Mutagenicity of nitrosamides and nitrosamidines in micro-organisms and plants

Effectiveness of repeated mutagenesis of sesame crosses for enhancing polygenic variability in F2M2 generation

The value of combining hybridization and mutagenesis in sesame was examined to determine if treating hybrid sesame plant material with mutagens generated greater genetic variability in four key productivity traits than either the separate hybridization or mutation of plant material. In a randomized block design with three replications, six F2M2 varieties, three F2varieties, and three parental varieties were assessed at Odisha University of Agriculture and Technology, Bhubaneswar, Odisha, India. The plant characteristics height, number of seed capsules per plant, and seed yield per plant had greater variability in the F2M2 generation than their respective controls (F2), however, the number of primary branches per plant varied less than in the control population. The chances for trait selection to be operative were high for all the characteristics examined except the number of primary branches per plant, as indicated by heritability estimates. Increases in the mean and variability of the characteristics examined indicted a greater incidence of beneficial mutations and the breakdown of undesirable linkages with increased recombination. At both phenotypic and genotypic levels strong positive correlations between both primary branch number and capsule number with seed yield suggest that these traits are important for indirect improvement in sesame seed yield. As a result of the association analysis, sesame seed yield and its component traits improved significantly, which may be attributed to the independent polygenic mutations and enlarged recombination of the polygenes controlling the examined characteristics. Compared to the corresponding control treatment or to one cycle of mutagenic treatment, two cycles of mutagenic treatment resulted in increased variability, higher transgressive segregates, PTS mean and average transgression for sesame seed yield. These findings highlight the value of implementing two EMS treatment cycles to generate improved sesame lines. Furthermore, the extra variability created through hybridization may have potential in subsequent breeding research and improved seed yield segregants may be further advanced to develop ever-superior sesame varieties.

Comparative mutagenicity of chemicals selected for test in the International Program on Chemical Safety's collaborative study on plant systems for the detection of environmental mutagens

A review has been made for the four compounds (maleic hydrazide, methyl nitrosourea, sodium azide, azidoglycerol) tested in the International Program on Chemical Safety's collaborative study on plant systems. Maleic hydrazide (MH) is a weak cytotoxic/mutagenic chemical in mammalian tissues and is classified as a class 4 chemical. In contrast, with few exceptions such as Arabidopsis, MH is a potent mutagen/clastogen in plant systems. The difference in its response between plant and animal tissue is likely due to differences in the way MH is metabolized. MH appears to be noncarcinogenic and has been given a negative NCI/NTP carcinogen rating. Methyl nitrosourea (MNU) is a toxic, mutagenic, radiomimetic, carcinogenic, and teratogenic chemical. It has been shown to be a mutagen in bacteria, fungi, Drosophila, higher plants, and animal cells both in vitro and in vivo. MNU is a clastogen in both animal and human cell cultures, plant root tips and cell cultures inducing both chromosome and chromatid aberrations as well as sister-chromatid exchanges. Carcinogenicity has been confirmed in numerous studies and involves the nervous system, intestine, kidney, stomach, bladder and uterus, in the rat, mouse, and hamster. MNU produces stage-specific teratogenic effects and also interferes with embryonic development. The experimental evidence that strongly indicates the mutagenic effects of MNU underlines the possible hazard of this compound to human beings. The experimental evidence for the stringent handling of this compound is clear. Sodium azide (NaN3) is cytotoxic in several animal and plant systems and functions by inhibiting protein synthesis and replicative DNA synthesis at low dosages. It is mutagenic in bacteria, higher plants and human cells and has been used as a positive control in some systems. In general, tests for clastogenicity have been negative or weakly positive. No evidence of carcinogenicity has been reported in a 2-year study seeking carcinogenic activity in male and female rats. Its advantages in comparison to other efficient mutagens are claimed to be a high production of gene mutations accompanied by a low frequency of chromosomal rearrangements and safer handling because of its nonclastogenic and noncarcinogenic action on humans. Azidoglycerol (AG) is a very potent mutagen in bacteria, yeast and higher plants including Arabidopsis and Tradescantia; however, it only slightly enhances the frequencies of recessive lethals in Drosophila. AG is at best a weak clastogen and is without effect in inducing chromosomal aberrations and SCEs in human peripheral lymphocytes in vitro. In microbial and plant systems, AG is considerably more potent than sodium azide in the maximal frequencies of mutation induced. In particular, in Saccharomyces cerevisae, AG is 3000-fold more mutagenic than sodium azide. Its carcinogenic and teratogenic properties are unknown.

Embryotoxicity induced by alkylating agents: 6. DNA adduct formation induced by methylnitrosourea in mouse embryos

Formation of DNA adducts in 11-day-old mouse embryos was studied by measuring the initial alkylation rates of the methylated purine bases 7-methylguanine, O6-methylguanine, and 3-methyladenine. In the first part of the studies the adduct rates were measured in the teratogenic dose range (ED10-ED90, 2.7-5.6 mg/kg). These results were compared with similar data obtained from studies with ethylmethanesulfonate and acetoxymethyl-methylnitrosamine. For the three investigated substances a correlation was found between the initial adduct rate of O6-alkylguanine in the DNA of the embryos and the teratogenic potency. In the second part of the study the rate of adduct formation was measured in the sub-teratogenic dose range. These data will be used for molecular dosimetry in a risk assessment of low doses.

Mechanism of inhibition of N-methyl-N'-nitro-N-nitrosoguanidine-induced mutagenesis by phenolic compounds

At non-toxic concentrations, 2 naturally occurring phenolic compounds, caffeic acid and chlorogenic acid, suppressed the mutagenic activity of the carcinogen N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) in Salmonella typhimurium strain TA1535. The inhibitory effect was observed only when the phenolic compound and the mutagen were administered concurrently. The interaction between phenolic compounds and MNNG was also studied in a cell-free system using a colorimetric method. The results are consistent with the assumption that phenolic compounds scavenge reactive electrophilic MNNG degradation products, thereby preventing their action on critical cellular targets.

Induction of forward mutation at the yg2 locus in maize by ethylnitrosourea

We analyzed the kinetics of mutation induction at the yg2 locus in Zea mays by N-ethyl-N-nitrosourea (ENU). ENU concentrations ranged from 0.1 to 10.7 mM and a linear, concentration-dependent response was observed for the induction of yg2 sectors in leaf 4 and leaf 5. For both leaf 4 and leaf 5 ENU exhibited one-hit kinetics consistent with either a point-mutational or chromosome-breakage mechanism of mutation induction. The mean induced mutation rate/mM ENU for the combined leaf 4 and leaf 5 data was 6.77 X 10(-4) mutations/locus/mM ENU. This value was used to compare the mutagenic effectiveness of ENU with ethyl methanesulfonate (EMS) and to calculate a rad equivalent for ENU. ENU was found to be 3.8-fold more effective than EMS. Based on their respective chemical reactivities with DNA, we propose that ENU predominantly induces point mutations at the yg2 locus. One rad of gamma radiation induced the same rate of mutation as an 8-hr exposure to a 6.71 X 10(-6) M concentration of ENU at 20 degrees C.

Carcinogenic N-nitroso compounds and their environmental significance

A short review is given on the potent and organ-specific carcinogenic effects of N-nitroso compounds. Like many other chemical carcinogens, nitrosamino compounds require activation in vivo via enzymatic alpha-hydroxylation to form alkylating agents as ultimate carcinogens. Since no epidemiological data in man are available, extrapolation of animal data to man are important as well as dose-response studies in risk evaluations. Important aspects of these problems are presented. Finally the environmental impact of N-nitroso compounds is summarized.

Determination of N-nitrosamines and N-nitrosamine precursors in rubber nipples from baby pacifiers by gas chromatography-thermal energy analysis

N-Nitrosamines and precursors are present in rubber products in which the accelerators and stabilizers used in the vulcanization process were derived from dialkylamines. Research was performed to develop data concerning the presence of N-nitrosamines and precursors so that the health significance of the exposure problem related to infant ingestion of these chemicals could be properly assessed. Volatile N-nitrosamines were determined in cut-up pacifier nipples by extraction with dichloromethane followed by concentration in a Kuderna-Danish evaporator, high-temperature mineral oil purge and trap, and analysis by gas chromatography--thermal energy analysis (GC-TEA). N-nitrosodibutylamine (NDBA) was the principal N-nitrosamine found, with concentrations up to 427 ppb. N-Nitrosamines and precursors in cut-up and intact nipples were determined by GC-TEA after a single extraction with artificial saliva. NDBA was the principal nitrosamine found, at levels up to 1040 ppb, while dibutylamine (DBA) was the principal precursor found, at levels up to 3890 ppb. The persistence of these compounds in intact nipples was determined by multiple artificial saliva extractions. Amounts of NDBA and DBA found after 15 artificial saliva extractions of intact pacifier nipples totalled 824 ppb and 15.6 ppm, respectively. N-Nitrosamine levels generally showed a gradual decrease in concentration with each extraction, whereas no consistent trend could be determined for concentrations of precursors.

Detoxification of nitrosamides and nitrosocarbamates in blood plasma and tissue homogenates

Nitrosamides and nitrosocarbamates exhibit relatively high mutagenic activity in Salmonella when compared with nitrosoureas. This high activity can be accounted for by activation of nitrosamides and nitrosocarbamates by cellular thiols, predominantly reduced glutathione, that are present intracellularly at concentrations in the millimolar range. In striking contrast to the in vitro mutagenicity tests, a number of studies have indicated that nitrosamides and nitrosocarbamates are less potent than nitrosoureas when tested in vivo in model systems such as the mouse. We extend here previous studies [Aukerman et al, 1983] that demonstrate striking chemical decomposition and inactivation of mutagenic activity of nitrosamides and nitrosocarbamates during exposure to murine blood plasma. Plasma glutathione concentrations are inadequate to account for the rapid inactivations noted. Furthermore, the predominant inactivating species is heat-sensitive, nondialyzable, and is greater than 25,000 daltons in size as judged by ultrafiltration experiments. Serum albumin has some inactivating capacity at the concentration found in undiluted plasma and could account for the very low but significant inactivating capacity of human plasma. On the other hand, serum albumin lacks the potency necessary to account for the extremely high levels of inactivating activity observed in rodent and rabbit plasma. Elsewhere we present evidence that carboxylesterase activity is the predominant inactivating species in mouse plasma [Aukerman et al, 1983; Aukerman, 1983; Brundrett and Aukerman, 1984]. Mouse liver, large intestine, kidney, and stomach have more activity per milligram protein under the assay conditions used than plasma itself. Rat liver S9 is also active at enhancing the decomposition of nitrosamides and nitrosocarbamates; most of this inactivating capacity resides in the microsomal fraction. The relatively rapid detoxification of these N-nitroso compounds by plasma and other tissues of rodents has important implications regarding the utility of rodents in assessment of tumorigenicity and/or antitumor activity of these classes of drugs in other animal species. Tests with Salmonella may be of use in estimating relative levels of protection that vary widely among mammalian species.

Comparison of mutagenicity and chemical properties of N-methyl-N'alkyl-N-nitrosoureas

The mutagenic activities of 11 N-methyl-N'-alkyl-N-nitrosoureas were tested on Salmonella typhimurium TA1535 and compared with chemical properties (alkylating activity and decomposition rate). In their relative mutagenicities the N-nitrosoureas that had a cyclic N'-alkyl groups showed far more mutagenic activity than those having a chain N;-alkyl group. M(1-A) NU and M(2-A)NU, which had the most bulky N'-alkyl group in this series, exhibited lethal effects at high concentrations. The mutagenicity showed a small positive correlation with decomposition rates but not with alkylating activities on 4-(p-nitrobenzyl)pyridine. The highest mutagenicity in this series was observed in N-methyl-N'cyclobutyl-N-nitrosourea. These results suggest that, in this series of N-methyl-N'-alkyl-N-nitrosoureas, structural differences in the N'-alkyl groups had great significance in mutagenicity.

A possible mechanism for an interaction between postreplication recombination repair and base misincorporation in mutation induction by N-methyl-N'-nitro-N-nitrosoguanidine in Haemophilus influenzae

Evidence in previous publications has suggested that treatment with monofunctional alkylating agents such as N-methyl-N'-nitro-nitrosoguanidine (MNNG) results in gaps being left in the DNA synthesized shortly after the treatment. This paper presents further evidence that suggests, though it does not conclusively prove, that there are indeed gaps. It shows that these events increase linearly with MNNG concentration, that they are formed mainly in DNA synthesized during the first hour after treatment, and that only a few are formed at later times. An hypothesis that involves the conversion by recombination repair of a single-strand base substitution, resulting from insertion of an incorrect base opposite an alkylated base, to a double-strand base substitution is proposed. It is suggested that most single-strand substitutions are removed by mismatch repair, leaving the double-strand substitutions as the main source of mutations. This hypothesis predicts that the mutation frequency will increase as the square of the exposure to MNNG, and this seems to be the case, at least at lower exposures at which complicating factors such as lengthened expression time are avoided. It also can explain a number of earlier observations on mutation fixation as detected by transformation. An attempt to show that the non-coding lesions causing the gaps were apurinic sites was unsuccessful.

An Escherichia coli mutant refractory to nitrosoguanidine mutagenesis

A newly-isolated Escherichia coli mutant suffers only about 10% as many mutations as normal strains on exposure to nitrosoguanidine. The responsible mutation, inm-1, maps at approximately minute 79 in the current E. coli genetic map. The mutant is normal for overall growth, nitrosoguanidine lethality, spontaneous mutagenesis, ultraviolet light lethality and mutagenesis, ethyl methanesulfonate lethality and mutagenesis, and the adaptive repair induced by alkylating agents. The existence of this mutation proves that nitrosoguanidine mutagenesis is not merely the result of reactions between the chemical and DNA, but requires specific cellular function(s), and underscores the peculiarity of nitrosoguanidine as a mutagen.

Genetic effects of dimethyl sulfate, diethyl sulfate, and related compounds

DMS and DES are monofunctional alkylating agents that have been shown to induce mutations, chromosomal aberrations, and other genetic alterations in a diversity of organisms. They have also been shown to be carcinogenic in animals. As an alkylating agent, DMS is a typical SN2 agent, attacking predominantly nitrogen sites in nucleic acids. DES is capable of SN1 alkylations as well as SN2 and thereby causes some alkylation on oxygen sites including the O6-position of guanine which is thought to be significant in mutagenesis by direct mispairing. The mutagenicity of DMS is better explained in terms of indirect, repair-dependent processes. With respect to both alkylating activity and genetic effects, striking similarities are found between DMS and MMS and between DES and EMS. In most systems where they have been tested, both DMS and DES are mutagenic. Results of many of the mutagenesis studies involving these compounds and other alkylating sulfuric acid esters are summarized in Tables 6, 7, 8, 9 and 10 of this review. Most data are consistent with these agents acting primarily as base-pair substitution mutagens. In the case of DES, strong specificity for G.C to A.T transitions has been reported in some systems but has not been clearly supported in some others. Low levels of frameshift mutations of the deletion type are also likely. In addition to the induction of mutations, recombinogenic and clastogenic effects have been described.

Mitochondrial mutagenesis in Saccharomyces cerevisiae. III. Nitrous acid

Nitrous acid (NA) induced mutations efficiently in mitDNA, conferring resistance to erythromycin and weakly induces mit- mutations. In some strains of yeast it also enhanced rho- mutations. The frequencies of nuclear and mitochondrial mutations induced with NA are compared.

Relationship between DNA alkylation and specific-locus mutation induction by N-methyl- and N-ethyl-N-nitrosourea in cultured Chinese hamster ovary cells (CHO/HGPRT system)

Chinese hamster ovary (CHO) cells in culture were utilized to determine the cytotoxicity, specific-locus mutation induction, and DNA alkylation which result from treatment of the cells with a range of concentrations of N-methyl-N-nitrosourea (MNU) or N-ethyl-N-nitrosourea (ENU). With [3H]MNU over the concentration range 0.43--13.7 mM, methylation of DNA was found to increase linearly, with a mean value of 56.7 pmol residue per mumol nucleoside per mM. With [1-3H]ENU over the concentration range 1.7--26.8 mM, ethylation was linear, with a mean value of 3.8 pmol residue per mumol nucleotide per mM. Mutation induction at the hypoxanthine-guanine phosphoribosyl transferase (HGPRT) locus was quantified by determination of the frequency of resistance to 6-thioguanine under stringently-defined selection conditions. The mutation frequency increased linearly with MNU or ENU concentration (0.01--2.0 mM); mean values were 2800 and 840 mutants per 10(6) clonable cells per mM, respectively. At equal levels of DNA alkylation, ENU was found to be approx. 4.5 times as mutagenic as MNU.

Mutagenicity of methyl-, ethyl-, propyl- and butylnitrosourea towards Escherichia coli WP2 strains with varying DNA repair capabilities

Methyl- (MNUA), ethyl- (ENUA), propyl- (PNUA) and butylnitrosourea (BNUA) have been tested for toxicity and mutation in a liquid suspension assay towards Escherichia coli WP2 and some of its repair deficient derivatives. A comparison of survival rates after nitrosourea exposure between WP2 and WP2 uvrA showed no difference between the two strains but a consistent difference in potency between the various nitrosoureas studied. Toxicity increased in the order MNUA less than PNUA less than ENUA less than BNUA. ENUA and PNUA induced a greater number of trp+ revertants in both strains than did MNUA and BNUA, particularly at low survival rates. None of these differences in biological potency could be accounted for by differences in rates of hydrolysis. ENUA, PNUA and BNUA were non-mutagenic towards WP2 lexA, WP2 recA and WP2 uvrA lexA, whereas MNUA did induce mutations. Ethyl methanesulphonate (EMS) was able to mutate WP2 lexA. These results are discussed in the light of current theories regarding the mechanism of action of these compounds.

Nitrous acid mutagenesis of duplex DNA as a three-component system

Purified native Hemophilus influenzae DNA is relatively insusceptible to nitrous acid (NA) mutagenesis in vitro, but is readily mutated following denaturation. NA mutagenicity for duplex DNA is significantly increased in the presence of various alcohols, glycols, phenols or primary amines. Phenol-extracted DNA contains dissociable contaminants of low molecular weight that enhance NA mutagenesis. Enhancement of NA mutagenesis by phenol and by spermine is due to the formation of unstable molecular species. We propose that reactive organic nitroso compounds are formed which then serve as delivery vehicles to promote mutagenicity of native DNA, perhaps via transnitrosation reactions. Similar reactions probably occur in vivo to promote NA-induced base substitution (but not frameshift) mutations in Salmonella typhimurium and in Escherichia coli. The possible significance of these observations to carcinogenesis is discussed.

Effects of N-nitrosopiperidine substitutions on mutagenicity in Drosophila melanogaster

N-Nitrosopiperidine (NP) and various derivatives were fed to Drosophila melanogaster males over a wide concentration range in order to assess their mutagenic potency in the induction of X-linked recessive lethals and chromosome loss. NP was effective in inducing lethals, as were its halogen and methyl-substituted derivatives, with the exception of 2,6-dimethyl NP. (Methyl substitutions at the alpha carbon atoms reduce or eliminate mutagenic activity.) Substitution of halogen groups on the piperidine ring enhanced the mutagenic activity, with the 3-chloro compound being the most mutagenic. In contrast, substitutions with a hydroxyl, carboxyl, or keto group resulted in a loss of mutagenicity. None of the compounds tested increased the frequency of chromosome loss or breakage in mature sperm.

Mutagenicity and cytotoxicity of congeners of two classes of nitroso compounds in Chinese hamster ovary cells

The induction of mutation by certain nitrosamidines and nitrosamides has been quantitated utilizing the hypoxanthine--guanine phosphoribosyl transferase (HGPRT) locus in Chinese hamster ovary cells. Dose--response relationships for cytotoxicity and mutagenicity are presented for N-methyl-N-nitrosourea (MNU), N-ethyl-N-nitrosourea (ENU), N-butyl-N-nitrosourea (BNU), N-methyl-N'-nitro-N-nitrosoguanidine (MNNG), and N-ethyl-N'-nitro-N-nitrosoguanidine (ENNG). Based on the concentration of each agent required to kill 90% of the cells, the following order of cytotoxicity was observed: MNNG greater than ENNG greater than MNU greater than ENU greater than BNU. This is the same order of potency as observed for mutation induction per unit concentration of mutagen.

The relation of repair phenomena to mutation induction in bacteria

The relation of various processes to mutation induction by radiation and chemicals is discussed for for various species of bacteria. A variety of repair processes have been identified at the molecular level that can eliminate many kinds of potentially mutagenic lesions before they can be converted to final mutation. Fixation often but not always occurs at replication. A number of mutagens, including UV light, ionizing radiation, and a number of chemicals, induce an error-prone process, perhaps a modification of the proof-reading system, that allows bacteria to survive after potentially lethal damage at the expense of making errors. Some mutagens, notably monofunctional alkylating agents and base analogues, produce mutations by other processes. Even in these cases, repair processes play an important role. There is some evidence that error-free as well as error-prone repair processes can be induced. A brief discussion is given of the relation of these findings to the practical problems of hazards estimations.

Antimutagenic effects of caffeine during nitrosoguanidine-induced mutagenesis of Salmonella typhimurium cells and phages

The effect of caffeine on nitrosoguanidine-induced mutagenesis of Salmonella typhimurium and its P22 and L phages was studied. The detected mutations included phage "clear" mutations, reversions of phage "amber" mutation, and prototrophic reversions of the his- auxotroph of Salmonella typhimurium. Neither the recA mutation of the host nor the erf mutation of the phage genome were found to affect the nitrosoguanidine-induced mutagenesis of the phage during vegetative growth. Beginning with a concentration of 0.2 mg/ml, caffeine decreased the frequency of mutants by 30--60%, attaining a maximum effect at 1.5 mg/ml and retaining this effect even at higher concentrations. A similar antimutagenic effect was observed with the mutagenesis of the host cells. The nitrosoguanidine-induced mutagenesis does not seem to be related to the function of the recA cell gene or the erf phage gene. The mechanism of mutagenesis by nitrosoguanidine probably has two components, one of them caffeine sensitive, the other caffeine-resistant.

Induction of dominant-lethal mutations after administration of ethylenethiourea in combination with nitrite of the n-nitroso-ethylenethiourea in mice

Mutagenic potentials of ethylenethiourea (ETU) in combination with sodium nitrite or of N-nitroso-ETU, a nitrosation product of ETU in vitro, were investigated in the mouse dominant-lethal test. Simultaneous 5-day p.o. administration of ETU (150 mg/kg) and sodium nitrite (50 mg/kg) caused a significant decrease in the percentage pregnancy and the number of implants in weeks 5 and 6 of testing, although no effects were obtained on these indices when the chemicals were applied separately. However, in the group treated with 30 mg ETU plus 10 mg sodium nitrite per kg no dominant-lethal mutations were induced. 5-day oral administration of 100 mg of N-nitroso-ETU per kg also exhibited similar effects to those observed after treatment with 150 mg ETU plus 50 mg sodium nitrite per kg.

Physiological modification of alkylating agent induced mutagenesis. II. Influence of the numbers of chromosome replicating forks and gene copies on the frequency of mutations induced in Escherichia coli

The frequency of reversions induced in Escherichia coli K-12 trpA58 by any of five different monofunctional alkylating agents increased as the growth rate of the organism was raised prior to mutagen treatment. The increase in mutation frequency did not correlate with growth rate-dependent changes in cell area or total cellular protein and DNA. After treatment of cells with N-methyl-N-nitrosourea (MNUA), no growth rate-dependent change was observed in the total DNA alkylation or percentage of O6-methylguanine present in the DNA extracted. The frequency of reversions induced by one mutagen, methyl methanesulphonate (MMS), increased in proportion to the average number of trpA gene copies per cell, whereas the frequency of reversions induced by the other compounds was dependent on the average number of chromosome replicating forks per cell. This difference was attributed to the different ratios of DNA base alkylation products observed, formed after treatment with MMS, an SN2-type reagent, or after treatment with the SN1-type reagents ethyl methanesulphonate (EMS), N-methyl-N'-nitro-N-nitrosoguanidine (MNNG), MNUA and N-ethyl-N-nitrosourea (ENUA). Possible reasons for the dependence of mutation frequency on the number of replicating forks per cell are discussed.

Physiological modification of alkylating-agent induced mutagenesis. I. Effect of growth rate and repair capacity on nitrosomethylurea-induced mutation of Escharichia coli

The effects of repair capacity and growth rate on the induction of mutations by N-methyl-N-nitosourea (MNUA) was investigated using the trpE reversion system of Escherichia coli WP2 and some repair-deficient derivatives isogenic for this gene. In all these strains reducing the growth rate prior to MNUA-treatment caused a reduction in the mutational response, however major differences were observed between strains. In exrA and recA- bacteria stationary phase cells were 100 times less mutable than cells grown at a mean generation time (m.g.t.) of 30 min, whereas reductions of 12 and 25 times were observed in the uvrA- and wild-type strains respectively. In contrast the mutational response of the polA- mutant varied only slightly with growth rate; the increases at high MNUA concentrations being equal to the increase in the trpE gene number. These results show the increasing importance of the error-prone exrA+/recA+-dependent repair system in mutation-induction by MNUA as the growth rate of the culture is reduced and its relative unimportance for mutation induction in nutrient broth-grown cells (m.g.t. 30 min).