The two-step model of bacterial UV mutagenesis.

Recent results are discussed which have led to a two-step model for UV mutagenesis in excision-deficient Escherichia coli. After exposure to UV, the replication fork is assumed to continue until immediately before certain photoproducts where it stops and leaves a gap which cannot be dealt with by recombination repair. In the first (misincorporation) step, bases (a proportion of which are 'wrong') are postulated to be inserted opposite the photoproduct under the direct influence of the recA gene product. These misincorporated bases can be revealed as mutations by delayed photoreversal in umuD,C and lexA (ind-) bacteria. Their level is determined by the particular allele of recA that is present (recA441 greater than recA+ greater than recA430) and their rate of formation by the amount of recA protein in the cell and the degree of enrichment of the medium. No other protein needs to be synthesized for this step to occur. The second (bypass) step requires induced levels of the products of the umuD and C genes which are postulated to facilitate continued DNA synthesis on the priming end opposite the photoproduct. In principle, further errors could be made at this stage which might appear as 'hitch-hiking' rather than 'targeted' mutations.