Multiple opposing constraints govern chromosome interactions during meiosis

Abstract

Homolog pairing and crossing over during meiosis I prophase is required for accurate chromosome segregation to form euploid gametes. The repair of Spo11-induced double-strand breaks (DSB) using a homologous chromosome template is a major driver of pairing in many species, including fungi, plants, and mammals. Inappropriate pairing and crossing over at ectopic loci can lead to chromosome rearrangements and aneuploidy. How (or if) inappropriate ectopic interactions are disrupted in favor of allelic interactions is not clear. Here we used an in vivo "collision" assay in budding yeast to test the contributions of cohesion and the organization and motion of chromosomes in the nucleus on promoting or antagonizing interactions between allelic and ectopic loci at interstitial chromosome sites. We found that deletion of the cohesin subunit Rec8, but not other chromosome axis proteins (e.g. Red1, Hop1, or Mek1), caused an increase in homolog-nonspecific chromosome interaction, even in the absence of Spo11. This effect was partially suppressed by expression of the mitotic cohesin paralog Scc1/Mdc1, implicating Rec8's role in cohesion rather than axis integrity in preventing nonspecific chromosome interactions. Disruption of telomere-led motion by treating cells with the actin polymerization inhibitor Latrunculin B (Lat B) elevated nonspecific collisions in rec8Δ spo11Δ. Next, using a visual homolog-pairing assay, we found that the delay in homolog pairing in mutants defective for telomere-led chromosome motion (ndj1Δ or csm4Δ) is enhanced in Lat B-treated cells, implicating actin in more than one process promoting homolog juxtaposition. We suggest that multiple, independent contributions of actin, cohesin, and telomere function are integrated to promote stable homolog-specific interactions and to destabilize weak nonspecific interactions by modulating the elastic spring-like properties of chromosomes.

Figure 1. Inter-chromosomal collision assay.

A. The chromosomal location of loxP sites for the collision assay described in the text. The primer configurations for detection of recombinants by qPCR are diagrammed. B. Comparison of average allelic and ectopic collision levels in NTD80 versus ndt80Δ andndt80Δ spo11Δ. Collisions are the measured level of recombinants per meiosis (i.e. 4 chromatids). The variance for allelic collision levels was somewhat higher than for ectopic. This may be due to differences in primer sequence and primer pair concentrations optimized for the respective qPCR reaction conditions; the template DNA isolated from each individual time course for both cases is the same. C. Percent of allelic collisions (blue) and ectopic collisions (red) among total collisions measured (allelic and ectopic).

Figure 2. Rec8 promotes allelic collisions independent of its role in sister chromatid cohesion and prevents nonspecific collisions.

A. Analysis of allelic and ectopic collision levels as described in Figure 1. B. Heatmap of Sidak adjusted P-values comparing collision levels between all allelic collision levels (left) and all ectopic collision levels (right).

Figure 3. Elevated levels of nonspecific collisions in rec8Δ mutants do not require recombination initiation.

A. Analysis of collisions in spo11Δ, spo11Δ rec8Δ, spo11Δ prec8-SCC1, spo11Δ ndj1Δ, and spo11Δ ndj1Δ rec8Δ mutants with Lat B treatment. Allelic (blue) and ectopic (red) collision levels in untreated cultures (dark bars) and Lat B treatment (light bars). Asterisks denote significant differences as follows: (*), P-values between 0.05 and 0.01; (**), P-values between 0.01 and 0.001; (***), P-values <0.001 by a two-tailed Student's t-test. B. Heatmap of Sidak adjusted P-values from Student's t-test comparing collision levels between relevant mutants in untreated and Lat B treated cells.

Figure 4. Elevated levels of nonspecific collisions in rec8Δ do not require Ndj1-dependent telomere attachments to the NE.

A. Analysis of allelic and ectopic collision levels in rec8Δ, ndj1Δ, ndj1Δ rec8Δ, csm4Δ, and ndj1Δ csm4Δ mutants with Lat B treatment. Graph parameters are as described as in Figure 3. B. Heatmap of Sidak adjusted P-values from Student's t-test comparing collision levels between relevant mutants in untreated and Lat B treated cells.

Figure 5. Homolog pairing kinetics in wild-type, ndj1Δ, csm4Δ, and ndj1Δ csm4Δ cells with and without Lat B treatment.

Kinetics of pairing in cells with tetO arrays integrated at URA3, located 35 kb away from the centromere on chromosome V, and expressing tetR-GFP fusion protein. Homologs are considered paired if only one GFP can be visualized in the cells (n = 200) from each time point. Error bars represent the standard error of the percentage of cells paired for independent cultures for each mutant (n = 200). All strains carry the ndt80Δ mutation. A. Analysis of WT (center panel and left panel) and ndj1Δ (right panel and left panel) pairing kinetics in the presence or absence of Lat B. B. Analysis of csm4Δ (center panel and left panel) and ndj1Δ csm4Δ (right panel and left panel) pairing kinetics in the presence or absence of Lat B.

Figure 6. DSB–independent ectopic collision levels in mutants with defects in centromere coupling and bouquet formation.

A. Analysis of collisions in spo11Δ, spo11Δ ndj1Δ, spo11Δ zip1Δ, spo11Δ ndj1Δ zip1Δ, spo11Δ csm4Δ and spo11Δ csm4Δ ndj1Δ mutants with Lat B treatment. Graph parameters are as described in Figure 3. B. Heatmap of Sidak adjusted P-values from Student's t-test comparing collision levels between relevant mutants in untreated and Lat B treated cells.

Figure 7. A mechanical model for homolog pairing.

A hypothetical sequence of interactions between homologous chromosomes (shown as black or red) subjected to a coupled-spring oscillator (see text). In the sequence from left to right, homolog pairing becomes progressively stabilized as weak interactions are disrupted. The positive and negative forces of actin influence both homologs, with actin-based associations shown at telomeres (green; squares for the red chromosome and circles for the black homologous chromosome) and at interstitial sites (blue; squares for the red chromosome and circles for the black homologous chromosome). Arrows around the periphery of the nucleus indicate direction of movement for the telomeres (green) and the interstitial sites (blue). Grey arrows in the interior of the nucleus show “Brownian-like” motion/unknown forces on the chromosomes , . (i) In wild-type cells, segments of chromosomes that are in closer proximity have axial segments that are more compact. (ii) Compact segments of homologous chromosomes interact. (iii) Movement of chromosome attachment points on the nuclear envelope results in stretching of segments that remove unstable interactions between chromosomes. (iv) Stable interactions between allelic loci are those achieved up to the point of dHJ resolution as established using the Cre/loxP assay .