Sexual dimorphism in the meiotic requirement for PRDM9: A mammalian evolutionary safeguard

Single-cell and bulk transcriptional profiling of mouse ovaries reveals novel genes and pathways associated with DNA damage response in oocytes

Immature oocytes enclosed in primordial follicles stored in female ovaries are under constant threat of DNA damage induced by endogenous and exogenous factors. Checkpoint kinase 2 (CHEK2) is a key mediator of the DNA damage response (DDR) in all cells. Genetic studies have shown that CHEK2 and its downstream targets, p53, and TAp63, regulate primordial follicle elimination in response to DNA damage. However, the mechanism leading to their demise is still poorly characterized. Single-cell and bulk RNA sequencing were used to determine the DDR in wild-type and Chek2-deficient ovaries. A low but oocyte-lethal dose of ionizing radiation induces ovarian DDR that is solely dependent on CHEK2. DNA damage activates multiple response pathways related to apoptosis, p53, interferon signaling, inflammation, cell adhesion, and intercellular communication. These pathways are differentially employed by different ovarian cell types, with oocytes disproportionately affected by radiation. Novel genes and pathways are induced by radiation specifically in oocytes, shedding light on their sensitivity to DNA damage, and implicating a coordinated response between oocytes and pregranulosa cells within the follicle. These findings provide a foundation for future studies on the specific mechanisms regulating oocyte survival in the context of aging, therapeutic and environmental genotoxic exposures.

Checkpoint kinase 2 controls insulin secretion and glucose homeostasis

After the discovery of insulin, a century ago, extensive work has been done to unravel the molecular network regulating insulin secretion. Here we performed a chemical screen and identified AZD7762, a compound that potentiates glucose-stimulated insulin secretion (GSIS) of a human β cell line, healthy and type 2 diabetic (T2D) human islets and primary cynomolgus macaque islets. In vivo studies in diabetic mouse models and cynomolgus macaques demonstrated that AZD7762 enhances GSIS and improves glucose tolerance. Furthermore, genetic manipulation confirmed that ablation of CHEK2 in human β cells results in increased insulin secretion. Consistently, high-fat-diet-fed Chk2-/- mice show elevated insulin secretion and improved glucose clearance. Finally, untargeted metabolic profiling demonstrated the key role of the CHEK2-PP2A-PLK1-G6PD-PPP pathway in insulin secretion. This study successfully identifies a previously unknown insulin secretion regulating pathway that is conserved across rodents, cynomolgus macaques and human β cells in both healthy and T2D conditions.

Meiosis through three centuries

Meiosis is the specialized cellular program that underlies gamete formation for sexual reproduction. It is therefore not only interesting but also a fundamentally important subject for investigation. An especially attractive feature of this program is that many of the processes of special interest involve organized chromosomes, thus providing the possibility to see chromosomes "in action". Analysis of meiosis has also proven to be useful in discovering and understanding processes that are universal to all chromosomal programs. Here we provide an overview of the different historical moments when the gap between observation and understanding of mechanisms and/or roles for the new discovered molecules was bridged. This review reflects also the synergy of thinking and discussion among our three laboratories during the past several decades.

Patterns of recombination in snakes reveal a tug-of-war between PRDM9 and promoter-like features

In some mammals, notably humans, recombination occurs almost exclusively where the protein PRDM9 binds, whereas in vertebrates lacking an intact PRDM9, such as birds and canids, recombination rates are elevated near promoter-like features. To determine whether PRDM9 directs recombination in nonmammalian vertebrates, we focused on an exemplar species with a single, intact PRDM9 ortholog, the corn snake (Pantherophis guttatus). Analyzing historical recombination rates along the genome and crossovers in pedigrees, we found evidence that PRDM9 specifies the location of recombination events, but we also detected a separable effect of promoter-like features. These findings reveal that the uses of PRDM9 and promoter-like features need not be mutually exclusive and instead reflect a tug-of-war that is more even in some species than others.

Single-cell and bulk transcriptional profiling of mouse ovaries reveals novel genes and pathways associated with DNA damage response in oocytes

Immature oocytes enclosed in primordial follicles stored in female ovaries are under constant threat of DNA damage induced by endogenous and exogenous factors. Checkpoint kinase 2 (CHEK2) is a key mediator of the DNA damage response in all cells. Genetic studies have shown that CHEK2 and its downstream targets, p53 and TAp63, regulate primordial follicle elimination in response to DNA damage, however the mechanism leading to their demise is still poorly characterized. Single-cell and bulk RNA sequencing were used to determine the DNA damage response in wildtype and Chek2-deficient ovaries. A low but oocyte-lethal dose of ionizing radiation induces a DNA damage response in ovarian cells that is solely dependent on CHEK2. DNA damage activates multiple ovarian response pathways related to apoptosis, p53, interferon signaling, inflammation, cell adhesion, and intercellular communication. These pathways are differentially employed by different ovarian cell types, with oocytes disproportionately affected by radiation. Novel genes and pathways are induced by radiation specifically in oocytes, shedding light on their sensitivity to DNA damage, and implicating a coordinated response between oocytes and pre-granulosa cells within the follicle. These findings provide a foundation for future studies on the specific mechanisms regulating oocyte survival in the context of aging, as well as therapeutic and environmental genotoxic exposures.

The Recombination Hotspot Paradox: Co-evolution between PRDM9 and its target sites

Recombination often concentrates in small regions called recombination hotspots where recombination is much higher than the genome's average. In many vertebrates, including humans, gene PRDM9 specifies which DNA motifs will be the target for breaks that initiate recombination, ultimately determining the location of recombination hotspots. Because the sequence that breaks (allowing recombination) is converted into the sequence that does not break (preventing recombination), the latter sequence is over-transmitted to future generations and recombination hotspots are self-destructive. Given their self-destructive nature, recombination hotspots should eventually become extinct in genomes where they are found. While empirical evidence shows that individual hotspots do become inactive over time (die), hotspots are abundant in many vertebrates: a contradiction called the Recombination Hotspot Paradox. What saves recombination hotspots from their foretold extinction? Here we formulate a co-evolutionary model of the interaction among sequence-specific gene conversion, fertility selection, and recurrent mutation. We find that allelic frequencies oscillate leading to stable limit cycles. From a biological perspective this means that when fertility selection is weaker than gene conversion, it cannot stop individual hotspots from dying but can save them from extinction by driving their re-activation (resuscitation). In our model, mutation balances death and resuscitation of hotspots, thus maintaining their number over evolutionary time. Interestingly, we find that multiple alleles result in oscillations that are chaotic and multiple targets in oscillations that are asynchronous between targets thus helping to maintain the average genomic recombination probability constant. Furthermore, we find that the level of expression of PRDM9 should control for the fraction of targets that are hotspots and the overall temperature of the genome. Therefore, our co-evolutionary model improves our understanding of how hotspots may be replaced, thus contributing to solve the Recombination Hotspot Paradox. From a more applied perspective our work provides testable predictions regarding the relation between mutation probability and fertility selection with life expectancy of hotspots.

Hypoxia Inducible Factors (HIF1α and HIF3α) are differentially methylated in preeclampsia placentae and are associated with birth outcomes

Preeclampsia is a placental vascular pathology and hypoxia is known to influence placental angiogenesis. Hypoxia Inducible Factors (HIF1α and HIF3α) mediate the response to cellular oxygen concentration and bind to hypoxia response element of target genes. However the mechanism regulating above activity is not well-understood. We investigated if placental DNA methylation (DNAm) and expression of HIF1α and 3α genes are altered and associated with pre-eclampsia, placental weight and birth outcomes. Using a cohort comprising women with preeclampsia [N = 100, delivering at term (N = 43) and preterm (N = 57)] and normotensive controls (N = 100), we analysed DNAm in HIF1α and 3α, and their mRNA expression in placentae, employing pyrosequencing and quantitative real-time PCR, respectively. We observed significant hypermethylation at cg22891070 of HIF3α in preeclampsia placentae compared to controls (β = 1.5%, p = 0.04). CpG8 in the promoter region of HIF1α, showed marginally significant hypomethylation in preterm preeclampsia compared to controls (β = - 0.15%, p = 0.055). HIF1α expression was significantly lower in preterm preeclampsia compared to controls (mean ± SE = 10.16 ± 2.00 vs 4.25 ± 0.90, p = 0.04). Further, DNAm in HIF1α promoter region was negatively associated with its expression levels (β =  - 0.165, p = 0.024). Several CpGs in HIF1α were negatively associated with placental weight and birth outcomes including birth weight (β range =  - 0.224-0.300) and birth length [β range =  - 0.248 to - 0.301 (p < 0.05 for all)]. Overall, we demonstrate altered DNAm in HIF1α and HIF3α in preeclampsia placentae, also associated with various birth outcomes. Correlation of DNAm in HIF1α and its expression suggests a possible role in the pathogenesis of pre-eclampsia. Further investigations on interactions between HIF1α and HIF3α in preeclampsia would be interesting.

Reproductive genomics of the mouse: implications for human fertility and infertility

Genetic analyses of mammalian gametogenesis and fertility have the potential to inform about two important and interrelated clinical areas: infertility and contraception. Here, we address the genetics and genomics underlying gamete formation, productivity and function in the context of reproductive success in mammalian systems, primarily mouse and human. Although much is known about the specific genes and proteins required for meiotic processes and sperm function, we know relatively little about other gametic determinants of overall fertility, such as regulation of gamete numbers, duration of gamete production, and gamete selection and function in fertilization. As fertility is not a binary trait, attention is now appropriately focused on the oligogenic, quantitative aspects of reproduction. Multiparent mouse populations, created by complex crossing strategies, exhibit genetic diversity similar to human populations and will be valuable resources for genetic discovery, helping to overcome current limitations to our knowledge of mammalian reproductive genetics. Finally, we discuss how what we know about the genomics of reproduction can ultimately be brought to the clinic, informing our concepts of human fertility and infertility, and improving assisted reproductive technologies.

Using the Collaborative Cross and Diversity Outbred Mice in Immunology

The Collaborative Cross (CC) and the Diversity Outbred (DO) stock mouse panels are the most powerful murine genetics tools available to the genetics community. Together, they combine the strength of inbred animal models with the diversity of outbred populations. Using the 63 CC strains or a panel of DO mice, each derived from the same 8 parental mouse strains, researchers can map genetic contributions to exceptionally complex immunological and infectious disease traits that would require far greater powering if performed by genome-wide association studies (GWAS) in human populations. These tools allow genes to be studied in heterozygous and homozygous states and provide a platform to study epistasis between interacting loci. Most importantly, once a quantitative phenotype is investigated and quantitative trait loci are identified, confirmatory genetic studies can be performed, which is often problematic using the GWAS approach. In addition, novel stable mouse models for immune phenotypes are often derived from studies utilizing the DO and CC mice that can serve as stronger model systems than existing ones in the field. The CC/DO systems have contributed to the fields of cancer immunology, autoimmunity, vaccinology, infectious disease, allergy, tissue rejection, and tolerance but have thus far been greatly underutilized. In this article, we present a recent review of the field and point out key areas of immunology that are ripe for further investigation and awaiting new CC/DO research projects. We also highlight some of the strong computational tools that have been developed for analyzing CC/DO genetic and phenotypic data. Additionally, we have formed a centralized community on the CyVerse infrastructure where immunogeneticists can utilize those software tools, collaborate with groups across the world, and expand the use of the CC and DO systems for investigating immunogenetic phenomena. © 2022 Wiley Periodicals LLC.

Prdm9 deficiency of rat oocytes causes synapsis among non-homologous chromosomes and aneuploidy

Aneuploidy (abnormal chromosome number) accompanies reduced ovarian function in humans and mice, but the reasons behind this concomitance remain underexplored. Some variants in the human gene encoding histone-3-lysine-4,36-trimethyltransferase PRDM9 are associated with aneuploidy, and other variants with ovarian function reduced by premature ovarian failure (POF), but no link between POF and aneuploidy has been revealed. SHR/OlaIpcv rat females lacking PRDM9 manifest POF-a reduced follicle number, litter size, and reproductive age. Here, we explored this model to test how POF relates to oocyte euploidy. The mutant rat females displayed increased oocyte aneuploidy and embryonic death of their offspring compared to controls. Because rat PRDM9 positions meiotic DNA breaks, we investigated the repair of these breaks. Fertile control rodents carry pachytene oocytes with synapsed homologous chromosomes and repaired breaks, while sterile Prdm9-deficient mice carry pachytene-like oocytes with many persisting breaks and asynapsed chromosomes. However, most PRDM9-lacking rat oocytes displayed a few persisting breaks and non-homologous synapsis (NHS). HORMAD2 protein serves as a barrier to sister-chromatid repair and a signal for the synapsis and DNA repair checkpoints. NHS but not asynapsis was associated with HORMAD2 levels similar to the levels on rat pachytene chromosomes with homologous synapsis. NHS was accompanied by crossing-over decreased below the minimum that is essential for euploidy. We argue that the increased mutant rat aneuploidy is due to NHS, which allows some oocytes to pass meiotic checkpoints without one crossing-over per chromosomal pair, leading to segregation errors, and thereby NHS links POF to aneuploidy.

Differential effects of two catalytic mutations on full-length PRDM9 and its isolated PR/SET domain reveal a case of pseudomodularity

PRDM9 is a DNA-binding histone methyltransferase that designates and activates recombination hotspots in mammals by locally trimethylating lysines 4 and 36 of histone H3. In mice, we recently reported two independently produced point mutations at the same residue, Glu360Pro (Prdm9EP) and Glu360Lys (Prdm9EK), which severely reduce its H3K4 and H3K36 methyltransferase activities in vivo. Prdm9EP is slightly less hypomorphic than Prdm9EK, but both mutations reduce both the number and amplitude of PRDM9-dependent H3K4me3 and H3K36me3 peaks in spermatocytes. While both mutations cause infertility with complete meiotic arrest in males, Prdm9EP, but not Prdm9EK, is compatible with some female fertility. When we tested the effects of these mutations in vitro, both Prdm9EP and Prdm9EK abolished H3K4 and H3K36 methyltransferase activity in full-length PRDM9. However, in the isolated PRDM9 PR/SET domain, these mutations selectively compromised H3K36 methyltransferase activity, while leaving H3K4 methyltransferase activity intact. The difference in these effects on the PR/SET domain vs the full-length protein shows that PRDM9 is not an intrinsically modular enzyme; its catalytic domain is influenced by its tertiary structure and possibly by its interactions with DNA and other proteins in vivo. These two informative mutations illuminate the enzymatic chemistry of PRDM9, and potentially of PR/SET domains in general, reveal the minimal threshold of PRDM9-dependent catalytic activity for female fertility, and potentially have some practical utility for genetic mapping and genomics.

Higher Intercellular Variation in Genome-Wide Recombination Rate in Female Mice

Meiotic recombination affects fertility, shuffles genomes, and modulates the effectiveness of natural selection. Despite conservation of the recombination pathway, the rate of recombination varies among individuals and along chromosomes. Recombination rate also differs among cells from the same organism, but this form of variation has received less attention. To identify patterns that characterize intercellular variation in the genome-wide recombination rate, we counted foci of the MLH1 recombination-associated protein in oocytes and spermatocytes from a panel of wild-derived inbred strains of house mice. Females show higher intercellular variation in MLH1 focus count than males from the same inbred strains. This pattern is consistent across strains from multiple subspecies, including 2 strains in which the average MLH1 focus count is higher in males. The sex difference in genome-wide recombination rate we report suggests that selection targeting recombination rate will be more efficient in males than in females.

Pathogenic variants of meiotic double strand break (DSB) formation genes PRDM9 and ANKRD31 in premature ovarian insufficiency

Purpose

The etiology of premature ovarian insufficiency (POI) is heterogeneous, and genetic factors account for 20–25% of the patients. The primordial follicle pool is determined by the meiosis process, which is initiated by programmed DNA double strand breaks (DSB) and homologous recombination. The objective of the study is to explore the role of DSB formation genes in POI pathogenesis.

Methods

Variants in DSB formation genes were analyzed from a database of exome sequencing in 1,030 patients with POI. The pathogenic effects of the potentially causative variants were verified by further functional studies.

Results

Three pathogenic heterozygous variants in PRDM9 and two in ANKRD31 were identified in seven patients. Functional studies showed the variants in PRDM9 impaired its methyltransferase activity, and the ANKRD31 variations disturbed its interaction with another DSB formation factor REC114 by haploinsufficiency effect, indicating the pathogenic effects of the two genes on ovarian function were dosage dependent.

Conclusion

Our study identified pathogenic variants of PRDM9 and ANKRD31 in POI patients, shedding new light on the contribution of meiotic DSB formation genes in ovarian development, further expanding the genetic architecture of POI.