Pch2 links chromosome axis remodeling at future crossover sites and crossover distribution during yeast meiosis

Abstract

Segregation of homologous chromosomes during meiosis I depends on appropriately positioned crossovers/chiasmata. Crossover assurance ensures at least one crossover per homolog pair, while interference reduces double crossovers. Here, we have investigated the interplay between chromosome axis morphogenesis and non-random crossover placement. We demonstrate that chromosome axes are structurally modified at future crossover sites as indicated by correspondence between crossover designation marker Zip3 and domains enriched for axis ensemble Hop1/Red1. This association is first detected at the zygotene stage, persists until double Holliday junction resolution, and is controlled by the conserved AAA+ ATPase Pch2. Pch2 further mediates crossover interference, although it is dispensable for crossover formation at normal levels. Thus, interference appears to be superimposed on underlying mechanisms of crossover formation. When recombination-initiating DSBs are reduced, Pch2 is also required for viable spore formation, consistent with further functions in chiasma formation. pch2Delta mutant defects in crossover interference and spore viability at reduced DSB levels are oppositely modulated by temperature, suggesting contributions of two separable pathways to crossover control. Roles of Pch2 in controlling both chromosome axis morphogenesis and crossover placement suggest linkage between these processes. Pch2 is proposed to reorganize chromosome axes into a tiling array of long-range crossover control modules, resulting in chiasma formation at minimum levels and with maximum spacing.

Figure 1. Association of Pch2 with WT chromosomes at different stages of SC polymerization.

(A–H) Early zygotene nuclei; Zip1 (A,E), and HA-Pch2 (B,F). (I–L) Late zygotene nucleus. (M–P) Pachytene nucleus. Arrows indicate rDNA clusters, identified by Pch2 localization pattern and lack of Zip1 (K,O). Spread zygotene and pachytene nuclei are identified base on extend of Zip1 staining. Colors are indicated by the corresponding labels. Meiosis was induced at 33°C.

Figure 2. Association of Pch2 with CO designation marker Zip3 at different stages of synapsis.

(A–D) Spread nucleus during early zygonema from a strain carrying HA-Pch2 and Zip3-GFP, stained with antibodies against HA- and GFP-epitopes as well as Zip1. (E–H) Spread pachytene nucleus. Arrow indicates rDNA cluster (G). Colors are indicated by the corresponding labels. Meiosis was induced at 33°C.

Figure 3. Association of axis components Hop1 and Red1 with CO–designation marker Zip3 at different stages of synapsis.

(A–D) Spread pachytene nucleus stained with antibodies against Hop1 and Zip3-GFP. (E–H) Spread pachytene nucleus labeled with antibodies against Red1-HA and Zip3-GFP. (I–R) Spread nuclei from time course shown in (E–H) stained with antibody against Zip1 as well as Red1-HA and Zip3-GFP. (I–M) Spread zygotene nucleus. (N–R) Spread pachytene nucleus. Colors are indicated by the corresponding labels. Yellow colors in (A,E,I,N) indicate overlap between the indicated protein signals. Meiosis was induced at 33°C.

Figure 4. Effects of Dmc1, Pch2, and Ndt80 on Zip3 and Hop1 localization and SC length in WT and pch2Δ.

(A–P) Spread nuclei from dmc1Δ (A–C), WT (PCH2) (D–F), pch2Δ (G–I), PCH2ndt80Δ (J–L), pch2Δndt80Δ (M–O) were stained with antibodies against Hop1 and Zip3. Colors are indicated by the corresponding labels in the individual channels. Regions of overlap between Zip3 and Hop1 are indicated by yellow color. Note the patchy versus continuous Hop1 localization in WT (E,K) versus pch2Δ (H,N). White arrow in (H) indicates the Hop1 stained nucleolus in pch2Δ. (P) Numbers of Zip3 foci per nucleus determined in spread nuclei in dmc1Δ (t = 5 hrs), WT (t = 7 hrs), pch2Δ (t = 7 hrs), PCH2ndt80Δ (t = 8 hrs), pch2Δndt80Δ (t = 8 hrs), respectively. P-values are from two-sided Wilcoxon rank sum tests. Error bars represent 95% confidence intervals. (See also Figure S2 for analysis of Hop1 domains in WT and pch2Δ.) (Q–V) Nuclei from WT (Q–S) or pch2Δ (T–V) from parallel time courses were spread and stained with antibodies against Hop1 and Zip1. Note discontinuous staining of Hop1 and Zip1 in WT (Q,R) compared to continuous staining patterns for both proteins in pch2Δ (T,U). (W) Twenty well-spread nuclei were selected and the combined contour length of Zip1 and Hop1 was determined in WT and pch2Δ (see Materials and Methods). The average SC/axis length was 34 µm (±1.8 µm C.I.) in WT and 40 µm (±2.2 µm C.I.) in pch2Δ, indicating a significant increase in pch2Δ versus WT (p = 0.00023; two-sided Wilcoxon rank sum test). Error bars represent 95% confidence intervals. Meiosis was induced at 33°C.

Figure 5. Genetic map distances in WT and pch2Δ determined by tetrad analysis at 33°C.

(A) Test intervals on chromosomes III, VII, and VIII. Marker order, physical (kb) and genetic sizes (cM) of surveyed chromosome regions are shown. Ovals indicate centromeres; diamonds represent telomeres. Intervals are referred to by numbers 1–9 throughout the text. (B) Classes of tetrads for a given interval and contributing crossover (CO) events. Parental ditype tetrads (PD), generated in the absence of COs, and tetratype tetrads (TT), generated when a single CO occurs, are predominant tetrad classes. Nonparental ditype tetrads (NPD) are generated exclusively by double COs involving all four chromatids. Double COs involving three chromatids produce TTs, and those involving two chromatids produce PDs. (C) Genetic distances determined for intervals 1–9 (see A) at 33°C. Contributions of TTs and NPDs to map distances are indicated in different shades. Asterisks indicate significant differences between map distances in WT and pch2Δ strains. Error bars represent standard errors (see also Table 1). (D) Frequencies of non-Mendelian segregation ( = gene conversion) events (i.e. markers deviating from 2∶2 segregation) in WT and pch2Δ at 33°C.

Figure 6. Interference in WT and pch2Δ in nine intervals at 33°C.

(A) Modified coincidence analysis. Left column: Map distances for test intervals ( = Test) determined from tetrads with parental reference intervals (PD). Right column: Map distances for test intervals determined from tetrads with nonparental reference intervals (TT, NPD). Interval names are given as follows: Test intervals are indicated by the first number in bold, CO status of the reference interval is indicated by the letter in italic (P = parental; N = nonparental), and reference intervals are specified by the number in italic. For interval numbers see Figure 5A. Error bars are standard errors. Asterisks above bars indicate significant differences for map distances in bracketed intervals between genotypes. Asterisks on bars in right column indicate significant differences between map distances for test intervals within genotypes (indicating interference). White dots on bars in right column indicate lack of significant differences between the same pairs of map distances within a genotype (indicating absence of interference). (B) Strength of CO interference as a function of physical interval size. Combined interval length (kb) indicates the physical distance between the two outermost markers of intervals considered. Ratios of map distances (P/N) were determined and compared between WT and pch2Δ. (Ratios <1 indicate loss of interference in pch2Δ and ratios >1 indicate increased strength of interference in pch2Δ versus WT). Symbols: *, significant interference in WT and pch2Δ; ○, no significant interference in WT and pch2Δ; ▾, significant interference in WT, but not in pch2Δ; ▴, significant interference in pch2Δ, but not in WT.

Figure 7. Spore viability patterns at reduced DSB levels in WT and pch2Δ at 30°C.

Tetrads were dissected from strains with the indicated genotypes of spo11 alleles. Ratios 4∶0, 3∶1, 2∶2, etc. indicate the frequencies of four-spore viable, three-spore viable, two-spore viable, etc. tetrads. spo11 hypomorphic mutants form ∼80% (spo11-HA/”), ∼30% (spo11-HA/spo11yf), and ∼20% (spo11da/”) of WT DSB levels .

Figure 8. CO interference and spore viability in pch2Δ and WT at different temperatures.

(A) Modified coincidence analysis for the same strains shown in Figure 6, sporulated at 30°C. Note that the apparent lack of interference between intervals 7 and 8 is likely due to large standard errors in the relatively smaller data set (see Table S1). See Figure 6 legend for details. (B) Spore viability patterns indicative of homolog nondisjunction for the same strains analyzed in Figure 7, sporulated at 33°C. See Figure 7 legend for details.

Figure 9. Spore viability and crossovers in pch2Δ under different incubation conditions.

(A) Spore viability patterns in PCH2spo11da/” (left) and pch2Δspo11da/” (right), undergoing meiosis at 30°C (liquid medium), 27°C (liquid medium), or 30°C (solid medium). At least 30 tetrads were dissected under each condition. (B) Crossover levels at the HIS4LEU2 recombination hotspot. Meiotic cultures were split at t = 0 hrs, and crossover levels were determined after incubation at 30°C or 27°C for 24 hrs. Asci from the same cultures were dissected to determine effects on spore viability (see A).