarcoma virus. J Virol 42:253, 1982 21. Epstein J, Barlogie B, Jerry Katzmann, Alexanian R: Pheno- typic heterogeneity in aneuploid multiple myeloma indicates pre-B cell involvement. Blood 71:861, 1988 22. Barlogie B, Hall R, Zander High-dose melphalan with autologous for multiple myeloma. Blood 67:1298, A, Dicke K, Alexanian bone marrow transplantation 1986 R: 23. Krishan A, Sauerteig A, Gordon K, Swinkin C: Flow cytometric monitoring of cellular anthracycline accumulation in murine leukemic cells. Cancer Res 46:1768, 1986 24. Marchetti D, Van N, Thompson EB, Barlogie B: Dual flow cytometric analysis of glucocorticoid receptor expression in human leukemic cell lines using monoclonal antibody and fluoresceinated cortisol probes. Cancer Ress 1 988 (in press) 25. Tahara E, Yasui W, Taniyama K, Ochiai A, Yamamoto T, Nakajo 5, Yamamoto M: Ha-ras oncogene product in human gastric carcinoma: Correlation with invasiveness, metastasis or prognosis. Jpn J Cancer Res 77:517, 1986 26. Alexanian R, Barlogie B, Fritsche H: Beta-2-microglobulin multiple myeloma. Am J Hematol 20:345, 1985 in 27. Shen WPV, Aldrich TH, Venta-Perez Furth ME: Expression of normal and mutant acute leukemia. Oncogene 1:157, 1987 G, Franza BR Jr. ras proteins in human 28. Rodenhuis 5, Bos JL, Slater Smets LA: Absence of oncogene activation of N-ras in lymphoblastic 67:1698, 1986 RM, Behrendt H, van't Veer M, amplifications and occasional leukemia of childhood. BlOOd 29. Mavilio F, Sposi NM, Petrini M, Bottero DeRossi G, Amadori 5, Mandelli F, Peschle cellular oncogenes in primary cells from human Proc Natl Acad Sci USA 83:4394, 1986 L, Marinucci M, C: Expression of acute leukemias.