Epigenetic activation of meiotic recombination near Arabidopsis thaliana centromeres via loss of H3K9me2 and non-CG DNA methylation

Abstract

Eukaryotic centromeres contain the kinetochore, which connects chromosomes to the spindle allowing segregation. During meiosis, centromeres are suppressed for inter-homolog crossover, as recombination in these regions can cause chromosome missegregation and aneuploidy. Plant centromeres are surrounded by transposon-dense pericentromeric heterochromatin that is epigenetically silenced by histone 3 lysine 9 dimethylation (H3K9me2), and DNA methylation in CG and non-CG sequence contexts. However, the role of these chromatin modifications in control of meiotic recombination in the pericentromeres is not fully understood. Here, we show that disruption of Arabidopsis thaliana H3K9me2 and non-CG DNA methylation pathways, for example, via mutation of the H3K9 methyltransferase genes KYP/SUVH4 SUVH5 SUVH6, or the CHG DNA methyltransferase gene CMT3, increases meiotic recombination in proximity to the centromeres. Using immunocytological detection of MLH1 foci and genotyping by sequencing of recombinant plants, we observe that H3K9me2 and non-CG DNA methylation pathway mutants show increased pericentromeric crossovers. Increased pericentromeric recombination in H3K9me2/non-CG mutants occurs in hybrid and inbred backgrounds and likely involves contributions from both the interfering and noninterfering crossover repair pathways. We also show that meiotic DNA double-strand breaks (DSBs) increase in H3K9me2/non-CG mutants within the pericentromeres, via purification and sequencing of SPO11-1-oligonucleotides. Therefore, H3K9me2 and non-CG DNA methylation exert a repressive effect on both meiotic DSB and crossover formation in plant pericentromeric heterochromatin. Our results may account for selection of enhancer trap Dissociation (Ds) transposons into the CMT3 gene by recombination with proximal transposon launch-pads.

A pseudomolecule‐scale genome assembly of the liverwort Marchantia polymorpha

“…DNA methylation is known to be important in modulating recombination rates in the model plant A. thaliana , which is realized in a strong anticorrelation between recombination rates and DNA methylation (Yelina et al , ; Zhao et al , ; Choi et al , ; Dluzewska et al , ; Tock and Henderson, ; Underwood et al , ). In line with that, we found that recombination rate and the level of DNA methylation were significantly anticorrelated, both in M. polymorpha and P. patens .…”

“…Unequal distribution of epigenetic marks, including cytosine methylation, is assumed to primarily contribute to the large‐scale variability in recombination rates along the chromosomes of flowering plants (Underwood et al , ). In particular, cytosine methylation around the centromeric regions of A. thaliana is about four times higher than that in the middle of the chromosome arms (Underwood et al , ). Our analysis shows that this is not the case in M. polymorpha and P. patens (Lang et al , ).…”

Linked-read sequencing of gametes allows efficient genome-wide analysis of meiotic recombination

“…The distributions of the CO breakpoints in pollen and F 2 s were highly correlated in most regions of the genome (Pearson's r 0.80~0.86, both p-value<2.2e- 16), and exhibited only 13 regions with local differences across the entire genome ( Fig. 4; Materials and Methods).…”

“…Thus, the local CO rate governs the types of allelic combinations that can arise through sexual reproduction. Both environmental and genetic factors have been shown to affect CO rates and distributions; some examples are the local sequence divergence or rearrangements between the homologous chromosomes [12][13][14] , the chromatin context [15][16][17][18] , variation in DNA repair mechanisms [19][20][21][22][23][24][25][26] , and environmental stress 27,28,29 .…”

An ultra high-densityArabidopsis thalianacrossover map that refines the influences of structural variation and epigenetic features

“…The average read depth per library gave 1x to 2x genome-wide coverage per individual ( Figure S2 ). Of the 1,920 Col x Ler F 2 sequenced individuals, 95% had sufficient data quality and coverage to call COs. We added data from 363 previously analyzed Col x Ler F 2 individuals (Choi et al 2016;Underwood et al 2018) that were produced using another library preparation method (Rowan et al 2015(Rowan et al , 2017 . After processing and error correction for the entire data set (see Methods), we identified 17,077 COs in 2,182 individuals (File S2).…”