Umeå University's logo

Recent cryo-EM structures of human DNA polymerase ε (Pol ε) bound to PCNA position a Pol ε-specific thumb insertion (polymerase thumb insertion; PTI) adjacent to a PCNA protomer, suggesting a regulatory role in DNA synthesis. To define the functional contribution of this region, we generated alanine-substitution variants in the yeast Pol ε thumb domain, targeting the PTI (SLED1131–1134→AAAA; polε-SLED) and an adjacent conserved loop (PVTE1101–1104→AAAA; polε-PVTE and KPFN1096–1099→AAAA; polε-KPFN). polε-SLED displayed increased intrinsic processivity, efficient bypass of DNA secondary structures, and enhanced synthesis on long templates, consistent with reduced pausing. In contrast, a previous study extended this substitution to six amino acids, SLEDLD1131-1136→AAAAAA, and found a reversed effect, a reduced processivity, indicating that subtle perturbations in this insertion can have opposing functional consequences. polε-PVTE shifted polymerase activity toward exonuclease proofreading and was not fully rescued by PCNA on long templates, whereas polε-KPFN retained near–wild-type activity but showed increased sensitivity to secondary structures that was alleviated by PCNA. In vivo, the corresponding pol2-SLED allele caused a modest mutator phenotype, while pol2-PVTE and pol2-KPFN showed little or no increase. Together, these results indicate that the PTI fine-tunes intrinsic processivity and proofreading to maintain replication fidelity during leading-strand synthesis.

DNA replication relies on precise coordination between proteins, including the sliding clamp proliferating cell nuclear antigen (PCNA), which encircles DNA to interact with key players in replication and repair. While biochemical studies have demonstrated interactions between PCNA and DNA polymerases δ and ε during DNA synthesis, the functional role of the Pol ε–PCNA interaction in vivo, particularly during leading strand synthesis, remains to be elucidated. To address this question, we employed AlphaFold to model how PCNA interact with four-subunit yeast Pol ε. Our models revealed two distinct points of interaction between Pol ε and PCNA: one at the P-domain and another at a PIP-box, a classical PCNA interaction motif. To validate these findings, we generated mutants that disrupted the Pol ε–PCNA interaction interface. Biochemical assays demonstrated that the PIP-box is critical for this interaction, with the P-domain serving as a secondary contact point. Notably, introducing these mutants into yeast, caused no phenotype in a wild-type background. However, when fewer origins are firing, resulting in longer stretches of leading strand synthesis before forks converge, strains expressing a Pol ε mutant lacking interaction with PCNA showed slower growth. These findings suggest that PCNA enhances the processivity of Pol ε both in vitro and in vivo.

Two recent studies provide structural insights into how human DNA polymerase ε (Pol ε) interacts with PCNA to form a processive holoenzyme on the leading strand. A series of cryo-EM images offer structural information on the proofreading process, showing how DNA is transferred between the polymerase and exonuclease sites in human Pol ε.