Effect of nucleotide excision repair on hprt gene mutations in rodent cells exposed to DNA ethylating agents

The role of the nucleotide excision repair (NER) pathway inremoval of DNA ethylation damage was investigated by means ofhprt mutational spectra analysis in the NER-deficient Chinesehamster ovary cell line UV5, which lacks ERCC2/XPD, and in itsparental cell line AA8. Both cell lines were exposed to ethylmethanesulfonate (EMS) or N-ethyl-N-nitrosourea (ENU). EMS gavea similar dose-dependent increase in hprt mutants in UV5 comparedwith AA8. In both cell lines EMS-induced mutations in the hprtcoding region consisted almost exclusively of GC     

Spontaneous mutation spectrum in the hprt gene in human lymphoblastoid TK6 cells

A spectrum of 100 mutations in the endogenous hprt gene of thehuman lymphoblastoid TK6 cell line is presented. The majorityof the mutations originates in sequences outside the codingregion of the gene. Large deletions are a major cause of inactivationof the hprt gene (57% of the mutants). Mutations in the splicesites that result in several forms of aberrantly spliced mRNAare relatively frequently recovered (16%) compared with mutantscontaining alterations in the coding region of the hprt gene(27%). The majority, but not all, of the splice mutants containan alteration in the consensus sequences of the splice sites.A spectrum of mutations in the coding region of the hprt geneenlarged to a total of 42 mutants shows that basepair substitutionspredominate (71%) and that small deletions and insertions areless frequently recovered. Basepair substitutions arise slightlymore frequently at GC basepairs than at AT basepairs. ³To whom correspondence should be addressed     

Molecular spectrum of background mutation at the hprt locus in human T-lymphocytes

The molecular basis of somatic mutation at the hypoxanthineguaninephosphoribosyl-transerase (hprt) locus in human 6-thioguanineresistant T-cell clones from 17 individuals has been studiedby Southern blot analysis, multiplex PCR (polymerase chain reaction)and direct sequencing of PCR amplified hprt cDNAs or genomicDNA. Twenty-three novel mutations were detected, which in additionto previously described mutations provide a background mutationalspectrum based on a total of 45 hprt mutations in human T-cells.Twenty T-cell mutants had base substitutions in the coding regionleading to 15 missense and five nonsense mutations. In additionto five frameshift mutations caused by four small deletionsand one duplication, seven splice mutations, three of them withskipping of exon 8, were detected. Thirteen genomic structuralalterations have also been identified; one of these had a genomicexon 1 deletion with a GGCCGG-hexamer in both breakpoints.     

Spectrum of spontaneously occurring mutations in the HPRT gene of the Chinese hamster V79 cell mutant V-H4, which is homologous to Fanconi anemia group A

The mitomycin C (MMC)-hypersensitive Chinese hamster V79 cellmutant V-H4 has a cellular phenotype similar to Fanconi anemia(FA), and has been shown to be homologous to FA group A. Toexamine consequences of the defect in V-H4 cells on spontaneousmutagenesis, we studied the frequency and nature of spontaneousmutations at the hypoxanthine phosphoribosyltransferase (HPRT)locus in this mutant and the parental V79 cells. The mutationrates expressed as the number of mutations per cell per generationwere 8.7 X10^–7 and 3.7 X10^–7 for V-H4 and V79 cellsrespectively. The molecular spectrum of 42 spontaneous hprtmutants of V-H4 cells was determined and compared with the previouslydescribed spectrum of spontaneous mutations at the HPRT locusof Chinese hamster V79 cells. The spectra of spontaneous mutationsin the hprt gene of both cell lines are predominated by basepair substitutions and splice mutations. Among the base changes,V-H4 shows a larger frequency of transitions (13/42; 31%) thantransversions (3/42; 7%), whereas in V79 transversions are observedmore often than transitions (P < 0.001; Wilcoxon test). Thefrequency of splice mutations in V-H4 (17/42; 40%), which affectsexon 4 almost exclusively, is not significantly different fromV79. The fraction of deletions in V-H4 is low (6/42; 14%), andcomparable to the level in V79. This is in contrast with thepublished molecular spectrum of spontaneous hprt mutants inFA (group D) cells, which consists predominantly of deletions. ⁴To whom correspondence should be addressed at MGC-Department of Radiation Genetics and Chemical Mutagenesis     

DNA sequence analysis of in vivo hprt mutation in human T lymphocytes

We have determined the molecular basis of hypoxanthine-guaninephosphoribosyltransferase (hprt) mutations that arose in vivoin the T lymphocytes of a normal male subject. In previous studies16% (23/141) of the mutants from this individual analyzed bySouthern blot displayed large structural alterations in hprt.Thirty-two mutants without these large hprt structural alterationsproduced sufficient hprt cDNA for polymerase chain reactionamplification and DNA sequence analysis. Base substitutionsin hprt cDNA resulting in missense mutations and one mRNA splicingaberration (inclusion of intron sequences) were observed in18/32 of these these mutants; substitutions occurred at bothAT and GC base pairs. Small deletions (3/32), a tandem changeand a single base insertion were also observed among the hprtcDNAs. Exon skipping and inclusion of hprt intron sequencesin the hprt cDNA were observed in an additional 9/32 of themutants. Analysis of T cell receptor (TCR) gene rearrangementsrevealed that six of eight mutants with an identical hprt T—Atransversion displayed the same TCR rearrangement pattern, indicatingthat they were clonally related and arose from a single in vivomutational event. ³Present address: Department of Pathology, University of NorthCarolina Chapel Hill, NC 27514, USA ⁴To whom correspondence should be addressed      

The molecular nature of mutations induced by adriamycin at the hprt locus of V79 cells

Adriamycin (AM), a widely used chemotherapeutic drug, induceda broad spectrum of gene mutations at the hprt locus of V79cells. The frequency and distribution of AM-induced deletionswas analyzed with multiplex polymerase chain reaction in twoV79 cell lines, which differed considerably in their spontaneousdeletion frequency. Among AM-induced mutants, deletions predominatedin both cell lines. Apart from total deletions of the hprt gene,partial deletions were found which were distributed all overthe hprt gene with breakpoints in nearly all introns. Underthe same experimental conditions, chromosome aberrations wereinduced by AM which mainly represented chromatid-type aberrations.Neither the induction of gene mutations nor the induction ofchromosome aberrations was enhanced by the repair inhibitor3-aminobenzamide. These results are discussed in the contextwith our earlier findings on bleomycin-induced mutations andit is suggested that at least two mechanisms lead to the formationof gene deletions. One of them seems to be associated with amisrepair process of frank DNA double-strand breaks and relatedto chromosome aberrations while the other is not. ¹To whom correspondence should be addressed     

The repair of 4-nitroquinoline-1-oxide induced DNA adducts in hypersensitive Chinese hamster mutants: lack of repair of UV induced (6-4) photoproduct correlates with reduced repair of adducts at the N2 of guanosine.

UV sensitive Chinese hamster mutants belonging to ERCC groups 1, 2 and 6 together with one cross-link- and one X-ray-sensitive mutant have been examined for sensitivity to 4-nitroquinoline-1-oxide (4NQO) and the ability to repair 4NQO adducts at the N2 and C8 of guanosine. Despite the fact that all of the mutants examined were hyper-sensitive to 4NQO there was little difference between the mutants V-H1, V-H4, V-C4 and UV61 and the parental cell lines as regards the ability to remove these lesions from bulk DNA. The UV5 and UV20 mutants were both defective in the ability to remove N2 guanosine adducts yet repaired the C8 guanine adduct as normal. The fact that the mutants V-H1, V-H4, V-C4 and UV61 are 4NQO sensitive but repair the above adducts suggests that either some other lesion(s) is responsible for increased toxicity in these mutants, or that some regions of the genome may not be repaired as effectively as bulk DNA in these mutants, or that the quality of the repair is less than in the parental cells. Clearly the inability to remove UV induced pyrimidine dimers and the (6-4) photoproduct associated with the UV5 and UV20 mutants correlates with the inability to repair 4NQO-N2 guanosine adducts. However, mutants capable of (6-4) photoproduct repair but not dimer repair (VH-1 and UV61) can repair this lesion. Hence it is possible that the same domains in these repair proteins are required for the recognition of (6-4) photoproduct repair and 4NQO-N2 guanosine adducts.(ABSTRACT TRUNCATED AT 250 WORDS)     

AT base pairs are the main target for mutations at the hprt locus of rat skin fibroblasts exposed in vitro to the monofunctional alkylating agent N-ethyl-N-nitrosourea

Spectra of N-ethyl-N-nitrosourea (ENU)-induced mutations differwidely among various in vitro and in vivo mutational systems.To investigate possible reasons for these differences, a mutationalsystem is needed in which the same target gene is used for comparisonin the same type of cells in vitro and in vivo. In the presentstudy, this was achieved by analysing at the molecular level35 hprt mutant rat fibroblast clones obtained from cell populationsexposed in vitro to ENU and comparing the mutational spectrumwith the previously determined spectrum of ENU-induced hprtmutants in the same target cells exposed in vivo. Twenty-eightmutants contained a single base pair alteration in the hprtcoding sequence. Most of these changes were found at AT basepairs (19/28), the AT to TA transversion being the most frequentkind of mutation (12/19), which is probably caused by O²-ethylthymine.Transversions at AT base pairs showed all mutated T's to belocated in the nontranscribed strand of the hprt gene, suggestinga strand specific fixation of mutations induced by O²-ethylthymine,which appears to be a general feature of ENU- and ENNG-inducedhprt mutations in mammalian cells. GC to AT transitions, probablycaused by O⁶-ethylguanine, were detected at a lower frequency(7/28). This in vitro mutational spectrum was very similar tothat of the same target cells exposed in vivo to ENU. A comparisonof the mutational spectra in AGT-proficient and AGT-deficientrodent cells exposed to ethylating agents showed that in contrastto the situation in AGT-proficient rat fibroblasts, GC to ATbase pair changes (and not AT to TA) are the predominant mutationsin AGT-deficient hamster cells. ⁴To whom correspondence should be addressed     

Hypersensitivity to mutation and sister-chromatid-exchange induction in CHO cell mutants defective in incising DNA containing UV lesions

Five UV-sensitive mutant strains of CHO cells representing different genetic complementation groups were analyzed for their ability to perform the incision step of nucleotide excision repair after UV exposure. The assay utilized inhibitors of DNA synthesis to accumulate the short-lived strand breaks resulting from repair incisions. After 6 J/m², each of the mutants showed <10% of the incision rate of the parental AA8 cells. After 50 J/m², the rate in AA8 was similar to that at 6 J/m², but the rates in the mutants were significantly higher (20% of the rate of AA8). Thus by this incision assay the mutants were phenotypically indistinguishable. Each of the mutants were hypersensitive to mutation induction at both thehprt andaprt loci by a factor of 10, and in the one strain tested ouabain resistance was induced sevenfold more efficiently than in AA8 cells. Sister chromatid exchange was also induced with sevenfold increased efficiency in the two mutant strains examined. Thus, these CHO mutants resemble xeroderma pigmentosum cells in terms of their incision defects and their hypersensitivity to DNA damage by UV.     

Molecular nature of spontaneous mutations at the hypoxanthine-guanine phosphoribosyltransferase (hprt) locus in chinese hamster ovary cells

The hypoxanthine-guanine phosphoribosyltransferase (hprt) locus has been widely used as a selectable genetic marker for studies of mammalian cell mutagenesis. We report here the spontaneous mutation spectrum at the hprt locus in 64 independently isolated mutants of Chinese hamster ovary (CHO) cells. All nine hprt exons were simultaneously analyzed via multiplex polymerase chain reaction (PCR) for rapid detection of gene deletions or insertions. Structural point mutations were identified by direct sequence analysis of the PCR amplified cDNA. The molecular nature of RNA splicing errors and insertions was analyzed by solid-phase direct exon sequencing. Single base substitutions were found in 24 mutants (38%), of which 21 were missense and 3 were nonsense mutations. Transversions were about twice as frequent as transitions. Fifteen mutants (23%) had deletions involving either intragenic small fragments (2), single exons (9), or multiple exons (4). The majority of deletion breakpoints (71%) were located in regions surrounding exons 4, 5, and 6. RNA splicing mutations were observed in 15 mutants (23%) and affected exons 3–8; most (6/15) resulted in the loss of exon 7. Two insertion mutants, one with a 209 bp insert in exon 4 and the other with a 88 bp insert accompanied by a 24 bp deletion in exon 6, represent novel mutations reported for the first time in spontaneous mutants of the mammalian hprt gene. © 1995 Wiley-Liss, Inc.     

Strand-specific mutation induction by 1,2-dibromomethane at the hypoxanthine-guanine phosphoribosyltransferase locus of Chinese hamster ovary cells

The nature of mutations induced by 1,2-dibromoethane (DBE) atthe hprt (hypoxanthine-guanine phosphoribosyltransferase) genewas analysed in Chinese hamster ovary (CHO-9) cells. Molecularcharacterization of 36 hprt mutants at the cDNA level yielded19 GCAT transitions, two ATCG transversions, three frameshiftmutations, two identical small deletions and 10 exon deletions.Further analysis of the deletion mutants by amplification ofspecific exons from genomic DNA showed two more GCAT transitionsat splice sites and an {small tilde}70 bp deletion. Assumingthat the S-[2-(N7-guanyl)ethyl]glutathione adduct is responsiblefor the GCAT transitions, 90% of the affected guanines werelocated in the non-transcribed strand of the hprt gene, suggestinga strand bias in repair of this adduct Nearest neighbour analysisof induced GCAT transitions indicates a preference for a 5'-PyPuGDNA sequence, i.e. 15/21 mutated guanines were located in eithera TGG or a CAG DNA sequence. These molecular data on DBEinducedmutations showed similar features as data from a study by Graveset al (Mutagenesis, 11, 229–233, 1996) in which they analyzed13 hprt mutants induced by DBE in CHO-K1 cells. Six of the sevenGCAT mutations were on positions mutated more than once amongthe 36 hprt mutants in the present study. The combined findingssuggest that some positions seem to be hot spots for DBE-inducedmutations. Concerning the relevance of these in vitro studiesfor germ cell mutagenesis the conclusion may be that these datalend further support to the view that mutation spectra derivedfrom in vitro systems have little predictive value for the natureof mutations induced in post–meiotic germ cells in vivo,as demonstrated for other alkylating agents in both Drosophitaand mice. ¹To whom correspondence should be addressed. Tel: +31 071 5276145; Fax: +31 071 5221615; Email: nivard{at}rullf2.medfac.leidenuniv.nl     

Molecular analyses of in vivo hprt mutations in human T-lymphocytes I. Studies of low frequency 'spontaneous' mutants by Southern blots

Fifty wild-type and 164 in vivo-derived hprt mutant T-cell clonesobtained from eight non-mutagen-exposed adult males with mutantfrequency values in the normal range (usually < 10 x 10^–6)were studied by Southern blot analyses to determine frequencyand extent of gross structural alterations in the hprt gene.Sixteen (9.8%) of the mutant clones showed hprt changes. Nosite or type of lesion predominated. Relative frequencies ofgross structural alterations in the recovered hprt mutants didnot differ among the eight individuals, within limits detectableby the study. DNA from 201 of these 214 clones was also studiedwith a T-cell receptor (TCR) ß gene probe as a markerfor independence of in vivo-derived clones. Some clones werealso studied with a TCR gene probe. Ninety-four percent ofwild-type and 89% of the hprt mutants were found to originatefrom independent in vivo precursors. Therefore, most of therecovered hprt mutants in the study were presumably derivedfrom separate in vivo mutations. For non-mutagenized adultswith normal mutant frequencies, in vivo mutant frequencies arethus reasonable approximations of in vivo mutation frequencies,although elsewhere we show that this is not necessarily truefor individuals with grossly elevated mutant frequencies.     

Summary of complementation groups of UV-sensitive CHO cell mutants isolated by large-scale screening

A summary is given for the lineage and complementation groupassignments of 153 UV-sensitive mutants of the CHO AA8 cellline. The distribution of mutants among six complementationgroups was highly non-random, with the great majority of theisolates belonging to groups 1 and 2. This asymmetry is consistentwith the known hemizygosity of these two linked loci in CHOcells. The relative numbers of mutants induced in group 2 wasfound to depend greatly on the type of mutagen used. Mutagenesiswith UV radiation, ethyl methanesulfonate (EMS), N-methyl-N'-nitro-N-nitroso-guanidine and 7-bromomethylbenz[a]anthraceneproduced high frequencies of group 2 mutants. In contrast, ICR170and ICR191, which are thought to produce mostly frameshift mutations,yielded very few mutants in group 2. These results are of particularimportance in light of the recent finding that the human ERCC2gene, which corrects group 2 mutants, has very strong homologywith the yeast gene RAD3. RAD3 is an essential gene for viabilityin yeast, and the low recovery of group 2 mutants using theframeshift agents strongly suggests that frameshift mutationstend to be lethal in the hamster ERCC2 locus. Several mutagen-sensitivedouble mutants were isolated in two-step selections from EMS-,mitomycin C- or UV-sensitive parental cells, including the lineUVU1, the first mammalian line with two mutations that affectUV sensitivity. The first mutation inactivated excision repair,and the second mutation appears to have affected some otherrecovery process. UVU1 should be useful for studying recoveryprocesses that are separate from nucleotide excision repair. ¹To whom correspondence should be addressed     

The contribution of excision repair to the DNA effects seen in the alkaline single cell gel test (comet assay)

The alkaline single cell gel test (SCG test or comet assay)was used to study the contribution of excision repair activityto the observed DNA effect mutagen treatment. The cytotoxicityand genotoxicity of UV-irradiation and the chemical mutagens4-nitroquinoline-1-oxide (4NQO), benzo[a]pyrene (BP) and 7,12-dimethy-benz[a]anthracenentracene(DMBA) were compared in a normal human cell line (MRC5CV1) andan excision-deficient xeroderma pigment-osum (XP) cell line(XP12ROSV). The XP cells showed increased cell killing aftertreatment with all mutagens tested, but did not show a clearincrease in DNA migration in the comet assay. DNA effects inMRC5 cells were strongly enhanced by the repair inhibitor aphidicolin(APC), while under the same experimental conditions, APC hadon effect on the XP cell line. The enhancing effect of APC onDNA migration in MRC5 cells and the lack of effects in XP cellsindicate that the induced DNA effects of 4NQO, BP and DMBA inthe comet assay mainly represent the activity of an excisionrepair process. ¹To whom correspondence should be addressed     

Role of Human N-Acetyltransferase 2 Genetic Polymorphism on Aromatic Amine Carcinogen-Induced DNA Damage and Mutagenicity in a Chinese Hamster Ovary Cell Mutation Assay

Carcinogenic aromatic amines such as 4-aminobiphenyl (ABP) and 2-aminofluorene (AF) require metabolic activation to form electrophilic intermediates that mutate DNA leading to carcinogenesis. Bioactivation of these carcinogens includes N-hydroxylation catalyzed by CYP1A2 followed by O-acetylation catalyzed by arylamine N-acetyltransferase 2 (NAT2). To better understand the role of NAT2 genetic polymorphism in ABP- and AF-induced mutagenesis and DNA damage, nucleotide excision repair-deficient (UV5) Chinese hamster ovary (CHO) cells were stably transfected with human CYP1A2 and either NAT2*4 (rapid acetylator) or NAT2*5B (slow acetylator) alleles. ABP and AF both caused significantly (P < 0.001) greater mutagenesis measured at the hypoxanthine phosphoribosyl transferase (hprt) locus in the UV5/CYP1A2/NAT2*4 acetylator cell line compared to the UV5, UV5/CYP1A2, and UV5/CYP1A2/NAT2*5B cell lines. ABP- and AF-induced hprt mutant cDNAs were sequenced and over 80% of the single-base substitutions were at G:C base pairs. DNA damage also was quantified by γH2AX in-cell western assays and by identification and quantification of the two predominant DNA adducts, N-(deoxyguanosin-8-yl)-4-aminobiphenyl (dG-C8-ABP) and N-(deoxyguanosin-8-yl)-2-aminofluorene (dG-C8-AF) by liquid chromatography-mass spectrometry. DNA damage and adduct levels were dose-dependent, correlated highly with levels of hprt mutants, and were significantly (P < 0.0001) greater in the UV5/CYP1A2/NAT2*4 rapid acetylator cell line following treatment with ABP or AF as compared to all other cell lines. Our findings provide further clarity on the importance of O-acetylation in CHO mutagenesis assays for aromatic amines. They provide evidence that NAT2 genetic polymorphism modifies aromatic amine-induced DNA damage and mutagenesis that should be considered in human risk assessments following aromatic amine exposures. Environ. Mol. Mutagen. 61:235–245, 2020. © 2019 Wiley Periodicals, Inc.     

UV mutagenesis,cytotoxicity and split-dose recovery in a human—CHO cell hybrid having intermediate (6−4) photoproduct repair

Somatic cell hybrids constructed between UV-hypersensitive Chinese hamster ovary cell line UV20 and human lymphocytes were used to examine the influence of a human DNA repair gene, ERCCI, on UV photoproduct repair, mutability at several drug-resistance loci, UV cytotoxicity and UV split-dose recovery. In hybrid cell line 20HL21-4, which contains human chromosome 19, UV-induced mutagenesis at the APRT, HPRT and Na⁺/K⁺-ATPase loci was comparable to that in repair-proficient CHO AA8 cells, whereas cell line 20HL21-7, a reduced human-CHO hybrid not containing human chromosome 19, exhibited a hypermutable phenotype at all 3 loci indistinguishable from that of UV20 cells. The response of 20HL21-4 cells to UV cytotoxicity reflected substantial but incomplete restoration of wild-type UV cytotoxic response, whereas responses of UV20 and 20HL21-7 cell lines to UV cytotoxicity were essentially the same, reflecting several-fold UV hypersensitivity. Repair of UV-induced (5–6) cyclobutane dimers and (6−4) photoproducts was examined by radioimmunoassay; (6−4) photoproduct repair was deficient in UV20 and 20HL21-7 cell lines, and intermediate in 20HL21-4 cells relative to wild-type CHO AA8 cells. UV split-dose recovery in 20HL21-4 cells was also intermediate relative to AA8 cells. These results show that the human ERCCI gene on chromosome 19 is responsible for substantial restoration of UV survival and mutation responses in repair-deficient UV20 cells, but only partially restores (6−4) UV photoproduct repair and UV split-dose recovery.     

Establishment of lung fibroblastic cell lines from a non-human primate Tupaia belangeri and their use in a forward gene mutation assay at the hypoxanthing--guanine phosphoribosyl transferase locus

The cells obtained from a lung of a new-born male Tupaia belangeriwere maintained in mass culture for >400 days. After 55 populationdoubling levels (100 days in culture), three cell lines wereseparately established; these lines showed constant growth properties.One line, designated as T-23, was used for a mutation assay.The T-23 cells showed an absolute plating efficiency of 30–50%,and a population doubling time of 18–19 h in Dulbecco'smodified Eagle's medium containing 10% fetal bovine serum. Thecells had a modal chromosome number of 62 (pseudodiploid) withthe loss of a chromosome and the gain of an unidentified one.T-23 cells, like human cells, were much more susceptible toouabain than mouse cells but relatively less susceptible to8-azaguanine, while, unlike human cells, they were less sensitiveto 6-thioguanine (6TG). N-Methyl-N'-nitro-N-nitrosoguanidine(MNNG) was less, but 4-nitroquinoline-l-oxide (4NQO) was moretoxic to T-23 cells than to human or mouse cells. Benzo[a]pyrene-inducedtoxicity was almost comparable among the cell types. For themutation assay, we chose 6TG-resistance (100 µM) as amarker. The optimal expression time (8–13 days) and celldensity at selection to eliminate metabolic cooperation (2 x10⁴ cells/ 60-mm dish) were determined. Some of the cells selectedwith 6TG showed <0.4% of the total incorporation of [¹⁴C]hypox-anthineinto wild-type cells, suggesting the mutants under selectionwere affected at the hypoxanthine-guanine phos-phoribosyl transferaselocus. Following the optimal protocol, we performed a quantitativemutation assay with simple alky-lating agents, MNNG, N-ethyl-N'-nitro-N-nitrosoguanidine,methyl methanesulphonate, ethyl methanesulphonate and 4NQO.All induced mutants in a dose-dependent fashion. The abilityto induce mutants in T-23 cells by these mutagens was comparablewith that in a conventional V79 assay. This is the first reportof a gene mutation assay system using non-human primate cells. ¹To whom correspondence should be addressed     

Mutational spectrum of ICR-191 at the hprt locus in human lymphoblastoid cells

Human TK6 lymphoblasts were treated with the acridine derivative ICR-191, and mutants at the hprt locus were isolated. Mutant hprt cDNA was reverse-transcribed from mRNA, amplified by polymerase chain reaction (PCR), and sequenced. Additions of single G:C base pairs (+1 frameshift mutations) in repetitive G:C sequences were found in 82% (32/39) of the mutants. Sixteen of the +1 frameshifts analyzed were located in a single sequence of six consecutive guanine bases in exon 3. The remaining +1 frameshifts occurred at six different GGG sequences (14 mutants) and a single GGGG sequence (2 mutants) in other hprt exons. The repetitive guanine sequences that underwent frameshift mutagenesis were located in both the transcribed and nontranscribed strands of hprt. No single base deletions (-1 frameshift mutations) were observed. Base substitutions were observed in 13% (5/39) of the clones analyzed and occurred at both G:C and A:T bases. Loss of exon 4 from the cDNA was also observed in 5% (2/39) of the mutants. Hprt mutants containing seven consecutive guanines (produced from a +1 frameshift in a GGGGGG sequence) were treated with ICR-191 and wild-type revertants selected in CHAT medium. Revertants were recovered at a frequency of approximately 10^?7 and contained the wild-type sequence (GGGGGG) in all clones analyzed. The observed frequency of ICR-191-induced -1 frameshift reversion in the GGGGGGG sequence was ～500-fold lower than the estimated frequency of +1 frameshifts observed in the wild-type GGGGGG sequence following the same ICR-191 treatment. These results suggest that ICR-191 produces predominantly +1 frameshift mutations at the hprt locus in human cells. © 1994 Wiley-Liss, Inc.