Document gbYEwneqMLrZ0ymL77JLjq7OL

JOURNAL OF THE AMERICAN COLLEGE OF TOXICOLOGY Volume 9. Number 5, 1990 Mary Atm [Jebert. Inc.. Publishers Mechanisms of Asbestos Carcinogenicity PETER VOYTEK. MIRIAM ANVER. TODD THORSLUND, JILL CONLEY, and ELIZABETH ANDERSON ABSTRACT We present a brief review of different potential mechanisms at the molecular and cellular levels that may be involved in asbestos-induced carcinogenicity. The usefulness of considering such mechanisms in developing appropriate biologically based models to estimate carcinogenic risk at environmental levels of asbestos fibers is discussed. INTRODUCTION sbestos is a general commercial name for a group of mineral fibers of hydrated silicates with a variety of Aphysical characteristics and metal compositions.'n Occupational exposure to several types of asbestos fibers has been shown to have a causal relationship with the development of lung cancer and malignant pleural mesotheliomas.'2-31 However, animal studies have shown that morphological difference and size can result in different carcinogenic potencies.'4-61 Initially, workplace exposure to asbestos fibers was the primary concern for human health hazards; however, more recently, attention has turned to potential risks to the general public exposed to asbestos used in building materials in the construction ofschools and public and commercial buildings. There is a likely potential to release high fiber concentrations into the environment during the removal of asbestos from these buildings in order to reduce long-term asbestos exposures. Costs for removal have been estimated to be greater than 53 billion doliars." 1 Therefore, it is important to have the most accurate estimates ofcarcinogenic risks in making decisions whether removal is necessary and would actually result in an adequate reduction in risk. Regulatory agencies are obligated by law to protect the public from levels of carcinogens that would pose an unreasonable risk to people. This often involves using epidemiological studies in which cancers have been observed to be statistically elevated in an exposed population in comparison to an unexposed matched population and extrapolating to lower levels where the "acceptable" risk ofcancer would be in the cider of l out of 10s or 106. Although most epidemiological studies are inadequate for precise quantitative extrapolations, without better alternatives they are used to set environmental exposure limits for carcinogenic agents. In conducting such assessments, however, no information is used concerning the mechanisms by which carcinogens cause normal cells to become carcinogenic. Conservative, and possibly inaccurate, assumptions are made by assuming that the induction of cancer can result from one interaction or "hit" with DNA, that there is a linear relationship between external exposure and the number of interactions with DNA. and that there is no "safe" exposure level where the probability of causing cancer is zero. In 1986. the EPA used epidemiological studies and these assumptions to estimate that a lifetime exposure to 0.0001 fibers of asbestos per ml of air would result in 2.4 asbestos-associated deaths from lung cancer and mesothelioma per 100.000 people exposed.'*1 However, the accuracy of using such linear models to extrapolate to lower levels ofasbestos exposures,has not been verified in epidemiology studies, nor has there been any biological data that would indicate that asbestos-induced carcinogenicity is a simple "one-hit" mechanism. In the past two decades, many scientific studies revealed the complex nature of interactions of asbestos with macromolecules and cellular responses to asbestds exposures. These interactions can result in direct and/or indirect Clcmcm International Corporation. Fairfax, Virginia. 541 ASBESTOS CARCINOGENICITY lung or in human Ivmphoblastoid cell lines exposed to either crocidolite or chrysolite asbestos fibers with mean panicle lengths of 1.1 to 3.4 pm and mean diameters between 0.15 and 0.47 pm. Fomace'231 was unable to detect single-strand DNA breaks in human fibroblast cells that had been treated with amosite asbestos fibers at doses of low cellular toxicity. Mossman et al.1'41 have also shown that crocidolite and chrysolite asbestos fibers at nontoxic levels did not cause single-strand breaks in the DNA of hamster tracheal epithelial cells. Libbus et al.<1Sl demonstrated that crocidolite asbestos fibers of average length and diameter of 14.8 pm and 0.9 pm. respectively, caused single-strand DNA breaks in cultured rat embryo cells. These investigators used a nick translation assay rather than alkaline elution to measure breaks. Electron microscopic examinations of the treated cells indicated that strand breaks occuned both in cells that visibly contained fibers and in cells where no fibers were detected. The DNA damage appeared to be dose related but nonlinear. These studies show that asbestos fibers can enter cells, bind to molecular components, and cause chromosomal damage and numerical chromosomal aberrations in certain cell types by direct interaction or indirectly through some intermediate that can result in mutations. Asbestos apparently causes severe chromosomal damage rather than alterations in just single genes since asbestos fibers have not been shown to cause gene mutations in mammalian ceil*26-271 or in bacterial cell*141 test systems or stimulate unscheduled DNA synthesis in rat hepatocytes.1291 However. Yang et al.1301 have demonstrated that xeroderma pigmentosum fibroblast cell lines, deficient in excision repair of certain types of chemical or ultraviolet light-induced DNA damage, are more sensitive to chrvsotile. amosite. and crocidolite asbestos fibers than normal human fibroblasts. These findings suggest that some types of asbestos-induced DNA damage may involve small lesions that are reparable. Asbestos fibers famosite. anthophyiiite. crocidolite. Rhodesian A chrysotile. and Canadian B chrysolite) have been shown to facilitate the transfection of viral DNA into liver epithelioid cells from chimpanzee livers, rhesus monkey kidneys cells, human carcinoma, and NIH 3T3 mouse fibroblast cells.1311 However, the concentration of asbestos fibers needed to cause transfections was high (3-10 mg/ml). A linear dose-response relationship was observed at higher doses, but at 0.1 mg/ml. little or no transfection was seen. Appel et al.*321 also were able to introduce plasmid DNA vectors into cos-7 monkey cells using Canadian chrysolite sample B asbestos fibers with an average diameter of 7 pm and lengths in the range of 130-3 pm as the transfecting agent. Therefore, under experimental conditions using high concentrations of asbestos fibers, nucleic acids can be inserted into the genome of cells. Ke et a!.<33> successfully transfected normal human mesothelial cells with fragmented DNA from a transformed human mesothelial cell line and obtained transformed clones that were morphologically indistinguish able from the original transformed line. Hence, the inserted DNA can alterthe expression ofcellular oncogenes and antioncogenes. depending on the insertion sites, or the transfected DNA could itselfpossess oncogenes that may be translated after insertion into genomic DNA. EVIDENCE FOR INDIRECT GENOTOXICITY There are two potential mechanisms where asbestos fibers can cause the formation of reactive oxygen species. These reactive oxygen species, in turn, can act as second messengers in causing genotoxicity. One mechanism involves the transfer of electrons from asbestos fibers (chrysotile with mean fiber length of 6 pm) to cellular molecules which, in turn, can interact with DNA.*341 The second mechanism involves an asbestos-induced inflammatory response that causes the accumulation of macrophages and polymorphonuclear leukocytes, which can release active oxygen species such as superoxide free radical and hydrogen peroxide, which then interact with DNA to cause mutations.133-3*1 Wong et al.1391 suggested that the ferrous iron in certain asbestos fibers'Tnay reduce oxygen to form the superoxide free radical. The superoxide anion may then result in the formation of other reactive oxygen species such as hydrogen peroxide and the hydroxyl free radical. Eberhardt et a!.<401 and Weitzman and Graceffa1*11 have shown that asbestos fibers can catalyze the formation of both the superoxide radical and hydroxyl radical from hydrogen peroxide. Mossman et al.*3*1 reported that asbestosis in rats caused by inhaling crocidolite asbestos fibers could be inhibited if the animals also were given catalase, which converts hydrogen peroxide into water. Cytotoxicity of hamster tracheal epithelial cells induced by crocidolite and chrysotile asbestos fibers of lengths varying from > 10 pm to < 2 pm can be inhibited by superoxide dismutase. which converts superoxide radical to H2Oi. and by hydroxyl radical scavengers.1421 Catalase was ineffective in protecting the cells from asbestos 543 R&S147973 ASBESTOS CARCINOGENICITY capable of initiating a cancer response. Like asbestos fibers, phorbol esters stimulate the production of reactive oxygen species and induce DNA strand breaks.132"'3' and as with asbestos, the toxic responses, as well as the promotional activity of TPA. can be inhibited by antioxidants.1341 Phorbol esters have been studied extensively with respect to their carcinogenic promotion activity*32-34-34' and they have been shown to cause changes in cell membranes, generate reactive oxygen species, increase lipid peroxidation, elevate certain enzyme activities, enhance phospholipid synthesis, alter methylation of macromolecules, and cause chromosomal damage. Promot ers may act initially through interactions with cell membranes and stimulate lipid peroxidation, which in turn, through various intermittent steps, produce free radical species that reach and damage genetic material.'37-381 Yano'391 has demonstrated that chrysotile and crocidolite asbestos fibers can induce the formation of malondialdehyde, a product of free radical-induced lipid peroxidation in human polymorphonuclear neutrophil cells, guinea pig peritoneal macrophages, and alveolar lavage cells. Weinstein'341 has reviewed the current status of the effects of TPA on protein kinase C (PKC). PKC plays a key role in cell signal transduction involving activation of oncogenes, cellular growth, and tumor promotion. PKC is the primary receptor for TPA and the phorbol ester binds at an allosteric site oh the enzyme, enhancing its ability to phosphorylate protein substrate(s). Cox et al.'401 have reported that PKC phosphorylates a NADPH oxidase on the plasma membrane that catalyzes the reduction of oxygen to the superoxide anion. PKC stimulation also increases the levels of ornithine decarboxylase (ODC1. which is the rate-limiting enzyme in the biosynthesis of polyamines that are necessary for the initiation ofcell division. Marsh and Mossman'411 have shown that chrysotile and crocidolite asbestos fibers also induce ODC activity in cultured hamster tracheal epithelial cells, but that ODC activity is significantly reduced when calcium entry antagonists (verapamil or nifedipine) are added to the cultures. They found that the longer fibers were more effective in stimulating enzymatic activity. Furthermore, palmitoyi carnitine and 1 -{5-isoquinolinylsulfony! )-2-methylpiperazine. inhibitors of PKC. also were effective in blocking ODC activity that was stimulated by asbestos fibers. It seems unlikely that asbestos fibers are promoting cells, creating active oxygen species, and stimulating enzymes like ODC in a similar manner to TPA. TPA can bind directly to PKC and activate the enzyme. However, asbestos fibers are likely to interact at the plasma membrane, to cause an influx ofcalcium into the cell, which activates PKC and may stimulate membrane phospholipases.'421 The phospholipases hydrolyze membrane polyphosphatidylinositols to form diacylglycerol and inositol 1.4.5triphosphate. Diacylglycerol is an activator of PKC631 and inositol 1.4.5-triphosphate causes release of intracellular stores ofcalcium.'641 Both calcium and diacylglycerol bind at the regulatory site of PKC and induce a conformational change that enhances the catalytic activity ofthe enzyme. Roney and Holian,42> have examined the effects of the PKC inhibitors tfluphenazine or staurosporinei on blocking the production of superoxide anion production by chrysotile asbestos fibers and the promoter, phorbol 12.13-dibutvrate. in guinea pig alveolar macrophages. These PKC inhibitors all reduced the amount of superoxide anion production. The authors postulated that asbestos is most likely stimulating membrane phospholipase C to produce diacylglycerol and inositol 1.4.5-niphosphate. which causes an increase in cell calcium levels, which in turn increase the activity of PKC. PKC then is responsible for inducing membrane NADPH oxidase to produce superoxide anion that leads to other reactive oxygen species and causes activation of oncogenes that affect cell growth and differentiation. However, crocidolite. anthophyllite. and amosite did not stimulate (he production of superoxide anion as effectively as chrysotile fibers. Gabrielson et al.`43) exposed cultured human lung mesothelial cells to amosite asbestos fibers and were v.ar.; to detect any generation of free radicals, nor did free radical scavengers (glutathione. n-acetyl-c; D-alpha-tocopherol, or superoxide dismutase) alter the cytotoxic effects of asbestos. These findings are in contrast to other studies described above. The differences may be due in part to different cell types and experimental conditions as well as to differences in asbestos fiber characteristics. DISCUSSION Figure 1 summarizes six potential mechanisms for the induction and promotion of cancer by asbestos fibers. Mechanism I involves asbestos fibers penetrating ihe target cell, directly interacting with DNA. and causing chromosomal aberrations (mutations). Depending on the severity and location of the mutation, some cells will not survive, other cells may survive and continue to function normally. while others may have abnormal expression of 545 R&S147974 ASBESTOS CARCINOGENICITY DNA that would constitute the initiation step in the carcinogenic process. Asbestos fibers readily interact with proteins both outside and inside cells. These types of interactions actually may compete with genomic DNA for the asbestos fibers and consequently may serve to protect the cell from asbestos-induced mutations. Because of the many macromolecules available to bind to asbestos fibers, it seems that at low fiber concentrations few. if any, fibers could penetrate the nucleus and interact with DNA. Mechanism II illustrates the capability of asbestos fibers to interact with DNA fragments and to act as a carrier for the fragments into the cell for insertion into genomic DNA. In vitro experiments with asbestos fibers were done at concentrations of both DNA fragments and asbestos fibers to maximize transformation; it is questionable whether appropriate conditions exist in vivo in which insertion mutagenicity can occur by this process. However, recent findings*661 have shown that DNA fragments exist outside cell membranes and may constitute as much as 1 % of total cellular DNA. Furthermore, inflammatory responses induced by asbestos fibers result in the destruction of macrophages and mesothelial cells (Mechanism VI), which can release degraded genomic DNA fragments. Therefore, in vivo DNA fragments are available to bind asbestos fibers that may be inserted into the DNA of normal cells. The likelihood of this mechanism occurring in vivo and contributing to asbestos carcinogenicity at low exposure levels is questionable as the concentrations of both carrier and nucleic acid fragments at asbestos deposition sites may not be sufficient to cause transfection of fragmented DNA into genomic DNA. Asbestos fibers have been shown to interact with chromosomes and have been proposed to interact with cytoskeletal structural proteins and have been shown to induce aneuploidy and polyploidy. Chromosomal imbalanced cells are abnormal and may be involved in the process of asbestos-induced carcinogenicity. Gibas et al.,*671 Popescu et al..'6!l and Tiainen et al.1691 reported that high incidences of numerical chromosomal abnormalities are present in human mesothelial cells, and the frequencies of certain abnormalities may be nonrandom. Oshimura et al.*70' reported that exposure ofSyrian hamster embryo cells to asbestos fibers resulted in a nonlinear dose-dependent increase in aneuploid. tetraploid. and binucleated cells, and that the cyiogenic changes, primarily aneuploidy, correlated with cell transformation. Mechanism III in Figure 1 illustrates the binding of asbestos fibers to tubulin, a protein that makes up the spindle apparatus necessary for chromosomal separation during cell division. It is still unclear if chromosomal abnormalities are part of the induction process of asbestos-induced carcinogenicity or whether they may be nonspecific secondary alterations that evolve during the progression of malignant cell growth. Mechanism IV illustrates that asbestos fibers can transferelectrons to cellular molecules afterentering the cell or by interacting at membrane surfaces; the cellular molecules in turn interact with genomic DNA and cause mutations. Mechanism V demonstrates promotional activity ofasbestos in which asbestos binds to membrane sites and causes epigenetic effects that initially alter cellular growth and function. These cellular alterations in DNA expression in an already initiated cell (transformed celll may cause it to progress further toward a cancer cell and may. by stimulating reactive oxygen species, cause a second mutation t transformation) that may be necessary for the formation of a cancer cell. The extent of the promotional response depends upon the number of interactions of asbestos with "receptor" sites (if specific receptor sites exist) on cell membrane surfaces. The kinetics of asbestos binding to membrane "receptors" and the extent of promotional activity arc not available, but the kinetics would most likely be nonlinear, and a threshold level needed to stimulate sufficient promotional activity may exist. The asbestos-induced inflammatory mechanism (VI) is multifaceted in that asbestos fibers attract macrophages, some of which are destroyed and release reactive oxygen species and DNA fragments and cause the release of macrophage growth-stimulating factors that result in cell regeneration at the deposition sites. The release of reactive oxygen species may induce mutations in the neighboring dividing cells; the DNA fragments from destroyed macrophages may bind to asbestos fibers, and become transplanted into nearby cells where the DNA fragments may . be inserted into genomic DNA (Mechanism II). The dashed lines in Figure 1 represent the contributing effects that the inflammatory response may have on the other postulated mechanisms by inducing cell division and making more target cells available to be transformed and releasing DNA fragments that could be inserted into genomic DNA. All six mechanisms presented here potentially could be involved in asbestos-induced mesothelioma and lung cancer and the mechanisms may vary with respect to cell type. Certain mechanisms may realistically occur only at very high concentrations of asbestos fibers and arc more likely to occur under in vitro laboratory conditions. Mechanisms V and VI, however, have been demonstrated in vivo, but adequate dose-response relationships for use in quantitative risk assessment arc not yet available. Quantitative cancer risk assessments have always suffered from lack of information on mechanisms of action of 547 R&S147975 ASBESTOS CARCINOGENICITY 23. FORNACE. A.J., Jr. (1982). Detection of DNA single-suand breaks produced during the repair of damage by DNA-protein Cfoss linking agents. Cancer Res. 42, 145-149. 24. MOSSMAN. B.T.. EASTMAN. A.. LANDESMAN. J.M.. and BRESNICK. E. (1983). Effectos of crocidolite and chrysolite asbestos on cellular uptake and metabolism of benzo[a|pyrcne in hamster uacheaj epithelial cells. Environ. Health Penpect. 51. 331-335. 25. L1BBUS. B.L., ILLENYE. S.A., and CRAIGHEAD. J.E. (1989). Induction of DNA strand breaks in cultured rat embryo cells by crocidolite asbestos as assessed by nick translation. Cancer Res. 49.5713-5718, 26. REISS. B,, SOLOMON. S,, TONG. C.. LEVENSTEIN, M.. ROSENBERG. S.H.. and WILLIAMS. G.H. (1982). Absence of mutagenic activity of three forms of asbestos in liver epithelial cells. Environ. Res. 27,389-397. 27. KENNE. K., LJUNGQUIST. S.. and RINGERTZ. N.R. (1986). Effects ofasbestos fibets on cell division, cell survival, and formation of thioguanine-tesistam mutants in Chinese hamster ovary cells. Environ. Res. 39,448-464. 28. CHAMBERLAIN. M.. and TARMY. E.M. (1977). Asbestos and glass fibers in bacterial mutation tests. Mutai. Res. 43, 159-164. 29. DENIZEAU. F,, MARION. M.. CHEVALIER. G.. and COTE. M.G. (1985). Inability of chrysotile asbestos fibers to modulate the 2-acetylaminofluotene-induccd UDS in primary cultures of rat hepatocytes. Mutat. Res. 155,83-90. 30. YANG, L.L.. KOURI. R.E.. and CURREN. R.D. (1984). Xeroderma pigmentosum fibroblasts are more sensitive to asbestos fibers than are normal human fibroblasts. Carcinogenesis 5. 291-294. 31. DUBES, G.R.. and MACH. L.R. (1988). Asbestos-mediated transfection of mammalian cell cultures. In Vitro Cell Dev. Biol. 24. 175-182. 32. APPEL. J.D.. FASY. T.M.. KOHTZ. D.S.. KOHTZ. J.D., and JOHNSON. E.M. (1988). Asbestos fibers mediate transformation of monkey cells by exogenous plasmid DNA. Proc. Natl. Acad. $ci. (USA) 85.7670-7674. 33. KE. Y,, REDDEL. R.R..GERWIN. BT.. REDDEL. H.K.. SOMERS, A.N.A..MCMENAMIN. M.G.. IAVECK. M.A.. STAHEL. R.A.. LECHNER. J.F.. and HARRIS. C.C. (1989). Establishment of a human in vitro mesothelial cell model system for investigating mechanisms of asbestos-induced mesothelioma. Am. J. Pathol. 134.979-991. 34. VALENTINE. R.. CHANG. M.J.W.. HART. R.W.. FINCH. G.U and FISHER. G.L. (1983). Theimal modification o( chrysotile asbestos: evidence (or decreased toxicity. Environ. Health Peispcet..51,357-368. 35. BEGIN, R., MASSE. S.. and BUREAU. M.A. (1982). Motphologic features and function ofthe airways in early asbestosis in the sheep model. Am. Rev. Respir. Dis. 126,870-879. 36. BOZELKA. B.E. GAUMER. H.R.. NORDBERG. J.. and SALVAGGIO. J.E. (1983). Asbestos-induced alterations o( human lymphoid cell mitogenic responses. Environ. Res. 30,281-290. 37. WARHEIT. D.B.. CHANG. L.Y.. HILL. L.H.. HOOK. G.E.R.. CRAPO. J.D- and BRODY. A.R. (1984). Pulmonary macrophage accumulation and asbestos-induced lesions at sites of fiber deposition. Am. Rev. Respir. Dis. 129.301-310. 38. MOSSMAN. B.T.. MARSH. J.P.. HARDWICK. D.. GILBERT. R.. HILL. S.. SESKO. A.. SHATOS. M.. DOHERTY. J,. WELLER. A., and BERGERON. M. (1986). Approaches to prevention of asbestos-induced lung disease using polyethylene glycol (PEGVconjugated catalase. J. Free Radicals Biol. Med. 2,335-338. 39. WONG. S.F.. HALUWELL. B.. RICHMOND. R.. and SKOWRONECK. W.R. (1981). The role of superoxide and hydroxyl radicals in the degradation of hvaluronic acid induced bv metal ions and bv ascorbic acid. J. inore. Biochem. 14. 127-134. 40. EBERHARDT. M.K.. ROMAN-FRANCO. A.A.. and QUILES. M.R. (1985). Asbestos-induced decomposition of hydrogen peroxide. Environ. Res. 37, 287-292. 41. WEtTZMAN. S.A.. and GRACEFFA, P. (1984). Asbestos catalyzes hydroxyl and superoxide radical generation from hydrogen peroxide (communication). Arch. Biochem. Biophys. 228.373-376. 42. MOSSMAN. B.T.. MARSH, J.P., and SHATOS. M.A. (1986). Alteration of superoxide dismuiase activity in tracheal epithelial cells by asbestos and inhibition of cytotoxicity by antioxidants. Lab. Invest. 54.204-212. 43. FISHER. G.L.. MOSSMAN. B.T.. MCFARLAND. A.R.. and HART. R.W. (1987). A possible mechanism of chrysotile asbestos toxicity. Drug Chem. Toxicol. 10,109-131. 44. LEANDERSON. P.. SODERKVIST. P,, TAGESSON, C.. and AXELSON. O. (1988). Formation of 8-hydroxyguanosine by asbestos and man made mineral fibers. Br. J. Ind. Med. 45,309-311. > 45. MOALLI, P.A.. MACDONALD. J.L..GOODGLICK. LA., and KANE, A. B. (1987). Acute injury and regeneration ofthe mesothelium in response to asbestos fibers. Am. J, Pathol, 128,426-445. 46. BRODY. A.R.. HILL. L.H.. ADKINS. B.. and O'CONNOR. R.W. (1981). Chrysotile asbestos inhalation m rats: deposition patient and reaction of alveolar epithelium and pulmonary macrophages. Am. Rev. Respir. Dis. 123.670-679. 47. WARHEIT. D.B.. OVERBY. L.H.. GEORGE. G.. and BRODY. A.R. (1988). Pulmonary macrophages arc attracted to inhaled particles through complement activation, Exp. Lung Res. 14,51-66. 48. GOODGUCK. L.A.. and KANE. A.G. (1986). Role ofreactive oxygen metabolites in crocidolite asbestos toxicity to mouse macrophages. Cancer Res. 46, 5558- 5566. 49. WEITZMAN, S.A.. and STOSSEL. T.P, (1981). Mutation caused by human phagocytes. Science 212.546-547. 50. TOPPING. D.C.. NETTESHELM. P.. and MARTIN. D.H. (1980). Toxic and tumorigenic effects of asbestos on tracheal mucosa. J. Environ. Pathol. Toxicol. 3. 261-275. 549