Mutation of the ATPase Domain of MutS Homolog-5 (MSH5) Reveals a Requirement for a Functional MutSγ Complex for All Crossovers in Mammalian Meiosis

EXO1 promotes the meiotic MLH1-MLH3 endonuclease through conserved interactions with MLH1, MSH4 and DNA

The endonuclease activity of MLH1-MLH3 (MutLγ) is stimulated by MSH4-MSH5 (MutSγ), EXO1, and RFC-PCNA to resolve meiotic recombination intermediates such as double Holliday junctions (HJs) into crossovers. We show that EXO1 directly interacts with MLH1 via the EXO1 MIP motif and a patch centered around EXO1-I403. Disrupting this interaction unexpectedly only partially inhibited MutLγ. We found that EXO1 also directly interacts with MutSγ. Crucially, a single point mutation in EXO1 (W371E) impairs its interaction with MSH4 and completely abolished its ability to activate DNA nicking by MutLγ without affecting its intrinsic nuclease function. Finally, disrupting magnesium coordinating residues in the nuclease domain of EXO1 has no impact on MutSγ-MutLγ activity, while the integrity of EXO1 residues mediating interactions with double-stranded DNA (dsDNA) is important. Our findings suggest EXO1 is an integral structural component of the meiotic resolvase complex, supported by conserved interactions with MutSγ, MutLγ and dsDNA. We propose that EXO1 helps tether MutSγ-MutLγ to dsDNA downstream of HJ recognition to promote DNA cleavage.

Crossover Interference Mediates Multiscale Patterning Along Meiotic Chromosomes

The classical phenomenon of crossover interference is a one-dimensional spatial patterning process that produces evenly spaced crossovers during meiosis. Quantitative analysis of diagnostic molecules along budding yeast chromosomes reveals that this process also sets up a second, interdigitated pattern of related but longer periodicity, in a "two-tiered" patterning process. The second tier corresponds to a previously mysterious minority set of crossovers. Thus, in toto, the two tiers account for all detected crossover events. Both tiers of patterning set up spatially clustered assemblies of three types of molecules ("triads") representing the three major components of meiotic chromosomes (crossover recombination complexes and chromosome axis and synaptonemal complex components), and give focal and domainal signals, respectively. Roles are suggested. All observed effects are economically and synthetically explained if crossover patterning is mediated by mechanical forces along prophase chromosomes. Intensity levels of domainal triad components are further modulated, dynamically, by the conserved protein remodeler Pch2/TRIP13.

Uncovering a multitude of stage-specific splice variants and putative protein isoforms generated along mouse spermatogenesis

BACKGROUND

Mammalian testis is a highly complex and heterogeneous tissue. This complexity, which mostly derives from spermatogenic cells, is reflected at the transcriptional level, with the largest number of tissue-specific genes and long noncoding RNAs (lncRNAs) compared to other tissues, and one of the highest rates of alternative splicing. Although it is known that adequate alternative-splicing patterns and stage-specific isoforms are critical for successful spermatogenesis, so far only a very limited number of reports have addressed a detailed study of alternative splicing and isoforms along the different spermatogenic stages.

RESULTS

In the present work, using highly purified stage-specific testicular cell populations, we detected 33,002 transcripts expressed throughout mouse spermatogenesis not annotated so far. These include both splice variants of already annotated genes, and of hitherto unannotated genes. Using conservative criteria, we uncovered 13,471 spermatogenic lncRNAs, which reflects the still incomplete annotation of lncRNAs. A distinctive feature of lncRNAs was their lower number of splice variants compared to protein-coding ones, adding to the conclusion that lncRNAs are, in general, less complex than mRNAs. Besides, we identified 2,794 unannotated transcripts with high coding potential (including some arising from yet unannotated genes), many of which encode unnoticed putative testis-specific proteins. Some of the most interesting coding splice variants were chosen, and validated through RT-PCR. Remarkably, the largest number of stage-specific unannotated transcripts are expressed during early meiotic prophase stages, whose study has been scarcely addressed in former transcriptomic analyses.

CONCLUSIONS

We detected a high number of yet unannotated genes and alternatively spliced transcripts along mouse spermatogenesis, hence showing that the transcriptomic diversity of the testis is considerably higher than previously reported. This is especially prominent for specific, underrepresented stages such as those of early meiotic prophase, and its unveiling may constitute a step towards the understanding of their key events.

The DNA helicase FANCJ (BRIP1) functions in double strand break repair processing, but not crossover formation during prophase I of meiosis in male mice

Meiotic recombination between homologous chromosomes is initiated by the formation of hundreds of programmed double-strand breaks (DSBs). Approximately 10% of these DSBs result in crossovers (COs), sites of physical DNA exchange between homologs that are critical to correct chromosome segregation. Virtually all COs are formed by coordinated efforts of the MSH4/MSH5 and MLH1/MLH3 heterodimers, the latter representing the defining marks of CO sites. The regulation of CO number and position is poorly understood, but undoubtedly requires the coordinated action of multiple repair pathways. In a previous report, we found gene-trap disruption of the DNA helicase, FANCJ (BRIP1/BACH1), elicited elevated numbers of MLH1 foci and chiasmata. In somatic cells, FANCJ interacts with numerous DNA repair proteins including MLH1, and we hypothesized that FANCJ functions with MLH1 to regulate the major CO pathway. To further elucidate the meiotic function of FANCJ, we produced three new Fancj mutant mouse lines via CRISPR/Cas9 gene editing: a full-gene deletion, truncation of the N-terminal Helicase domain, and a C-terminal dual-tagged allele. We also generated an antibody against the C-terminus of the mouse FANCJ protein. Surprisingly, none of our Fancj mutants show any change in either MLH1 focus counts during pachynema or total CO number at diakinesis of prophase I. We find evidence that FANCJ and MLH1 do not interact in meiosis; further, FANCJ does not co-localize with MSH4, MLH1, or MLH3 in meiosis. Instead, FANCJ co-localizes with BRCA1 and TOPBP1, forming discrete foci along the chromosome cores beginning in early meiotic prophase I and densely localized to unsynapsed chromosome axes in late zygonema and to the XY chromosomes in early pachynema. Fancj mutants also exhibit a subtle persistence of DSBs in pachynema. Collectively, these data indicate a role for FANCJ in early DSB repair, but they rule out a role for FANCJ in MLH1-mediated CO events.

The DNA helicase FANCJ (BRIP1) functions in Double Strand Break repair processing, but not crossover formation during Prophase I of meiosis in male mice

During meiotic prophase I, recombination between homologous parental chromosomes is initiated by the formation of hundreds of programmed double-strand breaks (DSBs), each of which must be repaired with absolute fidelity to ensure genome stability of the germline. One outcome of these DSB events is the formation of Crossovers (COs), the sites of physical DNA exchange between homologs that are critical to ensure the correct segregation of parental chromosomes. However, COs account for only a small (~10%) proportion of all DSB repair events; the remaining 90% are repaired as non-crossovers (NCOs), most by synthesis dependent strand annealing. Virtually all COs are formed by coordinated efforts of the MSH4/MSH5 and MLH1/MLH3 heterodimers. The number and positioning of COs is exquisitely controlled via mechanisms that remain poorly understood, but which undoubtedly require the coordinated action of multiple repair pathways downstream of the initiating DSB. In a previous report we found evidence suggesting that the DNA helicase and Fanconi Anemia repair protein, FANCJ (BRIP1/BACH1), functions to regulate meiotic recombination in mouse. A gene-trap disruption of Fancj showed an elevated number of MLH1 foci and COs. FANCJ is known to interact with numerous DNA repair proteins in somatic cell repair contexts, including MLH1, BLM, BRCA1, and TOPBP1, and we hypothesized that FANCJ regulates CO formation through a direct interaction with MLH1 to suppress the major CO pathway. To further elucidate the function of FANCJ in meiosis, we produced three new Fancj mutant mouse lines via CRISPR/Cas9 gene editing: a full-gene deletion, a mutant line lacking the MLH1 interaction site and the N-terminal region of the Helicase domain, and a C-terminal 6xHIS-HA dual-tagged allele of Fancj. We also generated an antibody against the C-terminus of the mouse FANCJ protein. Surprisingly, while Fanconi-like phenotypes are observed within the somatic cell lineages of the full deletion Fancj line, none of the Fancj mutants show any change in either MLH1 focus counts during pachynema or total CO number at diakinesis of prophase I of meiosis. We find evidence that FANCJ and MLH1 do not interact in meiosis; further, FANCJ does not co-localize with MSH4, MLH1, or MLH3 during late prophase I. Instead, FANCJ forms discrete foci along the chromosome cores beginning in early meiotic prophase I, occasionally co-localizing with MSH4, and then becomes densely localized on unsynapsed chromosome axes in late zygonema and to the XY chromosomes in early pachynema. Strikingly, this localization strongly overlaps with BRCA1 and TOPBP1. Fancj mutants also exhibit a subtle persistence of DSBs in pachynema. Collectively, these data suggest a role for FANCJ in early DSB repair events, and possibly in the formation of NCOs, but they rule out a role for FANCJ in MLH1-mediated CO events. Thus, the role of FANCJ in meiotic cells involves different pathways and different interactors to those described in somatic cell lineages.

Fancm has dual roles in the limiting of meiotic crossovers and germ cell maintenance in mammals

Meiotic crossovers are required for accurate chromosome segregation and producing new allelic combinations. Meiotic crossover numbers are tightly regulated within a narrow range, despite an excess of initiating DNA double-strand breaks. Here, we reveal the tumor suppressor FANCM as a meiotic anti-crossover factor in mammals. We use unique large-scale crossover analyses with both single-gamete sequencing and pedigree-based bulk-sequencing datasets to identify a genome-wide increase in crossover frequencies in Fancm-deficient mice. Gametogenesis is heavily perturbed in Fancm loss-of-function mice, which is consistent with the reproductive defects reported in humans with biallelic FANCM mutations. A portion of the gametogenesis defects can be attributed to the cGAS-STING pathway after birth. Despite the gametogenesis phenotypes in Fancm mutants, both sexes are capable of producing offspring. We propose that the anti-crossover function and role in gametogenesis of Fancm are separable and will inform diagnostic pathways for human genomic instability disorders.

Synaptonemal & CO analyzer: A tool for synaptonemal complex and crossover analysis in immunofluorescence images

During the formation of ova and sperm, homologous chromosomes get physically attached through the synaptonemal complex and exchange DNA at crossover sites by a process known as meiotic recombination. Chromosomes that do not recombine or have anomalous crossover distributions often separate poorly during the subsequent cell division and end up in abnormal numbers in ova or sperm, which can lead to miscarriage or developmental defects. Crossover numbers and distribution along the synaptonemal complex can be visualized by immunofluorescent microscopy. However, manual analysis of large numbers of cells is very time-consuming and a major bottleneck for recombination studies. Some image analysis tools have been created to overcome this situation, but they are not readily available, do not provide synaptonemal complex data, or do not tackle common experimental difficulties, such as overlapping chromosomes. To overcome these limitations, we have created and validated an open-source ImageJ macro routine that facilitates and speeds up the crossover and synaptonemal complex analyses in mouse chromosome spreads, as well as in other vertebrate species. It is free, easy to use and fulfills the recommendations for enhancing rigor and reproducibility in biomedical studies.

A Homozygous Loss-of-Function Mutation in MSH5 Abolishes MutSγ Axial Loading and Causes Meiotic Arrest in NOA-Affected Individuals

Non-obstructive azoospermia (NOA), characterized by spermatogenesis failure and the absence of sperm in ejaculation, is the most severe form of male infertility. However, the etiology and pathology between meiosis-associated monogenic alterations and human NOA remain largely unknown. A homozygous MSH5 mutation (c.1126del) was identified from two idiopathic NOA patients in the consanguineous family. This mutation led to the degradation of MSH5 mRNA and abolished chromosome axial localization of MutSγ in spermatocytes from the affected males. Chromosomal spreading analysis of the patient's meiotic prophase I revealed that the meiosis progression was arrested at a zygotene-like stage with extensive failure of homologous synapsis and DSB repair. Therefore, our study demonstrates that the MSH5 c.1126del could cause meiotic recombination failure and lead to human infertility, improving the genetic diagnosis of NOA clinically. Furthermore, the study of human spermatocytes elucidates the meiosis defects caused by MSH5 variant, and reveals a conserved and indispensable role of MutSγ in human synapsis and meiotic recombination, which have not previously been well-described.

Robust designation of meiotic crossover sites by CDK-2 through phosphorylation of the MutSγ complex

Crossover formation is essential for proper segregation of homologous chromosomes during meiosis. Here, we show that Caenorhabditis elegans cyclin-dependent kinase 2 (CDK-2) partners with cyclin-like protein COSA-1 to promote crossover formation by promoting conversion of meiotic double-strand breaks into crossover–specific recombination intermediates. Further, we identify MutSγ component MSH-5 as a CDK-2 phosphorylation target. MSH-5 has a disordered C-terminal tail that contains 13 potential CDK phosphosites and is required to concentrate crossover–promoting proteins at recombination sites. Phosphorylation of the MSH-5 tail appears dispensable in a wild-type background, but when MutSγ activity is partially compromised, crossover formation and retention of COSA-1 at recombination sites are exquisitely sensitive to phosphosite loss. Our data support a model in which robustness of crossover designation reflects a positive feedback mechanism involving CDK-2–mediated phosphorylation and scaffold-like properties of the MSH5 C-terminal tail, features that combine to promote full recruitment and activity of crossover–promoting complexes.

Diet effects on mouse meiotic recombination: a warning for recombination studies

Meiotic recombination is a critical process for sexually reproducing organisms. This exchange of genetic information between homologous chromosomes during meiosis is important not only because it generates genetic diversity, but also because it is often required for proper chromosome segregation. Consequently, the frequency and distribution of crossovers are tightly controlled to ensure fertility and offspring viability. However, in many systems, it has been shown that environmental factors can alter the frequency of crossover events. Two studies in flies and yeast point to nutritional status affecting the frequency of crossing over. However, this question remains unexplored in mammals. Here, we test how crossover frequency varies in response to diet in Mus musculus males. We use immunohistochemistry to estimate crossover frequency in multiple genotypes under two diet treatments. Our results indicate that while crossover frequency was unaffected by diet in some strains, other strains were sensitive even to small composition changes between two common laboratory chows. Therefore, recombination is both resistant and sensitive to certain dietary changes in a strain-dependent manner and, hence, this response is genetically determined. Our study is the first to report a nutrition effect on genome-wide levels of recombination. Moreover, our work highlights the importance of controlling diet in recombination studies and may point to diet as a potential source of variability among studies, which is relevant for reproducibility.

Cyclin N-Terminal Domain-Containing-1 Coordinates Meiotic Crossover Formation with Cell-Cycle Progression in a Cyclin-Independent Manner

During mammalian meiotic prophase I, programmed DNA double-strand breaks are repaired by non-crossover or crossover events, the latter predominantly occurring via the class I crossover pathway and requiring the cyclin N-terminal domain-containing 1(CNTD1) protein. Using an epitope-tagged Cntd1 allele, we detect a short isoform of CNTD1 in vivo that lacks a predicted N-terminal cyclin domain and does not bind cyclin-dependent kinases. Instead, we find that the short-form CNTD1 variant associates with components of the replication factor C (RFC) machinery to facilitate crossover formation, and with the E2 ubiquitin conjugating enzyme, CDC34, to regulate ubiquitylation and subsequent degradation of the WEE1 kinase, thereby modulating cell-cycle progression. We propose that these interactions facilitate a role for CNTD1 as a stop-go regulator during prophase I, ensuring accurate and complete crossover formation before allowing metaphase progression and the first meiotic division.

•

CNTD1 associates with sites of crossing over in meiosis, co-localizing with MutLγ

•

In the testis, CNTD1 does not interact with CDKs or with known crossover regulators

•

CNTD1 regulates crossing over via interactions with the replication factor C complex

•

CNTD1 regulates cell-cycle progression via interactions with the SCF complex

Interruption of an MSH4 homolog blocks meiosis in metaphase I and eliminates spore formation in Pleurotus ostreatus

Pleurotus ostreatus, one of the most widely cultivated edible mushrooms, produces high numbers of spores causing severe respiratory health problems for people, clogging of filters and spoilage of produce. A non-sporulating commercial variety (SPOPPO) has been successfully introduced into the market in 2006. This variety was generated by introgression breeding of a natural mutation into a commercial variety. Our cytological studies revealed that meiosis in the natural and derived sporeless strains was blocked in metaphase I, apparently resulting in a loss of spore formation. The gene(s) underlying this phenotype were mapped to an 80 kb region strongly linked to sporelessness and identified by transformation of wild type genes of this region into a sporeless strain. Sporulation was restored by re-introduction of the DNA sequence encoding the P. ostreatus meiotic recombination gene MSH4 homolog (poMSH4). Subsequent molecular analysis showed that poMSH4 in the sporeless P. ostreatus was interrupted by a DNA fragment containing a region encoding a CxC5/CxC6 cysteine cluster associated with Copia-type retrotransposons. The block of meiosis in metaphase I by a poMSH4 null mutant suggests that this protein plays an essential role in both Class I and II crossovers in mushrooms, similar to animals (mice), but unlike in plants. MSH4 was previously shown to be a target for breeding of sporeless varieties in P. pulmonarius, and the null mutant of the MSH4 homolog of S. commune (scMSH4) confers an extremely low level of spore formation. We propose that MSH4 homologs are likely to be a breeding target for sporeless strains both within Pleurotus sp. and in other Agaricales.

MutS homologue 4 and MutS homologue 5 Maintain the Obligate Crossover in Wheat Despite Stepwise Gene Loss following Polyploidization

Crossovers (COs) ensure accurate chromosome segregation during meiosis while creating novel allelic combinations. Here, we show that allotetraploid (AABB) durum wheat (Triticum turgidum ssp. durum) utilizes two pathways of meiotic recombination. The class I pathway requires MSH4 and MSH5 (MutSγ) to maintain the obligate CO/chiasma and accounts for ∼85% of meiotic COs, whereas the residual ∼15% are consistent with the class II CO pathway. Class I and class II chiasmata are skewed toward the chromosome ends, but class II chiasmata are significantly more distal than class I chiasmata. Chiasma distribution does not reflect the abundance of double-strand breaks, detected by proxy as RAD51 foci at leptotene, but only ∼2.3% of these sites mature into chiasmata. MutSγ maintains the obligate chiasma despite a 5.4-kb deletion in MSH5B rendering it nonfunctional, which occurred early in the evolution of tetraploid wheat and was then domesticated into hexaploid (AABBDD) common wheat (Triticum aestivum), as well as an 8-kb deletion in MSH4D in hexaploid wheat, predicted to create a nonfunctional pseudogene. Stepwise loss of MSH5B and MSH4D following hybridization and whole-genome duplication may have occurred due to gene redundancy (as functional copies of MSH5A, MSH4A, and MSH4B are still present in the tetraploid and MSH5A, MSH5D, MSH4A, and MSH4B are present in the hexaploid) or as an adaptation to modulate recombination in allopolyploid wheat.

Regulated Proteolysis of MutSγ Controls Meiotic Crossing Over

Crossover recombination is essential for accurate chromosome segregation during meiosis. The MutSγ complex, Msh4-Msh5, facilitates crossing over by binding and stabilizing nascent recombination intermediates. We show that these activities are governed by regulated proteolysis. MutSγ is initially inactive for crossing over due to an N-terminal degron on Msh4 that renders it unstable by directly targeting proteasomal degradation. Activation of MutSγ requires the Dbf4-dependent kinase Cdc7 (DDK), which directly phosphorylates and thereby neutralizes the Msh4 degron. Genetic requirements for Msh4 phosphorylation indicate that DDK targets MutSγ only after it has bound to nascent joint molecules (JMs) in the context of synapsing chromosomes. Overexpression studies confirm that the steady-state level of Msh4, not phosphorylation per se, is the critical determinant for crossing over. At the DNA level, Msh4 phosphorylation enables the formation and crossover-biased resolution of double-Holliday Junction intermediates. Our study establishes regulated protein degradation as a fundamental mechanism underlying meiotic crossing over.

A mutation in the endonuclease domain of mouse MLH3 reveals novel roles for MutLγ during crossover formation in meiotic prophase I

During meiotic prophase I, double-strand breaks (DSBs) initiate homologous recombination leading to non-crossovers (NCOs) and crossovers (COs). In mouse, 10% of DSBs are designated to become COs, primarily through a pathway dependent on the MLH1-MLH3 heterodimer (MutLγ). Mlh3 contains an endonuclease domain that is critical for resolving COs in yeast. We generated a mouse (Mlh3DN/DN) harboring a mutation within this conserved domain that is predicted to generate a protein that is catalytically inert. Mlh3DN/DN males, like fully null Mlh3-/- males, have no spermatozoa and are infertile, yet spermatocytes have grossly normal DSBs and synapsis events in early prophase I. Unlike Mlh3-/- males, mutation of the endonuclease domain within MLH3 permits normal loading and frequency of MutLγ in pachynema. However, key DSB repair factors (RAD51) and mediators of CO pathway choice (BLM helicase) persist into pachynema in Mlh3DN/DN males, indicating a temporal delay in repair events and revealing a mechanism by which alternative DSB repair pathways may be selected. While Mlh3DN/DN spermatocytes retain only 22% of wildtype chiasmata counts, this frequency is greater than observed in Mlh3-/- males (10%), suggesting that the allele may permit partial endonuclease activity, or that other pathways can generate COs from these MutLγ-defined repair intermediates in Mlh3DN/DN males. Double mutant mice homozygous for the Mlh3DN/DN and Mus81-/- mutations show losses in chiasmata close to those observed in Mlh3-/- males, indicating that the MUS81-EME1-regulated crossover pathway can only partially account for the increased residual chiasmata in Mlh3DN/DN spermatocytes. Our data demonstrate that mouse spermatocytes bearing the MLH1-MLH3DN/DN complex display the proper loading of factors essential for CO resolution (MutSγ, CDK2, HEI10, MutLγ). Despite these functions, mice bearing the Mlh3DN/DN allele show defects in the repair of meiotic recombination intermediates and a loss of most chiasmata.