Bacterial mutagenicity screening in the pharmaceutical industry.

Genetic toxicity testing is used as an early surrogate for carcinogenicity testing. Genetic toxicity testing is also required by regulatory agencies to be conducted prior to initiation of first in human clinical trials and subsequent marketing for most small molecule pharmaceutical compounds. To reduce the chances of advancing mutagenic pharmaceutical candidates through the drug discovery and development processes, companies have focused on developing testing strategies to maximize hazard identification while minimizing resource expenditure due to late stage attrition. With a large number of testing options, consensus has not been reached on the best mutagenicity platform to use or on the best time to use a specific test to aid in the selection of drug candidates for development. Most companies use a process in which compounds are initially screened for mutagenicity early in drug development using tests that require only a few milligrams of compound and then follow those studies up with a more robust mutagenicity test prior to selecting a compound for full development. This review summarizes the current applications of bacterial mutagenicity assays utilized by pharmaceutical companies in early and late discovery programs. The initial impetus for this review was derived from a workshop on bacterial mutagenicity screening in the pharmaceutical industry presented at the 40th Annual Environmental Mutagen Society Meeting held in St. Louis, MO in October, 2009. However, included in this review are succinct summaries of use and interpretation of genetic toxicity assays, several mutagenicity assays that were not presented at the meeting, and updates to testing strategies resulting in current state-of the art description of best practices. In addition, here we discuss the advantages and liabilities of many broadly used mutagenicity screening platforms and strategies used by pharmaceutical companies. The sensitivity and specificity of these early mutagenicity screening assays using proprietary compounds and their concordance (predictivity) with the regulatory bacterial mutation test are discussed.

Comprehensive assessment of the photomutagenicity, photogenotoxicity and photo(cyto)toxicity of azulene.

The polycyclic aromatic hydrocarbon azulene and its naturally occurring derivative guaiazulene (1,4-dimethyl-7-isopropylazulene) are known to absorb light in the UV-vis region of the spectrum. Both compounds were reported to be mutagenic in the Salmonella typhimurium bacterial mutagenicity assay (Ames test) in strain TA102, and to cause DNA damage in the comet assay in vitro upon exposure to UVA light. In contrast, another study reported a photoprotective effect in vitro of guaiazulene. We present here a comprehensive assessment of the photo(cyto)toxicity (3T3 fibroblast Neutral Red uptake test), the photomutagenicity (Ames test) and photogenotoxicity (comet assay and micronucleus test in L5178Y cells in vitro) of azulene. In the Ames test, the mutagenicity of azulene was assessed in the presence and absence of UV light by use of the Salmonella strains TA102, TA104, TA2638 and E. coli WP2. Azulene was irradiated before being plated with bacteria (pre-irradiation), or concomitantly with the bacteria either after plating or while in suspension. Guaiazulene was included in some of the experiments. Neither in the photo-Ames test nor in the other photogenotoxicity tests, azulene or guaiazulene showed any photomutagenic or photogenotoxic activity. Weak photo(cyto)toxicity (estimate of PIF≥1.67) was observed with azulene in the 3T3 NRU test, the Alamar Blue test and the relative cell count, which may be due to the generation of reactive oxygen species, as reported recently.

Phototoxicity and photogenotoxicity of nine pyridone derivatives.

Nine structurally related pyridone derivatives were assayed for photogenotoxicity and phototoxicity in the Ames test, the chromosomal aberration test in V79 cells and the neutral red uptake (NRU) test in 3T3 cells. All nine compounds absorb light to a comparable degree at wavelengths between 380 and 430 nm. Seven of the nine compounds were found to produce high quantities of singlet oxygen (1O(2)) upon irradiation in the presence of oxygen. These seven compounds were highly phototoxic in the NRU test, three were clearly and two were marginally photomutagenic in the Ames test, five were assessed as clearly and two as equivocally photoclastogenic in the chromosomal aberration test. Two compounds showed substantially lower 1O(2) yields. The pyridone ring of these two compounds is attached to a non-aromatic ring, while for the seven other compounds the chromophore system including the pyridone ring consists of two or three aromatic rings. One of the two compounds with low 1O(2) yields was distinctly less phototoxic and did not induce photogenotoxic effects. The other, structurally an indolo derivative and not the common thieno derivative, was, however, similarly phototoxic as the seven compounds with high 1O(2) quantum yield and was also clearly photogenotoxic indicating that different action pathways, not involving singlet oxygen, have to be considered at least for this compound.

Photochemical mutagenesis: examples and toxicological relevance.

Induction of DNA damage as a consequence of exposure to UV light has been established as the major cause of skin cancer. DNA molecules absorb photon energy directly for wavelengths <320 nm, and lead to well-characterized mutagenic DNA damage. Alternatively, endogenous or exogenous chemicals (sensitizers) may absorb light with the potential of subsequent energy or electron transfer, and lead indirectly to DNA damage. A few light-absorbing pharmaceuticals have long been known to cause photo(geno)toxic effects. Notably, psoralen and chlorpromazine derivatives have been established as photomutagens and the reaction mechanisms have been identified; the fluoroquinolone antibiotics have more recently been recognized as being photomutagenic. The type of DNA damage and the modulation by antioxidants indicate the involvement of reactive oxygen species (ROS), but other mechanisms are also reported for, at least, some derivatives. In routine genotoxicity studies, we observed the photomutagenic activity of a compound (Ro 19-8022) under development as an anxiolytic agent in the Ames tester strain TA102 under normal laboratory illumination conditions. Further investigations showed strong photogenotoxic activity in tests for gene mutations and chromosomal aberrations in mammalian cells. The finding led to the termination of drug development. Another example of a pharmaceutical for which photogenotoxic properties were observed during development is Ro 47-7737, a bisquinoline derivative of the antimalaria compound chloroquine. Also in this case, the photochemical reactivity contributed to the termination of the development process. The risk/benefit assessment for the described compounds has to take into account the human exposure situation, for example, the ability to avoid light exposure during treatment. Consideration of photochemical mutagenesis is specifically important for sunscreen ingredients. The active components of sunscreen lotions are efficient UV absorbers. Consequently, they reduce the genotoxicity of UV light and thus may be considered antimutagens. However, photodegradation to reactive molecules or energy transfer to DNA is possible, in principle, as has been reported for para-aminobenzoic acid (PABA).

Considerations on photochemical genotoxicity: report of the International Workshop on Genotoxicity Test Procedures Working Group.

Recent toxicological observations have caused concern regarding the need to test, for example, pharmaceuticals and cosmetic products for photochemical genotoxicity. The objective of this report is to give assistance on how to adapt existing test methods to investigate the potential of light-absorbing compounds to induce genotoxic effects on photoactivation. In general, the Organization for Economic Co-Operation & Economic Development (OECD) draft guideline on in vitro phototoxicity testing served as a basis for consideration. Concomitant exposure of the cells to the test compound and solar simulated light was considered appropriate as the initial, basic test condition. Optimization of the exposure scheme, e.g., a change of the irradiation spectrum, might be indicated depending on the initial test results. Selection of test compound concentrations should be based on results obtained with the dark version of the respective test system but might have to be modified if phototoxic effects are observed. Selection of the irradiation dose has to be performed individually for each test system based on dose-effect studies. The irradiation should induce per se a small, reproducible toxic or genotoxic effect. The report includes a specification of necessary controls, discusses factors that might have an impact on the irradiation characteristics, and gives a rationale for the omission of an external metabolic activation system. It also addresses the question that physicochemical and pharmacokinetic properties might trigger the need to test a chemical for photochemical genotoxicity. Relevant experimental observations are presented to back up the recommendations. The working group did not reach a consensus as to whether a single, adequately perfomed in vitro test for clastogenicity would be sufficient to exclude a photogenotoxic liability or whether a test battery including a gene mutation assay would be needed for product safety testing regarding photochemical genotoxicity.

Oxidative DNA damage and mutations induced by a polar photosensitizer, Ro19-8022.

The oxidative DNA damage induced by the polar photosensitizer Ro19-8022 in the presence of light was studied and correlated with the associated mutagenicity. Both in isolated DNA and AS52 Chinese hamster ovary cells, photoexcited Ro19-8022 gave rise to a DNA damage profile that was similar to that caused by singlet oxygen: base modifications sensitive to the repair endonuclease Fpg protein, which according to high-performance liquid chromatography (HPLC) analysis were predominantly 8-hydroxyguanine (8-oxoG) residues, were generated in much higher yield than single-strand breaks, sites of base loss (AP sites) and oxidative pyrimidine modifications sensitive to endonuclease III. Fifty percent of the Fpg-sensitive modifications were repaired within 2 h. Under conditions that induced 10 Fpg-sensitive modifications per 10(6) bp (six 8-oxoG residues per 10(6) bp), approximately 60 mutations per 10(6) cells were induced in the gpt locus of the AS52 cells. A rather similar mutation frequency was observed when a plasmid carrying the gpt gene was exposed to Ro19-8022 plus light under cell-free conditions and subsequently replicated in bacteria. Sequence analysis revealed that GC-->TA and GC-->CG transversions accounted for 90% of the base substitutions. A significant generation of micronuclei was detectable in AS52 cells exposed to the photosensitizer plus light as well.

Genotoxicity testing of biotechnology-derived products. Report of a GUM task force. Gesellschaft für Umweltmutationsforschung.

Various aspects of genotoxicity testing of biotechnology-derived products are discussed based on information gathered from a questionnaire which was sent to about 30 predominantly European companies. Feedback was received from 13 companies on 78 compounds, mostly recombinant proteins but also on a number of nonrecombinant proteins, which had been assessed for genotoxicity in a total of 177 tests. Four of the 78 compounds appeared to elicit reproducible genotoxic effects. For one of these compounds, the activity could be related to a nonpeptidic linker molecule. No scientifically convincing rationale for the other three compounds could be established, although, at least for two compounds, their activity may be connected with the enzymatic/hormonal activity. In addition to the survey, published reports on genotoxicity testing of biotechnology products were reviewed. The data are discussed relative to whether genotoxicity testing is a valuable exercise when assessing potentially toxic liabilities of biotechnology-derived compounds. It is concluded that genotoxicity testing is generally inappropriate and unnecessary, a position which is in accordance with the available guidelines addressing this area. For the 'average' protein, electrophilic reactions are difficult to envision. Indirect reactions via DNA metabolism and growth regulation seem possible for only very specific proteins such as nucleases, growth factors, cytokines. No information on testing of different types of biotechnology-derived products (e.g., ribozymes, antisense-oligonucleotides, DNA vaccines) has been received in the questionnaires. Discussion of their potential to cause genotoxic changes was based on literature reports. Even for those products for which concerns of genotoxic/tumourigenic potential cannot be completely ruled out, e.g., because of their interaction with DNA metabolism or proliferation control, the performance of standard genotoxicity assays generally appears to be of little value. All information, including also information on the occurrence of genotoxic impurities, has been utilized to formulate a decision tree approach for the genotoxicity testing of biotechnology-derived products.

Suppression of mutagenic activity of a series of 5HT2c receptor agonists by the incorporation of a gem-dimethyl group: SAR using the Ames test and a DNA unwinding assay.

A positive result in the Ames test is generally taken as a strong indication for a genotoxic (i.e. DNA damaging) property of the test compound, often sufficient to cause termination of its development as a new therapeutic agent. A number of serotonin receptor ligands have been tested for their mutagenic potential in the Ames assay at an early stage of development. For several compounds increases in the number of revertant colonies were observed in strain TA1537. Consequently, structure-activity relationship investigations were undertaken to search for compounds without mutagenic liability. All compounds are three ringed heterocyclic structures consisting of a benzene ring, a central (generally non-aromatic) 5- or 6-membered ring and a pyrrole or pyrazole ring. Using a gel shift assay we provide evidence that the observed genotoxic effects are strongly influenced by the intercalating properties of the compounds. The highest mutagenic response was seen with a compound possessing a central aromatic ring. The mutagenic activity of the naphthaleno derivatives appears to be stronger when compared with the indeno compounds, probably because of the less curved structure. Dimethyl substitution of the indeno substructure reduces the intercalating ability of the compounds and leads to loss of mutagenic activity. Pyrazole analogues of both indeno and naphthaleno structures appear to produce stronger mutagenic responses than the pyrrole derivatives.

The photomutagenicity of fluoroquinolones and other drugs.

Induction of DNA damage as a consequence of exposure to UV light has been established as the major and still increasing cause of skin cancer. Absorption of the photon energy may be either directly by the DNA molecules (for wavelengths < 320 nm) or may be by endogenous or exogenous chemicals (sensitizers) with the potential of energy or electron transfer to DNA. Oxygen-mediated reactions (often called type II reactions) appear to be the most important mechanism since molecular oxygen is a good and abundant substrate for triplet excited sensitizers. Energy transfer to molecular oxygen is possible for wavelengths in the near UV and in the visible part of the solar spectrum since the energy of the excited oxygen molecule ((1)O2*) is comparatively low. A few light-absorbing pharmaceuticals have long been known to cause photo(geno)toxic effects. Notably psoralene and chlorpromazine derivatives have been established as photomutagens and the reaction mechanisms have been identified. The fluoroquinolone antibiotics have more recently been recognized as being photomutagenic. The type of DNA damage and the modulation by antioxidants indicate the involvement of reactive oxygen species (ROS) but other mechanisms are also reported at least for some derivatives. In routine genotoxicity studies we observed a photomutagenic activity of a compound under development as an anxiolytic agent in the Ames tester strain TA102 at 'normal laboratory illumination' conditions. Further investigations showed strong photogenotoxic activity in tests for gene mutations and chromosomal aberrations in mammalian cells. The compound proved to be a potent (1)O2-producer. The finding led to termination of development but in the course of studies several structural analogues have been tested for which structure activity relationships will be described. The relevance of photogenotoxic properties of drugs for predicting adverse effects in man will be discussed.

Antimalarial activity of the bisquinoline trans-N1,N2-bis (7-chloroquinolin-4-yl)cyclohexane-1,2-diamine: comparison of two stereoisomers and detailed evaluation of the S,S enantiomer, Ro 47-7737.

The S,S enantiomer of the bisquinoline trans-N1,N2-bis(7-chloroquinolin-4-yl)cyclohexane-1,2-diamine, Ro 47-7737, is significantly more potent against chloroquine-resistant Plasmodium falciparum than the R,R enantiomer and the previously described racemate. Both the enantiomers and the racemate are more potent inhibitors of heme polymerization than chloroquine, and their activities are probably mediated by inhibition of this parasite-specific process. The S,S enantiomer, Ro 47-7737, was studied in more detail and proved to be a potent antimalarial in the treatment of P. vivax ex vivo and P. berghei in vivo. Its suppression of P. berghei growth in a mouse model (50% effective dose, 2.3 mg/kg of body weight) was equal to that of chloroquine and mefloquine, and Ro 47-7737 was found to be more potent than these two drugs in the Rane test, in which the curative effect of a single dose is monitored. The dose at which 50% of animals were permanently cured (34 mg/kg) was markedly superior to those of chloroquine (285 mg/kg) and mefloquine (> 250 mg/kg). When administered orally at 50 mg/kg, Ro 47-7737 also showed a faster clearance of parasites than either chloroquine or mefloquine, and unlike the other two compounds, Ro 47-7737 showed no recrudescence. In a study to compare prophylactic efficacies of oral doses of 50 mg/kg, Ro 47-7737 provided protection for 14 days compared to 3 days for mefloquine and 1 day for chloroquine. The good curative and prophylactic properties of the compound can be explained in part by its long terminal half-life. The ability to generate parasite resistance to Ro 47-7737 was also assessed. With a rodent model, resistance could be generated over eight passages. This rate of resistance generation is comparable to that of mefloquine, which has proved to be an effective antimalarial for many years. Toxicity liabilities, however, ruled out this compound as a candidate for drug development.

Review of the genotoxic properties of chlorpromazine and related phenothiazines.

Chlorpromazine and related phenothiazine drugs have been used in human and veterinary medications for more than 40 years, predominantly as psychotropic agents. Genotoxicity reports are in many cases of relatively antiquated test design. Overall there appears to be no genotoxic activity associated with these drugs when tested under standard conditions. Limited evidence for the potential to form mutagenic nitrosation products and some indication for the ability to modulate the genotoxic action of various mutagens have been presented in the literature. UV irradiation of chlorpromazine and other chlorinated derivatives produces reactive free radicals which possess DNA damaging properties. Induction of gene mutation and chromosomal aberrations have been observed in appropriately designed photomutagenesis experiments. Enhancement but also reduction of UV induced skin tumour formation by chlorpromazine have been found. The decisive factor for the discrepant actions has not been recognized. It is clearly advisable to avoid extensive UV exposure during therapy with these drugs.

Mutation induction and mutation spectra of S. typhimurium TA100 after exposure to isohistidines.

L-Isohistidine and D,L-isohistidine, but not D-isohistidine, caused an increase of the number of mutant colonies in S. typhimurium strain TA100. Spontaneous and also sodium azide or 2-aminoanthracene induced mutant numbers were enhanced by L-isohistidine and by an isomeric mixture of D,L- and L-isohistidine. These effects could not be attributed to a growth-enhancing property. The colony probe hybridization procedure was used to investigate the effects of the histidines on the spontaneous and azide-induced spectra of the hisG46 allele in strain TA100. D,L-Isohistidine, but not the D-isomer, caused and increase of transitions (CCC-->CTC) and transversions (CCC-->CAC) in the spontaneous spectrum. Sodium azide alone induced a strong increase of CCC-->CTC transitions; combination with the D,L-isohistidine led to a further enhancement of this type of base substitutions, whereas with the L-isomer, no such effect was observed. This supports the hypothesis that the activity of D,L-isohistidine is probably not due to DNA-damaging properties, but rather to indirect mechanisms, such as enhancement of the infidelity of DNA replication and/or interference with DNA-repair or proofreading functions.

The photomutagenicity of fluoroquinolones in tests for gene mutation, chromosomal aberration, gene conversion and DNA breakage (Comet assay).

The ability of fluoroquinolones to cause light-induced adverse effects has been established in experimental studies and clinical observations. The formation of active oxygen species appears to be responsible for this activity. Photomutagenicity tests with bacterial, lower eukaryotic and mammalian cells were performed with three fluoroquinolones (Fleroxacin, Ciprofloxacin and Lomefloxacin). After concomitant irradiation with simulated solar light (with a reduced UVB component), weak increases in the number of revertants were observed in Salmonella typhimurium TA104 and TA100. No photomutagenic activity was detected in Saccharomyces cerevisiae D7. In the chromosomal aberration (CA) test with Chinese hamster V79 cells the number of aberrant metaphases was markedly increased. In the Comet assay with mouse lymphoma cells, evidence of extensive DNA breakage was obtained. All three compounds showed similar potencies in the Comet and Ames assays while there was a clear gradation of potencies in the CA assay (Lomefloxacin > Fleroxacin > Ciprofloxacin), which conformed with reports on the relative potencies regarding phototoxicity. The oxygen radical scavengers catalase, superoxide dismutase and N, N'-dimethylurea modulated the photoclastogenicity and phototoxicity but had no influence on the effects in the Comet and Ames tests. It thus appears that different kinds of mechanism are responsible for toxicity and clastogenicity on the one side and DNA breakage and gene mutation on the other side. Pre-irradiation of the test articles did not lead to enhanced genotoxicity, indicating the involvement of very short lived genotoxic agents. The results endorse the advice to avoid excessive light exposure during antibiotic therapy with fluoroquinolones.

Synergistic/comutagenic action in the Ames test as an indication of irrelevant positive findings.

A number of structurally very diverse compounds which cause weak positive effects in the Ames test by evident or suspect irrelevant mechanisms is discussed. As a unifying observation we describe synergistic effects in combination with known mutagens in the responsive strains and comutagenic effects in initially unresponsive strains. We argue that the compounds enhance the formation of spontaneous (or mutagen-induced) revertant colonies by test-specific mechanisms likely to be of no relevance to multicellular eukaryotic organisms rather than possessing intrinsic genotoxic (i.e. DNa-damaging) properties in the Ames test.

Genotoxicity of 17 gyrase- and four mammalian topoisomerase II-poisons in prokaryotic and eukaryotic test systems.

The genotoxic potency of certain classes of topoisomerase II poisons is correlated with their affinity to the topoisomerase protein rather than with the presence of 'classical' structural alerts for DNA reactivity: bacterial topoisomerase II poisons (specifically named gyrase inhibitors) are highly genotoxic in prokaryotic systems; mammalian topoisomerase II poisons are potent mutagens/clastogens in eukaryotic systems. Studies with bacterial, lower eukaryotic and mammalian genotoxicity tests were performed to draw structure-activity conclusions and address risk-benefit considerations for the class of quinolone gyrase inhibitors. All 17 gyrase inhibitors investigated in this study showed genotoxic activity in Salmonella typhimurium strain TA102 and the SOS test. The genotoxic and the toxic activities increased in a highly parallel fashion from the parent compounds, nalidixic acid and oxolinic acid, to the new generation fluoroquinolones. Generally, the most potent fluoroquinolones also show clear-cut positive effects in eukaryotic test systems, although at concentrations 100-1000-fold higher than those effective in bacteria and also 100-1000-fold higher than the minimal genotoxic concentrations of antitumour topoisomerase II inhibitors (ellipticine, teniposide, mAMSA) used as reference compounds. However, subtle structural modifications of the quinolones can strongly diminish the preferential genotoxicity in the prokaryotic test systems.

Evaluation of the micronucleus test in vitro using Chinese hamster cells: results of four chemicals weakly positive in the in vivo micronucleus test.

A rapid and simple procedure for the micronucleus test (MNT) in vitro using Chinese hamster ovary (CHO) cells was established in our laboratory. The assay is intended to quickly screen chromosomal aberrations in vitro within the framework of industrial genotoxicity studies. To test the sensitivity of the assay in the experiments described here, four substances, classified as noncarcinogens but reported as weak inducers of micronuclei (MN) in bone-marrow cells of mice, were evaluated in the MNT in vitro. Of the four compounds, ascorbic acid, phenol, and 2,6-diaminotoluene proved to be genotoxic in the MNT in vitro. Titanium dioxide, which could not be dissolved in the culture medium, did not induce MN. The MNT in vitro proved to be quick and relatively simple and to yield highly reproducible results when testing the four chemicals.

Ro 42-1611 (arteflene), a new effective antimalarial: chemical structure and biological activity.

The discovery of the natural peroxides qinghaosu (arteannuin A, artemisinin) (1) and yingzhaosu A (3) from traditional Chinese herbal medicines was a major advance in the search for new antimalarials (Fig. 1). Whereas qinghaosu can be produced from natural sources and has been well studied, yingzhaosu A has never been available for full evaluation as anti-malarial. We have designed a synthesis of the novel ring system present in yingzhaosu A, the 2,3-dioxabicyclo[3.3.1]nonane and prepared a series of yingzhaosu A analogues which were tested against Plasmodium berghei in mice. Structure-activity rules could be established and used for lead optimization. The best anti-malarial activity was observed for analogues having a keto group within the ring system and an aliphatic or aromatic lipophilic tail as ring substituent. The optimized analogues possessed activity comparable to qinghaosu. In spite of the presence of a peroxide ring, the new compounds were chemically stable against common reagents. In contrast to qinghaosu and its derivatives, they were also stable against hydrolytic decomposition and could therefore be expected to show improved pharmacokinetic properties. As one of the best compounds, Ro 42-1611 (arteflene) (26n, Fig. 2) was selected for detailed preclinical evaluation. Ro 42-1611 (arteflene) was found negative in a battery of mutagenicity tests. It had low acute toxicity after oral or subcutaneous administration. In a 4-week oral tolerance study in rats, doses of up to 400 mg/kg/day were well tolerated.(ABSTRACT TRUNCATED AT 250 WORDS)

Recommendations for the performance of bacterial mutation assays.

At the International Workshop on the Standardisation of Genotoxicity Test Procedures, in Melbourne (27-28 February 1993), the current international guidelines for the correct conduct of bacterial mutation assays were considered, and the major differences between them were examined. An attempt was made to construct a scientifically based, internationally harmonized protocol. The main points of agreement were as follows. The consensus opinion was that there are currently insufficient data to justify a preference for either the preincubation or plate-incorporation methodologies as the initial test. Whichever method is used there was consensus agreement that the bacterial test battery should consist of S. typhimurium TA1537, TA1535, TA98 and TA100. There was also consensus that the 3 strains TA97a, TA97 and TA1537 could be used interchangeably. Although it was not possible to achieve a consensus, the majority of the working group members agreed that strains for the detection of mutagens acting specifically on AT base pairs should be routinely included within the test battery. These strains may be S. typhimurium TA102 or E. coli WP2 strains (WP2 pKM101 and WP2 uvrA or WP2 uvrA pkM101). With regard to study design it was universally agreed that 5 doses of test compound should be used in each experiment, and a majority agreement was obtained for 3 plates per dose. The use of 2 plates per dose is acceptable ONLY if the experiment is repeated. It is recommended that the negative controls may consist of solvent control alone provided that historical data are available to demonstrate lack of effect of the solvent in question. Positive control compounds should be included in all experiments, although the nature of these control compounds need not be specified in the guidelines. There was consensus agreement that for non-toxic freely soluble test agents, an upper limit of 5 mg/plate should be tested (5 microliters per plate for liquids). For insoluble or toxic compounds, the recommendations were the same as those for other in vitro tests (see appropriate paper). A consensus agreement was reached on the need to carry out further tests if equivocal results are obtained in the initial test, although it was generally agreed that the design of the repeat study should be left flexible. As there are little or no data to support the use of an exact repeat assay, a majority of the group recommended that negative results in the first test should be further investigated by either conducting a modified repeat (e.g. S9 titration) or by conducting the alternative methodology.(ABSTRACT TRUNCATED AT 400 WORDS)