The following HTML text is provided to enhance online
readability. Many aspects of typography translate only awkwardly to HTML.
Please use the page image
as the authoritative form to ensure accuracy.
Colloquium on Links Between Recombination and Replication: Vital Roles of Recombination
might completely block the leading strand polymerase, but allow lagging strand synthesis to continue some way beyond the block (15, 17). Thus, a block on the leading strand template may be one type of event that leads to a stalled replication fork having a local structure similar to the fork used in Fig. 3A, lanes 5–8. It may be no coincidence that this fork is bound preferentially by RecG. Recent studies suggest that stalled replication forks can be processed by formation of a Holliday junction to create a substrate that can be branch migrated and cleaved by the RuvABC complex (5). The free DNA end generated by this cleavage [Fig. 1A (i)] can then be acted on by RecBCD and RecA to generate a D-loop. This D-loop can then act as a target for PriA-mediated assembly of a new replication fork at the D-loop (1). Alternatively, the free DNA end spooled out by regression of a stalled fork may be acted on by RecBCD directly to allow RecA-catalyzed D-loop formation, followed by subsequent cleavage of the Holliday junction by RuvABC (5, 13). However, replication can also be reestablished in a manner that requires not only PriA but also RecG, and that can proceed without the need for RuvABC-directed cleavage of the regressed fork (8). How can replication be restarted from a stalled fork without formation of a D-loop? PriA preferentially binds to forks with the 3′ end of a leading strand present at the branch point (32). PriA can also assemble a competent replication complex that can utilize this 3′ end for priming of replication (11). However, in the absence of the 3′-OH group of a leading strand at a stalled fork there would be no means to prime leading strand synthesis. The conclusion that RecG preferentially binds forks that possess a lagging strand, whereas PriA has a higher affinity for forks with a leading strand, suggests that RecG may facilitate PriA-dependent replisome reloading when the stalled fork does not initially possess a 3′-OH group at the junction point to prime leading strand synthesis. How this might be achieved is not known. However, it has been suggested that RecG may promote a template switching reaction in which formation of a Holliday junction by RecG allows extension of the stalled leading strand by using the nascent lagging strand as a template (ref. 8; Fig. 1). The ability of RecG to unwind the leading and lagging strands at fork structures, together with the high initial binding affinity of RecG for forks possessing a lagging strand, support this model. Branch migration of the Holliday junction in the reverse direction would regenerate a fork that now had a leading strand 3’OH for binding by PriA and subsequent priming of leading strand synthesis by DNA polymerase III. Thus, the opposing binding affinities of RecG and PriA at fork structures might reflect the ability of RecG to bind and unwind stalled forks that cannot be directly targeted by PriA to reload an active replisome. We are currently investigating whether such a mechanism underlies the observed genetic interaction between RecG and PriA (33).
We thank Lynda Harris for outstanding technical support and Gary Sharples for critical reading of the manuscript. This work was supported by a program grant from the Medical Research Council to R.G.L. and Gary Sharples. P.M. is a Lister Institute-Jenner Research Fellow.
1. Sandler, S.J. & Marians, K.J. (2000) J. Bacteriol.182, 9–13.
2. Lindahl, T. (1996) Philos. Trans. R. Soc. London B351, 1529–1538.