Meiotic DNA exchanges in C. elegans are promoted by proximity to the synaptonemal complex

Abstract

During meiosis, programmed double-strand DNA breaks are repaired to form exchanges between the parental chromosomes called crossovers. Chromosomes lacking a crossover fail to segregate accurately into the gametes, leading to aneuploidy. In addition to engaging the homolog, crossover formation requires the promotion of exchanges, rather than non-exchanges, as repair products. However, the mechanism underlying this meiosis-specific preference is not fully understood. Here, we study the regulation of meiotic sister chromatid exchanges in Caenorhabditis elegans by direct visualization. We find that a conserved chromosomal interface that promotes exchanges between the parental chromosomes, the synaptonemal complex, can also promote exchanges between the sister chromatids. In both cases, exchanges depend on the recruitment of the same set of pro-exchange factors to repair sites. Surprisingly, although the synaptonemal complex usually assembles between the two DNA molecules undergoing an exchange, its activity does not rely on a specific chromosome conformation. This suggests that the synaptonemal complex regulates exchanges-both crossovers and sister exchanges-by establishing a nuclear domain conducive to nearby recruitment of exchange-promoting factors.

Figure 1.. The SC-CR promotes sister exchanges when it localizes to unpaired chromosomes.

(A) Exchanges are elevated on unpaired chromosomes in ieDf2 animals, but not in him-8 and zim-2 animals. (Data for WT animals are not shown because all chromosomes are paired.) Comparison between him-8 and zim-2 was not significant (Pearson’s chi-square). Comparison between him-8 and ieDf2 animals was significant (P = 0.001, Pearson’s chi-square). him-8 data are taken from Almanzar et al (2021). Diagrams of the different genotypes are shown to the right. Chromosomes are shown in blue, the axes in maroon, and the SC-CR in green. Unpaired chromosomes are shown with an orange background. Note that the unpaired chromosomes in him-8 and zim-2 animals (the X chromosome and chromosome V, respectively) are not associated with the SC-CR, whereas all chromosomes are unpaired and associated with the SC-CR in ieDf2 animals. (B) Representative images and interpretive diagrams of exchange and non-exchange chromosomes in ieDf2 and zim-2 animals. Yellow arrows denote exchange chromatids, white arrows denote non-exchange chromatids, and asterisks denote unlabeled chromatids (not scored). Interpretive diagrams of chromosomes surrounded by dashed white boxes are shown to the right. Red, axis (anti-HTP-3 antibodies); green, EdU; and blue, DNA (DAPI). Note the EdU signal crossing the axis in exchange chromatids. Scale bars = 1 µm.

Figure S1.. RAD-51 foci are more numerous in zim-2 compared with him-8 or WT animals.

(A) Representative images of mid-pachytene nuclei. Red, axis (anti-HTP-3 antibodies); and yellow, RAD-51. The clusters of RAD-51 foci in him-8 and zim-2 nuclei are likely on the unpaired chromosomes (X and V, respectively [MacQueen et al, 2005]). Scale bars = 1 µm. (B) Average number of RAD-51 foci per nucleus in mid-pachytene. zim-2 exhibited significantly higher numbers of foci compared with WT and him-8 animals (P < 0.0001, t test), consistent with a higher number of DSBs. This number is likely an underestimate of the actual number of DSBs in zim-2 animals because the resolution of confocal microscopy does not allow the resolution of all the individual foci within the large patches of RAD-51.

Figure 2.. Sister exchanges correspond to recombination nodules.

(A) Average number of GFP-COSA-1 foci in late-pachytene nuclei of animals of the designated genotype. (“+” indicates animals that have no genetic alterations except for the COSA-1-GFP transgene, which serves as a WT control). Results were significantly different from WT for all genotypes (P < 0.000001, t test). n value indicates the number of nuclei counted. Data for syp-2 are from Cahoon et al (2019); those for syp-1^K42E are from Gordon et al (2021); and those for syp-3(me42) are from Almanzar et al (2021). Light gray shading indicates conditions where only zero or two chromosomes are unpaired; dark gray shading indicates all chromosomes are unpaired. (B) Total number of sister exchanges extrapolated from the exchange number quantified in Figs 1A and 2D and Almanzar et al (2021). Shading is as in panel (A). Data for syp-3(me42) are from Almanzar et al (2021). (A, B) Note correspondence between the values in panels (A, B) for conditions where all chromosomes are unpaired. (C) Representative images of late-pachytene nuclei in WT, him-8, ieDf2, and syp-1 animals that also carry the GFP-COSA-1 transgene. Red, axis (anti-HTP-3 antibodies); green, GFP-COSA-1 (anti-GFP antibodies); and blue, DNA (DAPI). Scale bars = 1 µm. Diagrams of the different genotypes are shown to the right. Chromosomes are shown in blue, the axes in maroon, and the SC-CR in green. Unpaired chromosomes are shown with an orange background. Note that although all chromosomes are unpaired in both scenarios, they are associated with the SC-CR only in syp-1^K42E animals. (D) Representative images of exchange and non-exchange chromatids in syp-1 animals. Yellow arrows denote exchange chromatids, white arrows denote non-exchange chromatids, and asterisks denote unlabeled chromatids (not scored). Interpretive diagrams of chromosomes surrounded by dashed white boxes are shown to the right. Red, axis (anti-HTP-3 antibodies); green, EdU; and blue, DNA (DAPI). Scale bar = 1 µm. (E) Sister exchanges are elevated upon complete removal of the SC-CR. Pairwise comparisons between him-8 and syp-1 or syp-2 animals were significant (P = 0.002 and P = 0.008, respectively, Pearson’s chi-square test).

Figure 3.. Most sister exchanges in syp-1^K42E animals depend on ZHP-3.

(A) Representative pachytene nuclei in syp-1^K42E zhp-3-FLAG-AID with and without auxin (right and left, respectively). Note the absence of recombination nodules—foci of ZHP-3—when grown on auxin. Red, axis (anti-HTP-3 antibodies); green, ZHP-3 (anti-FLAG antibodies); and blue, DNA (DAPI). Scale bars = 1 µm. (B) Most sister exchanges in the syp-1^K42E background are dependent on ZHP-3. Sister exchanges in syp-1^K42E animals grown with or without auxin are significantly different (P < 0.001, Pearson’s chi-square test). Data for zhp-3(+) animals include data from Almanzar et al (2021). (C) Representative images of exchange and non-exchange chromosomes in syp-1^K42E zhp-3-FLAG-AID with and without auxin. Yellow arrows denote exchange chromatids, white arrows denote non-exchange chromatids, and asterisks denote unlabeled chromatids (not scored). Interpretive diagrams of chromosomes surrounded by dashed white boxes are shown to the right. Red, axis (anti-HTP-3 antibodies); green, EdU; and blue, DNA (DAPI). Scale bars = 1 µm.

Figure 4.. Sister exchanges and crossovers are regulated together.

(A) Combined immunofluorescence and FISH of pachytene nuclei from GFP-COSA-1 (control) and nT1/+; GFP-COSA-1 animals. The 5S locus is located on chromosome V within the homologously-paired portion in nT1/+. Gray, SC-CR (SYP-2); green, crossovers (GFP-COSA-1); and magenta, 5S locus. The crossover on chromosome V (judged by tracing the 5S-containing SC-CR signal) is marked with an arrowhead. Chromosomes IV and V are shown in two shades of blue in the diagrams to the right. Scale bars = 5 µm. (B) Distance along chromosome V between the 5S locus and the crossover (P < 0.0001, t test). (C) Representative partial projections of non-exchange chromosomes in nT1/+ animals and an interpretive cartoon. Arrowheads indicate the location of the crossover. Red, axis (anti-HTP-3 antibodies); green, EdU; and blue, DNA (DAPI). Scale bars = 1 µm. (D) Sister exchanges remain rare when a majority of chromosomes IV and V can only undergo sister exchanges. Pairwise comparisons of exchanges on paired chromosomes between WT, him-8, zim-2, and nT1/+ animals were not significant (P > 0.05 for all comparisons, Pearson’s chi-square).

Figure S2.. nT1/+ animals harbor one GFP-COSA-1 focus per chromosome.

(A) Representative images of late-pachytene nuclei in nT1/+ animals that also carry the GFP-COSA-1 transgene. Interpretive diagrams of chromosomes surrounded by dashed white boxes are shown to the right. Red, axis (anti-HTP-3 antibodies); green, GFP-COSA-1 (anti-GFP antibodies); and blue, DNA (DAPI). Scale bar = 1 µm. (B) Quantification of GFP-COSA-1 foci in WT and nT1/+ animals. The difference between the two genotypes was not significant (P = 0.21, t test). N value indicates the number of nuclei counted.

Figure 5.. SC-CR in various conformations can promote sister exchanges.

(A) Representative STED images of pachytene nuclei from WT, ieDf2, rec-8, and syp-1^K42E worms. Red, SYP-5; and green, HTP-3. Scale bars = 1 µm. HTP-3 fluorescence is normalized such that maximum fluorescence is 1. Dashed white arrow denotes line scan in the merged WT image, with the normalized fluorescence plot shown above. The SC-CR localizes between the axes in WT, ieDf2, and rec-8 worms, but the SC-CR and axis signals colocalize in syp-1^K42E worms. Overlaid line scans show similar distribution in WT (n = 12) and ieDf2 (n = 27) worms—two axis peaks separated by 150–160 nm. Rec-8 animals (n = 27) exhibit a somewhat smaller inter-axis distance—120 nm—whereas syp-1^K42E worms (n = 18) show a single peak. Averages are shown in green. Diagrams of chromosome architecture are shown to the right, with the two homologs shown in two shades of blue. (B) Representative confocal images of pachytene nuclei from WT and ieDf2 animals. Gray, SC-CR (SYP-2); and magenta, 5S locus. Diagrams of chromosome architecture are shown to the right, with the homologous chromosome V shown in two shades of blue. Because the homologs are not paired in ieDf2, each nucleus harbors two 5S foci. (C) In WT animals, foci span the SC-CR, indicating the four sisters (two from each homolog) are aligned. In contrast, most foci in ieDf2 animals are to one side of the SC-CR (P > 0.00001, Pearson’s chi-square).

Figure 6.. Recombination nodules do not form exclusively in the middle of the SC-CR.

(A) STED images of nuclei from the WT and ieDf2 worms immunolabeled for GFP-COSA-1 (magenta) and HTP-3 (green). Right, intensity distribution extracted from line profiles of merged images (insets). Black lines denote fits to either one (GFP-COSA-1) or a sum of two (HTP-3) Gaussian distribution plus constant background. Scale bars = 1 µm. (B) Histograms of GFP-COSA-1 focus center positions for WT (n = 94) and ieDf2 (n = 112). Values of 1 or −1 indicate localization on the meiotic axes, whereas a value of 0 indicates localization at the center of the SC-CR. Magnitudes >1 or <−1 (lighter bars) indicate GFP-COSA-1 center position outside the axes. Blue curves represent a Gaussian fit to the data, whereas the gray distribution represents estimated localization error (Fig S3); for clarity, curves are displayed with the same maximum value as each histogram. Bottom, schematic diagram of synaptonemal complex geometry, with axes marked in green and the GFP-COSA-1 focus in magenta.

Figure S3.. Localizing SYP-2 and GFP-MSH-5 relative to the axes.

(A) STED images of a nucleus from WT worms immunolabeled for SYP-2 (red) and HTP-3 (green). Middle, intensity distribution extracted from line profiles of merged image (insets). Black lines denote fits to either one (SYP-2) or a sum of two (HTP-3) Gaussian distribution plus constant background. Scale bars = 1 µm. Right, histogram of SYP-2 focus center positions (n = 10). Values of 1 or −1 indicate localization on the meiotic axes, whereas a value of 0 indicates localization at the center of the SC-CR. Blue curve represents a Gaussian fit to the data. Bottom, schematic diagram of synaptonemal complex geometry, with axes marked in green and the antibodies targeting SYP-2 in red. (B) STED images of a nucleus from GFP-MSH-5 worms immunolabeled for GFP-MSH-5 (magenta) and HTP-3 (green). Middle, intensity distribution extracted from line profile of the merged image (insets). Black lines denote fits to either one (GFP-MSH-5) or a sum of two (HTP-3) Gaussian distribution plus constant background. Scale bars = 1 µm. Right, histograms of GFP-MSH-5 focus center positions (n = 112). Values of 1 or −1 indicate localization on the meiotic axes, whereas a value of 0 indicates localization at the center of the SC-CR. Magnitudes >1 or <−1 (lighter bars) indicate GFP-MSH-5 center position outside the axes. Blue curves represent a Gaussian fit to the data, whereas the gray distribution represents estimated localization error (from panel (A)). Bottom, schematic diagram of synaptonemal complex geometry, with axes marked in green and the GFP-MSH-5 focus in magenta.