A toolkit to visualise and interfere with ubiquitin-dependent DNA damage bypass

Abstract

Genome replication is a fundamental process for growth and proliferation. Therefore, timely resolution of replication problems protects daughter cells from genomic instability. In eukaryotes, DNA damage bypass safeguards cells against replication stress. This pathway is regulated by ubiquitylation of the replicative clamp, proliferation cell nuclear antigen (PCNA). Monoubiquitylation of PCNA activates error-prone translesion synthesis. Alternatively, attachment of K63-linked polyubiquitin chains on PCNA promotes error-free template switching. Whereas the specifics of translesion synthesis are well characterised, activation of template switching is still puzzling. In this project, I was interested to understand how cells choose one pathway over the other in response to genotoxic stress and discover yet unknown factors involved in polyubiquitin-dependent template switching. Our collaborator developed PCNA-specific reagents that bind to mono- and polyubiquitylated PCNA in vitro. These probes contain a PCNA-interacting peptide box and ubiquitin-binding domains (UBD) that selectively recognise ubiquitin modification with a relevant geometry. I adapted the probes for in vivo tools using various experimental strategies, e.g. microscopy, genomics, and genetic screening in the budding yeast Saccharomyces cerevisiae. In the first a pproach, I converted the probes i nto fluorescent se nsors to perform live-cell im aging. By using an inducible system in wild type and mutant yeast cells deficient in PCNA ubiquitylation, I showed that these sensors bind PCNA exclusively and act in a pathway-specific m anner. U pon replication stress, ubiquitylated PCNA forms distinct nuclear foci that emerge during early S phase and resolve later in the G2/M phase. Analysis of the localisation and appearance of ubiquitylated PCNA by tracing fluorescently tagged replisomes and repair factors revealed that PCNA ubiquitylation is activated mainly behind replication forks and close to postreplicative repair territories (PORTs). Unlike DNA doublestrand breaks and collapsed replication forks, damaged DNA marked by ubiquitylated PCNA does not reside at the nuclear pores or the nuclear periphery. Moreover, by studying the relative distribution of ubiquitylated PCNA between nuclei and nucleoli, I demonstrated that bypass events are less frequent in nucleoli. In the second experimental strategy, I used the novel PCNA sensors in a genomics approach to investigate the genome-wide association of ubiquitylated PCNA. Although the sensors enriched on chromatin in a DNA damage specific manner, their enrichment was independent of PCNA ubiquitylation. I showed that UBDs of the sensors were responsible for the unspecific binding to damaged chromatin. In the third and complementary approach, I utilised the probes to search for interactors of polyubiquitylated PCNA. Overexpression of the polyubiquitin-specific probe inhibited error-free template switching and rendered cells sensitive to DNA damage, likely by competing with factors that bind to polyubiquitinated PCNA. I expected that an overexpression genetic screen would identify factors that suppress the sensitising effects of the probe. Performing such a screen unveiled that Rad5’s ubiquitin ligase and helicase functions were necessary to suppress the probe.