Mutation Research, 239 (1990) 17-27

17

Elsevier MUTREV 07279

Mutagenicity, carcinogenicity and teratogenicity of cobalt metal and cobalt compounds A. Lronard

1 and R. Lauwerys 2

t Teratogenicity and Mutagenicity Unit and 2 Industrial Toxicology and Occupational Medicine Unit, Faculty of Medicine, Catholic University of Louoain, B-1200 Brussels (Belgium)

(Received 5 October 1989) (Revision received19 December1989) (Accepted 20 December 1989)

Keywords: Cobalt; Mutagenicity; Carcinogenicity;Genotoxicity; Metal; Teratogenicity

Summary Cobalt metal and cobalt compounds are extensively used for the production of high-temperature alloys, diamond tools, cemented carbides and hard metals, for the production of various salts used in electroplating and as catalysts, drying agents in paints, additives in animal feeds and pigments. Cobalt oxides are used not only in the enameling industry and for pigments, but also in catalytic applications. There is no indication that cobalt metal and cobalt compounds constitute a health risk for the general population. Allergic reactions (asthma, contact dermatitis) can be induced by certain cobalt compounds. Interstitial fibrosis has also been observed in workers exposed to high concentrations of dust containing cobalt, tungsten, iron, etc., mainly in the cemented carbides and the diamond-polishing industries. Several experiments have demonstrated that single or repeated injections of cobalt metal powder or some forms of cobalt salt and cobalt oxide may give rise to injection site sarcoma in rats and in rabbits but the human health significance of such data is questionable. Intratracheal administration of a high dose of one type of cobalt oxide induces lung tumors in rats but not in hamsters. In the latter long-term inhalation of cobalt oxide (10 m g / m 3) did not increase the incidence of lung cancer. The human data are too limited to assess the potential carcinogenic risk for workers. Co 2÷ interacts with protein and nucleic acid synthesis and displays only weak mutagenic activity in microorganisms. Some cobalt salts have been reported to enhance morphological transformation of Syrian hamster embryo cells. Cobalt chloride displays some limited mutagenic activity in yeast and some cobalt compounds are able to produce numerical and structural chromosome aberrations in plant cells. Cobalt and its salts appear to be devoid of mutagenic and clastogenic activity in mammalian cells. Cobaltous acetate and cobaltous chloride have not been found to be teratogenic in hamsters and rats respectively.

Cobalt in the environment

Correspondence: Dr. A. L~.onard, Teratogenicity and Mutagenicity Unit, UCL 72 37, Avenue E. Mounier 72, B-1200 Brussels (Belgium).

Cobalt (atomic number 27; atomic weight 58.9332; melting point 1495°C; boiling point

0165-1110/90/$03.50 © 1990 ElsevierScience Publishers B.V. (Biomedical Division)

18 2900°C) forms approximately 0.001-0.002% of the earth's crust. Classified in group VIII of the periodic table between iron and nickel, cobalt is magnetic and closely resembles these metals in both its pure and its combined state but is harder. Ionic cobalt can exist in either a bivalent or a trivalent form. The most important cobalt minerals are sulfides and oxidized compounds such as carrollite (CuCo2S4), linnaeite (Co3S4), cattierite ((Co, Ni)S2), trienite ( ( C o 2 0 3 ) 2 • C u O • 6 H 2 0 ), sphaerocobaltite (COCO3). Some cobalt arsenides such as skutterudite ((Co,Ni,Fe)As3) and cobaltite ((CoAs)S) are also mined. There is only one naturally occurring isotope (59Co). More than 25,000 metric tons of cobalt are produced annually. The leading producing countries are Zaire (50%), Zambia (13%), Cuba, Australia, Canada, U.S.S.R., Japan, Finland and Morocco. About 55% of the world's output is used for the production of alloys (cobalt and nickel superalloys, permanent magnets, tool steels, medical implant alloys), cemented carbides, diamond tools and for the production of various salts used in electroplating and as catalysts, drying agents in paints, additives in animal feeds and pigments in glass and pottery industries (NIOSH, 1981). Cobalt oxides are used not only in the enameling industry and for pigments, but also in catalytic applications. Radioactive cobalt-60 (6°Co) is produced from ordinary 59Co by neutron bombardment. Since 6°Co can be encapsulated compactly, it has replaced radium in experimental medicine and cancer radiotherapy. 6°Co emits fl-rays and becomes 6°Ni, an emitter of energic y-rays. Its halflife is 5.3 years. The annual emission of cobalt into the environment has been estimated at 50 000 tons, of which 40000 come from natural sources (extraction by plants and weathering of rocks and soils) and 10000 from anthropogenic emissions. From the latter about 5000 tons result from the burning of fossil fuels and cobalt-containing materials (Merian, 1985). Presently there is no indication of any risk of environmental pollution by cobalt compounds. Soils contain 1-40 ppm (mean value 8-10 ppm), soils from basic igneous rock or argillaceous sediments 20-100 ppm, coal 1 ppm, inland waters and

drinking waters 0.1-0.5/~g Co/1. The atmospheric concentrations of cobalt in the general environment varies from less than 0.01 n g / m 3 to 4 n g / m 3 in industrial areas but levels up to 40 n g / m 3 have been reported occasionally (Koivistainen, 1980). In industry, the airborne concentration of cobalt is quite variable depending on the type of activity, workplaces and the selected monitoring system (area or personal sampling). Concentrations ranging from 0.005 to 0.3 m g / m 3 have been reported but peaks up to several m g / m 3 have also been observed (NIOSH, 1981; Morgan, 1983; Sprince et al., 1984; Meyer-Bisch et al., 1986). The new OSHA (U.S.A.) standard for cobalt metal fume and dust at the workplace is 0.05 m g / m 3 averaged over an 8-h work shift (TWA). N I O S H (1981), however, has concluded that there is insufficient evidence to warrant recommending an exposure limit (Centers for Disease Control, 1988). The American Conference of Governmental Industrial Hygienists (ACGIH, 1987) has recommended a TWA of 0.05 m g / m 3 for metal fume and dust. No maximum concentration at the workplace has been established by the Deutsche Forschungsgemeinschaft (1988) but a technical guiding concentration (TRK) of 0.5 m g / m 3 of cobalt in the form of cobalt metal, cobalt oxide and cobalt sulfide has been set up for the production of cobalt powder, catalysts, hard metal (tungsten carbide) and magnet production (processing of powder, machine pressing and mechanical processing of inserted articles) and 0.2 m g / m 3 for other activities. The TWA for cobalt metal, fume and dust amounts to 0.05 m g / m 3 in Sweden, 0.1 m g / m 3 in the U.K. and 0.5 m g / m 3 in U.S.S.R.

Metabolism and toxicology It should be stressed that the physical and chemical characteristics play an important role in the toxicokinetic and toxicity properties of the Co-containing substances. Since these characteristics have frequently not been reported in the papers analyzed in the present review, we were not always in a position to take this important aspect into account. The human adult body contains about 1 - 2 mg of cobalt: 14% in bone, 13% in muscle, and the remainder in other soft tissues (Venugopal and

19 Luckey, 1978). Cobalt is an essential nutrient to humans as an integral and necessary component of cyanocobalamin or vitamin B12 where it is bound in a porphyrin-like skeleton and to the cyanide ion. The presence of the cyanide group is an artefact of the method of isolation of the vitamin B12 crystals. In this form, vitamin B12 has no biological activity. The biochemical reactions of the cobalamins are dkle to the vitamin B12 coenzymes, methylcobalamin, in which the cyanide group of vitamin B12 is replaced by a methyl group, and adenosylcobalamin, in which the cyanide is replaced by the adenosyl group (Donaldson et al., 1988). No other function of cobalt in humans has been established. A daily intake of about 50 #g (of which 40/~g are in the form of vitamin B12) maintains cobalt equilibrium in the human. In contrast to humans, ruminant animals and some other monogastric mammals have an intestinal microflora which can utilize cobalt in the formation of vitamin B12. In subjects not occupationally exposed to cobalt the blood concentration is below 2.0/zg/1 whereas in urine it rarely exceeds 1/zg/1 (Elinder, 1984). Cobalt in the form of cobalt oxide or cobalt chloride is taken up by mammals after both inhalation and ingestion. The degree of respiratory absorption is unknown in humans, but in hamsters it has been estimated that about one third of inhaled cobalt oxide is absorbed (Elinder and Friberg, 1986). The gastrointestinal absorption of soluble cobalt compounds, such as cobalt chloride, is around 25% with large inter-individual variation. There are indications that cobalt and iron compete for the same transport mechanism in the duodenum (Taylor and Marks, 1978). After absorption from the lung or the gut cobalt is distributed mainly to the liver and kidney. Excretion takes place predominantly via the urine and secondarily via the feces probably after biliary excretion (Gregus and Klaassen, 1986). The short-term biological half-life is rapid, within days, but data indicate that a proportion of cobalt is retained with a biological half-life in the order of a year (Elinder et al., 1988). The measurement of cobalt concentrations in urine can be used for the assessment of current exposure. The acute oral toxicity of cobalt compounds is not high. Their acute LDs0 in rats ranges from 80

to 6400 m g / k g body weight (NIOSH, 1987). Cobalt can bind to thiol groups and can interfere with the activity of a number of enzymes in vivo. In the form of CoC12 • 6H20, it inhibits glucose oxidation in mice in vivo as determined by respiratory excretion of 14CO2 (Isom and Way, 1974). Large doses of Co also inhibit some enzymes like cytochrome oxidase, succinic dehydrogenase, glucose-6-phosphate dehydrogenase (Beskid, 1967). Cobalt can bind to calcium-binding proteins and blocks the calcium channel in squid axons (Baker et al., 1973; Phillips, 1980). High doses of cobalt given to rats can also deplete cytochrome P450 concentrations in the liver through depression of 8-aminolevulinic acid synthetase activity and activation of heme oxygenase (De Matteis and Gibb, 1977; Maines and Kappas, 1976; Sinclair et al., 1982). However, repeated administration of cobalt (10 m g / k g ) as cobalt chloride to rats resulted in an increase of hemoglobin concentration (Murdock, 1959). In workers chronically exposed to cobalt metal and cobalt salts, the 2 main target organs are the skin and the respiratory tract (Lauwerys, 1989). Cobalt is an allergen capable, in some subjects, of producing contact dermatitis, rhinitis, asthma and possibly allergic alveolitis. Such persons, once sensitized, may develop clinical manifestations even after exposure to minute amounts of cobalt. A chronic lung parenchymal disease called 'hard metal disease' and characterized by the presence of some interstitial fibrosis and restrictive ventilatory impairment has also been reported mainly among workers exposed to dust containing a.o. cobalt in the cemented carbide and the diamondpolishing industries. Pathologic findings have ranged from intense alveolitis, resembling desquamative or giant-cell interstitial pneumonitis, to end-stage non-specific pulmonary fibrosis. The initial signs of the disease appear after a period of exposure ranging from a few months to several years. It should be stressed that workers who developed hard metal disease had sustained exposures to cobalt mixed with other substances such as tungsten carbide, iron, diamond, etc. Polycythemia has also been observed in a few hard metal workers (Payne, 1977). Outbreaks of cardiomyopathy among heavy beer drinkers were reported from several areas throughout the world

20 (Canada, U.S.A. and various European countries such as Belgium) after breweries introduced cobalt chloride hexahydrate (up to 1.1 mg/1) to improve foaming and disappeared when the use of cobalt was discontinued (Wiberg et al., 1969). The synergistic effect of simultaneous exposure to cobalt and a poor alcohol nutritional status for the development of cardiomyopathy has been confirmed in animal experiments (Wiberg et al., 1969; Venugopal and Luckey, 1979). Some victims of beerdrinker's cardiomyopathy also developed a goiter a n d / o r hypothyroidism (Taylor and Marks, 1978). Toxic effects in various organs (i.e., liver, kidney, pancreas) have been observed in animals which usually had been treated with high doses of cobalt but the relevance of these observations for humans has not been documented.

Carcinogenicity Experimental data Although there exists no evidence for the carcinogenicity of cobalt metal and cobalt compounds administered by the oral route in mammals, several animal experiments have demonstrated that single or repeated i.m., s.c. or i.p. injections of cobalt metal powder or some cobalt c o m p o u n d s may give rise to injection-site rhabdosarcoma, pleomorphic rhabdomyofibrosarcoma and fibrosarcoma. In the experiments of Heath (1956), 10 male and 20 female rats were given one intramuscular injection of 28 mg cobalt metal powder into the high muscle. 5-12 months later, 17 of the animals developed tumors at the site of injection; 11 were rhabdomyosarcomas. Comparable results were reported with compounds such as cobalt sulfide and cobalt oxide (COO) by Gilman and Ruckerbauer (1962). Weaver et al. (1956) produced tumors in the thyroid gland and at the site of injection with Co metal powder. Thomas and Thiery (1953) observed transplantable liposarcomas and hyperplasia of adipose tissue after injection of Co metal dust to rabbits. In the experiment of Gilman (1962), 32 Wistar rats were injected with a single 20-mg dose into both the left and the right muscle whereas C3H and Swiss mice received injections of 5 mg per thigh. Twelve of the 24 rats surviving 90 days after the treatment developed tumors. The sulfide of cobalt

proved markedly more active than the oxide in numbers of tumors induced although no difference in the average latent period was found. In mice, however, in agreeing with the previous resuits of H e a t h (1959) and G i l m a n and Ruckerbauer (1962), CoO failed to induce tumors. Heath et al. (1971) and Swanson et al. (1973) were able to produce such neoplasms in rats with fine particles from surgical prqstheses made from a cobalt-chromium alloy. In rats, repeated intratracheal administration of cobalt oxide at 2 and 10 m g / k g every other week to a total dose of 78 and 390 mg/kg, respectively, induced lung tumors in 2 and 6% of the treated animals respectively. A similar study in hamsters provided negative results (Steinhoff, 1987). Wehner et al (1977) exposed hamsters to cobalt oxide in air at a concentration of 10 m g / m 3, 7 h / d a y for 5 days a week in a lifelong experiment. There was no increase in the incidence of lung cancer in exposed animals compared to controls. Negative results were also obtained by Stoner et al. (1976) who examined the production of lung adenomas in strain A mice receiving 95-475 m g / k g cobalt acetate in 19 intraperitoneal injections. In rhabdosarcoma induced in rats by intramuscular implantation of powdered cobalt, the metal slowly dissolves and disappears from the injection site (Heath and Webb, 1967; Heath et al., 1969). On the basis of experiments performed in vitro, the same authors have shown that metallic cobalt in powdered form slowly reacts with serum proteins to form soluble complexes. They have suggested that the cobaltprotein complexes may enter the cell by endocytosis and that subsequent digestion of the carrier proteins by lysosomal proteinases results in an intracellular liberation and redistribution of cobalt. At least 50% of the metal accumulated in the nuclei of the tumorous cells is bound to the nucleoli, the remainder being distributed in the nuclear sap, chromatin and mitochondria (Daniel et al., 1967). One study only was performed on the carcinogenicity of soluble cobalt salts. In 1977, Shabaan et al. gave 20 rats subcutaneous injections of cobalt chloride (form not mentioned) daily at a dose of 40 m g / k g in 2 courses of 5 days separated by a 9-day interval. Of the 11 animals that survived for 12 months, 8 developed fibrosarcoma. Four of

21 these neoplasms were not at the site of injection. The experiment was subsequently repeated on 20 animals and it was confirmed that a high prevalence of sarcoma in exposed rats appeared after only 8 months (6 sarcomas in 16 surviving rats).

Epidemiological data Despite heavy occupational exposure to cobalt in the past, few cases of cancer have been observed in workers (Kusaka et al., 1984). In nickel extraction plants in the U.S.S.R. an increased mortality from lung cancer was found among workers who had been occupied in the cobalt recovery shops as well as in the nickel-processing departments (Saknyn and Sabynina, 1973). However, there was simultaneous exposure to other genotoxic agents that could explain this finding (Kazantzis, 1981). Mur et al. (1987) have studied the mortality between 1950 and 1980 of a cohort of 1143 French workers in an electrochemical plant producing cobalt and iodium. Among cobalt production workers, there was a relative increase in deaths especially from lung cancers (standard mortality ratio: 4.66; 4 cases only). The authenticity of the occupational origin of the risk could not be established due to the low number of cases and because the role of tobacco consumption and simultaneous exposure to arsenic and nickel could not be taken into account. A nonstatistically significant increase in lung cancer deaths (7 observed, 3.1 expected) was found among a cohort of 1901 Swedish male hard metal workers with at least 5 years' employment (Hogstedt and Alexandersson, 1987). However, no dose-response relationship was found. In conclusion, there is insufficient evidence to indicate whether under current occupational exposure conditions, cobalt metal a n d / o r some of its derivatives increase the risk of cancer in workers. It can be added that no fibrosarcomas were observed in hundreds of patients with prosthetic implants made of an alloy containing cobalt, chromium and molybdenum (McKee, 1971).

Mutagenicity Co(II) interacts with proteins and nucleic acids (Jacobson and Turner, 1980) and functions as cofactor for D N A polymerase during D N A replication and the accompanying phosphate-bond

cleavage and formation. According to Eichhorn and Shin (1968), although Co(II) can bind to DNA, it causes no or little D N A damage in vitro. However, Moorhouse et al. (1985) suggested that Co(II) ion catalyzes hydroxyl radical and possible 'crypto-hydroxyl' radical formation under physiological conditions and Yamamoto et al. (1989) found that Co(II) induced strong D N A cleavage in the presence of hydrogen peroxide even without alkali treatment. The study of Yamamoto et al. (1989) was performed on plasmid pbcNI which carries a 6.6-kb BamHI chromosomal D N A segment containing the human c-Ha-ras-1 protooncogen. Co(II) has the ability to substitute for Mg(II) as activator of D N A polymerase (Sirover et al., 1979) and decreases the fidelity of D N A synthesis in vitro (Sirover and Loeb, 1976; Zakour et al., 1981). The concentration required to increase the error frequency was 4 mM (Sirover and Loeb, 1976; Miyaki et al., 1977). Cobaltous ion, Co(II), combines reversibly with histidine (Burk et al., 1946) and can also form a complex with adenine (Liquier-Milward, 1951). In vitro, 0.25 or 1 m M cobalt chloride failed to produce apurinic sites in calf thymus D N A (Schaaper et al., 1987), which may be a common intermediate in mutagenesis by diverse chemical agents (Loeb, 1985). Only weak crosslinking of proteins to D N A was observed by Wedrychowski et al. (1986) in Novikoff ascites hepatoma cells exposed for 8 h to cobalt chloride. Concentrations of cobalt chloride ranging from 1 to 1000 M neither modify the biological activity of single-stranded ~X174 D N A exposed to the metal followed by transfection into E. coli spheroplasts (Schaaper et al., 1987). The same compound failed to induce prophage in Escherichia coli WP2 s (X) (Rossman et al., 1984). The highest concentration tested (3.2 × 10 -5 M) inhibited cell growth. Cobaltous nitrate at a concentration of 600 g g / m l failed also to produce gene mutations in the T4 bacteriophage system (Corbett et al., 1970). Several experiments suggest that cobalt salts display only weak mutagenic activity in microorganisms. In preliminary studies Nishioka (1975) obtained negative results in the rec-assay on Bacillus subtilis treated with 0.05 M cobalt chloride but in the same assay on the H17 (Recc +, arg- try-) and M45 (Rec-, arg- try- ) strains Kanematsu et

22

al. (1980) and Kada et al. (1980) obtained a positive response with 0.05 M cobalt chloride and a mild one with 0.25 M cobalt hydroxide, 0.05 M cobalt sulfate and 0.05 M 2COCO 3 • 3Co(OH)2. Doses of 0.03-300 n m o l e / m l of Co(NO3) 2 - 6 H 2 0 gave negative results in the bacterial colorimetric SOS chromotest assay (Olivier and Marzin, 1987). Cobalt chloride and cobalt hydroxide failed also to produce reverse mutations in Escherichia coli and Salmonella typhimurium (Kanematsu et al., 1980; Tso et Fung, 1981; Arlauskas et al., 1985). Schultz et al. (1982), however, obtained positive results with some of the 15 hexacoordinate cobalt(III) compounds tested for DNA-damaging capabilities using an E. coli differential repair assay and for the ability to induce reverse mutations in S. typhimurium. According to the authors the genetic activity of the tested cobalt compounds seemed to be particularly dependent upon the structure of the ligands coordinated about the metal ion. Cobalt chloride at concentrations up to 5 /~M had no effect on UV mutagenesis in E. coli WP2 (Rossman and Molina, 1986). Cobalt chloride in post-treatment media has even been shown to diminish markedly the reverse mutations induced by N-methyl-N'-nitro-N-nitrosoguanidine, as well as "t-rays, in Escherichia coli WP2 B / r trp (Kada et al., 1979). The same compound was also found to display antimutagenic properties in a Bacillus subtilis mutator having an altered D N A polymerase III (Inoue et al., 1981), on the induction of reverse mutations in Salmonella typhimurium TA98 and TA1538 by Trp-P-1 (Mochizuki and Kada, 1982), and by Kuroda and Inoue (1988) studying the effect of CoC12 • 6 H 2 0 at concentrations of 20 and 40 m g / m l on spontaneous mutation rates in Bacillus subtilis NIGI125 his rout-1. Cobalt chloride, however, enhances the mutagenicity of 9aminoacridine (9AA) and can induce mutation when combined with the non-mutagens 4-aminopyridine (4AP), 4-aminoquinoline (4AQ) and harman (HM) (Ogawa et al., 1986). According to the authors, these compounds could act as carrier of the Co(II) cation which is released after penetration and may, thereafter, exert its mutagenic effect. This hypothesis was confirmed by additional experiments: (C2Hs)4N 2 • CoBr 4 and (C2Hs)4N 2 • Coi4, which possess the ligands with compara-

tively weak chelating affinity and no mutagenic activity, induced apparent increases in the number of revertant colonies in strain TA2637; by contrast, C0(C5H702) 2 (H20)2 , Co(CIoH8N2) 3 C12 • 6H20, Co(C12HsN2) 3 CI" 8 H 2 0 , Co(C10HsN2) 3 (Ca0HsN2) 3 (C104)2, and Co(C12H8N2)3 (C104)2,

which have ligands with a stronger affinity to the Co(II) ion, could not induce mutation. Co(III) complexes, such as Co(C10HsN2) 3 C13-3H20, CO(CloHsN2)3 (C104)3 • 3 H 2 0 , Co(C12HsN2)3 C13 • 7 H 2 0 , Co(C12H8N2) 3 (C104) 3 • 2 H 2 0 ,

Co(NH3)5X Y (where X, Y = N H 3, C12; C1, C1; N02, C12) , Co(C2H8N2)2X Y (where X, Y =

C2HsN 2, C1; C204, C1; CO 3, C1), and cisCo(C2H8N2) 2 C1 X Y (where X, Y = NH3, C12; C1, CI) were inactive. This inactivity could be the result of their kinetic stability, although the possibility remains that these compounds may not be transported across the cell membranes. Some cobalt salts have been shown to enhance morphological transformation of Syrian hamster embryo cells by a simian adenovirus, SA7 (Casto et al., 1979; Di Paolo and Casto, 1979). For Co(C2H302)2, at the lowest concentration tested that was positive (0.2 mM), the enhancement ratio was 7.2, the corresponding values being 0.25 m M and 6.1 for CoMoO 4. Comparable results have been reported by Pavlova et al. (1979) and Heck et al. (1982). Costa et al. (1982) reported that crystalline CoS 2 particles were actively phagocytosed by cells and potently induced morphological transformation of Syrian hamster embryo cells in a concentration-dependent fashion whereas amorphous CoS was not actively phagocytosed and induced considerably less morphological transformation. It should be pointed out, however, that from a review of the literature Brookes (1981) concluded that little or no evidence exists that a change of cell morphology in culture equates with a transformed genotype. Cobalt chloride is known to induce respiratory deficiency (Lindegren et al., 1958) and to cause 'petite' (mitochondrial) mutations in yeast (Prazmo et al., 1975) but no mitochondrial erythromycin-resistant mutations (Putrament et al., 1977). Cobalt chloride exhibited a weak convertogenic activity in Saccharomyces cerevisiae and was only marginally active for reverse mutations (Singh, 1983; Kharab and Singh, 1985). Increasing

23 concentrations (0.75, 1.5, 3.0 and 6.0 m g / m l ) resulted in 1.4-, 2.1-, 3.1-, 3.3-fold increases and 1.5-, 2.0-, 2.1-, 2.4-fold increases over spontaneous rates for gene conversion and reversion, respectively. The great majority of carcinogenic as well as non-carcinogenic metal salts have been shown to be able to produce numerical and structural chromosome aberrations in plant cells (Lronard, 1988). The experimental results reported in the literature confirm that this is also true for some cobalt compounds. Komezynski et al. (1963) reported the capability of cobalt salts to induce spindle inhibition and chromosomal aberrations in that system. Herich (1965) observed clumping and stickiness of chromosomes in Vicia faba treated with cobalt salts. Santanu et al. (1984) also obtained positive results in root meristems of Allium cepa treated for 1-5 h with 0.10 m g / m l to 1 mg aqueous solution of Co(BSOP)(NH3) 2 NO 3 and Co(BSOP)(PY)2 NO3, where B S O P = d i a n i o n of N,N-ortho-phenylenebis(salicylaldimine) and PY = pyridine. With the same compounds a marked effect was noted by Tribedi et al. (1984) following a 5-h treatment with 0.25 m g / m l solution. No aberrations persisted after a recovery of 5 - 7 days. CoC12 • 6 H 2 0 has been shown to decrease the cloning efficiency of C H O cells (Hsie et al., 1978; Tan et al., 1984). The concentration required to reduce the CE to 50% was 140/~M. Cobalt chloride as CoC12 • 6H20, however, failed to induce trifluorothymidine-resistant ( T F T Res) mutants in L 5 1 7 8 / T K +/- mouse lymphoma cells by directly exposing cells to doses of 5.69-57.11 / t g / m l (Amacher and Pailled, 1980). Cobalt chloride at a dose of 0.2 m M very slightly increased the mutation frequency at the hypoxanthine-guanine phosphoribosyl transferase ( H G P R T ) locus in Chinese hamster V79 cells (Miyaki et al., 1979) and comparable results are reported by Morita et al. (1985) for the production of mutations at the H G P R T locus in mouse FM3A cells by cobalt chloride and [(C2Hs)4N]2[CoC12] in the concentration range of 2 - 3 x 10 -4 M with a 48-h treatment. McLean et al. (1982) reported that the DNA-strand breaks were increased in human white blood cells exposed to low concentrations of cobalt chloride and Robison et al. (1982) observed the production of D N A single-strand breaks in C H O cells exposed

to 1 0 / ~ g / m l of cobalt sulfide for 24 h. HamiltonKoch et al. (1986) observed that cobalt chloride reduced the cloning efficiency of C H O cells more than that of human diploid fibroblasts, an effect which was shown to be correlated only partly to differences in cellular uptake but mainly to true strand-break induction. Cobalt salts are, however, apparently devoid of clastogenic properties in animal cells as shown by the results reported by Paton and Allison (1972). In their study, subtoxic doses ( < 8.0 × 10 -7 M) of cobalt nitrate added to leukocyte cultures 48 h before fixation and to fibroblast cultures 24 h before fixation failed to induce structural chromosome aberrations although the compound was toxic to the cells at 8.0 x 10-7 M. The results obtained by Sanyal et al. (1980) in their in vivo study therefore appear rather surprising. The authors claim, indeed, that repeated oral administration of cobalt chloride at a rate of 15 m g / k g body weight over a period of 21 days produced an increase in the number of dividing cells in male albino rats and chromosomal aberrations such as stickiness, breaks or gaps. Cytogenetic analysis of bone marrow cells is generally considered a valuable method to estimate the clastogenic potential of chemicals. However, several factors such as the solubility of the chemical and its toxicity have been shown to influence greatly the quality of the preparations. This argument has been frequently used to explain the contradictions observed between different authors or between the results obtained in vitro and in vivo as is the case for some cobalt compounds. For this reason, the in vivo micronucleus test is actually preferred to chromosome examination. The same remarks can probably be applied to the results of Nehe'z et al. (1982) who report briefly, in an abstract, on an increase of chromosome aberrations in the bone marrow of mice treated with cobalt blue.

Teratogenieity Relatively few studies have been performed to evaluate the effects of cobalt salts on developing organisms. Kury and Crosby (1968) injected the chick yolk sac on the 4th day with 0.4-0.5 mg of CoC12. They observed anemia, thyroid epithelial hyperplasia as well as gross abnormalities in eyes and lower extremities among the survivors.

24 The effects of a prenatal exposure to cobalt salts appear to be less marked for mammals. A considerable reduction in collagen synthesis was reported b y Srivastava et al. (1976) in e m b r y o n i c rat calvaria maintained in tissue culture and exposed for 24 h to a 10 -4 M concentration of cobalt chloride. This concentration is in the same range as that f o u n d in tissues surrounding implanted metallic materials. At a dose of 5 m g / k g cobaltous acetate ( C o ( C H 3 C O O ) 2 . 4 H 2 0 ) failed to induce any embryocidal or teratogenic effect when administered alone to golden hamsters and also did not protect from the teratogenic effect of c a d m i u m ( F e r m and Carpenter, 1968). Kasirsky et al. (1967, 1969) showed that a single i.p. injection on day 10 or 11 of either cobaltous chloride (25 m g / k g ) or sodium cobalt nitrite (50 m g / k g ) induces cleft lips and palates in the CF-1 strain of mice but protects against the p r o d u c t i o n of cleft palate b y cortisone. C y a n o c o b a l a m i n , sodium cobalt nitrite, and cobaltous chloride also protect against the teratogenic effects of phenytoin in CF-1 mice (Mitola et al., 1978). In the experiments of Wide (1984) 5 m M of CoC12 • 6 H 2 0 was administered in the tail veins of mice either late during day 3 of pregnancy or early on d a y 8 of pregnancy. Examination of the fetuses, 2 days before birth, on the 17th day of pregnancy, showed only an interference of the metal with fetal skeletal ossification. D o m i n g o et al. (1985) gave 12, 24 or 48 m g / k g / d a y of cobalt chloride to Wistar rats from the 14th day of gestation through 21 days of lactation. T h e y observed a reduction in the n u m b e r of litters, a dose-dependent delay in the growth of living pups but no significant differences in organ weights. N o external malformations were observed in any case. Paternain et al. (1988) gave daily oral doses of 0, 25, 50 or 100 m g / k g CoC12 to S p r a g u e - D a w l e y rats on days 6 - 1 5 of gestation. Females were killed on d a y 20. Maternal effects including significant reductions in weight gain and food cons u m p t i o n were f o u n d in the 1 0 0 - m g / k g / d a y group. N o treatment-related changes were recorded in the n u m b e r of c o r p o r a lutea, total implants, resorptions, the n u m b e r of live and dead fetuses, fetal size parameters or fetal sex distribution data. Examination of fetuses for gross external

abnormalities, skeletal malformations, or ossification variations revealed that CoC12 did not produce teratogenicity or significant fetotoxicity in the rat at doses as high as 100 m g / k g / d a y . In an abstract G ~ r b a n et al. (1986) reported that intraperitoneal injection of cobalt chloride (10 × 10 -2 g%) in pregnant female rats o n days 7 / 8 and 1 5 / 1 6 of p r e g n a n c y induced changes in biochemical homeostasis, in somatometric indices and in embryonic and fetal mortality.

Conclusions In spite of the fact that several experiments have demonstrated that single or repeated injections of cobalt metal p o w d e r or cobalt c o m p o u n d s m a y give rise to injection-site tumors there is insufficient evidence that cobalt and its comp o u n d s represent a carcinogenic risk. for humans. In that respect it is of interest to point out that cobalt metal and its c o m p o u n d s display only weak mutagenic properties. In m a m m a l s , cobalt acetate and cobalt chloride seem to be devoid of teratogenic activity.

References ACGIH (1988) Threshold Limit Values and Biological Exposure Indices for 1988-1989, American Conference of Governmental Industrial Hygienists, Cincinnati, OH. Amacher, D.A., and S.C. Paillet (1980) Induction of trifluorothymidine-resistant mutants by metal ions in L5178 Y/TK +/- cells, Mutation Res., 78, 279-288. Arlauskas, A., R.S.U. Baker, A.M. Bonin, R.K. Tandon, P.T. Crisp and J. Ellis (1985) Mutagenicity of metal ions in bacteria, Environ. Res., 36, 379-388. Baker, P.F., H. Me,eves and E.B. Ridgway (1973) Effects of manganese and other agents on the calcium uptake that follows depolarization of squid axons, J. Physiol., 231, 511-526. Beskid, M. (1967) The action of cobalt on kidneys of the guinea pig, Folia Histochem. Cytochem., 5, 33-72. Brookes, P. (1981) Critical assessment of the value of in vitro cell transformation for predicting in vivo carcinogenicity of chemicals, Mutation Res., 86, 233-242. Burk, D., J. Hearon, L. Caroline and A.L. Shade (1946) J. Biol. Chem., 165, 723-726. Casto, B.C., J. Meyers and J.A. Di Paola (1979) Enhancement of viral transformation for evaluation of the carcinogenic or mutagenic potential of inorganic metal salts, Cancer Res., 39, 193-198. Centers for Disease Control (1988) Recommendations for Occupational Safety and Health Standard, NMWR, 37 (Suppl. 5-7), 1-30.

25 Corbett, T.H., C. Heidelberger and W.F. Dove (1970) Determination of the mutagenic activity to bacteriophage T4 of carcinogenic and non-carcinogenic compounds, Mol. Pharmacol., 6, 667-679. Costa, M., J.D. Heck and S.H. Robison (1982) Selective phagocytosis of crystalline metal sulfide particles and DNA strand breaks as a mechanism for the induction of cellular transformation, Cancer Res., 42, 2757-2763. Daniel, M.R., J.C. Heath and M. Webb (1967) Respiration of metal induced rhabdomyosarcomata, Br. J. Cancer, 21, 780-786. De Matteis, F., and A.H. Gibb (1977) Inhibition of haem synthesis caused by cobalt in the rat liver, Biochem. J., 162, 213-216. Deutsche Forschungsgemeinschaft (1988) Maximum Concentrations at the Workshop and Biological Tolerance Values for Working Materials, Verlagsgesellschaft mbH, Weinheim. Di Paolo, J.A., and B.C. Casto (1979) Quantitative studies of in vitro morphological transformation of Syrian hamster cells by inorganic metal salts, Cancer Res., 39, 1008-1013. Domingo, J.L., J.L. Paternain, J.M. Llobet and J. Corbella (1985) Effects of cobalt on postnatal development and late gestation in rats upon oral administration, Rev. Esp. Fisiol., 41,293-295. Donaldson, J.D., S.J. Clark and S.M. Grimes (1980) Vitamin B12 coenzymes, in: The Cobalamins in Cobalt in Medicine, Agriculture and the Environment, Ch. 6, The Cobalt Development Institute, London. Eichhorn, G.L., and Y.A. Shin (1968) Interaction of metal ions with polynueleotides and related compounds. XII. The relative effect of various metal ions on DNA helicity, J. Am. Chem. Soc., 90, 7323-7328. Elinder, C.G. (1984) Health hazards from exposure to cobalt, with special reference to carcinogenic, mutagenic and teratogenie effects, Toxicol. Environ. Chem., 7, 251-256. Elinder, C.G., and L. Friberg (1986) Cobalt, in: L. Friberg, G.F. Nordberg and V.B. Vouk (Eds.), Handbook on the Toxicology of Metals, Elsevier/North-Holland Biomedical Press, Amsterdam, New York, Oxford, pp. 211-232. Elinder, C.G., L. Gerhadson and G. Oberdoester (1988) Overview, in: J.W. Clarkson (Ed.), Biological Monitoring of Toxic Metals, Plenum Press, New York, pp. 1-72. Ferm, V.H., and S.J. Carpenter (1968) The relationship of cadmium and zinc in experimental mammalian teratogenesis, Lab. Invest., 18, 429-432. GSrban, Z., G. Daranyi, I. Borza, F. Jianu, M. Brazda, R. Nemes and I. Eremia (1986) Homeostasis changes induced by cobalt action on the materno-fetal complex in rats. I. Biochemical and somatometric aspects, Teratology, 33, 17A-18A. Gilman, J.P.W. (1962) Metal carcinogenesis. II. A study on the carcinogenic activity of cobalt, copper, iron, and nickel compounds, Cancer Res., 22, 158-165. Gilman, J.P.W., and G.M. Ruckerbauer (1962) Metal carcinogenesis. I. Observations on the carcinogenicity of a refinery dust, cobalt oxide, and colloidal thorium dioxide, Cancer Res., 22, 152-157.

Gregus, Z., and C.D. Klaassen (1986) Disposition of metals in rats: a comparative study of fecal, urinary and biliary excretion and tissue distribution of eighteeen metals, Toxicol. Appl. Pharm., 85, 24-38. Hamilton-Koch, W., R.D. Snyder and J.M. Lavelle (1986) Metal-induced DNA damage and repair in human diploid fibroblasts and Chinese hamster ovary cells, Chem.-Biol. Interact., 59, 17-28. Heath, J.C. (1956) The production of malignant tumors by cobalt in the rat, Br. J. Cancer, 10, 668-673. Heath, J.C. (1959) Attempts to induce tumors with powdered metallic cobalt in other sites and species, Annu. Rep. Br. Emp. Cancer Campaign, 322-323. Heath, J.C., and M. Webb (1967) Content and intracellular distribution of the inducing metal in the primary rliabdomyosarcomata induced in the rat by cobalt, nickel and cadmium, Br. J. Cancer, 21,768-779. Heath, J.C., M. Webb and M. Caffrey (1969) The interaction of carcinogenic metals with tissues and body fluids. Cobalt and horse serum, Br. J. Cancer, 23, 153-166. Heath, J.C., M.A.R. Freeman and S.A.V. Swanson (1971) Carcinogenic properties of wear particles from prostheses made in cobalt-chromium alloy, Lancet, i, 564-566. Heck, J.D., and M. Costa (1982) In vitro assessment of the metal compounds. I. Mammalian cell transformation, Biol. Trace Elem. Res., 4, 71-82. Herich, R. (1965) The effects of cobalt on the structure of chromosomes and on mitosis, Chromosoma, 17, 194-198. Hogstedt, C., and R. Alexandersson (1987) Mortality among hard-metal workers in Sweden, Scand. J. Work Environ. Health, 13, 177-178. Hsie, A.W., J.P. O'Neill, J.R. San Sebastian, D.B. Couch, P.A. Brimer, W.N.C. Sun, J.C. Fuscoe, N.L. Forbes, R. Machnaoff, J.C. Riddle and M.H. Hsie (1978) Quantitative mammalian cell genetic toxicology: study of the toxicity and mutagenicity of seventy individual environmental agents related to energy technologies and three subfractions of a crude synthetic oil on the C H O / H G P R T system, Oak Ridge, National Laboratory, Report, pp. 293-315. Inoue, T., Y. Ohta, Y. Sadaie and T. Kada (1981) Effect of cobaltous chloride on spontaneous mutation induction in a Bacillus subtilis mutator strain, Mutation Res., 91, 41-45. Isom, G.E., and J.L. Way (1974) Alteration of in vivo glucose metabolism by cobaltous chloride, Toxicol: Appl. Pharmacol., 27, 131-139. Jacobson, K.B., and J.E. Turner (1980) The interaction of cadmium and certain other metal ions with proteins and nucleic acids, Toxicology, 16, 1-37. Kada, T., T. Inoue, A. Yokoiyama and L.B. Russell (1979) Combined genetic effects of chemicals and radiation, in: S.O. Kada, M. Imamura, T. Terashima and H. Yamaguchi (Eds.), Radiation Research, Proc. 6th Int. Congr. Radiat. Res., Jpn. Ass. Radiat. Res., Tokyo, pp. 711-720 Kada, T., K. Hirano and Y. Shirasu (1980) Screening of environmental mutagens by the rec-assay system with Bacillus subtilis, in: F.J. de Serres and A. Hollaender (Eds.), Chemical Mutagens. Principles and Methods for their Detection, Vol. 6, Plenum Press, New York, pp. 149-171.

26 Kanematsu, N., M. Hara and T. Kada (1980) Rec-assay and mutagenicity studies on metal compounds, Mutation Res., 77, 109-116. Kasirsky, G., G.F. Gautieri and D.E. Mann (1967) Inhibition of cortisone-induced cleft palate in mice by cobaltous chloride, J. Pharm. Sci., 56, 1330-1332. Kasirsky, G., W.T. Sherman, R.F. Gautieri and D.E. Mann Jr. (1969) Cobalt-cortisone interrelationships in the induction and inhibition of cleft palate in mice, J. Pharm. Sci., 58, 766-767. Kazantzis, G. (1981) Role of cobalt ion, lead, manganese, mercury, platinum, selenium and titanium in carcinogenesis, Environ. Health Perspect., 40, 143-161. Kharab, P., and I. Singh (1985) Genotoxic effects of potassium dichromate, sodium arsenite, cobalt chloride and lead nitrate in diploid yeast, Mutation Res., 155, 117-120. Koivistainen, P. (1980) Mineral element composition of FinIrish foods, Acta Agric. Scand., 22, Suppl., 171 pp. Komczynski, L., H. Nowak and L. Reyniak (1963) Effect of cobalt, nickel and iron on mitosis in roots of broad bean (Vicia faba), Nature (London), 198, 1016-1017. Kuroda, Y., and T. Inoue (1988) Antimutagenesis by factors affecting DNA repair in bacteria, Mutation Res., 202, 387-391. Kury, G., and R.J. Crosby (1968) Studies on the development of chicken embryos exposed to cobaltous chloride, Toxicol. Appl. Pharmacol., 13, 199-206. Kusaka, Y., O. Kuwabara and M. Kitamura (1984) A case of diffuse lung disease associated with lung cancer in a hard metal worker, Ind. Health, 22, 804-808. Lauwerys, R. (1989) Toxicologie Industrielle et Intoxications Professionnelles, 3rd edn., Masson, Paris. L6onard, A. (1988) Mechanisms in metal genotoxicity: the significance of in vitro approaches, Mutation Res., 198, 321-326. Lindegren, C.C., S. Nagai and H. Nagai (1958) Induction of respiratory deficiency in yeast by manganese, copper, cobalt and nickel, Nature (London), 182, 446-448. Liquier-Milward, J. (1951) Evidence of a complex compound of cobalt with a purine base (adenine), Nature (London), 167, 1068-1069. Loeb, L.A. (1985) Apurinic sites as mutagenic intermediates, Cell, 40, 483-484. Maines, M.D., and A. Kappas (1976) Studies on the mechanism of induction of haem oxygenase by cobalt and other metal ions, Biochem. J., 154, 125-131. McKee, G.K. (1971) Carcinogenic properties of wear particles from prostheses made in cobalt-chromium alloy, Lancet, i, 750. McLean, J.R., R.S. McWilliams, J.G. Kaplan and H.S. Birnboim (1982) Rapid detection of DNA strand breaks in human peripheral blood cells and animal organs following treatment with physical and chemical agents, Prog. Mutation Res., 3, 259-263. Merian, E. (1985) Introduction on environmental chemistry and global cycles of chromium, nickel, cobalt, beryllium, arsenic, cadmium and selenium and their derivatives, in: E. Merian, R.W. Frei, W. Hardi and Ch. Schlatter (Eds.),

Carcinogenic and Mutagenic Metal Compounds, Gordon and Breach, New York, London, Paris, Montreux, Tokyo, pp. 3-32. Meyer-Bisch, C., Q.T. Phan, J.M. Mur, N. Massin, J.J. i o u l i n , D. Teculescu, B. Carton, F. Pierre and F. Baruthio (1986) Risque pulmonaire li6 h l'exposition aux poussirres de m&aux durs. Etude 6pidrmiologique, Cahiers Notes Document., 124, 319. Mitala, J.J., D.E. Mann Jr. and R.F. Gautieri (1978) Influence of cobalt (dietary), cobalamins, and inorganic cobalt saltson phenytoin- and cortisone-induced teratogenesis in mice, J. Pharm. Sci., 67, 377-380. Miyaki, M., I. Murata, M. Osabe and T. Ono (1977) Effect of metal cations on misincorporation by E. coli DNA polymerases, Biochem. Biophys. Res. Commun., 77, 854-860. Miyaki, M., N. Akamatsu, T. Ono and H. Koyama (1979) Mutagenicity of metal cations in cultured cells from Chinese hamster, Mutation Res., 68, 259-263. Mochizuki, H., and T. Kada (1982) Antimutagenic action of cobaltous chloride on Trp-l-induced mutations in Salmonella typhimurium TA98 and TA1538, Mutation Res., 95, 145-157. Moorhouse, C.P., B. Halhwell, M. Grootveld and J.M.C. Gutteridge (1985) Cobalt (II) ion as a promoter of hydroxyl radical and possible 'crypto-hydroxyl' radical formation under physiological conditions. Differential effects of hydroxyl radical scavengers, Biochim. Biophys. Acta, 843, 261-268. Morgan, L.G. (1983) A study into the health and mortality of men exposed to cobalt and oxide, J. Soc. Occup. Med., 33, 181-186. Morita, H., K. Noda and M. Umeda (1985) Mutagenicity of nickel and cobalt compounds in a mammalian cell line, Mutation Res., 147, 265-266. Mur, J.M., J.J. Moulin, M.P. Charruyer-Seinerra and J. Lafitte (1987) A cohort mortality study among cobalt and radium workers in an electrochemical plant, Am. J. Ind. Med., 12, 75-82. Murdock, H.R. Jr. (9152) Studies on the pharmacology of cobalt chloride, J. Am. Pharm. Ass., 48, 140-142. Nehe'z M., G. Berenesi, H.A. Freye, E. Mazzag, H. Scheufler and A. Selypes (1982) Teratogenic and mutagenic effects of several pigments and chemicals used in ceramics industry, a mouse experiment in vivo, Mutation Res., 97, 206-207. NIOSH (1981) Occupational Hazard Assessment. Criteria for Controlling Occupational Exposure to Cobalt, DHHS Publication 82-107, Cincinnati, OH, 95 pp. NIOSH (1987) Registry of Toxic Effects of Chemical Substances, DHHS Publication, 87-114, Rockville, MD, 5147 pp. Nishioka, H. (1975) Mutagenic activities of metal compounds in bacteria, Mutation Res., 31, 185-189. Ogawa, H.I., K. Sakata, T. Inouye, S. Jyosui, Y. Niyitani, K. Kakimoto, M. Morishita, S. Tsuruta and Y. Kato (1986) Combined mutagenicity of cobalt (II) salt and heteroaromatic compounds in Salmonella typhimurium, Mutation Res., 172, 97-104. Olivier, Ph., and D. Marzin (1987) Study of the genotoxic

27 potential of 48 inorganic derivatives with the SOS chromotest, Mutation Res., 189, 263-269. Paternain, J.L., J.L. Domingo and J. Corbella (1988) Developmental toxicity of cobalt in the rat, J. Toxicol. Environ. Health, 24, 193-200. Paton, G.R., and A.C. Allison (1972) Chromosome damage in human cell cultures induced by metal salts, Mutation Res., 16, 332-336. Pavlova, M.T., A.J. Lewis and B.J. Johnson (1979) Enhancement of adenovirns transformation by treatment of hamster embryo and lung cells with environmental pollutants, Abstr. Annual Meeting Am. Soc. Microbiol., 79, 223. Payne, L.R. (1977) The hazards of cobalt, J. Soc. Occup. Med., 27, 20-25. Prazmo, W., E. Balbin, H. Baranowski, A. Ejchart and A. Putrament (1975) Manganese mutagenesis in yeast. II. Conditions of induction and characteristics of mitochondrial respiratory deficient Saccharomyces cerevisiae mutants induced with manganese and cobalt, Genet. Res., 26, 21-29. Putrament, A., H. Baranowska, A. Eychart and W. Jachymczyk (1977) Manganese mutagenesis in yeast, Mol. Gen. Genet., 151,169-176. Robison, S.H., O. Cantoni and M. Costa (1982) Strand breakage and decreased molecular weight of DNA induced by specific metal compounds, Carcinogenesis, 3, 657-662. Rossman, T.G., and M. Molina (1986) The genetic toxicology of metal compounds: II. Enhancement of ultraviolet lightinduced mutagenesis in Escherichia coil WP2, Environ. Mutagen., 8, 263-271. Rossman, T.G., M. Molina and L.W. Meyer (1984) The genetic toxicology of metal compounds: 1. Induction of prophage in E. coil WP2 s (~), Environ. Mutagen., 6, 59-69. Saknyn, A.V., and N.K. Shabynina (1973) Some statistical data on carcinogenous hazards for workers engaged in the production of nickel from oxidized ores, Gig. Trud. Prof. Zabol., 17, 25-28. Santanu, T., B.A. Krishna, K. Dhananjoy, R. Satyabrata and D. Kalamendu (1984) Effects of cobalt (III) complexes of Schiff bases on root tip mitosis, Int. J. Environ. Stud., 22, 241-244. Sanyal, R., A.K. Gift, G. Talukder and A. Sharma (1980) Effects of cobalt and nickel on mammalian cellular systems, Biol. Bull. India, 2, 85-90. Schaaper, R.M., R.M. Koplitz, I.K. Tkeslehasvili and L.A. Loeb (1987) Metal-induced lethality and mutagenesis: possible role of apurinic intermediates, Mutation Res., 177, 179-188. Schultz, P.N., G. Warren, C. Kosso and S. Rogers (1982) Mutagenicity of a series of hexacoordinate cobalt (III) compounds, Mutation Res., 102, 393-400. Shabaan, A.A., V. Marks, M.C. Lancaster and G.N. Dufcu (1977) Fibrosarcomas induced by cobalt chloride (COC12) in rat, Lab. Anim., 11, 43-46. Sinclair, P.R., J.F. Sinclair, H.L. Bonhonsky, A.H. Gibbs and F. De Matteis (1982) Formation of cobalt protoporphyrin by chicken hepatocytes in culture, Biochem. Pharmacol., 31,993-999. Singh, I. (1983) Induction of reverse mutation and mitotic gene conversion by some metal compounds in Saccharomyces cerevisiae, Mutation Res., 117, 149-152. Sirover, M.A., and L.A. Loeb (1976) Infidelity of DNA synthe-

sis in vitro: screening for potential metal mutagens and carcinogens, Science, 194, 1434-1436. Sirover, M.A., D.K. Dube and L.A. Loeb (1979) On the fidelity of DNA synthesis. VIII. Metal activation of E. coil DNA polymerase, J. Biol. Chem., 254, 107-110. Spiejers, G.J.X., E.I. Krnajne, J.M. Berkvens and M.J. van Logten (1982) Acute oral toxicity of inorganic cobalt compounds in rats, Food Chem. Toxicol., 20, 311-314. Sprince, N.L., R.I. Chamberlin, C.A. Hales, A.L. Weber and H. Kazemi (1984) Respiratory disease in tungsten carbide production workers, Chest, 86, 549-557. Srivastava, R., N. Lefebvre and C. Onkelinx (1976) Effects of metal salts on collagen synthesis in embryonic rat calvaria, Toxicol. Appl. Pharmacol., 37, 229-235. Steinhoff, D. (1987) Personal communication. Stoner, G.D., M.B. Shimkin, M.C. Troxell, T.L. Thompson and L.S. Terry (1976) Test for carcinogenicity of metallic compounds by the pulmonary tumor response in strain A mice, Cancer Res., 36, 1744-1747. Swanson, S.A.N., M.A.R. Freeman and J.C. Heath (1973) Laboratory tests on total joint replacement protheses. J. Bone Joint Surg., 55B, 759-773. Tan, E.L., M.W. Williams, R.L. Schenley, S.W. Perdue, T.L. Hayden, J.E. Turner and A.W. Hsie (1984) The toxicity of sixteen metallic compounds in Chinese hamster ovary cells, Toxicol. Appl. Pharmacol., 74, 330-336. Taylor, A., and V. Marks (1978) Cobalt: a review, J. Hum. Nutr., 32, 165-177. Thomas, J.A., and J.P. Thiery (1953) Production 61ective de liposarcoma chez des lapins par les oligo616ments zinc et cobalt, C.R. Acad. Sci., 236, 1387-1389. Tribedi, S., A.K. Biswas, D. Koner, S. Ray and K. Dey (1984) Effect of cobalt (III) complexes of Schiff bases on root tip mitosis, Int. J. Environ. Stud., 22, 241-244. Tso, W.W., and W.P. Fung (1981) Mutagenicity of metallic cations, Toxicol. Lett., 8, 195-200. Venugopal, B., and T.K. Luckey (1978) Metal Toxicity in Mammals. 2. Chemical Toxicity of Metals and Metalloids, Plenum Press, New York and London. Weaver, J.C., V.M. Kolainsek and P.D. Richards (1956) Cobalt tumor of thyroid gland, Calif. Med., 85, 110-112. Wedrychowski, A., W.N. Schmidt and L.S. Hnilica (1986) DNA-protein crosslinking by heavy metals in Novikoff hepatoma, Arch. Biochem. Biophys., 251, 397-402. Wehner, A.P., R.H. Bush, R.J. Olson and D.U. Craig (1977) Chronic inhalation of cobalt oxide and cigarette smoke by hamsters, Am. Ind. Hyg. Ass. J., 38, 338-346. Wiberg, G.S., I.C. Murro, J.C. M6ranger, A.B. Morrison, H.C. Grice and H.A. Hoggtveit (1969) Factors affecting the cardiotoxic potential of cobalt, Clin. Toxicol., 2, 257-271. Wide, M. (1984) Effect of short-term exposure to five industrial metals on the embryonic and fetal development of the mouse, Environ. Res., 33, 47-53. Yamamoto, K., S. Inoue, A. Yamazaki, T. Yoshinaga and S. Kawanishi (1989) Site-specific DNA damage induced by cobalt (II) ion and hydrogen peroxide: role of singlet oxygen, Chem. Res. Toxicol., 2, 234-239. Zakour, R.A., T.A. Kunkel and L.A. Loeb (1981) Metal-induced infidelity of DNA synthesis, Environ. Health Perspect., 40, 197-205.