Document kDn65oBJ4amejo73Ro9gxOr0J

JENVWONHENTAL EE3EAJCCH 52, 164-177 (1990) On the Mechanism of Cell Internalization of Chrysotile Fibers: An Immunocytochemical and Ultrastructural Study W. Malorni, F. Iosi, M. Falchi, and G. Donelli Department of infrastructures, Istituto Superiore di Saniti, Viale Regina Elena, 299-00161 Rome, Italy Received October 6,1989 Human breast carcinoma cells (CG5) and human laryngeal carcinoma cells (HEp-2) were exposed to 10 and 50 |tg/ml of small (about 5 pm) chrysotile asbestos fibers. Morphological and ultrastructural changes were evaluated by means ofimmunocytochemistry and by scan- . ning and transmission electron microscopy. Our attention was focused on the mechanisms 1 of cell internalization and on transport of chrysotile fibers. The fibers appeared to penetrate l| \/[. the cell cytoplasm and to be translocated in proximity ofthe nucleus. Small chrysotile fibers could also be found inside the nucleus ofinterphase cells. Involvement of the main cyto- skeletal components, i.e., micmfilaments, intermediate filaments, and microUibles, in the . cytotoxicity of chrysotile fibers was also evaluated. Our findings suggest that after fiber \penetration, a rearrangement ofthe cytoskdetal apparatus occurs. It has also been observed .' that small fibers remain associated with the cytoskeletal framework, which can thus play a role in asbestos intracytoplasmic translocation in epithelial cells. Furthermore, after the cell , has completely recovered its morphology, fiber internalization ultimately seems to lead to the formation ofgiant multinuclealed cells. These data could be indicative of an interaction occurring between asbestos fibers and the normal mitotic process. The disturbance of the II cell cvtoskclcton and the dose nnrfw.r wynn may hr of .mporlqnr^. in hetweftft asbestosanri \ the well-known CardnOgCtUC effects of \ asbestos mineral fibers, o urn Academic rn. inc. INTRODUCTION The chrysotile fiber, the main component of mineral asbestos, is a hydrated magnesium silicate which forms a soft flexible and tensile fiber particularly in sistent to abrasion and heat (Allison et al,, 1975). In recent years, the mechanism of asbestos dust cytotoxicity and tumorigenicity has been widely investigated. General morphological characteristics such as size, shape, and diameter were considered to play an essential role in the fibrogenic arid carcinogenic potential of asbestos dusts (Kaw et al., 1982). '' The association between exposure to asbestos.and several human diseases, including lung and gastrointestinal malignancies, has been suggested (Neiiberger \Vet al., 1984). Mesotheliomas and carcinomas seem to be the-main forms of neo- PpplasiarretatetHo-the^xposure to breathableasbestosfibers (Environmental Health Criteria, 1986). Some of these in vivo studies have beeiiconfirmed in vitro using different cultured cell systems. In particular, the hemolytic capacities (Harington ' et al., 1971; Allison et al., 1975; Beck and Bignon, 1980; Hunt et al., 1981), the possible interaction with extracellular matrix and plasma membrane (Brody et al,, 1983), and the mechanisms of endocytosis and putative relationships with chro mosomes (Hesterberg et al.. 1985; Jaurand et al., 1986; Wang et al., 1987) have 0013-9351/90 $3.00 Copyright O 1990 by Academic Press. Inc. All rights of reproduction tn any form reserved. 164 PLAINTIFF'S EXHIBIT SA-49I /sotile Fibers: ruStural Study ; G. Donelh ' d, Viale Regina Elena, ; rcinoma cells (HEp-2) were ' testos fibers. Morphological icytochemistry and by scanfocused on the mechanisms fibers appeared to penetrate leus. Small chrysotile fibers olvement of the main cytotts, and microtubles, in the ngs suggest that after fiber rs. It has also been observed vork, which can thus play a . Furthermore, after the cell ultimately seems to lead to : indicative of an interaction ess. The disturbance of the sbestos fibers and the cell's own carcinogenic effects of 7. " - 7 * V i w } ;stos, is a hydrated :riSu fiber particularly re cent years, the mechanism, been widely investigated, shape, and diameter were d carcinogenic potential of, l several human diseases, ieen suggested (Neuberger. be the main forms of neoers (Environmental Health J:n confirmed in vitro using jf. lytic capacities (Harington ^ 80; Hunt et al., 1981), the !? 1a membrane (Brody et al., /e relationships with chro- 3 6; Wang et al., 1987) have CHRYSOTILE INTERNALIZATION PATTERN 165 been extensively studied in several cell models. Notwithstanding, complex mech anisms underlying fiber cytopathology are far from having been elucidated. A possible involvement rtf'free radicals causing lipid peroxidation and DNA break-, ages has been hypothesized as a result of toxic injury caused by intracellularW chrysotile fibers (Sincock and Seabright, 197S; Mossman et al., 1983b). Further-v more, cytoskeletal elements, which have been shown to play a pivotal role in cytoplasmic traffic, could also represent important carriers of small intracellular fibers (Ruttner et al., 1987). Mesothelial cell cytoskeleton has, in fact, been found to be associated with asbestos fibers, and a close contacTbetweenlthe nuclear membrane and chrysotile fibers has also been detected (Rflttner et al., 1987). However, the meclwmsms'Of chfysotile fiber internalization and transport seem to need further investigations. The purpose of this study was thus to try to better elucidate these peculiar mechanisms of cytotoxicity. MATERIALS AND METHODS Cell Cultures Human breast carcinoma cells (CG5), kindly provided by Dr. G. Sica (Univer sity Cattolica, Rome, Italy), and human laryngeal carcinoma cells (HEp-2), were cultured at 37C in Dulbecco's modified Eagle's medium (DMEM) supplemented with nonessential amino acids, L-glutamine, vitamins, 10% fetal calf serum plus 100 IU/ml penicillin and 100 pg/ml streptomycin. For light and scanning electron microscopy (SEM) both control and treated cells were grown on 13-mm-diam glass coverslips in separate wells and seeded at a density of approximate!jHO4 cells/ml. Cells- were subcultured in 25-cm2 Falcon plastic flasks at a density of approx imately 10s cells/ml for transmission electron microscopy (TEM). For the obser vation ofcytoskeleton by TEM, gold grids (200 mesh) were coated and adhered to -plastic flasks using 0.2% Formvar, in clorophorm. The flasks were seeded with cells at a density of 104 cells/ml UICC (Union International Contre le Cancer) standard reference samples of chrysotile fibers were suspended in culture medium (DMEM) dispersed by sonication and sterilized by autoclaving. Granulometric SEM studies showed that after 20 min of sonication more than 80% offibers were less than 5 pm in length. The resulting suspension was added to a final volume of 5 mi at a concentration of 50 (ig/ml. Immunocytochemical and electron microscopic observations were performed after 3 and 6 hr. For long-term treatment, chrysotile fibers were added to the cultures at a concentration of 10 pg/ml. After 24 hr, cells were washed with phosphate-buffered saline (pH 7.3) and incubated in fresh medium. Cells were then observed daily by phase contrast microscopy for at least a week. Cytochalasin B (CB) (Sigma) treatment was performed as follows: CB was dissolved in dimethyl sulfoxide (DMSO) and then added to the culture medium to obtain a final concentration of 1 \lM. After a 1-hr incubation with CB, chrysotile treatment was performed. Cultured cells treated with equal amounts of DMSO alone were con sidered as controls. Finally, immunocytochemical observations were performed. 166 MALOKNI ET AL. Fluorescence Microscopy CeBs grown on coverslips were fixed in 3.7% formaldehyde in phosphate ni. .. (pH 7.4) for 10 min at room temperature. After washing in the same buffer, forthe^ detection of cytoskeletal elements, ceUs were permeabQized with 0.5% Triton ^ X-100 (Sigma, T-6878) for 5 min at room temperature. For nuclei detection, the ceUs were incubated with 20 |tg/ml Hoechst solution (Sigma) at 37C for 10 min. Once through washing, coverslips were mounted with 50% glycerol-phosphate* buffered saline. For tubulin labeling, incubations with monoclonal antibody directed against a-tubulin (Amersham International) were carried out at 37C for 30 min. After washing, cells were incubated with a sheep anti-mouse IgG fluorescein-linked whole antibody (Amersham International). For actin detection, ceUs were stained with fluorescein-phalioidin (NBD-phaBacidin, Molecular Probes) at 37C for 30 min. For the detection of keratin filaments, ceUs were fixed with methanol for 5 min at room temperature and for 5 sec with acetone at -20C. CeUs were then incu bated with a polyclonal antibody directed against keratin (Ortho Diagnostic) and with a sheep anti-rabbit IgG fluorescein-linked whole antibody (Amersham Inter national). Electron Microscopy For transmission electron microscopy, ceUs grown on 25-cm2 flasks were fixed in 1.5% glutaraldehyde in 0.05 m cacodylate buffer, postfixed in 1% Os04, dehy drated in ascending grades of ethanols, and embedded in Agar 100 (Agar AIDS). Ultrathin sections were stained with uranyl acetate and lead citrate and were examined with a Zeiss EM 10 C electron microscope. Cytoskeleton Preparation for Electron Microscopy The procedure foflowed was an adaptation of the detergent extraction method previously described (Schliwa and Van Blerkom, 1981). AU of the steps were carried out at 37C. Before detergent extraction, the ceUs were rinsed briefly in PHEM (60 mM Pipes, 25 mM Hepes, 10 mM EGTA, and 2 mM MgCy, pH 6.9IS They were then dipped in PHEM buffer containing 0.75% Triton X-100 for 2 min. The detergent-permeabilized ceUs (cytoskeletons) were washed briefly in PHEM and then fixed for 10 min in a 10 mM sodium phosphate buffer containing 1% glutaraldehyde-0.2% tannic acid, pH 7.0. After fixation, salt was removed from the specimens by washing the cytoskeletons in distiUed water. The cytoskeleton preparation was then dehydrated by increasing concentrations of ethanol and critical-point-dried in C02. For transmission electron microscopy the specimens were freeze-dried in a Balzers BAF 300 freeze-etch unit at a stage temperature of -80C for 30 .min and then rotary-coated with platinum at a 45 angle and with carbon at a 90 angle. They were then examined with a Philips 430 electron microscope at 300 kV. For scanning electron microscopy the ceUs were goldcoated (<10-nm film thickness) by sputtering and examined with a Philips 515 scanning electron microscope. Both SEM and TEM microscopes are equipped f '^in phosphate buffer , ~ H Tsame buffer, for the' j jMsA with 0.5% Triton For nuclei detection, the! 'Sigma) at 37C for 10 min. h 50% glycerol-phosphate- antibody directed against . at 37C for 30 min. After use IgG fluorescein-linked etection, cells were stained ilar Probes) at 37C for 30 ed with methanol for 5 min J0C. Cells were then incuitin (Ortho Diagnostic) and antibody (Amersham Inter- 3 >n 25-cm2 flasks were fixed istfixed in 1% Os04, dehy- i l in Agar 100 (Agar AIDS), and lead citrate and were: t jjt extraction method 1 8l)V All of the steps were cells were rinsed briefly ind 2 mM MgCIJ, pH 6.915. 5% Triton X-100 for 2 min. e washed briefly in PHEM ihate buffer containing 1% on, salt was removed from? :d water. The cytoskeleton :entrations of ethanol and f microscopy the specimens . it at a stage temperature of ? urn at a 45 angle and with 'ith a Philips 430 electron' scopy the cells were goldamined with a Philips 515 microscopes are equipped CHRYSOTILE INTERNALIZATION PATTERN 167 with energy-despersive X-ray detection system (EDAX PV-9900). X-ray spectra were collected over 60 sec, using a 0-5-|im-diam circular probe. The accelerating voltage was 300 kV for TEM and 30 kV for SEM. RESULTS Chrysotile fiber internalization seems to occur by a sequence of events that can be analyzed by light microscopy and thin section TEM. In our experiments only small fibers, less than 5 pm in length, were considered. The longer ones (>10 pm) did not appear to be able to be actively internalized in the cell cytoplasm, but they seemed to easily induce cell degeneration and death. In fact, a quantitative eval uation performed by the trypan blue exclusion test showed that most of the nonviable cells (90%) included the longer fibers, whereas only viable cells were to be considered for this study. When cultured epithelial cells were analyzed by TEM after treatment with chrysotile fibers at a concentration of 10 pg/ml for 3 hr, small fibers could be found on the cell surface (Fig. la). However, after 6 hr of treatment, asbestos fibers were found in the cytoplasm of 70% of the cell sections examined. These fibers appeared to have no membrane envelope (Fig. lb) whereas cell plasma membrane and organelle integrity seemed to be maintained. Furthermore, when the nuclear region ultrastructure ofinterphase cells was examined, asbestos fibers were also detected. Fibers were found in the perinuclear region (Fig. lc), and smaller fibers ofabout 0.5 pm were observed inside the nuclear envelope (Fig. Id) in 20% of the thin sections examined. Moreover, fibers were also found in the cell nucleoli (Fig. le). Nevertheless, nuclear envelope ultrastructure seemed to be maintained. ___ To detect whether chrysotile fibers could interact with some cytoskeletal ele ments to be transported through epithelial cell cytoplasm, an immunocytochem- ical and a parallel electron microscopic analysis was performed. After treatment with detergents, as described under Materials and Methods, cytoplasmic and plasma membrane soluble proteins and phospholipids were extracted; only the cytoskeletal elements and the cell nucleus were detectable. Thus, the main cyto skeletal components were available for microscopic examination. In particular, actin microfilaments, microtubules, and keratin intermediate filaments were con sidered. Monolayer cultures of control epithelial cells revealed characteristic fea tures in actin (Fig. 2a), microtubule (Fig. 2b), and keratin (Fig. 2c> arrangement which appeared to be a well-represented network of fibers and filaments orga-, nized throughout the cell cytoplasm. When the same cells were observed after 3J hr of chrysotile treatment, remarkable changes in the arrangement of cytoskeletal] elements could be observed. In particular, wide breakages within the actin stress1 fiber network due to the penetration of fibers in the cells were detected (Fig. 3a)| Likewise, the microtubular apparatus and keratin intermediate filament network underwent alterations in their arrangements (Figs. 3b and 3d, respectively); such alterations were clearly related to the presence of asbestos fibers in the cyto plasm. In fact, when the same cells in Figs. 3b and 3d were observed by phased contrast microscopy, several asbestos fibers were detectable (Figs. 3c and 3e, respectively). On the other hand, when CB-treated cells were exposed to chryso- \ 168 MALOKNI EX AL. Fig. I. Transmission electron microscopy. Micrograph sequence showing the pattern of internal ization of small chrysolite fibers in CGS cultured cells. Fibeis were found to adhere to the cell surface (a), inside the cytoplasm devoid of a membrane envelope (b), in the perinuclear region close to the nuclear envelope (c), inside the nucleus (arrow) (d), and, finally,, small fibers can be detected in the nuclear matrix (arrow) (e). (a, *55,200; b and c, x 14,720; d, x 18.400; e, x23,000)t . tile fibers for 3 hr, actin depolimerization was observed and no fibers could be detected inside the cell cytoplasm (Fig. 5). However, the cytoskeletal element network appeared to have fully recovered after 6 hr treatment (Figs. 4a and 4b). Furthermore, when recovery experiments were carried out in fiber-free culture If ' medium for aHveek after the-beekming of exposure to asbestos, a remarkable jlj increase in the number of giant multinucleated cells could be observedTFtgs. 6, I 7a)7Thesepolynudeated celhTdisplayed large^rcasoF^oplasnTwitlfnd nuclei, and some regions filled with several nuclei (Fig. 7a). Small nuclei, probably de riving from nuclear segmentation, were also observed (Fig. 6). When analyzed by ce showing the pattern of internaP-j : found to adhere to the cell surface! the perinuclear region close to thel small fibers can be detected in the f ;,400; e, x23,000). rved and no fibers could be: er, the cytoskeletal element, treatment (Figs. 4a and 4b). ried out in fiber-free culture, e to asbestos, a remarkable' could be observed (Figs. 6, of cytoplasm with no nuclei,' . Small nuclei, probably de-f i (Fig. 6). When analyzed by Fig. 2. Fluorescence microscopy. CGS control cells. Staining (a) or immunostaining (b, c) showing the normal arrangement of actin fibers (a), microtubules (b), and cytoketatin filaments (c) in flat adhering cells (a. b, and c, x860). Fig. 3. Fluorescence microscopy. CGS cultured cells treated for 3 hr with chrysotile fibers. Actin staining (a) and a-iubulin immunostaining (b) of the same cells shown in (c) by brightfield. Morpho logical modifications in these cytoskeletal components, characterized by rearrangements and break ages in the asbestos-inserting regions is well evidenced when long fibers are considered (arrows). A cytokeralin rearrangement was also detectable in chrysotile-treated CGS cells (d, cytokeratin immu nostaining; e, the same cells observed by brightfield) (a-e, x 1720). I i Fig. 4. Light microscopy. Brightfield (a) and actin staining by fluorescein-phalloidin (b) of CGS cells. A complete recovery of the actin network is observed after 6 hr treatment with chrysotile fibers (X1680). Fig. 5. F-actin fluorescence microscopy of CB-treated CGS cells after asbestos exposure. Chrysotile fibers were not detectable in the cytoplasm (x 1680). . Fig. 6. CGS cell stained with Hoechst solution showing several clustered nuclei (x 1680). fr j>( ' phase contrast microscopy, asbestos fibers were visible inside the giant cells, which were sometimes as much as 50 times larger than control cells. Furthermore, cytoskeletal elements appeared normal in shape, having recovered their arrange ment (Figs. 7a-7d). After a week, the percentage of giant multinucleated cells was 30%, compared to approximately 5% in control cells. To better elucidate the possible mechanisms of interaction between chrysotile Fibers and the cytoskeletal apparatus in epithelial cells, cytoskeletal extracts, prepared as described above, were studied by transmission and scanning electron microscopy. Observations performed by TEM allowed us to visualize the cyto- CHRYSOTILE INTERNALIZATION PATTERN 171 fluorescein-phalloidin (b) of CGS hr treatment with chrysotilc fibers 5 cells after asbestos exposure. clustered nuclei (xl680). sible inside the giant cells, 1 control cells. Furthermore, ing recovered their arrangeant multinucleated cells was f eraction between chrysotile sells, cytoskeletal extracts, ission and scanning electron ed us to visualize the cyto- Fig. 7. The appearance of giant polynucleated cells is clearly evident in CGS cells treated for 6 hr with chrysotile fibers, when observed by brightfield after a week in fiber-free medium (a). Parallel fluorescence microscopy observations showed the normal arrangement of actin filaments (b), micro tubules (c), and cytokeratin intermediate filaments (d) in multinucleated flattened cells. A remarkable positivity for microtubules and cytokeratin filaments is visible in the perinuclear region (a, x 1260; b, c. and d, x 1680). skeletal network arrangement after treatment with chrysotile fibers, as well as to detect possible signs of interaction between small fibers and cytoskeletal elements (Fig. 8). In fact, after Triton X-100 extraction, a morphologic relationship between fibers and the cytoskeleton appeared to be maintained. The observation that 172 MALORNI ET AL. treatment with chrysotile fibers. Cytoskeletal elements are visualized. A small fiber, confirmed to be ofchrysatile type by the diffraction pattern (inset), appears in close relationship with thin cytoskeletal elements (arrows) (x 19,750). internalized asbestos fibers are not free in the cytoplasm and are not removed from the cell by detergent extraction can shed some light on an intimate connec tion between fibers and the cytoskeleton. Energy-dispersive X-ray microanalysis and the diffraction pattern (Fig. S, inset) confirmed that the electron-dense bodies were asbestos fibers of the chrysotile type in close relationship with cytoskeletal extracts. Parallel experiments performed using SEM enabled us to better visual ize, along with the simple interlacing of fibers, a direct interaction between as bestos and the cytoskeleton (Figs. 9a, 9b). In fact, high magnification (Fig. 9c) reveals a close relationship between a chrysotile fiber and some cytoskeletal elements. In consideration of their diameter they could probably belong to the microfilament system. . DISCUSSION The present results show that the small fibrogenic material added to epithelial CHRYSOTILE INTERNALIZATION PATTERN 173 :^4 * . **1 \ V.-v, 1 vi 'vte 3sr'fc> - * fe \ s observed after 6 hr of i i / tall fiber, confirmed to be rela'uonship with thin cytoskeletal. lasm and are not removed i ght on an intimate connec- i ersive X-ray microanalysis\ it the electron-dense bodies, ationship with cytoskeletal\ enabled us to better visual;ct interaction between as- ' igh magnification (Fig. 9c) _ 3er and some cytoskeletal " aid probably belong to the j material added to epithelial i- Fig. 9. Scanning electron microscopy. Triton-permeated CGS cells treated with chrysotile fibers for 6 hr. Cytoskeletal organization appears well preserved. Relationships between asbestos fibeis and cytoskeletal elements can be observed (a, b). At higher magnification (c) interaction seems to exist between some cytoskeletal elements and the internalized fiber, (a and c) Reverse contrast, (b) En largement of the outlined area shown in (a). Arrows: chrysotile fiber; arrowhead: cytoskeletal ele ments. (a. x800; b, x3500; c, x 14.000). ' 174 MALORNI ET AL. cell culture media appears to be associated with cytoskeletal elements. Internal-^ ized and transported fibers were detected in the nuclear envelope of intcrphasell) cells. Cell culture studies have been extensively used to investigate toxic and long term effects of asbestifonn minerals (Hext and Richards, 1976; Hext et al., 1977; Chamberlain et al., 1979; Wang et al., 1987). Results present considerable differ ences which depend upon fibrous material, cell type, fiber processing, and the f extent and duration of exposure. However, some general rules can be traced. Chrysotile stimulates collagen production, proteoglycans, and ah aggregation of pericellular fibronectin (Vartio et al., 1986), which is the prerequisite offibrosis in vivo (Rahman et al., 1975; Richards and Morris, 1979). Chrysotile asbestos also seems able to bind sialic acid groups to membranes (Brody et al., 1983), and the interaction between the positively charged fibers and cell membranes could be related to cell damage (Allison et al., 1975; Mossman et al., 1982, 1983a). In accordance with these observations, chrysotile fibers seem to enter the epithelial cell cytoplasm, damaging both the plasma membrane and the underlying cyto skeletal network. Nevertheless, our data seem to indicate that both plasma mem brane and cytoskeletal elements can undergo rapid rearrangement and recovery. The involvement of cytoskeletal elements in intracellular transport of asbestos fibers has recently been hypothesized (Riittner et al., 1987). In fact, as has been previously described (Richards et al., 1977; Joseph et al., 1983), ingested fibers . are not free in the cytoplasm, but seem to interact with some cytoskeletal ele ments. The results reported here not only confirm previous observations (Riittner - etal., 1987), but also suggest that the cytoskeletal apparatus plays a structural role in the interaction of chrysotOe fibers with cell cytoplasm. However, a functional role of cytoskeletal elements cannot be ruled out. Phagocytosed asbestos fibers that are not removed from the cell by detergent extraction are revealed by scanning and transmission electron microscopic ob servations; they are shown to be in close contact with some cytoskeletal compo nents. This could therefore shed some light on the close relationship between fibers and the cytoskeleton. Previous studies have shown that the uptake offibers causes an increase in the filamentous (polymerized) to nonfilamentous actin ratio in epithelial cells (Brody et al., 1983), and that individual actin filaments are probably attached to the fiber surfaces (Brody et al., 1986). In fact, after treatment I with cytochalasin B, a specific actin-depolimerizing agent, chrysotile fibers in Ij cytoskeletal extracts were not detectable. This could suggest an important role of the relationships between actin filaments and chrysotile asbestos fibers next to a passive "aspecific" internalization. Several in vitro studies on cell growth kinetics were also carried out in epithe lioid cell lines after treatment with asbestos fibers (Neugut et al., 1978; Reiss et j al., 1980; Kaw et al., 1982; Tilkes and Beck, 1983; Kenne et at., 19g6). Growth inhibition and a more prolonged mitosis, as well as a high sensitivity io chrysotile i fibers of cells undergoing mitosis, were suggested (Neugut et al., 1978). An in( crease in bi- or polinucleated cells was also observed by using another component of asbestos fibers, i.e., crocidolite, in treated Chinese hamster ovary cells (CHO) (Kenne er al., 1986). In fact, it is possible that many cells which try to undergo skeletal elements. .Inter lear-.envelope of interpli * /^tigate toxic and long? rdsf 1976; Hext et at., 1977; present considerable differ- fiber processing, and the eneral rules can be traced.' :ans, and an aggregation < he prerequisite offibrosis in' 9). Chrysotile asbestos also; Brody et at., 1983), and the d cell membranes could be tan et at., 1982, 1983a). In seem to enter the epithelial. ie and the underlying cytocate that both plasma mem:arrangement and recovery., ellular transport of asbestos , 1987). In fact, as has been ] it at., 1983), ingested fibers 3 with some cytoskeletal de vious observations (Ruttner aratus plays a structural role asm. However, a functional. 1 from the cell by detergent >n dectron microscopic ob ' cytoskeletal compo- . relationship between'i jwh mat the uptake of fibers] o nonfilamentous actin ratiol dividual.actin filaments are! 986). In fact, after treatments ; agent, chrysotile fibers ini suggest an important role ofj tile asbestos fibers next to; -e also carried out in epithe-Y Seugut et at., 1978; Reiss et 't ECenne et at., 1986). Growth high sensitivity to chrysotileNeugut et at., 1978). An in-; by using another component. ; hamster ovary cells (CHO) / cells which try to undergo? CHRYSOTILE INTERNALIZATION PATTERN 175 mitosis Ml to complete cytoplasmic division; as a consequence, they end up forming giant muldnudeated cells with several micronuclei which could derive from a chromosome fragment (Grate and Revell, 1972). These giant cells undergo DNA replication with no cell division, remaining in the culture for long periods of time (up to 1 month) and growing in size before they die. Moreover, nuclear division may also occur as a result of the so-called amitotic process, previously described by other authors (Clegg, 1963; De Martino et at., 1985). In fact, the higher-order structure of the interphase eukariotic cell nucleus is still a matter of conjecture. The increased nuclear foldings, particularly along the axis of the nu cleus, together with the increased eccentricity and reduced size of the nucleolus, point to amitotic division (Clegg, 1963; Herv&s et aL, 1985). Previous studies also hypothesized that changes in the contractile microfilaments and microtubules may result in a constriction ofthe nucleus which can lead to nuclear segmentation and fragmentation (Ghadially, 1982; Neftel et at., 1983). In addition, several studies were performed on the direct interactions between chrysotile fibers and chromosomes. In particular, metaphase and anaphase ab normalities were detected in different cultured cells (Hesterberg and Barret, 1985). Bridges, abnormalities, and misaggregations of chromosomes resulting in aneuploidy could also be induced by abnormalities in nuclear division (Herv&s et at., 1985). On the other hand, a direct interaction between chrysotile fibers and the cytoskeletal apparatus, mainly microtubules and microfilaments, could also contribute to chromosomal misaggregaiion via an impairment of the mitotic spin dle. In fact, the sensitivity of epithelial cells in vitro has been related to the plasticity of its cytoskeleton resulting in chromosomal damage and aneuploidy (Haugen and Hams, 1982). In conclusion, our resultrsupport the hypothesis ofa direct, active involvement of cytoskeletal elements in chrysotile fiber translocation in the cytoplasm of cul tured epithelial cells. These findings, together with those on the "stored" fibers detected in interphase nuclei, might suggest an active contribution of epithelial cells to their own pathological or carcinogenic potential. REFERENCE Allison, A. C., Harington, J. S., and Badami, D. V. (1975). Mineral fibres: Chemical physiochemical and biological properties. Adv. Pharmacol. Chemotker. 12, 291-402. Beck. E. G., and Bignon, J. (1980). In vitro effects of mineral dusts. NATO ASI Ser.. Brody, A. R., Gcrwyn, G., and HOI, L. H. (1983). Interactions of chrysotile and crocidplite asbestos with red blood cell membranes: Chrysotile binds to sialic acid. Lab. Invest. 49, 4S&-475. Brody, A. R., Hill, L. H., Barrett, J. C., and Adler, K. B. (1986). Intracellular translocation of inor ganic particles. In "The Cytoskeleton. a Target for Toxic Agents" (T. W. Clarkson, P. R. Sager, and T. L. M. Syversen, Eds.), pp. 221-227. Plenum, New York. Chamberlain, M., Brown, R. C., Davies, R., and Griffits, D. M. (1979). In vitro prediction of the pathogenicity of mineral dusts. Brit. J. Exp. Pathol. 60, 320-327. Clegg. E. J. (1963). Studies on artificial cryptorchidism: Morphological and quantitative changes in the Sertoli cells of the rat testis. J. Endocrinol. 26, 567-574. De Martino, C.,- Marcante, M. L., Malomi, W., Citro, G.. Floridi, A., Scorza-Barcellona, P.. and Capanna;' E. (1985). Morphological and karyological studies on in vitro Sertoli cells of adult Crab-EatingMacaques (Macaca fascictdaris). Amer. J. Printatol. 9, 173-187. Environmental Health Criteria 53 (1986). "Asbestos and Other Natural Mineral Fibres." World Health Organization, Geneva. vt i V\ t 176 MALORNI ET AL. Ghadially, F. N. (1982). Radial segmentation ofthe nucleus. In "Ultrastructuial Pathology ofthe Cell and Matrix'* (F. N. Ghadially, Ed.), 2nd ed., p. 8. Butterworths, London. Grate, S. J-, and Revell, S. H. (1972). Condathm of chromosome damage and colony-forming ability in Syrian Hamster cells in culture irradiated in Gt. Cun. Top. Radial. Res. Q. 7, 303-309. Harington, J. S., Miller, K., and Macnab, G. (1971). Hemolysis by asbestos. Environ. Res. 4,95-98. Haugen, A., and Harris, C. (1982). Asbestos carcinogenesis: Asbestos interactions and epithelial lesions in cultured human tracheobronchial tissues and cells. Recent Results Cancer Res. 82, 32-41. HervAs, J. P., Lbpez-SAez, J. F., and Gimtnez-Martln, G. (1985). Proliferation of multinucleate cells with unbalanced nuclei. Cell Biol. Int. Rep. 9,91-101. Hesterbcrg, T. W., and Barret, J. C. (1985). Induction by asbestos fibers of anaphase abnormalities: Mechanism for aneupkridy induction and possibly carcinogenesis. Carcinogenesis 6,473-475. Hext,P. M., and Richards, R. I. (1976). Biochemical effects ofasbestiform minerals on lung fibroblast cultures. Brit. J. Exp. Pathol. 57,281-286. Hext, P. M., Hunt, J., Dogdson, K. S., and Richards, R. J. (1977). The effects oflong-term exposure oflung fibroblast strains to chrysotile asbestos. Brit. J. Exp. Pathol. 58, 160-167. Hunt, J., Pooley, F. D., and Richards, R. J. (1981). Biological reactivity of calcium silicate compos ites: In vitro studies. Environ. Res. 26,51-55. . Jaurand, M. C., Kheuang, L., Magne, L,, and Bignon, J. (1986). Chromosomal changes induced by chrysotile fibres or benzo-3,4-pyrene in rat pleural mesotheiial cells. Mutat. Res. 169, 141-148. Joseph, L. B., Stephens, R. E., Ottolenghi, A. C., Lipetz, P. D., and Newman, H. A. I. (1983). Morphological transformation in vitro ofnormal human fibroblasts by chrysotile. Environ. Health Perspect. 51,17-22. Kaw, J. L., TOkes, F., and Beck, E. G. (1982). Reaction ofcells cultured in vitro to different asbestos dusts of equal surface area but different fibre length. Brit. J. Exp. Pathol. 63, 109-115. Kenne, K., Ljungquist, S., and Ringertz, N. R. (1986). Effects of asbestos fibers on cell division, cell survival, and formation of thioguanine-resistant mutants in Chinese Hamster Ovary cells. Envi ron. Res. 39,448-464. . Mossman, B. T., Adler, K. B., Jean, L. J., and Craighead, J. E. (1982). Mechanisms of hypersecre tion rodent tracheal explants after exposure to chrysotile asbestos. Chest 815, 235-255. Mossman, B. T., Jean, L. J., and Landesman, J. M. (1983a). Studies using lectins to determine min eral interactions with cellular membranes. Environ. Health Perspect. 51.23-25. Mossman, B. T., Eastman, A., Landesman, J. M., and Bresnick. E. (1983b). Effects ofcrocidolite and chrysotile asbestos on cellular uptake, metabolism, and DNA after exposure of Hamster tracheal epithelial cells to benzo(a)pyrene. Environ. Health Perspect. 51, 331-336. Neftel, K. A., Stahel.'R., Mailer, O. M., Morel], A., and Arrenbrecht, S. (1983). Radial segmentation of nuclei (Reider cells): A morphological marker of T cell neoplasms? Acta Haematol. 70, 213 223. Neuberger, M., Kundi, M,, and Friedl, H. P. (1984). Environmental asbestos exposure and cancer mortality. Arch. Environ. Health 39, 261-265. Neugut, A. I., Eisenberg, D.. Silverstein, M., Pulkrabek, P., and Weinstein, J. B. (1978). Effects of asbestos on epithelioid cell lines. Environ. Res. 17, 256-265. ' Rahman, O., Beg, M. W,, Viswanathan, P. N., and Zaidi, S.H._(1975). Biochemical changes in the lungs of rats caused by asbestos dust. Scand. J. Work Environ. Health 1, 50-53. Reiss, B., Solomon, S., Weisbuiger, J. H., and Williams. G. M. (1980). Comparative toxicities of different forms of asbestos in a cell culture assay. Environ. Res. 22, 109-129. Richards. R. J., Hext, P. M.. Desai, R., Tetley, T., Hunt, J., Presley. R., and Dogdson. K. S. (1977). Chrysotile asbestos: Biological reaction potential. In "Inhaled Particulates" (J. JValton and M. C. McGovern, Eds.), pp. 477-493. Pergamon. New York. ' Richards. R. J., and Morris. T. C. (1979). Collagen and mucopolysaccharide production in growing lung fibroblasts exposed to chrysotile asbestos. Life Sci. 12, 441-451. Ruttner. J. R.. Lang, A. B-. Gut. D. R., and Wydler, M. U. (1987). Morphological aspects of inter action between asbestos fibers and human mesotheiial cell cytoskeleton. Exp. Cell Biol. 55, 285-294. (Jltrastnictural Pathology ofthe < rths, London. f and colony-fanning l Res. Q. 7, 303-309. ^. Jgtos. Environ. Res. 4,9i \sbcstos interactions and epitheli Is. Recent Results Cancer Res. i. Proliferation of multinucleate < os fibers of anaphase abnonnalitie nesis. Carcinogenesis 6,473-475. bcstifonn minerals on lung fibrobh ). The effects of long-term exposu Pathol. 58, 160-167. activity of calcium silicate compos-1 . Chromosomal changes induced by] lal cells. Mutat. Res. 169, 141-148. D., and Newman, H. A. I. (1983). ilasts by chrysotile. Environ. Heah :ultured in vitro to different asbestos] Exp. Pathol. 63, 109-115. asbestos fibers on cell division, cell] Chinese Hamster Ovary cells. Env (1982). Mechanisms of hypersecre-] asbestos. Chest 815, 235-255. '.dies using lectins to determine min-] *erspect. 51, 23-25F '*983b). Effects ofcrocidolite and] Exposure of Hamster tracheal] }j-336. (1983). Radial segmentation] :oplasms? Acta Haematol. 70, 213 ental asbestos exposure and cancer] d Weinstein. J. B. (1978). Effects < (1975). Biochemical changes in I on. Health 1, 50-53. i. (1980). Comparative toxicities i Rer. 22, 109-129. sley. R.. and Dogdson, K. S. (1977).] i Particulates" (J. Walton and M. C.. ilysaccharide production in growing] 441-451. 87). Morphological aspects of inter-] 1 cytoskeleton. Exp. Cell Biol. 55, f CHRYSOTHJE INTERNALIZATION PATTERN 177 Schliwa, M., and Van Blerkom, J. (1961). Structural interaction at cytoskeletal components. J. Cell Biol. 90,222-235. Sincocic, A. M., and Seabright, M. (1975). Inductionofchromosome changes in Chinese Hamstercells by exposure to asbestos fibers. Nature (London) 257,56-58. Tilkes. F., and Beck. E. G. (1983). Influence of well-defined mineral fibres on proliferating ceils. Environ. Health Perspect. 51,275-279. Vartio. T., Virtanen, I., and Lehto, V. P. (1986). Aggregation of fibronectin by asbestos fibers in the pericellular matrix of cultured human fibroblasts. Acta Pathol. Microbiol. Immunol. Scand. Sect. A 94, 169-175. Wang, M. D., Jaurand, M. C., Magne, L., Kheuang, L-, Pinchon, M. C., and Bignon. M. D. (1987). The interactions between asbestos fibers and metaphase chromosomes of rat pleural mesothelial cells in culture. A scanning and transmission electron microscopic study. Amer J Pathol. 126, 343-349. -