Mcm10 deficiency causes defective-replisome-induced mutagenesis and a dependency on error-free postreplicative repair

Abstract

Mcm10 is a multifunctional replication factor with reported roles in origin activation, polymerase loading, and replication fork progression. The literature supporting these variable roles is controversial, and it has been debated whether Mcm10 has an active role in elongation. Here, we provide evidence that the mcm10-1 allele confers alterations in DNA synthesis that lead to defective-replisome-induced mutagenesis (DRIM). Specifically, we observed that mcm10-1 cells exhibited elevated levels of PCNA ubiquitination and activation of the translesion polymerase, pol-ζ. Whereas translesion synthesis had no measurable impact on viability, mcm10-1 mutants also engaged in error-free postreplicative repair (PRR), and this pathway promoted survival at semi-permissive conditions. Replication gaps in mcm10-1 were likely caused by elongation defects, as dbf4-1 mutants, which are compromised for origin activation did not display any hallmarks of replication stress. Furthermore, we demonstrate that deficiencies in priming, induced by a pol1-1 mutation, also resulted in DRIM, but not in error-free PRR. Similar to mcm10-1 mutants, DRIM did not rescue the replication defect in pol1-1 cells. Thus, it appears that DRIM is not proficient to fill replication gaps in pol1-1 and mcm10-1 mutants. Moreover, the ability to correctly prime nascent DNA may be a crucial prerequisite to initiate error-free PRR.

Keywords: DNA replication; DRIM; Mcm10; PCNA ubiquitination; Pol1; sumoylation; translesion synthesis.

Figure 1.pol1–1 and mcm10–1 mutations stimulate mono-ubiquitination of PCNA. (A and B) Cultures of mcm10–1, pol1–1, and the corresponding parental strains were grown to mid-log phase at 25 °C and shifted to varying temperatures as indicated for 3 h. Total protein was precipitated with TCA and fractionated by SDS-PAGE for western blot analysis with an anti-PCNA antibody. The asterisk indicates a PCNA form with a low molecular weight post-translational modification (or a non-specific band) visible in darker exposures. (C) mcm10–1 and wild-type parental strains containing no vector DNA (−), empty vector (pEV), or a wild-type MCM10 transgene expressed from the endogenous MCM10 promoter (pMCM10) were cultured to mid-log phase at 25 °C. Cultures were then split and shifted to 25 °C or 35 °C as indicated for 3 h before harvesting. Unmodified and ubiquitinated PCNA were monitored as mentioned above. (D) Ten-fold serial dilutions of strains from (C) were grown on synthetic complete medium lacking uracil for 3 d at the indicated temperatures.

Figure 2. Ubiquitination and sumoylation patterns of PCNA in pol1–1 and mcm10–1 mutants. (A and B) mcm10–1, pol1–1, and the corresponding parental strains expressing His₆-tagged PCNA were grown to mid-log phase at 25 °C and shifted to 35 °C for 3 h. Cultures were treated with MMS immediately prior to temperature shift where indicated. PCNA was purified under denaturing conditions and the eluates fractionated by SDS-PAGE for western blot analysis with anti-PCNA, anti-ubiquitin, and anti-SUMO antibodies as indicated. Ubiquitinated forms of PCNA are denoted as Ub₁ and Ub₂ for mono- and di-ubiquitin, respectively. SUMO attachment is indicated as S_K164 for K164 and S_K127 for K127. Poly-sumoylated species are represented by S_poly.

Figure 3. Elevated mutation rates in pol1–1 and mcm10–1 mutants are PCNA–K164-dependent. Bars indicate the mutation rates in pol1–1 (A), mcm10–1 (B), and the corresponding parental strains expressing either wild-type PCNA (POL30) or PCNA carrying a substitution in K164R (K164R). Each mutation rate represents the median of at least 12 independent measurements. Significance was determined by the Mann–Whitney U test and is indicated by an asterisk.

Figure 4. Elevated mutation rates in pol1–1 and mcm10–1 are dependent on Rev1 and Rev3. Bars indicate the mutation rates in pol1–1 (A), mcm10–1 (B), and the corresponding parental strains, carrying deletions of REV1, REV3, or RAD30 as indicated. Each mutation rate represents the median of at least 12 independent measurements. Significance was determined by the Mann–Whitney U test and is indicated by an asterisk.

Figure 5.pol1–1 and mcm10–1 mutations do not impact the normal TLS response to UV damage. Bars indicate the mutation frequencies in pol1-1 (A), mcm10–1 (B), and the corresponding parental strains in the presence or absence of UV light (10 J/m²) treatment. Each bar represents the median of at least 12 independent measurements. Significance was determined by the Mann–Whitney U test and is indicated by an asterisk.

Figure 6. Error-free PRR but not TLS suppresses the temperature sensitivity of mcm10–1. (A and B) Serial 10-fold dilutions of the indicated strains were grown on YPD plates for 2 d at the indicated temperatures. (C) Serial 10-fold dilutions of the indicated strains were grown on SC plates lacking uracil for 3 d at the indicated temperatures.

Figure 7. Origin activation defects do not trigger PCNA ubiquitination in asynchronous cultures. (A) Asynchronous cultures of wild-type, mcm10–1, and dbf4–1 cells were grown to mid-log phase and then shifted to either 25 °C or 35 °C for 3 h in the presence or absence of 0.02% MMS. Cells were then harvested and analyzed for DNA content. (B) Protein extracts were prepared from cells treated as described in (A). Modified and unmodified forms of PCNA were detected using an anti-PCNA antibody. (C) Serial 10-fold dilutions of the indicated strains were grown on YPD plates for 2 d at various temperatures as marked. (D) Wild-type, mcm10–1, and dbf4–1 cells were harvested from the cultures analyzed in (A), and protein extracts were prepared. Modified and unmodified forms of Rad53 were detected with an anti-Rad53 antibody. Phosphorylated histone H2A was detected using a phospho-S129 specific anti-histone H2A antibody. Tubulin served as a loading control.

Figure 8. Primer quality affects PRR pathway choice. The cartoon depicts the inherent differences between pol1–1 and mcm10–1 mutants in PRR pathway dynamics. (Left) In pol1–1, intrinsically inaccurate primer synthesis by the Pol1–1 enzyme leads to mismatches unsuitable for pol-δ extension. These primers are likely extended, at least initially, by pol-ζ, which is efficient in adding to terminal mismatches. It is possible that mismatched primers are also unsuitable to initiate template switching, explaining why the Rad5-dependent error-free pathway is not efficiently utilized in these mutants. (Right) Primers in mcm10–1 are synthesized by a wild-type Pol1 enzyme, allowing for engagement in template switching, which promotes cell survival. The dashed line denotes the actual switch from the lagging strand template to the nascent leading strand. *Although the translesion polymerase pol-ζ is activated in both pol1–1 and mcm10–1 mutants, it does not impact survival of either strain.