(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Canonical and noncanonical roles of Hop1 are crucial for meiotic prophase in the fungus Sordaria macrospora [1]

['Emeline Dubois', 'Université Paris-Saclay', 'Commissariat À L Énergie Atomique Et Aux Énergies Alternatives', 'Cea', 'Centre National De La Recherche Scientifique', 'Cnrs', 'Institute For Integrative Biology Of The Cell', 'Gif-Sur-Yvette', 'Stéphanie Boisnard', 'Henri-Marc Bourbon']

Date: 2024-07

We show here that in the fungus Sordaria macrospora, the meiosis-specific HORMA-domain protein Hop1 is not essential for the basic early events of chromosome axis development, recombination initiation, or recombination-mediated homolog coalignment/pairing. In striking contrast, Hop1 plays a critical role at the leptotene/zygotene transition which is defined by transition from pairing to synaptonemal complex (SC) formation. During this transition, Hop1 is required for maintenance of normal axis structure, formation of SC from telomere to telomere, and development of recombination foci. These hop1Δ mutant defects are DSB dependent and require Sme4/Zip1-mediated progression of the interhomolog interaction program, potentially via a pre-SC role. The same phenotype occurs not only in hop1Δ but also in absence of the cohesin Rec8 and in spo76-1, a non-null mutant of cohesin-associated Spo76/Pds5. Thus, Hop1 and cohesins collaborate at this crucial step of meiotic prophase. In addition, analysis of 4 non-null mutants that lack this transition defect reveals that Hop1 also plays important roles in modulation of axis length, homolog-axis juxtaposition, interlock resolution, and spreading of the crossover interference signal. Finally, unexpected variations in crossover density point to the existence of effects that both enhance and limit crossover formation. Links to previously described roles of the protein in other organisms are discussed.

Funding: This work was supported by grants from the Agence Nationale de la Recherche (ANR-20-CE12-0006) to E.E., the National Institutes of Health, R35 GM136322, to N.K., via a subcontract collaboration with D.Z., and the CNRS (UMR 9198, I2BC). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Copyright: © 2024 Dubois et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

The present study includes identification of Sordaria Hop1 protein and documentation of its chromosomal loading patterns in wild type and diverse pairing and recombination mutants. Analysis of a complete HOP1 deletion mutation and of 6 mutant alleles that express different portions or mutated forms of the Hop1 protein, allowed detailed elucidation of multiple component roles of Hop1 in: (i) loading of the protein; (ii) modulation of axis length; (iii) resolution of chromosome entanglements formed during pairing; and (iv) crossover patterning and interference. We found also that Sordaria Hop1 is specifically required during the transition from homolog coalignment to synapsis for the maintenance of the cohesin-associated protein Spo76/Pds5 (but not for maintenance of the cohesin Rec8) along chromosome axes. Maintenance of Spo76/Pds5 along the chromosomes is correlated with initiation of the SC because its axis localization is maintained in absence of recombination and Sme4/Zip1 and thus SC formation. Mutations in REC8 and SPO76/PDS5 confer analogous effects. These 3 axis-associated molecules are thus required collaboratively to ensure regular progression from leptotene to pachytene. Implications of these and other findings as well as relationships to Hop1 roles in other organisms are discussed.

In the current work, we have exploited the power of the fungus Sordaria macrospora as a system for visualizing whole chromosome dynamics in relationship with the recombination process to further investigate the role(s) of this organism’s HORMA-domain component Hop1 during meiotic prophase. Previous studies in other organisms indicate that this protein presents diverse patterns of localization throughout synapsis. Hop1 is also involved differently among different organisms with respect to key meiotic events like formation of DNA double-strand breaks (DSBs), crossovers, and synaptonemal complex (SC) (review and references in [ 1 – 3 , 8 ]). For example, in budding yeast, mouse, and C. elegans, Hop1 is required for initiation of recombination while, in Arabidopsis thaliana, DSB formation is Hop1-independent [ 9 – 12 ]. Our current findings add to this diversity, as described below.

During the meiotic program of most organisms, recombination at the DNA level, recombination-mediated pairing of maternal and paternal chromosomes (homologs), as well as progressive organization of higher-order structure along chromosomes, all occur in direct functional linkage. This linkage is mediated in large part via direct physical association of recombination complexes with chromosome axes that comprise: meiotic and mitotic cohesin complexes, cohesin-associated proteins like Pds5/Spo76 and Wapl, condensins, Topoisomerase II and meiosis-specific axis proteins. The latter include one or several HORMA domain-containing proteins (budding-yeast Hop1, mammalian HORMAD1 and HORMAD2, plant ASY1/PAIR2 and ASY2, and Caenorhabditis elegans HIM-3, HTP-1, HTP-2 and HTP-3) and a coiled-coil domain protein (budding-yeast Red1, mammalian SYCP2/SYCP3 and plant ASY3/PAIR3/DSY2 and ASY4) that binds directly to Hop1 in vitro and in cells (review and references in [ 1 – 3 ]). Many gaps still exist in our understanding of exactly how axis structure arises, what are the natures of the molecular and functional interactions among axis components and of axes with their associated recombination complexes (e.g. [ 2 , 4 – 7 ], review in [ 3 , 8 ]).

Results

Axis association of Hop1 does not require DSBs or recombination proteins Hop1 localization was first analyzed in 2 mutants spo11Δ and mer2Δ which do not form DSBs and remain thus asynaptic until end pachytene [25]. The haploid number of chromosomes being 7 in Sordaria, Hop1 is visible along the axes of the 14 un-synapsed chromosomes (shown by different colors) in both mutants from leptotene throughout pachytene indicating that Hop1 localization on axis is independent of both of these molecules and of the DSB formation itself (Fig 2A and 2B; n = 20 nuclei for each mutant). Secondly, Hop1 localizes also normally in absence of post-DSB recombination proteins in mer3Δ and msh4Δ (Fig 2C and 2D; n = 20 nuclei for each mutant) which make DSBs but exhibit defects in later stages as manifested in aberrant coalignment and synapsis (illustrated in Fig 2D by un-synapsed yellow and blue homologs compared to perfectly coaligned red and cyan homolog) (see also [28]). Thirdly, Hop1 localization is also independent of the Zip2-Zip4 complex (Fig 2E) which plays a central role in SC nucleation [28]. In accord with normal axis formation in these mutants, they all exhibit mean WT-like axis lengths as measured by Hop1-GFP (52.8 ± 3 microns in spo11Δ versus 53.7 ± 4 microns in WT; n = 20 and 150 nuclei, respectively; ± indicates the standard deviation, here and below). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. Hop1 localizes along axes in recombination mutants but synapsis is defective when absent. (A, B) Hop1 localizes all along the 14 unsynapsed chromosome axes in absence of DSBs in spo11Δ (A) and mer2Δ mutants (B) as confirmed by corresponding drawings. (C–E) Same localization of Hop1-GFP in mer3Δ (C), msh4Δ (D), and zip2Δ (E) mutants with defective coalignment. (F) Sme4/Zip1-GFP localization in WT and hop1Δ with corresponding drawings (middle) and DAPI (right). SC installation occurs in both WT (left) and hop1Δ (right) but instead of the 7 SCs seen in WT (left), 14 segments of SCs are seen in the mutant (right). (G) Similar contrast in SC formation between WT (top) and hop1Δ (bottom) observed with the SC central-component Ecm11. Scale bars: 2 μm. (H) Comparison of the distribution of the number of Sme4 (blue) and Ecm11 (magenta) SC segments in hop1Δ vs. WT SCs (large blue dot). (I) Comparison of the SC lengths between WT (left) and hop1Δ (right). (H, I): n = 39, 35, and 19 nuclei, respectively. Mean and error bar (SD) are indicated for each set. Statistical significance was assessed using two-tailed Student’s t test; ns, not significant (P ≥ 0.05); ****P < 0.0001. The raw data underlying panels 2H and 2I are available in S1 Data. DSB, double-strand break; SC, synaptonemal complex; WT, wild-type. https://doi.org/10.1371/journal.pbio.3002705.g002

In the absence of Hop1, the synaptonemal complex initiates normally at zygotene but is never complete from telomere to telomere In hop1Δ, as in WT, SC installation (observed by the SC transverse filament Sme4/Zip1-GFP) initiates at zygotene in several small nucleation sites (S3A Fig and below). However, at pachytene, in contrast to WT where SCs are formed along all the seven homologs (Fig 2F, left), hop1Δ nuclei exhibit only short segments of SCs (Fig 2F, right). The same contrast between WT and mutant is also observed with the SC central-region protein Ecm11 (Fig 2G, top and bottom, respectively), implying that the defect pertains to the entire tripartite SC structure. As expected, the number of Sme4 and Ecm11 segments is significantly similar: two-tailed Student’s test; P ≥ 0.05; Fig 2H; details in corresponding legend. The total per-nucleus length of hop1Δ SC segments is significantly different from WT (60% of the full WT SC length: 31.7 ± 9.1 (Sme4) and 33.8 ± 7.6 (Ecm11) microns, versus 53.7 ± 4 microns in WT, but again significantly similar (P ≥ 0.05) for the 2 proteins (Fig 2I). Although incomplete, SC formation in the hop1Δ mutant exhibits the same kinetics and functional dependencies as in WT: (i) SCs appear at zygotene as small Sme4/Zip1-GFP segments (S3A Fig, left versus right and see below) and disappear progressively as the homologs desynapse until, by the diffuse stage, they are no longer detectable except in a few remaining segments (illustrated by Ecm11-GFP + Hei10-GFP, S3B Fig). (ii) SC formation is DSB dependent: like in the single spo11Δ mutant, only univalents are visible in the hop1Δ spo11Δ double mutant (S3C Fig, left and right, respectively; n = 15 and 20 nuclei). (iii) Also, just as in the sme4Δ single mutant, homologs are coaligned at 200 nm distance in the hop1Δ sme4Δ double mutant, but do never progress beyond this stage and, thus, do not form SCs (S3D Fig, left and right, respectively; n = 40 and 20 nuclei).

Analysis of haploid meiosis shows that the SC and Spo76/Pds5 defects seen in hop1Δ are both dependent on Hop1 per se and not an indirect consequence of aberrant homolog pairing Identification of synapsis and Spo76 localization defects along hop1Δ homologs during pachytene (above) raised the possibility that these defects might be a secondary consequence of a prior mutant defect in coalignment/pairing. To exclude this possibility, we examined prophase events during haploid meiosis where, in absence of a homolog, no homolog pairing occurs. We showed previously that a mutant lacking the Slp1 SUN-domain protein is defective for the nuclear fusion (“karyogamy”) that precedes meiotic prophase in Sordaria, but not for regular progression of meiotic events in the 2 unfused twin-haploid nuclei [33]. Specifically, in this mutant, the entire program of SC formation and recombination occurs efficiently between sister chromatids in each of the 2 nuclei. Also, just as in WT, in slp1Δ, Spo76-GFP loads as smooth lines along the 7 chromosomes (S4B Fig) and remains present in this condition throughout pachytene. We now find that absence of Hop1 confers the same constellation of defects in these twin slp1Δ haploid nuclei as in diploid meiosis. (i) In the double hop1Δ slp1Δ mutant, Spo76-GFP forms continuous lines at leptotene by ascus size (S4C Fig) as in the hop1Δ single mutant (S4B Fig). However, at later stages, a mixture of short and longer Spo76 lines is present in all double hop1Δ slp1Δ nuclei (arrows in S4D Fig; n = 20 nuclei) just as in the diploid hop1Δ single mutant (Fig 4E), implying that the protein is being lost from some chromosome regions. (ii) In the single slp1Δ mutant, the SC central component Ecm11-GFP exhibits continuous seven lines in both nuclei with Hei10 foci in most of the lines (arrows in S4E Fig). In the double hop1Δ slp1Δ mutant, Ecm11-GFP (with Hei10 foci, arrows in S4F Fig) exhibits more numerous and shorter lines (S4F Fig, middle) than in the single slp1Δ mutant (S4E Fig), implying that Ecm11 and thus the SC is never complete along the seven chromosomes (n = 20 nuclei). These results show clearly that defective SC formation in diploid hop1Δ and correlated loss of Spo76/Pds5 from axes which are not linked by an SC, cannot reflect aberrancies in the pairing process in the mutant because the hop1Δ slp1Δ double mutant shows similar defects between the sister chromatids.

The C-terminal domain of Sordaria Hop1 is not required for Hop1 loading but is crucial for both pairing and chromosome-entanglement resolution, despite absence of a clear “closure motif” By sequence comparison, the last 63aa of Sordaria Hop1 correspond to both the type of charged and hydrophobic amino acids found in other organisms and the expected length of a “closure motif,” found at Hop1 C-termini in many organisms [24]. We created a Hop1-CTD mutant protein, with these 63 aa removed, and tagged it by mCherry. Hop1-CTD-mCherry colocalizes continuously with Spo76/Pds5-GFP along chromosome axes from leptotene throughout pachytene (Fig 7A) just as does Hop1 in WT (Fig 1C). The deleted domain is thus not required either to anchor the protein to the chromosome axis or for maintenance of normal axis integrity during pachytene. This phenotype is particularly notable because in some organisms, the C-terminal “closure motif” is known to mediate Hop1 axis localization (e.g., [2,3,6]). Interestingly, however, A. thaliana homolog ASY1 has a canonical C-terminal CM but does also not require this motif for axis association of the protein [23]. Thus, for both Sordaria and A. thaliana, the mechanism by which Hop1 becomes incorporated into axes, without benefit of a closure motif, remains to be defined. One possibility could be a direct interaction of Hop1 with cohesins. Such interactions are known in several organisms but involvement of the closure motif per se, remains unknown (e.g., [2]). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 7. Coalignment, SC and CO/Hei10 phenotypes of the 4 non-null hop1 mutants. (A–D) hop1-CTD mutant. (A) Pachytene nucleus illustrating the perfect colocalization of Hop1-CTD-mC with Spo76-GFP (“green” Hei10 foci are visible along all bivalents). (B) Late leptotene coalignment stage (axes marked by Spo76-GFP). The 7 WT homologs are synchronously coaligned (left and middle) while the hop1-CTD mutant shows highly irregular coalignment with some ends synapsed (arrowheads) when others (arrows) are still largely separated. (C) Irregularities persist into zygotene: most ends are synapsed (arrowheads) but one homolog end (arrow and yellow in drawing) is still widely open. (D) Three pachytene nuclei. Left: example of a nucleus with full synapsis (indicated by Sme4/Zip1-GFP). Middle and right: mid- and late pachytene nuclei (and corresponding drawings) with large open regions (arrowheads in middle nucleus) and interlockings (arrows in both). (E–G) hop1-K14R and hop1-K259R mutants. (E) Sites of point mutations are indicated by a black cross on red lines. (F) Pachytene nucleus with perfect colocalization of Hop1-K14R-mC with Spo76-GFP. Arrow indicates the site where 2 bivalents are trapped in the non-synapsed region of a third one. (G) Three examples of delayed/abnormal coalignment/synapsis. Arrows point to interlocking sites (left and right nuclei) and arrowheads indicate synapsed regions in the middle nucleus. (H, I) hop1-CTD mutant. (I) At pachytene, the mutant protein localizes along all axes (left nucleus). The mutant exhibits also regular synapsis as shown by Sme4-GFP (middle nucleus and corresponding drawing) and regularly spaced Hei10 foci (right nucleus). Scale bars: 2 μm. CTD, C-terminal domain; SC, synaptonemal complex; WT, wild-type. https://doi.org/10.1371/journal.pbio.3002705.g007 Using the advantage of continuous axis localization of Hop1-CTD-mC and its perfect colocalization with Spo76-GFP, we next analyzed in detail the pairing and synapsis phenotypes of the hop1-CTD mutant. During WT leptotene, the 7 homologs synchronously co-align first at 400 nm and then at 200 nm separation distance (Fig 7B, left) before proceeding to synapsis. In hop1-CTD, coalignment is severely defective, with some homologs still largely separated (arrows in Fig 7B, right) and others partially paired or synapsed (arrowheads in Fig 7B, right). In accord with these configurations, the mutant exhibits also more or less dramatic per-chromosome and per-region defects at zygotene with most ends synapsed but also un-synapsed ends (respectively arrowheads and arrow in Fig 7C). Moreover, even at pachytene, SC formation (tested by Sme4/Zip1-GFP) is complete from telomere to telomere along the 7 homologs in only 15% of the 81 analyzed pachytene nuclei (Fig 7D, left). The other 85% nuclei show 1 to 5 non-synapsed regions per nucleus (arrowheads in Fig 7D, middle and in corresponding drawing; more examples in S6B Fig). All bivalents are affected but the 3 longer chromosomes are more often partially non-synapsed (in 50% to 40% of the nuclei) than the smaller chromosomes (30% to 18%). The presence of fully synapsed chromosomes, as well as the continuous localization of Spo76-GFP along the axes (above), implies that non-synapsed regions in this mutant are due to defects in homologous pairing and not to abnormal axis status as seen in hop1Δ. Interestingly, the mutant most impressive defect is the presence of a high number of topological irregularities (“interlocks/entanglements”) visible throughout pachytene (arrows in Figs 7D, middle and right, and S5B). Chromosome entanglements do also occur in 20% of WT zygotene nuclei (n = 50 nuclei), but they are absent at pachytene (n = 120), implying active resolution [36]. In contrast, interlock configurations are still present in 41% (33/80) of hop1-CTD late pachytene nuclei (Figs 7D, right and S5B, right), implying a specific defect in interlock resolution. Due to weak Rec8 pachytene “staining” and absence of Spo76 in non-synapsed regions (above), we were not able to count the number of interlockings in hop1Δ, but the fact that all analyzed zygotene nuclei showed at least 1 entanglement (Fig 3C and 3D), could indicate that this defect is also present in the null mutant. Moreover, we showed previously that in absence of Mlh1, 45% of the late pachytene nuclei exhibited at least 1 bivalent with an open region and/or an interlocking, implicating a role of Mlh1 in interlock resolution [36]. To test the role of Mlh1 in the hop1-CTD mutant background, we analyzed synapsis in the hop1-CTD mlh1Δ double mutant (S6C Fig). It exhibits even a greater number of interlocks than each of the single mutants at mid-late pachytene: 80% in hop1-CTD mlh1Δ (n = 25 nuclei) compared to 41% in hop1-CTD (n = 80), 45% in mlh1Δ (n = 300), and 0% in wild type (n = 120). Since regular homolog coalignment is required to avoid interlock formation, the role of Hop1 in pairing could explain this synergy. However, Mlh1 being present in the hop1-CTD mutant, there must be other factors/effects that impede removal of interlocks in the mutant and thus also likely in WT.

The 2 putative SUMO sites are also required for efficient pairing and interlocking resolution but the conserved S/TJQ cluster domain (SCD) is not We also mutated two of the identified putative SUMO sites creating point-mutations in the hop1-K14R and hop1-K259R alleles (red lines in Figs 6A and 7E). Axis integrity is normal: both mutant proteins (tagged by mCherry) and Spo76/Pds5-GFP colocalize continuously along chromosome axes from leptotene throughout pachytene in a hop1Δ background (Figs 7F and S6D) just as in WT (Fig 1C). Synchronous coalignment is, however, defective, with some homologs still largely separated and others partially synapsed (arrows and arrowheads in Fig 7G, middle and right). In accord with these defects, both mutants exhibit complete SC formation only in, respectively, 44% (n = 50 nuclei) and 65% (n = 89) of the analyzed pachytene nuclei and again, the 2 longer chromosomes are more often un-synapsed than the 4 smaller chromosomes (e.g., 82% and 64% in hop1-K14R). Also, as in hop1-CTD nuclei, interlock configurations are visible at zygotene (arrow in Fig 7G, left) and persist throughout late pachytene (arrow in Fig 7F) in 38% (19/50) of the hop1-K14R nuclei and in 13% (12/89) of the hop1-K259R nuclei. In contrast, the hop1-SCD mutant, in which the putative Zn-finger motif (comprising 54 aa) is deleted from the HOP1 gene (Figs 6A and 7H), exhibits none of the chromosome defects characteristic of the null mutant phenotypes: (i) Hop1-SCD-mC is continuous along all axes from leptotene to pachytene (Fig 7I, left), where it colocalizes with Spo76-GFP (S6E Fig); (ii) 96% of nuclei show fully regular SCs (by Sme4/Zip1-GFP, Fig 7I, middle; n = 50 nuclei) implying that, both pairing and interlock resolution are WT-like and that the SCD motif is not required for these features. (iii) Hei10 foci form along all 7 bivalents (Fig 7I, right) like in WT, but further analyses show defects in their localization (e.g., closely spaced foci) and number, indication of a role of this domain in the crossover patterning (below). Unexpectedly, when compared to WT, this mutant still produces a reduced number of 8-spored asci in the hop1Δ background (S2B Fig). As we observed no obvious precocious homolog separation in the 20 metaphase I nuclei of the mutant, we have no clear explanation for this phenotype. It was shown that in budding yeast, the pericentromeric regions are under-enriched for Hop1 [45]. We do not know the status of Hop1 at the Sordaria centromeres, but if the same pattern is true, we cannot exclude an alteration of chromatid cohesion in the mutant.

All 4 “non-null” mutants exhibit increased axis length In the 2 hop1 point mutants, as well as in the hop1-SCD and hop1-CTD mutants (hereafter referred to collectively as the 4 “non-null” mutants), axes assemble correctly with WT-like localization of Spo76-GFP and Rec8-GFP. Moreover, SC forms along the lengths of all homologs. However, we also discovered that all 4 mutants exhibit 25% longer axes/SCs than WT (Fig 7A): 65.1 ± 5.7 microns in hop1-SCD (n = 54), 69.2 ± in hop1-CTD (n = 49), 65.1 ± 4.1 in hop1-K14R (n = 53 nuclei), and 64.5 ± 6.5 in hop1-K259R (n = 45), when compared to the 52.2 ± 5.1 microns in WT (n = 146). Interestingly, while significantly different from WT (Brown–Forsythe ANOVA P < 0.0001), the observed increases are not significantly different among the 4 mutants (P > 0.05). Furthermore, mutant axis lengths are already longer at the coalignment stage, thus before SC formation, indication of an early role of Hop1. Axes lengths at late zygotene and early pachytene in hop1Δ (by Rec8-GFP) are also longer than in WT: 70.7 ± 3.8 microns (only 10 nuclei measured, due to the weak Rec8 signal). These results reveal that Hop1 has a discrete function that impacts axis length during axis assembly.

Non-null mutant phenotypes reveal a direct role for Hop1 in crossover interference In the hop1-SCD mutant, where SC formation is highly regular (above Fig 7I), crossover interference can be examined without complexities due to incomplete synapsis. In this mutant, crossover interference is defective by 2 criteria. First, analysis of crossover interference by coefficient of coincidence (CoC) reveals a substantial increase in the probability of closely spaced double crossovers in both long and short bivalents (Fig 8B). Second, the gamma distribution of distances between adjacent Hei10 foci for 350 hop1-SCD bivalents exhibits a ν value of 2.71 ± 0.46 as compared to the 5.03 ± 0.18 seen for foci along 658 SCs of WT, consistent with defective interference. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 8. SC lengths, crossover number, and distribution in the 4 non-null mutants. (A) Comparison of the SC length per nucleus (in microns) of the 4 non-null mutants vs. WT SC lengths, indicates that their length is different from WT (gray, left) but not significantly different among the 4 mutants (Brown–Forsythe ANOVA P > 0.05; ns = not significant). Mean and error bar (SD) are indicated for each set. (B, C) CoC relationships for positions of correlated Hei10 foci/crossovers along (B) hop1-SCD and (C) hop1-K14R, hop1-K259R and hop1-CTD pachytene bivalents (longer 1 and 2 bivalents left, 3 to 7 bivalents right). (B) The strength of crossover interference is provided by the inter-interval distance at which CoC = 0.5. This “interference distance” (see details in Methods) is 1.30 microns in WT (black) and 0.6 microns in hop1-SCD (blue). These values correspond approximately to the average distance between adjacent crossovers. (C) The same difference between WT (black) and mutants (red, green, and orange lines) is seen in the CoC curves of the corresponding mutants. CoC curves indicate that the 4 mutants show interfering COs but with smaller than WT interference distances. (D) Number and distribution of Hei10 foci along the hop1-SCD SCs of the 2 long bivalents (bivalents 1 and 2; left) and the other shorter bivalents (bivalents 3–7; right); red arrows indicate the absence of E0 bivalents, which means that all bivalents exhibit at least 1 Hei10 focus. (E) Comparison of the number of Hei10 foci per nucleus in the 4 non-null mutants vs. WT (gray). The number of foci is not significantly different among the 4 mutants. Only the 2 SUMO-site mutants (right) show slightly higher numbers of foci than WT (ANOVA P < 0.0004; see text for the number of nuclei analyzed). Mean and error bar (SD) are indicated for each set. The raw data underlying all panels from 8A to 8E are available in S1 Data. CoC, coefficient of coincidence; SC, synaptonemal complex; WT, wild-type. https://doi.org/10.1371/journal.pbio.3002705.g008 Furthermore, the same effect is observed in the hop1-K14R, hop1-K259R, and hop1-CTD mutants, even though pairing/synapsis is often compromised (above): their gamma ν values are respectively of 2.7 ± 0.44, 2.53 ± 0.41, and 2.69 ± 0.36 (versus 5.03 ± 0.18 in WT) and CoC analysis reveals, a substantial increase in double crossovers that are closely spaced in both long and short chromosomes (Fig 8C). Moreover, CoC curves show that crossover interference is similarly reduced in the hop1-K259R mutant, in nuclei with full SCs and nuclei with open regions in their SCs (due mostly to non-resolved interlockings; S6F Fig). On the other hand, there is no increase in the frequency of bivalents that fail to exhibit even 1 Hei10 focus, even in the smaller bivalents (Fig 8D, red arrows), indicating that occurrence of the so-called “obligatory crossover” remains intact, in contrast with the phenotype of the A. thaliana asy1 mutant [46]. We note that the effects observed in the non-null mutants reflect likely a loss of normal Hop1 function rather than a gain of interfering function by the mutant proteins. This interpretation is supported by the fact that all 4 non-null mutants exhibit the same basic defects, many of which are also mirrored in the null mutant. Interestingly, two of the mutants (hop1-K14R and hop1-K259R) carry point mutations in two of the identified putative Hop1 SUMOylation sites, suggesting that SUMOylation of Hop1 is required for WT-like axis length and crossover patterns.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002705

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.