- Email: [email protected]
The variation on the mutagenicity of cnp during anaerobic biodegradation
PII: S0043-1354(00)00547-9
Wat. Res. Vol. 35, No. 11, pp. 2589–2594, 2001 # 2001 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0043-1354/01/$ - see front matter
THE VARIATION ON THE MUTAGENICITY OF CNP DURING ANAEROBIC BIODEGRADATION TAKU MATSUSHITA1*, SATORU SAKUMA1, KATSUHIKO NAKAMURO2 and YOSHIHIKO MATSUI2 2
1 Department of Civil Engineering, Gifu University, 1-1 Yanagido, Gifu, Gifu 501-1193, Japan and Faculty of Pharmaceutical Sciences, Setsunan University, 45-1, Nagaotoge-cho Hirakata, Osaka 5730101, Japan
(First received 9 June 2000; accepted in revised form 29 October 2000) Abstract}The mutagenicity of water, including herbicide CNP, and its time-variation during anaerobic biodegradation were studied through Ames assay using strains with or without. S9 mix: TA98, TA100, YG1021, YG1024, YG1026, and YG1029. The bacteria, for the anaerobic biodegradation, was obtained from a paddy field, and preincubated for a month. The CNP was decomposed in an anaerobic culture inoculated with the bacteria, and finally yielded CNP-amino as one of the CNP metabolites. About 16% of the initial CNP was transformed into CNP-amino by the 14th day. The mutagenicities to TA98, YG1024, and YG1029 strains with S9 mix increased with cultivating time, the latter two showed the strongest sensitivity to CNP-amino. The contribution of CNP to the mutagenicity decreased as the chemical decomposed, while the contribution of CNP-amino increased. However, the increased mutagenicity was not limited to the contribution of CNP-amino, but also to the contribution of other metabolites. The contributions of other CNP metabolites were 67% of total mutagenicity to the TA98 strain and 30% to the YG1029 strain. These unknown mutagenic metabolites were the indirect frameshift mutagens which did not have nitro- and amino-substituents, and the indirect base-pair mutagens which might possibly have some amino-substituents. # 2001 Elsevier Science Ltd. All rights reserved Key words}CNP, herbicide, ames assay, mutagenicity, anaerobic biodegradation, YG strain
INTRODUCTION
Chlornitrofen (CNP, IUPAC name: 2,4,6-trichlorophenyl 4-nitrophenyl ether, Fig. 1) was extensively used to control paddy weeds in Japan because of its low impact on rice plant. Draper and Casida investigated mutagenicity of several diphenyl ether herbicides including CNP and their substituted compounds by Ames assay using TA100 strain, and reported that CNP was non-mutagenic with or without metabolic activation ( S9 mix) (Draper and Casida, 1983). However, after a relatively high incidence of gallbladder cancer in Niigata, Japan was related to CNP in drinking water as reported by Yamamoto et al. in an epidemiological study, the safety of CNP has attracted attention (Yamamoto et al., 1993). After applied to a paddy field, CNP is decomposed by soil bacteria and transformed into several compounds (Kuwatsuka, 1976; Oyamada and Kuwatsuka, 1979, 1989). Kuwatsuka found that the nitro group in CNP was biologically deoxidized into
*Author to whom all correspondence should be addressed. Tel.: +81-58-294-2444; fax: +81-58-230-1891; e-mail: [email protected]
an amino group, and that CNP-amino (2,4,6trichlorophenyl 4-aminophenyl ether, Fig. 1) was produced in paddy soil during the flooded period (Kuwatsuka, 1976). Kuwatsuka also reported that CNP was mainly decomposed into CNP-amino, CNP-methylamino (2,4,6-trichlorophenyl 4-metylaminophenyl ether), CNP-acetylamino (2,4,6-trichlorophenyl 4-acetylaminophenyl ether), CNPpropylamino (2,4,6-trichlorophenyl 4-propylaminophenyl ether) and CNP-hydroxyl (2,4,6-trichlorophenyl 4-hydroxylphenyl ether) (Kuwatsuka, 1976). These studies suggest that the risk assessment of CNP should be judged in conjunction with the CNP metabolites. Draper and Casida reported CNPamino to be mutagen with S9 mix (Draper and Casida, 1983). This study suggested that the biological degradation could possibly produce other mutagens and not necessarily decrease toxicity. Nevertheless, the toxicities of CNP metabolites, except CNP-amino, have not been investigated. All metabolites of CNP will hardly ever be identified, or their toxicity will hardly ever be evaluated during biodegradation by simply adding the toxicity of each identified metabolite. This methodology cannot estimate if the toxicity will increase or decrease when CNP is applied in the environment.
2589
2590
T. Matsushita et al.
Fig. 1. Chemical structure of CNP (IUPAC name: 2,4,6trichlorophenyl 4-nitrophenyl ether) and CNP-amino (IUPAC name: 2,4,6-trichlorophenyl 4-aminophenyl ether).
extract was dried over anhydrous sodium sulfate, and 200 mL of the extract was directly injected into HPLC. The separation column for the HPLC was 4.6 mm inside diameter 250 mm long filled with Wakosil-II 5C18-100 (Wako Pure Chemicals Industries, Ltd., Osaka, Japan). The mobile phase was acetonitrile: water (70 : 30, v : v). The flow rate was 0.3 mL/min, and detection was recorded at 205 nm for CNP and at 254 nm for CNP-amino. Twenty liters of the sample medium for Ames assay was extracted with 2 L of n-hexane. The extract was dried over anhydrous sodium sulfate, and then evaporated to obtain solid residue. The evaporation was performed using a rotary evaporator at room temperature, under reduced pressure in order to prevent the sample from being transformed by heating. The residual organic matter was redissolved in DMSO of 6 mL. Anaerobic biodegradation
Instead of analyzing the toxicity of CNP and each CNP metabolite as above, whole culture medium was used to determine the toxicity of the CNP solution under the aerobic biological degradation (Tanaka et al., 1996). Tanaka et al. reported that the mutagenicity of the test solution increased during the biological degradation. However, the distinction was not made whether the mutagenicity was due to CNP, CNP metabolites (including CNP-amino), or other constituents in the river water from which the CNP solution was prepared. Matsushita et al. applied chromosomal aberration test to a CNP solution under aerobic biological degradation and found that the induced chromosomal aberration activity did not decrease to its control level even when CNP was not detected in the solution (Matsushita et al., 1997). Accordingly, our aims in this research were the following: to investigate the mutagenicity of the culture media, including CNP, during anaerobic biodegradation, and to classify the mutagenicity into the contributions of CNP, CNP-amino and other unknown metabolites.
MATERIALS AND METHODS
Regents All reagents were purchased from the following sources; S9 (liver homogenates prepared from male Sprague–Dawley rats with pheno-barbital and 5,6-bezoflavone) and co-factor from Oriental Yeast Co., Ltd., Tokyo, Japan; CNP, CNPamino, n-hexane, sodium sulfate, dimethyl sulfoxide (DMSO), acetonitrile and other chemicals from Wako Pure Chemicals Industries, Ltd., Osaka, Japan. A stock solution of CNP was prepared by dissolving the reagent CNP into acetone (4000 mg/L). The CNP solution was then prepared by diluting this stock solution with water: undissolved particles of CNP were removed by filtration through a paper filter of No. 5C (Advantec Toyo, Tokyo, Japan). Methods of measurements CNP and CNP-amino were quantified by reverse-phase high-performance liquid chromatography (HPLC, 600 Controller, 717 Plus Autosampler and 486 Tunable Absorbance Detector, Waters Co., Milford, MA, USA) after extracting using n-hexane. Hundred milliliters of the sample medium was extracted with 10 mL of n-hexane. The
Anaerobic bacteria originating from a paddy field soil on the west side of Gifu University were used for the CNP biodegradation experiments after preincubation. The soil was collected under flooded conditions in spring, 1999. The wet soil was stored at 48C until use. One gram of the wet soil and the CNP solution was added to a 500 mL bottle. The culture medium described below was then added to the bottle. After introducing nitrogen gas into the bottle, the mouth of the bottle was covered with a frosted glass lid and sealed with Parafilm. The bottle was under anaerobic condition at 208C in the dark for a month for preincubating. We conducted the CNP degradation experiments with the preincubated bacteria: the bacteria was inoculated by injecting a small portion of the culture solution from the bottle. The CNP degradation was proceeded in 10 L glass vessel filled with liquid medium. The vessel was placed in the dark. The liquid medium was K2HPO4: 350 mg/L, KH2PO4: 270 mg/L, NH4Cl: 530 mg/L, MgCl26H2O: 100 mg/L, CaCl22H2O: 75 mg/L, NaHCO3: 1200 mg/L, Na2S9H2O: 500 mg/L, yeast extract: 30 mg/L. During the biodegradation, the oxidation-reduction potential was maintained below 470 mV. Thus, we consider the culture medium to be under anaerobic condition. Ames assay Assays were performed in the presence and absence of a rat liver activation system: the method was according to the 20-min preincubation procedure described by Maron and Ames (Maron and Ames, 1983). The test strains used in this assay were YG1021, YG1024, YG1026 and YG1029 in addition to the classical strains, TA98 and TA100. These YG strains with an elevated level of classical nitroreductase or O-acetyltransferase were developed by cloning the corresponding gene into pBR322 plasmids and introducing those plasmids carrying the gene into TA98 and TA100 (Watanabe et al., 1989; 1990). Further details about these strains are described in Table 1. These bacteria were grown in 20 mL of Oxoid nutrient broth No. 2 and incubated for 16 h at 378C in a shaking water bath in order to insure adequate aeration. 2-Nitrofluorene was used as the positive control chemical in the absence of S9 mix, and 2-aminoanthracene was used in the presence of S9 mix. S9 was used as the external metabolic activation system at a final concentration of 10% in a co-factor mix. Treatments without S9 mix were conducted by adding either 0.2 mL of DMSO or 0.0125-0.05 mL of a solution of the test sample to 0.5 mL of Na–K buffer. Treatments with metabolic activation were conducted by adding 0.5 mL of S9 mix to the test sample instead. These reagents were mixed and incubated in a tube with 0.1 mL of the test-strains for 20 min in a shaking water bath at 378C. After adding 2 mL
Mutagenicity of CNP during anaerobic biodegradation
2591
Table 1. Salmonella typhimurium strains used in this study Strain
Description
TA98 YG1021 YG1024 TA100 YG1026 YG1029
his D3052 (pKM101) TA98 (pYG219): a nitroreductase-ovrproducing strain TA98 (pYG219): an O-acetyltransferase-overproducing strain his G46 (pKM101) TA100 (pYG216): a nitroreductase-overproducing strain TA100 (pYG219): an O-acetyltransferase-overproducing strain
of the top agar to the tube, the resulting mixture was plated on a glucose minimal agar plate. Revertant colonies (His+ revertants) were counted after incubating at 378C for 48 h. The number of His+ revertants for each sample was given as the mean value of the two plates. To quantify the magnitude of the mutagenicity, we calculated the ratio of the number of His+ revertants per plate to the number of spontaneous revertants per plate. In this paper, this ratio is referred to as ‘the mutagenicity intensity’. Compounds of which mutagenicity intensities were larger than the two which were considered to be mutagenic.
RESULTS AND DISCUSSION
Mutagenicities of CNP and CNP-amino Figure 2 shows the mutagenicities of CNP and CNP-amino to TA and YG strains. CNP was nonmutagenic in the standard Ames test with test-strains that detect frame shift (TA98) and base-pair substitution (TA100) either in the absence or presence of S9 mix. This result is in accordance with the results obtained by other researchers (Takeda et al., 1989). Although CNP has one nitro group, no mutagenicity was detected using YG1021 and YG1026 strains, which were developed to enhance sensitivity to nitroarenes (nitroreductase-overproducing strain). Mutagenicity was detected using YG1024 and YG1029 strains, which have enhanced sensitivity to aminoarenes (O-acetyltransferase-overproducing strain). Einisto¨ et al. (1991) reported the same phenomenon for 1,8-dinitropyrene, which has two nitro groups (Einisto¨ et al., 1,8-dinitropyrene, out of more than 10 compounds with nitro groups, did not show exceptionally stronger mutagenicity to the YG1021 strain than the TA98 strain. On the contrary, a much stronger mutagenicity was detected to the YG1024. We believe this phenomenon was caused by the following processes: In Salmonella typhimurium, the CNP nitro group is not reduced by overproduced nitroreductase of YG1021 and YG1026, but by other nitroreductase which the classical strains already have. CNP, therefore, did not exhibit positive mutagenicity to the nitroreductase-overproducing strains (YG1021 and YG1026). Since reduced CNP is then metabolized by O-acetyltransferase which is enhanced in YG1024 and YG1029 strains, the strong mutagenicity is detected to the O-acetyltransferase-overproducing strains (YG1024 and YG1029 strains). On the other hand, CNP-amino exhibited positive mutagenic activity to the two classical strains in the
Fig. 2. Mutagenicities of CNP and CNP-amino to TA and YG strains. mean S9mix condition, means+S9mix condition. Spontaneous revertants are as follows: 17 (TA98 S9), 15 (TA98+S9), 42 (YG1021 S9), 35 (YG1021+S9), 40 (YG1024 S9), 31 (YG1024+S9), 17 (YG1041 S9), 22 (YG1041+S9), 88 (TA100 S9), 53 (TA100+S9), 114 (YG1026 S9), 162 (YG1026+S9), 152 (YG1029 S9), 144 (YG1029+S9).
presence of S9 mix. These results also agree with the suggestion of other researchers (Takeda et al., 1989). Moreover, the mutagenicity of CNP-amino was detected in all YG strains in the presence of S9 mix. Overall, the mutagenicity of CNP-amino was much stronger relative to CNP. Variation on the mutagenicity during anaerobic biodegradation Figure 3 shows concentration of CNP and CNPamino during anaerobic biodegradation. CNP was gradually decomposed by bacteria in 14 days, and
2592
T. Matsushita et al.
yielded CNP-amino as one of the metabolites. At the beginning of cultivation, the CNP concentration was 238 mg/L. After three days of the cultivation the CNP concentration was lowered to 206 mg/L, and produced 1 mg/L of CNP-amino. On the 14th day of the cultivation, the CNP was almost completely degraded, and the CNP-amino was 34 mg/L. The initial molar concentration of CNP and the final molar concentration of CNP-amino measured on the 14th day were 746 and 118 nmol/L, respectively. Thus, these bacteria transformed 16% of the CNP into CNP-amino, and the remaining CNP (84%) was transformed into unknown compounds. Meanwhile, the CNP concentration in the control, with no bacteria, did not change. Consequently, one can conclude that bacterial activity caused the CNP degradation.
Fig. 3. Decomposition of CNP and formation of CNPamino.
We measured the mutagenicities of the CNP solution at four occasions. At each measurement, the mutagenicity intensity increased linearly with increased dose between 0 and 167 mL/plate (the data are not presented in the paper). Accordingly, we evaluated the mutagenicity of the CNP solution as the mutagenicity intensity at 167 mL of dose. Figure 4 shows the CNP solutions mutagenicities variation during anaerobic biodegradation. In the presence of S9 mix (*), mutagenicities were found to TA98, YG1024 and YG1029 strains, and a weak mutagenicity was found to YG1021 strain. No mutagenicities were detected to TA100 and YG1026 at any cultivating time. While the mutagenicities to TA98, YG1024 and YG1029 strains increased with cultivating time, the mutagenicity to the YG1021 strain, which is not less sensitive than the TA98 strain, did not increase. We do not have an explanation at this time, why the mutagenicity to the YG1021 strain did not increase. On the other hand, no mutagenicity to any strain was detected in the absence of S9 mix (*). Although the CNP solution did not show any mutagenicity before biodegradation, mutagenicity was detected after biodegradation. This indicates that the CNP transformed into some stronger mutagens by biodegradation. To investigate the effect of the medium and its metabolites to the mutagenicity, a test without CNP was conducted: the procedure was the same as in the anaerobic biodegradation test with CNP. No
Fig. 4. Variation on the mutagenicities of the CNP solution to TA and YG strains. means S9mix condition. means+S9mix condition. Spontaneous revertants are as follows: 14 (TA98 S9), 13 (TA98+S9), 52 (YG1021 S9), 30 (YG1021+S9), 35 (YG1024 S9), 35 (YG1024+S9), 61 (TA100 S9), 47 (TA100+S9), 126 (YG1026 S9), 185 (YG1026+S9), 173 (YG1029 S9), 129 (YG1029+S9).
Mutagenicity of CNP during anaerobic biodegradation
mutagenicity was observed to any strain for the culture medium at any cultivating time. Therefore, one can conclude that only CNP and its anaerobic metabolites in the medium induced the mutagenicity. Contribution of CNP, CNP-amino and other metabolites to the mutagenicity We quantitatively estimated the CNP and CNPamino contribution to the mutagenicity of the CNP solution under biodegradation as follows. The CNP mutagenicity induced to the solution was calculated from the CNP concentration regression line from the linear increasing part of dose-response curves. The concentrations of CNP in the solution under biodegradation were in the range of the linear part. The mutagenicity induced by CNP-amino was calculated by the same procedure as CNP. In this research, we presumed only additive effect of CNP and CNP-metabolites to the mutagenicity, and this estimate did not account for possible synergic effects to the mutagenicity among CNP and CNP-metabolites. Figure 5 shows the calculated and observed mutagenicity intensities to TA98 strain (Fig. 5A), YG1024 strain (Fig. 5B) and YG1029 strain (Fig. 5C) in the presence of S9 mix with cultivating time. Two bars are illustrated for each cultivating time: the left one indicates the calculated mutagenicity intensity induced by CNP and CNP-amino, and the right one the observed mutagenicity intensity of the CNP solution. To the TA98 strain (Fig. 5A) and the YG1029 strain (Fig. 5B), the calculated and the observed mutagenicity intensities are almost of the same value until the 3rd day of cultivation. The observed value then became larger than the calculated value after the 7th day of cultivation. The difference may be attributed to unknown mutagens which could not be explained from the amount of CNP and
2593
CNP-amino. Except CNP-amino, we found other anaerobic metabolites inducing mutagenicity in the culture medium. The metabolites inducing mutagenicity to the TA98 strain (Fig. 5A) were indirect mutagen which caused frameshift mutagenicity. The mutagenicity intensity to the TA98 strain induced by these unknown metabolites constituted 47% of the measured value on the 7th day, and 67% on the 14th day. The unknown mutagens detected to the YG1029 strain (Fig. 5B) induced indirect base-pair substitution mutagenicity. Since the YG1029 strain is sensitive to compounds with amino groups, the latter unknown mutagens may possibly have amino-substituents. We estimated that the unknown metabolites contributions to the total mutagenicity intensity were 30% both on the 7th and the 14th day. The contribution of CNP-amino to the mutagenicity was 67% on the 7th and the 14th day. The CNP-amino was the main source of mutagenicity. On the other hand, to the YG1024 strain (Fig. 5C), the observed mutagenicity intensity drastically increased from the 3rd day to the 7th day of the cultivation. The calculated mutagenicity intensity was of the same magnitude as the observed mutagenicity intensity, except the result for the 7th day. On the 7th day, the calculated value was larger than the observed. Although we cannot offer any definite explanation at this moment, one of the possible reasons for this discrepancy is that the synergic effect among the CNP-metabolites decreased the total mutagenicity. Moreover, CNP-amino strongly contributed to the mutagenicity to this strain, in particular after the 7th day, thus, we did not consider other metabolites to affect the mutagenicity to this strain. According to the results of TA98 (Fig. 5A) and YG1024 (Fig. 5C) strains, unknown mutagenic CNP-metabolites induced indirect frameshift mutagenicity, and did not probably have nitro-substituents and amino-substituents. However, as discussed above, there are some exceptional cases in functional
Fig. 5. Contribution of CNP and CNP-amino to the total mutagenicities of the CNP solution in the presence of S9 mix. , CNP-aminto (calculated value); , CNP (calculated value); &, spontaneous revertants; &, observed value.
2594
T. Matsushita et al.
group relationships to a mutagenic sensitivity, and further research is needed. The CNP-amino contributed the most to the whole in those strains which showed positive mutagenicity. Other unknown metabolites did also have a nonnegligible induction. This means that it is important to evaluate the mutagenicity of the whole solution. Otherwise, we must take time and cost-consuming effort to identify all metabolites, and conduct mutagenicity assay on each metabolite. These unknown mutagenic metabolites are classified into two types. One is the indirect frameshift mutagen which does not have nitro-substituents and amino-substituents (Fig. 5A). The other is the indirect base-pair mutagen which may possibly have some aminosubstituents (Fig. 5B). CONCLUSIONS
We measured the mutagenicities of the culture medium during anaerobic biodegradation of CNP. We conclude: 1. CNP was anaerobically biodegraded into CNPamino and other metabolites. 2. Sixteen percent of the initial CNP transformed into CNP-amino. 3. The mutagenicities to the TA98, YG1024 and YG1029 increased with decreasing CNP. Although the CNP solution did not have any mutagenicity before biodegradation, mutagenicity was detected after the biodegradation. Which indicates that CNP transformed into some stronger mutagens by biodegradation. Non-mutagenic compounds have a possibility to change into mutagens by biodegradation. The change in mutagenicity of a compound during biodegradation should be tested regardless of its original mutagenicity. 4. The CNP-amino contributed the most to the mutagenicities. 5. The observed mutagenicity intensity was lager than the mutagenicity intensity calculated from the CNP and CNP-amino concentrations. The difference indicated that other metabolites nonnegligibly contributed to the mutagenicities. If all mutagens cannot be identified, it is important to evaluate the mutagenicity of whole solution. 6. These unknown mutagenic metabolites are classified into two types. One is the indirect frameshift mutagen which does not have nitro-substituents and amino-substituents (Fig. 5(A)). The other is
the indirect base-pair mutagen which may possibly have some amino-substituents (Fig. 5(B)). Acknowledgements}Tomofumi Okuno (Faculty of Pharmaceutical Sciences, Setsunan University) instructed us in Ames assay. Takehiko Nohmi (National Institute of Health Sciences) provided us the Salmonella typhimurium strains. Lars B. Reutergardh reviewed this manuscript. This research was partially supported by the Ministry of Education, Science, Sports and Culture, Grant-in-Aid for Encouragement of Young Scientists, 12750496, 2000.
REFERENCES
Draper W. M. and Casida J. E. (1983) Diphenyl ether herbicides and related compounds: structure–activity relationships as bacterial mutagens. J. Agric. Food Chem. 31(6), 1201–1207. Einisto¨ P., Watanabe M., Ishidate M. Jr. and Nohmi T. (1991) Mutagenicity of 30 chemicals in Salmonella typhimurium strains possessing different nitroreductase or O-acetyltransferase activities. Mutat. Res. 259, 95–102. Kuwatsuka S. (1976) Decompose of herbicide in the soil. Agrichem. Sci. 3(3), 107–122 (in Japanese). Maron D. M. and Ames B. N. (1983) Revised methods for the Salmonella mutagenicity test. Mutat. Res. 113, 173–215. Matsushita T., Nakamichi H., Itoh S. and Sumitomo H. (1997) Evaluating safety of aerobic biodegradation of herbicide CNP. Environ. Engng. Res. 34, 29–34 (in Japanese). Oyamada M. and Kuwatsuka S. (1979,) Degradation of CNP, a diphenyl ether herbicide, in flooded soil under oxidative and reductive conditions. J. Pestic. Sci. 4(2), 157–163. Oyamada M. and Kuwatsuka S. (1989) Microbial metabolism of the herbicide chlornitrofen and its amino derivative. J. Pestic. Sci. 14, 329–335. Takeda A., Ando M., Sekida H., Jinno T. and Matsuda H. (1989) The degradation through the chlorination and the mutagenicity of diphenyl ether herbicide. Proceedings of Safety Evaluation and Fate of Chemicals in the Public Water, pp. 107–126 (in Japanese). Tanaka Y., Iwasaki H. and Kitamori S. (1996) Biodegradation of herbicide chlornitrofen (CNP) and mutagenicity of its degradation products. Water Sci. Technol. 34(7/8), 15–20. Watanabe M., Ishidate M. Jr. and Nohmi T. (1989) A sensitive method for the detection of mutagenic nitroarenes: construction of nitroreductase-overproducing derivatives of Salmonella typhimurium strains TA98 and TA100. Mutat. Res. 216, 211–220. Watanabe M., Ishidate M. Jr. and Nohmi T. (1990) Sensitive method for the detection of mutagenic nitroarenes and aromatic amines: new derivatives of Salmonella typhimurium tester strains possessing elevated O-acetyltransferase levels. Mutat. Res. 234, 337–348. Yamamoto M., Endo K., Nakadaira H. and Mano H. (1993) Epidemiology of biliary tract cancer in Japan: analytical studies. Acta Med. Biol. 41, 127–138.
Wat. Res. Vol. 35, No. 11, pp. 2589–2594, 2001 # 2001 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0043-1354/01/$ - see front matter
THE VARIATION ON THE MUTAGENICITY OF CNP DURING ANAEROBIC BIODEGRADATION TAKU MATSUSHITA1*, SATORU SAKUMA1, KATSUHIKO NAKAMURO2 and YOSHIHIKO MATSUI2 2
1 Department of Civil Engineering, Gifu University, 1-1 Yanagido, Gifu, Gifu 501-1193, Japan and Faculty of Pharmaceutical Sciences, Setsunan University, 45-1, Nagaotoge-cho Hirakata, Osaka 5730101, Japan
(First received 9 June 2000; accepted in revised form 29 October 2000) Abstract}The mutagenicity of water, including herbicide CNP, and its time-variation during anaerobic biodegradation were studied through Ames assay using strains with or without. S9 mix: TA98, TA100, YG1021, YG1024, YG1026, and YG1029. The bacteria, for the anaerobic biodegradation, was obtained from a paddy field, and preincubated for a month. The CNP was decomposed in an anaerobic culture inoculated with the bacteria, and finally yielded CNP-amino as one of the CNP metabolites. About 16% of the initial CNP was transformed into CNP-amino by the 14th day. The mutagenicities to TA98, YG1024, and YG1029 strains with S9 mix increased with cultivating time, the latter two showed the strongest sensitivity to CNP-amino. The contribution of CNP to the mutagenicity decreased as the chemical decomposed, while the contribution of CNP-amino increased. However, the increased mutagenicity was not limited to the contribution of CNP-amino, but also to the contribution of other metabolites. The contributions of other CNP metabolites were 67% of total mutagenicity to the TA98 strain and 30% to the YG1029 strain. These unknown mutagenic metabolites were the indirect frameshift mutagens which did not have nitro- and amino-substituents, and the indirect base-pair mutagens which might possibly have some amino-substituents. # 2001 Elsevier Science Ltd. All rights reserved Key words}CNP, herbicide, ames assay, mutagenicity, anaerobic biodegradation, YG strain
INTRODUCTION
Chlornitrofen (CNP, IUPAC name: 2,4,6-trichlorophenyl 4-nitrophenyl ether, Fig. 1) was extensively used to control paddy weeds in Japan because of its low impact on rice plant. Draper and Casida investigated mutagenicity of several diphenyl ether herbicides including CNP and their substituted compounds by Ames assay using TA100 strain, and reported that CNP was non-mutagenic with or without metabolic activation ( S9 mix) (Draper and Casida, 1983). However, after a relatively high incidence of gallbladder cancer in Niigata, Japan was related to CNP in drinking water as reported by Yamamoto et al. in an epidemiological study, the safety of CNP has attracted attention (Yamamoto et al., 1993). After applied to a paddy field, CNP is decomposed by soil bacteria and transformed into several compounds (Kuwatsuka, 1976; Oyamada and Kuwatsuka, 1979, 1989). Kuwatsuka found that the nitro group in CNP was biologically deoxidized into
*Author to whom all correspondence should be addressed. Tel.: +81-58-294-2444; fax: +81-58-230-1891; e-mail: [email protected]
an amino group, and that CNP-amino (2,4,6trichlorophenyl 4-aminophenyl ether, Fig. 1) was produced in paddy soil during the flooded period (Kuwatsuka, 1976). Kuwatsuka also reported that CNP was mainly decomposed into CNP-amino, CNP-methylamino (2,4,6-trichlorophenyl 4-metylaminophenyl ether), CNP-acetylamino (2,4,6-trichlorophenyl 4-acetylaminophenyl ether), CNPpropylamino (2,4,6-trichlorophenyl 4-propylaminophenyl ether) and CNP-hydroxyl (2,4,6-trichlorophenyl 4-hydroxylphenyl ether) (Kuwatsuka, 1976). These studies suggest that the risk assessment of CNP should be judged in conjunction with the CNP metabolites. Draper and Casida reported CNPamino to be mutagen with S9 mix (Draper and Casida, 1983). This study suggested that the biological degradation could possibly produce other mutagens and not necessarily decrease toxicity. Nevertheless, the toxicities of CNP metabolites, except CNP-amino, have not been investigated. All metabolites of CNP will hardly ever be identified, or their toxicity will hardly ever be evaluated during biodegradation by simply adding the toxicity of each identified metabolite. This methodology cannot estimate if the toxicity will increase or decrease when CNP is applied in the environment.
2589
2590
T. Matsushita et al.
Fig. 1. Chemical structure of CNP (IUPAC name: 2,4,6trichlorophenyl 4-nitrophenyl ether) and CNP-amino (IUPAC name: 2,4,6-trichlorophenyl 4-aminophenyl ether).
extract was dried over anhydrous sodium sulfate, and 200 mL of the extract was directly injected into HPLC. The separation column for the HPLC was 4.6 mm inside diameter 250 mm long filled with Wakosil-II 5C18-100 (Wako Pure Chemicals Industries, Ltd., Osaka, Japan). The mobile phase was acetonitrile: water (70 : 30, v : v). The flow rate was 0.3 mL/min, and detection was recorded at 205 nm for CNP and at 254 nm for CNP-amino. Twenty liters of the sample medium for Ames assay was extracted with 2 L of n-hexane. The extract was dried over anhydrous sodium sulfate, and then evaporated to obtain solid residue. The evaporation was performed using a rotary evaporator at room temperature, under reduced pressure in order to prevent the sample from being transformed by heating. The residual organic matter was redissolved in DMSO of 6 mL. Anaerobic biodegradation
Instead of analyzing the toxicity of CNP and each CNP metabolite as above, whole culture medium was used to determine the toxicity of the CNP solution under the aerobic biological degradation (Tanaka et al., 1996). Tanaka et al. reported that the mutagenicity of the test solution increased during the biological degradation. However, the distinction was not made whether the mutagenicity was due to CNP, CNP metabolites (including CNP-amino), or other constituents in the river water from which the CNP solution was prepared. Matsushita et al. applied chromosomal aberration test to a CNP solution under aerobic biological degradation and found that the induced chromosomal aberration activity did not decrease to its control level even when CNP was not detected in the solution (Matsushita et al., 1997). Accordingly, our aims in this research were the following: to investigate the mutagenicity of the culture media, including CNP, during anaerobic biodegradation, and to classify the mutagenicity into the contributions of CNP, CNP-amino and other unknown metabolites.
MATERIALS AND METHODS
Regents All reagents were purchased from the following sources; S9 (liver homogenates prepared from male Sprague–Dawley rats with pheno-barbital and 5,6-bezoflavone) and co-factor from Oriental Yeast Co., Ltd., Tokyo, Japan; CNP, CNPamino, n-hexane, sodium sulfate, dimethyl sulfoxide (DMSO), acetonitrile and other chemicals from Wako Pure Chemicals Industries, Ltd., Osaka, Japan. A stock solution of CNP was prepared by dissolving the reagent CNP into acetone (4000 mg/L). The CNP solution was then prepared by diluting this stock solution with water: undissolved particles of CNP were removed by filtration through a paper filter of No. 5C (Advantec Toyo, Tokyo, Japan). Methods of measurements CNP and CNP-amino were quantified by reverse-phase high-performance liquid chromatography (HPLC, 600 Controller, 717 Plus Autosampler and 486 Tunable Absorbance Detector, Waters Co., Milford, MA, USA) after extracting using n-hexane. Hundred milliliters of the sample medium was extracted with 10 mL of n-hexane. The
Anaerobic bacteria originating from a paddy field soil on the west side of Gifu University were used for the CNP biodegradation experiments after preincubation. The soil was collected under flooded conditions in spring, 1999. The wet soil was stored at 48C until use. One gram of the wet soil and the CNP solution was added to a 500 mL bottle. The culture medium described below was then added to the bottle. After introducing nitrogen gas into the bottle, the mouth of the bottle was covered with a frosted glass lid and sealed with Parafilm. The bottle was under anaerobic condition at 208C in the dark for a month for preincubating. We conducted the CNP degradation experiments with the preincubated bacteria: the bacteria was inoculated by injecting a small portion of the culture solution from the bottle. The CNP degradation was proceeded in 10 L glass vessel filled with liquid medium. The vessel was placed in the dark. The liquid medium was K2HPO4: 350 mg/L, KH2PO4: 270 mg/L, NH4Cl: 530 mg/L, MgCl26H2O: 100 mg/L, CaCl22H2O: 75 mg/L, NaHCO3: 1200 mg/L, Na2S9H2O: 500 mg/L, yeast extract: 30 mg/L. During the biodegradation, the oxidation-reduction potential was maintained below 470 mV. Thus, we consider the culture medium to be under anaerobic condition. Ames assay Assays were performed in the presence and absence of a rat liver activation system: the method was according to the 20-min preincubation procedure described by Maron and Ames (Maron and Ames, 1983). The test strains used in this assay were YG1021, YG1024, YG1026 and YG1029 in addition to the classical strains, TA98 and TA100. These YG strains with an elevated level of classical nitroreductase or O-acetyltransferase were developed by cloning the corresponding gene into pBR322 plasmids and introducing those plasmids carrying the gene into TA98 and TA100 (Watanabe et al., 1989; 1990). Further details about these strains are described in Table 1. These bacteria were grown in 20 mL of Oxoid nutrient broth No. 2 and incubated for 16 h at 378C in a shaking water bath in order to insure adequate aeration. 2-Nitrofluorene was used as the positive control chemical in the absence of S9 mix, and 2-aminoanthracene was used in the presence of S9 mix. S9 was used as the external metabolic activation system at a final concentration of 10% in a co-factor mix. Treatments without S9 mix were conducted by adding either 0.2 mL of DMSO or 0.0125-0.05 mL of a solution of the test sample to 0.5 mL of Na–K buffer. Treatments with metabolic activation were conducted by adding 0.5 mL of S9 mix to the test sample instead. These reagents were mixed and incubated in a tube with 0.1 mL of the test-strains for 20 min in a shaking water bath at 378C. After adding 2 mL
Mutagenicity of CNP during anaerobic biodegradation
2591
Table 1. Salmonella typhimurium strains used in this study Strain
Description
TA98 YG1021 YG1024 TA100 YG1026 YG1029
his D3052 (pKM101) TA98 (pYG219): a nitroreductase-ovrproducing strain TA98 (pYG219): an O-acetyltransferase-overproducing strain his G46 (pKM101) TA100 (pYG216): a nitroreductase-overproducing strain TA100 (pYG219): an O-acetyltransferase-overproducing strain
of the top agar to the tube, the resulting mixture was plated on a glucose minimal agar plate. Revertant colonies (His+ revertants) were counted after incubating at 378C for 48 h. The number of His+ revertants for each sample was given as the mean value of the two plates. To quantify the magnitude of the mutagenicity, we calculated the ratio of the number of His+ revertants per plate to the number of spontaneous revertants per plate. In this paper, this ratio is referred to as ‘the mutagenicity intensity’. Compounds of which mutagenicity intensities were larger than the two which were considered to be mutagenic.
RESULTS AND DISCUSSION
Mutagenicities of CNP and CNP-amino Figure 2 shows the mutagenicities of CNP and CNP-amino to TA and YG strains. CNP was nonmutagenic in the standard Ames test with test-strains that detect frame shift (TA98) and base-pair substitution (TA100) either in the absence or presence of S9 mix. This result is in accordance with the results obtained by other researchers (Takeda et al., 1989). Although CNP has one nitro group, no mutagenicity was detected using YG1021 and YG1026 strains, which were developed to enhance sensitivity to nitroarenes (nitroreductase-overproducing strain). Mutagenicity was detected using YG1024 and YG1029 strains, which have enhanced sensitivity to aminoarenes (O-acetyltransferase-overproducing strain). Einisto¨ et al. (1991) reported the same phenomenon for 1,8-dinitropyrene, which has two nitro groups (Einisto¨ et al., 1,8-dinitropyrene, out of more than 10 compounds with nitro groups, did not show exceptionally stronger mutagenicity to the YG1021 strain than the TA98 strain. On the contrary, a much stronger mutagenicity was detected to the YG1024. We believe this phenomenon was caused by the following processes: In Salmonella typhimurium, the CNP nitro group is not reduced by overproduced nitroreductase of YG1021 and YG1026, but by other nitroreductase which the classical strains already have. CNP, therefore, did not exhibit positive mutagenicity to the nitroreductase-overproducing strains (YG1021 and YG1026). Since reduced CNP is then metabolized by O-acetyltransferase which is enhanced in YG1024 and YG1029 strains, the strong mutagenicity is detected to the O-acetyltransferase-overproducing strains (YG1024 and YG1029 strains). On the other hand, CNP-amino exhibited positive mutagenic activity to the two classical strains in the
Fig. 2. Mutagenicities of CNP and CNP-amino to TA and YG strains. mean S9mix condition, means+S9mix condition. Spontaneous revertants are as follows: 17 (TA98 S9), 15 (TA98+S9), 42 (YG1021 S9), 35 (YG1021+S9), 40 (YG1024 S9), 31 (YG1024+S9), 17 (YG1041 S9), 22 (YG1041+S9), 88 (TA100 S9), 53 (TA100+S9), 114 (YG1026 S9), 162 (YG1026+S9), 152 (YG1029 S9), 144 (YG1029+S9).
presence of S9 mix. These results also agree with the suggestion of other researchers (Takeda et al., 1989). Moreover, the mutagenicity of CNP-amino was detected in all YG strains in the presence of S9 mix. Overall, the mutagenicity of CNP-amino was much stronger relative to CNP. Variation on the mutagenicity during anaerobic biodegradation Figure 3 shows concentration of CNP and CNPamino during anaerobic biodegradation. CNP was gradually decomposed by bacteria in 14 days, and
2592
T. Matsushita et al.
yielded CNP-amino as one of the metabolites. At the beginning of cultivation, the CNP concentration was 238 mg/L. After three days of the cultivation the CNP concentration was lowered to 206 mg/L, and produced 1 mg/L of CNP-amino. On the 14th day of the cultivation, the CNP was almost completely degraded, and the CNP-amino was 34 mg/L. The initial molar concentration of CNP and the final molar concentration of CNP-amino measured on the 14th day were 746 and 118 nmol/L, respectively. Thus, these bacteria transformed 16% of the CNP into CNP-amino, and the remaining CNP (84%) was transformed into unknown compounds. Meanwhile, the CNP concentration in the control, with no bacteria, did not change. Consequently, one can conclude that bacterial activity caused the CNP degradation.
Fig. 3. Decomposition of CNP and formation of CNPamino.
We measured the mutagenicities of the CNP solution at four occasions. At each measurement, the mutagenicity intensity increased linearly with increased dose between 0 and 167 mL/plate (the data are not presented in the paper). Accordingly, we evaluated the mutagenicity of the CNP solution as the mutagenicity intensity at 167 mL of dose. Figure 4 shows the CNP solutions mutagenicities variation during anaerobic biodegradation. In the presence of S9 mix (*), mutagenicities were found to TA98, YG1024 and YG1029 strains, and a weak mutagenicity was found to YG1021 strain. No mutagenicities were detected to TA100 and YG1026 at any cultivating time. While the mutagenicities to TA98, YG1024 and YG1029 strains increased with cultivating time, the mutagenicity to the YG1021 strain, which is not less sensitive than the TA98 strain, did not increase. We do not have an explanation at this time, why the mutagenicity to the YG1021 strain did not increase. On the other hand, no mutagenicity to any strain was detected in the absence of S9 mix (*). Although the CNP solution did not show any mutagenicity before biodegradation, mutagenicity was detected after biodegradation. This indicates that the CNP transformed into some stronger mutagens by biodegradation. To investigate the effect of the medium and its metabolites to the mutagenicity, a test without CNP was conducted: the procedure was the same as in the anaerobic biodegradation test with CNP. No
Fig. 4. Variation on the mutagenicities of the CNP solution to TA and YG strains. means S9mix condition. means+S9mix condition. Spontaneous revertants are as follows: 14 (TA98 S9), 13 (TA98+S9), 52 (YG1021 S9), 30 (YG1021+S9), 35 (YG1024 S9), 35 (YG1024+S9), 61 (TA100 S9), 47 (TA100+S9), 126 (YG1026 S9), 185 (YG1026+S9), 173 (YG1029 S9), 129 (YG1029+S9).
Mutagenicity of CNP during anaerobic biodegradation
mutagenicity was observed to any strain for the culture medium at any cultivating time. Therefore, one can conclude that only CNP and its anaerobic metabolites in the medium induced the mutagenicity. Contribution of CNP, CNP-amino and other metabolites to the mutagenicity We quantitatively estimated the CNP and CNPamino contribution to the mutagenicity of the CNP solution under biodegradation as follows. The CNP mutagenicity induced to the solution was calculated from the CNP concentration regression line from the linear increasing part of dose-response curves. The concentrations of CNP in the solution under biodegradation were in the range of the linear part. The mutagenicity induced by CNP-amino was calculated by the same procedure as CNP. In this research, we presumed only additive effect of CNP and CNP-metabolites to the mutagenicity, and this estimate did not account for possible synergic effects to the mutagenicity among CNP and CNP-metabolites. Figure 5 shows the calculated and observed mutagenicity intensities to TA98 strain (Fig. 5A), YG1024 strain (Fig. 5B) and YG1029 strain (Fig. 5C) in the presence of S9 mix with cultivating time. Two bars are illustrated for each cultivating time: the left one indicates the calculated mutagenicity intensity induced by CNP and CNP-amino, and the right one the observed mutagenicity intensity of the CNP solution. To the TA98 strain (Fig. 5A) and the YG1029 strain (Fig. 5B), the calculated and the observed mutagenicity intensities are almost of the same value until the 3rd day of cultivation. The observed value then became larger than the calculated value after the 7th day of cultivation. The difference may be attributed to unknown mutagens which could not be explained from the amount of CNP and
2593
CNP-amino. Except CNP-amino, we found other anaerobic metabolites inducing mutagenicity in the culture medium. The metabolites inducing mutagenicity to the TA98 strain (Fig. 5A) were indirect mutagen which caused frameshift mutagenicity. The mutagenicity intensity to the TA98 strain induced by these unknown metabolites constituted 47% of the measured value on the 7th day, and 67% on the 14th day. The unknown mutagens detected to the YG1029 strain (Fig. 5B) induced indirect base-pair substitution mutagenicity. Since the YG1029 strain is sensitive to compounds with amino groups, the latter unknown mutagens may possibly have amino-substituents. We estimated that the unknown metabolites contributions to the total mutagenicity intensity were 30% both on the 7th and the 14th day. The contribution of CNP-amino to the mutagenicity was 67% on the 7th and the 14th day. The CNP-amino was the main source of mutagenicity. On the other hand, to the YG1024 strain (Fig. 5C), the observed mutagenicity intensity drastically increased from the 3rd day to the 7th day of the cultivation. The calculated mutagenicity intensity was of the same magnitude as the observed mutagenicity intensity, except the result for the 7th day. On the 7th day, the calculated value was larger than the observed. Although we cannot offer any definite explanation at this moment, one of the possible reasons for this discrepancy is that the synergic effect among the CNP-metabolites decreased the total mutagenicity. Moreover, CNP-amino strongly contributed to the mutagenicity to this strain, in particular after the 7th day, thus, we did not consider other metabolites to affect the mutagenicity to this strain. According to the results of TA98 (Fig. 5A) and YG1024 (Fig. 5C) strains, unknown mutagenic CNP-metabolites induced indirect frameshift mutagenicity, and did not probably have nitro-substituents and amino-substituents. However, as discussed above, there are some exceptional cases in functional
Fig. 5. Contribution of CNP and CNP-amino to the total mutagenicities of the CNP solution in the presence of S9 mix. , CNP-aminto (calculated value); , CNP (calculated value); &, spontaneous revertants; &, observed value.
2594
T. Matsushita et al.
group relationships to a mutagenic sensitivity, and further research is needed. The CNP-amino contributed the most to the whole in those strains which showed positive mutagenicity. Other unknown metabolites did also have a nonnegligible induction. This means that it is important to evaluate the mutagenicity of the whole solution. Otherwise, we must take time and cost-consuming effort to identify all metabolites, and conduct mutagenicity assay on each metabolite. These unknown mutagenic metabolites are classified into two types. One is the indirect frameshift mutagen which does not have nitro-substituents and amino-substituents (Fig. 5A). The other is the indirect base-pair mutagen which may possibly have some aminosubstituents (Fig. 5B). CONCLUSIONS
We measured the mutagenicities of the culture medium during anaerobic biodegradation of CNP. We conclude: 1. CNP was anaerobically biodegraded into CNPamino and other metabolites. 2. Sixteen percent of the initial CNP transformed into CNP-amino. 3. The mutagenicities to the TA98, YG1024 and YG1029 increased with decreasing CNP. Although the CNP solution did not have any mutagenicity before biodegradation, mutagenicity was detected after the biodegradation. Which indicates that CNP transformed into some stronger mutagens by biodegradation. Non-mutagenic compounds have a possibility to change into mutagens by biodegradation. The change in mutagenicity of a compound during biodegradation should be tested regardless of its original mutagenicity. 4. The CNP-amino contributed the most to the mutagenicities. 5. The observed mutagenicity intensity was lager than the mutagenicity intensity calculated from the CNP and CNP-amino concentrations. The difference indicated that other metabolites nonnegligibly contributed to the mutagenicities. If all mutagens cannot be identified, it is important to evaluate the mutagenicity of whole solution. 6. These unknown mutagenic metabolites are classified into two types. One is the indirect frameshift mutagen which does not have nitro-substituents and amino-substituents (Fig. 5(A)). The other is
the indirect base-pair mutagen which may possibly have some amino-substituents (Fig. 5(B)). Acknowledgements}Tomofumi Okuno (Faculty of Pharmaceutical Sciences, Setsunan University) instructed us in Ames assay. Takehiko Nohmi (National Institute of Health Sciences) provided us the Salmonella typhimurium strains. Lars B. Reutergardh reviewed this manuscript. This research was partially supported by the Ministry of Education, Science, Sports and Culture, Grant-in-Aid for Encouragement of Young Scientists, 12750496, 2000.
REFERENCES
Draper W. M. and Casida J. E. (1983) Diphenyl ether herbicides and related compounds: structure–activity relationships as bacterial mutagens. J. Agric. Food Chem. 31(6), 1201–1207. Einisto¨ P., Watanabe M., Ishidate M. Jr. and Nohmi T. (1991) Mutagenicity of 30 chemicals in Salmonella typhimurium strains possessing different nitroreductase or O-acetyltransferase activities. Mutat. Res. 259, 95–102. Kuwatsuka S. (1976) Decompose of herbicide in the soil. Agrichem. Sci. 3(3), 107–122 (in Japanese). Maron D. M. and Ames B. N. (1983) Revised methods for the Salmonella mutagenicity test. Mutat. Res. 113, 173–215. Matsushita T., Nakamichi H., Itoh S. and Sumitomo H. (1997) Evaluating safety of aerobic biodegradation of herbicide CNP. Environ. Engng. Res. 34, 29–34 (in Japanese). Oyamada M. and Kuwatsuka S. (1979,) Degradation of CNP, a diphenyl ether herbicide, in flooded soil under oxidative and reductive conditions. J. Pestic. Sci. 4(2), 157–163. Oyamada M. and Kuwatsuka S. (1989) Microbial metabolism of the herbicide chlornitrofen and its amino derivative. J. Pestic. Sci. 14, 329–335. Takeda A., Ando M., Sekida H., Jinno T. and Matsuda H. (1989) The degradation through the chlorination and the mutagenicity of diphenyl ether herbicide. Proceedings of Safety Evaluation and Fate of Chemicals in the Public Water, pp. 107–126 (in Japanese). Tanaka Y., Iwasaki H. and Kitamori S. (1996) Biodegradation of herbicide chlornitrofen (CNP) and mutagenicity of its degradation products. Water Sci. Technol. 34(7/8), 15–20. Watanabe M., Ishidate M. Jr. and Nohmi T. (1989) A sensitive method for the detection of mutagenic nitroarenes: construction of nitroreductase-overproducing derivatives of Salmonella typhimurium strains TA98 and TA100. Mutat. Res. 216, 211–220. Watanabe M., Ishidate M. Jr. and Nohmi T. (1990) Sensitive method for the detection of mutagenic nitroarenes and aromatic amines: new derivatives of Salmonella typhimurium tester strains possessing elevated O-acetyltransferase levels. Mutat. Res. 234, 337–348. Yamamoto M., Endo K., Nakadaira H. and Mano H. (1993) Epidemiology of biliary tract cancer in Japan: analytical studies. Acta Med. Biol. 41, 127–138.