ABSTRACT
Saccharomyces cerevisiae Mek1 is a CHK2/Rad53-family kinase that regulates meiotic recombination and progression upon its activation in response to DNA double-strand breaks (DSBs). The full catalog of direct Mek1 phosphorylation targets remains unknown. Here, we show that phosphorylation of histone H3 on threonine 11 (H3 T11ph) is induced by meiotic DSBs in S. cerevisiae and Schizosaccharomyces pombe. Molecular genetic experiments in S. cerevisiae confirmed that Mek1 is required for H3 T11ph and revealed that phosphorylation is rapidly reversed when Mek1 kinase is no longer active. Reconstituting histone phosphorylation in vitro with recombinant proteins demonstrated that Mek1 directly catalyzes H3 T11 phosphorylation. Mutating H3 T11 to nonphosphorylatable residues conferred no detectable defects in otherwise unperturbed meiosis, although the mutations modestly reduced spore viability in certain strains where Rad51 is used for strand exchange in place of Dmc1. H3 T11ph is therefore mostly dispensable for Mek1 function. However, H3 T11ph provides an excellent marker of ongoing Mek1 kinase activity in vivo. Anti-H3 T11ph chromatin immunoprecipitation followed by deep sequencing demonstrated that H3 T11ph was highly enriched at presumed sites of attachment of chromatin to chromosome axes, gave a more modest signal along chromatin loops, and was present at still lower levels immediately adjacent to DSB hotspots. These localization patterns closely tracked the distribution of Red1 and Hop1, axis proteins required for Mek1 activation. These findings provide insight into the spatial disposition of Mek1 kinase activity and the higher order organization of recombining meiotic chromosomes.
bioRxiv version 2 (June 2017) One major experimental change was incorporated into the revised manuscript: We repeated the anti-H3 T11ph ChIP-seq experiment on larger scale, including two meiotic time points from each of two wild type cultures and one time point from a spo11-Y135F mutant culture. To facilitate comparison of different samples, we used meiotic S. pombe cells as a spike-in control for all samples for both anti-H3 and anti-H3 T11ph ChIP-seq. Most conclusions described in the first bioRxiv submission were confirmed, but the improved datasets allowed us to derive more detailed information in particular about H3 T11ph patterns around DSB sites.
bioRxiv version 3 (October 2017) The following experimental changes were incorporated, along with more minor changes in response to reviewer comments:
We added previously unpublished ChIP-seq data for Red1 protein, generated by Masaru Ito and Kunihiro Ohta, who have been added as coauthors.
We repeated key experiments with the H3-T11V single point mutant. No conclusions were changed relative to prior experiments with the H3-S10, T11V mutant.
We repeated the analysis of spore viability in a dmc1 rad54-T132A background using a more appropriate isogenic control, and recapitulated the original conclusion that the H3-T11V mutation modestly decreases spore viability in this sensitized background.
