Break Induced Replication and Genome Rearrangments

The homology-dependent repair of DSBs usually occurs by a conservative gene conversion mechanism, preventing extensive loss of heterozygosity (LOH) or chromosome rearrangements. Breaks that present only one end for repair, for example at eroded uncapped telomeres or when homology is limited to one side of the DSB, are thought to repair by strand invasion into a homologous duplex DNA followed by replication to the chromosome end (break-induced replication, BIR). As BIR from one of the two ends of a DSB would result in extensive LOH it suggests BIR is suppressed when DSBs have two ends in order for repair to occur by a more conservative HR mechanism. Furthermore, our studies have shown that the replication intermediate formed during BIR is unstable and the invading end can switch to a different template resulting in a translocation. BIR and a related mechanism, fork stalling and template switching (FoSTeS), are thought to be responsible for many of the genome rearrangements that give rise to non-reciprocal translocations and copy number variation associated with human disease.

 We developed a plasmid transformation assay to study BIR. The chromosome fragmentation vector (CFV1) contains the URA3 selectable marker, SUP11, CEN4, an ARS, a tract of (G1-3T)n to provide a site for telomere addition and a unique DNA segment from the left arm of chromosome III (BUD3). The CFV is linearized between BUD3 and the telomere seeding sequences with a restriction endonuclease and used to transform yeast, selecting for Ura+ colonies. Most transformants (95%) arise by de novo telomere addition to heal one end of the CFV and strand invasion at the other end into the endogenous yeast locus to copy the entire chromosome arm yielding a stable 110 kb chromosome fragment (CF). The presence of the CF is verified by PFGE. We found BIR was completely dependent on RAD52, and reduced by 20-100 fold by mutation of RAD51, RAD54, RAD55 or RAD57. Thus, BIR require the same factors as gene conversion suggesting the initial strand invasion step is catalyzed by a common set of proteins. In contrast to gene conversion repair, BIR requires the non-essential subunit of the DNA polymerase delta complex, Pol32. Dependency of recombination events on Pol32 is frequently used to determine the contribution of BIR to repair.

To measure template switching during BIR we generated diploid strains with two different mutant alleles of ade2 or leu2 so that template switching could be detected by formation of Ade+ or Leu+ recombinants during BIR. In addition, template switching at repeated sequences generates CFs with aberrant mobility by pulsed-field gel electrophoresis. These methods are currently being used to identify genes that promote or inhibit template switching.