FANCD2-dependent mitotic DNA synthesis relies on PCNA K164 ubiquitination.

Ubiquitination of proliferating cell nuclear antigen (PCNA) at lysine 164 (K164) activates DNA damage tolerance pathways. Currently, we lack a comprehensive understanding of how PCNA K164 ubiquitination promotes genome stability. To evaluate this, we generated stable cell lines expressing PCNAK164R from the endogenous PCNA locus. Our data reveal that the inability to ubiquitinate K164 causes perturbations in global DNA replication. Persistent replication stress generates under-replicated regions and is exacerbated by the DNA polymerase inhibitor aphidicolin. We show that these phenotypes are due, in part, to impaired Fanconi anemia group D2 protein (FANCD2)-dependent mitotic DNA synthesis (MiDAS) in PCNAK164R cells. FANCD2 mono-ubiquitination is significantly reduced in PCNAK164R mutants, leading to reduced chromatin association and foci formation, both prerequisites for FANCD2-dependent MiDAS. Furthermore, K164 ubiquitination coordinates direct PCNA/FANCD2 colocalization in mitotic nuclei. Here, we show that PCNA K164 ubiquitination maintains human genome stability by promoting FANCD2-dependent MiDAS to prevent the accumulation of under-replicated DNA.

Mitotic DNA Synthesis in Untransformed Human Cells Preserves Common Fragile Site Stability via a FANCD2-Driven Mechanism That Requires HELQ.

Faithful genome duplication is a challenging task for dividing mammalian cells, particularly under replication stress where timely resolution of late replication intermediates (LRIs) becomes crucial prior to cell division. In human cancer cells, mitotic DNA repair synthesis (MiDAS) is described as a final mechanism for the resolution of LRIs to avoid lethal chromosome mis-segregation. RAD52-driven MiDAS achieves this mission in part by generating gaps/breaks on metaphase chromosomes, which preferentially occur at common fragile sites (CFS). We previously demonstrated that a MiDAS mechanism also exists in untransformed and primary human cells, which is RAD52 independent but requires FANCD2. However, the properties of this form of MiDAS are not well understood. Here, we report that FANCD2-driven MiDAS in untransformed human cells: 1) requires a prerequisite step of FANCD2 mono-ubiquitination by a subset of Fanconi anemia (FA) proteins, 2) primarily acts to preserve CFS stability but not to prevent chromosome mis-segregation, and 3) depends on HELQ, which potentially functions at an early step. Hence, FANCD2-driven MiDAS in untransformed cells is built to protect CFS stability, whereas RAD52-driven MiDAS in cancer cells is likely adapted to prevent chromosome mis-segregation at the cost of CFS expression. Notably, we also identified a novel form of MiDAS, which surfaces to function when FANCD2 is absent in untransformed cells. Our findings substantiate the complex nature of MiDAS and a link between its deficiencies and the pathogenesis of FA, a human genetic disease.