Final Report: Health Risk of the Trihalomethanes Found in Drinking Water Carcinogenic Activity and Interactions

EPA Grant Number: R825384
Title: Health Risk of the Trihalomethanes Found in Drinking Water Carcinogenic Activity and Interactions
Investigators: Pereira, Michael A.
Institution: Medical College of Ohio
EPA Project Officer: Packard, Benjamin H
Project Period: January 9, 1997 through January 8, 2000
Project Amount: $442,347
RFA: Drinking Water (1996) RFA Text |  Recipients Lists
Research Category: Drinking Water , Water


The use of chlorine to disinfect water for the purpose of drinking produces in the finished product low levels of various disinfection by-products (DBPs) with the trihalomethanes (THM) and haloacetic acids (HAA) being the most common. Although the concentrations of these DBPs are low, their consumption in chlorinated drinking water does result in daily and chronic exposure to a mixture of halogenated organics. Furthermore, interest in THM and HAA stems from their carcinogenic activity in laboratory animals. The extrapolation to humans of the carcinogenic activity of THM is complicated by their activity being demonstrated when administered by oral gavage in corn oil. Therefore, the extrapolation of the results of chronic bioassays to human exposure from drinking water not only has to be assessed across species but also across vehicle and route of administration. Also, potential interactions between and among the THM and HAA exist that can result in additive, synergistic, or antagonistic effects and complicate the extrapolation.

The objective of this grant was to decrease the uncertainties in the risk assessment of the DBPs especially related to the use of results from animal studies. The understanding of the mechanism by which the THM and HAA induce cancer would be used to decrease the uncertainties. Therefore, we investigated the proposed mechanism of DBPs that includes enhancement of cell proliferation and decreased methylation (hypomethylation) of DNA and genes (i.e., the c-myc protooncogene). Hypomethylation is an early event in human and animal cancers resulting in decreased control of the expression (activity) of genes. Our working hypothesis was that the DBPs cause hypomethylation of DNA by preventing the methylation of newly synthesized daughter strands of DNA. DNA is methylated by DNA methyltransferase (DNA MTase) with s-adenosyl-methionine (SAM) as the methyl donor to form 5-methyl-cytosine. This requires that the DBP both enhance cell proliferation and produce toxicity. The impact of the route of administration on the mechanism of DBPs was included in our investigations. Thus, we compared the drinking water exposure, the route of human exposure by oral gavage, and the route used in the carcinogenesis bioassays of the THM. Because the THM and HAA are present in the mixture of DBPs in drinking water, we determined the effect of concurrent administration of chloroform (representative of the THM) on the carcinogenic activity and mechanism of dichloroacetic acid (DCA) and trichloroacetic acid (TCA), representatives of the HAA.

Summary/Accomplishments (Outputs/Outcomes):

The ability of THM and HAA to induce cell proliferation, hypomethylation of the c-myc protooncogene, and toxicity was determined. The c-myc protooncogene was chosen because increased expression is associated with increased cell proliferation, and its expression is, in part, controlled by methylation. Dichloroethane (DCA), trichloroethane (TCA), and trichloroethylene (TCE) were demonstrated to produce hypomethylation of DNA and the promoter region of the c-myc gene. DCA- and TCA-promoted tumors also contained hypomethylated DNA and c-myc. The expression of the mRNA and protein of c-myc and c-jun (another protooncogene) was increased in mouse liver by DCA, TCA, and trichloroethylene and in tumors produced by DCA and TCA. DNA MTase activity was increased in tumors produced by DCA and TCA while decreased in non-involved liver. Mouse and human tumors have hypermethylation of tumor suppressor genes that results in the down regulation of the mRNA of these genes. The increased DNA MTase activity in DCA- and TCA-promoted tumors could increase the susceptibility of tumor suppressor genes to hypermethylation, furthering the neoplastic transformation. In summary, liver tumors from DCA- and TCA-treated mice contained decreased methylation in the promoter regions and increased expression of the mRNA and protein of the c-jun and c-myc protooncogenes in the presence of increased DNA MTase activity.

We propose that THM, DCA, TCA, and TCE induced hypomethylation of genes by decreasing the availability of SAM. The concentration of SAM could be decreased by the toxicity of the DBP resulting in a decrease in ATP and/or by a decrease in glutathione that is utilized in the metabolism or toxicity of the DBP. Methionine is proposed to prevent the decrease in the concentration of SAM and thus prevent the hypomethylation of the genes. We have demonstrated that methionine prevented the DCA-, TCA-, and TCE-induced decrease in the methylation of c-jun and c-myc genes and their induced increase in the level of the mRNA and protein of the two protooncogenes. The affect of methionine was dose dependent. The prevention of DCA-, TCA-, and TCE-induced DNA hypomethylation by methionine supports the hypothesis that the DBP caused hypomethylation by depleting the availability of SAM. Hence, methionine would prevent DNA hypomethylation by maintaining the level of SAM. Furthermore, the results suggest that the dose of DCA, TCA, and TCE must be sufficient to decrease the level of SAM in order for these carcinogens to be active.

In another study, we compared the effect of four THMsCchloroform, bromodichloromethane, chlorodibromomethane, and bromoformCon cell proliferation, toxicity, and DNA methylation in the liver of female B6C3F1 mice at dose levels used in the National Cancer Institute/National Toxicology Program bioassay and when administered in drinking water at 80 percent saturation. When administered by oral gavage to female B6C3F1 mice, at the high-dose level, all four THMs enhanced cell proliferation. Chloroform and bromoform were the most active. Interestingly, there appeared to be a break in the dose-response curves, especially for chloroform and bromoform and to a lesser extent by bromodichloromethane, suggesting a threshold. When administered in the drinking water only, bromoform and chloroform produced a statistically significant increase in cell proliferation. Furthermore, the activity of the THMs administered at 80 percent saturation in the drinking water was similar to the activity of the low dose administered by gavage (i.e., much less than the high dose by gavage).

With respect to liver toxicity, the high dose of all four THMs administered by gavage induced toxicity; chloroform and bromodichloromethane were the most active. The low dose was less active especially for chloroform and bromodichloromethane. The THMs administered in drinking water induced liver toxicity with a severity similar to their low dose administered by gavage, suggesting that exposure to THM in drinking water would be no more carcinogenic than the low dose used in the oral gavage bioassays.

When administered by gavage, the THM induced a dose-dependent decrease in the methylation of the promoter region of the c-myc protooncogene. The four THMs were of similar potency although chlorodibromomethane appeared to be a little less active than the other three. When administered in drinking water bromoform appeared to be slightly more active than the other three THMs which were of similar activity.

The results of this study demonstrated that the THM enhanced cell proliferation, induced toxicity and caused hypomethylated c-myc protooncogene. The ability of the THM to induce toxicity, cell proliferation, and hypomethylation of c-myc did not correspond. Bromodichloromethane was the most toxic and induced the greatest decrease in methylation followed closely by chloroform. Furthermore, bromoform was equally potent in causing as great an enhancement of cell proliferation as chloroform, although it produced the least toxicity and caused less hypomethylation than chloroform and bromodichloromethane. However, the results are consistent in supporting our hypothesis that hypomethylation resulted from the prevention of the methylation of newly synthesized DNA and requires both enhanced cell proliferation and toxicity. Enhancement of cell proliferation is required because adult mouse liver, as used in this study, has an extremely low level of cell proliferation (i.e., ~0.1 percent of the hepatocytes). The results also indicate that the dose-response for enhanced cell proliferation induced by the THM administered by gavage contains a threshold, or is at least concave.

Chloroform, DCA, and TCA are found together in chlorinated drinking water and have demonstrated carcinogenic activity in mouse liver; therefore, the carcinogenic activity of mixtures containing them was evaluated. Female and male B6C3F1 mice were administered 30 mg/kg N-methyl-N-nitrosourea on day 15 of age. At 5 weeks of age, they started to receive in their drinking water 3.2 g/L DCA or 4.0 g/L TCA with 0, 800, or 1600 mg/L chloroform until sacrificed at 36 weeks of age. Chloroform reduced the yield of liver foci and tumors promoted by DCA, but not TCA. Chloroform appeared to preferentially reduce the yield of DCA-promoted eosinophilic foci and tumors in contrast to basophilic lesions. DCA administered alone did not promote kidney tumors. However, in male but not female mice the yield of kidney tumors in DCA-treated mice was greatly increased by chloroform. TCA administered alone increased the yield of kidney tumors in male but not female mice. The co-administration of chloroform did not affect the yield of kidney tumors in TCA-treated mice. Hence, chloroform prevented the promotion of liver tumors by DCA, enhanced the promotion of kidney tumors by DCA in male mice, and did not affect the promotion by TCA of liver and kidney tumors.

The effect of chloroform administered in the drinking water on the methylation of the promoter region of the c-myc gene was determined. Chloroform administered in drinking water at 1600 and 800 mg/L, but not at 400 mg/L, induced hypomethylation of c-myc. When chloroform was co-administered with DCA in the drinking water, it prevented the hypomethylation of c-myc induced by DCA. Thus, DCA in the presence of 0 and 400 mg/L chloroform resulted in the same amount of hypomethylated c-myc gene. However, when co-administered with 800 and 1600 mg/L chloroform, there was a dose-related decrease in the amount of hypomethylated c-myc. When chloroform was co-administered with TCA, it did not alter the extent of hypomethylation induced by TCA. Hence, there was a correlation between the effect of co-administering chloroform in the drinking water on hypomethylation of c-myc and its effect on the yield of liver tumors. That is, chloroform decreased the ability of DCA both to induce hypomethylation of c-myc and to promote liver tumors while not affecting the ability of TCA to induce either effect.


In summary, our results demonstrate that the carcinogenic activity of even a binary mixture of chemical carcinogens is difficult to predict. The mixtures in our study demonstrated synergism and antagonism that were target organ and sex specific. Thus, chloroform prevented DCA promotion of liver cancer, enhanced DCA promotion of kidney cancer in male but not female mice, and had no effect in either sex on the promotion by TCA of liver and kidney cancer. Therefore, it is very important to validate using carcinogenesis bioassays the models used in risk assessment of mixtures. Otherwise unrealistically high or low estimates could result that would do a disservice to society and to the protection of people and the environment.

Journal Articles on this Report : 7 Displayed | Download in RIS Format

Other project views: All 7 publications 7 publications in selected types All 7 journal articles
Type Citation Project Document Sources
Journal Article Coffin JC, Ge R, Yang S, Kramer PM, Tao L, Pereira MA. Effect of trihalomethanes on cell proliferation and DNA methylation in female B6C3F1 mouse liver. Toxicological Sciences 2000;58(2):243-252. R825384 (Final)
  • Abstract from PubMed
  • Full-text: Toxicological Sciences
  • Other: Toxicological Sciences PDF
  • Journal Article Ge R, Yang S, Kramer PM, Tao L, Pereira MA. The effect of dichloroacetic acid and trichloroacetic acid on DNA methylation and cell proliferation in B6C3F1 mice. Journal of Biochemical and Molecular Toxicology 2001;15(2):100-106. R825384 (Final)
  • Abstract from PubMed
  • Journal Article Pereira MA, Kramer PM, Conran PB, Tao L. Effect of chloroform on dichloroacetic acid and trichloroacetic acid-induced hypomethylation and expression of the c-myc gene and on their promotion of liver and kidney tumors in mice. Carcinogenesis 2001;22(9):1511-1519. R825384 (Final)
    R828083 (2000)
    R828083 (2001)
    R828083 (Final)
  • Abstract from PubMed
  • Full-text: Carcinogenesis
  • Other: Carcinogenesis
  • Journal Article Tao L, Kramer PM, Ge R, Pereira MA. Effect of dichloroacetic acid and trichloroacetic acid on DNA methylation in liver and tumors of female B6C3F1 mice. Toxicological Sciences 1998;43(2):139-144. R825384 (Final)
  • Abstract from PubMed
  • Full-text: Toxicological Sciences PDF
  • Journal Article Tao L, Ge R, Xie M, Kramer PM, Pereira MA. Effect of trichloroethylene on DNA methylation and expression of early-intermediate protooncogenes in the liver of B6C3F1 mice. Journal of Biochemical and Molecular Toxicology 1999;13(5):231-237. R825384 (Final)
  • Abstract from PubMed
  • Journal Article Tao L, Yang S, Xie M, Kramer PM, Pereira MA. Effect of trichloroethylene and its metabolites, dichloroacetic acid and trichloroacetic acid, on the methylation and expression of c-Jun and c-Myc protooncogenes in mouse liver:prevention by methionine. Toxicological Sciences 2000;54(2):399-407. R825384 (Final)
  • Abstract from PubMed
  • Full-text: Toxicological Sciences
  • Other: Toxicological Sciences PDF
  • Journal Article Tao L, Yang S, Xie M, Kramer PM, Pereira MA. Hypomethylation and overexpression of c-jun and c-myc protooncogenes and increased DNA methyltransferase activity in dichloroacetic and trichloroacetic acid-promoted mouse liver tumors. Cancer Letters 2000;158(2):185-193. R825384 (Final)
  • Abstract from PubMed
  • Full-text: Science Direct
  • Other: Science Direct PDF
  • Supplemental Keywords:

    water, drinking water, exposure, risk, risk assessment, effects, health effects, human health, metabolism, dose-response, carcinogen, mutagen, animal, mammalian, enzymes, chemicals, toxics, solvents, organics, decisionmaking, cost benefit, environmental chemistry, biology, histology, genetics, pathology, zoology, modeling., RFA, Health, Scientific Discipline, Waste, Water, Environmental Chemistry, Health Risk Assessment, Chemistry, chemical mixtures, Risk Assessments, Biochemistry, Biology, Drinking Water, cancer risk, monitoring, trihalomethanes, exposure and effects, chemical byproducts, disinfection byproducts (DPBs), exposure, community water system, carcinogens, human exposure, treatment, haloacetic acids, carcinogenicity, tumors, colon cancer, cancer risk assessment, drinking water contaminants, drinking water system

    Progress and Final Reports:

    Original Abstract
  • 1997
  • 1998