Environmental and Molecular Mutagenesis

20:277-288 (1 992)

Analysis of DNA Strand Breaks Induced in Rodent liver In Vivo, Hepatocytes in Primary Culture, and a Human Cell line by Chlorinated Acetic Acids and Chlorinated Acetaldehydes Lina W. Chang, F. B. Daniel, and Anthony B. DeAngelo Environmental Monitoring Systems Laboratory, U.S. Environmental Protection Agency, Cincinnati, Ohio (L. W.C., F.B.D.) and Health Effects Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina (A.B.D.) An alkaline unwinding assay was used to quantitate the induction of DNA strand breaks (DNA SB) in the livers of rats and mice treated in vivo, in rodent hepatocytes in primary culture, and in CCRF-CEM cells, a human lymphoblastic leukemia cell line, following treatment with tri- (TCA), di- (DCA), and mono- (MCA) chloroacetic acid and their corresponding aldehydes, tri- (chloral hydrate, CH), di- (DCAA) and mono- (CAA) chloroacetaldehyde. None of the chloroacetic acids induced DNA SB in the livers of rats at 4 hr following a single administration of 1-10 mmole/kg. TCA (1 0 mmole/kg) and DCA (5 and 10 mmole/kg) did produce a small amount of strand breakage in mice (7% at 4 hr) but not at 1 hr. N-nitrosodiethylamine (DENA), an established alkylating agent and a rodent hepatocarcinogen, produced DNA SB in the livers of both species. TCA, DCA, and MCA also failed to induce DNA strand breaks in splenocytes and epithelial cells derived from the stomach and duodenum of mice treated in vivo. None of the three chloroacetaldehydes induced DNA SB in either mouse or rat liver. The continuous exposure of mice to 5 g/L DCA in the drinking water for 7 and 14 days did not induce appreciable hepatic DNA SB (< 10% at 14 days), although peroxisome proliferation, as evidenced by an increased cyanide-insensitive palmitoyl CoA oxidase (PCO) activity, was stimulated to 490% (7

days) and 652% (1 4 days) of control. Under this protocol, DENA (0.1 g/L) produced DNA damage after both 7 days (73% of control) and 14 days (57% of control). Similarly, long-term exposure of rats (30 weeks) to 2 g/L DCA in the drinking water, a level that increased PCO activity to 364% of the control value, exhibited no DNA damage. Both the chloroacetic acids and the chloroacetaldehydes were ineffective in inducing DNA SB in cultured rat and mouse hepatocytes at concentrations below .those that yielded cytotoxicity. The chloroacetic acids were also ineffective in the CCRF-CEM cells. However, two of the chloroaldehydes, DCAA and CAA, did induce DNA SB in the CCRF-CEM cells at concentrations that did not decrease the cell viability after 2 hr of treatment. Prior incubation of DCAA and CAA with a rat S9 liver homogenote eliminated much of the DNA damaging activity. These studies provide further evidence that the chloroacetic acids lack genotaxic activity not only in rodent liver, a tissue in that they induce tumors, but in a variety of other roden tissues and cultured cell types. Two of the chloroacetaldehydes, DCAA and CAA, are direct acting DNA damaging agents in CCRF-CEM cells, but not in liver or splenocytes in vivo or in cultured hepatocytes. CH showed no activity in any system investigated. 0 1992 Wiley-Liss, Inc.

Key words: genotoxins in drinking water, chlorination by-products, DNA damage, chlaroacetic acids chloroacetaldehydes, chloral hydrate, CCRF-CEM cells, rodent liver, hepatocytes

INTRODUCTION

Received October 3, 1991; revised and accepted August 5 . 1992

The chloroacetic acids, and to a lesser extent their corresponding chloroacetaldehydes, are formed when water containing organic materials is treated with chlorine [Kopfler, 1985; Seeger et al., 19851, and both are widely found in finished drinking water in the ng/L range [Uden and Miller, 1983; Krasner et al., 19891. Recent studies [DeAngelo and McMillan, 1989; Bull et al., 1990; DeAngelo et al., 19911 have demonstrated the carcinogenicity of dichloroacetic (DCA) and trichloroacetic (TCA) acid when administered in

Address reprint requests to Anthony B. DeAngelo, Health Effects Research Laboratory, MD 68, U.S. Environmental Protection Agency, Research Triangle Park, NC 277 I 1 .

0 1992 Wiley-Liss, Inc.

Presented in part at the Annual Meeting of the American Association for Cancer Research, San Francisco, CA, May 2&27, 1989. This document has been reviewed in accordance with U . S . Environmental Protection Agency policy and approved for publication. Approval does not signify that the contents necessarily reflect the views or policy of the agency nor does the mention of trade names or commercial products constitute an endorsement or recommendation for use.

27%

Chang et al.

the drinking water to male B6C3Fl mice, and experiments recently completed in our laboratory shows that both DCA and TCA are hepatocarcinogens in female mice (unpublished observations). DCA is also carcinogenic in the male rat [DeAngelo and Daniels, 19921, and monochloroacetic acid (MCA), 2-chloroacetaldehyde (CAA) and trichloroacetaldehyde (chloral hydrate, CH) produce tumors in male mice [Daniel et al., 19911. The involvement of the chloroacetic acids as initiators of carcinogenesis, i.e., their ability to directly damage DNA, is uncertain since these chemicals have generally failed to exhibit genotoxic activity when tested in a variety of in vivo and in vitro tests [Waskel, 1978, Herbert et al., 1980; Rapsonet al., 1980; Meier and Blazak, 19911. The chloroacetaldehydes, in contrast, were capable of inducing genotoxicity in several test systems. CH was reported to be mutagenic in the Salmonella tester strain TA 100 [ Bignami et al., 1980; Haworth, 1983; Waskell, 19781 and fungi [Bronzetti et a]., 1984; Carere et al., 19851 and to induce sister chromatid exchanges in human lymphocytes [Gu et al., 19811 and aneuploidy in yeast [ Sora et al., 19871, hamster cells IDegrassi and Tanzorella, 19881. mouse spermatatogonia [Russo et al., 19841, and human lymphocytes IVagnarelli et al., 19901. CAA was a potent inhibitor of DNA synthesis [Kandala et al., 19901, formed interstrand cross-linkages with DNA in vitro [Spengler and Singer, 19881 and modified DNA conformation [Singhal and Landes, 19881. It was shown to be mutagenic in bacteria [McCann et al., 19751 and Chinese hamster cells [Huberman et al., 19751 and induced aneuploidy in Asperigilfus [Crebelli et al., 19901. The direct measurement of DNA damage is a useful parameter for assessing the genotoxicity of environmental contaminants [Kohn, 1983; Daniel et al., 1985, 1989; Garberg et al., 1988; Chang et al., 1991). One type of DNA damage, DNA strand breakage (DNA SB), often occurs as an early consequence of the interaction between genotoxic agent and mammalian DNA. A particularly effective method of quantifying DNA SB is the DNA alkaline unwinding assay (DAUA), which has been employed to measure cellular DNA damage induced by various chemical and physical agents [Rydberg, 1975; Kanter and Schwartz, 1979; Ahnstrom and Erixson, 1981; Daniel et al., 1985, 19891. Recently, Nelson and Bull [ 19881 found that TCA, DCA, and CH increased the level of DNA SB in the livers of mice and rats at doses that produced no hepatotoxic effects as assessed by serum enzyme levels. The slopes of the doseresponse curves and the order of potency differed significantly between rats and mice, suggesting that different mechanisms of DNA damage induction may be involved in the two species. They also noted that in mice the induction of DNA SB occurred before there was any evidence of peroxisome proliferation [Nelson et al., 1989). We report here the results of investigations undertaken to determine the ability of TCA, DCA, MCA, and the corresponding aldehydes, CH, dichloroacetaldehyde (DCAA) and CAA to in-

duce DNA SB in intact rodent liver, in primary cultures of rat and mouse hepatocytes and in a human cell line using a DAUA developed previously in our laboratory [Daniel et al., 19851.

MATERIALS AND METHODS Chemicals MCA, DCA, TCA, CAA, dichloroacetaldehyde diethyl acetal, CH (trichloroacetaldehyde, in water exists as chloral hydrate), sodium lauryl sarcosinate (SLS), dithiothreital (DTT), ethylenediaminetetracetic acid (EDTA), collagenase Type IV, and dexamethasone were purchased from Sigma Chemical Co. (St. Louis, MO). Methyl methanesulfonate (MMS), N-nitroso-diethylamine (DENA), N-nitrosodimethylamine (DMNA), and dimethylsulfate (DMS) were obtained from Aldrich Chemical Co. (Milwaukee, WI). Williams Medium E (WME), RPM Medium 1640, fetal bovine serum (FBS), Hepes buffer ( I M), and all other cell culture products were purchased from GIBCO (Grand Island, NY). Hydroxylapatite gel (HTP-DNA grade) was obtained from BioRad Laboratories (Richmond, CA). DNA specific Hoechst dye 33258 (Bisbenzimide) was purchased from Calbiochem Behring Corp. (La Jolla, CA). DCAA was prepared by the distillation of dichloroacetaldehyde diethyl acetal with concentrated sulfuric acid [Huntress, 19481 and was judged to be at least 98% pure by gas chromatographic analysis. All chloroacetic acid stock solutions were made up in distilled water and neutralized to pH 7.G7.4 with NaOH prior to use.

Animals and Treatments All aspects of this research were carried out under guidelines approved by the American Association of Accreditation for Laboratory Animal Care. Male Fischer 344 rats (F344-25G300 g) and male B6C3Fl mice (30-35 g) were purchased from Charles River (Wilmington, MA) and, upon arrival, were quarantined for 2 weeks. The animals were maintained on Purina Laboratory Chow and water ad libitum at 22 +- 2°C and 40-60% humidity under a 12-hr light-dark cycle. Doses of the test chemicals up to 1/4 to 1/3 the LD,, were chosen based upon earlier studies demonstrating biological activity [DeAngelo et al, 19891, carcinogenic activity [DeAngelo et al., 19911, or induction of DNA damage [Nelson and Bull, 19881. Each test animal received a single dose of test chemical in distilled water, whereas the control animals received the vehicle only. Animals were sacrificed 4 hr after treatment. The rats and mice treated for extended periods of time (1, 2, or 30 weeks) received the test chemicals in the drinking water. The concentration of the test chemical in the drinking water solutions was analyzed by capillary gas chromatography as previously described [Daniel et al., 1991; DeAngelo et al, 1991). Rats were killed by decapitation and 1.5 g of the superioranterior lobe of the liver was quickly removed, rinsed, and

DNA Strand Breaks by Chloroacetic Acids and Chloroacetaldehydes

placed in a beaker containing 12 ml of ice cold modified Seligmann Balanced Salt Solution (SBSS: 0.13 M NaCl, 2.68 mM KCl, 17.8 mM sodium acetate, 0.36 mM NaH,PO,, 0.73 mM KH,PO,, 5.55 mM D-Glucose, 0.14 mM NaHCO,, 17 mM ascorbic acid, pH 7.2) freshly supplemented with 10 mM DTT. The livers were pressed through a cold, stainless steel tissue press with 1.0 mm diameter sieve holes from Harvard Apparatus (Millis, MA) to remove connective tissues. The disrupted livers were mixed gently with 12 ml SBSS and filtered through 8-layer cotton gauze pads. A 1 ml aliquot of the suspended tissue was centrifuged at 100 X g for 5 min, resuspended in 1 ml ice cold phosphate-buffered saline (PBS) containing 20 mM EDTA and was assayed in the DAUA. Mice were killed by cervical dislocation. The entire liver and spleen were quickly removed, rinsed, and placed in 12 ml (liver) or 5 ml (spleen) ice cold SBSS. The pressed liver preparations were handled as described above for the rat liver. The mouse spleens were forced through a No. 60 mesh stainless steel screen using a syringe piston and then rinsed with the same 5 ml SBSS. After allowing the large tissue fragments to settle, the suspended splenocytes were pelleted at 200 X g for 3 min and resuspended in 2 ml of 0.83% NH,Cl for 5 min at room temperature for lysing. The cell pellet was rinsed in SBSS and resuspended in 1 ml ice cold PBS/EDTA and assayed in the DAUA. The isolation of duodenum and stomach epithelial cells from mice was modified from the method of Rydberg and Johanson [ 19751. Briefly, a 5 cm segment of duodenum was cut off and flushed with 5 ml of ice cold buffer (0.114 M Tris, 0.077 M NaCl, pH 9) at 0°C with a 10 ml syringe. The segment was then cut open and spread out on an ice cold glass petri dish. The mucus membrane from the intestine was scraped off using a spatula, mixed with 5 ml of pH 9 buffer, and then drawn thru a 12-gauge needle five times. One ml of this cell suspension was used in DAUA directly. Whole stomach was dissected out, cut open, and the contents flushed out under running tap water. The isolation and preparation of epithelial cells from the stomach for DAUA were the same as described above for the intestinal cells. Portions of livers to be used for the assay of the peroxisome marker enzyme, cyanide-insensitive palmitoyl co-enzyme A (PCO) activity were frozen in liquid nitrogen and stored at -70°C until use. Cell Culture and Treatment

The human lymphoblastic cell line, CCRF-CEM, was obtained from the American Type Culture Collection (Rockville, MD) and grown in culture RPM 1640 medium supplemented with 10% FBS, 25 mM Hepes, 20 mM L-glutamine, and 50 pg/ml gentamicin. The cells were centrifuged and suspended in culture medium without FBS at a concentration of 2 x lo6 celldml. Aliquots (2 ml) of the cell suspension was placed into 16 x 125 mm polystyrene culture tubes to which the test chemicals were added in 20 pl. The cells

279

were placed on a rotary rack and incubated at 37°C under a 5% CO, atmosphere. After 2 hr, 200 p1 of the cell suspension was removed and tested for viability using the trypan blue exclusion method. The remainder of the cells were centrifuged at 200 X g for 3 min, the cell pellets were resuspended in 1 ml of ice cold PBS/EDTA, and were immediately subjected to the DAUA. Isolation and Culture of Rodent Hepatocytes

Hepatocytes were prepared by the two-step liver perfusion method of Seglen [ 19731as modified by Williams et al. [ 19771. Isolated hepatocytes were suspended in Williams Medium E (WME) containing 10% FBS, 25 mM hepes, 2 mM L-glutamine, 30 nM dexamethasone, 0.5 pM insulin, and 50 pg/ml gentamicin. The viability of the hepatocytes as measured by the trypan blue exclusion method was routinely greater than 90%. The hepatocytes (2 X lo6cells/ml) in the WME were plated into 60 mm culture dishes and allowed to attach for 2 hr at 37°C. The unattached cells were removed and the plates were gently washed. The test chemicals in 5 ml culture medium without FBS were added, and the plates were incubated for 4 hr. Following treatment, the medium was removed and tested for lactate dehydrogenase activity (LDH; Sigma Kit No. 228) as an index of cytotoxicity. The cell cultures were rinsed two times with ice cold PBS, and the hepatocytes were removed with a rubber policeman into 3 ml of ice-cold modified SBSS freshly supplemented with 10 mM DTT and 0.1%nonidet P-40. The cells were disrupted using three gentle strokes of a Dounce homogenizer. Cell pellets were collected by centrifugation and resuspended in 1 ml PBS/EDTA for the DAUA.

DNA Alkaline Unwinding Assay (DAUA) The DAUA was performed according to the procedure of Daniel et al. [ 19851. The assay estimates the extent of DNA damage (strand breaks) via the assessment of the percentage of DNA remaining double stranded after alkaline unwinding. Briefly, the resuspended cell pellets liver and spleen in PBS, or intestinal and stomach cells in pH 9 buffer are lysed in 1 ml of 0. I N NaOH (the alkaline unwinding is achieved with the lysis). The time employed for the unwinding varies from 1 hr for CCRF-CEM cells and splenocytes to 45 min for the liver and intestinal and stomach cells. The solutions were neutralized with 1 ml of 0.1N HCl and 0.25 ml of 2% SLS/20 mM EDTA followed by 5 sec of sonication. Single(ss) and double- (ds) stranded DNA were separated on a hydroxylapatite column at 60°C. Hoechst Dye 33258 was added to a concentration of 4 X IO-’M and the amount of DNA in each fraction was determined fluorimetrically using the Shimadzu RF-5000U Spectrofluorometer (Giangarlo Scientific Co., Pittsburgh, PA) set at Ex 350 nm and Em 465 nm. The fraction (F) of ds-DNA remaining after alkaline unwinding in the sample was calculated by dividing the amount of ds-DNA by the total (ss- plus ds-) DNA.

280

Chang et al.

TABLE 1. DNA Strand Breaks in Stomach and Duodenal Epithelial Cells From B6C3F, Mice Treated With Chloroacetic Acidst Treat me nt ~~~~~

N"

Stomach

4 4 4 4 2

0.93 0.01' 0.72 k 0.06" 0.90 0.02 0.89 5 0.03 0.90 ? 0.02

Duodenum

LDH (U/L)

~~

Control 1 mmol/kg MMS 10 mmol/kg TCA 10 mmol/kg DCA 10 mmollkg MCA

* *

0.93 ? 0.01 0.82 5 0.01* 0.94 0.01 0.91 2 0.01 0.92 ? 0.01

*

308

5 93 274 67 279 5 90 338 ? 50

*

tMale B6C3F, mice were given the test chemicals in one dose p.0. and killed 4 hr later. Blood was collected and the sera prepared. Stomach and duodenal epithelial cells were isolated and DNA single-strand breaks were assayed as described in Materials and Methods. "Number of animals. hFrdction of dsDNA remaining after 45 min of alkaline unwinding. 'Mean S.E.M. for each treatment. Cell preparations were assayed in triplicate. *P =s0.01 when tested for significance by Dunnett's multiple comparison test.

*

Peroxisomal Enzyme Assay

Fig. 1. Hepatic DNA strand breaks in mice. Animals were given a single treatment p.0. of (A) tri- (TCA). di- (DCA) or mono- (MCA) chloroacetic acid or (B) tri- (CH), di- (DCAA), or mono- (CAA) chloroacetaldehyde and killed 4 hr later. N-nitrosodiethylamine (DENA) and methyl methanesulfonate (MMS) at I mmole/kg were the positive control chemicals. Preparations of liver cell nuclei were assdyed for the fraction of double-stranded DNA remaining after a 45-min alkaline unwinding period. Each bar represents the mean from four animals assayed in triplicate. P G 0.05 (*) or S 0.01 (**) when tested for significance by Dunnett's multiple comparison test.

The relationship between the F-value and the number of DNA strand breaks induced per cell (NJ as a function of the concentration of the test chemical has been determined previously for the CCRF-CEM cells [Daniel et al., 1985; 1989): Ni = - 6.1 X lo4 [ln (FT/Fo)],where FT and F,, are the F values for the treated cells and control cells respectively. (The decrease in the natural logarithm of the F-value is inversely proportional to the increase in the number of induced DNA S B . ) Due to the quantitative and qualitative differences in the activity of the DNA repair enzymes between cell and tissue types, this equation should not be used to estimate (based on F-value) the absolute number of DNA strand breaks in other cell lines. Nevertheless, the decreasing F-value indicates increasing DNA strand breakage in all cell types.

Portions of the frozen livers were homogenized ( 1 :I0 w/v) in a buffer containing 0.25 M sucrose, 0.05 M sodium EDTA, and 0.02 M Tris-HC1, pH 7.4. The homogenates were centrifuged at 800 X g for 5 min, the fatty layers were removed by aspiration, and the extracts stored at -70°C until assay. Previous work had shown that the enzyme activities in the frozen extracts did not differ significantly from the activities in liver extracts not frozen prior to assay. The cyanide-insensitive PCO activity was measured according to the method of Osumi and Hashimoto [ 19781. Statistical Analysis

The data were analyzed for statistical significance by use of the Dunnett's test [ 19641 for multiple comparison of treatment means with a control mean 119641. The level of significance was set at alpha (a)< 0.05.

RESULTS The fraction of hepatic DNA remaining double-stranded (ds-DNA) after alkaline unwinding from the liver of mice killed 4 hr following a single oral administration of each chloroacetic acid or chloroacetaldehyde is shown in Figure 1. None of the chloroacetic acids (TCA, DCA, or MCA) induced a dose-dependent increase in DNA SB (Fig. 1 A), but the highest doses (5-10 mmole/kg) of DCA and TCA did give a 7% (P < 0.05) reduction of dsDNA when compared to the control animals (F = 0.95 0.01). The chloroacetaldehydes, even when tested at the highest dose not causing death (5 mmole/kg), did not significantly reduce the F-value: 0.91 5 0.01 (CH), 0.90 k 0.01 (DCAA), and 0.92 ? 0.01 (CAA) as compared to the control value of

*

DNA Strand Breaks by Chloroacetic Acids and Chloroacetaldehydes

281

TABLE 11. Effects of Continuous Exposure of Mice and Rats to Dichloroacetic Acid in the Drinking Water on Liver DNA Strand-Breaks? Drinking water concentration

F"

Mice

PCO Activityh 7 Days

2 g/L NaCl 0.5 g/L DCA 5 g/L DCA 0.1 g/L DENA Rats

0.93 f 0.01" 0.89 f 0.01 0.89 2 0.01 0.68 2 0.02**'

2 g/L NaCl 0.05 g/L DCA 0.5 g/L DCA 2 g/L DCA

0.79 f 0.01 0.82 f 0.01 0.79 f 0.01 0.83 0.01

F

PCO Activity 14 Days

I .65 f 0. I8 1.67 0.22 10.76 f 0.78** N.D.'

*

0.86 zk 0.01 0.83 f 0.01 0.78 2 0.03* 0.49 k 0.06**

2.02 f 0.21 2.48 f 0.21 9.90 f 0.65** N.D.

30 Weeks 2.23 f 0.10 2.33 2 0.10 2.62 f 0.29 8.12 0.36**

*

*

"Male B6C3F, mice were exposed to DCA in the drinking water for 7 or 14 days; Male F344 rats were exposed for 30 weeks. Liver cell nuclei were prepared from three mice and five rats and assayed in triplicate for DNA strand breaks. Cyanide-insensitive palmitoyl COA oxidase activity was assayed in low speed cell supernatant extracts prepared from five mice and rats. See Materials and Methods. "Fraction of dsDNA remaining after 45 min alkaline unwinding. bEnzyme activity is expressed as nmoles of NAD reduced/min/mg protein. 'Mean f S.E.M. *P S 0.05 (*) and SO.01 (**) when tested for significance by the Dunnett's multiple comparison test. N.D. = not done.

0.92 t 0.005 (Fig. IB). No increase in DNA SB were measured in splenocyte DNA from these same animals by any of the test compounds (data not shown). The positive control chemicals induced significant DNA SB (P < 0.01) in the appropriate target organs, as evidenced by the lowered F-values: liver, 0.53 5 0.01 and 0.63 5 0.01; spleen, 0.58 t 0.03 and 0.69 0.03 for MMS and liver, 0.38 5 0.01 and 0.42 0.02 for DENA. The corresponding F-values observed in untreated animals were liver, 0.95 t 0.01 and 0.92 k 0.005, and spleen, 0.97 0.005 and 0.98 0.003. Similarly, analysis of the DNA SB in mice killed 1 hr after a single dose of DCA did not cause DNA damage: F = 0.85 t 0.01, 0.84 f 0.01, and 0.84 2 0.01 for 1, 5 , and 10 mmole/kg DCA, respectively. In contrast, DNA damage was readily detected 1 hr after administration of a single dose of 1 mmole/kg DENA or 1 mmolelkg MMS, F t 0.40 0.02 (P < 0.01) and 0.63 t 0.01 (P < 0.01), respectively, when compared to the control value of 0.85 0.01 (data not shown). The three chloroacetic acids also failed to induce DNA SB in epithelial cells that were isolated from the stomach and duodenum of mice 4 hr after administration of a single 10 mmole/kg dose of the three chloroacetic acids (Table 1). In contrast, 2 mmole/kg MMS decreased the F-value (increased DNA SB) for both cell types. None of the chloroacetic acids (Fig. 2A) or the chloroacetaldehydes (Fig. 2B) produced detectable DNA damage in the livers of F344 rats killed 4 hr after a single gavage treatment. TCA, DCA, CH, and DCAA were administered at several doses covering a dose range of 1-10 mmole/kg.

*

*

*

*

*

*

The two monochloro compounds, MCA and CAA, were toxic at the higher doses (5-10 mmole/kg), and the rats receiving these did not survive. As with mice, the positive control for these experiments, DENA, induced significant levels of DNA damage (0.45 f 0.01 and 0.36 k 0.01; P < 0.01) compared to the untreated animals (0.80 0.01 and 0.83 k 0.01). Chronic exposures of mice to DCA in the drinking water (mice 7 and 14 days; rats 30 weeks) did not induce appreciable hepatic DNA damage (< 10% for 5 g/L at 14 days). However, 5 g/L DCA induced cyanide-insensitive PCO activity in mice to 652% (7 days) and 490% (14 days) of the control value. Likewise, in rats exposed to 5 g/L DCA for 30 weeks PCO activity was increased to 364% of the control value. As with the single dose experiments, drinking water exposure to 0.1 g/L DENA produced DNA damage as evidenced by lowered F-values (27% at 7 days; 43% at 14 days) in the treated mice as compared to the untreated animals. In order to determine if the chloroacetic acids and chloroacetaldehydes could damage liver cell DNA by direct exposure, primary cultures of rat hepatocytes were treated for 4 hr with 1-10 mM of the three chloroacetic acids and the three chloroacetaldehydes. Identical plates of cells were treated with MMS (0.2 mM) and DENA (2 mM) as positive control chemicals. Figure 3 shows that DCA and TCA (Fig. 3A) and DCAA and CH (Fig. 3B) failed to induce DNA damage in rat hepatocytes at any of the concentrations used when compared to control values. The two monochloro compounds (MCA and CAA; Fig. 3A.B) did decrease the F-value at 5 mM (0.75 0.02, MCA; 0.37 0.02 MCAA) and 10 mM (0.37 2 0.02, MCA; 0.22 f 0.02,

*

*

*

282

Chang et al.

TCA were not cytotoxic at these concentrations, as the amount of cellular LDH released into the media by all treatments did not exceed 12% vs. 9% for the untreated cultures (Fig. 4C, D). MCA at 10 mM did decrease the F-value (0.60 + 0.06; p < 0.01) compared to the untreated cultures (0.84 0.03; Fig. 4A). However, as with the rat hepatocytes, this concentration was cytotoxic as evident by an elevated LDH release (54%). In contrast, DMNA at 2 mM induced DNA SB (F = 0.49 ? 0.01 and 0.54 0.005; P < 0.01) without evidence of cytotoxicity (LDH release < 10%). The chloroacetic acids also did not induce DNA SB in human CCRF-CEM cells following a 2-hr treatment (Fig. 5A). The number of DNA SB induced, Ni was calculated as described by Daniel et al. [ 19851. At the 10 mM MCA concentration, a small increase (7%) in the Ni measured 7.2 x 10’ vs. 4.4 x 10’ DNA SB/cell/pM for control cells (P < 0.05). None of these concentrations of chloroacetic acids decreased cell viability to below 95% (Fig. 5C). a value that did not differ significantly from the untreated cells (99%). In contrast to the acids, two of the three chloroacetaldehydes DCAA ( 1-10 mM) and CAA (0.1-1 mM) did increase the level of DNA SB in the CCRF-CEM cells (Fig. 5B). In addition, this increased level of DNA damage occurred in the absence of cytotoxicity. CH was both noncytotoxic and nongenotoxic (Fig. 5A, B). The Ni in the control cells was 6.4 X lo’, whereas treatment with I mM DCAA increased the Ni to 10.5 X 10’ (P < 0.05) and 10 mM DCAA induced 29.2 X 10’ DNA SB (P < 0.01). The cell viability (Fig. 5D) did not fall below 96% (vs. 97% for the control culture). Treatment with CAA also induced DNA SB/cell/kM: 16.3 X 10’ (P < O.Ol), 50.7 X 10’ (P < 0.01) and 94.4 x 10’ (P < 0.01) at 0.1, 1, and 10 mM, respectively. The corresponding cell viabilities were 92%. 86% (P < 0.01) and 76% (P < 0.01). Figure 6A compares the induction of DNA SB produced by treating CCRF-CEM cells with 1 mM DCAA and CAA for 2 hr with that observed after a 22-hr recovery period. A 2-hr DCAA treatment doubled Ni from 5.8 X 10’ to 11.4 x 10’ DNA SB/cell/kM (P < 0.01) without decreasing cell viability (95% vs. 97% for control; Fig. 6C), whereas CAA increased the Ni to 73.4 X 10’ DNA SB/ cell/pM (P < 0.01) and decreased cell viability to 78%. Removing the chemicals from the cells and allowing them to incubate for an additional 22 hr resulted in a large increase in the number of DNA SB in the DCAA treated cells (Ni = 84.6 X 10’ vs. 9.2 X 10’ DNA SB/cell/p.M forcontrol cells, P < 0.01; Fig. 6A). The Ni in the CAA treated cells remained high (54.4 X 10’ DNA SB/cell/kM, P < 0.01). Cell viability for both DCAA and MCAA declined over the 22-hr “repair” period to 14% and 5%, respectively, compared to a 98% viability in the control cells. After 2 hr, 0.5 mM MMS induced 35.4 X 10’ DNA SB without detectable cytotoxicity (97% viability). Over the subsequent 22-hr recovery period after the MMS was re-

*

*

Fig. 2. Hepatic DNA strand breaks in rats. Animals were given a single treatment p.0. of (A) tri- (TCA), di- (DCA) or mono- (MCA) chloroacetic acid or (B) tri- (CH), di- (DCAA). or mono- (CAA) chlorodcetdldehyde and killed 4 hr later. N-nitrosodiethylamine (DENA)at 1 mmolelkg was the positive control chemical. Preparations of liver cell nuclei were assayed for the fraction of double-stranded DNA remaining after a 45-min alkaline unwinding period. Each bar represents the mean from four animals assayed in triplicate. P 0.01 (**) when tested for significance by Dunnett’s multiple comparison test.

CAA) control valves were 0.81 ? 0.01 and 0.89 & 0.01. However, the strand breaks appeared to occur secondarily to cytotoxicity as demonstrated by the concurrent LDH released into the culture medium (49-100%. MCA, Fig. 3C; 77-82%, CAA, Fig. 3D) relative to the control cultures (13-25%). The reduced F-values for MMS (0.55 2 0.01; 0.25 k 0.03) and DENA (0.61 0.01; 0.79 +- 0.02) indicate that a significant number of DNA SB occurred with these compounds at noncytotoxic concentrations ( 15-27% of the LDH released). As with rat cultures, treating primary cultures of mouse hepatocytes with TCA (Fig. 4A) or DCA (Fig. 4B) for 4 hr did not induce DNA SB. The F-values for the cells treated with TCA were 0.81 ? 0.01 (0.1 mM), 0.82 t 0.03 ( 1 mM) and 0.82 t 0.02 (10 mM) compared to 0.84 0.03 for the control cells. The corresponding F-values ranged between 0.85 0.01 and 0.87 t 0.01 for 1-20 mM DCA compared to 0.87 0.005 for the untreated cells. DCA and

*

*

*

_+

DNA Strand Breaks by Chloroacetic Acids and Chloroacetaldehydes

283

Fig. 3. DNA strand breaks in rat hepatocytes. Primary cultures of hepatocytes were prepared as described in Materials and Methods and treated for 4 hr with tri- (TCA), di- (DCA) and mono- (MCA) chloroacetic acid or tri-, (CH), di- (DCAA), and mono- (CAA) chloroacetaldehyde. N-nitrosodiethylamine (DENA; 2 mM) and methyl methanesulfonate (MMS; 0.2 mM) were the positive control chemicals. The cultures were assayed for the fraction of double-stranded DNA remaining after a 45-min alkaline un-

winding period (A, chloroacetic acids; B, chloroacetaldehydes) and the amount of lactate dehydrogenase, LDH, (C, chloroacetic acids; D,chloroacetaldehydes) released into the media. Each bar represents the mean from two experiments; each test concentration was assayed in triplicate. P 6 0.05 (*) or sz 0.01 (**) when tested for significance by Dunnett's multiple comparison test.

moved, the DNA SB were largely eliminated (Ni = 15.2 X lo3 vs. 9.2 X lo3 DNA SB/cell/pM forcontrol cells, P < 0.01) and the viability in these cultures remained high (99%). Preincubation of DCAA and CAA ( 1 mM) with rat liver S9 homogenate fraction for 30 min before adding them to the CCRF-CEM cells effectively eliminated the DNA damaging capabilities of the two aldehydes (Fig. 6B). For example, 1 mM CAA induced 14.4 X lo3 DNA SB/cell/pM compared to 7.2 X lo3 (P < 0.01) for the untreated cultures. Neither of the two chloroacetaldehydes lowered the cell viability measured at 2 hr after treatment (Fig. 6D; 92% and 95% for DCAA and CAA, respectively, vs. 99% for control cells). After a 22-hr incubation following the removal of the chemicals from the cells, only CAA showed a slight increase in the number of DNA SB (Ni = 17.5 X lo3 vs. 1 1.4 X lo3 DNA SB/cell/pM for control cells; Fig. 6B). No decrease in cell viability was seen at 24 hr for DCAA (96%) and CAA (91 %) when compared to the control cells

(98%; (Fig. 6D). In contrast, preincubation with rat S9 fraction had no effect on the ability of MMS to induce DNA SB at 2 hr (Ni = 47.4 X lo3 vs. 7.2 X 1 0 3 DNA SB/ cell/pM for control cells, P < 0.01) or the ability of the CCRF-CEM cells to repair the DNA damage 22 hr after the chemical was removed (Ni = 14.4 X lo3 vs. 11.4 X 10' DNA SB/cell/p,M; Fig. 6B). Treatment with the MMS resulted in no loss of cell viability at 2 hr (92%) or 22 hr (94%).

DISCUSSION Previously we demonstrated that the DAUA employed here [Daniel et al., 1985, 1989; Changet al., 19911 is useful for quantifying DNA SB induced by chlorinated chemicals in CCRF-CEM cells, a human lymphoblastoid line. The assay can also be used to measure DNA damage in cells obtained from intact animals treated with these agents [Daniel et al., 1989; Chang et al., 19911. These experiments

284

Chang et al.

Fig. 4. DNA strand breaks in mouse hepatocytes. Primary cultures of hepatocytes were prepared as described in the Materials and Methods section and treated for 4 hr with tri- (TCA), di- (DCA), and mono- (MCA) chloroacetic acid. N-nitrosodimethylamine (DMNA: 2mM) was the positive control chemical. The cultures were assayed for the fraction of double-

stranded DNA remaining after a 45-min alkaline unwinding period ( A and B) and the amount of lactate dehydrogenase (LDH: C and D) released into the media. Each bar represents the mean from two experiments: each test concentration was assayed in triplicate. P S 0.01 (**) when tested for significance by Dunnett’s multiple comparison test.

demonstrate that chlorinated acetic acids and chlorinated acetaldehydes, several of which have been shown to be hepatocarcinogenic to rodents, fail to induce (or induced marginal levels of) DNA SB in either the target organ (liver) or other organs in rats or mice given the compounds at pharmacologic (and hepatocarcinogenic) dose levels and under both acute or chronic protocols. In addition, the chlorinated acetic acids were also inactive when applied to the aforementioned CCRF-CEM cells; however, two of the chlorinated aldehydes, CAA and DCAA, appeared to have DNA strandbreaking activity in these cells. In both cases, however, it was difficult in these experiments to unambiguously separate frank genotoxicity from those DNA damaging actions that might be induced by secondary effects such as cytotoxicity and subsequent cell death. For example, CAA and DCAA were extremely cytotoxic to both CCRFCEM cells as well as to primary hepatocyte cultures. In our hands, DCA and TCA did not show activity in rodent liver. The small amount of DNA damage (< 7%) observed at 4 hr after treatment with 5 and 10 mM/kg, although statistically significant, we interpret to be without

biological relevance since strand breaks were not observed after I hr and were not dose related. These results could not be confirmed in further studies (data not shown). Increasing the doses above 10 mM/kg produced frank toxicity by 4 hr and the death of some treated mice. These results stand in contrast to that observed earlier by Bull and coworkers [Nelson and Bull, 1988; Nelson et al., 19891. They reported dose-dependent increases in rodent liver DNA SBs induced by DCA, TCA, and CH beginning at doses s 1 mmol/kg, levels 10-100 times lower than were used in this study. It should be pointed out that these workers used a different DNA unwinding assay [Morris and Shertzer, 19851 than that employed in this study. In addition, we did not observe the induction of DNA SB in any of the systems employed in our studies by CH. This compound has been shown to exhibit mutagenic and clastogenic responses in several in vitro test systems [Waters and Black, 19761. Our failure to detect DNA SB in the mice administered TCA are in agreement with recently published data by Styles et al. [19911 who were unable to detect liver DNA SB in

DNA Strand Breaks by Chloroacetic Acids and Chloroacetaldehydes A

285

B 70

o( 10-3)

(~10-3)

I

-

40 Ni 30 20 60

l00

1001

I

1

2

3

4

o 6

e

7

0

a

-

80 -

TCA

-&-

DCA

4

MCA

0

MU8

80

-

Ni 40-

..

I

0

I

T Y

CAA

0

MMS

1

2

9

4

7

6

5

8

9

10

mM

C

A B I

DCM

-e-

mw

mM

y

CH

D

-

80%80%40%-

TCA

-&-

DCA

%-

MCA

0

MMS

I

'

1

2

'

'

'

I

'

'

'

3

4

5

8

7

8

0

10

mM

4

I

cn

4D C M

I

T y

20%0%' 0

A B

4 c M

yi

6

MMS

20%

0%' 0

"

'

1

3

2

'

4

'

5

' 6

'

'

'

'

7

8

0

10

mM

Fig. 5. DNA strand breaks in cultured human cells. CCRF-CEM cells were grown and maintained as described in Materials and Methods and treated for 2 hr with tri- (TCA), di- (DCA) and mono- (MCA) acetic acid or tri- (CH), di- (DCAA), and mono- (CAA) chloroacetaldehyde. Methyl methanesulfonate (MMS) was the positive control chemical. The cultures were assayed for the fraction of double-stranded DNA remaining after a

I-hr alkaline unwinding period and the number of DNA strand breaks (Ni) was calculated (A, chloroacetic acids; B, chloroacetaldehydes).Cell viability was assayed by the trypan blue exclusion method (C, chloroacetic acids; D, chloroacetaldehydes). Each bar represents the mean from three incubations. P 6 0.05 (*) or S 0.01 (**) when tested for significance by Dunnett's multiple comparison test.

mice following I , 2, and 3 day dosings of 500 mg/kg TCA. The differences between our work and that of Nelson et al. [ 19891, notwithstanding, would appear that within the detection limits of the DAUA used here the subject compounds are not producing DNA strand breakage since treatment of additional animals with the direct- (MMS) and indirectacting (DENA) alkylating agents clearly resulted in DNA SB in the absence of cytotoxic effects. Previously, in experiments employing radiolabeled alkylating agents, we demonstrated the DAUA will routinely detect damage levels of 100-1,OOO DNA SB per cell and that in CCRF-CEM cells at least we observe approximately one strand break per 10 DNA methylations [Daniel et al., 19891. It is possible, but not likely, that all of these chlorinated agents are producing a form of DNA damage (e.g., stable, unrepairable adducts) that are not readily detected by the DAUA employed here. Our observation of DNA SB by CAA is not unexpected as this compound is a strong mutagen [Huberman et al., 1975;

McCann et al., 1975; Bartsch et al., 19761 and has been shown by several laboratories, including our own, to be highly DNA reactive [Barrio et al., 1975; Spengler and Singer, 1988, Kandala et al., 19901. The work presented here would suggest that the analog, DCAA, might also be a mutagen and form adducts with DNA. Obviously, more research into the mechanism of action of this class of chemicals, some or all of which are rodent hepatocarcinogens, is required. Previously our laboratory [DeAngelo et al., 19891 and others [Elliot and Elcombe, 19871 have shown that the DCA and TCA are potent peroxisome proliferators. The latter have been proposed as a unique class of rodent hepatocarcinogens [Reddy et al., 19801; however, their mechanism for this action is not known [DeAngelo et al., 19891. One common theory is that the heightened production of hydrogen peroxide, a mild stable oxidant, and its subsequent metabolism (resulting from the enhanced levels of peroxisomal enzymes) gives

Chang et al.

286

-SQ 10-3)

A

+SQ 10-3)

B

(X

(X 24 HR

2 HR

100

I

80

8o

80

24 HR

2 HR

100

tl

80

Ni

Ni 40

40

20

20

-

n CONTROL MMS

c

DCM

CAA

CONTROL MMS

1

D 2HR

v

7 L ;

Y

0

CAA

-59

100%

I A

DCAA

80%

80%

J ..

40*

lCAA MCAA

20%

0% JNTROLMMS

DCAA MCAA

CONTROLMMS

CONTROL YMS

DCM

CAA

CONTROL YYS

DCAA

CAA

+SQ 2 HR

24 HR

v

I A B I L I T Y CONTROLMMS

DCAA MCAA

CONTROLMMS

DCAA MCAA

Fig. 6. The effect of S9 preincubation on the chloroacetaldehyde induction of DNA strand breaks in CCRF-CEM cells. The cells were treated for 2 hr with 1 mM di- (DCAA) or mono- (CAA) chloroacetaldehyde with (A and C) and without (B and D) a prior 30-min incubation of the test chemicals with rat liver S9 fraction. Methyl methane sulfonate (MMS; 0.5 mM) was the positive control chemical. Following treatment, one-half of the cultures was assayed for the fraction of double-stranded DNA remaining after a I-hr alkaline unwinding period from which the number of DNA

strand breaks was calculated (A and B). Cell viability ( C and D) was assayed by the trypan blue exclusion method. The remaining cells were removed from the test media by centrifugation. resuspended in media minus the test chemicals, and incubated for an additional 22 hr after which DNA stand breaks and cell viability were assayed. Each bar represents the mean from three incubations. P G 0.01 (**)when tested for significance by Dunnett’s multiple comparison test.

rise to more reactive, DNA-damaging species such as the hydroxy radical [Chance et al., 19791. These species might be expected to initialize DNA damages such as strand breaks and cross-links that might trigger amelorative cellular repair processes leading, in turn, to gene mutation and chromosomal aberrations [Cadet and Teoule, 19781. Our studies do not rule out such mechanisms of hepatocarcinogenicity for these chlorinated compounds, but they do show that the process of peroxisome proliferation per se did not result in DNA strand breakage. In the experiments described here, the administration of 5 g/L DCA in drinking water to rats and mice clearly induced peroxisomes at 7 (652%) and 14 days (490%) in mice and 30 weeks (364%) in rats with no induction of detectable DNA SB. It should be noted that this dose is hepatocarcinogenic in the males of both species [DeAngelo et al., 1991; DeAngelo and Daniel,

19921. Further, the administration of 0.1 g/L DENA, also an hepatocarcinogen, did not induce peroxisomes, whereas it did induce DNA SB. Similar conclusions have been expressed by Elliot and Elcombe (1987) who reported that clofibrate, methyl clofenopate, and di-(2-ethylhexy1)phthalate, also powerful peroxisome proliferators, do not induce DNA SB. Many chemicals that induce peroxisome proliferation also possess mitogenic and hepatotrophic properties [Grasso and Sharratt, 19911. Cell proliferation plays a critical role in the carcinogenic process [Goodman et al., I99 1 1. We have reported that DCA and, to a lesser extent, TCA are effective hepatotrophic agents [DeAngelo and Daniel, 1990; DeAngelo et al., 19911. The degree to which liver enlargement was due to hypertrophy or hyperplasia is not firmly established. However, we found that DCA and TCA did not

DNA Strand Breaks by Chloroacetic Acids and Chloroacetaldehydes

significantly increase the hepatocyte labeling index during the chronic exposure periods. Newer studies [Carter et al., 1991; DeAngelo et al., 19911 are showing that DCA and TCA significantly depress cell division during the first 14 days of treatment and provide evidence that hypertrophy underlies the early liver weight gains. In summary, our studies show that the chlorinated acetic acids and the chlorinated acetaldehydes did not produce DNA SB in rodent liver cells even when administered at levels up to 50&1,000 times higher that those seen in finished drinking water. CAA and DCAA do appear to be capable of producing DNA SB in CCRF-CEM cells but not in fresh primary hepatocytes (at noncytotoxic concentrations). One major difference between these cell types is that CCRF-CEM cells do not contain high levels of P450 oxidative enzymes [Kanter and Schwartz, 19791, whereas the primary hepatocytes do. Further evidence to support this view is the observation that the incubation of the two chlorinated acetaldehydes with a rat liver homogenate before application to the CCRF-CEM cells destroys the DNA damaging potential of the two compounds. More work is required to elucidate the biochemical action of these chemicals. In addition to the carcinogenic activities of DCA and TCA, our laboratory has recently shown that MCA [Daniel et al., unpublished observations], CAA, and CH [Daniel et al., 19911 also induced liver cancer in the male B6C3F1 mouse. In addition, the pattern of lesions that appear with all five compounds is very similar. For example, in all cases we observe the occurrence of hepatocellular hyperplastic nodules in addition to clearly defined hepatocellular tumors. We have recently summarized our evidence for our contention that these hyperplastic nodules are the precursors to the tumors, e.g., hepatocellular adenomas and hepatocellular carcinomas [Richmond et al., 19911. The primary reason for this conjecture is the observation that many of these proliferative nodules contain ‘‘nests” of cells that respond in situ to five immunohistochemical markers for neoplasia. The frequency of multiple expression of these markers in the nests of cells is similar to that frequency of multiple expression seen in the frank tumors.

ACKNOWLEDGMENTS The authors thank Mr. Paul Mystkowski for his excellent technical assistance and Mr. Carl Chavis for preparing the DCAA.

REFERENCES Ahnstrom G, Erixon K ( 198I ): The measurement of strand breaks by DNA unwinding in alkali and hydroxylapdtite chromatography. In Friedberg EC, Hanawalt PC (eds): “DNA Repair: A Laboratory Manual of Research Procedures.” New York: Marcel Dekker, pp. 1403-1418. Barrio J, Secrist J, Leonard N (1972): Fluorescent adenosine and cytidine derivatives. Biochem Biophys Res Comm, 46(2):597-604. Bartsch H, Malaveille C. Barbin A, Bresil H, Tomatis L, Montesano R

287

(1976): Mutagenicity and metabolism of vinyl chloride and related compounds. Environ Health Perspect 17:193-1 98. Bignami M, Conti G. Conti L, Crebelli R, Misuraca F. Puglia A, Randazzo R, Sciandrello G, Carere A (1980): Mutagenicity of halogenated aliphatic hydrocarbons in Sulmonellu ryphimurium. Strepromyces coelicolcir and Aspergillus nidulans. Chem Biol Interact 30:%23. Bronzetti G, Galli A, Corsi C, Cundari E, Del Carratone RD, Nieri R. Paolini M (1984): Genetic and biochemical investigations on chloral hydrate in vitro and in vivo. Mutat Res 141: 19-22. Bull RJ. Sanchez IM, Nelson MA. Larson JL, Lansing AJ (1990): Liver tumor induction in B6C3FI mice by dichloroacetate and trichloracetate. Toxicol63:341-359. Cadet J, Teoule R (1978): Comparative study of oxidation of nucleic acid components by hydroxyl radical, singlet oxygen, and superoxide anion radicals. Photochem Photobiol28:661-667. Carere A, Conti G. Conti L, Crebelli R (1985): Assays in Aspergillus niduluns for induction of forward mutation in haploid strain 35 and for mitotic nondisjunction, haploidization and crossing over in diploid strain PI. Prog Mutat Res 5:307-3 12. Carter JH. Carter HW, DeAngelo AB. Daniel FB (1991): Morphometric evaluation of the short term effects of chloroacetic acids on hepatomeglia in the male B6C3FI mouse. Proc Amer Assoc Cancer Res 32:84. Chance B, Sies H, Boveris A (1979): Hydroperoxide metabolism in mammalian organs. Physiol Rev 595274105, Chang LW. Daniel FB, DeAngelo AB (1991: DNA strand breaks induce in cultured human and rodent cells by chlorohydroxyfuranones. mutagens isolated from drinking water. Teratotgen Carcinogen and Mutagen 11:103-114. Daniel FB, DeAngelo AB, Stober JA. Olson FR, Page NP (1991): Hepatocarcinogenicity chloral hydrate and 2-hloroacetaldehyde in the male B6C3FI mouse. Fund Appl Toxicol (in press). Daniel FB. Chang LW, Schenck KM, DeAngelo AB, Skelly MF (1989): The further application to hazards identification for contaminants from environmental samples. Toxicol lndust Health 5:647465. Daniel FB, Haas DL, Pyle SM (1985): Quantation of chemically induced DNA strand breaks via an alkaline unwinding assay. Anal Biochern 144:390-402. DeAngelo AB, McMillan LP ( 1989): Carcinogenicity of chlorinated acetic acids. In: J o k y , RL et al. (eds): “Water Chlorination: Chemistry, Environmental Impact and Health Effects,” Vol. 6. Chelsea, MI: Lewis, pp 193-199. DeAngelo AB, Daniel FB (1990): Comparative carcinogenicity of dichloroacetic (DCA) and trichloroacetic (TCA) acid in the male B6C3FI mouse. Toxicologist 10:148. DeAngelo AB, Chavis C (1991): Early changes in liver DNA synthsis and hepatocyte turnover during dichloroacetic acid (DCA) and trichloroacetic acid (TCA) carcinogenesis. Proc Amer Assoc Cancer Res 32:84. DeAngelo AB, Daniel FB (1992): An evaluation of the carcinogenicity of the chloroacetic acids in the male F344 rat. Toxicologist, 12:206. DeAngelo AB, Daniel FB, McMillan L, Wernsing P, Savage Jr. RE (1989): Species and strain sensitivity to the induction of peroxisome proliferation by chloroacetic acids. Toxicol Appl Pharmacol 101:285-298. DeAngelo AB. Daniel FB. Stober JA, Olson, GR (1991): The carcinogenicity of dichloroacetic acid in the male B6C3FI mouse. Fund Appl Toxicol 16:337-347. Degrassi F and Tanzarella, C (1988): Immunofluorescent staining of kinetochores in rnicronuclei: A new assay for detection of aneuploidy. Mutat Res 203:339-345. Dunnett CW (1964): New tables for multiple comparisons with a control. Biometrics 20:482491. Elliot BM. Elcombe CR ( I 987): Lack of DNA damage or lipid peroxidation measured in vivo in the rat liver following treatment with peroxisome proliferators. Carcinogenesis 8: 12 13-12 18.

288

Chang et al.

Garberg P, Akerblom E-L, Bolcsfoldi G ( 1988): Evaluation of a genotoxicity test measuring DNA-strand breaks in mouse lymphoma cells by alkaline unwinding and hydroxylapatite elution. Mutat Res 203:155-176. Goodman JI, Ward JM, Popp JA. Launig JE. Fox. TR (1991): Mouse liver carcinogenesis: Mechanisms and Relevance. Fund Appl Toxicol 17:65 1 4 5 5 . Grasso P, Sharratt M (1991): Role of persistent. non-genotoxic tissue damage in rodent cancer and relevance to humans. Annu Rev Pharmacol Toxicol3 1:253-287. Gu WZ, Sele B, Jalben P, Vincent M. Chmara D, Faure J, Marka C. ( 198I ): Induction of sister chromatid exchange by trichloroethylene and its metabolites. Toxicol Eur Res 3:63-67. Haworth S, Lawlor T. Mortelmans K . Speck W. Zeiger E. (1983): Salrnonella mutagenicity results for 250 chemicals. Environ Mutagen Suppl 1:3-142. Herbert V. Gardner A. Coleman N (1980): Mutagenicity of dichloroacetate. an ingredient of some formulations of pangamic acid (Tradename “vitamin B15”). Amer J Clin Nutr 33:l 179-1 182. Huberman E. Bartsch H, Sachs L ( 1975): Mutation induction in Chinese hamster V79 cells by two vinyl chloride metabolites, chloroethylene oxide and 2-chloroacetaldehyde. Int J Cancer 15539. Huntress EH ( 1948): “The Preparation. Properties. Chemical Behavior, and Identification of Organic Chlorine Compounds.” New York: John Wiley & Sons, pp 6 2 M 2 1 . 6 5 8 . Kandala JC, Mrema JEK, DeAngelo A. Daniel FB, Guntaka RV ( 1990): 2-chloroacetaldehyde and 2-chloroacetal are potent inhibitors of DNA synthesis in animal cells. Biochem and Biophys Res Comm. I67:457463. Kanter PM, Schwartz HS (1979): A hydroxylapatitc batch assay for quantitation of cellular DNA damage. Anal Biochem 97:77-84. Kohn NH ( 1 983): The significance of DNA damage assays in toxicity and carcinogenicity assessment. Ann NY Acad Sci 407:10&1 18. Kopfler FC, Ringhand HP. Coleman WE, Meier JR (1985): Reactions of chlorine in drinking water with humic acids and in vivo. In Jolly RL (ed): “Water Chlorination, Environmental Impact and Health Effects,”Vol. 5. Chelsea, MI: Lewis. pp 161-173. Krasner SW. McGuire MJ. Jacangelo JG. Patania NL. Reagen KM. Aieta EM (1989): The occurrence of disinfection by-products in US drinking water. J Amer Water Works Assoc 81:41-53. McCann J , Simmon V. Streitwieser D, Ames BN (1975): Mutagenicity of chloroacetaldehyde. a possible metabolic product of I .2-dichloroethane (ethylene dichloridc). chloroethanol (ethylene chlorohydrin). and cyclophosphamide. Proc Nat Acad Sci, USA 72(8):3 19@3 193. Meier JR. Blazak WR (1991): Evaluation of genotoxicity oldichloroacetic and trichloroacetic acid. Mutat Res (in press). Morris SR. Shertzer HG (1985): Rapid analysis of DNA strand breaks in soft tissues. Environ Mutagen 7:871-880. Nelson MA, Bull RJ (1988): Induction o f strand breaks in DNA by trichloroethylene and metabolites in rat and mouse liver in vivo. Toxicol Appl Pharmacol94:45-54. Nelson MA, Lansing AJ. Sanchez IM, Bull RJ, Springer DL (1989): Dichloroacetic acid and trichloroacetic acid-induced DNA strand breaks are independent of peroxisome proliferation. Toxicol 58:239-248.

Osumi T. Hashimoto T (1978): Enhancement of fatty acyl-Coa oxidizing activity in rat liver peroxisomes by di-(2-ethylhexyl)phathalate.J Biochem 83: 1361-1365. Rapson WH, Nazar MA, Butsky VV (1980): Mutagenicity produced by aqueous chlorination of organic compounds. Bull Environ Contam Toxicol24:590-597. Reddy JK. Azamoff DL. Hignite CE ( 1980): Hypolipidemic hepatic peroxisome proliferators form a novel class of chemical carcinogens. Nature 283:397-398. Richmond RE. DeAngelo AB. Potter CL. and Danie FB (1991):The role of hyperplastic nodules in dichloroacetic acid-induced hepatocarcinogenesis in B6C3FI male mice. Carcinogenesis 12: 1383-1 387. Russo A, Pacchierotti F. Metalli P (1984): Nondisjunction induced in mouse spermatogenesis by chloral hydrate. a metabolite of trichloroethylene. Environ Mutagen 6:695-703. Rydberg B ( 1975): The rate of strand separation in alkali in alkali of DNA of irradiated mammalian cells. Radiation Res 61 :274-287. Rydberg B, Johanson KJ (1975): Radiation-induced DNA strand breaks and their rejoining in crypt and villous cells of the small intestine of the mouse. Radiation Res 64:281-292. Seeger DR. Moore LA, Stevens AA (1985): Formation of acidic trace organic byproducts from chlorination of humic acids. In Jolly RL (ed): “Water Chlorination. Environmental Impact and Health Effects.” Vol. 5. Chelsea, MI: Lewis, pp 859-873. Seglen PO (1973): Preparation of rat liver cells. 111. Enzymatic requirements for tissue dispersion. Exp Cell Res 82:391-398. Sora S, Carbone MLA ( 1987): Chloral hydrate, methymercury hydroxide and ethidium bromide affect chromosomal segregation during meiosis of Succharomvces cerevisiue. Mutat Res 190: 13-1 7. Spengler SJ. Singer B (1988): Formation of interstrand cross-links in chloroacctaldehyde-treated DNA demonstrated by ethidium bromide fluorescence. Cancer Res 48(7):48044806. Styles JA. Wyatt I , Coutts C (199 I ): Trichloroacetic acid: studies on uptake and effects on hepatic DNA and liver growth in mouse. Carcinogenesis 12:1715-1719. Uden PC, Miller JW (1983): Chlorinated acids and chloral in drinking water. J Am Water Works Assoc 75524-527. Vagnarelli P., DeSario A, DeCarli L. (1990): Aneuploidy induced by chloral hydrate detected in human lymphocytes with the Y97 probe. Mutagen. 5591-592. Waskel L (1978): Study of the mutagenicity of anesthetics and their metabolites. Mutat Res 57:141-153. Waters EM, Black SA (1976): Trichloroethylene I & 11. An abstracted literature collection, 1907-1976. Report ORNLITIRC-7612. Springfield, VA National Technical Information Service. Williams GM. Bermudez E, Scaramuzzino D (1977); Rat hepatocyte primary cell cultures. 111. Improved dissociation and attachment techniques and the enhancement of survival by culture medium. In Vitro 13:809-817.

Accepted byC.S. Aaron