Escherichia coli possesses three SOS-inducible DNA polymerases (Pol II, IV, and V) that were recently found to participate in translesion synthesis and mutagenesis. Involvement of these polymerases appears to depend on the nature of the lesion and its local sequence context, as illustrated by the bypass of a single N-2-acetylaminofluorene adduct within the NarI mutation hot spot. Indeed, error-free bypass requires Pol V (umuDC), whereas mutagenic (؊2 frameshift) bypass depends on Pol II (polB). In this paper, we show that purified DNA Pol II is able in vitro to generate the ؊2 frameshift bypass product observed in vivo at the NarI sites. Although the ⌬polB strain is completely defective in this mutation pathway, introduction of the polB gene on a low copy number plasmid restores the ؊2 frameshift pathway. In fact, modification of the relative copy number of polB versus umuDC genes results in a corresponding modification in the use of the frameshift versus error-free translesion pathways, suggesting a direct competition between Pol II and V for the bypass of the same lesion. Whether such a polymerase competition model for translesion synthesis will prove to be generally applicable remains to be confirmed. NarI mutation hot spot ͉ translesion synthesis ͉ slippage mutagenesis ͉ N-2-acetylaminofluorene ͉ umuDC (Pol V) P oint mutations are formed during DNA replication either as genuine replication errors or as a consequence of the presence of damage in the parental DNA. By changing the chemical structure of the bases, damaging agents are often effective blocks to the progression of replicative DNA polymerases. It has now become clear that these blocks are overcome by specialized enzymes (translesional DNA polymerases) that can read through damaged bases in a process known as translesion synthesis (TLS) (1-5). Because of the presence of the lesion, this process is inevitably less accurate than normal replication and will thus trigger increased mutation rates in the newly synthesized strand opposite the damaged base.Recently it was shown that in Escherichia coli, all three SOS-inducible DNA polymerases, Pol II (polB), Pol IV (dinB), and Pol V (umuDC), can be involved in TLS, depending on the nature of the DNA damage and its sequence context (6). We have identified an intriguing situation where the bypass of a given lesion, N-2-acetylaminofluorene (AAF), located within a frameshift mutation hot spot, the NarI site, is mediated by two genetically distinct pathways. Indeed, in that sequence context, the bypass of an AAF guanine adduct requires Pol V for nonslipped elongation, yielding error-free TLS but Pol II for slipped elongation, thus producing Ϫ2 frameshift TLS (6). The nonslipped TLS pathway obeys the rules previously described for base substitution mutagenesis induced by UV light or abasic sites, i.e., Pol V and RecA* dependence (for a recent review, see ref. 7). In contrast, the involvement of DNA Pol II in the slipped Ϫ2 frameshift pathway is more intriguing. Although a series of phenotypes related to DNA repair...