DNA replication and replication termination in Escherichia coli

Abstract

A prerequisite for successful cell division is the generation of an accurate copy of the entire genome as well as faithful segregation into the daughter cells. In the bacterium Escherichia coli, replication of the circular chromosome is initiated at a single origin (oriC) where two replication forks are assembled and proceed bi-directionally until they converge within a defined termination region opposite oriC and fork fusion takes place. This region is flanked by ter sequences, which, when bound by Tus protein, form a replication fork trap that allows forks to enter but not to leave. While the events associated with initiation as well as the elongation of replication have been extensively studied, the molecular details associated with the fusion of two replisomes are far less well characterised. The data presented here significantly extend our understanding of the molecular mechanics associated with the fusion of two replisomes in E. coli. My results strongly support the idea that RecG is a key player in processing intermediates that arise as two replication forks fuse. In the absence of RecG, over-replication of the chromosome is initiated at fork fusion intermediates, a process that can take place outside of the native termination area if forks are forced to fuse in an ectopic location. RecG has also been implicated in processing recombination intermediates. Over-replication in the absence of RecG is dependent on recombination and my data support the idea that RecG is important in limiting replication that initiates at recombination intermediates, some of which arise as a consequence of fork fusion events. In contrast, my data do not support the notion that the over-replication of chromosomal DNA in cells lacking RecG is in any way triggered by R-loops, as suggested previously in the literature. The observation that over-replication is taking place in cells lacking exonuclease I strongly supports the idea that 3’ single-stranded DNA structures are a key intermediate of fork fusion events. My data demonstrate that 3’ flaps accumulating in the absence of ExoI can be converted into 5’ flaps and degraded by 5’ exonucleases such as ExoVII and RecJ. RecG helicase is likely to be involved in this conversion. Thus, the data presented in this thesis highlight the complexity of replication fork fusion events and demonstrate that multiple protein activities are required to process fork fusion intermediates in order to allow DNA replication to be completed with high accuracy and enable faithful segregation of complete chromosomes into daughter cells.