Hierarchy and heterogeneity in the recruitment of <i>B. subtilis</i> replication-restart proteins to stalled forks

Abstract The restart of replication forks that have become stalled or disintegrated during the replication cycle is vital for all organisms, and in many bacterial species involves the conserved and essential DNA helicase PriA. PriA has been shown to physically interact with the C-terminus of SSB, which also binds to several other proteins involved in DNA repair and restart. It has been proposed that PriA is enriched at all replication forks in Bacillus subtilis via SSB interaction, such that it is instantly present to respond to a requirement for restart. Using single molecule tracking, we show that SSB and PriA are comprised of populations having very different diffusion constants, ruling out that PriA is co-migrating with fork-bound SSB. Indeed, PriA was only enriched at a subset of cells in exponentially growing cells, dependent on the C-terminus of SSB, but largely showed confined motion through the entire genome, searching for target sites in a transcription factor-like manner. Upon stalling of forks, SSB became highly enriched in all cells, suggesting a first line of response. PriA was also visibly enriched at forks following replication stress, in contrast to primosome proteins DnaD and DnaI, who showed only moderate changes in localization or in single molecule motion. PriA dwell times were affected by the lack of the SSB C-terminus, and also by the absence of RecG helicase, which is involved in recombination events. Heterogeneity of restart proteins at replication forks also extends to translesion DNA polymerases PolY1 and PolY2. Both proteins are low-abundant such that a considerable fraction of cells is devoid of any molecule. Our findings show that SSB accumulation is an initial response to replication stress, and that translesion synthesis and lesion skipping are less frequent events than fork remodelling.