Site- and strand-specific mismatch repair of human H-ras genomic DNA in a mammalian cell line.

recognition complexes (1). Recent in vitro experiments haveDefective mismatch repair has recently been implicated indicated that at least seven different biochemical activitiesas the major contributor towards the mutator phenotype are required for mammalian mismatch repair (2). Mechanismsobserved in tumour cell lines derived from patients of ‘new’ strand discrimination prior to mismatch repair withindiagnosed with hereditary non-polyposis colon cancer mammalian cells have not yet been elucidated. However,(HNPCC). Cell lines from other cancer-prone syndromes, experiments using E.coli or human cell extracts have demon-such as xeroderma pigmentosum, have been found to be strated strand-specificity of mismatch repair by the creation ofdefective in nucleotide excision repair of damaged bases. a pre-existing nick on the same strand of DNA as the incorrectSome genetic complementation groups are defective speci- base, located either 59 or 39 from the mismatch (3–6).fically in transcription-coupled excision repair, although Within mammalian cells, there is more than one repairthis type of repair defect has not been associated with system for G:T or G:U mispairs putatively arising fromcancer proneness. Mechanisms contributing to the high endogenous 5-methylcytosine or cytosine deamination events,incidence of activating point mutations in oncogenes (such respectively (7,8). One well-described repair mechanism foras H-ras codon 12) are not understood. It is possible that the G:T mispair is a base excision repair process using anovel mechanisms of misrepair or misreplication occur at specific thymine glycosylase that does not require strand-these sites in addition to the above DNA repair mechanisms. discrimination (9). In addition, it has recently been reportedIn this study, we have compared the rate of strand-directed that the GTBP polypeptide of the hMutSαcomplex appearsmismatch repair of four mispairs (G:A, A:C, T:C and G:T) to bind preferentially with G:T mispairs (10). One moreat the H-ras codon 12, middle G:C position. Our results mismatch repair complex, hMutLα(composed of hMLH1 andindicate that, although this location is not a ‘hot spot’ for hPMS2) can restore strand-discriminatory repair activity forbacterial mismatch repair, it is a ‘hot spot’ for decreased all eight mismatches in nuclear extracts of a human tumourrepair of specific mismatched bases within NIH 3T3 cells. cell line (6). It is now evident that all combinations ofNIH 3T3, unlike Escherichia coli, have an extremely low mismatches can be repaired in mammalian cells; however, G:Trepair rate of the G:A mispair (35%), as well as the A:Cmispairs appear to be corrected with higher efficiency thanmispair (58%) at this location. NIH 3T3 also have aother mispairs, perhaps due to a wider repertoire of repairmoderately low repair rate of the T:C mispair (80%) atmechanisms (1,11).the codon 12 location. Conversely, NIH 3T3 repair ofThe majority of patients who have been diagnosed withG:T (100%) is comparable to E.coli repair (94%) of thishereditary non-polyposis colon cancer (HNPCC*) have germ-mismatch. These results demonstrate that a mismatchline mutations in one or more of the human mismatch repaircontaining an incorrect adenine on either strand at thehomologues. Studies using tumour cell lines developed fromH-ras codon 12 middle base pair location is most likely toundergo a mutational event in NIH 3T3 cells. Conversely, these patients have demonstrated a distinct mutator phenotypea mismatch containing an incorrect thymine in the tran- that appears to have a substantial increase in replication errors(RER