Integrated in silico and in vitro genotoxicity assessment of thirteen data-poor substances.

The Canadian Domestic Substances List (DSL) contains chemicals that have not been tested for genotoxicity as their use pre-dates regulatory requirements. In the present study, (quantitative) structure-activity relationships ((Q)SAR) model predictions and in vitro tests were conducted for genotoxicity assessment of 13 data-poor chemicals from the DSL (i.e. CAS numbers 19286-75-0, 13676-91-0, 2478-20-8, 6408-20-8, 74499-36-8, 26694-69-9, 29036-02-0, 120-24-1, 84696-48-9, 4051-63-2, 5718-26-3, 632-51-9, and 600-14-6). First, chemicals were screened by (Q)SAR models in Leadscope® and OASIS TIMES; two chemicals were excluded from (Q)SAR as they are complex mixtures. Six were flagged by (Q)SAR as potentially mutagenic and were subsequently confirmed as mutagens using the Ames assay. Of nine chemicals with clastogenic (Q)SAR flags, eight induced micronuclei in TK6 cells. Benchmark dose analysis was used to evaluate the potency of the chemicals. Four chemicals were bacterial mutagens with similar potencies. Three chemicals were more potent in micronuclei induction than the prototype alkylating agent methyl methanesulfonate and three were equipotent to the mutagenic carcinogen benzo[a]pyrene in the presence of rat liver S9. Overall, 11 of the 13 DSL chemicals demonstrated at least one type of genotoxicity in vitro. This study demonstrates the application of genotoxic potency analysis for prioritizing further investigations.

Computational identification of structural factors affecting the mutagenic potential of aromatic amines: study design and experimental validation.

A grid-based, alignment-independent 3D-SDAR (three-dimensional spectral data-activity relationship) approach based on simulated 13C and 15N NMR chemical shifts augmented with through-space interatomic distances was used to model the mutagenicity of 554 primary and 419 secondary aromatic amines. A robust modeling strategy supported by extensive validation including randomized training/hold-out test set pairs, validation sets, "blind" external test sets as well as experimental validation was applied to avoid over-parameterization and build Organization for Economic Cooperation and Development (OECD 2004) compliant models. Based on an experimental validation set of 23 chemicals tested in a two-strain Salmonella typhimurium Ames assay, 3D-SDAR was able to achieve performance comparable to 5-strain (Ames) predictions by Lhasa Limited's Derek and Sarah Nexus for the same set. Furthermore, mapping of the most frequently occurring bins on the primary and secondary aromatic amine structures allowed the identification of molecular features that were associated either positively or negatively with mutagenicity. Prominent structural features found to enhance the mutagenic potential included: nitrobenzene moieties, conjugated π-systems, nitrothiophene groups, and aromatic hydroxylamine moieties. 3D-SDAR was also able to capture "true" negative contributions that are particularly difficult to detect through alternative methods. These include sulphonamide, acetamide, and other functional groups, which not only lack contributions to the overall mutagenic potential, but are known to actively lower it, if present in the chemical structures of what otherwise would be potential mutagens.

Safety evaluation of mulberry leaf extract: Acute, subacute toxicity and genotoxicity studies.

Mulberry (Morus alba L.) leaves are of broad popular use for food or remedy purposes due to their bioactive properties, especially antidiabetic activity and antioxidative activity. The present study aimed to assess the toxicological profile of mulberry leaf extract (MLE), through acute, subacute toxicity and genotoxicity tests. Male and female rats received by gavage 15.0 g/kg bw of MLE in the acute toxicity test, and 0, 1.88, 3.75 and 7.50 g/kg bw/d of MLE for subacute toxicity test. In the acute toxicity study, no mortality or behavioral changes were observed, indicating the LD50 is higher than 15.0 g/kg bw. In the subacute toxicity test, no significant changes were observed in hematological, biochemical or histopathological parameters in the animals exposed. The no-observed-adverse-effect level in the subacute toxicity study was considered to be 7.50 g/kg bw/d, the highest dose tested. In the genotoxicity study, MLE showed no mutagenic activity in the Ames assay and no evidence of potential to induce chromosome aberrations or sperm abnormalities in mice exposed to 10 g/kg bw. Collectively, aqueous extract of mulberry leaves could be considered safe, and the results support the application of MLE as novel food ingredient or product.

Antioxidant and Antimutagenic Activities of Different Fractions from the Leaves of Rhododendron arboreum Sm. and Their GC-MS Profiling.

In this era of urbanization and environmental pollution, antioxidants and antimutagens derived from plants are promising safeguards for human health. In the current investigation, we analyzed the antioxidant and antimutagenic effects of the hexane, chloroform, and ethyl acetate fractions of Rhododendron arboreum Sm. leaves and determined their chemical composition. The different fractions inhibited lipid peroxidation, repressed the production of nitric oxide radicals, and prevented deoxyribose degradation. The antimutagenic activity of the leaf fractions was analyzed against 4-nitro-O-phenylenediamine, sodium azide and 2-aminofluorene mutagens in two test strains (TA-98 and TA-100) of Salmonella typhimurium. The experiment was conducted using pre- and co-incubation modes. The best results were obtained in the pre-incubation mode, and against indirect acting mutagen. The presence of a number of bioactive constituents was confirmed in the different fractions by GC-MS analysis. The study reveals the strong antioxidant and antimutagenic activity of R. arboreum leaves. We propose that those activities of R. arboreum might correspond to the combined effect of the phytochemicals identified by GC-MS analysis. To the best of our knowledge, this is the first report on the antimutagenic activity of R. arboreum leaves.

The various aspects of genetic and epigenetic toxicology: testing methods and clinical applications.

Genotoxicity refers to the ability of harmful substances to damage genetic information in cells. Being exposed to chemical and biological agents can result in genomic instabilities and/or epigenetic alterations, which translate into a variety of diseases, cancer included. This concise review discusses, from both a genetic and epigenetic point of view, the current detection methods of different agents' genotoxicity, along with their basic and clinical relation to human cancer, chemotherapy, germ cells and stem cells.

Resolution of contradiction between in silico predictions and Ames test results for four pharmaceutically relevant impurities.

The ICH M7 Guideline requires low level control of mutagenic impurities in pharmaceutical products to minimize cancer risk in patients (ICHM7, 2014). Bacterial mutagenicity (Ames) data is generally used to determine mutagenic and possible carcinogenic potential of compounds. Recently, a publication on experiences of using two in silico systems to identify potentially mutagenic impurities highlighted the importance of performing a critical review of published Ames data utilized as part of a mutagenicity assessment of impurities (Greene et al., 2015). Four compounds (2-amino-5-hydroxybenzoic acid, 2-amino-3-chlorobenzoic acid, methyl 2-amino-4-chlorobenzoate and 4-morpholinopyridine) reported mutagenic were identified in a two system in silico assessment and expert review of the structuresas non-mutagenic. Likely reasons for mutagenicity could not be identified and the purity of the compounds tested was proposed. In the current investigation, the purest available sample of the four compounds was tested in an OECD-compliant Ames test. The compounds were all found to be non-mutagenic. Possible reasons for the discrepancy between previously reported and current results are discussed. Additionally, important points to consider when conducting an expert review of available Ames data are provided particularly in cases where reported Ames results are discrepant with a two system in silico assessment.

Liposomal encapsulation masks genotoxicity of a chemotherapeutic agent in regulatory toxicology assessments.

The burgeoning application of nanotechnology to a variety of industries including cosmetics, food, medicine and materials has led to the exploration of nanotoxicology as a trending subject of research. However the role of a nanovector, in affecting the mutagenicity of its therapeutic payload has not yet been investigated. In this study, we compare the mutagenicity of the free drug - doxorubicin hydrochloride with its nanoencapsulated form - doxorubicin loaded liposome, using conventional methods required for regulatory approval. Contrary to free doxorubicin, doxorubicin encapsulated liposome expressed a significantly lower mutant frequency in the Ames assay, and was non-genotoxic in the in vitro micronucleus assay. Further investigation of the systems' cytotoxicity and their interaction with the bacterial cell envelope, suggests that the modification of the test parameters and release of the encapsulated drug prior to the Ames test show comparable mutagenic potential of the nanotherapeutic system to a free drug.

Use of low density polyethylene membranes for assessment of genotoxicity of PAHs in the Seine River.

The genotoxicity of river water dissolved contaminants is usually estimated after grab sampling of river water. Water contamination can now be obtained with passive samplers that allow a time-integrated sampling of contaminants. Since it was verified that low density polyethylene membranes (LDPE) accumulate labile hydrophobic compounds, their use was proposed as a passive sampler. This study was designed to test the applicability of passive sampling for combined chemical and genotoxicity measurements. The LDPE extracts were tested with the umu test (TA1535/pSK1002 ± S9) and the Ames assay (TA98, TA100 and YG1041 ± S9). We describe here this new protocol and its application in two field studies on four sites of the Seine River. Field LDPE extracts were negative with the YG1041 and TA100 and weakly positive with the TA98 + S9 and Umu test. Concentrations of labile mutagenic PAHs were higher upstream of Paris than downstream of Paris. Improvement of the method is needed to determine the genotoxicity of low concentrations of labile dissolved organic contaminants.

Dermal Exposure to Cumene Hydroperoxide: Assessing Its Toxic Relevance and Oxidant Potential.

Cumene hydroperoxide (CHP) is a high production volume chemical that is used to generate phenol and acetone. Dermal exposure to CHP was hypothesized to result in systemic tissue toxicity, production of free radicals, and consequent decrease in plasma antioxidant levels. To evaluate the hypothesis and characterize the toxicity of CHP, male and female B6C3F1/N mice and F344/N rats were exposed to varying doses of CHP applied topically for 14 or 90 days. No significant changes in survival or body weight of mice and rats were observed following 14 days of exposure. However, 90 days of CHP exposure at the high dose (12 mg/kg) triggered a significant decrease (-15%) in the body weight of the male rat group only. Irritation of the skin was observed at the site of application and was characterized by inflammation and epidermal hyperplasia. In treated animals, histology of liver tissue, free radical generation, and antioxidant levels in blood plasma were not significantly changed as compared to the corresponding controls. Consistent with the lack of systemic damage, no increase in micronucleated erythrocytes was seen in peripheral blood. In conclusion, topical CHP application caused skin damage only at the application site and did not cause systemic tissue impairment.

In vitro and in vivo toxicological evaluations of methyl ferulate, methyl p-coumarate, and pulegone 1,2-epoxide.

CONTEXT: Toxicological screening of natural compounds for medicinal purposes. OBJECTIVES: The objective of this study is to evaluate the toxicity of methyl ferulate (MF), methyl p-coumarate (MpC), and pulegone 1,2-epoxide (PE) with in vitro and in vivo assays. MATERIALS AND METHODS: The in vitro toxicity of MF, MpC, and PE was assessed at a concentration of 10 mg/ml with the Ames assay using two strains of Salmonella typhimurium TA98 and TA100. Human red blood cells (RBC) were used to determine the hemolytic activity of these compounds. The cytotoxicity of above compounds was determined with brine shrimp lethality bioassay (BSLB) at the concentrations of 0.1-20 mg/ml. While dermal and ocular irritation studies were conducted on healthy rabbits (n = 8) for 96 and 12 h post-topical application of test compounds, respectively. RESULTS: PE produced 6-8% hemolysis of RBCs at all the tested concentrations while MF and MpC produced 10-5% hemolysis up to 20 mg/ml, and 50-85% hemolysis at concentrations of 40 and 80 mg/ml, respectively. The Ames assay indicated that MF, MpC, and PE were non-mutagenic as the test values were not significantly higher as compared with background values of the assay. BSLB suggested the lethal concentration (LC50) values of MF, MpC, and PE as 4.38, 6.74, and 25.91 mg/ml, respectively. In vivo ocular and dermal irritation scores of MF, MpC, and PE were comparable with ethanol (control) in rabbits indicating the non-irritant nature of these natural compounds. CONCLUSION: The present studies suggest that these compounds are non-toxic/non-irritant and might be used for medicinal purposes.

Evaluation of in vitro and in vivo safety profile of the Indian traditional medicinal plant Grewia tiliaefolia.

Grewia tiliaefolia Vahl., is a subtropical tree used in Ayurvedic formulations and also by tribal communities in India for treatment of wounds, urinary infection and skin diseases. Despite of its medicinal use, the plant has not been previously tested for its safety. This study evaluated the toxicological profile of the methanolic extract of G. tiliaefolia leaves (MEGT) through in vitro (cytotoxic, mutagenic, genotoxic) and in vivo (acute and sub-acute) studies. In vitro assays conclude that there is no significant (p<0.05) level of cytotoxicity and genotoxicity at the tested concentrations (250-2000 µg/ml) in peripheral blood mononuclear cells. Ames assay in the tested Salmonella typhimurium strains revealed no significant (p<0.05) increase in the mutagenic index both in the presence and absence of metabolic activation. Alternatively, no change in mortality, physiological, behavioural, haematological and biochemical parameters were observed under in vivo acute (250, 500, 1000, 2000 mg/kg b.w) and sub-acute (100, 200, 1000 mg/kg b.w) toxicity studies in Wistar rats. Histopathological examinations of the animals did not reveal any treatment related changes and showed normal architecture, even at the highest concentration (1000 mg/kg b.w) following 28 days treatment. Overall, the results show that MEGT is non-toxic at the tested concentrations and is safe to be used as herbal medicine.

Mutagenicity and clastogenicity evaluation of metaphenoxy benzyl chloride by ames and micronucleus assays.

Metaphenoxy Benzyl Chloride (3-PBC; CAS # 53874-66-1) is a commonly used intermediate in organic synthesis and is a widely used chemical building block. There is likelihood of human contact with 3-PBC in the process of organic or pesticide synthesis and even through as impurity in the product. Further, this chemical has been reported to be found in water reservoirs causing a threat to aqueous flaura and fauna. Genotoxic alert has been suspected for the chemical 3-PBC based on its structure hence, the impetus of present work was to assess the mutagenic potential of 3-PBC using Ames bacterial reverse-mutation assay and in vitro micronucleus assay. 3-PBC was tested for mutagenic potential at six different concentrations, with 0.05 (without metabolic activation) and 0.50 (with metabolic activation) µL/plate as the highest concentrations, followed by five lower concentrations with 2-fold spacing. In clastogenic evaluation, 3-PBC was tested at concentrations ranged from 0.31 to 4.88 µL/mL to assess micronucleus induction in mammalian cells viz. Chinese Hamster Ovarian-K1 cells (CHO-K1 cells). In the Ames assay, 3-PBC did not show mutagenicity in all five tester strains of Salmonella typhimurium viz. TA98, TA100, TA102, TA1535 and TA1537 both in the presence and absence of a metabolic activation system and found to be nonclastogenic when evaluated in "CHO-K1 cells".

Antimutagenic potential and modulation of carcinogen-metabolizing enzymes by ginger essential oil.

Essential oil extracted from ginger (GEO) was evaluated for its mutagenicity to Salmonella typhimurium TA 98, TA 100, TA 102, and TA 1535 strains with and without microsomal activation. GEO was found to be non-mutagenic up to a concentration of 3 mg/plate. It was also assessed for antimutagenic potential against direct acting mutagens such as sodium azide, 4-nitro-o-phenylenediamine, N-methyl-N'-nitro-N-nitrosoguanidine, tobacco extract, and 2-acetamidoflourene, which needs microsomal activation. GEO significantly inhibited (p < 0.001) the mutagenicity induced by these agents in a concentration-dependent manner. The effect of GEO to modulate the action of phase I carcinogen-metabolizing enzymes was investigated by studying its effect on various isoforms of microsomal cytochrome P450 enzymes. Significant inhibition of CYP1A1, CYP1A2, and CYP2B1/2, aniline hydroxylase (an indicator of CYP2E1 activity), and aminopyrine-N-demethylase (indicator of CYP1A, 2A, 2B, 2D, and 3A activity) was shown by GEO both in vitro and in vivo. GEO gave an IC50 value of 30, 57.5, and 40 µg for CYP1A1, CYP1A2, and CYP2B1/2, respectively, 55 µg for aniline hydroxylase, and 37.5 µg for aminopyrene-N-demethylase. GEO also significantly increased the levels of phase II carcinogen-metabolizing enzymes uridine 5'-diphospho-glucuronyl transferase and glutathione-S-transferase in vivo indicating the use of GEO as an antimutagen and as a potential chemopreventive agent.

Genotoxicity and cytotoxicity evaluation of a heat-not-burn product.

'Heat-not-burn' products (HnBP) contain lower levels of harmful substances than traditional cigarettes, but the use of these products warrants further toxicological evaluation. We have compared the cytotoxicity and genotoxicity of a heat-not burn product with conventional cigarettes, in vivo and in vitro. Male Sprague Dawley rats were exposed to mainstream smoke from conventional cigarettes or a HnBP, for 4 or 28 days, followed by isolation of bone marrow polychromatic erythrocytes (PCE) and histological examination of the testes. Chinese hamster lung fibroblast cells were exposed in vitro to total particulate matter from cigarette smoke obtained through Cambridge filters. The cytotoxicity and genotoxicity of total particulate matter were assessed by the neutral red uptake assay, chromosome aberration assay, in vitro micronucleus test, comet assay, and Ames assay. In the short-term exposure rat models, only the conventional-cigarettes group showed a significant increase in the ratio of micronuclei to total PCE. There was no significant difference in rat testis histology in the long-term exposure models. In vitro, in the neutral red uptake assay, the HnBP product showed lower cytotoxicity than conventional cigarettes. Conventional cigarettes showed greater genotoxicity in the chromosome aberration assay, high-dose Ames tests with exogenous metabolic activation, and micronucleus tests. In summary, our results suggest that HnBP have lower cytotoxicity and genotoxicity than conventional cigarettes.

Genetic toxicity of high-boiling petroleum substances.

There are several specific types of high-boiling petroleum substances (HBPS) having final boiling points >343°C), in which genetic toxicity can be related to the content of polycyclic aromatic compounds (PACs), specifically crude oils, gas oils, heavy fuel oils, lubricant base oils, waxes and aromatic extracts. Evaluation of optimized Salmonella tests covering over 250 samples from 43 types of HBPS revealed that gene mutation can be determined for these substances using a protocol optimized for the detection of mutagenic PAC. The outcomes of modified Salmonella assays can be predicted using HBPS compositional information as input to a newly developed statistical model. The general outcome of the optimized Salmonella assay can be predicted for an untested substance based on its Aromatic Ring Class (ARC) profile. Review of the results from numerous cytogenetic tests showed that although a few positive study results have been reported, most HBPS do not produce chromosomal effects when tested in rodent bone marrow assays or in in vitro chromosomal aberration assays. Results of both bacterial and cytogenetic studies can be used to satisfy genetic toxicity endpoints for the HBPS category substances.

Analysis of chemical structures and mutations detected by Salmonella TA98 and TA100.

As part of an analysis performed under the auspices of the International Workshop on Genotoxicity Testing (IWGT) in 2017, we and others showed that Salmonella frameshift strain TA98 and base-substitution strain TA100 together + /- S9 detected 93% of the mutagens detected by all the bacterial strains recommended by OECD TG471 (Williams et al., Mutation Res. 848:503081, 2019). We have extended this analysis by identifying the numbers and chemical classes of chemicals detected by these two strains either alone or in combination, including the role of S9. Using the Leadscope 2021 SAR Genetox database containing > 21,900 compounds, our dataset containing 7170 compounds tested in both TA98 and TA100. Together, TA98 and TA100 detected 94% (3733/3981) of the mutagens detected using all the TG471-recommended bacterial strains; 39% were mutagenic in one or both strains. TA100 detected 77% of all of these mutagens and TA98 70%. Considering the overlap of detection by both strains, 12% of these mutagens were detected only by TA98 and 19% only by TA100. In the absence of S9, sensitivity dropped by 31% for TA98 and 29% for TA100. Overall, 32% of the mutagens required S9 for detection by either strain; 9% were detected only without S9. Using the 2021 Leadscope Genetox Expert Alerts, TA100 detected 18 mutagenic alerting chemical classes with better sensitivity than TA98, whereas TA98 detected 10 classes better than TA100. TA100 detected more chemical classes than did TA98, especially hydrazines, azides, various di- and tri-halides, various nitrosamines, epoxides, aziridines, difurans, and half-mustards; TA98 especially detected polycyclic primary amines, various aromatic amines, polycyclic aromatic hydrocarbons, triazines, and dibenzo-furans. Model compounds with these structures induce primarily G to T mutations in TA100 and/or a hotspot GC deletion in TA98. Both TA98 and TA100 + /- S9 are needed for adequate mutagenicity screening with the Salmonella (Ames) assay.

Combination of Multidimensional Instrumental Analysis and the Ames Test for the Toxicological Evaluation of Mineral Oil Aromatic Hydrocarbons.

Mineral oil aromatic hydrocarbons (MOAHs) include mutagenic and carcinogenic substances and are considered a potential health risk. Current methods address the total MOAH content but cannot address the actual toxicological hazard of individual components. This work presents a combined methodology closing those gaps: high-performance liquid chromatography (HPLC) coupled to gas chromatography with flame ionization detection was used to determine the MOAH content. To characterize present substance classes, comprehensive two-dimensional gas chromatography with time-of-flight mass spectrometry was applied. Preparative HPLC separated MOAHs into subgroups, which were tested with a miniaturized Ames test evaluating DNA reactivity of isolated fractions. Combining these methods allowed a correlation between present subgroups and DNA reactivity. The developed approach was applied to a mineral oil and distinguished between not DNA-reactive mono- and diaromatics and DNA-reactive tri- and polyaromatics, providing a proof of concept. Hereinafter, it will be applied to diverse sample matrices including mineral oils, food, and food contact materials.

Mutagenicity Evaluation of Nanoparticles by the Ames Assay.

The Ames assay is a classic and robust method for identifying and evaluating chemical mutagens that reverse the mutations of Salmonella typhimurium and/or Escherichia coli bacteria strains with amino acid synthesis defects. It is also called the bacterial reverse mutation assay. Ames assay has been widely used for detecting genetic toxicity of many chemicals and gained increased applications in risk assessment of emerging environmental pollutants such as nanomaterials. In this chapter, we presented a detailed step-by-step method using the Ames assay to detect potential mutagenicity of metal oxide nanoparticles. The strategy to use the liver S9 fraction for bioactivation and a preincubation procedure is recommended. This method is easy to use to test genetic toxicity of other environmental contaminants and new chemicals.

Direct Comparison of the Lowest Effect Concentrations of Mutagenic Reference Substances in Two Ames Test Formats.

The Ames assay is the standard assay for identifying DNA-reactive genotoxic substances. Multiple formats are available and the correct choice of an assay protocol is essential for achieving optimal performance, including fit for purpose detection limits and required screening capacity. In the present study, a comparison of those parameters between two commonly used formats, the standard pre-incubation Ames test and the liquid-based Ames MPF™, was performed. For that purpose, twenty-one substances with various modes of action were chosen and tested for their lowest effect concentrations (LEC) with both tests. In addition, two sources of rat liver homogenate S9 fraction, Aroclor 1254-induced and phenobarbital/ß-naphthoflavone induced, were compared in the Ames MPF™. Overall, the standard pre-incubation Ames and the Ames MPF™ assay showed high concordance (>90%) for mutagenic vs. non-mutagenic compound classification. The LEC values of the Ames MPF™ format were lower for 17 of the 21 of the selected test substances. The S9 source had no impact on the test results. This leads to the conclusion that the liquid-based Ames MPF™ assay format provides screening advantages when low concentrations are relevant, such as in the testing of complex mixtures.

Are all bacterial strains required by OECD mutagenicity test guideline TG471 needed?

The International Workshop on Genotoxicity Testing (IWGT) meets every four years to seek consensus on difficult or conflicting approaches to genotoxicity testing based upon experience, available data, and analysis techniques. At the 2017 IWGT meeting in Tokyo, one working group addressed the sensitivity and selectivity of the bacterial strains specified in the Organization for Economic Cooperation and Development (OECD) Test Guideline TG471 to recommend possible modification of the test guideline. Three questions were posed: (1) Although TA100 is derived from TA1535, does TA1535 detect any mutagens that are not detected by TA100? (2) Among the options of Salmonella TA1537, TA97 or TA97a, are these strains truly equivalent? (3) Because there is a choice to use one of either E. coli WP2 uvrA, E. coli WP2 uvrA pKM101, or Salmonella TA102, are these strains truly equivalent? To answer these questions, we analyzed published bacterial mutation data in multiple strains from large (>10,000 compound) databases from Leadscope and Lhasa Limited and anonymized data for 53 compounds tested in TA1535 and TA100 provided by a pharmaceutical company. Our analysis involved (1) defining criteria for determining selective responses when using different strains; (2) identifying compounds producing selective responses based upon author calls; (3) confirming selective responses by visually examining dose-response data and considering experimental conditions; (4) using statistical methods to quantify the responses; (5) performing limited additional direct-comparison testing; and (6) determining the chemical classes producing selective responses. We found that few mutagens would fail to be detected if the test battery did not include Salmonella strains TA1535 (8/1167), TA1537 (2/247), TA102 (4/46), and E. coli WP2 uvrA (2/21). Of the mutagens detected by the full TG471 strain battery, 93% were detected using only strains TA98 and TA100; consideration of results from in vitro genotoxicity assays that detect clastogenicity increased this to 99%.