Revealing a possible function for Chi Sites in avoiding genomic rearrangement

Chi sites are 8 bp sequences that are strongly over-represented in genomes since E. coli have ~ 1000 Chi sites, whereas in a random genome there would be ~ 150 copies of an 8 bp sequence.   It has long been known that the function of RecBCD changes when it encounters a Chi site sequence (c site), but it had been unclear why that change of function was advantageous.   Similarly, it was unclear why RecBCD would remove tens of thousands of bases that then have to be replaced by a DNA polymerase since such a process consumes energy and time.  Work by the Prentiss group provided a possible explanation for both of these previously mysterious phenomena.

Importantly, bacterial genomes contain many long-repeated sequences. Approximately ~ 5% of the E. coli genome consists of repeats longer than 20 bp, and ~ 1 % of the genome consists of repeats longer than 100 bp.  Genomic rearrangement can result if during repair of a double strand break in DNA RecA mediated homology recognition joins different copies of a repeated sequence.  ; however, the Prentiss group showed that Chi sites are preferentially located outside of repeated sequences.  As a result,  the positions of Chi sites in genomes greatly reduces the probability that RecA mediated homologous recombination will result in fatal genomic rearrangement.


Furthermore, in collaboration with the Godoy-Carter group the Prentiss group showed that homology stringency is enhanced if synthesis by a slow polymerase immediately follows RecA mediated homologous recombination.

Very recent work has studied how the length of ssDNA-RecA filaments might influence homologous recombination by allowing multiple homologous contacts between the ssDNA-RecA filament and the dsDNA being searched.