INVITRO EVIDENCE THAT UV-INDUCED FRAMESHIFT AND SUBSTITUTION MUTATIONS AT T-TRACTS ARE THE RESULT OF MISALIGNMENT-MEDIATED REPLICATION PAST A SPECIFIC THYMINE DIMER

A previous study of UV-induced (254 nm) mutations in the lacI gene of Escherichia coli found that frameshift mutations accounted for about 35% of the observed mutations and that these mutations occurred predominantly at A(n) . T(n) sequences [Miller, J. H. (1985) J. Mol. Biol. 182,48-65]. Because A(n) . T(n) sequences are hotspots for cis-syn thymine dimer formation [Brash, D. E., & Haseltine, W. A. (1982) Nature 298, 189-192], it would appear that UV-induced frameshift mutations are the result of an error during replicative bypass of a thymine dimer within such a sequence. To test the validity of such a proposal, replication experiments were carried out on templates containing cis-syn thymine dimers at each of the five possible sites of a T6 tract. The 59-mer templates were prepared by ligating oligonucleotides containing an EcoRI site to the 5'-end of decamers containing the cis-syn thymine dimer and oligonucleotides containing the primer site to the 3'-end. Primer-extension reactions were then carried out on these templates with a 3' --> 5' exonuclease-deficient (exo-) Klenow fragment of E. coli polymerase I and an exo- T7 polymerase (Sequenase Version 2.0). The replicative bypass products were cleaved with EcoRI to rigorously establish and quantify the presence of frameshift mutations. Both polymerases were able to bypass dimers at all sites, but only the exo- T7 polymerase led to detectable frameshifts, both -1 (approximately 30%) and -2 (approximately 5%), and only with the template containing a cyclobutane dimer at the second site from the 5'-end of the T6 tract. Sequencing of the T7 polymerase-catalyzed bypass products of all templates demonstrated that within the limits of discrimination only As were introduced opposite the dimer-containing T tracts. The only exception was for the template with the dimer at the second site which led to a readily detectable amount of a substitution mutation (approximately 30%) opposite the 5'-thymine of the T6 tract. A mechanism involving a competition between reversible misalignment and realignment steps and irreversible elongation steps is proposed to explain the origin of both the frameshift and the substitution mutations. The implications of this work to the mechanism of UV-induced frameshift and substitution mutations at T tracts in vivo are discussed.