Ž .Mutation Research 412 1998 251–260

Chemicals mutagenic in Salmonella typhimurium strain TA1535but not in TA100

Michael J. Prival a,), Errol Zeiger b

a ( )Genetic Toxicology Branch HFS-236 , Food and Drug Administration, Washington, DC 20204, USAb EnÕironmental Toxicology Program, National Institute of EnÕironmental Health Sciences, Research Triangle Park, NC 27709, USA

Received 24 July 1997; revised 3 November 1997; accepted 4 November 1997

Abstract

Ž .The standard Salmonella mutagenicity test uses two strains of Salmonella typhimurium TA1535 and TA100 containingŽ .the same base pair substitution mutation hisG46 . These strains differ only in that strain TA100 contains the plasmid

pKM101, whose mucAB gene products enhance SOS mutagenesis. This makes strain TA100, in general, the more sensitiveof the two for mutagen detection, raising the question as to whether or not to include strain TA1535 in the core battery ofstrains in routine testing. Out of 659 chemicals judged as mutagens in the S. typhimurium assay when subjected to the

Ž .National Toxicology Program’s screening protocol, 36 5% were evaluated as positive in strain TA1535 but not in strainTA100. Of these, 23 were judged as negative and 13 as equivocal in strain TA100, and 5 were positive or equivocal in at

Ž .least one other strain TA97 or TA98 . In general, the data on these chemicals indicate that the absolute increases inrevertants per plate induced in strain TA1535 were too small to have been judged as positive if similar increases occurred in

Žstrain TA100, which has a much higher spontaneous background. For three chemicals acetaldehyde oxime, 6-mercap-.topurine, and 1,3-butadiene the absolute increases in revertants in strain TA1535 greatly exceeded those in strain TA100.

Evaluation of the reproducibility of these findings and of the mechanisms and relevance of unique TA1535 positives shouldbe useful when decisions are made as to whether this strain should be kept as a part of the core battery of strains in the S.typhimurium assay. q 1998 Elsevier Science B.V.

Keywords: Salmonella typhimurium; Ames test; TA1535; TA100; Base pair substitution mutation

1. Introduction

The Salmonella typhimurium mutagenicity assayw xdeveloped by Ames et al. 1 and Maron and Ames

) Ž .Corresponding author. Genetic Toxicology Branch HFS-236 ,Food and Drug Administration, 200 C Street SW, Washington,DC 20204, USA. Tel.: q1-202-205-4347; fax: q1-202-205-4758;E-mail: mprival@bangate.fda.gov

w x2 is the most widely used test for assessing thegenetic toxicity of chemicals. The bacterial strainsused in this test have mutations in genes coding forenzymes required for the biosynthesis of the aminoacid histidine. Several of these strains containframeshift mutations and thus respond to chemicalsthat induce frameshift mutations; others have basepair substitutions and are used to detect chemicalsthat induce this type of mutation.

1383-5718r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.Ž .PII S1383-5718 97 00196-4

( )M.J. PriÕal, E. ZeigerrMutation Research 412 1998 251–260252

It is generally accepted that a standard Salmonellamutagenicity test will include at least twoframeshift-sensitive strains, usually strain TA98 and

Ž .either strain TA1537 or TA97 or its variant, TA97a .Originally, the standard test also included strainTA1538, which contains the same hisD3052 muta-tion as strain TA98. Unlike strain TA1538, strainTA98 contains the plasmid pKM101, and the prod-ucts of the mucAB genes on this plasmid enhanceSOS mutagenesis, thereby making strain TA98 moreresponsive to certain mutagens than is strain TA1538w x2 . As a result, strain TA1538 is now seldom used inroutine screening of chemicals for mutagenicity.

Similarly, the original test battery of S. ty-phimurium strains included strain TA1535, contain-ing the base pair substitution mutation hisG46. ThisAT™GC mutation leads to the insertion of prolineŽ . Ž .codon CCC rather than leucine CTC in thepolypeptide chain of ATP-phosphoribosyltransferase,the first enzyme in the pathway of histidine biosyn-thesis. Hisq revertants can arise from a variety of

Ž .intragenic transitions GC™AT and transversionsŽ .GC™TA and GC™CG and by extragenic sup-

Ž .pressor mutations AT™CG in two threonine-tRNAw xgenes 3 . When the plasmid pKM101 was intro-

duced into strain TA1535, the resulting strain,TA100, became the most sensitive of all of the S.

w xtyphimurium tester strains 4 . However, the originalbase pair substitution strain, TA1535, continued tobe part of the test battery because it was thought thatsome mutagenic chemicals would be detected in thisstrain but not detected when strain TA100 was used.

To help evaluate the usefulness of the continuedinclusion of strain TA1535 in the standard Salmonellatester strain battery, we have reviewed and summa-rized the data for all of the chemicals that werescored as positive in strain TA1535 but negative in

Ž .TA100 in the National Toxicology Program’s NTPmutagenicity testing program from its inception in1979 through December, 1996.

2. Materials and methods

As of December 1996, there were 1865 chemicalstested for mutagenicity in Salmonella by the NTP.All testing is done on coded chemicals by using

standard protocols. All test results, including thosenot previously reported, are from pre-incubation as-

w xsays performed as previously described 5,6 . Onechemical, 1,3-butadiene, was tested in desiccators as

w xdescribed in Ref. 6 . The current protocol in theNTP testing program calls for the testing of chemi-cals in strains TA98 and TA100 with and without S9before other strains are used. If positive results areobtained in either of these two strains, the test isrepeated to confirm the positive finding, and thechemical is evaluated as mutagenic. Further testing isgenerally not done. If the chemical is not mutagenicin the initial test with strains TA98 and TA100, it istested in two additional strains, TA97 and TA1535.This strategy was adopted after an evaluation of theeffectiveness of the different strains for detecting

w xmutagens 4 . As a result of this sequential testingstrategy, many chemicals found to be mutagenic instrain TA100 were never tested by the NTP in strainTA1535. However, all 1657 chemicals tested in strainTA1535 were also tested in strain TA100.

3. Results

A summary of the evaluations of the 1657 chemi-cals that were tested both in strain TA1535 andTA100 is shown in Table 1. Not included in thistable are chemicals that were tested in the NTPProgram by using strains TA100 and TA98 andfound to be mutagenic in one or both of these strainsand were therefore not tested in strain TA1535. Ofthe 173 chemicals evaluated as mutagenic in strain

Ž . Ž .TA1535, 23 13% were called negative and 13 8%equivocal in strain TA100. Five of the chemicals

Table 1Summary of results obtained on 1657 chemicals that were testedfor mutagenicity in strains TA100 and TA1535 by the NationalToxicology Program

TA100 TA1535

q y ?

q 137 199 13y 23 1213 10? 13 46 3

?sevaluated as equivocal. For this table, chemicals judged muta-Ž . Ž .genic q or weakly mutagenic qw were combined as q.

( )M.J. PriÕal, E. ZeigerrMutation Research 412 1998 251–260 253

called positive in strain TA1535 but not in TA100were judged mutagenic or equivocal in TA97 androrTA98. Thus, out of the total of 659 chemicals judged

Ž .to be mutagenic in the NTP program, 36 5% werepositive in strain TA1535 but negative or equivocal

Ž .in strain TA100 while 31 5% were uniquely posi-tive in strain TA1535.

Table 2 is a list of the chemicals that were judgedto be mutagenic in strain TA1535 but not in strain

TA100 in the NTP testing program, along withliterature citations containing the data from thesestudies. The results from six of these chemicalsŽacetaldehyde oxime, butanal oxime, 1,3-dimethyl-2-nitrobenzene, methyl ethyl ketoxime, p-nitro-

.benzamide, and tricaprylin have not previously beenpublished, and are therefore displayed in figuresbelow. Because strains TA1535 and TA100 bothcontain the same hisG46 mutation and differ only by

Table 236 chemicals evaluated as mutagenic in strain TA1535 but negative or equivocal in strain TA100 by the National Toxicology Program

a bChemical CAS No. Overall evaluation Other strains References

TA100 TA1535

w x2-Amino-4-chlorophenol 92-85-2 y qw 23w xAzoxymethane 25843-45-2 y qw 6w x1,2,3-Benzotriazole 95-14-7 y qw 7w x1,3-Butadiene 106-99-0 y q 24w xCalcium cyanamide 156-62-7 y qw 5w xCyclohexanone oxime 100-64-1 y qw 25w xC.I. direct green 1 3626-28-6 y qw 7w xEosin 17372-87-1 y q TA97 6w x2,3-Epoxybutane 3266-23-7 y q 6w xN-Ethylmorpholine 100-74-3 y qw 7w xHydrazine sulfate 10034-93-2 y q 26w xMaleic hydrazide, diethanolamine salt 5716-15-4 y qw 26w x6-Mercaptopurine monohydrate 6112-76-1 y q 6

Methyl ethyl ketoxime 96-29-7 y qp-Nitrobenzamide 619-80-7 y qw

w xPhenobarbital 50-06-6 y qw 27w x2-Phenyl-2-ethylmalondiamide 80147-40-6 y qw 23w xPrimaclone 125-33-7 y q 26

Ž . w xS- n-Propyl chlorothioformic acid 13889-92-4 y q 6w x1,3-Propylene oxide 503-30-0 y q 23w xTetraethyl dithiopyrophosphate 3689-24-5 y q 23w xToluene ethylsulfonamide 8047-99-2 y qw 6

Tricaprylin 538-23-8 y qAcetaldehyde oxime 107-29-9 ? q

w xAcetin 26446-35-5 ? q 5w xN-Acetyl-o-toluidine 120-66-1 ? qw 5

Butanal oxime 110-69-0 ? qww xChlorinated trisodium phosphate 56802-99-4 ? qw TA97 23w x2-Chloroethanol 107-07-3 ? q 5w x1-Chloro-2-propanol 127-00-4 ? q 7w xDimenhydrinate 523-87-5 ? q 7

1,3-Dimethyl-2-nitrobenzene 81-20-9 ? qww xC.I. disperse blue 1 2475-45-8 ? q TA97; TA98 23w xN-Hydroxyethyl ethylenediamine 111-41-1 ? qw 7w x4-Nitrophthalic anhydride 5466-84-2 ? q TA97; TA98 23w xPotassium bromate 7758-01-2 ? q TA97 6

aOverall evaluation: q, mutagenic; qw, weakly mutagenic; ?, equivocal response; y, non-mutagenic.bOther strains: Salmonella strains, other than TA100 or TA1535, in which the chemical was mutagenic or induced an equivocal response.

( )M.J. PriÕal, E. ZeigerrMutation Research 412 1998 251–260254

the presence of plasmid pKM101 in strain TA100,there are two possible reasons why a chemical maybe mutagenic in strain TA1535 but not in TA100.First, the chemical may induce a low level of muta-tions via a non-SOS dependent pathway, for exam-ple, by inducing DNA lesions that have non-pairingcharacteristics. If the absolute number of revertantsinduced in this manner is small and similar in thetwo bacterial strains, then the mutagenicity of thetest chemical will, in general, be more easily de-tected in strain TA1535 because the background

Žmutant count per plate is much lower generally. Ž .8–20 in strain TA1535 than it is in TA100 90–160 .

For example, a chemical that causes an increase of20 revertants over a background of 10–20 in TA1535would be judged a mutagen, whereas the same 20revertants over a background of G90 in strain TA100would not be considered as indicative of a mutagenic

Ž .Fig. 1. Mutagenicity of acetaldehyde oxime. A Hamster liver S9.Ž . Ž . Ž .B Rat liver S9. - - - 10% S9 in S9 mix. — 30% S9. Testsperformed by SRI International, Menlo Park, CA. Chemical ob-tained from Aldrich Chemical, 99% purity; dissolved in water.

Ž . Ž .Fig. 2. Mutagenicity of acetin. — Without S9. - - - 10%Ž .hamster liver S9. — - - — 10% rat liver S9.

response. The second possible explanation for find-ing a mutagen in strain TA1535 but not in TA100 isthat a chemical may induce more mutations in strainTA1535 than it does in strain TA100, though it isdifficult to imagine a molecular mechanism for suchdifferential activity.

When we analyzed the data on the 36 chemicalsthat were judged mutagenic in strain TA1535 andnon-mutagenic or equivocal in TA100, we found that

Ž23 of them showed mutagenic responses increase in.revertants per plate in strain TA1535 that were so

small that they could easily have been missed if thesame increases had occurred in the higher back-ground strain TA100. The data on the remaining 13chemicals indicated that actual increases in mutantcounts in strain TA1535, at least under some testconditions, may have exceeded those seen in TA100,

Fig. 3. Mutagenicity of azoxymethane with 30% hamster liver S9.

( )M.J. PriÕal, E. ZeigerrMutation Research 412 1998 251–260 255

Ž .Fig. 4. Mutagenicity of 1,2,3-benzotriazole. - - - 10% hamsterŽ .liver S9. — 10% rat liver S9.

Ž . Ž .Fig. 5. Mutagenicity of 1,3-butadiene. — Without S9. - - -Ž .30% hamster liver S9. — - - — 30% rat liver S9. Tests

performed in desiccator at Microbiological Associates, Rockville,MD. Chemical obtained from Matheson Gas Products; purityinformation not available; tested as a gas.

Ž .Fig. 6. Mutagenicity of cyclohexanone oxime. - - - 10% hamsterŽ .liver S9. — 30% hamster liver S9. Tests performed by SRI

International, Menlo Park, CA. Chemical obtained from AldrichChemical, 97% purity; dissolved in DMSO.

Ž . Ž .Fig. 7. Mutagenicity of eosin. - - - 10% hamster liver S9. —30% hamster liver S9.

though in most cases no firm conclusion on thispoint could be drawn from the available data.

Figs. 1–13 show mutagenicity data on these 13chemicals. In the figures, responses evaluated as

Ž .Fig. 8. Mutagenicity of 6-mercaptopurine monohydrate. AŽ . Ž . Ž . Ž .— Without S9. B - - - 30% hamster liver S9. — 30% ratliver S9.

( )M.J. PriÕal, E. ZeigerrMutation Research 412 1998 251–260256

Fig. 9. Mutagenicity of methyl ethyl ketoxime using 30% hamsterliver S9. Tests performed by Microbiological Associates,Rockville, MD. Chemical obtained from Pfaltz and Bauer, )99%purity; dissolved in DMSO.

Ž .Fig. 10. Mutagenicity of 1,3-propylene oxide. - - - 30% hamsterŽ .liver S9. — 30% rat liver S9.

Ž .Fig. 11. Mutagenicity of S- n-propyl chlorothioformic acid with30% hamster liver S9.

Ž .Fig. 12. Mutagenicity of tetraethyl dithiopyrophosphate. - - -Ž .10% hamster liver S9. — 30% hamster liver S9.

Fig. 13. Mutagenicity of tricaprylin with 30% rat liver S9.

Ž .Fig. 14. Mutagenicity of butanal oxime. - - - 10% rat liver S9.Ž .— 30% rat liver S9. Tests performed by SRI International,Menlo Park, CA. Chemical obtained from Pfaltz and Bauer,)99% purity; dissolved in DMSO.

( )M.J. PriÕal, E. ZeigerrMutation Research 412 1998 251–260 257

Fig. 15. Mutagenicity of 1,3-dimethyl-2-nitrobenzene with 30%hamster liver S9. Tests performed by SRI International, MenloPark, CA. Chemical obtained from Aldrich, 99% purity; dissolvedin DMSO.

Ž .non-mutagenic are designated y , responses evalu-Ž .ated as positive as q , those evaluated as weakly

Ž .positive as qw , and those evaluated as equivocalŽ . Žas ? . Only 3 of these 13 chemicals acetaldehyde

.oxime, 1,3-butadiene, and 6-mercaptopurine showclear evidence of a greater induction of mutants instrain TA1535 than in TA100.

Figs. 14–16 show data obtained with butanaloxime, 1,3-dimethyl-2-nitrobenzene, and p-nitro-benzamide, where the increases observed with strainTA1535 were not great enough to have resulted in apositive response if the same absolute increases hadoccurred in strain TA100. The chemicals selected for

Ž .Fig. 16. Mutagenicity of p-nitrobenzamide. - - - 10% hamsterŽ .liver S9. — 30% hamster liver S9. Tests performed by Microbi-

ological Associates, Rockville, MD. Chemical obtained fromAldrich, 98% purity; dissolved in DMSO.

these figures are those for which data from theNTP-sponsored studies have not been previouslypublished.

4. Discussion

We have presented data on chemicals that havebeen evaluated as mutagenic when tested in the NTPmutagenicity testing program by using S. ty-phimurium strain TA1535, but not when tested usingstrain TA100. The evaluation method used by the

w xNTP, as described previously 6,7 , involves the useof subjective judgment, and the distinctions between

Žcategories non-mutagenic vs. equivocal; equivocalvs. weakly mutagenic; weakly mutagenic vs. muta-

.genic are not based on restrictive or statistical rules.The figures presented here and the references citedin Table 2 will enable interested readers to reevalu-ate the data according to their own methods orcriteria.

It might be asserted that some chemicals evalu-ated as mutagenic in strain TA1535 or TA100 do notmeet the widely accepted ‘two-fold’ criterion formutagenicity. Of course, the two-fold rule is, like theNTP evaluation method, quite arbitrary and of un-known statistical relevance since the underlying dis-tribution of replicate control mutant counts has not

w xbeen well defined 8 . It has been shown that thetwo-fold rule is, by statistical criteria, overly conser-vative for strains with high background rates, such asTA100, but allows too many false positives with

w xstrains with low background rates 8 . Furthermore, ithas been found that the use of evaluation criteriamore stringent than those applied by the NTP doesnot improve the predictivity of the Salmonella assaywith regard to the carcinogenicity of the test chemi-

w xcals 9 .Strain TA100 generally has a higher sensitivity to

mutagenic insult as compared to strain TA1535 dueto the enhancement of SOS mutagenesis facilitatedby the mucAB genes on the plasmid pKM101 that is

w xpresent in TA100 10 . However, cases in whichstrain TA1535 is more sensitive than strain TA100 tothe mutagenic effects of chemicals have been re-ported previously by others. Even in the originalpublication on the construction of strain TA100,

( )M.J. PriÕal, E. ZeigerrMutation Research 412 1998 251–260258

w xMcCann et al. 11 reported that diethylsulfate wasmuch more potent as a mutagen in strain TA1535than in strain TA100. Similarly, previous studies bythe NTP found that, for example, 1-aziridine ethanolinduced more mutants in strain TA1535 than in

w x w xstrain TA100 5 . More recently, Maragos et al. 12found that glyceryl trinitrate induces reversion instrain TA1535 in the plate incorporation assay, butnot in strain TA100. They suggest that this phe-nomenon is indicative of efficient repair of DNAdamage along with equal or greater cytotoxicity inthe presence of pKM101.

w xAshby et al. 13 reported that the rat liver car-cinogen methapyrilene was mutagenic in strainTA1535, but not in TA100; the increases in mutantyield seen in TA1535 were at concentrations thatwere toxic to TA100. For the three chemicals re-ported on here in which mutagenicity in strainTA1535 was much greater than that seen in strain

ŽTA100 acetaldehyde oxime, 1,3-butadiene, and 6-.mercaptopurine , there was no indication of toxicity

in strain TA100 except for acetaldehyde oxime testedŽ .with hamster liver S9 Fig. 1A . In certain other

Ž .cases e.g., azoxymethane, Fig. 3; and eosin, Fig. 7 ,however, decreasing revertant counts in strain TA100at doses that were judged to be mutagenic in strainTA1535 may indicate that the chemicals were, infact, more toxic to the pKM101-bearing strain.

It is presumed that the DNA damage caused bychemicals that induce equivalent numbers of mutantsin strains TA100 and TA1535 does not lead totranslesion DNA repair synthesis associated withSOS mutagenesis. Presumably, these chemicals ei-ther alkylate DNA bases in such a way as to result in

Žsimple mispairing during DNA synthesis e.g., ac-.etaldehyde oxime and 1,3-butadiene , or are them-Žselves incorporated into the DNA e.g., 6-mercap-

.topurine , resulting in similar, simple mispairing.Such mechanisms would be expected to result inmutagenic potencies that are similar in strainsTA1535 and TA100, rather than significantly greaterin strain TA1535, as has been found in the experi-ments reported here. Thus, the reason for the highersensitivity of TA1535 seen with these chemicalsremains to be elucidated.

Although there were 13 chemicals whose absoluteŽmutagenic activity in terms of induced revertants

.per plate at particular doses may have been greater

Ž .in strain TA1535 than in strain TA100 Figs. 1–13 ,such differences were striking in only three cases:

Ž . Ž .acetaldehyde oxime Fig. 1 , 1,3-butadiene Fig. 5 ,Ž .and 6-mercaptopurine Fig. 8 .

In contrast to the results presented here, Rogers-w xBack et al. 14 reported that acetaldehyde oxime

was non-mutagenic in several strains of S. ty-phimurium, including TA1535 and TA100, at dosesup to 10 mg per plate. It is possible that their use ofthe standard plate incorporation procedure rather thana pre-incubation protocol could account for the dif-ference between their findings and those shown inFig. 1.

The higher sensitivity of strain TA1535 to themutagenic effects of 1,3-butadiene, as compared tostrain TA100, has been reported by de Meester et al.w x w x15 and Araki et al. 16 . Experiments by de Meester

w x Žet al. 17 with strain TA1530 which, like strains.TA1535 and TA100, contains the hisG46 allele ,

showed that 1,3-butadiene induced mutants on plateslacking S9 metabolic activation only when theseplates were incubated in a desiccator along withplates containing S9. This phenomenon apparentlyresulted from the migration of volatile mutagenicmetabolites of 1,3-butadiene from plates with S9 toplates without S9 within the desiccator. In the NTP

Ž .studies Fig. 5 , the mutagenicity observed in theabsence of S9 was presumably caused by suchvolatile metabolites, because the plates with andwithout S9 were incubated in the same desiccator. Apresumed oxidative metabolite, butadiene monoxideŽ .1,2-epoxy-3-butene , a direct-acting mutagen in bothstrains TA1535 and TA100, was, like 1,3-butadieneitself, found to induce more mutants per plate in

w xstrain TA1535 than in TA100 15 .Ž .The NTP results with 6-mercaptopurine Fig. 8

w xare similar to the findings of Marhan 18 , who hasreported that 6-mercaptopurine was mutagenic instrain TA1535, inducing over 400 revertants perplate at a dose of 100 mg, but was not mutagenic in

Ž .strain TA100 by the ‘two-fold rule’ criterion in theplate incorporation assay with and without metabolicactivation. Others, however, have found that 6-mercaptopurine induced more revertants in strainTA100 than in strain TA1535 in the pre-incubation

w xassay 19 and in the standard Ames plate incorpora-w xtion assay 20 . This array of results raises the ques-

tion of reproducibility. It should be noted that only

( )M.J. PriÕal, E. ZeigerrMutation Research 412 1998 251–260 259

one test was performed in TA100 under each S9Ž .condition in the NTP experiments Fig. 8 , as dis-

cussed below.For the ten other chemicals whose results are

depicted in Figs. 1–13, the higher sensitivity ofstrain TA1535, as compared to strain TA100, wasnot as pronounced as it was for acetaldehyde oxime,1,3-butadiene, and 6-mercaptopurine. For these tenchemicals, it is difficult to conclude, based on thesestudies alone, that strain TA1535 is more sensitivethan strain TA100. For example, in the data pre-

Ž .sented on 1,2,3-benzotriazole Fig. 4 , with the ex-ception of one point in one test using strain TA1535in the presence of rat liver S9, the increase inrevertants per plate induced in strain TA1535 maynot be any greater than that in strain TA100. In a

w xprevious collaborative study on this chemical 21 ,Ž .this chemical under the name 1-H-benzotriazole

was found to be mutagenic or questionable in strainTA1535 in the presence of S9 in two laboratories butnegative in two other laboratories, while the resultsin strain TA100 were scored as non-mutagenic in allcases. Even the published data from this large collab-orative study do not clearly answer the question as towhether benzotriazole can induce more mutants perplate in strain TA1535 than in strain TA100.

For all but two of the 13 chemicals for which themutagenic responses in strain TA1535 exceeded thosein TA100, tests were performed under the relevantS9 conditions only one time in strain TA100. Thisleaves open the possibility that the differences seenin the responses between the two strains might notbe reproducible, though it has previously been foundthat fewer than 1% of chemicals evaluated as nega-tive upon initial testing in strain TA100 in the NTPprogram gave a different result upon repeat testingw x22 . The two chemicals for which replicate testswere performed in strain TA100 were acetaldehyde

Ž . Ž .oxime Fig. 1 and methyl ethyl ketoxime Fig. 9 .The data summarized here should be useful when

decisions are made as to which strains to include inthe test battery for mutagenicity screening of chemi-cals in bacteria. In particular, these data should shedlight on the usefulness of S. typhimurium strainTA1535. We would conclude, based on the datapresented here, that there is not, at the present time,a sufficient basis for eliminating strain TA1535 fromthe standard test battery. However, additional data or

further analysis may justify a different conclusion inthe future.

References

w x1 B.N. Ames, J. McCann, E. Yamasaki, Methods for detectingcarcinogens and mutagens with the Salmonellarmam-malian–microsome mutagenicity test, Mutation Res. 31Ž .1975 347–364.

w x2 D. Maron, B.N. Ames, Revised methods for the SalmonellaŽ .mutagenicity test, Mutation Res. 113 1983 173–215.

w x3 E. Kupchella, W. Koch, T.A. Cebula, Mutant alleles oftRNAThr genes suppress the hisG46 missense mutation in

Ž .Salmonella typhimurium, Environ. Mol. Mutagen. 23 199481–88.

w x4 E. Zeiger, K.J. Risko, B.H. Margolin, Strategies to reducethe cost of mutagenicity screening with the Salmonella assay,

Ž .Environ. Mutagen. 7 1985 901–911.w x5 S. Haworth, T. Lawlor, K. Mortelmans, W. Speck, E. Zeiger,

Salmonella mutagenicity test results for 250 chemicals, Envi-Ž .ron. Mutagen. 5 1983 3–142, suppl. 1.

w x6 E. Zeiger, B. Anderson, S. Haworth, T. Lawlor, K. Mortel-mans, Salmonella mutagenicity tests: V. Results from the

Ž .testing of 311 chemicals, Environ. Mol. Mutagen. 19 19922–141, suppl. 21.

w x7 E. Zeiger, B. Anderson, S. Haworth, T. Lawlor, K. Mortel-mans, W. Speck, Salmonella mutagenicity tests: III. Resultsfrom the testing of 255 chemicals, Environ. Mutagen. 9Ž .1987 1–110, suppl. 9.

w x8 N.F. Cariello, W.W. Piegorsch, The Ames test: the two-foldŽ .rule revisited, Mutation Res. 369 1996 23–31.

w x9 M.J. Prival, V.C. Dunkel, Reevaluation of the mutagenicityand carcinogenicity of chemicals previously identified asfalse positives in the Salmonella typhimurium mutagenicity

Ž .assay, Environ. Mol. Mutagen. 13 1989 1–24.w x10 W.H. Koch, R. Woodgate, The SOS response, in: J.A.

Ž .Nickoloff, M.F. Hockstra Eds. , DNA Damage and Repair,Vol. 1; DNA Repair in Prokaryotes and Lower Eukaryotes,Humana Press, Totowa, NJ, 1998, pp. 107–134.

w x11 J. McCann, N.E. Spingarn, J. Kobori, B.N. Ames, Detectionof carcinogens as mutagens: bacterial tester strains with R

Ž .factor plasmids, Proc. Natl. Acad. Sci. U.S.A. 72 1975979–983.

w x12 C.M. Maragos, A.W. Andrews, L.K. Keefer, R.K. Elespuru,Ž .Mutagenicity of glyceryl trinitrate nitroglycerin inŽ .Salmonella typhimurium, Mutation Res. 298 1993 187–195.

w x13 J. Ashby, R.D. Callander, D. Paton, E. Zeiger, F. Ratpan,Weak and unexpected mutagenicity to Salmonella of the rathepatocarcinogen methapyrilene, Environ. Mol. Mutagen. 12Ž .1988 243–252.

w x14 A.M. Rogers-Back, T.E. Lawlor, T.P. Cameron, V.C. Dunkel,G enotoxicity of 6 oxim e com pounds in theSalmonellarmammalian–microsome assay and mouse lym-

qry Ž .phoma TK assay, Mutation Res. 204 1988 149–162.w x15 C. de Meester, F. Poncelet, M. Roberfroid, M. Mercier,

( )M.J. PriÕal, E. ZeigerrMutation Research 412 1998 251–260260

Mutagenicity of butadiene and butadiene monoxide, Biochem.Ž .Biophys. Res. Commun. 80 1978 298–305.

w x16 A. Araki, F. Noguchi, F. Kato, T. Matsushima, Improvedmethod for mutagenicity testing of gaseous compounds by

Ž .using a gas sampling bag, Mutation Res. 307 1994 335–344.w x17 C. de Meester, F. Poncelet, M. Roberfroid, M. Mercier, The

mutagenicity of butadiene towards Salmonella typhimurium,Ž .Toxicol. Lett. 6 1980 125–130.

w x18 J. Marhan, Mutagenicity of cytostatic drugs in a bacterialŽ .system: I. Ames test, Folia Microbiol. 40 1995 457–461.

w x19 I. Nepomnaschy, S.E. Sommer, R. Nagel, Effect of the ratliver S9 fraction on the mutagenicity of azathioprine in theSalmonellarmammalian microsome assay, Experientia 40Ž .1984 370–371.

w x20 T.S. Matney, T.V. Ngyuen, T.H. Connor, W.J. Dane, J.C.Theiss, Genotoxicity classification of anti-cancer drugs, Ter-

Ž .atog. Carcinog. Mutag. 5 1985 319–328.w x21 V.C. Dunkel, E. Zeiger, D. Brusick, E. McCoy, D. McGre-

gor, K. Mortelmans, H.S. Rosenkranz, V.F. Simmon, Repro-ducibility of microbial mutagenicity assays: II. Testing ofcarcinogens and non-carcinogens in Salmonella typhimurium

Ž .and Escherichia coli, Environ. Mutagen. 7 1985 1–248,suppl. 5.

w x22 D.S. Gatehouse, Haworth, T. Cebula, E. Gocke, L. Kier, T.Matsushima, C. Melcion, T. Nohmi, T. Ohta, S. Venitt, E.

Zeiger, Recommendations for the performance of bacterialŽ .mutation assays, Mutation Res. 312 1994 217–233.

w x23 E. Zeiger, B. Anderson, S. Haworth, T. Lawlor, K. Mortel-mans, Salmonella mutagenicity tests: IV. Results from the

Ž .testing of 300 chemicals, Environ. Mol. Mutagen. 11 19881–158, suppl. 12.

w x24 DHHS, Toxicology and carcinogenesis studies of 1,3-butadiene in B6C3F1 mice, inhalation studies, CAS No.106-99-0, NTP Technical Report No. 434, US Department ofHealth and Human Services, May 1993.

w x25 DHHS, NTP technical report on toxicity studies of cyclohex-anone oxime administered by drinking water to B6C3F1mice, CAS No. 100-64-1, Toxicity Report Series No. 50, USDepartment of Health and Human Services, NIH Publication96-3934, April 1996.

w x26 K. Mortelmans, S. Haworth, T. Lawlor, W. Speck, B. Tainer,E. Zeiger, Salmonella mutagenicity tests: II. Results from the

Ž .testing of 270 chemicals, Environ. Mutagen. 8 1986 1–119,suppl. 7.

w x27 E. Zeiger, S. Haworth, Tests with a pre-incubation modifica-tion of the Salmonellarmicrosome assay, in: J. Ashby, F.J.deSerres, M. Draper, M. Ishidate Jr., B.H. Margolin, B.

Ž .Matter, M.D. Shelby Eds. , Collaborative Study of Short-Term Tests for Carcinogens, Progress in Mutation Research,Vol. 5, Elsevier, Amsterdam, 1985, pp. 187–199.