Mutation Research, 244 (1990) 55-60 Elsevier
55
MUTLET 0342
D N A damage and prophage induction and toxicity of nitrofurantoin in Escheriehia coli and Vibrio cholerae cells Sukla Sengupta, Md.S. Rahman, U. Mukherjee, Jayasri Basak, A.K. Pal and S.N. Chatterjee Biophysics Division. Saha ln.~titute ~f Nuclear Physics. Cah'utta - 700 037 (India)
(Accepted 21 December 1989)
Keyword~'."Nitrofurantoin; Toxicity; DNA damage; Prophage induction
Summary Repair-deficient and repair-proficient strains of E. coli K12 were sensitive to nitrofurantoin treatment to varying degrees with the double mutant strain (uvrA 6, recA 13) being most sensitive. Ultraviolet absorption data and thermal chromatography through a hydroxyapatite column revealed that nitrofurantoin treatment of V. cholerae strain O G A W A 154 produced a maximal amount of 55°70 reversibly bihelical D N A at a nitrofurantoin dose of 120 #.g/ml/h, which indicated the formation of inter-strand cross-links in DNA. Nitrofurantoin also produced prophage-~, induction in E. coli K I2 strain GY 5027: e n v A , uvrB, a m p A 1, strA 0~), in a dose-dependent manner, the maximum induction being highly significant ( P < 0.(301). Previously published mutation data coupled with the prophage induction data presented here suggest that the genotoxic properties of nitrofurantoin are mediated through the SOS pathway.
Several of the nitrofuran derivatives, including nitrofurantoin, are widely used as antibacterial agents. Nitrofurantoin was reported to be unusually toxic to bacteria that are deficient in DNA recombination repair, particularly those with a mutated recA 13 gene (Jenkins and Bennett, 1976; Lu et ai., 1979). Lu et al. (1979) reported the relative toxicities of several nitrofurans including nitrofurantoin and showed that the mutagenic ac-
Correspondence: Dr. S.N. Chatterjee, Biophysics Division, Saha Institute of Nuclear Physics, 37 Belgachia Road, Calcutta - 700 037 (India).
tivities of these drugs, in general, paralleled toxicity. Nitrofurantoin was also shown to be rec-test positive (Kada et al., 1972; Tazima et al., 1975) in different bacterial systems and was also mutagenic in E. coli WP2 strains. Although nitrofurantoin was reported not to be carcinogenic in the rat (Erturk et al., 1970), the results were not conclusive since an inadequate number of animals were used for the test (McCalla, 1983). This communication examines the nitrofurantoin sensitivity of a further series of E. coli strains harboring mutations in genes involved in excision, recombination, and SOS repair pathways. In view of the reported broad antibacterial spectrum of the
3165-7992/90/$ 03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
56 drug, this study has used 2 bacterial systems, E. coli and V. cholerae, to study the killing and also induction of DNA damage by nitrofurantoin. We have used another short-term test, the prophage induction assay, to obtain further evidence about its mutagenic potential. The phage induction assay was carried out with the specially constructed permeable strain of E. coli (Moreau et al., 1976). This modified induction assay has detected many carcinogens (Moreau and Devoret, 1977; Rojanapo et al., 1981). This communication also presents evidence of the nature and extent of DNA damage produced by nitrofurantoin in a bacterial cell. Materials and methods
Bacteria The bacterial strains used in the study were: Escherichia coli K12 strains AB 1157 (repairproficient), AB 1899 (Ion 1), AB 2470 (recB 21), AB 1884 (uvrC 34), AB 1885 (uvrB 5), AB 1886 (uvrA 6), AB 2463 (recA 13), AB 2480 (uvrA 6, recA 13). These strains were obtained through the kind courtesy of Dr. Barbara J. Bachmann, E. coil Genetic Stock Center (U.S.A.); Vibrio cholerae (classical) strain OGAWA 154 was obtained through the kind courtesy of Dr. S.C. Pal, Director, National Institute of Cholera and Enteric Diseases, Calcutta. The strains used for the prophage induction study were E. coli K12 GY 5027 (envA, uvrB, ampA 1, strA 0~)) (a specially constructed permeable cell) and GY 4015 (ampA 601, ~, indicator) and were obtained through the kind courtesy of Dr. R. Devoret (France). Media The culture media used were: (i) NB medium containing 1 g Bacto Peptone (Difco), 1 g Lab Lemco powder (Oxoid), 0.5 g NaC1 and 100 ml water, pH being adjusted to 7.2 for E. coil and 8.0 for V. cholerae; (ii) Nutrient agar (NA) plates prepared by adding 1.3070 Bacto-Agar (Difco) to the NB medium; (iii) LB medium containing 5 g yeast extract (Oxoid), 10 g tryptone (Oxoid), 10 g NaC1, distilled water 1 liter; (iv) LBE containing LB medium supplemented with 0.2°7o glucose and
20O'/o medium E of Vogel and Bonner (1956); (v) GT-ampicillin plates containing 8 g Bacto Peptone (Difco), 5 g tryptone (Oxoid), 5 g NaCI, 13 g BactoAgar (Difco) and I liter distilled water, supplemented with 10 ~g/ml o-ampicillin; and (vi) soft agar containing 7 g Bacto-Agar (Difco) in 1 liter distilled water.
Chemicals Chemically pure nitrofurantoin was obtained from Sigma Chemical Co. (U.S.A.). All other chemicals used were of analytical grade. Nitrofurantoin treatment and assay o f bacterial viability Aliquots (1 ml) of log phase cells were inoculated into 20 ml of NB medium containing various concentrations of drug, incubated at 37°C in the dark for a fixed period of 1 h and then immediately assayed for viability on drug-free NA plates after appropriate serial dilution in saline (0.85°7o). Colonies were scored after overnight incubation at 37°C in dark. 37o7osurvival doses (D37) were determined from a semilogarithmic plot of bacterial survival against drug concentration. The term 'dose' is used in this paper to mean exposure dose and is expressed in #g/ml/h. DNA isolation and estimation o f inter-strand cross-links DNA was isolated from drug-treated or untreated bacterial cells and the purity assayed following the methods described earlier (Marmur, 1961; Chatterjee et al., 1983). The isolated DNA was subjected to thermal denaturation-renaturation treatment (DRT) after brief shearing by ultrasonic radiation (Cole, 1970: Chatterjee et al., 1983). Cross-linked DNA was estimated in terms of the fraction of isolated DNA rendered reversibly bihelical by DRT as monitored by UV absorption spectrophotometry (Rutman et al., 1969; Chatterjee et al., 1987) or by thermal chromatography through a hydroxyapatite column (Miyazawa and Thomas, 1965; Chatterjee et al., 1983, 1987). Brief shearing by sonication degrades the DNA molecules into smaller fragments. When a DNA
57
fragment bearing one or more inter-strand covalent links (cross-links) is denatured by heating, the 2 strands do not separate completely, unlike native DNA, but remain joined to each other by the covalent link(s). On rapid or quench cooling, the 2 strands of the DNA fragment get bridged to each other all along their length as if by a zippering action and the DNA fragments thus regain the native bihelical structure. DNA acquiring this reversibly bihelical nature was described for the first time by Geiduschek (1961) as characteristic of DNA bearing inter-strand cross-links.
Prophage-X induction Prophage k induction by nitrofurantoin was assessed turbidimetrically and in a plaque-forming unit assay (Moreau et al., 1976; Pal and Chatterjee, 1985). Results The repair-deficient and repair-proficient strains of E. coil K12 were sensitive to nitrofurantoin treatment in varying degrees (Fig. l). The double mutant strain (uvrA 6, recA 13) was most sensitive, and the D37 values of the E. coil Kl2 strains in order of sensitivity to nitrofurantoin were repairproficient (17.0) < Ion 1 (7.0) < recB 21 (4.9) < lex 1 (4.4)
55
MUTLET 0342
D N A damage and prophage induction and toxicity of nitrofurantoin in Escheriehia coli and Vibrio cholerae cells Sukla Sengupta, Md.S. Rahman, U. Mukherjee, Jayasri Basak, A.K. Pal and S.N. Chatterjee Biophysics Division. Saha ln.~titute ~f Nuclear Physics. Cah'utta - 700 037 (India)
(Accepted 21 December 1989)
Keyword~'."Nitrofurantoin; Toxicity; DNA damage; Prophage induction
Summary Repair-deficient and repair-proficient strains of E. coli K12 were sensitive to nitrofurantoin treatment to varying degrees with the double mutant strain (uvrA 6, recA 13) being most sensitive. Ultraviolet absorption data and thermal chromatography through a hydroxyapatite column revealed that nitrofurantoin treatment of V. cholerae strain O G A W A 154 produced a maximal amount of 55°70 reversibly bihelical D N A at a nitrofurantoin dose of 120 #.g/ml/h, which indicated the formation of inter-strand cross-links in DNA. Nitrofurantoin also produced prophage-~, induction in E. coli K I2 strain GY 5027: e n v A , uvrB, a m p A 1, strA 0~), in a dose-dependent manner, the maximum induction being highly significant ( P < 0.(301). Previously published mutation data coupled with the prophage induction data presented here suggest that the genotoxic properties of nitrofurantoin are mediated through the SOS pathway.
Several of the nitrofuran derivatives, including nitrofurantoin, are widely used as antibacterial agents. Nitrofurantoin was reported to be unusually toxic to bacteria that are deficient in DNA recombination repair, particularly those with a mutated recA 13 gene (Jenkins and Bennett, 1976; Lu et ai., 1979). Lu et al. (1979) reported the relative toxicities of several nitrofurans including nitrofurantoin and showed that the mutagenic ac-
Correspondence: Dr. S.N. Chatterjee, Biophysics Division, Saha Institute of Nuclear Physics, 37 Belgachia Road, Calcutta - 700 037 (India).
tivities of these drugs, in general, paralleled toxicity. Nitrofurantoin was also shown to be rec-test positive (Kada et al., 1972; Tazima et al., 1975) in different bacterial systems and was also mutagenic in E. coli WP2 strains. Although nitrofurantoin was reported not to be carcinogenic in the rat (Erturk et al., 1970), the results were not conclusive since an inadequate number of animals were used for the test (McCalla, 1983). This communication examines the nitrofurantoin sensitivity of a further series of E. coli strains harboring mutations in genes involved in excision, recombination, and SOS repair pathways. In view of the reported broad antibacterial spectrum of the
3165-7992/90/$ 03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
56 drug, this study has used 2 bacterial systems, E. coli and V. cholerae, to study the killing and also induction of DNA damage by nitrofurantoin. We have used another short-term test, the prophage induction assay, to obtain further evidence about its mutagenic potential. The phage induction assay was carried out with the specially constructed permeable strain of E. coli (Moreau et al., 1976). This modified induction assay has detected many carcinogens (Moreau and Devoret, 1977; Rojanapo et al., 1981). This communication also presents evidence of the nature and extent of DNA damage produced by nitrofurantoin in a bacterial cell. Materials and methods
Bacteria The bacterial strains used in the study were: Escherichia coli K12 strains AB 1157 (repairproficient), AB 1899 (Ion 1), AB 2470 (recB 21), AB 1884 (uvrC 34), AB 1885 (uvrB 5), AB 1886 (uvrA 6), AB 2463 (recA 13), AB 2480 (uvrA 6, recA 13). These strains were obtained through the kind courtesy of Dr. Barbara J. Bachmann, E. coil Genetic Stock Center (U.S.A.); Vibrio cholerae (classical) strain OGAWA 154 was obtained through the kind courtesy of Dr. S.C. Pal, Director, National Institute of Cholera and Enteric Diseases, Calcutta. The strains used for the prophage induction study were E. coli K12 GY 5027 (envA, uvrB, ampA 1, strA 0~)) (a specially constructed permeable cell) and GY 4015 (ampA 601, ~, indicator) and were obtained through the kind courtesy of Dr. R. Devoret (France). Media The culture media used were: (i) NB medium containing 1 g Bacto Peptone (Difco), 1 g Lab Lemco powder (Oxoid), 0.5 g NaC1 and 100 ml water, pH being adjusted to 7.2 for E. coil and 8.0 for V. cholerae; (ii) Nutrient agar (NA) plates prepared by adding 1.3070 Bacto-Agar (Difco) to the NB medium; (iii) LB medium containing 5 g yeast extract (Oxoid), 10 g tryptone (Oxoid), 10 g NaC1, distilled water 1 liter; (iv) LBE containing LB medium supplemented with 0.2°7o glucose and
20O'/o medium E of Vogel and Bonner (1956); (v) GT-ampicillin plates containing 8 g Bacto Peptone (Difco), 5 g tryptone (Oxoid), 5 g NaCI, 13 g BactoAgar (Difco) and I liter distilled water, supplemented with 10 ~g/ml o-ampicillin; and (vi) soft agar containing 7 g Bacto-Agar (Difco) in 1 liter distilled water.
Chemicals Chemically pure nitrofurantoin was obtained from Sigma Chemical Co. (U.S.A.). All other chemicals used were of analytical grade. Nitrofurantoin treatment and assay o f bacterial viability Aliquots (1 ml) of log phase cells were inoculated into 20 ml of NB medium containing various concentrations of drug, incubated at 37°C in the dark for a fixed period of 1 h and then immediately assayed for viability on drug-free NA plates after appropriate serial dilution in saline (0.85°7o). Colonies were scored after overnight incubation at 37°C in dark. 37o7osurvival doses (D37) were determined from a semilogarithmic plot of bacterial survival against drug concentration. The term 'dose' is used in this paper to mean exposure dose and is expressed in #g/ml/h. DNA isolation and estimation o f inter-strand cross-links DNA was isolated from drug-treated or untreated bacterial cells and the purity assayed following the methods described earlier (Marmur, 1961; Chatterjee et al., 1983). The isolated DNA was subjected to thermal denaturation-renaturation treatment (DRT) after brief shearing by ultrasonic radiation (Cole, 1970: Chatterjee et al., 1983). Cross-linked DNA was estimated in terms of the fraction of isolated DNA rendered reversibly bihelical by DRT as monitored by UV absorption spectrophotometry (Rutman et al., 1969; Chatterjee et al., 1987) or by thermal chromatography through a hydroxyapatite column (Miyazawa and Thomas, 1965; Chatterjee et al., 1983, 1987). Brief shearing by sonication degrades the DNA molecules into smaller fragments. When a DNA
57
fragment bearing one or more inter-strand covalent links (cross-links) is denatured by heating, the 2 strands do not separate completely, unlike native DNA, but remain joined to each other by the covalent link(s). On rapid or quench cooling, the 2 strands of the DNA fragment get bridged to each other all along their length as if by a zippering action and the DNA fragments thus regain the native bihelical structure. DNA acquiring this reversibly bihelical nature was described for the first time by Geiduschek (1961) as characteristic of DNA bearing inter-strand cross-links.
Prophage-X induction Prophage k induction by nitrofurantoin was assessed turbidimetrically and in a plaque-forming unit assay (Moreau et al., 1976; Pal and Chatterjee, 1985). Results The repair-deficient and repair-proficient strains of E. coil K12 were sensitive to nitrofurantoin treatment in varying degrees (Fig. l). The double mutant strain (uvrA 6, recA 13) was most sensitive, and the D37 values of the E. coil Kl2 strains in order of sensitivity to nitrofurantoin were repairproficient (17.0) < Ion 1 (7.0) < recB 21 (4.9) < lex 1 (4.4)
