Drinking water contaminants and their experimentally-induced mutagenic and toxic effects

1. Drinking water contaminants and theirexperimentally-induced mutagenic and toxic effects G. Mancaş1, T. Vartiainen2, P. Rantakokko2, T. Navrotescu1, M. Vasilov1, D. Mancaş3, D. Diaconu3 1Institute of Public Health, Iaşi, Romania 2National Institute of Public Health, Kuopio, Finland 3Agilrom Scientific Company, Bucureşti, Romania

2. Objective • The disinfection of drinking water provides protection against microbial diseases but also produces chemical by-products that may pose other types of risks to public health. Among these chemicals, the halogenated hydroxyfuranones, although present at concentrations lower than 0.1 μg/L in drinking water, can be responsible for more than half of Ames test mutagenicity. The so called “mutagen X”(MX), the most potent of these chlorohydroxyfuranones, has been identified as a strong carcinogen in rats. • In this context, the identification and measurement of these risks generated by drinking water from different sources is a realistic goal for public health protection. • Based on these data, our work aimed to identify the chemical pattern of drinking water contaminants, to measure the MX levels in water extracts and to assess their possible toxic effects, by Ames test of mutagenicity of water sample extracts and, respectively, by short-term toxicity tests on different aquatic organisms. An attempt to relate chemical exposure levels with experimentally-found toxic effects was also our purpose.

3. Methodological approach • In this study (2000-2002) drinking water was sampled from the water-work Delea of Vaslui town, situated in the northeastern Romania. The selection of this drinking water source was accounted for some chemical and biological pollution episodes recorded in the recent years and the high frequency of inadequate organic matter levels found in the yearly surveys of drinking water quality. • As raw water, the Delea water-work uses surface water, mainly from Soleşti lake. About 60,000 inhabitants of Vaslui town use this drinking water (about 14,000 cubic meters of water/day). • The treatment process consists of prechlorination, coagulation, sedimentation, filtration and final chlorination. The prechlorination step is not always carried out and the prechlorine doses are never measured. After the prechlorination, the measurement of chlorine level in water is made, but, usually it takes 7-8 days untill the results are obtained. For final chlorination, 4-6 kg of chlorine/hour are used and 5-7 kg of chlorine when there is no prechlorination step. • In these uncontrolled conditions, the inadequate quality of drinking water was recorded in the past years, with unknown consequences on population health status.

4. Methodological steps • Twice a year sampling of large volumes (over 200 liters each)) of raw water (RW) from Soleşti lake (source water) and chlorinated water (CW) supplied by Delea water-work • Assessment of chemical and biological quality parameters of water samples • XAD-2 resin adsorbtion/diethyl ether extraction of water samples to concentrate the non-volatile organic chemicals • Measurement of MX levels in water extracts by GC-MS technique • GC-MS qualitative analysis of RW- and CW-extracts (the identification of chemicals was performed based on Willey library. Values over 90% were considered among the identified compounds). • Toxicity tests using water extracts: - Ames test for mutagenicity (without enzymatic activation) - Short-term toxicity tests on aquatic organisms belonging to various systemic groups (vertebrates and invertebrates, animals and plants): - Acute tests(24 h) on green algae sp. Scenedesmus quadricauda (A-test) - Acute tests(24 şi 48 ore) on microcrustaceans sp. Daphnia magna Straus - Acute and subacute tests (24, 48, 72, 96 h and 7 days) on fish sp. Carassius auratus GibelioBloch The designs of toxicity tests were achieved based on the assumption that the RW- and CW-extracts contain the same chemicals as RW and CW samples, respectively.

5. Toxic effects tests aimed at: • assessing the sublethal effects of some small concentrations of water chemical contaminants (as extracts) at levels normally found in raw and chlorinated water; • determining the toxic effects of water chemical contaminants at gradually increased concentrations; • comparing the toxic effects induced by CW extracts with those by RW extracts and with controls treated with adequate amounts of diethyl ether; The experimental conditions were the following: • Environmental air temperature: 20.0 – 21.5° C • Water temperature: 20-22° C • Light intensity: 968-1532 lux • Concentration of oxygen dissolved in water: 9.8 – 10.0 mg/l • Level of ammonia in the water: 0 • Level of nitrites in the water: 0 • Total hardness: 18.7 G • Alkalinity: 4.0-4.7 ml HCL n/10 • Parasites: absent • Organisms showing impurification: absent

6. Results A significant increase of organic compounds levels (Figure 1) measured as permanganate consumption up to 21.17 mg/L in CW samples was noted in the study period (MAC = 10.0 mg/L). MAC Fig. 1. Organic matter concentrations found in drinking water in different points of distribution network in Vaslui in 2000-2002 period

7. Phenolic compounds (MAC in drinking water – 0.001 mg/L) Phenolic compounds (up to 0.083 mg/L) and anionic detergents levels (up to 0.35 mg/L) were found in high concentrations in raw water, as Figure 2 shows. 2000 2001 Anionic detergents levels (MAC in drinking water – 0,2 mg/L) 2000 2001 Fig. 2 Organic pollutants levels in RW and CW of Delea water work in the study period

8. The water biological quality revealed the following major aspects: - higher frequency (13.04 %) of ferobacteria in water samples of distribution network compared with chlorinated water sampled at Delea water–work (7.89%) - higher frequency (10.52%) of Giardia sp.cysts in chlorinated water sampled at water-work in comparison with only 8.69% found in distribution network samples. These results are consequences of both inadequate treatment process of raw water in the Delea water-work and deficiencies of distribution network. The qualitative GC-MS analysis of water extracts has reliably identified mixtures of chemicals both in CW and RW extracts (Tables 1 and 2). The chemicals identified have complex structures and further investigation are necessary to confirm them using reference standards.

9. Table 1. List of organics identified in RW-extract sample by GC-MS analysis

10. Table 2. List of organics identified in CWextract sample by GC-MS analysis (a)

11. Table 2. List of organics identified in CWextract sample by GC-MS analysis (b)

12. The MX levels and the results of mutagenicity test are indicated in Table 3. The water samples of June 2000 had higher levels of MX than those ones of November 2001. In addition, the levels measured for this compound were higher in CW- than those found in RW-extracts. These results are the first ones measured in water samples of Romania. Levels of MX higher than those found in our study were reported in Finland (15, 16), UK (17) and USA (18,19). Maximum MX levels measured in water samples of Netherlands (20), Spain (21), Japan (22,23), Germany (24) and Russia (25) were lower than those ones reported here. Table 3. MX levels in water sample extracts and the results of mutagenity test The range of mutagenicity for different raw or chlorinated water samples tested worldwide is quite large, but, basically, higher values in chlorinated water than their corresponding raw waters were reported. There is substantial evidence that most of mutagenic activity in drinking water originates from the reaction of disinfectants, especially chlorine, with the organic matter present in source waters. The water sampled in June 2000 showed also higher mutagenicity in Ames Salmonella assay (Table 3). At the same time, the mutagenicity of CW sample was over 3-fold higher than that of the corresponding RW-sample. <100 = result is lower than the limit of quantitation in the mutagenity analysis <LOQ = result is lower than the limit of quantitation in MX and BMX – analysis

13. Results of toxicity tests on aquatic organisms The results of the tests on green algae are presented in table 4. Tabelul 4. Results of toxicity tests (A-test) on green algae – sp. Scenedesmus quadricauda, for raw water (RW) and chlorinated water (CW) samples (%) DEE – diethyl ether

14. By comparing the results obtained with CW extracts with the RW ones, stronger toxic effects under the action of CW-extracts were found, the differences ranging from 5.16% (concentration factor x1), 12.72% (concentration factor x2), 25.35% (concentration factor x3) to 50.42% (concentration factor x5). Figure 3. Results of acute toxicity tests on green algae (%)

15. 120 100 80 A-test (%) 60 40 20 0 1 2 3 4 5 6 Concentration factor r = - 0.98 (p<0.001) y = 104.15 – 14.88x Figure 4. Correlation between different concentrating levels of organics in CW and results of A-test on green algae sp. Scenedesmus quadricauda, The exposure-effect relationship (Fig. 4) indicated by regression curve, has showed statistical significant negative strong correlation. These results demonstrate that, in our experimental conditions, the higher the organic compounds concentration the lower the biostimulation photosynthesis activity.

16. Toxicity tests on microcrustaceans Tabelul 5. Results of lethalities (%) for microcrustaceans sp. Daphnia magna Straus, in acute toxicity tests using CW and RW-extracts DEE – diethyl ether The results of acute toxicity tests on Daphnia are presented in Table 5.

17. 24 h 100 RW CW Control 80 60 Lethality % 40 20 0 x 1 x 2 x 3 x 5 x 10 Concentration factor Figure 5. Results of lethalities (%) for microcrustaceans sp. Daphnia magna Straus, in acute toxicity tests In the experiments with a 3-fold concentration of water chemical pollutants, lethality of the organisms after 24 h-exposure in CW was 50% as compared to only 20% in RW, that is a 2.5 times increase in lethality. After a 48 h-exposure, the results also revealed higher lethalities, with CW being more toxic that the corresponding RW. When comparing the results of the experiments with the two types of extracts, it was noticed that CW extract caused after 48 hours a higher lethality than the corresponding RW, starting even with the samples with concentration factor x1, that is the two water samples as such, without added extract (10% versus 0%).

18. The mean tolerance limit for the tested species is in the range 34-68 μl RW extract in 25 ml RW (that is a concentration factor 3-5) and in the range 17-34 μl CW extract in 25 ml CW (2-3 concentration factor 2-3). r = 0.87 (p<0.001) y = 9.17 + 11.30x r = 0.92 (p<0.001) y = -0.05 + 11.62x Figure 6. Correlations between different concentrating levels of organics in CW and lethalities of microcrustaceans sp. Daphnia magna Straus after acute toxicity tests

19. Fish exposure in CW with concentration factor of 1 yielded a 10% lethality after 7 days, higher than those resulted in fish 7 days-exposure in RW (0%). • Fish exposure in CW with concentration factor x 20 yielded higher lethalities: 40%, 60% and 80% after 24h-, 72 h- and 7 days-exposures, respectively, • in comparison with the lethalities recorded in fish exposed in RW with the same concentration factor and for similar experimental periods: 20%, 40% and 60%, respectively. Figure 7. Lethality of fish sp. Carassius auratus – Gibelio Bloch treated with gradually increased concentrations of chemicals in RW and CW during experimental periods

20. Exposure-effect relationshipswere calcu-lated based on fish lethality percentages as indicator of toxic effect of gradually increased concentrations of chemi-cals in CW and RW, for acute (24 h) and sub-acute (7 days) expe-rimental exposure peri-od. The results are indicated in Figure 8. The relationships were statistically significant in acute exposure to CW and RW. 24 h-exposure in CW 24 h-exposure in RW 7 days-exposure in CW 7 days-exposure in RW Figure 7. Exposure-effect relationship in acute and subacute experiments on fish

21. These experimental results on aquatic organisms are in agreement with both the findings for mutagenicity and the MX levels measured in the two types of water samples. • Conclusions • Analyses of water samples of Delea water-work in Vaslui town revealed high levels of organic compounds in both raw and chlorinated water samples and an inadequate biological pollution, especially in chlorinated water samples, basically due to high degree of organic pollution of raw water and unappropriate application of water treatment procedures. • Water samples extract analysis has reliably identified complex mixtures of chemicals in both raw and chlorinated water samples. • Higher level of potent mutagenic chlorinated compound (MX) in chlorinated water compared with raw water samples was found. • Findings of water extract-mutagenicity assay were in agreement with MX levels measured in raw and chlorinated water extracts.

22. Tests carried out in controlled laboratory conditions on three types of aquatic organisms belonging to different systemic groups revealed more marked toxic effects and higher lethalities under the influence of exposure in CW than in the corresponding RW. The inhibition of biostimulation of photosynthesis process in experimentally-exposed green algae and the lethality percentages of microcrustaceans and fish in gradually-increased concentrations of chemicals have been higher in the experimental tests with chlorinated water than those carried out with raw water. • The exposure-effect relationships between gradually increased concentrations of chemicals and the toxic effects were statistically significant for acute exposures in raw and, respectively, chlorinated water. • Further similar studies are also necessary for other water-works which use surface water as source water for drinking water.