Franšais      print      e-mail    


Le Curieux F., Erb F. and D. Marzin (1998) Identification of genotoxic compounds in drinking waters. Rev. Sci. Eau 11 (nSpecial 10th anniversary) : 103-118. [article in French]

Original title: Identification de composés génotoxiques dans les eaux de boisson.

Full text (PDF)


In 1974, two independent studies - one in the Netherlands and the other in the United States - demonstrated the occurrence of trihalomethanes in drinking water. Following studies showed that these chemicals were common contaminants of drinking water and that chloroform, i.e. one of these trihalomethanes, was carcinogenic in rodents. Further investigations demonstrated that extracts of chlorinated drinking water induced significant mutagenicity in the Ames/Salmonella assay. In the present paper we will fist discuss the methods used to detect the genotoxic activity of drinking water and, then, the methods developed to identify the compounds responsible for this activity. After this, we will present the main genotoxic chemicals identified in drinking water, before finally considering several propositions to limit the exposure of populations to these genotoxic compounds.

Drinking water is usually produced through a multistage process which includes one or several chlorination steps. It is now widely accepted that the genotoxic activity of drinking water mainly originates from the reaction of chlorine with humic substances present in raw water. Humic substances are natural organic matters (resulting from the degradation of plants and animal tissues) of very complex structure with most chemical functions arranged in aromatic rings or aliphatic chains. The identification of a genotoxic activity in drinking water usually requires concentration of the water samples. Even though such a process implies a probable qualitative/quantitative alteration of the constituents of water samples, the extremely low amounts of genotoxic compounds in drinking water require concentration steps. Among the many genotoxicity tests carried out, the Ames test (which detects reverse mutations in bacteria Salmonella typhimurium) is the assay which was the most frequently used in the field of drinking water mutagenicity. Other tests were performed on eucaryotic cells. Assays detecting micronuclei or chromosomal aberrations in plants, or mutations in mold, yeast, or maize enabled the detection of genotoxic effects of drinking water extracts. Tests on mammal cells also showed that drinking water extracts induced point mutations, sister chromatid exchanges, chromosomal aberrations and micronuclei. In vivo tests on aquatic organims such as newt or mussels demonstrated the micronuclei inducing effect of unconcentrated drinking water samples.Regarding the identification of the compounds responsible for the genotoxicity, it is obviously not possible to identify all of the thousands of chemicals that may be involved. But such a process is important in order to evaluate the specific genotoxicity and the risk associated with (at least) the main chemicals occurring in drinking water. The identification process usually follows three steps:

  1. concentration of the sample can be performed using reverse osmosis, freeze drying, liquid-liquid extraction, and/or adsorption on non ionic resin followed by extraction with organic solvent;
  2. the purification step uses one or a combination of chromatographic techniques (TLC, packed column liquid chromatography, HPLC or GC);
  3. structural identification of the chemical is performed using data from mass spectrometry, and proton and carbon NMR, or UV spectroscopy.

The analysis of the genotoxic compounds of drinking water showed that they are rather non-volatile, quite acid and not stable at high pH, rather polar, and with a mean molecular weigh around 200.

Turning now to the identity of these compounds, it is considered that the genotoxicity of drinking water is mainly due to organohalogenated chemicals. Some inorganic chemicals (this class of chemicals is usually not recovered in drinking water extracts) which induce genotoxic or carcinogenic effects must, however, be recalled. Arsenic, nitrates, bromates and radon are natural or human-activity-related drinking water contaminants which are responsible for cancers in rodents or in humans. Among the many genotoxic or carcinogenic organohalogenated compounds identified in drinking water, the most abundant chemicals are chlorinated and/or brominated trihalomethanes. Other important groups of compounds are chlorinated and/or brominated derivatives of acetic acids, acetonitriles, ketones, phenolic compounds. The chlorinated hydroxyfuranones, although present at concentrations lower than 0.1 µg/l in drinking water, can be responsible for more than half of the Ames mutagenicity. MX, the most potent of these chlorohydroxyfuranones, has been submitted to intensive toxicological studies worldwide and was very recently identified as a potent carcinogen in rats.

Now that the presence of genotoxic compounds in drinking waters is a well documented and accepted fact, the perspectives lies in the better identification of the impact of these drinking water contaminants. The development of more sensitive tests such as the Comet assay (detection of DNA strand breaks) or the 32P postlabelling assay (detection of DNA adducts) should be pursued. Moreover, the interaction between genotoxic compounds and DNA must be investigated more thoroughly, including the identification of adduct structures. More globally, it is of interest to better assess the impact of these agents on public health and on the occurrence of specific human cancers. At present, even though a few individual water contaminants are classified as human probable carcinogens, the chlorinated drinking water (in itself) is not considered as carcinogenic to humans. Exposure to these potentially harmful agents can be limited with

  1. improving drinking water quality - i.e. decreasing the formation of genotoxins - by using raw water containing lower amounts of organic matter; and
  2. modifying the water chemical treatment by using lower amounts of chlorine and/or combining chlorine with other disinfectants.

The public health can also be protected by the setting of guidelines for drinking water: each compound identified as dangerous would be given a concentration threshold which should never be exceeded. The Environmental Protection Agency in the U.S.A. and the World Health Organisation are authorities setting such guidelines. Finally, we believe it is important to limit the concentration of genotoxic compounds in drinking water as much as possible, and one way to do so is to use chlorine in smaller amounts and in a more efficient way. But it is of paramount importance to keep in mind that the disinfection process (in which chlorine still plays a major role) and the providing of a microbiologically safe drinking water should never be jeopardized.


Identification, genotoxic compounds, drinking water, chlorination, humic substances, organohalogenated.

Corresponding author

D. Marzin, Laboratoire de Toxicologie, Institut Pasteur de Lille, 1 rue du Professeur Calmette, BP 245, F-59019 Lille Cedex, FRANCE

Email : Daniel.Marzin@pasteur-lille.fr

Franšais      print      e-mail    

Update: 2006-12-19
© INRS Eau, Terre et Environnement