Decreased frequency and highly aberrant spectrum of ultraviolet-induced mutations in the hprt gene of mouse fibroblasts expressing antisense RNA to DNA polymerase zeta.

In the budding yeast Saccharomyces cerevisiae, DNA polymerase zeta (pol zeta) is responsible for the great majority of mutations generated during error-prone translesion replication of DNA that contains UV-induced lesions. The catalytic subunit of pol zeta is encoded by the Rev3 gene. The orthologue of Rev3 has been cloned from higher eukaryotic cells, including human, but its role in mutagenesis and carcinogenesis remains obscure. Investigation into the cellular function of pol zeta has been hindered by the fact that Rev3 knockout mice do not survive beyond midgestation, and embryonic stem cells used to derive these mice are not genetically stable. We have generated a transgenic mouse that expresses antisense RNA transcripts to mRev3 endogeneous RNA. These mice are viable, have greatly reduced levels of Rev3 transcript, and have reduced levels of B cells and impaired development of high-affinity memory B cells. Here, we report that exposure of fibroblasts derived from these mice to UV resulted in a 4-5-fold reduction in mutant frequency at the hprt locus at every dose examined, and the mutation spectrum was highly aberrant compared with the control cells. In the control cells, 80% of the mutations were transitions and approximately 75% of these arose from photoproducts in the putative leading strand template. Strikingly, in transgenic cells, most of the mutations were transversions and there was a complete loss of strand bias. This mutation spectrum is highly aberrant and is similar to that induced by UV in human xeroderma pigmentosum variant cells, which lack polymerase eta. These data indicate that most UV-induced mutations are dependent on DNA pol zeta, a function that has been conserved from yeast to higher eukaryotic cells. However, in mammalian cells, other DNA polymerase(s) may accomplish error-prone translesion replication and are responsible for residual UV mutagenesis observed in the absence of pol zeta. Further, these data support a central role for DNA polymerase eta in the error-free bypass of UV photoproducts. The antisense Rev3 mice should be a useful model to study mutagenic lesion bypass by pol zeta in mammalian cells and to investigate the role this polymerase plays in carcinogenesis.