Document 44n8Zp3dvMwZVmqDvLvaVZJV1

Industrial Health. 1991, 29, 37- 56 PLAINTIFF'S EXHIBIT SA-523 37 ORIGINAL ARTICLES Cell Toxicity, Hemolytic Action and Clastogenic Activity of Asbestos and Its Substitutes Kimiko KOSHI, Norihiko KOHYAMA, Toshihiko MYOJO and Kazuo FUKUDA National Institute of Industrial Health. 21-1, Sagao 6-chome, Tama-ku, Kawasaki 214, Japan (Received November 28, 1990 and in revised form March 4, 1991) Abstracts: The cell toxicity, hemolytic and clastogenic activity were examined in various kinds of asbestos and some asbestos substitues with reference to the their mineralogies! and physicochemical characteristics. There were thirty-five fibrous and con-fibrous samples including L'lCC chrysotile, size-selected samples of UICC chrysotile, chrysotile altered by heating and grinding, Yamabe (Japan) chrysotile with long and short fibers. Coalings (U.S. A.) chrysotile with short fibers, UICC crocidolite, amosite, and 19 non-asbestos samples sach as, glass fibers, calcium silicates, sepiotites and some ciay minerals. The cell toxicity and the hemolytic and clastogenic activity of asbestos were the strongest for chrysotile among all of the asbestos samples tested, and their strengths varied with fiber length and with the conditions of grinding and heating. These cellular effects of Yamabe chrysotile with long fibers and size-selected UICC chrysotile with long fibers were stronger than those of chrysotile of the same origin but with short fibers. These effects were weaker in chrysotile altered by beating and grinding. Among the asbestos substitutes, the cell toxicity, hcmoiytic and clastogenic activities of thin glass fibers were more marked than those of thick glass fibers. The four types of sepiolite were strongly hemolytic, but their cell toxicity and clastogenicity varied according to their grade of crystallinity and/or fiber size. These effects of calcium silicates and some clay minerals were generally low but varied with mineral species. In general, the cell toxicity, hemolytic and clastogenic activities of the asbestos substitutes tested here were mild compared with those of asbestos. Key words: Asbestos--Long fiber--Short fiber Altered asbestos--Asbestos substitutes Cell toxicity--Hemolysis Chromosome aberrations To whom all correspondence should be addressed: Dr. Kimiko KOSHI. Industrial Physician, Central Motor Co.. Ltd, 4-12. Ohyama-cho, Sagami- "' fiara, TCanagawa 229. Japan I Q..3nw.^y.V.y.V.>^V.A>y.VV.yhi:AViX..r..l.A.AA\.VV\Vl\^MVTYT-.i-rvy>>i%.^yf^'[^1^pir^^ -T.-fcAB 38 K, K.OSHI. st aL Introduction Asbestos inhalation is strongly associated with an increased risk of asbestosis, carcinoma of the lung and pleural and peritoneal mesothelioma.I !1 These adverse health effects have been confirmed by many epidemiologic studies and by animal experiments. Pleural mesothelioma can be induced in rats by intrapleural injec tion of asbestos fibers, and pulmonary carcinoma by inhalation exposure.1 Sl General recognition of the hazardous effects of asbestos for human health has led several (inter)national agencies to recommend the prohibition or restriction of asbestos uses. These policies have promoted worldwide development and produc tion of various asbestos substitutes. However, their toxic effects are not well known. Stanton et al.<s> and Pott'' reported that long, thin fibers were more carcinogenic than short, thick fibers, and that fiber carcinogenicity depends more on the fibers' durability than their physicochemical properties. If so. the dimen sions of the asbestos substitutes become an important factor in their carcinoge nicity. However, whether the other factors, such as chemical composition and the surface properties, affect the carcinogenicity or not is still obscure. In 1988, the Japanese Environmental Agency reported" that in the air of the areas along motorways and some arterial roads, asbestos fibers shorter than 2 /im or asbestos fibers with altered structures were several times more abundant than hi other areas such as residential zones. Braking by automobilies was found to be responsible for the increased air levels of both the short asbestos fibers ar.d the altered non-crystalline dusts whose original asbestos morphology and crystal structure had been destroyed by the heating and grinding during braking. These studies show the importance of further study to reveal the biological effects of various kinds and sizes of fibers, such as short fibers less than 2-3 /im in length, altered asbestos, and asbestos substitutes. In thte investigation, we prepared samples of asbestos with long fibers and short fibers, and of alteredasbestos and asbestos substitutes such as glass fibers, calcium silicates and sepiolites, and assessed their in vitro biological effects, such as the ceil toxicity and chromosome aberration in Chinese hamster lung cells, and the hemolysis of human erythrocytes. Materials and Methods 1. Mineral samples of asbestos and asbestos substitutes and their method of preparation. Thirty-five kinds of fibrous and non-fibrous mineral samples including asbestos, glass fiber, calcium silicate fiber, sepioiite and some clay minerals, were prepared for in vitro experiments as shown in Table 1. The asbestos samples prepared were chrysotile UICC-B (from Canada); its size-selected samples and altered samples "which were heated and/or ground; Coalinga chrysotile from California, :bestosis, adverse / animal al injec. > ;> alth has iction of producnot weil re more ds more ; dimenrcinogeion and r of the an 2/zm =nt than ound to ad the crystal giological 2-3 fim ion, we alteredd sepio'ity and lysis of of .sbestos, ,repared .repared altered lifomia, IN VITRO EFFECTS OF ASBESTOS AND ITS SUBSTITUTES 39 Table 1 Samples and the origins prepared for in vitro experiment. SAMPLE ORION PREPARATION ASBESTOS 1. Chrysotile 2. ' 3. " 4. " 5. " 6. " 7. " 8. " 9. " 10. " 11. " 12. " 13. " 14. " 15. Crocidolite 16. Amosite CLAY MINERALS 17. Antigolite 18. " 19. Halloysite 20. Kaolinite 21. Mg-Chlorite 22. Hemalite CLASS FIBER 23. MN S 100 24. MN # 104 25. MN # 108A 26. MN#108B CALCIUM SILICATE 27. Gyroiite 28. 11A Tobermolite 29. a-dicalchum silicate hydrate 30. Xonotlitc 31. Wollastonite SEPIOLITE 32. Sepiolite 33. " 34. " 35. ' UICC-B, CAN UICC-A, Rhodesia Coalinga 3778, Ca, USA Y-A, Yamabe, JPN Y H(3), Yamabe. JPN Y-C, Yamabe, JPN UICC-B-#200 UICC-B- #635 UICC B -R UICC-B-600'C UlCC-B-80O:C UICC-B-G2 UICC-B-G3 UICC-B-800G UICC UICC Ant-Ko-C, Komon, JPN Ant-Ko-F. Kfimori, JPN Argeria KGa-1. Georgia, USA Wanibuchi, Tottori, JPN Brucite, Quebec, CAN Manville, USA Manville, USA Manville, USA Manville, USA Synthesized by M Synthesized by M Synthesized by M Synthesized by M Quebec, CAN Sepio-C, China Scpio-K, Kuzuu, JPN Sepio-S, Spain Sepio-T, Turky 0 0 0 1 2 1 3 3 3 4 5 6 */ 8 0 0 9 1 I 1 I 10 u 11 II a 12 12 12 12 9 13 I 13 13 U.S.A.; Yamabe chrysotile from Hokkaido, Japan; UICC-amoist; and UICCcrocidolite. The glass fibers, calcium silicate fibers and sepiolite studied in this experiment have been widely used as asbestos substitutes in modem industrial countries. The clay minerals in Table 1 are of non-fibrous forms except for halloysite and nemalite. The sample preparation methods, keyed to the numbers for each mineral in Table 1, were as follows 1 40 K. KOSHI, ei at. 0: The samples, i.e. UICC standard asbestos (chrysotile-A and B, amosite and crocidolite) and chrysotile from the Coalinga mine, California, were used as obtained. Details of the sample preparation method have been published.*'''J) 1: Fiber-sizes were separated by an elutriation method using a suspension of crushed original mineral to produce particles with < 10 fim equivalent spher ical diameter (<10/im e.s.d.). 2: After the above elutriation procedure, the fiber sample was stirred with a magnetic stirrer to remove a small amount of coexisting magnetite. 3: Fiber-sizes were separated using a dry size-selective procedure with a 2 -component fluidized bed, wire screens (200 and 635 mesh) and virtual impactor. 4: Heating for 1 h at 600cC. 5: Heating for lh at 800C. 6: Grinding for 1 h with ball mill. 7: Grinding for 2 h with a ball mill. 8: Heating for 1 h at 800 C and then grinding for 1 h with a ball mill. 9: Grinding by hand in an agate mortar. 10: Grinding for 30 min in an ethanol solution using a ball mill. 11: Crashing at an impact pressure of 25 ton/inch! using an oil-pressure appara tus. 12: Synthesis from a mixture of CaO and silicic acid by a hydrothermal forma tion procedure. 13: Pulverizing with an automatic milling machine and collection of small fractions by an air-stream separation method. These samples were analyzed by X-ray diffraction, electron microscopy, ther mal, chemical, and surface charge analysis. Some characteristics of each sample were as follows (Further details will appeare elsewhere.): [J] Chrysotile UICC-B: Chrysotile UICC-B consisted mostly of long fibers, as shown in the transmission electron micrograph in Fig. la, and did not contain any other fibrous minerals. However, there were small amounts of platy miner als, such as brucite, pyroaurite, lizardite and antigolite. 12] Size-selected chrysotile (UICC-B- # 200. UICC-B-i 635, and UICC-B-R): Chrysotile UICC-B was separated into various size fractions by a dry procedure using a 2-component fluidized bed, wire screens (200 and 635 mesh) and a virtual impactor with a cut-off point of 2/4m.111 The size distribution of the fraction sizes is shown in Fig, 2. Eighty-one percent of the short-fiber fraction (UICC-B-SS 635) consisted of fibers less than 5 /zm long and 96% less than 10/iin long. Eighty percent of the middle fiber fraction (UICC-B- it 200) consisted of fibers less than lO.uin long. The residue fraction consisted mostly of long fibers and was designated UICC-B-R. X-ray diffraction analysis revealed that the crystalline state of these fractions was not markedly changed. _3_ ' Chrysotile from Yamabe i Y-A, Y-H(3), and Y-C : Chrysotile from site and used as shed.''"" ision of t spher- with a ilh a 2 virtual i!l. appara- forma t small y, thersample bers, as contain miner- :~R): icedure and a of the racfion iO/fm sted of : fibers tat the from 3N ViTRO EFFECTS OF ASBESTOS AND ITS SUBSTITUTES 41 Fig. TEM photographs of the asbestos and asbestos substitutes used in this study, ta: Chrysolite UICC--B lb; Chrysolite Y-A r 1c: -Chrysolite y4|(3) Id: Glass fiber MN i 100 le: Xonotulite sythcsizeil If: Sepiolitc Sepi C Scale indicates 5 u 42 K. KOSHI, <*i at FIBER LENGTH (^m) Fig. 2. Distribution of filter length in size-selected chrysotile B Yamabe, Hokkaido, Japan, was categorized into 3 different types, the long-fiber type (Y-A), the short-fiber type (Y-H(3)3 , the short fiber type with lowcrystallinity (Y-C) containing 4 different polytypes.,2> TEM photographs show that Y-A consisted of very long fibers and Y-H(3) very short fibers mostly less than 2 jim long (Figs, lb and lc). A small amount of magnetite contamination in Y-H(3) was mostly removed with a magnetic stirrer. Chrysotile (Y-C) was an unusual type showing a greenish-white color and containing one-layer clino-, 2 -layer clino-, 2-layer ortho-, and 6-layer ortho-chrysotiels. The fiber length was mostly under 2.um, similar to Y-H(3), but the crystallinity was low according to X-ray diffraction analysis. //I Chrysotile samples altered by heating and grinding: Chrysotile UICC-B was. heated at 600 C and 8G0C for Iff (UICC-B-600 and UICC-B-800, respec tively) and was ground for I hand 2 h (UICC-B-G2 and UICC-B-G3, respec mg'fiber th lowis show stly less .lination C) was rlino-, 2 gth was ding to ' JICC-B respecrespec- IN VITRO EFFECTS OF ASBESTOS AND ITS SUBSTITUTES 43 tively). The typical X-ray diffraction pattern of chrysotile still remained in UICC-B-600, but the peak intensity was slightly weakened. The fibrous form of chrysotile was still evident by TEM observation. The chrysotile was transformed into forsterite in UICC-B-800. The fibrous form, however, still remained in this sample, but was very brittle and changed easily into small particles when ground in the agate mortar. Some fibrous form of chrysotile still remained in UICC-B-G2, but in UICC-B-G3, it had completely disappeared and changed into the fine particles of less than 0.1 pm. [J] Glass fiber (MNP 100, MNP 104, MNp 108A and MNP 108B): The glass fibers used were made commercially by Manville Co., U.S.A., code ff 100, having a nominal diameter of 0.29-0.32/tm; code P 104, 0.39-0.53/rm; code $ 108A, 0.69-1.1/tm; and code P 108B, 1.2-2.4/rm. The fiber samples were crushed by the impact pressure method. In the crushed sample of MN P 100, 90% were less than 5 pm long and 95% were less than 10/Um long (Fig. Id), the other crushed samples (MN S 104. MN108 P A and MN P 108B) were of similar fiber length. X-ray diffraction analysis showed that these samples were in an amorphous state and did not contain any mixed or crystalline materials. [6] Calcium silicate fibers: in this experiment, some synthetic calcium silicates such as xonotolite, alpha-dicalcium silicate hydrate, 11A-tobemolite and gyrolite were examined, as well as a natural wollastonite. According to TEM analysis, the 1 lA-tobermolite had a thin sheet-like or elongated hexagonal platy form a few pm in diameter. The xonotolite had a fibrous form about 5 pm in length and less than 0.1 trm in diameter (Fig. le). The gyrolite had a thin platy form and the alpha-dicalcium silicate hydrate had various forms ranging from gel-like to fibrous. The natural calcium silicate, wollastonite, from Quebec, Canada, had a one-layer triclinic structure and a wide range of fiber sizes, mostly long and thick fibers. [7] Sepiolite (Sepio-C, Sepio-K, Sepio-S and Sepio-T): Sepiolite samples from four different natural sources were used. They were obtained from China (SepioC), Kuzuu, Japan (Sepio-K), Spain (Sepio-S) and Turkey (Sepio-T). X-ray diffraction analysis revealed that the crystallinity of these sepiolites descended in the order Sepio-C, Sepio-K. Sepio-S and Sepio-T. Small amounts of talc and calcite were detected as contaminants in sample Sepio-C, but never in the other samples. The sample of Sepio-C consisted mostly of fibers 1-10 ^m long and 0.05 -0.1 /im in diameter, but long and thick fibers over 20 pm long and over l,am in diameter were also present (Fig. If). The fibers of Sepio-K were mostly 3-7pm long and 0.01-0.07 pm in diameter. The fibers of Sepio-S and Sepio-T were mostly 3-5pm long and 0.01pm in diameter. The order of crystallinity of these sepiolite samples were well coordinated with the fiber size (length and diameter). r,S] Clay minerals: To compare the in vitro experiments with non-fibrous mineral? with those of the fibrous minerals above, some clay minerals with a platy form were selected. These were antigolite from the Komori mine, Kyoto, 44 K. KOSHI, et al. Japan (with a iath shape), kaolinite from Georgia, U.S.A. (with a hexagonal platy form), and magnesium chlorite from Wanibuchi mine, Tottori, Japan (with an irregular platy form). In addition, halloysite, a kind of kaolin mineral with a short fibrous form, and nemalite, a kind of brucite with a fibrous form, were also prepared. 2. Cell toxicity Chinese hamster lung (CHL) cells were obtained from Dainippon Pharmaceuti cal Co. Ltd. CHL cells were grown in minimal essential medium (MEM) (GIBCO) sup plemented with 10% fetal calf serum (GIBCO) and with penicillin (50 units/ml) and streptomycin (50 /rg/ml) (GIBCO). Survival of CHL cells in the absence or presence of the mineral dusts was determined by colony-forming efficiency from a single-cell suspension. Approxi mately 200 cells were seeded into 60-mm-diameter Petri dishes (Falcon) along with the test dust. The surviving cells were allowed to grow and form colonies for 7 days. They were then fixed with 95% ethanol and stained with Giemsa solution. 3. Hemolysis Human blood was taken from the cubital vein, and blood cells were washed three times with veronal-buffered saline. An erythrocyte suspension (approximate ly 2%) was prepared by making up 2 ml of packed erythrocytes to 100 ml of veronal-buffered saline. To 1 ml of this suspension, 1 ml of veronai-bufFered saline containing various amounts of mineral dusts was added, and the mixture was incubated for 1 h at 37C with shaking at 40 strokes for 1 min. Then, 0.2 ml of a 2.5% solution of glutaraldehyde was added to each sample to stop hemolysis. The samples were then centrifuged at 2000 r.p.m. for 10 min and the optical density of the clear supernatant was read at 541 nm using a spectropho tometer. Complete lysis was obtained by preparing a 2% erythrocyte suspension in distilled water with two drops of Triton X-100 (Sigma Chemical Co. U.S.A.) and 0.2 ml of 2.5% glutaraldehyde solution. The hemolytic activity of the test dusts was calculated from the average optical density of three replicates and was expressed as a percentage of the optical density of the totally lysed control (100%). 4. Cultures and chromosome preparation A pseudo-diploid CHL cell line was used. The cells were grown in MEM supplemented with 10% fetal calf serum. The cells (4 X 10s cells) were seeded in a 25-cmz Falcon plastic flask, and various amounts of the test dusts were added Jo the cultures. All culturing was at 37C m a 5% C02 atmosphere for 46 h and 0.05 jj.% of colcemid/ml was added for the final 2 h. After incubation, rxagonai in (with ral with in, were maceuti- 0) supnits/ml) ists was \pproxi1) along colonies Giemsa washed ximateJO ml of buffered mixture hen, 0.2 to stop and the ctrophospension U.S.A.) the test and was control . MEM : seeded .ts were here for ubation. IN VITRO EFFECTS OF ASBESTOS AND ITS SUBSTITUTES +5 the cells were collected by scraping with a spatula and treated with 0,075 M calcium chloride for 15 min, followed by fixation with a mixture of ethanol and acetic acid (3:1) at 0C. The fixed cells were spread on a glass slide then stained with Giemsa solution. The Giemsa-stained slides were examined for chromosome and chromatid aberrations, each type of aberration in one cell being scored separately. Results 1. Cytotoxicity The toxic doses (TD*,: the amounts of dust necessary to inhibit the colony forming efficiency of CHL cells by 50%) of mineral samples are shown in Table 2. Among the UICC reference asbestos samples, chrysotile was the most toxic, followed by amosite and crocidolite. The toxic doses of chrysotile A, chrysotile B, amosite and crocidolite were respectively, 1.6, 1.2, 4.5 and 7/zg/ml. As to the relationship between cell toxicity and fiber length, Yamabe chrysotile with long fibers (Y-A) was more toxic than the same chrysotile with short fibers (Y-H (3); . The toxic dose of Yamabe chrysotile with short fibers was 3.7/rg/ml and that of Yamabe chrysotile with long fibers was 0.2 tig/ml. Among the size- selected chrysotile UICC-B, the short-fiber fraction (UICC-B- 635) had the weakest toxicity for cells and the toxicity increased with increasing fiber length. The toxic dose of the long-residure fiber fraction (UICC-B-R) was 0.3/ig/ml, those of the middle-fiber (UICC-B- $ 200) and short-fiber fraction (UICC-B- $ 635) were 0.9,g/ml and l.ljug/ml, respectively. The toxic dose of Coalinga chrysotile with short fibers was 0.9 /rg/ml. The toxic dose of chrysotile altered by grinding for 2 h was 25/zg/ml and that of the chrysotile heated at 800 C for 1 h and ground for 1 h was 67/zg/ml. The toxic doses ofglass fiber codes If 100, 104, 108A and 108B (Manville Co.) were 10, 11, 18 and 27 /rg/ml, respectively. The calcium silicate samples used were synthetic materials such as fibrous xonotolite, a-dicalcium silicate hydrate, 11A tobermolite, gyrolite and natural wollastonite. The toxic doses of fibrous xonotolite, a-dicalcium silicate hydrate and natural wollastonite were more than 50/tg, respectively, and that of 11 A tobermolite was 45tzg/ml. The toxic dose for gyrolite was 8/zg/ml. The toxic dose of sepiolite from China was 0.9!g/ml, which was the same level as that of UICC chrysotile. The toxic dose of sepiolite from Kuzuu was 2.5/tg/ml. Those of sepiolites from Turkey and Spain were 30/rg/ml and 50fig /ml, respectively. > 2. Hemolysis studies _ The .hemolytic doses* (HDK; dust dose necessary to cause 50% hemolysis of human erythrocytes) of various fibrous samples are shown in Table 3. r 3 i ....... 46 V.V.'.V.V.V.W 11ITIT.tfrtWi K. KOSHI, et ai T D 3(): The amounts o f dust necessary to inhibit the colony-forming efficiency o f C H L cells by 50;ifiv S IN VITRO EFFECTS OF ASBESTOS AND ITS SUBSTITUTES 47 The amounts o f dust necessary to inhibit the colony-forming efficiency o f C H L cells by 50%. sc; i2iI ! o o w w V (l G C <0 V , XS <<Z; 5 xi .2 M 1> CJ *rJ S 'a *> 2 p .2 3 2 5 O a -5 X .:1.Hk3*. *y~* ,11 !H J <o 8 Si ^5t <GC <Mma ;U y ^ ip e 1 rn :; GQ ~O lu -.| \V -a) !"> A: 1S020 . 3 u >> yi D -E- -C ] fn o >< " ^> jj ! " IS as Jj X o T o >- q ,\J. i:rA. - j>i. y -h v ? !Z Jlf >s? <, 2tu.0 o .Hiy iaL <m 2 8 b? az 3o .11 .o o < u ___I S C! Sp Q< C X X CS5? i Vi 48 K. KOSHI, et at. To achieve 50% hemolysis, less than 200/Ug/ml of UICC chrysotile A and B, Yamabe chrysotile with long fibers (Y-A) and Coalinga chrysotile was needed. UICC amosite and crocidolite rarely produced 50% hemolysis at any concentra tion tested. Yamabe chrysotile with long fibers (Y-A) was slightly more hemolytic than chrysotile with short fibers (Y-H(3)). The residue fraction (UICC-B-R) of the size-selected UICC chrysotile was more hemolytic than the middle-fiber (UICC-B$ 200) or short-fiber fraction (UICC-B- f? 635). The hemolytic activity of chr ysotile altered by grinding was reduced with increased grinding time. Also, the UICC chrysotile altered by grinding for 1 h and heating for 1 h at 800C (UICC-B-800G) was less hemolytic than chrysotile altered by grinding only. Among the asbestos substitutes, the hemolytic potential of sepiolite was the stronger than glass fiber, calcium silicates and other clay minerals. In the sepiolite group, the sample from Kuzuu was the most hemolytic, followed by sepiolite from Spain and Turkey. Their hemolytic doses were 110, 195 and 210 tig/ml, respectively. The sepiolite from China was the least hemolytic. The sepiolite from China, showing the highest cytotoxic activity, had the weakest hemolytic activity, 350/ig/ml. Among the glass fibers, the hemolytic dose of the thinest fiber (Code 100) was 800/Ug/ml, and those of thicker glass fibers (Code If 104, 108A and B) were above 1000/zg/ml. To achieve 50% hemolysis, 210/ig /ml of gyrosite was required, whereas 260 trg/ml of a-dicalcium silicate hydrate, 380/ig/ml of xonotolite and more than 1000/rg/mI of 1IA tobermoiite were needed. Among clay mineral fibers, the hemolytic doses of halloysite and Mg-chlorite were 225 and 350 Mg/ml, respectively. Those of antigolite C, antigolite F (prepared under 5 ft), kaolinite, and fibrous nemalite were more than lOOOjUg/ml.. 3. Chromosomal aberrations and induction of polyploidy UICC chrysotile, crocidolite and amosite induced structural chromosomal aber ration and polyploidy. As shown in Table 4, chrysotile was the most clastogenic, followed by crocidolite and amosite. The polyploidy occurred frequently in the amosite sample, whereas the structural chromosomal aberrations were mainly chromatid gaps. The aberrations in the cultures treated with chrysotile-UICC-B included gaps, chromosome and chromatid breaks, chromatid exchanges, dicen trics and polyploidy. Upon treatment of the cultures with size-selected chrysotile-UICC-B, the longresidue fractions (UICC-B-R) had the highest clastogenicity followed by the middle-fiber fraction (UICC-B- <f 200) and the short-fractions (UICC-B- ff 635) (Table 5). Table 5 shows that long-residue fraction induced high levels of polyploidy and structural chromosome aberrations. The aberrations in the cul tures .-treated with the long-residue fraction included gaps, chromatid breaks, dicentrics and chromatid exchanges. Those in the cultures treated with the \ and B. > needed, oncentra- ytic than l) of the UICC-Bv of chrAlso, the t 800 C only, was the In the owed by and 210 ic. The weakest se of the rs (Code S 210/ig hydrate, zc were site and C, antiore than nal aberstogenic, !y in the mainly UTCC-B s, dicen- he longby the i-If 635) levels of the cut- breaks, >Wth the IN VITRO EFFECTS OF ASBESTOS AND ITS SUBSTITUTES 49 Table 4. Chromosomal aberrations In CHL cells treated with UICC asbestos. Treatment Control Chrystotile Crocidolite Amositc Dose (Vg/ml) 0 5 10 30 100 10 30 100 10 30 100 Abnormality (%) Gaps Breaks Dicentrics Exchanges 20 70 12 3 S2 not observed 40 12 0 not observed 20 20 60 0 2 2 2 2 0 0 0 0 0 l 2 1 0 0 0 0 0 Others 0 0 0 0 Polyploids (%) 0 38 34 50 0 24 0 32 0 18 0 36 0 42 Table 5. Chromosomal aberrations in CHL cells treated with slze-selectcd chrysotile. Treatment Control Short-fraction (UICC-B- i 635) Middle-fraction (UICC-B- i 200) Residure-fraction (UICC-B-R) Dose (//g/ml) . Gaps 02 10 4 30 4 100 ' 4 10 6 30 4 10 4 30 12 Abnormality {%) Breaks Dicentrics Exchanges 00 00 00 00 04 24 26 06 0 0 0 0 0 0 4 6 Others 0 0 0 0 0 0 0 0 Poiypioids (%) 0 34 34 38 46 38 66 65 short-fiber fraction included only gaps and low levels of polyploidy. In the Yamabe chrysotile sample with long fibers (Y-A), the aberrations observed also included gaps, chromatid breaks, dicentrics, chromatid exchanges and rings, whereas the Yamabe chrysotile with short fibers (Y-H(3)) and Coalinga chrysotile did not induce any significant increase in the incidence of such chromosome abnormalities, but mainly increased the incidence of chromatid gaps, isochromatid gaps and polyploidy. Table 6 shows the frequency of chromosome abnormalities found in metaphase spreads of CHL cells exposed to Yamabe chrysotile with long and short fibers. Table 7 shows that chrysotile altered by grinding for 2 h (UICC-B-G3) did not induce structural chromosome abnormalities, but only low levels of polyploidy. The results obtained from treatment with asbestos substitutes are shown in Table 8. Glass fibers, codes # 100 and 1104, did not significantly increase structural chromosome aberrations. They caused only relatively high levels of polyploidy. Code # 108A induced neither structural nor numerical chromosome aberrations up to dosage of 30//g/ml. Among calcium silicates, synthetic UA tobermolite, a-dicalcium silicate hydrate, gyrolite and natural wollastonite did not significantly increase polyploidy and structual chromosome aberration. But in the xonotolite induction of polyploidy and structual chromosome abberation, they 50 K. KOSHI. et aL Table 6. Chromosomal aberrations in CIIL cells treated with long nber-chrysotile and short fiber-chrysotite. Treatment Dose Cag/ml) Control Yamabe (Y-A) (long) Yamabe (Y-H(3)) (short) Coalinga (short) 0 5 10 30 10 30 100 10 30 Abnormality {%) Gaps Breaks Dicentrics Exchanges 0 10 4 10 0 not observed 20 40 42 20 80 0 4 4 0 0 3 0 0 0 2 2 0 0 1 0 0 Others 0 2 (ring) 0 0 0 0 0 0 Polyploids (%) 0 43 54 18 30 40 22 34 Table 7. Chromomomal aberrations in CHL cells treated with chrysotile altered by grinding and heating. Treatment Dose Cug/ml) Control Chrysotile (grinding for 2 h) Chrysotile (heating at 800" C, grinding for 1 h) 0 10 30 100 300 10 30 100 300 Gaps 2 0 0 1 3 0 0 0 0 Abnormality (%) Breaks Dicentrics Exchanges 00 00 00 00 00 00 00 00 00 0 0 0 0 0 0 0 0 0 Others 0 0 0 0 0 0 0 0 0 Polyploids (%) 0 4 4 4 4 2 0 2 2 were stronger than the other calcium silicates tested. The aberrations in the cultures treated with xonotolite included gaps and dicentics. Among the sepiolite samples, the sepiolite from Kuzuu, Spain and Turkey did not significantly increase polyploidy and structual chromosome abberation up, to a dosage of 100//g/ml. But the sepiolite from China induced a relatively high level of polyploidy, approximately the same as glass fiber code i? ICO. The aberrations in the cultures treated with the sepiolite from China included exchanges only by the addition of lOytg/ml. In the additions of 30 and 100/ig/ml they caused only chromatid gaps. Figure 3 shows photographs of numerical and structual chromosome aberra tions in the cells treated with fibers. Discussion In an earlier study, we'5> demonstrated that among various asbestos and the related mineral samples, chrysotile had the highest hemolytic and cytotoxic effects on erythrocytes and macrophages. To estimate the cell toxicity of asbestos, several primary and cultured cell lines have been used as experimental materials since the 1970's. The results obtained from the experiments using macrophages lie loids (%) 0 48 54 18 30 40 22 3+ grinding ploids (c/o) o 4 4 4 4 2 o 2 2 us in the : sepiolite .nificantly losage of level of rations in ily by the used only ie aberra- . and the xic effects asbestos, materials rrophages IN VITRO EFFECTS OF ASBESTOS AND fTS SUBSTITUTES 51 Tabic 8. Chromosomal aberrations in CHI. cells treated with asbestos substitutes. Treatment Dose (/rg/ml) Control Glass fiber Code $ 100 Code # 108A Ca. silicate Xonotolite 11 A. Tobermolite a-dicalcium silicate hydrate Gyrolite Wollastonite Sepiolite China X u/.uu Spain Turkey 0 10 30 100 300 10 30 100 300 10 30 100 300 10 30 100 300 10 30 100 300 10 30 100 300 10 30 100 300 10 30 100 300 10 30 100 300 10 30 100 300 10 30 100 300 Gaps 2 Abnormality (%) Breaks Dicentrics Exchanges 00 0 20 20 20 30 20 40 20 20 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 20 00 20 20 00 00 00 00 00 00 00 00 20 40 20 20 20 20 20 20 0 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 40 40 not observed 00 20 00 00 00 00 20 40 00 20 00 00 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 4 0 0 0 0 0 2 0 0 0 1 0 0 0 0 Others 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Polyploids (%) . 0 lb 22 26 32 4 4 14 18 8 10 12 20 4 6 4 5 4 4 4 4 6 4 4 4 2 2 4 12 21 26 32 6 6 8 8 6 8 6 16 4 6 6 8 52 K. KOSHI, et at. 1) * * -- 4) Fig. 3. Photograph* of chromosome of CHL ceils 1) Chromosome of CHL cells (no addition of fibers, control) 2) Polypoid (addition of Yamabe chrysotlle with short fibers) 3) Polypoid including dicentric (addition of UICC chrysotlle) 4) Polypoid including chromatid exchange and break (addition of Yamabe chrysotlle with long fibers) 5) Polypoid Including gap (addition of Yamabe chrysotlle with short fibers) % *  \ U of CHE cells , addition of "fysotile with lion of UICC *change and tie with long of Yamabe IN VITRO EFFECTS OF ASBESTOS AND ITS SUBSTITUTES S3 and cultured cells showed almost the same tendencies. However, it seems difficult to infer the fibrogenicity and carcinogenicity of asbestos directly from their cellular effects. In 1978, Chamberlain and Brown'41 referring to data of Wagner1' who conducted tumorigenicity experiment of the mineral dusts, found that there is a qualitative relationship between the tumorigenicity and in vitro cytotoxicity of fibrous minerals. That is, the activity of fibrous dust in reducing the colony forming efficiency of V79-4 cells was related to the tumorigenic potential of the dusts. Therefore, using the same method, we tested the amounts of dust necessary for inhibiting the colony-forming efficiency of Chinese hamster lung cells (CHL cell line) by 50%. Chrysotile and crocidolite asbestos have been reported to induce chromosome aberrations in Syrian hamster embryo cells1511' and Chinese hamster ovary cells,171'' V79-4 cells,15' human mesothelial cells,"' and rat pleural mesothelial cells.211 However, there has been no comparative study on the clastogenicity of asbestos with short fibers vs long fibers, originating from the same sample. In this study we obtained samples having different fiber size distributions originating from the same sources. To obtain a chrysotile sample with short fibers, we had not ground a long fiber sample with a ball-mill or an agate mortar. The short-chrysotile fibers Were separated from the original (UICC) sample by aerodynamically22', or collected from the same geological environment in a mine (Yamabe, Hokkaido, Japan), where also producing long chrysotile fibers.-There fore, the crystal structure, especially the surface property of the short fibers was considered to be unchanged or to be equivalent with the long fiber sample. It is significant that the cytotoxicity, clastogenic and hemolytic activities of the short-fiber chrysotiles, such as the short-fiber fraction of the size-selected UICC chrysotile or the short-fiber Yamabe chrysotile, were weaker than those of the middle-fiber and long-residue-fiber fractions of the size-selected UICC chrysotile and Yamabe chrysotile with long fibers. And in particular both the polyploidy and structual chromosmal aberrations induced by the long-fiber chrysotile were more marked than those induced by the short-fiber chrysotile. That is, dicentrics, chromatid exchanges and chromosome breaks were induced by the long-fiber chrysotile, whereas the short-fiber chrysotile induced mostly chromatid gaps. There have been two reports of reduced cytotoxicity in the macrophages22' and V-79 cells1*' as a result of grinding asbestos. Also, it has been reported that the clastogenic activity of ground crocidolite in CHO-K1 cells17'1'1 was reduced in comparison with the original crocidolite. In our study, the crystal structure of chrysotile disappeared with increasing grinding time, and the cytotoxicity and the clastogenic and hemolytic effects were reduced. A ball milling was also not used in this study to obtain short fibers of asbestos substitutes. Glass fibers were crushed using an oil pressure apparatus and their fiber shapes were kept af shown in Fig. 1-d. The average diameter of the code # 100 glass fiber was approximately the same as that of the chrysotile fibers. The 54 K. KOSHI, et al. average length of the code # 100 was simillar to short fraction of size-selected UICC chrysotile and was slightly longer than the Yamabe chrysotile with short fibers. The cytotoxicity, dastogcnicity and hemolytic action of the Code Sf 100 were slightly weaker than these chrysotile samples. Further, the cytotoxicity and the biological effects of the glass fibers were reduced with increasing diameter. Among the sepiolite samples, the fiber size decreased in the order of their sources from China, Kuzuu, Spain and Turkey. The levels of hemolytic action of sepiolites from Kuzuu and Spain were similar to those of UICC chrysotile samples and were generally stronger than that of the other mineral fibers tested. The order of cell toxicity and clastogenic activities of the sepiolites seems to be well correlated with crystallinity and/or fibe size. The clastogenicity of the sepiolite from China was moderate, but that of the other sepiolite samples was weak. Since the* clastogenicity varied with the sepiolite samples, the tumorigenic potential would varied according to the degree of crystallinity and fiber size, although Wagner"' has reported that a sepiolite from the U.S.A. did not induce mesothelioma after intrapleural injection into F344 rats. However, judging from these results, we supposed that a sepiolite fibers having good crystallinity and/or long and thin fiber-shape will be more carcinogenic and will induce mesothelioma in an animal experiment, and we25' produced mesothelioma in female SD rats' pleura by injection of sepilite from China which is composed of long fibers, but no mesothelioma was produced by sepiolite from Spain with short fibers. Recent ly, Pott et ai."' reported similarly that sepioiite from Finland induced abdominal tumors (sarcomas or mesotheliomas excluding tumors of the uterus) after intraperitoneal injection in female Wister rats, but sepiolite from Unicaluaro under the same conditions did not. Among the calcium silicate samples, such as xonotolite, 11A tobermolite, gyrolite and a-dicalcium silicate hydrate, and natural wollastonite, the cytotox icity, hemolytic action and clastogenicity were weaker than thos'e of the chrysotile sample. Although some reports have shown that natural wollastonite is more toxic to macrophages27' or erythrocytes2*' than the synthetic calcium silicates, the cellular effects of the wollastonite tested in our study were no worse than those of synthetic calcium silicates. Stanton et al.2'' have reported that the probability of developing pleural tumor was best correlated with the number of fibers with diameters less than 0.25 jum and lengths greater than 8 /m. In our study, the frequency of polypoidy induction was related to the length of fibers in various fibrous samples. Acknowledgment The authors are very grateful to Mrs. K. Suzuki for her skilled technical assistance. c-selected ith short de S100 .icity and iameter. of their ic action ehrysoti'le rs tested, ms to be y of the iples was norigenic iber size, it induce .ing from y and/or ithelioma SD rats' bers, but Recent'.ominal after inro under ermolite, cytotoxhrysotile is more ;ates. the an those al tumor 0.25 fica olypoidy technical fN VITRO EFFECTS OF ASBESTOS AND ITS SUBSTITUTES 55 References 1) Becklare MR. Asbestos-related diseases of the lung and other organs. Am Res Respir Disease 1976: 114: 187-227. 2) SelikofF iJ. I.ung cancer and mesotheriema during prospective surveillance of 1249 asbestos insulation workers. 1963-1974, Ann N.Y. Acad Sci 1976; 271: 448-56. 31 Reeves Ai, Puro HE. Smith RG. Vonvald AJ. Experimental asbestos carcinogenesis. Environ Res 1971: 4: 496-511. 4) Stanton MF, Wrench C. Mechanisms of mesotherioma induction with asbestos and fibrous glass. J Natl Cancer Inst 1972: 48: 797-821 5) Wagner JC, Berry G, Timbrcll V. Mesothelioma in rats after inoculation with asbestos and other materials. Brit J Cancer 1973: 28: 173-185. 6) Stanton MF, Layard M, Tegerls A et al. Relation of particle dimension to carcinogenicity in amphibole asbestos and other fibrous minerals. 3IC1 1981; 67 : 965 75 71 Pott F. Some aspects on the dosimetry of the carcinogenic potency of asbestos and other mineral dusts. Staub-Reinhait. tuft 1978: 38: 486-90. 8) Asbestos abatement manual Japanese Evironmental Agency ed. Tokyo: Gyosei, 1988 1.37-51. (In Japanese) 91 Timbrel! V, Gilson JC, Webster I. UICC standard reference samples of asbestos. Int J Cancer 1968; 3: 406-8. 101 Mumpton FA, Thompson CS. Mineralogy and origin of the Coalinga asbestos deposit. Clays and Clay Minerals 1975; 23: 131-43. ' U) Myojo T, Kohyama N. Separation of asbestos fibers by length-procedure for obtaining differentlength samples for biological experiments. J Society of Powder Technology 1990 (In Japanesel 12) Kohyama N. unpublished 13) Koshi K, Hayashi H, Sakabe H. Cell toxicity and hemolytic action of asbestos dust. Ind Health 1968; 6: 69-79. 14) Chamberlaim M, Brown RC. The cytotoxic effects of asbestos and other mineral dust in tissue cultuer cell lines. Brit J Exptl Pathol 1978; 59: 183-9, 15) Lappa KS, Fu MM, Epstein SS. Cytogenetic studies on chrysntilc asbestos. Environ Res 1975; 10: 165-173. 16) Oshimura M, Hesterberg TW, Tsatsvi T et al. Correlation of asbestos-induced cytogenetic effects with ceil transformation of Syrian humster embryo cells in culture. Cancer Res 1984; 44: 5017 -22. 17) Sincock AM. Seabright M. Induction of chromosome change in Chinese hamster cells by exposure to asbestos fibers. Nature 1975; 257: 56-8. 18) Sincock AM. Preliminary studies of the in vitro cellular effects of asbestos and fine glass dusts. In: Hiatt HH et al eds. The origins of human cancer. Vol. B. Cold Spring Harbor, 1977; 941 -54. 19) Babu KA, Lakkad BC, Nigam SK et al. In vitro eytoiogica! and cytogenetic effects of an Indian variety of chrysotile asbestos. 1980; Environ Res 21: 416 22. 20) Price-Jones MT, Gubbing G, Chamberlain M. The genetic effects of crocidolite asbestos; Comparison of chromosome abnormalities and sister chromatid exchanges. 1980; Mutat Res 79: 331-6. 21) Lecher JF, Tokiwa T, Yeager Jr H. et al. Asbestos-associated chromosomal changes in human mesothclial eetis. In i Beck EG and Bignon J, eds. In vitro effects of mineial dusts. Berlin Heidelberg: Springer-Verlag, 1985; 197-202. 22) Myojo T. A fibrous aerosol generator using a two-component fluidized bed. Ind Health 1983; 21: 79-89. 23) Davies R, Griffiths DM, Preece AW. Factors involved in the cytotoxicity of UICC amositc toward macrophages. Environ Health Perspective 1983; 51: 245-8. 24) Wagner, CJ. Health hazards of substitutes, Proceedings of symposium on asbestos Montreal: 1982; 244-267. 25) _Fukuda K, Koshi K, Kohyama N. Myojo T. Biological effects of asbestos and its substitutes- 56 K. KOSHI, et al. Fibrogenicity and carcinogenicity in mice and rats. Environ Res in Japan 1987; 93.1-18. (In Japanese) 26) Pott F, Bellmann B, Muhle H, Rodilsperger K, Rippe RM, Roller M, Rosenburch M. Intraperitoneal injection studies for the evaluation of the carcinogenicity of fibrous phyllosilieates. In: Bignon J, ed. NATO ASI. Series, VoL G21, Health related effects of phyilosilicates. Berlin Heidelberg: Springer-Veriag, 1990; 319-29. 27) Skaug V, Gytseth B. Hemolytic activity of five different calcium silicates. Environ Health Perspectives 1983 ; 51: 195-205. 28) Skaug V, Gylseth B. In vitro macrophage cytotixicity of five calcium silicates. Brit J Ind Health 1984; 41: 116-21. 29) Stanton MF, Layard M, Tegeris A, Milter E. May M, Kent E. Carcinogenicity of fibrous glass: pleural response in the rat in relation to fiber dimension. J Natl Cancer Inst 1977; 58: 587-603.