Mechanisms employed by Escherichia coli to prevent ribonucleotide incorporation into genomic DNA by Pol V

Abstract

Escherichia coli pol V (UmuD'(2)C), the main translesion DNA polymerase, ensures continued nascent strand extension when the cellular replicase is blocked by unrepaired DNA lesions. Pol V is characterized by low sugar selectivity, which can be further reduced by a Y11A "steric-gate" substitution in UmuC that enables pol V to preferentially incorporate rNTPs over dNTPs in vitro. Despite efficient error-prone translesion synthesis catalyzed by UmuC_Y11A in vitro, strains expressing umuC_Y11A exhibit low UV mutability and UV resistance. Here, we show that these phenotypes result from the concomitant dual actions of Ribonuclease HII (RNase HII) initiating removal of rNMPs from the nascent DNA strand and nucleotide excision repair (NER) removing UV lesions from the parental strand. In the absence of either repair pathway, UV resistance and mutagenesis conferred by umuC_Y11A is significantly enhanced, suggesting that the combined actions of RNase HII and NER lead to double-strand breaks that result in reduced cell viability. We present evidence that the Y11A-specific UV phenotype is tempered by pol IV in vivo. At physiological ratios of the two polymerases, pol IV inhibits pol V-catalyzed translesion synthesis (TLS) past UV lesions and significantly reduces the number of Y11A-incorporated rNTPs by limiting the length of the pol V-dependent TLS tract generated during lesion bypass in vitro. In a recA730 lexA(Def) ΔumuDC ΔdinB strain, plasmid-encoded wild-type pol V promotes high levels of spontaneous mutagenesis. However, umuC_Y11A-dependent spontaneous mutagenesis is only ~7% of that observed with wild-type pol V, but increases to ~39% of wild-type levels in an isogenic ΔrnhB strain and ~72% of wild-type levels in a ΔrnhA ΔrnhB double mutant. Our observations suggest that errant ribonucleotides incorporated by pol V can be tolerated in the E. coli genome, but at the cost of higher levels of cellular mutagenesis.

Figure 1. Effect of ΔrnhA and ΔrnhB on UV survival of recA730 lexA(Def) ΔumuDC ΔdinB strains expressing pol V variants.

10 µl of 10-fold serial dilutions of overnight cultures were spotted onto the surface of rectangular LB agar plates and exposed to 40 J/m² 254 nM UV-light (panels A and C) and 20 J/m² 254 nM UV-light (panels B and D). Both unirradiated (−) and UV-irradiated (+) plates were incubated overnight at 37°C. In each panel, UV survival is shown for the recA730 lexA(Def) ΔumuDC ΔdinB strains either harboring pGB2 vector, or expressing pol V variants. The main observation of these experiments is that the UV-resistance of cells expressing umuC_Y11A increase dramatically in strains lacking rnhB, whereas survival of cells equipped with wild-type pol V, umuC_F10L, or umuC_Y11F is largely unaffected by the status of rnhB.

Figure 2. Quantitative UV survival and mutagenesis assays.

A: Survival. Exponentially growing cells were exposed to various doses of UV-light and serial dilutions spread on LB plates containing spectinomycin. The number of viable colonies was determined after overnight incubation at 37°C. Error bars indicate the standard error of the mean. Consistent with the semi-quantitative UV-survival assay shown in Figure 1, UV-resistance of strains expressing Y11A_UmuC increased significantly in the ΔrnhB background, while there was no change in UV-survival of the strain harboring vector, pGB2, or expressing wild-type pol V in the rnhB ^+/− strains. B: Mutagenesis. UV-induced mutagenesis was determined by exposing exponentially growing cells to 20 J/m² UV light. Cell viability was in the range of 85–90% survival for wild-type pol V and ∼60–70% for vector control, pGB2, and the Y11A mutant. The average number of His⁺ revertants per 10⁸ surviving cells ± standard error of the mean is indicated on the graph. The rnhB ⁺ strains are indicated by navy-colored bars, while ΔrnhB strains are indicated by the gold-colored bars. As observed, the UmuC_Y11A-expressing cells exhibited an ∼9-fold increase in UV mutagenesis compared to the rnhB ⁺ strain.

Figure 3. In vitro translesion synthesis past a TT-CPD lesion catalyzed by mixtures of pol IV and pol V.

Translesion DNA synthesis was performed using a circular DNA template with a running-start primer with its 3′ end located 5 bases before the 3′T of the CPD. Primer extension reactions catalyzed by pol IV (80 nM, lane 1 or 800 nM, lane 6), wild-type pol V (80 nM, lanes 2 and 7), pol V (UmuC_Y11A) (60 nM, lanes 4 and 9), or a combination of pol IV (80 nM, lane 3 and 5 or 800 nM, lane 8 and 10) with either wild-type (80 nM, lanes 3 and 8) or polV (UmuC_Y11A) (60 nM, lanes 5 and 10) were performed for 30 sec as described in the Methods section. Part of the template sequence and position of the gel wells and a CPD lesion are indicated to the right of the gel panel. As clearly observed, when present in a 10-fold excess (similar to SOS induced conditions), pol IV inhibits TLS catalyzed by pol V.

Figure 4. Role of NER in strains expressing pol V variants.

10 µl of 10-fold serial dilutions of overnight cultures were spotted onto the surface of rectangular LB agar plates and exposed to 40 J/m² 254 nM UV-light (Panel A), or 1 J/m² 254 nM UV-light (Panels B and C). Both unirradiated (−) and UV-irradiated (+) plates were incubated overnight at 37°C. In each panel, UV survival is shown for the recA730 lexA(Def) ΔumuDC ΔdinB strains either harboring pGB2 vector, or expressing pol V variants. Panel A (uvr ⁺ strain) is reproduced from Figure 1A for direct comparison to the isogenic ΔuvrA (Panel B) and ΔuvrC (panel C) strains. The main observation of these experiments is that while the uvr ⁻ strains are considerably more UV-sensitive than the isogenic uvr ⁺ strain, the relative sensitivity of the strains expressing pol V variants changes in the uvr ⁻ background, with UmuC_Y11A promoting an increase in UV-survival to a similar extent as wild-type pol V.

Figure 5. Effect of ΔrnhA and ΔrnhB on spontaneous mutagenesis in recA730 lexA(Def) ΔumuDC ΔdinB strains expressing pol V variants.

Spontaneous mutagenesis was measured by assaying reversion of the hisG4 ochre allele (leading to histidine prototophy) as described in Materials and Methods . The average number of His⁺ revertants per plate ± standard error of the mean is indicated in the table. Since the extent of mutagenesis promoted by wild-type pol V differed in the various strains, we have expressed the level of mutagenesis promoted by the variants as a percentage of wild-type mutagenesis. As clearly observed, umuC_Y11A-dependent mutagenesis increased in the ΔrnhB strain and was further elevated in the ΔrnhA ΔrnhB double mutant. In contrast, umuC_F10L gave consistently low levels of mutagenesis in all strains, and umuC_Y11F higher than wild-type levels in all strains.

Figure 6. Model for the effect of RNase HII and NER proteins on UV sensitivity of strains proficient for ribonucleotide incorporation.

Translesion replication catalyzed by Y11A mutant produces a TLS tract containing multiple ribonucleotides. NER excises UV-induced lesions and produces gaps on the template strand. RNase HII initiating removal of multiple rNMPs incorporated during TLS produces nicks on the daughter strand. The concerted action of both these repair pathways results in formation of persistent double strand breaks ultimately leading to cell death. Inactivation of either repair pathway selectively improves UV-resistance of cells expressing UmuC_Y11A.