The conserved LEM-3/Ankle1 nuclease is involved in the combinatorial regulation of meiotic recombination repair and chromosome segregation in Caenorhabditis elegans

Abstract

Homologous recombination is essential for crossover (CO) formation and accurate chromosome segregation during meiosis. It is of considerable importance to work out how recombination intermediates are processed, leading to CO and non-crossover (NCO) outcome. Genetic analysis in budding yeast and Caenorhabditis elegans indicates that the processing of meiotic recombination intermediates involves a combination of nucleases and DNA repair enzymes. We previously reported that in C. elegans meiotic joint molecule resolution is mediated by two redundant pathways, conferred by the SLX-1 and MUS-81 nucleases, and by the HIM-6 Bloom helicase in conjunction with the XPF-1 endonuclease, respectively. Both pathways require the scaffold protein SLX-4. However, in the absence of all these enzymes, residual processing of meiotic recombination intermediates still occurs and CO formation is reduced but not abolished. Here we show that the LEM-3 nuclease, mutation of which by itself does not have an overt meiotic phenotype, genetically interacts with slx-1 and mus-81 mutants, the respective double mutants displaying 100% embryonic lethality. The combined loss of LEM-3 and MUS-81 leads to altered processing of recombination intermediates, a delayed disassembly of foci associated with CO designated sites, and the formation of univalents linked by SPO-11 dependent chromatin bridges (dissociated bivalents). However, LEM-3 foci do not colocalize with ZHP-3, a marker that congresses into CO designated sites. In addition, neither CO frequency nor distribution is altered in lem-3 single mutants or in combination with mus-81 or slx-4 mutations. Finally, we found persistent chromatin bridges during meiotic divisions in lem-3; slx-4 double mutants. Supported by the localization of LEM-3 between dividing meiotic nuclei, this data suggest that LEM-3 is able to process erroneous recombination intermediates that persist into the second meiotic division.

Fig 1. Genetic interaction between LEM-3, MUS-81, SLX-1 and SLX-4 endonucleases.

Embryonic lethality in % was determined by counting number of dead eggs/total number of eggs laid. Error bars represent standard deviation of the mean. Sample sizes of indicated genotype are as follows: wild type n = 328, lem-3 (tm3468) n = 527, lem-3 (mn155) n = 357, xpf-1 n = 269, lem-3 (mn155); xpf-1 n = 264, slx-1 n = 143, slx-1; lem-3 (mn155) n = 210, mus-81 n = 181, mus-81 lem-3 (mn155) n = 250, slx-4 n = 377, lem-3 (tm3468); slx-4 n = 407, lem-3 (mn155); slx-4 n = 233.

Fig 2. Mutation of lem-3 in mus-81 and slx-4 mutants causes increased apoptosis.

(A) Quantification of apoptotic cells per gonad in the indicated genotypes. The apoptotic cells were measured using DIC microscopy. N = 5 gonad arms for each genotype. Error bars represent standard deviation of the mean. (B) Representative images of DAPI-stained germline in wild type, lem-3, mus-81, slx-4, mus-81 lem-3, lem-3; slx-4 and mus-81 lem-3; spo-11 mutants. Red arrowheads indicate pyknotic cells with abnormally condensed nuclei in the pachytene stage. Scale bars: 15 μm.

Fig 3. Comparison of RAD-51 and RMH-1 foci localization in wild type, lem-3, mus-81 single mutants and mus-81 lem-3 double mutant.

(A) Quantification of RAD-51 profiles over the course of meiotic prophase. C. elegans gonads were divided into seven equal zones. We determined the number of RAD-51 foci in each zone. Quantifications were done based on three representative gonads for each genotype. Error bars are standard error of the mean. (B) Quantification of RMH-1::GFP foci in wild type, lem-3, mus-81, xpf-1,mus-81 lem-3 and mus-81; xpf-1 mutants in early Pachytene, mid Pachytene and late Pachytene stages. Quantifications were done for three gonads per genotype. Asterisks indicate statistical significance as determined by student T test. P Values below 0.05 were considered as significant, p < 0.05 is indicated with *, p < 0.01 with **, p < 0.005 with *** and p < 0.0001 with ****. (C) Representative images of gonads stained with DAPI. Yellow numbers represent those nuclei with RMH-1::GFP foci between 1 and 10, pink numbers represent nuclei with RMH-1::GFP counts between 11 and 25, and white numbers represent nuclei with RMH-1::GFP foci above 25. Scale bars: 5 μm. (D) Representative close-up images of mid and late Pachytene nuclei with different numbers of GFP::RMH-1 foci from a wild type gonad arm.

Fig 4. Delayed removal of COSA-1 foci in mus-81 lem-3 double mutants.

(A) Crossover designation is normal in mus-81 lem-3 and lem-3; slx-4 double mutants. DAPI staining of representative pachytene nuclei containing GFP::COSA-1 foci. Scale bars: 2 μm. (B) Quantification of nuclei with indicated number of COSA-1 foci. (C) Projections of representative nuclei from diakinesis oocytes of wild type, lem-3, mus-81, slx-4, mus-81 lem-3 and lem-3; slx-4 mutants stained with α-HTP-3, a component of the C. elegans axial element (red) and DAPI (blue). The persistent GFP::COSA-1 foci localized between two homologous pairs are highlighted by white arrowheads. (D) Scheme depicting normal bivalent from wild type and dissociated bivalent with COSA-1 focus at junction from slx-4 mutant at the diakinesis stage. (E) Quantification of the diakinesisi nuclei with indicated number of COSA-1 foci. The sample size (n) indicates the number of germline examined for each genotype.

Fig 5. Depletion of LEM-3 and MUS-81 leads to formation of dissociated bivalents.

(A) Images of DAPI-stained chromosomes in –1 oocytes at diakinesis in wild type, lem-3, mus-81 and mus-81 lem-3 mutants. Red arrows indicate dissociated bivalents. Chromosome fragment is highlighted with a red arrowhead. Scale bars: 2 μm. (B) Quantification of bivalents, ‘dissociated bivalents’ and fragments observed in indicated genotypes. Overlapping chromosomes that could not be assigned to the above categories were scored as “n/d”. Sample sizes of indicated genotype are as follows: wild type n = 40; lem-3 n = 36; mus-81 n = 36; mus-81 lem-3 n = 42.

Fig 6. Univalent formation in rmh-1 mus-81 lem-3 triple mutants can be reduced by depletion of brd-1.

(A) Quantification of univalents in -1 oocytes of wild type, rmh-1 and rmh-1 mus-81 lem-3 mutants. The number of univalents in rmh-1 and rmh-1 mus-81 lem-3 mutants was counted. Asterisks indicate statistical significance as determined by Z-score. P Values below 0.05 were considered as significant, p < 0.05 is indicated with *, p < 0.01 with **, p < 0.005 with *** and p < 0.0001 with ****. (B) Representative image of a diakinesis nucleus of rmh-1 mus-81 lem-3. The majority of univalents classify as “peculiar univalents” (indicated by white arrow) stained by LAB-1 (green) and phsopho-histone H3 (pH3, red). Scale bars: 5 μm. (C) Quantification of DAPI-stained bodies in –1 oocytes at diakinesis in the indicated genotypes. The sample size (n) is indicated as follows: wild type (n = 23), rmh-1 (n = 88), mus-81 (n = 32), lem-3 (n = 26), brd-1 (n = 30), mus-81 lem-3 (n = 23), rmh-1 mus-81 (n = 46), rmh-1 lem-3 (n = 40), rmh-1 mus-81 lem-3 (n = 73), rmh-1 mus-81 lem-3; brd-1 (n = 50). (D) Quantification of RAD-51 profiles over the course of meiotic prophase. C. elegans gonads were divided into seven equal zones and RAD-51 foci were counted in each nucleus of each zone. Quantifications were done based on three representative gonads per genotype. Error bars are standard error of the mean.

Fig 7. Localization of LEM-3 and its role in meiotic chromosome segregation.

Localization of GFP::LEM-3 in mitotic zone (A) and pachytene stage (B). (C) LEM-3 foci (green) do not colocalize with crossover precursor marker ZHP-3 (red). (D) Colocalization of LEM-3 with AIR-2 between dividing nuclei during meiosis II. We note that the female pronucleus is already partially decondensed. (E) Schematic depiction of LEM-3 localization during meiotic division. (F) Representative images taken from time-lapse recordings of mCherry::Histone H2B expressing embryos during meiotic division. Red arrowheads indicate the chromatin linkages. (G) Quantification of DNA linkage formation during meiotic division in the indicated genotypes.