Characterization and therapy of fertilization failure in murine and human models with HNRNPR mutations

TL;DR

This study identifies HNRNPR mutations as a cause of male infertility and shows that treatments like artificial oocyte activation can restore fertilization.

Contribution

HNRNPR mutations are newly linked to infertility, with a novel mechanism involving PLCζ regulation and therapeutic interventions.

Findings

HNRNPR mutations disrupt PLCζ expression and oocyte activation in both human and mouse models.

Artificial oocyte activation and NusA-PLCζ treatment effectively restore fertilization in HNRNPR-mutant models.

Loss of hnRNPR function leads to abnormal calcium oscillations in oocytes following ICSI.

Abstract

Oocyte activation is essential for successful fertilization and subsequent embryonic development. However, only a few disease-causing genes have been associated with sperm-derived oocyte activation failure, and the underlying molecular mechanisms and therapeutic approaches remain largely unknown. Here, we identified pathogenic mutations in HNRNPR from three infertile patients whose partners repeatedly failed to achieve transferable embryos despite undergoing both in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI). Remarkably, artificial oocyte activation (AOA, Srcl₂) combined with ICSI successfully restored fertilization. Whole-exome sequencing revealed HNRNPR mutations shared among affected families. To establish causality, we generated a knock-in mouse model, in which males exhibited phenotypes consistent with those observed in patients. Mechanistically, ICSI with sperm from Hnrnpr-mutated mice was unable to induce normal calcium oscillations in oocytes, while spermatozoa from both humans and mice exhibited reduced expression and mislocalization of phospholipase C zeta (PLCζ). Further analyses demonstrated that hnRNPR regulates Plcz1 splicing in an m6A-dependent manner. Beyond Srcl₂ treatment, we also developed NusA-PLCζ to effectively restore oocyte activation. Collectively, these findings reveal a previously unrecognized molecular mechanism by which HNRNPR mutations cause sperm-borne oocyte activation failure and male infertility, while highlighting targeted therapeutic strategies to restore fertilization.

Pathogenic HNRNPR mutations were identified as a novel cause of sperm-borne oocyte activation failure. Mechanistic and therapeutic insights reveal RNA splicing-dependent regulation of PLCζ and potential clinical interventions.

Pathogenic HNRNPR mutations were discovered in infertile men whose partners repeatedly failed to achieve fertilization after IVF and ICSI.A knock-in mouse model carrying the human Hnrnpr mutation recapitulated sperm-borne oocyte activation failure.Disrupted Plcz1 splicing and reduced PLCζ expression in sperm were observed due to loss of hnRNPR function.Aberrant calcium oscillations in oocytes were demonstrated following ICSI with Hnrnpr-mutant sperm.Fertilization was effectively restored by artificial oocyte activation (Srcl2) or treatment with recombinant NusA-PLCζ protein.

Pathogenic HNRNPR mutations were discovered in infertile men whose partners repeatedly failed to achieve fertilization after IVF and ICSI.

A knock-in mouse model carrying the human Hnrnpr mutation recapitulated sperm-borne oocyte activation failure.

Disrupted Plcz1 splicing and reduced PLCζ expression in sperm were observed due to loss of hnRNPR function.

Aberrant calcium oscillations in oocytes were demonstrated following ICSI with Hnrnpr-mutant sperm.

Fertilization was effectively restored by artificial oocyte activation (Srcl2) or treatment with recombinant NusA-PLCζ protein.

Pathogenic HNRNPR mutations were identified as a novel cause of sperm-borne oocyte activation failure. Mechanistic and therapeutic insights reveal RNA splicing-dependent regulation of PLCζ and potential clinical interventions.

Figures

Genetic mutation of *HNRNPR* leads to sperm-borne oocyte activation failure.(A) Flowchart illustrating the clinical process of sperm-borne oocyte activation failure and the assisted oocyte activation (AOA) treatment protocol. (B) Summary of assisted reproductive outcomes, including in vitro fertilization (IVF), intracytoplasmic sperm injection (ICSI), and ICSI combined with AOA, from three clinical patients. (C) Pedigree analysis of families carrying *HNRNPR* variants associated with male infertility, identified through whole-exome sequencing. Pink arrows indicate affected males, and a double

Mutations in sperm *Hnrnpr* cause sperm-borne oocyte activation failure in mice.(A) In vitro fertilization (IVF) of sperm from wild-type (WT) and knock-in (KI) mice. Scale bar = 40 μm. (B) Quantification of the percentage of viable zygotes following IVF. Data are presented as mean ± s.d.; statistical significance was determined using a two-sided Student’s *t* test. (C, D) Schematic diagrams illustrating the experimental procedures for intracytoplasmic sperm injection (ICSI) and ICSI combined with assisted oocyte activation (ICSI + AOA). (E) Differential interference contrast (DIC) images showi

Mutated *Hnrnpr* sperm induce disordered calcium oscillations in oocytes following ICSI.(A) Circular diagram illustrating genes associated with oocyte activation failure and their molecular mechanisms. Five sperm-borne and twenty-six oocyte-borne genes are listed as pathogenic contributors. (B) Fluorescence imaging of calcium signals in oocytes using the Fura-2 probe after ICSI with sperm from wild-type (WT) and knock-in (KI) male mice. The diffusible Fura-2 AM ester probe predominantly localizes to the cytoplasm after entering the cell. Scale bar = 10 μm. (C) Relative quantification of calciu

Genetic mutation of *HNRNPR* leads to sperm-borne oocyte activation failure.(A) Flowchart illustrating the clinical process of sperm-borne oocyte activation failure and the assisted oocyte activation (AOA) treatment protocol. (B) Summary of assisted reproductive outcomes, including in vitro fertilization (IVF), intracytoplasmic sperm injection (ICSI), and ICSI combined with AOA, from three clinical patients. (C) Pedigree analysis of families carrying *HNRNPR* variants associated with male infertility, identified through whole-exome sequencing. Pink arrows indicate affected males, and a double

Mutations in sperm *Hnrnpr* cause sperm-borne oocyte activation failure in mice.(A) In vitro fertilization (IVF) of sperm from wild-type (WT) and knock-in (KI) mice. Scale bar = 40 μm. (B) Quantification of the percentage of viable zygotes following IVF. Data are presented as mean ± s.d.; statistical significance was determined using a two-sided Student’s *t* test. (C, D) Schematic diagrams illustrating the experimental procedures for intracytoplasmic sperm injection (ICSI) and ICSI combined with assisted oocyte activation (ICSI + AOA). (E) Differential interference contrast (DIC) images showi

Mutated *Hnrnpr* sperm induce disordered calcium oscillations in oocytes following ICSI.(A) Circular diagram illustrating genes associated with oocyte activation failure and their molecular mechanisms. Five sperm-borne and twenty-six oocyte-borne genes are listed as pathogenic contributors. (B) Fluorescence imaging of calcium signals in oocytes using the Fura-2 probe after ICSI with sperm from wild-type (WT) and knock-in (KI) male mice. The diffusible Fura-2 AM ester probe predominantly localizes to the cytoplasm after entering the cell. Scale bar = 10 μm. (C) Relative quantification of calciu

Abnormal transcriptomic and proteomic profiles in oocytes and sperm.(A) Flowchart illustrating the SMART-Seq experimental workflow performed after ICSI and ICSI combined with AOA. (B) Volcano plot showing differentially expressed genes (DEGs) between WT-ICSI and KI-ICSI groups. *P* values were calculated using a Welch’s *t* test. (C) Volcano plot showing DEGs between KI-ICSI and KI-ICSI + AOA groups. *P* values were calculated using a Welch’s *t* test. (D) Venn diagram depicting the overlap of DEGs identified in (B, C). (E) Gene Ontology (GO) enrichment analysis of the common DEGs identified i

hnRNPR regulates *Plcz1* splicing in an m6A-dependent manner.(A) Summary chart of alternative splicing (AS) events and affected genes across five major categories significantly altered by *Hnrnpr* mutation in testes, based on RNA-seq analysis using rMATS (replicate Multivariate Analysis of Transcript Splicing). (B) Pie charts showing the distribution of AS events and associated genes among different splicing types. (C) RNA-seq read coverage maps of WT and KI mouse testes, with bar charts indicating the proportion of reads mapped to intergenic, intronic, and exonic regions. (D) Gene Ontology (G

The combined treatments can restore the fertilization ability of sperm in mice with *Hnrnpr* mutations.(A, B) Computer-assisted sperm analysis (CASA) was used to evaluate total motility (A) and progressive motility (B) among untreated (WT, KI), mEVs-treated, and mEVs-SKAP2-treated groups after 3-h in vitro co-incubation. Data are presented as mean ± s.d., and statistical significance was determined using a two-sided Student’s *t* test. (C) Quantification of morphologically abnormal sperm (500 spermatozoa per mouse). Results are shown as mean ± s.d., with “NS” indicating no significant differen

Full text

The paper explainedProblemOocyte activation failure (OAF) is a major cause of male-factor infertility, frequently resulting in fertilization failure even after intracytoplasmic sperm injection (ICSI). However, the genetic and molecular mechanisms underlying sperm-borne OAF and its therapeutic strategies remain poorly understood.ResultsPathogenic mutations in HNRNPR were identified in three infertile men whose partners repeatedly failed to produce transferable embryos following IVF and ICSI. A knock-in mouse model carrying the corresponding Hnrnpr mutation reproduced the human infertility phenotype. Mechanistic analyses revealed that Hnrnpr mutations disrupt Plcz1 splicing in an m6A-dependent manner, resulting in reduced expression and mislocalization of PLCζ and consequently abnormal calcium oscillations during oocyte activation. Artificial oocyte activation with SrCl₂ or treatment with a recombinant NusA-PLCζ protein effectively restored fertilization.ImpactThis study identifies HNRNPR as a novel genetic cause of sperm-borne oocyte activation failure and male infertility. It also provides a mechanistic understanding of how RNA splicing defects impair oocyte activation and introduces targeted therapeutic approaches, such as artificial oocyte activation (SrCl_2_) or PLCζ, to enable successful fertilization in affected individuals.

Introduction

Oocyte activation is a fundamental biological process that marks the transition from a mature metaphase II-arrested oocyte to a developmentally competent zygote (Sugita et al, 2024). This event is initiated before fertilization, when sperm entry triggers a cascade of tightly regulated molecular and cellular changes that release the oocyte from meiotic arrest and initiate embryonic development (Shafqat et al, 2022). Central to this process is the generation of repetitive intracellular calcium (Ca²⁺) oscillations, which act as universal signaling cues to orchestrate downstream developmental events (Yeste et al, 2017). These oscillations are driven primarily by the sperm-specific enzyme phospholipase C zeta (PLCζ; gene name Plcz1), which hydrolyzes phosphatidylinositol 4,5-bisphosphate (PIP₂) to produce inositol 1,4,5-trisphosphate (IP₃) (Zhao et al, 2023). The resulting IP₃-mediated Ca²⁺ release from the endoplasmic reticulum triggers a sequence of essential processes, including cortical granule exocytosis to prevent polyspermy, resumption of meiosis, pronuclear formation, and activation of the embryonic genome (Lin et al, 2023). Defects in calcium oscillations and oocyte activation represent a major cause of fertilization failure in assisted reproductive technologies (ART), including IVF and ICSI (Sun and Yeh, 2021). In cases of genetic male infertility, such as mutations in ACTL7A (Xin et al, 2020) and IQCN (Dai et al, 2022), failure to achieve successful fertilization has been directly linked to abnormalities in the expression, subcellular localization, or enzymatic activity of the sperm-specific PLCζ, the principal trigger of repetitive calcium oscillations in the oocyte. Despite its central role, the molecular pathways regulating PLCζ synthesis, localization, and functional activity during spermatogenesis remain incompletely understood.

Emerging evidence indicates that RNA-binding proteins (RBPs) play a critical role in spermatogenesis by modulating post-transcriptional processes (Morgan et al, 2021), such as mRNA stability, alternative splicing, and translational control, thereby influencing PLCζ expression in haploid spermatids and mature spermatozoa. Dysregulation of these RBPs can compromise fertility and human health (Tao et al, 2024), while RBPs regulating production or proper localization of PLCζ, ultimately leading to sperm-borne oocyte activation failure, remain unexplored. Thus, elucidating these post-transcriptional regulatory mechanisms of Plcz1 is essential not only for advancing our understanding of the fundamental biology of oocyte activation but also for informing the development of targeted interventions, such as artificial oocyte activation and RBP-modulating strategies, aimed at rescuing fertilization defects in ART. To overcome sperm-borne oocyte activation deficiencies, artificial oocyte activation (AOA) strategies have been developed (Travnik et al, 2023). These approaches aim to mimic the physiological calcium (Ca²⁺) oscillations normally triggered by sperm-delivered PLCζ, thereby rescuing fertilization in cases of oocyte activation failure (Nicholson et al, 2024). Commonly employed methods include the use of Ca²⁺ ionophores, such as ionomycin or A23187, which facilitate rapid Ca²⁺ influx into the oocyte; electrical stimulation, which transiently depolarizes the oocyte membrane to induce Ca²⁺ release from the endoplasmic reticulum; and recombinant PLCζ protein, which directly reproduces the enzymatic activity of sperm PLCζ to initiate IP₃-mediated Ca²⁺ oscillations (Antonova et al, 2024). These strategies not only restore the resumption of meiosis and pronuclear formation but also re-establish downstream signaling events essential for embryonic genome activation and subsequent embryo development.

In this study, we identify mutations of HNRNPR as a novel cause of sperm-borne oocyte activation failure. Mechanistically, hnRNPR regulates m6A-dependent alternative splicing of Plcz1, thereby influencing PLCζ expression and downstream Ca²⁺ oscillations critical for oocyte activation. To functionally restore activation in cases of PLCζ insufficiency, we employed Srcl₂-induced Ca²⁺ release to trigger endoplasmic reticulum-mediated signaling, successfully rescuing fertilization and embryo development in both human and mouse oocytes. Furthermore, we developed extracellular vesicle-based SKAP2 delivery (EVs-SKAP2) to restore sperm motility, and demonstrated that recombinant PLCζ protein can rescue sperm-borne activation failure with Hnrnpr mutations in mice. Collectively, our findings reveal HNRNPR mutations as a new genetic marker for male-derived oocyte activation failure and infertility, while also providing translational insights into targeted therapeutic strategies, such as Sr²⁺-mediated activation and recombinant PLCζ supplementation, to overcome fertilization barriers following ICSI.

Results

Mutations of HNRNPR gene cause oocyte activation failure and male infertility

We conducted a clinical investigation of a cohort of 572 individuals with asthenoteratozoospermia on three patients who experienced fertilization failure following both in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI) procedures. Remarkably, artificial oocyte activation (AOA) using Srcl₂ successfully rescued fertilization failure in these patients (Fig. 1A). A total of 75% (12/16) of oocytes achieved successful fertilization, and 50% (6/12) of zygotes developed into good-quality embryos (Fig. 1B; Table EV1). Following the transfer of three high-quality embryos, pregnancy was achieved by the patient’s partner. To elucidate the genetic basis underlying the observed oocyte activation failure, whole-exome sequencing (WES) was performed. We identified a homozygous missense variant (c.1540 G > A) in HNRNPR in two affected brothers from a consanguineous family, and a compound heterozygous missense variant (c.1280 A > C and c.1369 G > A) in a proband from a non-consanguineous family (Fig. 1C). In silico analyses using SIFT, PROVEAN, CADD, and MutationTaster predicted all three variants to be deleterious. Furthermore, PhastCons and phyloP conservation analyses revealed a high degree of nucleotide conservation at these loci (Fig. 1D). To assess the functional consequences of the identified HNRNPR variants, computer-assisted sperm analysis (CASA) showed that the patient’s sperm concentration was within the normal range; however, the motility and proportion of morphologically normal spermatozoa were significantly reduced (Fig. EV1A; Table EV2). Scanning electron microscopy (SEM) revealed structural abnormalities, including tapered or thin sperm heads and bent flagella in two affected individuals (Fig. EV1B). Transmission electron microscopy (TEM) further demonstrated disruption of the characteristic “9 + 2” microtubule arrangement in the midpiece, principal piece, and endpiece of the sperm flagellum (Fig. EV1C). Quantitative analysis confirmed a significant increase in both sperm head abnormalities and flagellar defects associated with the HNRNPR mutations (Fig. EV1A; Table EV2).Figure 1. Genetic mutation of HNRNPR leads to sperm-borne oocyte activation failure.(A) Flowchart illustrating the clinical process of sperm-borne oocyte activation failure and the assisted oocyte activation (AOA) treatment protocol. (B) Summary of assisted reproductive outcomes, including in vitro fertilization (IVF), intracytoplasmic sperm injection (ICSI), and ICSI combined with AOA, from three clinical patients. (C) Pedigree analysis of families carrying HNRNPR variants associated with male infertility, identified through whole-exome sequencing. Pink arrows indicate affected males, and a double horizontal line denotes consanguineous marriage. M mutation, W wild-type. (D) Schematic representation of HNRNPR variant characteristics and computational prediction of their pathogenicity. Source data are available online for this figure.

Mutation of Hnrnpr causes oocyte activation failure and male infertility in mice

Before exploring the functional role of hnRNPR, we first examined its expression profile. Hnrnpr mRNA was found to be highly enriched in the testis compared with other adult organs in wild-type (WT) mice (Appendix Fig. S1A). Immunofluorescence analysis of P56 testicular sections confirmed that hnRNPR was expressed throughout multiple stages of germ cell development within the seminiferous tubules (Appendix Fig. S1B). Further analysis across various postnatal stages revealed that Hnrnpr expression markedly increased from P21 onward (Appendix Fig. S1C). To investigate the hnRNPR function in vivo, we generated a Hnrnpr knock-in (KI) mouse line targeting the RGG domain (Appendix Fig. S2A). Homozygous KI mice exhibited normal viability, with survival rates comparable to WT mice from 2 months (97.84% vs. 98.66%) to 8 months (94.81% vs. 95.67%) of age (Appendix Fig. S2B), indicating that hnRNPR is not essential for general survival. Consistently, body weight and testis-to-body weight ratios showed no significant differences between WT and KI mice (Appendix Fig. S2C,D). Despite normal testicular weight, Hnrnpr-mutant males were completely infertile (Appendix Fig. S2E). Breeding trials confirmed that KI males failed to sire offspring, consistent with the infertility phenotype observed in affected human patients. Computer-assisted sperm analysis (CASA) revealed comparable sperm counts between WT and KI mice, but a significant reduction in sperm motility and an increased proportion of morphologically abnormal spermatozoa in KI males (Appendix Fig. S2F–I). To determine whether hnRNPR also functions in Sertoli cells, we generated Sertoli cell knockout mice (Dhh-Cre; Hnrnpr^flox/Del^, designated DcKO) (Appendix Fig. S2J). Unlike KI males, DcKO mice exhibited normal testis size and fertility (Appendix Fig. S2K–M), indicating that hnRNPR is essential for germ cell function and male fertility, whereas its presence in Sertoli cells is dispensable for reproductive competence.

To further assess the role of hnRNPR in sperm head morphogenesis, stepwise analysis of spermatid development revealed disrupted acrosome biogenesis and nuclear condensation from step 9 to step 16 (Fig. EV2A,B). Transmission electron microscopy (TEM) of WT testes showed normal spermatid differentiation, whereas in KI mice, the acroplaxome, a structure anchoring the acrosome to the nucleus, was disorganized at the cap phase, as evidenced by an enlarged space between the inner acrosomal membrane (IAM) and the nuclear envelope (NE) (Fig. EV2C). The proportion of spermatids exhibiting a loosened acroplaxome was significantly higher in KI mice than in WT controls, and the acroplaxome was markedly thicker in KI spermatids (Fig. EV2D,E). In WT mice, elongated spermatids at the maturation phase displayed condensed nuclei with a well-defined acrosome tightly apposed to the nuclear surface. In contrast, KI spermatids lacked a normal acrosome, and epididymal sperm exhibited acrosomal detachment from the nuclear envelope accompanied by nuclear deformation (Fig. EV2F–H). Notably, both IVF and ICSI failed to rescue the fertilization defect in KI spermatozoa (Fig. 2A–F), whereas artificial oocyte activation (AOA) successfully overcame the fertilization failure and embryonic development arrest (Fig. 2D–F), consistent with the phenotype observed in affected human patients. Collectively, these findings highlight a critical role for hnRNPR in regulating spermiogenesis, oocyte activation, and overall fertilization competence.Figure 2. Mutations in sperm Hnrnpr cause sperm-borne oocyte activation failure in mice.(A) In vitro fertilization (IVF) of sperm from wild-type (WT) and knock-in (KI) mice. Scale bar = 40 μm. (B) Quantification of the percentage of viable zygotes following IVF. Data are presented as mean ± s.d.; statistical significance was determined using a two-sided Student’s t test. (C, D) Schematic diagrams illustrating the experimental procedures for intracytoplasmic sperm injection (ICSI) and ICSI combined with assisted oocyte activation (ICSI + AOA). (E) Differential interference contrast (DIC) images showing embryonic development after ICSI and ICSI + AOA in different groups. Scale bar = 40 μm. (F) Quantitative analysis of embryonic development rates following ICSI and ICSI + AOA treatments. Data are presented as mean ± s.d.; statistical significance was determined using a two-sided Student’s t test. Data in (A, B) represent results from six independent experiments (two technical and three biological replicates, n = 6). Data in (C–F) represent results from three independent experiments (three biological replicates, n = 3). Source data are available online for this figure.

Mutations of Hnrnpr in sperm lead to abnormal calcium oscillations in oocytes

To elucidate the molecular mechanism underlying the oocyte activation failure caused by Hnrnpr mutations, we systematically surveyed previously reported pathogenic mechanisms associated with oocyte activation failure. Among all identified genes, only five were of sperm origin, whereas twenty-six were of oocyte origin (Fig. 3A). Notably, all five sperm-derived genes function in oocyte activation through regulation of the calcium signaling pathway (Fig. 3A). Based on these findings, we hypothesized that Hnrnpr mutations might disrupt calcium signaling within oocytes. To test this hypothesis, we performed immunofluorescence staining to monitor calcium dynamics in oocytes following ICSI with WT or KI spermatozoa. Under physiological calcium conditions (2 mM), oocytes injected with KI sperm failed to exhibit periodic calcium oscillations (Fig. 3B–E). Increasing the extracellular calcium concentration to 5 mM induced only a transient, low-amplitude calcium rise in KI oocytes, which remained significantly weaker than that observed in WT controls (Fig. 3B–E). These results indicate that Hnrnpr-mutant sperm are incapable of triggering the normal pattern of calcium oscillations required for oocyte activation.

Because calcium oscillations in oocytes mainly originate from the endoplasmic reticulum (ER) (Wei et al, 2024), we next examined whether Hnrnpr mutations affected ER calcium release. Oocytes were treated with the IP₃ receptor agonist histamine, which successfully induced ER calcium release and normal cytoplasmic calcium oscillations, demonstrating the functional integrity of IP₃ receptors (Fig. 3F,G). Moreover, supplementation with Srcl₂ restored cytoplasmic calcium oscillations in oocytes injected with KI sperm. To confirm that the increase in cytoplasmic calcium originated from ER stores, ER calcium dynamics were monitored using the ER-specific calcium indicator Cal-520ER™ AM. The cytoplasmic calcium increase was accompanied by a corresponding decrease in ER calcium, supporting the notion of ER calcium release (Fig. 3H–J). In addition to Srcl₂ being able to restore the calcium oscillations of oocytes, immunofluorescence analyses revealed that Srcl₂ treatment rescued subsequent fertilization, as evidenced by the formation of male and female pronuclei (Fig. EV3A,B). Collectively, these results demonstrate that sperm carrying Hnrnpr mutations fail to initiate normal calcium oscillations in oocyte, leading to oocyte activation failure, while artificial oocyte activation with Srcl₂ effectively restores fertilization and developmental competence.Figure 3. Mutated Hnrnpr sperm induce disordered calcium oscillations in oocytes following ICSI.(A) Circular diagram illustrating genes associated with oocyte activation failure and their molecular mechanisms. Five sperm-borne and twenty-six oocyte-borne genes are listed as pathogenic contributors. (B) Fluorescence imaging of calcium signals in oocytes using the Fura-2 probe after ICSI with sperm from wild-type (WT) and knock-in (KI) male mice. The diffusible Fura-2 AM ester probe predominantly localizes to the cytoplasm after entering the cell. Scale bar = 10 μm. (C) Relative quantification of calcium signal intensity. Data are presented as mean ± s.d.; NS not significant. P values were determined using a two-sided Student’s t test. (D, E) Response of Ca²⁺ oscillations to changes in extracellular calcium concentrations (2 mM and 5 mM) across three independent experiments. (D) Representative traces. (E) Quantification of oscillation frequency under the indicated conditions. Data are shown as mean ± s.d.; NS not significant. P values determined using a two-sided Student’s t test. (F, G) Ca²⁺ oscillation imaging following histamine (MedChemExpress (Monmouth Junction, NJ, USA; Cat# HY-B1204)) stimulation (0 μM and 10 μM). (F) Representative traces. (G) Quantified oscillation frequencies, shown as mean ± s.d.; NS not significant. P values were calculated using a two-sided Student’s t test. (H–J) Ca²⁺ oscillation imaging in oocytes subjected to WT-ICSI, KI-ICSI, and KI-ICSI + AOA treatments. (H) Representative Fura-2 traces. (I) Quantification of oscillation frequency, shown as mean ± s.d.; NS not significant. P values were determined using a two-sided Student’s t test. (J) Representative traces obtained using Cal-520ER™ AM fluorescence imaging. Data in (B–J) represent results from three independent experiments (three biological replicates, n = 3). Source data are available online for this figure.

Abnormal RNA transcript in oocytes after ICSI and PLCζ expression in haploid spermatids

To elucidate the molecular mechanisms underlying the oocyte activation failure caused by Hnrnpr mutations in sperm, we first performed Smart-seq transcriptome analysis following ICSI (Fig. 4A). Comparative analysis revealed 261 differentially expressed genes (DEGs) between WT and KI oocytes after ICSI, and 278 DEGs between KI-ICSI and KI-ICSI + AOA groups (Fig. 4B,C). Venn diagram analysis identified 81 overlapping DEGs, and Gene Ontology (GO) enrichment indicated that these genes were primarily associated with calcium homeostasis and fertilization pathways (Fig. 4D,E). These results suggest that disruption of Hnrnpr in sperm impairs the transcriptional response in the oocytes and is essential for oocyte activation. Due to Hnrnpr mutation in sperm, further single-cell RNA sequencing (scRNA-seq) was conducted on testes from WT and Hnrnpr knock-in (KI) mice. This analysis identified 16 distinct cell clusters, which were grouped into six principal cell populations using marker genes, and the transcriptional profiles of these clusters were largely comparable between WT and KI testes (Appendix Figs. S3A–E and S4). Quantitative analysis confirmed that overall cell-type proportions were also not significantly altered (Appendix Fig. S3F). UMAP visualization and marker gene expression analysis further verified the similar presence of spermatogonia, spermatocytes, round spermatids, and elongating spermatids in both groups (Appendix Fig. S5A–D). Pseudotime trajectory and cell cycle reconstruction revealed preserved developmental progression from undifferentiated germ cells to mature spermatids in both genotypes (Appendix Fig. S5E–J).Figure 4. Abnormal transcriptomic and proteomic profiles in oocytes and sperm.(A) Flowchart illustrating the SMART-Seq experimental workflow performed after ICSI and ICSI combined with AOA. (B) Volcano plot showing differentially expressed genes (DEGs) between WT-ICSI and KI-ICSI groups. P values were calculated using a Welch’s t test. (C) Volcano plot showing DEGs between KI-ICSI and KI-ICSI + AOA groups. P values were calculated using a Welch’s t test. (D) Venn diagram depicting the overlap of DEGs identified in (B, C). (E) Gene Ontology (GO) enrichment analysis of the common DEGs identified in (D). P values were calculated using a Fisher’s exact test. (F) Differential gene expression and GO enrichment analysis of haploid spermatids from WT and KI mice. P values were calculated using a Fisher’s exact test. (G) Left, Venn diagram comparing quantitative proteomics of human and mouse epididymal sperm. Right, GO enrichment analysis of shared differentially expressed proteins. P values were calculated using a Fisher’s exact test. (H) Violin plots showing the expression distribution of variable genes within haploid spermatid clusters. Data are shown as mean ± s.d.; statistical significance was determined using a two-sided Student’s t test. (I) Quantitative analysis of Plcz1 mRNA expression in haploid spermatids from WT and KI mice by qPCR. Data are presented as mean ± s.d.; P values were calculated using a Mann–Whitney U test. (J) Representative western blot showing PLCζ protein expression in haploid spermatids from WT and KI mice, with GAPDH as the loading control. (K) Quantification of PLCζ protein levels shown in (J). Data are presented as mean ± s.d.; significance determined using a two-sided Student’s t test. Data in (B, C, H) represent results from three independent biological replicates (n = 3). Data in (I–K) represent results from six independent experiments (two technical and three biological replicates, n = 6). Source data are available online for this figure.

Despite the preservation of general spermatogenic processes, scRNA-seq revealed distinct transcriptomic alterations at the haploid spermatid stage. Specifically, DEGs were enriched in pathways related to fertilization, suggesting functional disruption during sperm maturation (Fig. 4F). Notably, the expression of Plcz1, which encodes phospholipase C zeta (PLCζ), a key sperm-derived factor responsible for initiating oocyte activation, was markedly reduced in KI haploid spermatids (Fig. 4F). To support these transcriptomic findings, we performed quantitative proteomic analysis of epididymal sperm from both humans and mice. A total of 145 differentially expressed proteins (DEPs) were identified in humans and 201 in mice, with 98 proteins shared across both species (Fig. 4G). GO analysis of these DEPs revealed significant enrichment in fertilization-related processes. Importantly, both transcriptomic and proteomic analyses consistently demonstrated significant downregulation of PLCζ in the KI group (Fig. 4H–K). Visualization of Plcz1 expression within the scRNA-seq dataset by violin plots confirmed reduced expression in haploid spermatids from KI mice (Fig. 4H). RT-qPCR and immunoblotting further validated these findings (Fig. 4I–K). Moreover, immunofluorescence analysis of mature spermatozoa from both patients and KI mice demonstrated abnormal subcellular localization of PLCζ (Fig. 5A,B). Immunoblotting corroborated that PLCζ expression was diminished in mature epididymal spermatozoa (Fig. 5C–F). Collectively, these results establish that hnRNPR is a pivotal regulator of Plcz1 gene expression. Mutation of HNRNPR disrupts the synthesis and localization of PLCζ, impairing the sperm’s ability to trigger calcium oscillations and oocyte activation, ultimately leading to male infertility.Figure 5HNRNPR mutation disrupts the expression and localization of PLCζ in epididymal spermatozoa.(A) Immunofluorescence detection of PLCζ localization in spermatozoa from healthy individuals and patients carrying HNRNPR mutations. Scale bar = 2 μm. (B) Immunofluorescence analysis of PLCζ localization in sperm from wild-type (WT) and knock-in (KI) mice. Scale bar = 2 μm. (C) Immunoblotting showing PLCζ protein expression levels in sperm from normal individuals and HNRNPR-mutated patients, with GAPDH used as the loading control. (D) Quantitative analysis of PLCζ protein expression from (C). Data are presented as mean ± s.d. P values were calculated using a two-sided Student’s t test. (E) Immunoblotting showing PLCζ protein expression in sperm from WT and KI mice. (F) Quantification of PLCζ protein levels in WT and KI sperm. Data are shown as mean ± s.d.; statistical significance was determined using a two-sided Student’s t test. Data in (A–F) represent results from six independent experiments (two technical and three biological replicates, n = 6). Source data are available online for this figure.

hnRNPR regulates pre-mRNA alternative splicing of Plcz1 in an m6A-dependent manner

As a well-characterized splicing regulator, hnRNPR binds precursor mRNAs and modulates alternative splicing (AS) across multiple tissues (Hu et al, 2024; Wang et al, 2025). To explore whether hnRNPR exerts a similar regulatory role in the testis and male germ cells, we analyzed RNA sequencing (RNA-seq) data from the testes of postnatal day 56 (P56) wild-type (WT) and Hnrnpr knock-in (KI) mice. Comparative transcriptomic profiling revealed that Hnrnpr mutation profoundly altered the splicing landscape, leading to 1671 differential AS events and 1202 differentially spliced genes in diploid spermatids (Fig. 6A). Among these, exon skipping was the most prominent alteration, accounting for 61.4% (1026/1671) of total AS events and 60.57% of affected genes across three biological replicates (Fig. 6B). Other detected splicing changes included mutually exclusive exons (209 events), alternative 5′ splice sites (133 events), alternative 3′ splice sites (148 events), and retained introns (155 events), collectively indicating broad disturbances in splice site selection. Mapping analyses further revealed that Hnrnpr mutation decreased exonic read density while increasing intronic and intergenic reads (Fig. 6C), suggesting impaired transcript maturation and increased splicing errors or transcript instability. Gene Ontology (GO) enrichment analysis of differentially spliced genes highlighted strong associations with fertilization, acrosome formation, and mRNA stability (Fig. 6D). These findings indicate that hnRNPR-dependent splicing is essential for the correct expression of fertilization-related genes, and its disruption may underlie the oocyte activation defects and male infertility observed in Hnrnpr KI mice.

To experimentally validate these observations, we examined AS events in the key fertilization-related gene Plcz1 using Sashimi plot visualization and semiquantitative RT-PCR analysis (Fig. 6E–G). RT-PCR with isoform-specific primers confirmed that Hnrnpr mutation induces aberrant splicing characterized by exon skipping, including exon 3 in Plcz1 transcripts (Fig. 6G). Such mis-splicing events give rise to noncanonical or truncated transcript isoforms, potentially compromising Plcz1 protein expression and its function in triggering oocyte activation. To determine whether hnRNPR directly interacts with precursor mRNAs and modulates Plcz1 splicing in an m6A-dependent manner during spermiogenesis, we performed RNA-binding protein immunoprecipitation sequencing (RIP-seq) of hnRNPR together with m6A RNA immunoprecipitation sequencing (meRIP-seq) using round spermatids from postnatal day 35 (P35) mouse testes. Integrative multi-omics analysis, combining single-cell RNA sequencing (scRNA-seq), RNA-seq, m6A-seq, and RIP-seq datasets, identified 46 overlapping genes that were differentially expressed and alternatively spliced, while also carrying both m6A modifications and hnRNPR-binding peaks (Fig. 6H,I). GO enrichment of these targets revealed strong enrichment for biological processes linked to fertilization and sperm acrosome development (Fig. 6I). Among these, Plcz1 emerged as a key functional target due to its indispensable role in initiating calcium oscillations during oocyte activation. To verify the direct regulatory relationship between hnRNPR and Plcz1, RNA immunoprecipitation followed by PCR (RIP-PCR) confirmed that hnRNPR binds directly to Plcz1 precursor mRNA, with strong binding affinity validated by RIP-qPCR (Fig. 6J). Integrative Genomics Viewer (IGV) visualization further demonstrated that hnRNPR-binding sites coincide with m6A-enriched splice junctions, supporting the notion that hnRNPR regulates Plcz1 splicing through an m6A-dependent mechanism (Fig. 6K). Collectively, these results reveal that hnRNPR orchestrates Plcz1 pre-mRNA alternative splicing via m6A-mediated regulation, thereby ensuring proper expression of Plcz1, a key sperm-derived factor critical for oocyte activation and fertilization success.Figure 6hnRNPR regulates Plcz1 splicing in an m6A-dependent manner.(A) Summary chart of alternative splicing (AS) events and affected genes across five major categories significantly altered by Hnrnpr mutation in testes, based on RNA-seq analysis using rMATS (replicate Multivariate Analysis of Transcript Splicing). (B) Pie charts showing the distribution of AS events and associated genes among different splicing types. (C) RNA-seq read coverage maps of WT and KI mouse testes, with bar charts indicating the proportion of reads mapped to intergenic, intronic, and exonic regions. (D) Gene Ontology (GO) analysis of aberrantly spliced genes showing significant enrichment in biological processes related to fertilization and acrosome formation. P values were calculated using a Fisher’s exact test. (E) Sashimi plots generated using the Integrative Genomics Viewer (IGV), displaying representative differential splicing events of Plcz1 between WT and KI testes. (F) Diagram illustrating the alternative splicing pattern and calculation scheme. (G) Semiquantitative RT-PCR validation of Plcz1 aberrant splicing in testes. Differentially spliced exons are labeled as “▲exon number.” Schematic diagrams below each gel image illustrate exon inclusion or exclusion, while histograms on the right display percent spliced-in (PSI) values (mean ± s.d.). Statistical significance was determined using a two-sided Student’s t test. (H) The distribution of hnRNPR binding sites in the RNAs of isolated round spermatids identified by RIP-seq (RIP-sequencing). (I) Left, Venn diagram showing 46 overlapping genes among m6A-modified genes (m6A-seq), AS-regulated genes, hnRNPR-bound genes, and differentially expressed genes. Right, GO enrichment analysis of these 46 overlapping genes. P values were calculated using a Fisher’s exact test. (J) RNA immunoprecipitation (RIP) using hnRNPR antibody followed by PCR and gel electrophoresis demonstrating direct binding of hnRNPR to differentially spliced Plcz1 mRNA. (K) IGV browser tracks displaying Plcz1 transcript features derived from RNA-seq (green), hnRNPR RIP-seq (red), and meRIP-seq (purple). Red shading and yellow arrowheads highlight regions of alternative splicing, hnRNPR binding, and m6A modification, respectively. Data in (G, J) represent results from six independent experiments (two technical and three biological replicates, n = 6). Source data are available online for this figure.

Therapeutic rescue of fertilization failure caused by HNRNPR mutations

In our study, the Hnrnpr mutation not only leads to a decline in sperm motility but also causes oocyte activation failure. Thus, we first aimed to improve the sperm motility and found Src kinase-associated phosphoprotein 2 (SKAP2) plays a crucial role in cytoskeletal remodeling and sperm motility. To explore its translational therapeutic potential, we developed a gene therapy-inspired strategy utilizing SKAP2-enriched extracellular vesicles (mEVs-SKAP2). Leveraging the higher loading efficiency and intrinsic cargo-delivery capacity of extracellular vesicles (EVs) (Fig. EV4A), we conducted in vitro co-incubation assays by mixing mEVs-SKAP2 with epididymal spermatozoa in humans (Table EV3) and mice (Table EV4). Post-treatment assessment using computer-assisted sperm analysis (CASA) revealed a significant improvement in sperm motility, while correction of morphological and axonemal abnormalities remained limited (Figs. 7A–D and EV4B–E and EV5A–C). Moreover, both F-ACTIN polymerization and phosphate levels were markedly upregulated following treatment (Figs. 7E–G and EV5D–G), indicating enhanced cytoskeletal dynamics and improved sperm bioenergetic activity.

Traditionally, viable testicular or epididymal sperm can be used to achieve fertilization through in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI). To evaluate the advantages of our EV-SKAP2 strategy over these conventional approaches, we assessed fertilization competence following IVF (Fig. 7H). Furthermore, given that sperm harboring Hnrnpr mutations fail to trigger oocyte activation upon entry, we applied Srcl₂ treatment to induce artificial activation (Fig. 7H). Remarkably, combined treatment with EVs-SKAP2 and Srcl₂ successfully restored fertilization competency, enabling normal two-cell embryo development and the eventual birth of live offspring (Fig. 7I–K). As a critical downstream effector of hnRNPR, Plcz1 is essential for triggering calcium oscillations during oocyte activation and fertilization. To evaluate the therapeutic potential of Plcz1 restoration, we synthesized a NusA-tagged recombinant human PLCζ (NusA-PLCζ) protein. Following ICSI, 0.01 mg/ml of NusA-PLCζ was microinjected into metaphase II (MII) oocytes (Fig. 7L). Post-treatment analyses demonstrated significantly improved embryo developmental rates compared to NusA control groups (Fig. 7M,N), confirming the functional rescue of oocyte activation and early embryogenesis. Collectively, these findings demonstrate that mEVs-SKAP2 enhances sperm motility primarily by modulating F-ACTIN dynamics and ATP metabolism, while Srcl₂ and NusA-PLCζ supplementation effectively restores calcium oscillation and oocyte activation competence, establishing a coherent mechanistic and translational framework for the development of targeted therapeutic strategies to overcome fertilization failure.Figure 7. The combined treatments can restore the fertilization ability of sperm in mice with Hnrnpr mutations.(A, B) Computer-assisted sperm analysis (CASA) was used to evaluate total motility (A) and progressive motility (B) among untreated (WT, KI), mEVs-treated, and mEVs-SKAP2-treated groups after 3-h in vitro co-incubation. Data are presented as mean ± s.d., and statistical significance was determined using a two-sided Student’s t test. (C) Quantification of morphologically abnormal sperm (500 spermatozoa per mouse). Results are shown as mean ± s.d., with “NS” indicating no significant difference; P values were calculated using a two-sided Student’s t test. (D) The percentage of abnormal axonemes in sperm flagella was quantified. Data are presented as mean ± s.d., and significance was assessed by two-sided Student’s t test, with “NS” indicating no significant difference. (E) Representative western blot showing F-ACTIN and G-ACTIN protein levels and significant differences in actin polymerization rates in sperm following in vitro treatment. (F) Densitometric analysis of the F-ACTIN/G-ACTIN ratio from (E), expressed as mean ± s.d.; statistical significance was evaluated using a two-sided Student’s t test. (G) The concentration of free phosphate, reflecting ATP content and energy released after hydrolysis, was measured across different groups. Results are expressed as mean ± s.d., and P values were determined using a two-sided Student’s t test. (H) The schematic flowchart illustrates the IVF procedure, in which MII oocytes and sperm underwent EVs-SKAP2 incubation followed by artificial oocyte activation (Srcl₂ treatment). (I) Representative images showing two-cell development of oocytes from WT and KI mice under different treatment protocols. The purple arrow indicates a successfully developed two-cell embryo; scale bar = 50 μm. (J) Quantification of two-cell stage oocytes from WT and KI mice subjected to the indicated treatments. Data are expressed as mean ± s.d., and statistical significance was evaluated using a two-sided Student’s t test. (K) Offspring analysis following combined mEVs-SKAP2 and Srcl₂ therapy in WT and KI male mice. Data represent mean ± s.d., with P values obtained by a two-sided Student’s t test. (L) Flowchart of the PLCζ treatment protocol: after ICSI, 0.01 mg/ml of PLCζ was microinjected into the oocyte. (M) Representative differential interference contrast (DIC) images of two-cell embryos from NusA and NusA-PLCζ groups, with a scale bar of 50 μm. (N) Quantification of the percentage of two-cell embryos following ICSI-AOA, as shown in (M). Data are presented as mean ± s.d., and P values were calculated using a two-sided Student’s t test. Data in (A–D, F, G, J, N) represent results from nine independent experiments (three technical and three biological replicates, n = 9). Source data are available online for this figure.

Discussion

Our study identifies HNRNPR mutations as a previously unrecognized genetic cause of sperm-borne oocyte activation failure, thereby expanding the molecular landscape of male infertility. Successful fertilization depends on precise and sustained calcium oscillations initiated by sperm-delivered phospholipase C zeta (PLCζ), which triggers the release of intracellular calcium from the oocyte endoplasmic reticulum to initiate meiosis resumption and early embryonic development. Here, we demonstrate that hnRNPR regulates this essential process through m6A-dependent alternative splicing of Plcz1, ensuring the correct maturation and translation of PLCζ mRNA. Abnormal hnRNPR function disrupts this splicing-dependent regulatory axis, leading to reduced or aberrant PLCζ expression, defective calcium signaling, and ultimately oocyte activation failure.

These findings reveal a previously unrecognized connection between RNA splicing regulation and oocyte activation, underscoring that post-transcriptional control of sperm-borne factors is as critical as their transcriptional regulation for successful fertilization. Our results indicate that hnRNPR acts as a pivotal RNA-binding protein (RBP) that safeguards the integrity and expression of key fertilization determinants through precise modulation of mRNA processing. However, hnRNPR protein is also highly expressed in spermatocytes, suggesting that hnRNPR not only functions during sperm maturation and fertilization, but also may play a role in the early stages of spermatogenesis. Given that PLCζ deficiency is a major cause of oocyte activation failure and male infertility (Hachem et al, 2017) and previous studies have shown that certain genes, such as ACTRT1 (Zhang et al, 2024b), IQCN (Dai et al, 2022), ACTL7A (Xin et al, 2020), and ACTL9 (Dai et al, 2021) can also cause the failure of oocyte activation by influencing the morphology of the sperm acrosome. Our findings further position hnRNPR as an upstream regulatory node controlling the availability of functional Plcz1 mRNA and protein. Disruption of hnRNPR thus leads to aberrant Plcz1 transcript maturation, reduced PLCζ expression, and abnormal sperm acrosome, ultimately impairing oocyte activation and fertilization. Beyond HNRNPR mutation-related infertility, this discovery provides a conceptual framework for exploring how the RBP network governs sperm transcriptome remodeling, protein synthesis, and fertilization competence. Importantly, the identification of hnRNPR-Plcz1 regulation opens new avenues for the development of diagnostic biomarkers and targeted therapeutic strategies, such as restoring PLCζ levels or modulating RBP activity, to rescue fertilization failure in affected patients.

To address the functional consequences of impaired PLCζ activity, we tested multiple artificial oocyte activation (AOA) strategies. Treatment with Srcl₂, which induces endoplasmic reticulum calcium release through activation of the IP₃R signaling cascade (Han and Gao, 2013), effectively restored oocyte activation and fertilization in both mouse and human models (Fawzy et al, 2018). This provides strong clinical support for Sr²⁺-based AOA as a therapeutic intervention for sperm-borne activation deficiency. In previous studies, Srcl₂ stimulation is frequently used in mouse models and, in selected human cases, as a gentler and more physiologically relevant inducer of Ca²⁺ oscillations. Sr²⁺ acts by activating the phospholipase C-IP₃ signaling cascade, promoting repetitive Ca²⁺ transients that closely resemble those initiated by PLCζ (Kashir et al, 2022; Travnik et al, 2023). In addition to AOA, we explored complementary therapeutic strategies aimed at correcting upstream sperm defects associated with HNRNPR mutations. The use of extracellular vesicle-mediated SKAP2 delivery (EVs-SKAP2) effectively enhanced sperm motility by restoring cytoskeletal organization and ATP metabolism, thereby improving overall fertilization competence. Recent studies have also explored recombinant PLCζ protein or mRNA microinjection as a physiological alternative to restore Ca²⁺ oscillations in PLCζ-deficient sperm (Parrella et al, 2024; Saleh et al, 2024). This strategy directly targets the molecular defect underlying activation failure and has shown promising results in both animal models and preliminary human applications. In our study, microinjection of recombinant NusA-tagged PLCζ protein successfully rescued oocyte activation failure in mice, confirming that direct restoration of PLCζ activity can re-establish endogenous signaling and normal fertilization.

Nonetheless, these studies collectively emphasize that no single artificial oocyte activation (AOA) protocol is universally effective. Instead, individualized selection of activation strategies based on the underlying molecular defect, such as PLCζ deficiency, Ca²⁺ signaling disruption, or sperm structural abnormalities, appears to offer the most favorable therapeutic outcomes (Zhao et al, 2023). In this context, the identification of HNRNPR mutations as a novel cause of defective oocyte activation further underscores the importance of genotype-guided AOA approaches that directly target the molecular etiology of fertilization failure. However, the translational application of Sr²⁺ requires careful consideration, as its long-term effects on epigenetic reprogramming, embryonic genome activation, and offspring development remain incompletely understood. Comprehensive longitudinal and safety studies are therefore essential before its widespread clinical adoption (Dai et al, 2024). Future research should also aim to systematically evaluate the clinical efficacy and developmental safety of both Sr²⁺- and recombinant PLCζ-based interventions through well-designed prospective trials.

Collectively, our findings establish HNRNPR mutations as a novel diagnostic marker for sperm-derived oocyte activation failure, revealing a previously unrecognized mechanistic link between RNA splicing dysregulation and fertilization incompetence. The identification of hnRNPR as a critical regulator of Plcz1 mRNA maturation and stability underscores the pivotal role of post-transcriptional control in ensuring the availability of functional sperm-borne oocyte activation factors. By integrating molecular characterization with successful rescue of fertilization through artificial oocyte activation, this work establishes a framework for precision reproductive medicine, in which genotype-guided interventions, such as targeted AOA or molecular correction of Plcz1, can be designed to restore fertilization capacity in affected individuals.

Methods

Reagents and tools tableReagent/resourceReference or sourceIdentifier or catalog number Experimental models HNRNPR mutation patientsZhejiang universityK2024235Normal spermatozoa samplesZhejiang universityK2024235Hnrnpr Konck-in mutation (M.musculus)This studyThis studyDhh-Cre (M.musculus)Shanghai Model Organisms Center, Inc.NM-KI-225049C57BL/6 J (M.musculus)Cyagen BiosciencesKOAI241107YZ3 Recombinant DNA hyeA3A-BE4maxAddgeneCat# 157944pCold-NusAMiaoling plasmidCat# P8497pGEX-6P-1Miaoling plasmidCat# P0005 Antibodies Rabbit anti-hnRNPRABclonalCat# A21321Rabbit anti-GAPDHProteintechCat# 10494-1-APHRP Goat anti-mouse IgGFineTestCat# FNSA-0003HRP Goat anti-rabbit IgGFineTestCat# FNSA-0004Dylight 488 Goat anti-mouse IgGAbbkineCat# A23210Dylight 594 Goat anti-rabbit IgGAbbkineCat# A23420Rabbit anti-γH2AXABclonalCat# AP0099TUBA1A Monoclonal AntibodyCUSABIOCat# CSB-MA754656A0mRabbit anti-PLCZ1BiossCat# bs-5378RRabbit anti-alpha TubulinMedChemExpressCat# HY-P86200Rabbit anti-PLCZ1CovalabCat# PAB0367Rabbit anti-SKAP2AffinityCat# DF12085 Oligonucleotides and other sequence-based reagents PCR primersThis studyTable EV5ApexHF HS DNA Polymerase FSACCURATE BIOTECHNOLOGY (HUNAN) CO., LTDCat# AG12201 Chemicals, enzymes, and other reagents Clomiphene Citrate TabletsCodal Synto Ltd.Cat# 517853508MenotropinsLizhu PharmaceuticalCat# H10940097Human chorionic gonadotropinLizhu PharmaceuticalCat# S20210010Pregnant Mare Serum GonadotropinNingbo Second Hormone FactoryCat# 63017888Chorionic gonadotrophinNingbo Second Hormone FactoryCat# 63041888Western and IP cell lysis bufferNew Cell & Molecular BiotechCat# P70100SurePAGE™GenScript CorporationCat# M00657His tag Nanoselector Magnetic beadsHUABIOCat# 004-101-003His tag Nanoselector AgaroseHUABIOCat# 004-101-002Phosphate BufferProcellCat# PB1803275x loading bufferX-BlotCat# 0025HistamineMedChemExpressCat# HY-B1204Penicillin-StroptomycinAmizona Scientific LLCCat# AP1001-100Cal-520ER™ AMAAT BioquestCat# 21149 Software GraphPad Prism https://www.graphpad.com/ Version 10.1.2Xiantao xueshu https://www.xiantaozi.com/ Version 3.0ImageJ https://imagej.net/software/nih-image Version 1.48q Other Bradford Protein Assay KitabsinCat# abs580304Seamless DNA Assembly Plus KitMedChemExpressCat# HY-K1041Malachite Green Phosphate Detection KitBeyotimeCat# S0196MF-actin/G-actin In Vivo Assay KitCytoskeleton IncCat# BK037Microscopic operation and injection systemEppendorf&Olympus&HHAMILTONOosight&IX 73&LYKOS&METAZeiss900 upright laser confocal microscopeZEISSLSM900QuantStudio^TM^ 5 Real-TimePCR SystemThermo Fisher ScientificQuant Studio^TM^ 5

Whole-exome sequencing and data analysis

Human samples were collected after obtaining informed consent from all participants with severely reduced progressive sperm motility (<32%) and a high frequency of morphologically abnormal spermatozoa (<4% normal forms). Genomic DNA was extracted using the Blood Genomic DNA Extraction Kit (Qiagen, Germany), followed by whole-exome sequencing (WES) (Wu et al, 2025; Xu et al, 2025). The captured regions included coding exons and their flanking intronic sequences from more than 20,000 genes. Paired-end sequencing (2 × 100 bp) was performed on the Illumina HiSeq 2500 platform (Illumina, USA). The mean sequencing depth across the targeted regions exceeded 170×, with more than 95% of bases covered at a minimum depth of 30× Sequence alignment and variant calling were conducted using standard bioinformatics pipelines, and BAM files were generated for visualization in the Integrative Genomics Viewer (IGV; Broad Institute, USA). Variants were filtered according to a minor allele frequency (MAF) < 2%, predicted pathogenicity, and inheritance pattern. All candidate variants identified by WES were further validated using Sanger sequencing. This study was conducted in accordance with the principles of the WMA Declaration of Helsinki, the Department of Health and Human Services Belmont Report, and approved by the Institutional Ethics Committee of the Fourth Affiliated Hospital, Zhejiang University School of Medicine (Approval No.: K2024235, date: 19 Dec 2024). The Biobank declaration and national-level authorization (KY-2024-228) were also obtained, and all Ethics Committees involved adhered to the guidelines of the Declaration of Helsinki.

Ovarian stimulation and ICSI in humans

Women underwent ICSI treatment at our center as clinically indicated. Ovarian stimulation was carried out using a mild stimulation protocol consisting of clomiphene citrate (CC; Clomiphene Citrate Tablets, Codal Synto Ltd., Limassol, Cyprus) in combination with human menopausal gonadotropin (hMG; Menotropins for Injection, Lizhu Pharmaceutical, China). Oocyte retrieval was performed 35 h after human chorionic gonadotropin (hCG, Lizhu Pharmaceutical, China) administration for final oocyte maturation. Prior to ICSI, cumulus cells were mechanically removed. Fertilization and subsequent embryonic development were continuously assessed using a time-lapse monitoring system. All media for ICSI and embryo culture were obtained from Vitrolife (Gothenburg, Sweden; G5 series plus).

Assisted oocyte activation

For mice, oocytes were exposed to 5 mM Srcl₂ (Sigma-Aldrich, St. Louis, MO, USA) in KOSM medium for 6 h following ICSI. After AOA, the oocytes were washed twice with fresh KOSM medium and subsequently cultured in the same medium at 37 °C under 5% CO₂ and 95% humidity. For humans, the activation medium was prepared by equilibrating 1 ml of G1 medium supplemented with 5 mM EGTA (to chelate Ca²⁺) in an incubator for at least 2 h. Immediately before ICSI, 10 mM Srcl₂ was added to this medium. After ICSI, oocytes were incubated in G-MOPS at 37 °C with 5% CO₂ for 30 min, then transferred into the prepared activation buffer for 4 h. Following AOA, the oocytes were thoroughly washed several times in G1 medium and cultured according to standard embryo culture protocols.

Generation of Hnrnpr knock-in and Sertoli cell-specific knock-out mice

All mouse experiments were performed using animals maintained on a C57BL/6 J inbred genetic background. The Hnrnpr point mutation knock-in mice were generated using the base editor hyeA3A-BE4max, provided by Addgene (Cat# 157944). To achieve Sertoli cell-specific knockout, Hnrnpr^flox/flox^ mice were also crossed with Dhh-Cre mice, which express Cre recombinase in Sertoli cells, generating Hnrnpr^flox/flox^; Dhh-Cre mice (referred to as “Hnrnpr-DcKO” in the text or “DcKO” in the figures). Genotype identification of mice using ApexHF HS DNA Polymerase FS (AG12201, ACCURATE BIOTECHNOLOGY (HUNAN) CO., LTD, ChangSha, China) and SurePlus 100 bp DNA Ladder (Hangzhou Yangming Biotechnology Co., Ltd., Amizona, AMG2001-100). Male mice at various developmental stages (from birth through adulthood) were used for phenotypic analyses, while mice aged 56 days postnatally were selected specifically for Sertoli cell isolation. All mice were housed under specific pathogen-free (SPF) conditions at the Zhejiang University Animal Center, with standard environmental controls for temperature, humidity, and light cycle. All experimental procedures were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of Zhejiang University (Approval No.: IACUC-ZJU34082, date: Mar 5, 2024), and carried out in accordance with institutional and international ethical regulations. The Ethics Committee operates under the principles of the International Council for Laboratory Animal Science (ICLAS).

Histological analysis

Testes and epididymides from wild-type (WT) and knock-in (KI) mice were based on the randomization process and fixed overnight at 4 °C in either Bouin’s solution (Sigma-Aldrich, HT10132) for histological examination or 4% paraformaldehyde (PFA) for immunostaining. For histological analysis, fixed tissues were dehydrated, embedded in paraffin, and sectioned at a thickness of 5 µm using a CryoStar NX50 cryostat (Thermo Fisher Scientific). Following deparaffinization and rehydration, sections were stained with periodic acid-Schiff (PAS) reagent using the PAS Staining Kit (Beijing Solarbio Science & Technology Co., Ltd, G1281) and Modified Hematoxylin-Eosin Staining Kit (Beijing Solarbio Science & Technology Co., Ltd, G1121). Images were acquired using a light microscope (Axio Scope.A1; Zeiss, Germany).

Immunofluorescence

Tissues and cells were seeded onto circular coverslips in 24-well plates and cultured to ~70% confluence over 24 h. After the indicated treatments, samples were fixed with 4% paraformaldehyde (PFA) for 15 min at room temperature (RT), permeabilized with 0.25% Triton X-100 in PBS for 10 min, and blocked with 5% BSA for 1 h at RT (Leismann et al, 2025). Primary antibodies, diluted in 5% BSA, were applied overnight at 4 °C. After three washes of 10 min each with PBS, tissues and cells were incubated with fluorophore-conjugated secondary antibodies (1:500) for 2 h at RT. Nuclei were counterstained with DAPI, followed by three additional PBS washes. Coverslips were mounted using antifade mounting medium and sealed with nail polish. Fluorescence images were acquired using a Zeiss LSM900 confocal microscope. We extend our gratitude to Bioss for the antibody of PLCZ1 Rabbit pAb (bs-5378R, 1:100) and alpha Tubulin Antibody from MedChemExpress (Monmouth Junction, NJ, USA; Cat# HY-P86200, 1:200). Tubulin was determined using TUBA1A Monoclonal Antibody (1:200, CSB-MA754656A0m, CUSABIO, https://www.cusabio.com/). We also extend our gratitude to ABclonal Technology (WuHan, China) for their antibodies of γH2AX (Cat# AP0099, 1:100) and hnRNPR (Cat# A21321, 1:200).

Epididymal sperm analysis

The caudal epididymis was isolated from 8-week-old mice and finely minced to release sperm. The tissue was incubated in pre-warmed phosphate-buffered saline (PBS) at 37 °C for 10–20 min. For sperm morphology assessment, smears were prepared and stained with hematoxylin and eosin (H&E; Servicebio, G1076) according to the manufacturer’s instructions. For scanning electron microscopy (SEM), freshly collected sperm were resuspended in electron microscopy fixative (Servicebio, G1102), fixed for 2 h at room temperature (RT), and then stored at 4 °C for 8 h. Samples were post-fixed with 1% osmium tetroxide (OsO₄) in 0.1 M PBS (pH 7.4) for 1.5 h at RT. Following dehydration through a graded ethanol series and 15 min in isoamyl acetate (Sinopharm, 10003128), specimens were dried using a critical point dryer (Quorum, K850), coated with gold particles, and imaged using a scanning electron microscope (S-3400N; Hitachi, Tokyo, Japan). For transmission electron microscopy (TEM), adult mouse testes were sectioned into small fragments and fixed overnight in 0.1 M cacodylate buffer (pH 7.4) containing 2.5% glutaraldehyde and 3% paraformaldehyde. After three washes in the same buffer, samples were post-fixed in 1% OsO₄ for 1 h at 4 °C. Ultrathin sections (60–70 nm) were stained with uranyl acetate and lead citrate, and observed using a JEM-1400 transmission electron microscope (JEOL) (Stathatos et al, 2024).

Isolation of spermatogenic cells

Spermatogenic cells, including spermatogonia, spermatocytes, round spermatids, and elongating spermatids, were isolated using the STA-PUT velocity sedimentation technique. Testes from postnatal day 35 (P35) mice were enzymatically dissociated with collagenase IV (Sigma, Cat. No. C5138-100 mg) and trypsin (Sigma, Cat. No. 9002-07-7) to generate a single-cell suspension. Following enzymatic digestion and washes with Phosphate Buffer (Procell, PB180327), the cells were filtered and loaded onto a linear bovine serum albumin (BSA) gradient in custom STA-PUT glassware. This method separates cells based on differences in sedimentation velocity, which reflect variations in cell size and density, thereby enabling the enrichment of specific germ cell populations. Only fractions with ≥90% purity were collected and used for downstream analyses.

Quantitative real-time PCR (RT-qPCR)

Total RNA was extracted from samples using TRIzol reagent (Invitrogen, 15596-025) according to the previous instructions (Gan et al, 2020). For RNA extraction and reverse transcription from isolated germ cells, a single-cell sequence-specific amplification kit (Vazyme, P621) was used following the provided protocol. Residual genomic DNA was removed using RNase-free DNase (Roche, 69182). Subsequently, 1 μg of total RNA was reverse-transcribed into cDNA. Quantitative real-time PCR was performed using SYBR Green master mix and gene-specific primers on a StepOnePlus Real-Time PCR System (Applied Biosystems, 4309155). Equal amounts of total RNA from the testes of each genotype were analyzed. Gene expression levels were quantified using the 2^-ΔΔCt method, with GAPDH serving as the internal control. The primer sequences required for the research are all listed in Table EV5.

Western blotting

Samples were washed twice with cold PBS and lysed using western and IP cell lysis buffer (New Cell and Molecular Biotech, Cat# P70100). A suitable volume of 5× loading buffer (X-Blot, Xblot-0025) was added to each lysate to achieve a final 1× concentration, and samples were denatured at 98 °C for 10 min. Proteins were separated on FuturePAGE™ 10% 15 Wells (Cat# ET15010, ACE, China), and transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore, IPVH00010). Membranes were blocked with 10% skim milk (Biosharp, BS102) in TBST (50 mM Tris-Cl, pH 7.5; 150 mM NaCl; 0.1% Tween-20) for 1 h at room temperature. Following blocking, membranes were incubated overnight at 4 °C with primary antibodies diluted in blocking solution with gentle agitation (Jeanpierre et al, 2025). After three washes of 10 min each with TBST, membranes were incubated with HRP-conjugated secondary antibodies (1:2000) in TBST for 1.5 h at room temperature. Protein bands were visualized using ECL reagent (Bio-Rad, 1705061) and detected with the ChemiDoc XRS+ imaging system (Bio-Rad). SKAP2 antibody from Affinity Biosciences (Cat# DF12085, 1:1000). PLCZ1 antibody from Covalab (1:2000). GAPDH from Proteintech (1:5000).

RNA-immunoprecipitation (RIP) sequencing

Round spermatids were isolated from three testes of postnatal day 35 (P35) mice and homogenized in ice-cold lysis buffer containing 50 mM Tris-HCl (pH 7.4), 100 mM NaCl, 0.5% NP-40, a protease inhibitor cocktail (1:100), and an RNase inhibitor. The homogenate was incubated on ice for 20 min to ensure complete cell lysis, followed by centrifugation at 15,000 × g for 10 min at 4 °C. Ten percent of the lysate was set aside as an input control, while the remaining supernatant was subjected to immunoprecipitation using an anti-hnRNPR antibody or rabbit IgG as a negative control. RNA was extracted from both input and RIP samples using TRIzol reagent (Invitrogen). Stranded RNA-seq libraries were generated using the KC-Digital™ Stranded mRNA Library Prep Kit for Illumina®, according to the manufacturer’s instructions. This kit minimizes PCR and sequencing duplication bias by incorporating unique molecular identifiers (UMIs) to label pre-amplified cDNA molecules. Library fragments of ~200 bp were enriched, quantified, and sequenced on the Illumina NovaSeq 6000 platform using paired-end 150 bp (PE150) sequencing (Xiong et al, 2025).

Recombinant protein PLCζ synthesis and microinjection of oocytes

Human PLCζ was expressed as a NusA-6×His fusion protein and purified by His tag Nanoselector Magnetic beads and His tag Nanoselector Agarose (HUABIO, Catalog# 004-101-003, 004-101-002). The pCold-NusA plasmid (P8497) was obtained from MiaoLingBio, China. The NusA-His-tagged PLCζ was tested for its ability to induce Ca²⁺ oscillations in oocytes prior to use and was microinjected at a pipette concentration of 0.01 mg/ml. As a control, NusA protein was purified in the same manner and injected separately at 0.01 mg/ml in KCl/Hepes buffer. For microinjection, mouse oocytes were injected with sperm using custom-made ICSI pipettes. Sperm-containing pipettes were advanced through the oocyte plasma membrane using piezo pulses delivered with a Prime Tech piezo manipulation system. Recombinant PLCζ was delivered into oocytes via fine-tip micropipettes inserted through the plasma membrane using pressure pulses. The processing of NusA and NusA-PLCζ is based on a blinding.

SKAP2 preparation and seminiferous tubules microinjection

The pGEX-6P-Skap2 plasmid was constructed using the Seamless DNA Assembly Plus Kit from MedChemExpress (Monmouth Junction, NJ, USA; Cat# HY-K1041) (Gan et al, 2025). The pGEX-6P-1 plasmid (P0005) was obtained from MiaoLingBio, China. BL21(DE3) competent E. coli was transformed with human or mouse Skap2 cDNA cloned into the pEGX-6P expression vector. Transfer the suspension to a 15 mL centrifuge tube (Bioland, ATS05-15). DMEM high-glucose culture medium from Abbkine (Cat# BMC1010). Fetal bovine serum (Cat No. 209111) and cell chamber culture plate (Cat No. 725431) were procured from NEST Biotechnology Co., Ltd. (Wuxi, China). Fetal bovine serum (Cat# FBS-300) was procured from Inner Mongolia Jinyuankang Biotechnology Co., Ltd. Penicillin–stroptomycin from Hangzhou Yangming Biotechnology Co., Ltd. (Amizona, Cat No. AP1001-100). Transformed cells were cultured for 12 h to induce protein expression and harvested by ultracentrifugation at 18,000 × g for 20 min. The supernatant containing soluble SKAP2 protein was collected into fresh 50 mL tubes. To construct the SKAP2 delivery system, milk-derived extracellular vesicles (mEVs) were isolated from raw milk via differential centrifugation and subsequently loaded with SKAP2 via electroporation using the CUY21EDIT II system (BEX, Japan). SKAP2 protein and mEVs were mixed at a 1:1 mass ratio in PBS, yielding a final mEV concentration of 0.1 mg/mL. The mixture was transferred to pre-chilled 0.4 cm electroporation cuvettes and subjected to 10 electroporation cycles under the following conditions: perforation voltage of 110 V, pulse duration of 6 ms, interval of 10 ms, secondary voltage of 25 V, and capacitance of 940 μF. After electroporation, the samples were incubated at 37 °C for 50 min to allow vesicle membrane recovery. Protein concentrations were determined using the Bradford Protein Assay Kit (Cat# abs580304, absin). Finally, the mEVs loaded with SKAP2 were microinjected into seminiferous tubules via the efferent ducts (Wang et al, 2023).

In vitro fertilization

Female donor mice were injected intraperitoneally with 5–10 IU of pregnant mare serum gonadotropin (PMSG, NSHF) at 18:00 on day 1 to stimulate follicular development, followed 44–48 h later by 5–10 IU of chorionic gonadotropin (CG, NSHF) to induce ovulation (Zhang et al, 2024a). Human tubal fluid (HTF) medium was pre-equilibrated in a CO₂ incubator at 37 °C on the same day. Spermatozoa were collected from the cauda epididymides of euthanized male mice and placed into HTF droplets under paraffin oil, then incubated at 37 °C in a CO₂ incubator for 1 h to undergo capacitation. Mature oocytes were co-incubated with capacitated sperm for 4–6 h. Fertilized embryos were identified in the fertilization medium, carefully isolated, washed thoroughly with fresh culture medium, and transferred to new drops of culture medium for subsequent development.

Intracytoplasmic sperm injection (ICSI)

ICSI was performed as previously described (Gan et al, 2024). Briefly, spermatozoa were collected from the epididymides of wild-type (WT) and knock-in (KI) mice and subjected to sonication to separate the heads from the tails. Individual sperm heads were injected into oocytes using a micromanipulator equipped with a Piezo-electric actuated pipette. Following injection, oocytes were cultured in KSOM medium (MR-107-D, Millipore) under mineral oil at 37 °C in a 5% CO₂ atmosphere. Fertilization outcomes were assessed by examining and imaging the oocytes 5–8 h post-injection.

F-actin/G-actin assay

The ratio of filamentous actin (F-actin) to globular actin (G-actin) was determined using the F-actin/G-actin In Vivo Assay Kit (Cytoskeleton Inc., Cat# BK037), following the previous instructions. Tissue samples were homogenized in a lysis buffer containing F-actin stabilization agents and incubated at 37 °C for 10 min. The lysates were then subjected to ultracentrifugation at 100,000 × g for 1 h at 37 °C using a pre-warmed Beckman Coulter Optima MAX-XP ultracentrifuge, separating the supernatant (containing G-actin) from the pellet (containing F-actin). The pellets were resuspended in depolymerization buffer and incubated on ice for 1 h to convert F-actin into G-actin. Samples were then mixed with 5× SDS sample buffer, heat-denatured at 95 °C, and resolved by SurePAGE™ (GenScript Corporation, Cat# M00657). Proteins were transferred to PVDF membranes (Merck) using a Bio-Rad wet transfer system at 300 mA for 2 h at 4 °C. After blocking with 5% non-fat milk in 1× TBST for 1 h at room temperature, the membranes were incubated overnight at 4 °C with the primary antibody provided in the kit. Detection was carried out using a HRP Goat anti-mouse IgG from FineTest (Cat# FNSA-0003), and chemiluminescent signals were visualized using Pierce™ Enhanced Chemiluminescent Substrate (Thermo Fisher) on a GE Healthcare Amersham imaging system.

Statistical analysis

The Shapiro-Wilk test determines whether the data follows a Normal Distribution. The F test checks whether the variances of different sample groups are equal. All quantitative data are expressed as mean ± standard deviation (s.d.). When the data follows a normal distribution and has homogeneity of variance, the two-sided Student’s t test should be used; when the data does not follow a normal distribution or lacks homogeneity of variance, the Mann–Whitney U test should be employed, with GraphPad Prism version 10.1.2. Details of statistical significance are provided in the Figures and legends.

Supplementary information

Table EV1 Table EV2 Table EV3 Table EV4 Table EV5 Appendix Peer Review File Source data Fig. 1 Source data Fig. 2 Source data Fig. 3 Source data Fig. 4 Source data Fig. 5 Source data Fig. 6 Source data Fig. 7 Expanded View Figures