Abstract

   Rationale: Spermatogenesis is a highly organized cell differentiation
   process in mammals, involving mitosis, meiosis, and spermiogenesis.
   DIS3L2, which is primarily expressed in the cytoplasm, is an RNA
   exosome-independent ribonuclease. In female mice, Dis3l2-deficient
   oocytes fail to resume meiosis, resulting in arrest at the germinal
   vesicle stage and complete infertility. However, the role of DIS3L2 in
   germ cell development in males has remained largely unexplored.

   Methods: We established a pre-meiotic germ cell conditional knockout
   mouse model and investigated the biological function of DIS3L2 in
   spermatogenesis and male fertility through bulk RNA-seq and scRNA-seq
   analyses.

   Results: This study unveils that conditional ablation of Dis3l2 in
   pre-meiotic germ cells with Stra8-Cre mice impairs spermatogonial
   differentiation and hinders spermatocyte meiotic progression coupled
   with cell apoptosis. Such conditional ablation leads to defective
   spermatogenesis and sterility in adults. Bulk RNA-seq analysis revealed
   that Dis3l2 deficiency significantly disrupted the transcriptional
   expression pattern of genes related to the cell cycle, spermatogonial
   differentiation, and meiosis in Dis3l2 conditional knockout testes.
   Additionally, scRNA-seq analysis indicated that absence of DIS3L2 in
   pre-meiotic germ cells causes disrupted RNA metabolism, downregulated
   expression of cell cycle genes, and aberrant expression of
   spermatogonial differentiation genes, impeding spermatogonial
   differentiation. In meiotic spermatocytes, loss of DIS3L2 results in
   disturbed RNA metabolism, abnormal translation, and disrupted meiotic
   genes that perturb meiotic progression and induce cell apoptosis,
   leading to subsequent failure of spermatogenesis and male infertility.

   Conclusions: Collectively, these findings highlight the critical role
   of DIS3L2 ribonuclease-mediated RNA degradation in safeguarding the
   correct transcriptome during spermatogonial differentiation and
   spermatocyte meiotic progression, thus ensuring normal spermatogenesis
   and male fertility.

   Keywords: spermatogonia, meiosis, spermatogenesis, infertility,
   scRNA-seq

Introduction

   Spermatogenesis is a complex process of cell differentiation that
   results in the production of mature spermatozoa, sustaining male
   fertility throughout adulthood. It consists of three distinct phases:
   mitosis, meiosis, and spermiogenesis [41]^1^, [42]^2. Spermatogonial
   stem cells undergo amplification, generating a population of
   undifferentiated spermatogonia, which further differentiate into
   differentiated spermatogonia. Type B spermatogonia give rise to two
   primary spermatocytes. These cells undergo processes including DNA
   replication, synapsis, and meiotic recombination and ultimately divide
   twice to produce haploid round spermatids, initiating spermiogenesis
   and forming mature spermatozoa [43]^3.

   Considerable investigations have focused on discovering developmentally
   regulated transcriptional networks required for mouse spermatogenesis
   [44]^4^-[45]^7. Multiple RNases and co-factors are involved in
   developmentally the correct transcriptome to ensure normal
   spermatogenesis in mice [46]^8^-[47]^11. For example, the catalytic
   ribonuclease EXOSC10 (yeast homolog, RRP6) mediates RNA decay in the
   nucleus and nucleolus [48]^12^, [49]^13. EXOSC10 is associated with
   epigenetic chromosome silencing and is essential for male germ cell
   proliferation and development [50]^8. During the early leptotene and
   zygotene stages of male germ cell meiosis, transcription remains
   inactive when synapsis occurs [51]^14. Transcription resumes at the
   pachytene stage, coinciding with meiotic recombination
   [52]^15^-[53]^17. A recent study has reported that the zygotene to
   pachytene transition is not only associated with transcription
   resumption but also with a programmed wave of mRNA degradation, which
   is essential for meiotic progression [54]^9. Thus, investigations of
   the transcriptome networks provide insights into the underlying
   molecular mechanisms that support spermatogenesis and ensue male
   fertility in mammals.

   DIS3-like exonuclease 2 (DIS3L2), a homolog of DIS3, is an RNA-exosome
   independent ribonuclease primarily expressed in the cytoplasm [55]^18^,
   [56]^19. Germline mutations in DIS3L2 have been reported to lead to
   Perlman syndrome, characterized by overgrowth, and increased
   susceptibility to Wilms tumor [57]^19. Recently, it has been
   demonstrated that DIS3L2-mediated decay safeguards endoplasmic
   reticulum-targeted mRNA translation and maintains calcium ion
   homeostasis [58]^20. Additionally, DIS3L2 has been shown to regulate
   cell proliferation and tissue growth in both humans and flies [59]^21.
   Furthermore, recent studies have delved into the role of DIS3L2 in
   female reproduction, revealing that Dis3l2-deficient oocytes nearly
   arrest at the germinal vesicle stage, resulting in total infertility in
   female mice [60]^22. Upon DIS3L2 depletion, uridylated-poly(A) RNAs
   remain intact and become prevalent in the transcriptome of
   Dis3l2-deficient oocytes. These uridylated-poly(A) RNAs possess shorter
   poly(A) tails, significantly reducing their translational activity
   [61]^22. Despite the emerging understanding of DIS3L2's functions in
   female reproduction and associated human diseases, its significance in
   male reproduction remains unclear.

   In this study, germ cell Dis3l2 conditional knockout (cKO) mice were
   established, and the results revealed that the loss of DIS3L2 severely
   impairs spermatogonial differentiation and hampers spermatocyte meiotic
   progression. This results in notable defects in spermatogenesis and,
   ultimately, leads to male sterility. Comparative RNA-seq and scRNA-seq
   analyses revealed severe dysregulation of the transcriptome during
   spermatogonial differentiation and spermatocyte meiosis in Dis3l2 cKO
   testes. Altogether, the data of this study strongly suggest that DIS3L2
   plays a pivotal role in sculpting the transcriptome during
   spermatogonial differentiation and spermatocyte meiotic progression,
   ensuring normal spermatogenesis and male fertility.

Results

Dis3l2 expression in the mouse testes

   It has previously been reported that DIS3L2 functions as an RNA-exosome
   independent ribonuclease and is primarily expressed in the cytoplasm.
   To investigate the role of DIS3L2 in mouse testes, Dis3l2 transcript
   levels were initially assessed using recently published scRNA-seq data
   from adult mouse testes [62]^4. Dis3l2 transcripts were detected in
   various cell types within the adult mouse testes, including
   spermatogonia, spermatocytes, spermatids, Sertoli cells, and Leydig
   cells ([63]Figure S1A, B). To further delineate the expression pattern
   of Dis3l2 transcripts in testicular cell types, given the absence of
   commercially available antibodies for immunofluorescence, an in situ
   hybridization mRNA analysis of Dis3l2 was conducted in the testes of
   P8, P12, and P35 wild-type mice. These analyses conclusively
   demonstrated that Dis3l2 was indeed expressed in the spermatogonia,
   spermatocytes, and round spermatids ([64]Figure S1C). These findings
   indicate the presence of Dis3l2 in pre-meiotic germ cells and
   spermatocytes, thus providing the foundation for a conditional ablation
   of Dis3l2 in the testes to elucidate the function of DIS3L2 in mouse
   spermatogenesis.

DIS3L2 is essential for spermatogenesis and male fertility

   To investigate the role of DIS3L2 in spermatogenesis, Dis3l2 was
   conditionally disrupted by crossing Dis3l2^Floxed/Floxed (Dis3l2^F/F)
   mice ([65]Figure S2A) with Stra8-Cre mice (Figure [66]1A). In these
   mice, CRE recombinase is initially expressed at P3, affecting part of
   spermatogonia through all pre-leptotene stage spermatocytes [67]^23.
   Stra8-Cre; Dis3l2^F/- mice (referred to as Dis3l2 cKO) and their
   littermate controls were obtained and confirmed through genomic PCR
   assay ([68]Figure S2B). In situ hybridization mRNA analysis of Dis3l2
   in P12 control and Dis3l2 cKO mouse testes revealed the absence of
   Dis3l2 in pre-meiotic germ cells and spermatocytes (Figure [69]1B).
   Moreover, quantitative RT-PCR assay revealed a significant reduction in
   the Dis3l2 mRNA expression level in P12 Dis3l2 cKO testes compared with
   those of controls ([70]Figure S1D). In search of a commercially
   available antibody of DIS3L2, one antibody was finally identified and
   was available for immunoblot. Immunoblot assay documented a significant
   reduction in DIS3L2 protein abundance in the P12 Dis3l2 cKO testes
   compared with control testes ([71]Figure S1E). These results indicate
   the successful establishment of a conditional KO mouse line with
   specific ablation of Dis3l2 in pre-meiotic germ cells and meiotic
   spermatocytes. Subsequently, the reproductive phenotypes of Dis3l2 cKO
   male mice were investigated. Over a 6-month mating period, Dis3l2 cKO
   males proved to be completely sterile (Figure [72]1C) and exhibited
   significantly smaller testes with a reduced ratio of testis to body
   weight at 3 months (Figure [73]1D, E). The histological analysis
   documented a range of anomalies in Dis3l2 cKO testes (Figure [74]1F).
   The great majority of the seminiferous tubules were devoid of germ
   cells and lined only with Sertoli cells (asterisks), some tubules
   featured multinucleated giant Sertoli cells (arrows), a few tubules
   exhibited reduced numbers of spermatogenic cells and vacuoles, and very
   few tubules displayed normal spermatogenesis with all stages of
   spermatogenic cells (star). To further validate the observed defects in
   spermatogenesis in Dis3l2 cKO testes as shown by PAS staining, the
   dual-immunofluorescence staining of DEAD-box helicase 4 (DDX4, a germ
   cell-specific marker) and Wilms tumor 1 (WT1, a Sertoli cell-specific
   marker) was performed (Figure [75]1G). The co-immunostaining results
   revealed that only a single layer of DDX4-positive cells was present at
   the basement membrane of the epithelium, and WT1-positive cells were
   fused, forming multinucleated giant Sertoli cells (dashed circles) in
   some mutant tubules, consistent with observations in testicular
   sections after PAS staining. In Dis3l2 cKO testes with only Sertoli
   cells, a single layer of germ cells was observed at the epithelium
   (Figure [76]1G), indicating disruption in spermatogenic cell
   differentiation. To identify these germ cells at the epithelium, DDX4
   and promyelocytic leukemia zinc finger, PLZF (official name ZBTB16, a
   marker for undifferentiated spermatogonia) [77]^24, were co-stained
   (Figure [78]1H). The results indicated that nearly all DDX4-positive
   cells were also PLZF-positive, suggesting they were undifferentiated
   spermatogonia. Co-immunostaining of PLZF and Cyclin D1 (a marker for
   mitotically active spermatogonia) demonstrated that these PLZF-positive
   spermatogonia displayed high mitotic activity in adult Dis3l2 cKO
   testes ([79]Figure S3A), implying normal maintenance of
   undifferentiated spermatogonia. As Stra8-Cre-mediated recombination
   starts from type A[1] spermatogonia, co-expression of GDNF family
   receptor alpha 1 (GFRA1) and Cyclin D1 indicated that GFRA1-positive
   undifferentiated spermatogonia (potential spermatogonial stem cells)
   were mitotically active ([80]Figure S3B), revealing normal self-renewal
   and proliferation of spermatogonial stem cells in adult Dis3l2 cKO
   testes. Immunofluorescence staining with antibodies to DDX4 and stem
   cell growth factor receptor (KIT, a marker for differentiated
   spermatogonia) documented that KIT-positive spermatogonia were barely
   detected in adult Dis3l2 cKO testes, whereas a population of
   KIT-positive spermatogonia was present in adult control testes
   ([81]Figure S3C), suggesting compromised spermatogonial differentiation
   in adult Dis3l2 cKO testes. Collectively, these results imply a
   significant compromise in spermatogonial differentiation and
   spermatogenic lineage development in adult Dis3l2 cKO testes.
   Consistently, almost no spermatozoa were present in the cauda
   epididymis of Dis3l2 cKO mice (Figure [82]1I). Taken together, these
   findings underscore the essential role of DIS3L2 in mouse
   spermatogenesis, with its ablation resulting in dramatic loss of
   spermatogenic cells and subsequent male infertility.

Figure 1.

   [83]Figure 1
   [84]Open in a new tab

   DIS3L2 is required for spermatogenesis and male fertility. (A)
   Schematic diagram showing the establishment of the Dis3l2 cKO male
   mouse model. (B) In situ hybridization mRNA analysis of Dis3l2 in P12
   control and Dis3l2 cKO mouse testes. Scale bar, 50 μm. (C) Fertility
   test showing the sterility of Dis3l2 cKO male mice. Mean number of
   litters ± s.d. ***P < 0.001. (D, E) Morphology of testes (D) and the
   ratio of testis to body weight (E) in control and Dis3l2 cKO mice at 3
   months. Scale bar, 1 mm. Mean ± s.d., n = 4 biologically independent
   testes from four different animals. ***P < 0.001. (F) Testicular
   sections of control and Dis3l2 cKO mice collected at 3 months and
   stained with periodic acid-Schiff and hematoxylin. Asterisks indicate
   Sertoli cell-only tubules, star indicates normal seminiferous tubule,
   and arrows indicate multinucleated giant Sertoli cells. Scale bar, 50
   μm. (G) Immunofluorescence staining of testis tissue sections of
   control and Dis3l2 cKO mice at 3 months after staining with antibodies
   against DDX4 (left) or WT1 (middle), followed by staining with Hoechst
   33342 to detect DNA (right). Scale bar, 50 μm. Dashed circles indicate
   multinucleated giant Sertoli cells. (H) Dual-immunofluorescence
   staining with DDX4 and PLZF antibodies in the testis tissue of control
   and Dis3l2 cKO mice at 3 months. Dashed rectangles indicate enlarged
   areas. Scale bar, 50 μm. (I) Cauda epididymides of control and Dis3l2
   cKO mice at 3 months and stained with periodic acid-Schiff and
   hematoxylin. Scale bar, 50 μm. Representative of n = 3 (B, F-I)
   independent biological replicates with similar results per condition.

Dis3l2 deletion causes severe spermatogenic cell loss during the first wave
of spermatogenesis

   To examine the earliest stage of spermatogenic defects, this study
   compared and analyzed testes isolated from P3 to P35 control and Dis3l2
   cKO mice using immunohistochemistry with an anti-DDX4 antibody.
   Immunohistochemical assessments revealed that Dis3l2 cKO testes
   displayed normal germ cell numbers at P3 but exhibited a significant
   reduction in germ cell numbers by P7, indicating a defect in
   spermatogonial development (Figure [85]2A, C). From P10 to P35, control
   testes exhibited normal meiotic progression and completed the first
   wave of spermatogenesis (Figure [86]2A). Conversely, Dis3l2 cKO testes
   showed reduced numbers of spermatocytes during the first wave of
   spermatogenesis (Figure [87]2A, C). In P21 and P35 Dis3l2 cKO testes,
   DDX4-positive germ cells in the center of the seminiferous tubules were
   noticeably absent, leaving only a single layer of DDX4-positive germ
   cells at the epithelium (Figure [88]2A). This suggests defective
   spermatogenic lineage development, consistent with the defect observed
   in adult Dis3l2 cKO testes. To delve deeper into the defects in
   spermatocyte development, testes from P10, P11, P12, P13, and P17
   control and Dis3l2 cKO mice were analyzed.

Figure 2.

   [89]Figure 2
   [90]Open in a new tab

   DIS3L2 is essential for spermatocyte survival and the first wave of
   spermatogenesis. (A) Histological and immunohistochemical analyses of
   DDX4 in the testicular tissue sections of control and Dis3l2 cKO mice
   from P3 to P35. Scale bar, 50 μm. (B) Immunohistochemical analysis
   using a polyclonal DDX4 antibody was performed in the testis tissues of
   control and Dis3l2 cKO mice from P10 to P17. Some of the images in (B)
   are enlarged from those in (A). Arrowheads indicate germ cells with
   pyknotic nuclei. Scale bar, 50 μm. (C) Statistical analysis of
   DDX4-positive cells per tubule in the testis tissue of control and
   Dis3l2 cKO mice from P3 to P17. Mean ± s.d., n = 3 biologically
   independent testes from three different animals; ns, no significance,
   **P < 0.01, and ****P < 0.0001. Representative of n = 3 (A, B)
   independent biological replicates with similar results per condition.

   Histological and immunohistochemical results indicated that Dis3l2 cKO
   spermatogonia were able to enter meiosis at P10, but experienced
   significant spermatocyte loss from P11 to P13 (Figure [91]2B). Pyknotic
   nuclei and shrunken cytoplasm of germ cells were frequently observed
   during this period (Figure [92]2B, arrowheads), indicating severe germ
   cell apoptosis. By as early as P17, germ cells in the central lumen of
   the seminiferous tubules had vanished in Dis3l2 cKO testes (Figure
   [93]2B), reflecting impaired meiosis. Quantitative analysis of germ
   cell numbers further corroborated the drastic reduction of
   spermatogenic cells during meiosis (Figure [94]2C). In summary, these
   findings demonstrate the critical role of DIS3L2 in the development of
   primary spermatocytes and highlight that the ablation of DIS3L2 leads
   to substantial germ cell death and significant loss of spermatogenic
   cells during the first wave of spermatogenesis.

DIS3L2 ablation impairs spermatogonial differentiation

   Stra8-Cre induces the expression of the CRE enzyme in advanced germ
   cells before meiosis [95]^23 and a significant decline of germ cells
   was observed in P7 Dis3l2 cKO testes. Thus, spermatogonial development
   was examined. First, the overall number of germ cells and
   undifferentiated spermatogonia were determined by whole-mount staining
   using antibodies to DDX4 and PLZF. The number of DDX4-positive germ
   cells and PLZF-positive spermatogonia were significantly reduced in P7
   and P10 Dis3l2 cKO tubules (Figure [96]3A, B). This finding was
   confirmed by immunofluorescence staining in P7 testicular
   cross-sections that documented a significant reduction in the number of
   PLZF-positive cells per tubule in Dis3l2 cKO testes compared with
   control testes (Figure [97]3C, E). Next, spermatogonial differentiation
   was investigated in Dis3l2 cKO tubules. Immunofluorescence staining
   with antibodies targeting DDX4 and KIT documented that the number of
   KIT-positive spermatogonia per tubular cross-section was remarkably
   decreased in P7 Dis3l2 cKO testes compared with controls (Figure
   [98]3D, F). Additionally, TUNEL-positive signals were significantly
   increased in Dis3l2 cKO testes compared with the controls at P7
   ([99]Figure S4). Statistical analyses revealed a significant increase
   in the percentage of PLZF-positive cells and a dramatic decrease in the
   percentage of KIT-positive cells in P7 Dis3l2 cKO tubules (Figure
   [100]3G, H), suggesting that DIS3L2 is required for spermatogonial
   differentiation. Whole-mount co-staining with antibodies to DDX4 and
   KIT demonstrated a significant decline in the number of KIT-positive
   cells in P10 Dis3l2 cKO tubules (Figure [101]3I). Collectively, these
   findings indicate that conditional disruption of Dis3l2 in pre-meiotic
   germ cells causes a remarkable decline in the number of spermatogonia,
   especially the differentiated spermatogonial population.

Figure 3.

   [102]Figure 3
   [103]Open in a new tab

   DIS3L2 is required for spermatogonial development. (A, B)
   Immunofluorescence of whole-mount testes from P7 (A) and P10 (B)
   control and Dis3l2 cKO mice after staining with antibodies to DDX4
   (left) or PLZF (middle) and merged (right). Scale bar, 50 μm. (C)
   Immunofluorescence staining of P7 control and Dis3l2 cKO testicular
   sections with antibodies to DDX4 (left) or PLZF (middle) and merged
   with Hoechst 33342 to stain the DNA (right). Scale bar, 50 μm. (D) Same
   as (C), but with antibodies to DDX4 and KIT. (E, F) Statistical
   analysis of PLZF-positive cells (E) and KIT-positive cells (F) per
   tubule from P7 control and Dis3l2 cKO testes. Mean ± s.d., n = 3
   biologically independent testes from three different mice. **P < 0.01
   and ****P < 0.0001. (G, H) Statistical analysis of the ratio of
   PLZF-positive cells to DDX4-positive cells (G) and KIT-positive cells
   to DDX4-positive cells (H) in P7 control and Dis3l2 cKO testes. Mean ±
   s.d., n = 3 biologically independent replicates. *P < 0.05. (I)
   Whole-mount staining of testes from P10 control and Dis3l2 cKO mice
   after staining with antibodies to DDX4 (left) or KIT (middle) and
   merged (right). Scale bar, 50 μm. Representative of n = 3 (A-D, I)
   independent biological replicates with similar results per condition.

DIS3L2 is required for spermatocyte survival and meiosis

   To delve deeper into the meiotic defects in Dis3l2 cKO testes, the
   expression pattern of the meiotic marker gene, synaptonemal complex
   protein 3 (SYCP3), a vital component of the synaptonemal complex, was
   examined from P10 to P17 using immunohistochemistry analysis. In
   control mouse testes, germ cells progressed through leptotene,
   zygotene, and pachytene, and eventually reached the diplotene stage
   during meiotic progression from P10 to P17 (Figure [104]4A).
   Conversely, SYCP3 exhibited a significantly reduced expression pattern
   in Dis3l2 cKO testes (Figure [105]4A). Both the ratio of SYCP3-positive
   seminiferous tubules and the number of SYCP3-positive germ cells per
   tubule were markedly decreased in Dis3l2 cKO testes compared with
   control testes from P10 to P17 (Figure [106]4B, C). Notably, an initial
   increase in the ratio of SYCP3-positive tubules and numbers of
   SYCP3-expressing cells per tubule from P10 to P12 was observed in
   Dis3l2 cKO testes, followed by a substantial reduction from P13 to P17
   (Figure [107]4B, C). By P17, very few SYCP3-positive cells were
   observed in the seminiferous tubules of Dis3l2 cKO testes (Figure
   [108]4A). These results indicate that germ cells in Dis3l2 cKO testes
   initiate meiosis and progress into the zygotene stage at P12 and P13,
   but subsequently undergo increased spermatocyte apoptosis and loss
   shortly afterwards. This was further corroborated by
   immunohistochemistry using SYCP3 and TUNEL assay (Figure [109]4D, E).
   Numerous small vacuoles were observed in the seminiferous tubules
   (marked by white dashed lines), and a few SYCP3-positive cells
   (highlighted by yellow arrowheads) displayed characteristic pyknotic
   nuclei in the centre of the seminiferous tubules in Dis3l2 cKO testes
   at P12 and P13 (Figure [110]4D). TUNEL assays revealed a significant
   increase in apoptotic signals in Dis3l2 cKO testes at P12 and P13
   (Figure [111]4E). These findings unequivocally demonstrate that Dis3l2
   deficiency leads to severe spermatocyte loss, highlighting the
   essential role of DIS3L2 in normal meiosis.

Figure 4.

   [112]Figure 4
   [113]Open in a new tab

   DIS3L2 is required for male germ cell meiosis. (A) Histological and
   immunohistochemical analyses of SYCP3 in the testis tissue of control
   and Dis3l2 cKO mice from P10 to P17. Scale bar, 50 μm. (B) Statistical
   analysis of the ratio of SYCP3-positive tubules in the testis tissue of
   control and Dis3l2 cKO mice from P10 to P17. Mean ± s.d., n = 3
   biologically independent testes from three different mice. **P < 0.01,
   ***P < 0.001, and ****P < 0.0001. (C) Statistical analysis of the
   number of SYCP3-positive cells per tubule in the testis tissue of
   control and Dis3l2 cKO mice from P10 to P17. Mean ± s.d., n = 3
   biologically independent testes from three different mice. *P < 0.05,
   and ****P < 0.0001. (D) Immunohistochemical analysis of SYCP3 in the
   testis tissue of control and Dis3l2 cKO mice at P12 and P13. Some of
   the images in (D) are enlarged from those in (A). Scale bar, 50 μm.
   Yellow arrowheads indicate spermatocytes with pyknotic nuclei. Dashed
   circles represent tubules with vacuoles. (E) TUNEL assay in the testis
   tissue of control and Dis3l2 cKO mice at P12 and P13. DNA was stained
   with Hoechst 33342. Scale bar, 50 μm. The right panel shows the
   statistical analysis of the number of TUNEL-positive cells per tubule
   in the testis tissue of control and Dis3l2 cKO mice at P12 and P13.
   Mean ± s.d., n = 3 biologically independent testes from three different
   mice. **P < 0.01. Representative of n = 3 (A, D, and E) independent
   biological replicates with similar results per condition.

DIS3L2 is required for spermatocyte meiotic progression

   In male germ cells, meiosis is initiated around P10, with primary
   spermatocytes progressing through the stages of leptotene (P10),
   zygotene (P12), pachytene (P14), and reaching the diplotene stage at
   P17 (Figure [114]5A). γH2AX, a protein involved in double-strand break
   repair, exhibits robust expression in the nucleus from leptotene to
   zygotene stages and is enriched in the sex body (XY body) during the
   pachytene and diplotene stages. To characterize the meiotic deficiency
   in Dis3l2 cKO testes, this study performed co-staining of DDX4 and
   γH2AX in seminiferous tubules at P13 and P15. Immunofluorescence
   staining revealed that control testes contained numerous sex-body
   positive cells and tubules at P13 and P15 (Figure [115]5B, C),
   indicating that a substantial number of spermatocytes had progressed
   into the pachytene stage and beyond. However, in Dis3l2 cKO testes,
   sex-body positive spermatocytes were less prevalent in the tubules at
   P13 and nearly absent in the tubules at P15 (Figure [116]5B, C). The
   ratio of sex-body positive tubules was also significantly reduced in
   Dis3l2 cKO testes compared with control testes at P13 and P15 (Figure
   [117]5D). The majority of spermatocytes in P13 and P15 Dis3l2 cKO
   testes displayed strong γH2AX signals in the nuclei, indicating that
   leptotene/zygotene spermatocytes predominated in Dis3l2 cKO testes and
   their transition to the pachytene stage and beyond was severely
   disrupted.

Figure 5.

   [118]Figure 5
   [119]Open in a new tab

   DIS3L2 is required for spermatocyte meiotic progression. (A) Schematic
   view of the four stages including pre-leptotene/leptotene (PL/L),
   zygotene (Z), pachytene (P), and diplotene (D) of meiotic prophase I
   corresponding to their developmental time point. (B) Immunofluorescence
   analysis results of P13 testes from control and Dis3l2 cKO mice after
   co-staining with antibodies to DDX4 (germ cells) and γH2AX
   (spermatocytes) as well as Hoechst 33342 (DNA). Scale bar, 50 μm.
   Dashed circles indicate tubules containing sex-body-positive
   spermatocytes. (C) The same as (B), but at P15. (D) Statistical
   analysis of the ratio of tubules containing sex-body-positive
   spermatocytes in the testis tissue of control and Dis3l2 cKO mice at
   P13 and P15. Mean ± s.d., n = 3 biologically independent testes from
   three different mice. ****P < 0.0001. (E) Chromosome spreads of
   spermatocytes from P13 control and Dis3l2 cKO mice. Co-staining was
   performed with antibodies to SYCP3 and γH2AX. Scale bar, 10 μm. (F)
   Statistical analysis of the proportion of leptotene, zygotene, and
   pachytene spermatocytes in the P13 testis tissue of control and Dis3l2
   cKO mice. Mean ± s.d., n = 3 biologically independent testes from three
   different mice. *P < 0.05, **P < 0.01, and ****P < 0.0001.
   Representative of n = 3 (B, C, and E) biologically independent
   replicates with similar results per condition.

   To precisely delineate the meiotic progression defects, this study
   prepared chromosome spreads of spermatocytes isolated from control and
   Dis3l2 cKO mouse testes at P13 and stained them with SYCP3 and γH2AX
   antibodies.

   Control testes harbored approximately 20.67%, 30.98%, and 48.35% of
   spermatocytes at the leptotene, zygotene, and pachytene stages,
   respectively (Figure [120]5E, F). While all three stages were
   observable in the chromosome spreads of Dis3l2 cKO testes, only about
   9.17% of spermatocytes were in the pachytene stage, with 36.33% and
   54.50% of spermatocytes at the leptotene and zygotene stages,
   respectively (Figure [121]5E, F), suggesting impaired spermatocyte
   meiotic progression. In addition, chromosome synapsis was investigated
   and chromosome spreads of spermatocytes were prepared from P13 control
   and Dis3l2 cKO testes. Co-staining of SYCP3 and synaptonemal complex
   protein 1 (SYCP1) documented that SYCP3 and SYCP1 signals were
   completely co-localized on the autosomes in control pachytene
   spermatocytes, while discontinued SYCP1 signals were frequently
   observed on the autosomes in Dis3l2 cKO pachytene spermatocytes
   ([122]Figure S5A), suggesting abnormal chromosome synapsis in Dis3l2
   cKO testes. Consistently, statistical analysis indicated that the
   proportion of abnormal synaptonemal pachytene spermatocytes was
   significantly increased in Dis3l2 cKO testes compared with control
   testes ([123]Figure S5B). Collectively, these results unequivocally
   demonstrate impaired transition of Dis3l2-deficient spermatocytes from
   the leptotene/zygotene stage to the pachytene stage and beyond,
   underscoring the essential role of DIS3L2 in normal spermatocyte
   meiotic progression.

Dis3l2 deficiency causes dysregulation of transcripts

   To investigate the molecular repercussions of Dis3l2 depletion in
   spermatogenesis, bulk RNA-seq was performed in comparison with the
   transcriptome of control and Dis3l2 cKO testes at P12. Sample
   clustering using principal component analysis (PCA) and the heatmap of
   differentially expressed genes (DEGs) revealed striking similarity
   among the three biological replicates within each condition,
   underlining the significance of the differences between the two
   conditions (Figure [124]6A, B), thereby affirming the reliability of
   the sequence data. The heatmap and volcano plot distinctly portrayed
   the transcriptional patterns between the two conditions and unveiled a
   series of DEGs (Figure [125]6B, C). The RNA-seq analysis pinpointed
   6,023 DEGs, of which 3,277 transcripts were significantly upregulated
   and 2,746 genes were downregulated in Dis3l2 cKO testes, with an
   adjusted P-value of < 0.05 as the cutoff, as determined by DESeq2
   (Figure [126]6C). It was apparent that a large number of genes were
   dysregulated and the number of upregulated genes was higher than that
   of downregulated genes, suggesting defective RNA degradation in
   response to Dis3l2 deletion. The DEGs were categorized into various
   biotypes to discern the potential targets of DIS3L2 in the testes.
   Notably, long intergenic noncoding RNA, processed transcript,
   antisense, processed pseudogene, and protein coding genes (mRNA) were
   the primary biotypes among the DEGs ([127]Figure S6A). Collectively,
   these data indicate that conditional ablation of DIS3L2 ribonuclease
   causes significant dysregulation of transcripts in the testes.

Figure 6.

   [128]Figure 6
   [129]Open in a new tab

   DIS3L2 deletion disrupts the transcriptome in the testis tissue of
   Dis3l2 cKO mice. (A) Principal component analysis (PCA) of RNA-seq data
   from P12 control and Dis3l2 cKO mice. (B) Heatmap of differentially
   expressed genes (DEGs) between control and Dis3l2 cKO mice. (C) Volcano
   plot showing DEGs using an adjusted P-value of < 0.05 as the cutoff
   value by DESeq2. (D) Gene set enrichment analysis of biological process
   categories of Dis3l2 cKO testes in comparison with control testes (left
   panel) using gene set enrichment analysis (GSEA) software program and
   molecular signatures database (MSigDB) C5 GO gene sets. Examples of the
   enrichment plot for Gene Ontology terms of epithelial tube
   morphogenesis and meiotic cell cycle (right panel). (E) RNA-seq results
   of selected transcripts (log[2]-fold change) related to the
   spermatogenic process in the testis tissue of Dis3l2 cKO mice. (F)
   Quantitative RT-PCR assay for the validation of genes with
   downregulated expression involved in spermatogonial development,
   meiotic process, and spermatogenesis. For comparison, the abundance
   (relative to β-actin) of each gene in the testis tissue of control mice
   was set to 1. Data are presented as mean ± s.d. for n = 3 biologically
   independent samples per condition.

   Subsequently, gene set enrichment analysis (GSEA) was conducted to
   identify the enriched biological processes (gene ontology [GO] terms)
   and pathways in Dis3l2 cKO testes. GO enrichment analysis revealed that
   upregulated transcripts were primarily associated with processes such
   as angiogenesis, epithelial tube morphogenesis, cell-substrate
   adhesion, tissue migration, carbohydrate metabolic process, skeletal
   system development, regulation of actin filament-based processes, cell
   growth, and mesenchyme development (Figure [130]6D). By contrast,
   downregulated transcripts were significantly enriched in GO terms
   related to biological processes such as meiotic cell cycle, chromosome
   segregation, DNA repair, DNA recombination, DNA replication, regulation
   of DNA metabolic process, ribonucleoprotein complex biogenesis, germ
   cell development, mRNA processing, and RNA splicing (Figure [131]6D).
   Consistently, pathway enrichment analysis revealed that processes such
   as lipid metabolism, Rho GTPase cycle, extracellular matrix
   organization, hemostasis, development biology, carbohydrate metabolism,
   cellular response to stimuli, and fatty acid metabolism were notably
   elevated. By contrast, processes such as cell cycle, DNA double-strand
   break repair, chromosome maintenance, DNA replication, RNA metabolism,
   mRNA processing, regulation of TP53 activity, and mRNA splicing were
   markedly reduced in Dis3l2 cKO testes ([132]Figure S6B).

   The RNA-seq data highlighted a marked decrease in genes related to
   spermatogonial development, meiosis, and spermatogenesis in Dis3l2 cKO
   testes (Figure [133]6E). This study further validated the expression
   patterns of these transcripts through real-time RT-PCR, confirming a
   significant downregulated expression of genes associated with
   spermatogonial development (e.g., Plzf, Pramef12, Sohlh1, Sohlh2,
   Rhox10, and Dazl), meiotic processes (e.g., Mael, Cpeb1, Stra8, Rec8,
   Spo11, Sycp1, Sycp2, Sycp3, Syce1, Syce2, Hormad1, Dmc1, and Rad51),
   and spermatogenesis (e.g., Piwil1, Piwil2, and Tdrd9; Figure [134]6F).
   This underscores that genes crucial for spermatogonial development,
   meiosis, and spermatogenesis are significantly downregulated in the
   absence of DIS3L2. Collectively, the RNA-seq data suggest that the loss
   of DIS3L2 leads to a severe impairment in RNA degradation, disrupting
   the delicate balance between RNA transcription and degradation. This
   disruption, in turn, leads to perturbations in RNA metabolism and the
   transcriptome during spermatogonial development and meiosis, ultimately
   culminating in impaired spermatogonial differentiation and spermatocyte
   meiotic progression.

scRNA-seq defines the transcriptome of Dis3l2 cKO spermatogonia

   To determine the spermatogonial composition and transcriptome changes
   in Dis3l2 cKO testes, single cells were isolated from P15 control and
   Dis3l2 cKO testes, and scRNA-seq analysis was performed using the 10×
   Genomics platform. After quality control, a total number of 10,419
   control and 8,731 Dis3l2 cKO testicular cells were retained for
   subsequent analysis. Uniform Manifold Approximation and Projection
   (UMAP) and marker gene analyses were performed for cell type
   identification of the combined testicular cells. Eight cell types were
   identified based on the expression patterns of known marker genes in
   the mouse testis, including spermatogonia (SPG), spermatocytes
   (Scytes), and six somatic cell populations: endothelial cells,
   macrophages, myoid cells, Leydig cells, stromal cells, and Sertoli
   cells ([135]Figure S7A, B). Statistical analysis revealed significant
   reductions in the cell number and percentage of both spermatogonia and
   spermatocytes in P15 Dis3l2 cKO testes compared with control testes
   ([136]Figure S7C, D), revealing abnormalities in spermatogonial
   development and spermatocyte meiosis in Dis3l2 cKO testes.

   To define how DIS3L2 ribonuclease affects cellular heterogeneity and
   transcriptome-wide signatures of spermatogonia, spermatogonia from P15
   testes were re-clustered. After filtering out cells of poor quality,
   3,743 spermatogonia were analyzed, and based on UMAP and marker gene
   analyses, four distinct spermatogonial subtypes (SPG1, SPG2, SPG3 and
   SPG4) were identified (Figure [137]7A, B). Based on the expression
   pattern of spermatogonial marker genes, SPG1 cells were assigned as
   SSCs, SPG2 as progenitor/undifferentiated spermatogonia, SPG3 as early
   differentiated spermatogonia, and SPG4 as late differentiated
   spermatogonia (Figure [138]7B). Monocle pseudotime analysis provided
   the developmental trajectory of spermatogonial cells from SPG1 to SPG4
   (Figure [139]7C). The results revealed that 7.27%, 21.16%, 26.83%, and
   44.75% cells were sorted into SPG1, SPG2, SPG3, and SPG4 subtypes,
   respectively, in control testes and 15.54%, 39.82%, 20.25%, and 24.40%
   cells were present in SPG1, SPG2, SPG3, and SPG4 subtypes,
   respectively, in Dis3l2 cKO testes (Figure [140]7D, E). A remarkable
   increase in the percentage of the SPG1 and SPG2 subtypes and a decrease
   in the percentage of the SPG3 and SPG4 subtypes in Dis3l2 cKO testes
   was observed. This indicates impairment of spermatogonial
   differentiation in the absence of DIS3L2 ribonuclease.

Figure 7.

   [141]Figure 7
   [142]Open in a new tab

   scRNA-seq analysis of Dis3l2 cKO spermatogonia. (A) UMAP plots of
   combined (left panel), control (middle panel), and Dis3l2 cKO (right
   panel) spermatogonia that had four distinct spermatogonial subtypes
   (SPG1, SPG2, SPG3, and SPG4). (B) Dot plot for the expression of
   selected marker genes across all identified cell types. (C) Pseudotime
   trajectory of the combined four spermatogonial subtypes. (D and E)
   Summary of detailed cell numbers and percentages of spermatogonia in
   each cell cluster in control and Dis3l2 cKO testes. (F) Volcano plots
   of DEGs in Dis3l2 cKO SPG3 (left panel) and SPG4 cells (right panel)
   using a cutoff of P < 0.05 and log[2] fold-change of > 0.2. (G) Venn
   diagrams of upregulated genes (top panel) and downregulated genes
   (bottom panel) between SPG3 and SPG4 cells. (H) GO analysis of
   downregulated transcripts in the SPG3 subtype (top panel) and the SPG4
   subtype (bottom panel) using Metascape.

   To uncover the underlying molecular causes of the perturbation in
   spermatogonial differentiation, differential gene expression analyses
   were conducted in SPG2, SPG3, and SPG4 cells. Using a cutoff of P <
   0.05 and log[2] fold-change > 0.2, a total of 519 DEGs, including 337
   upregulated and 182 downregulated genes, were identified in Dis3l2 cKO
   SPG2 cells, 806 DEGs (481 upregulated and 325 downregulated genes) in
   SPG3 cells, and 1,026 DEGs (406 upregulated and 620 downregulated) in
   SPG4 cells (Figure [143]7F and [144]Figure S8A). A majority of genes
   were dysregulated in response to DIS3L2 ablation in SPG2, SPG3, and
   SPG4 cells, suggesting defective RNA degradation. GO analysis of DEGs
   in SPG2 cells revealed that upregulated genes were predominantly
   enriched in processes related to stem cell population maintenance, the
   respiratory electron transport chain, and cell division, while
   downregulated transcripts were primarily involved in DNA replication,
   cell cycle phase transitions, mRNA processing, and translation
   ([145]Figure S8B). A large number of DEGs were overlapped between the
   SPG3 and SPG4 subtypes (Figure [146]7G), and the top 20 upregulated and
   downregulated transcripts of SPG3 and SPG4 subtypes were shown in
   Supplementary [147]Figure S8C-E. In assessing the enriched GO terms in
   SPG3 cells, transcripts with increased abundance were primarily
   involved in oxidative phosphorylation, cell division, and mRNA
   metabolic processes ([148]Figure S8C), and downregulated transcripts
   were significantly enriched in cell cycle, DNA replication, and
   metabolism of RNA (Figure [149]7H). Similar to SPG3, GO terms
   associated with energy-coupled proton transmembrane transport, cell
   cycle, mRNA metabolic processes, and translation were significantly
   enriched ([150]Figure S8D), while terms related to the cell cycle and
   RNA metabolism were markedly decreased in the SPG4 subtype of Dis3l2
   cKO cells (Figure [151]7H). Collectively, these results imply that the
   downregulation of cell cycle and DNA replication genes may be directly
   related to the observed spermatogonial differentiation defect in Dis3l2
   cKO testes. Therefore, several genes related to spermatogonial
   differentiation and cell cycle in P8 control and Dis3l2 cKO testes were
   examined. RT-qPCR analysis revealed that several spermatogonial
   differentiation genes (e.g., Stra8, Sohlh1, Sohlh2) and cell cycle
   genes (e.g., Cenpa, Cdk2ap2) were significantly decreased in Dis3l2 cKO
   testes ([152]Figure S9). Altogether, these findings suggest that
   alterations in spermatogonial differentiation genes and cell cycle
   genes upon loss of DIS3L2 ribonuclease may cause defective
   spermatogonial differentiation.

Transcriptome signatures of Dis3l2 cKO spermatocytes

   Differential gene expression analysis was performed on spermatocytes
   from scRNA-seq data and the enriched GO terms and pathways were
   determined in Dis3l2 cKO spermatocytes. In total, 1,722 upregulated and
   1,196 downregulated transcripts were identified in Dis3l2 cKO
   spermatocytes using a cutoff of P < 0.05 and log[2] fold-change of >
   0.2. When restricted to a more stringent P < 0.05 and log[2]
   fold-change of > 0.5, 690 genes were upregulated and 521 genes were
   downregulated (Figure [153]8A). The number of upregulated genes was
   higher than that of downregulated genes, suggesting insufficient RNA
   degradation upon DIS3L2 deletion. The top 20 of increased and decreased
   transcripts are shown in Figure [154]8B and D. GO analysis documented
   that increased transcripts were largely involved in the metabolism of
   RNA and regulation of translation, whereas downregulated genes were
   mainly enriched in meiosis I and mRNA processing in Dis3l2 cKO
   spermatocytes (Figure [155]8C, E). The downregulation of meiotic genes
   (e.g., Mael, Cpeb1, Stra8, Rec8, Spo11, Sycp1, Sycp2, Sycp3, Syce1,
   Syce2, Hormad1, Dmc1, and Rad51) had been verified by RT-qPCR analysis
   in P12 Dis3l2 cKO testes (Figure [156]6F). Immunoblot assay also
   confirmed the significant reduction of three meiosis-related proteins,
   namely RPA2, RAD51, and DMC1, in P12 Dis3l2 cKO testes ([157]Figure
   S10). Overall, these results indicate that Dis3l2 deficiency in
   spermatocytes causes aberrant RNA metabolism, disrupted translation,
   and dysregulated meiotic genes, leading to impaired progression of
   spermatocyte meiosis.

Figure 8.

   [158]Figure 8
   [159]Open in a new tab

   Transcriptome signatures of Dis3l2 cKO spermatocytes. (A) Volcano plot
   of DEGs in Dis3l2 cKO spermatocytes using a cutoff of P < 0.05 and
   log[2] fold change of > 0.2 (left panel) and P < 0.05 and log[2] fold
   change of > 0.5 (right panel). (B) Top 20 transcripts of upregulated
   DEGs. (C) Enriched GO terms and pathways of the upregulated genes
   analyzed using Metascape. (D) Top 20 downregulated transcripts. (E)
   Metascape analysis of GO terms and pathways for downregulated
   transcripts.

Comparison of RNA-seq data with scRNA-seq data

   To further investigate the transcriptome changes resulting from Dis3l2
   depletion in male germ cells, bulk RNA-seq data and scRNA-seq data were
   compared and analyzed. As Stra8-Cre-mediated recombination starts from
   type A[1] spermatogonia, we first compared the bulk RNA-seq data from
   testes with the scRNA-seq data from SPG3 cells. Among upregulated
   genes, 167 of 3,227 genes were shared among RNA-seq data (5.10%) and
   481 genes among scRNA-seq data (34.72%) ([160]Figure S11A). Among
   downregulated transcripts, 120 genes of a total of 2,746 genes were
   shared among RNA-seq data (4.37%) and 325 genes among scRNA-seq data
   (36.92%) ([161]Figure S11A). GO analysis documented that the overlapped
   genes were significantly enriched for cell division, cell cycle, DNA
   replication, mRNA processing, and translation ([162]Figure S11A). The
   abnormalities of these biological processes may directly relate to the
   observed defect in spermatogonial differentiation. Next, the bulk
   RNA-seq data from testes were compared with the scRNA-seq data from
   spermatocytes. The 531 increased transcripts that were shared
   represented 16.20% (3,277) and 30.84% (1,722), respectively, of the
   upregulated genes in the RNA-seq data and the scRNA-seq data
   ([163]Figure S11B). Likewise, the 617 transcripts that were decreased
   in both datasets reflected 22.47% (2,746) and 51.59% (1,196) of
   downregulated genes in the RNA-seq data and the scRNA-seq data,
   respectively ([164]Figure S11B). GO analysis revealed that the common
   DEGs were primarily enriched in meiosis I and male gamete generation
   ([165]Figure S11B). The disruption of genes involved in meiosis and
   male gamete generation may be responsible for impaired meiotic
   progression of spermatocytes. Taken together, the disrupted genes
   involved in cell cycle, mRNA processing, translation, meiosis, and male
   gamete generation are possible indirect DIS3L2-targeted substrates and
   may account for the observed defects in spermatogonial differentiation
   and spermatocyte meiotic progression.

Model of DIS3L2 in driving spermatogonial differentiation and meiotic
progression

   The findings indicate that DIS3L2-mediated cytoplasmic RNA decay is
   essential for normal spermatogonial differentiation and spermatocyte
   meiotic progression (Figure [166]9A). Conditional disruption of Dis3l2
   in pre-meiotic germ cells causes defective RNA metabolism,
   downregulation of cell cycle genes, and aberrant expression of
   spermatogonial differentiation genes that impede spermatogonial
   differentiation (Figure [167]9B). In spermatocytes, Dis3l2 deficiency
   results in abnormal RNA metabolism, aberrant translation, and disrupted
   meiotic genes that perturb spermatocyte meiotic progression and induce
   cell apoptosis (Figure [168]9B), leading to subsequent failure of
   spermatogenesis and male sterility.

Figure 9.

   [169]Figure 9
   [170]Open in a new tab

   Hypothetical model of DIS3L2 function in pre-meiotic and meiotic germ
   cells. (A) DIS3L2-mediated RNA decay is required for normal
   spermatogonial differentiation and spermatocyte meiosis. (B) Dis3l2
   deficiency causes abnormal RNA metabolism, disrupted cell cycle genes,
   and disturbed spermatogonial differentiation genes, leading to
   impairment of spermatogonial differentiation. In spermatocytes, loss of
   DIS3L2 causes abnormal RNA metabolism, disrupted translation, and
   dysregulated meiotic genes, resulting in impaired meiotic progression
   and increased cell apoptosis.

Discussion

   Many investigations have been dedicated to elucidating the
   transcriptional networks essential for mouse spermatogenesis [171]^4^,
   [172]^6^, [173]^25^-[174]^27. Numerous RNases and co-factors have been
   identified as regulators of the correct transcriptome in mice, crucial
   for ensuring normal spermatogenesis [175]^8^-[176]^11. For instance,
   EXOSC10, the yeast homolog of RRP6, is a catalytic ribonuclease
   associated with the RNA exosome complex. It has been demonstrated to
   play a pivotal role in both male and female reproduction. In male mice,
   EXOSC10 is indispensable for germ cell development, and the conditional
   deletion of Exosc10 in pre-meiotic germ cells results in
   spermatogenesis defects and subfertility [177]^8. Disruption of RNA
   degradation in Exosc10 conditional null oocytes leads to abnormalities
   in meiotic resumption and pre-implantation embryo development [178]^28.
   Recently, DIS3L2 ribonuclease has been shown to degrade
   terminal-uridylated RNA, ensuring oocyte maturation and female
   fertility [179]^22. However, whether DIS3L2 exerts a similar role in
   male spermatogenesis and fertility remains unclear.

   The Perlman syndrome is often associated with a high neonatal mortality
   rate, and there are few reports of long-term survivors. Most of the
   infants with DIS3L2 mutations develop respiratory distress and/or renal
   anomalies, and die within the few days of life [180]^29. Only ten
   patients with DIS3L2 mutations have been identified by the end of 2017,
   and most of them die before adulthood [181]^29. Therefore, no relevant
   clinical findings regarding reproductive abnormalities with DIS3L2
   mutations have been reported to date. Dis3l2^Δ11/Δ11 mice (exon
   11-deleted) display perinatal lethality, with no homozygous animals
   surviving the first postnatal day [182]^30. Whether DIS3L2 is involved
   in germ cell development in males remains unknown. The present study
   generated a conditional Dis3l2 KO mouse model using Stra8-Cre to
   investigate the function of DIS3L2 in spermatogenesis and male
   fertility. Our findings demonstrate that DIS3L2 ribonuclease is
   essential for normal spermatogenesis and male fertility. The
   conditional disruption of Dis3l2 in pre-meiotic germ cells disrupts
   spermatogonial differentiation and impairs spermatocyte meiotic
   progression. Bulk RNA-seq data provide compelling evidence that
   perturbations in RNA degradation lead to disruptions in RNA abundance,
   thereby dysregulating gene expression associated with cell cycle in
   differentiated spermatogonia and meiosis in spermatocytes. scRNA-seq
   analysis further revealed that the absence of DIS3L2 in differentiated
   spermatogonia causes abnormal RNA metabolism and dysregulated cell
   cycle genes, and the loss of DIS3L2 in spermatocytes leads to aberrant
   RNA metabolism and disrupted meiotic genes. Dis3l2 deficiency not only
   causes significant accumulation of transcripts but also results in the
   downregulation of numerous genes. Elevated RNA substrates or pathways
   in Dis3l2 cKO may serve as repressors of gene expression. Therefore,
   the downregulated transcripts may be an indirect effect of the loss of
   DIS3L2 ribonuclease. Collectively, these disturbances ultimately result
   in defects in spermatogonial differentiation, meiotic progression, and
   subsequent spermatogenic lineage development.

   Uridylation, the addition of uridine residues to the 3′ end of
   shortened poly(A) tails, has been associated with RNA degradation.
   DIS3L2 is known to selectively degrade uridylated RNAs, including
   uridylated pre-let-7 miRNAs and mRNAs [183]^18^, [184]^31^-[185]^33.
   TUT4 and TUT7 (collectively referred to as TUT4/7) are the primary
   cellular terminal uridylyltransferases responsible for mediating miRNA
   and mRNA 3′ uridylation [186]^34^-[187]^36. Previous reports have
   highlighted the crucial role of 3′ uridylation in shaping both male and
   female germline transcriptomes [188]^37. TUT4/7-mediated 3′ uridylation
   is indispensable for timely maternal mRNA clearance and the
   maternal-to-zygotic transition [189]^38. A recent study further
   underscores the significance of TUT4/7-mediated uridylation in clearing
   numerous transcripts expressed during the zygotene stage in pachytene
   spermatocytes. Deletion of both TUT4 and TUT7 in pre-meiotic germ cells
   by using Stra8-Cre mice leads to spermatogenic arrest in the late
   pachytene stage, accompanied by cell apoptosis, ultimately resulting in
   spermatogenic failure and male infertility [190]^9. Gene expression
   analysis documents that a total number of 857 DEGs (732 upregulated and
   125 downregulated genes) are identified in Tut4/7 cKO pachytene
   spermatocytes, suggesting defective RNA decay when removing an RNA
   degradation signal by 3′ uridylation. TAIL-seq analysis further
   indicates that deficiency of TUT4/7 reduces terminal uridylation of
   genes not upregulated in the Tut4/7 cKO pachytene spermatocytes by
   5.8-fold, whereas the uridylation of upregulated transcripts is
   completely dependent on TUT4/7 [191]^9. This highlights the critical
   role of 3′ uridylation-primed mRNA degradation in facilitating male
   meiotic progression.

   DIS3L2 exoribonuclease is responsible for targeting uridylated RNAs.
   The present study demonstrated that conditional ablation of Dis3l2 in
   pre-meiotic germ cells similarly leads to defective meiotic
   progression, accompanied by cell apoptosis. The transition from
   leptotene/zygotene to the pachytene stage is significantly impeded by
   the loss of DIS3L2. Comparative bulk RNA-seq and scRNA-seq data provide
   evidence that the absence of DIS3L2 ribonuclease severely disrupts the
   transcriptome and the transcriptional expression pattern of genes
   related to meiosis. Differential gene expression analysis from
   scRNA-seq data identified 1,722 upregulated and 1,196 downregulated
   transcripts in Dis3l2 cKO spermatocytes, suggesting defective RNA decay
   upon DIS3L2 deletion. These findings, in conjunction with prior
   research, affirm that TUT-DIS3L2-mediated RNA degradation is vital for
   meiotic progression during male germ cell meiosis. However, it is
   noteworthy that the deletion of DIS3L2 in spermatocytes results in an
   earlier defect in the transition from leptotene/zygotene to the
   pachytene stage, in contrast to Tut4/7-deficient spermatocytes that
   undergo arrest specifically at the pachytene stage. This implies that
   DIS3L2 may have functions independent of TUT4/7 in licensing meiotic
   progression.

   The fate of uridylated transcripts is modulated by tail editing for
   either degradation by DIS3L2 or readenylation. A previous study reports
   that terminal oligo uridylation is often fewer than five residues in
   mouse oocytes through TAIL-seq analysis [192]^37. Additionally,
   transcripts that have uridine at the 3′ termini are difficult to be
   isolated by the Poly(A) inclusive RNA isoform sequencing (PAIso-seq)
   method because the capture of RNA by PAIso-seq depends on the sequence
   of poly(A) tails. By employing various analyses such as Ribo-Minus
   RNA-seq, poly(A) RNA-seq, and PAIso-seq, a recent study indirectly
   investigates the fates of uridylated transcripts in Dis3l2 cKO
   (Zp3-Cre; Dis3l2^F/F) oocytes [193]^22. Upon the depletion of DIS3L2
   ribonuclease in mouse oocytes, uridylated-poly(A) RNAs accumulate
   significantly. Additionally, TUT4/7 uridylation may potentially
   elongate (from less than 5 to approximately 24 nucleotides) if not
   recognized and degraded by DIS3L2 ribonuclease. Furthermore, the
   uridylated transcripts in Dis3l2 cKO oocytes can be re-adenylated using
   the same polyadenylation signal sequences as in control oocytes.
   Although the re-adenylated transcripts can recruit poly(A)-binding
   proteins, the length of the poly(A) tails of the uridylated-poly(A)
   RNAs is shorter, which may disrupt the recruitment of poly(A)-binding
   proteins and compromise the initiation of translation. Overall,
   uridylated-poly(A) RNAs exhibit increased stability and reduced
   translation in Dis3l2 cKO oocytes. Our current study indicates that the
   transcriptome undergoes severe disruption upon DIS3L2 deletion. We
   speculate that uridylated-poly(A) RNAs accumulate and the global
   translation activity is compromised due to the features of the
   uridylated transcripts and the translation machinery proteins that have
   been insufficiently translated in Dis3l2 cKO germ cells. However,
   whether the fates of uridylated transcripts in Dis3l2-deficient
   spermatogonia and spermatocytes show similar patterns to those in
   Dis3l2 cKO oocytes needs to be further determined.

   In Dis3l2 cKO oocytes, several meiotic regulatory genes (e.g., Rad21l,
   Meioc, Nr2c2, Sycp3, and Tubgcp6) were dysregulated after DIS3L2
   deletion [194]^22. Some of these genes have known meiosis-related
   promoting or repressive roles [195]^39^-[196]^41. The disruption of
   meiosis genes and disrupted translation may cause impairment of meiotic
   resumption and germinal vesicle arrest in Dis3l2 cKO oocytes. Similar
   dysregulations have been observed in Dis3l2 cKO spermatocytes. The
   dysregulated transcripts were largely involved in RNA metabolism,
   translation regulation, and meiosis I in Dis3l2 cKO spermatocytes. The
   results, together with previous findings, highlight the crucial role of
   DIS3L2 ribonuclease in safeguarding the transcriptome to ensure meiotic
   progression of mouse spermatocytes and oocytes necessary for fertility.

   In summary, our investigations underscore that DIS3L2-mediated RNA
   degradation is indispensable for spermatogonial differentiation,
   spermatocyte meiotic progression, and male fertility. Upon DIS3L2
   deletion, the transcriptome in differentiated spermatogonia and meiotic
   spermatocytes undergoes severe disruption, with notable dysregulation
   in cell cycle genes in spermatogonia and meiosis genes in
   spermatocytes. The data pinpoint DIS3L2 as a pivotal regulator in
   spermatogonial differentiation, male meiosis, and spermatogenesis,
   shaping the transcriptome during these critical developmental stages.
   Nevertheless, the precise molecular mechanisms governing the fates of
   uridylated transcripts in Dis3l2-deficient spermatogonia and
   spermatocytes warrant further investigation in future studies.

Materials and methods

Ethics statement

   All animal procedures were conducted in accordance with the guidelines
   approved by the Ethics Committee for Animal Research of the School of
   Life Sciences, Shandong University, China (Approval No.
   SYDWLL-2021-90). The Dis3l2 Floxed mice were procured from Cyagen,
   while the Star8-Cre mouse line [197]^23 was generously provided by
   Professor Xiao-Yang Sun.

Mouse genotyping

   Mouse tail tissue samples were subjected to lysis with the Direct PCR
   Lysis Reagent containing proteinase K at 55 °C overnight. Subsequently,
   the samples were incubated at 85 °C for 1 hour to deactivate proteinase
   K. Specific DNA fragments were amplified using the EmeraldAmp GT PCR
   Master Mix (Takara Bio USA) along with appropriate primers. The PCR was
   conducted with an annealing temperature of 58 °C for 35 cycles using
   the Mastercycler Pro (Eppendorf). The primers employed for genotyping
   are detailed in Supplementary [198]Table S1.

Fertility assay

   To evaluate fertility, one wild-type female mouse was housed together
   with either a control or a Dis3l2 cKO male mouse for a period of 6
   months. At least 3 mating cages were arranged for each genotype, and
   the average number of pups per litter was documented.

Histology assay

   Mouse testes and epididymides samples were immersed in Bouin's solution
   and left overnight at 4 °C. Subsequently, the samples were embedded in
   paraffin, sectioned (5 μm), and affixed onto slides before staining
   with periodic acid-Schiff (PAS) and hematoxylin. Finally, the samples
   were examined and captured under a microscope (Nexcope NE950).

Immunohistochemistry and immunofluorescence

   Following de-waxing, rehydration, and antigen retrieval using 0.01%
   sodium citrate buffer (pH 6.0), sections were blocked with blocking
   buffer containing 0.05% Tween-20 at room temperature for 1 hour,
   followed by an overnight incubation with primary antibodies
   (Supplementary [199]Table S2) at 4 °C. For immunohistochemistry (IHC),
   the ImmPRESS™ Polymer Detection Kit (Vector Laboratories) and
   diaminobenzidine (DAB) were utilized to visualize antibody binding. For
   immunofluorescence (IF) staining, specific secondary antibodies
   (Supplementary [200]Table S2) were employed to detect the antigen, and
   DNA was counterstained with Hoechst 33342. Images were captured with a
   fluorescent microscope (Nexcope NE950).

   For whole-mount staining, freshly isolated tissues were fixed in 4%
   paraformaldehyde (PFA) overnight at 4 °C, permeabilized in blocking
   buffer containing 5% donkey serum, 3% bovine serum albumin, 0.5% Triton
   X-100, and 0.05% Tween 20 overnight at 4 °C and then incubated with
   primary antibodies (Supplementary [201]Table S2) overnight at 4 °C.
   Secondary antibodies (Supplementary [202]Table S2) conjugated to Alexa
   Fluor were incubated overnight at 4 °C. After washing in
   phosphate-buffered saline (PBS), samples were mounted with PBS on the
   slides. Fluorescence images were captured by confocal microscopy (Leica
   STELLARIS 5).

In situ hybridization

   The sequence of the mouse Dis3l2 probe for in situ hybridizations was
   retrieved from the GenePaint database. Templates were generated via PCR
   from cDNA and subsequently subcloned into the pEASY Blunt Zero Cloning
   vector. This construct was then linearized and utilized as a template
   for in vitro transcription using T7 polymerase to produce a probe
   labeled with digoxigenin-11-UTP (Roche). Testis samples freshly
   isolated were fixed in 4% PFA overnight at 4°C, followed by a wash in
   PBS. Subsequently, the samples underwent sequential incubation in 15%
   and 30% sucrose in PBS at 4°C for 16 hours and were then embedded in
   Tissue-Tek O.C.T. on dry ice. Sections, 10 μm thick, were cut and
   air-dried at 37°C for 30 minutes. After treatment with 2 μg/ml
   proteinase K, the slices were incubated at 65°C overnight with the
   digoxigenin-labeled probe in a hybridization solution containing 50%
   (v/v) formamide, 5 × saline sodium citrate, 5 × Denharts, 250 μg/ml
   yeast RNA, and 500 ug/ml herring sperm DNA. The slides were
   subsequently washed in 0.2 × saline sodium citrate at 62°C, followed by
   an incubation with anti-digoxigenin antibody (Roche) at 4°C overnight.
   Signals were developed in NBT/BCIP stock solution (Roche) and captured
   using a fluorescent microscope (Nexcope NE950).

Meiotic chromosome spreads

   Mouse testes were decapsulated, placed in hypotonic buffer (30 mM Tris
   pH 7.5, 17 mM trisodium citrate, 5 mM EDTA, and 50 mM sucrose) for 30
   minutes, and seminiferous tubules were chopped to release germ cells in
   200 mM sucrose. The resulting cell suspensions were deposited on slides
   pre-coated with fixation buffer (1% PFA and 0.1% Triton-X100 in PBS),
   fixed for 3 hours in a humidifying chamber at room temperature, and
   air-dried. For immunostaining, the slides were incubated with primary
   antibodies (Supplementary [203]Table S2) at 4°C overnight and
   subsequently with Alexa Fluor secondary antibodies (Supplementary
   [204]Table S2) for 1 hour at room temperature. Images were captured
   using a fluorescent microscope (Nexcope NE950).

Immunoblot assay

   Total protein was extracted in 1 × LDS sample buffer with 1× NuPAGE
   Sample Reducing Agent (Thermo Fisher Scientific). Proteins were
   separated on 4%-12% Bis-Tris gels and electrophoretically transferred
   onto polyvinylidene difluoride membranes. The membranes were blocked
   with 5% nonfat milk in Tris-buffered saline containing 0.05% Tween 20
   (TBST) at room temperature (RT) for 1 h and probed with primary
   antibodies (Supplementary [205]Table S2) overnight at 4 °C. The
   membranes were washed three times with TBST and incubated (1 h, RT)
   with secondary antibodies (Supplementary [206]Table S2), followed by
   washing with TBST and developed using SuperSignal West Dura Extended
   Duration Substrate (Thermo Fisher Scientific). Signals were detected
   with Hyperfilm ECL (GE Healthcare) according to the manufacturer's
   instructions.

RNA isolation and RT-qPCR

   Total RNA was extracted from mouse testes using the AFTSpin Tissue/Cell
   Fast RNA Extraction Kit for Animals (ABclonal), and cDNA was
   synthesized with the SuperScript IV First-Strand Synthesis System
   (Thermo Fisher Scientific). The RT-qPCR assay was conducted using the
   iTaq Universal SYBR Green Supermix (Bio-Rad) on the QuantStudio 6 Flex
   Real-Time PCR System (Thermo Fisher Scientific). The primers utilized
   in these experiments are detailed in Supplementary [207]Table S3. The
   relative abundance of each transcript was calculated using 2^-ΔΔCt and
   normalized to endogenous β-actin expression [208]^42.

RNA-seq library preparation

   Total RNA was extracted using the RNeasy Micro Kit, and libraries were
   prepared using a Universal RNA-seq Library Preparation Kit (TECAN)
   according to the manufacturer's instructions. Briefly, double-strand
   cDNA was synthesized using a mixture of random and poly(T) primers,
   followed by double-stranded cDNA fragmentation, end repair, adaptor
   ligation, strand selection, targeted transcript depletion with
   AnyDeplete, and PCR amplification. The final PCR-amplified libraries
   were sequenced on an Illumina HiSeq 2500 platform, generating 50-bp
   single-end reads.

RNA-seq analysis

   The processed reads were aligned to the University of California Santa
   Cruz (UCSC) mm10 reference genome through the use of HISAT2.
   Subsequently, the quantification results obtained from "featureCount"
   were subjected to analysis with the Bioconductor package DESeq2, which
   applies a negative binomial distribution to estimate both technical and
   biological variability. A PCA plot was generated to identify any
   potential sample outliers. Comparative analyses were performed between
   cKO and control samples. A gene was considered differentially expressed
   when the false discovery rate (FDR) adjusted p-value for differential
   expression was less than 0.05. The correlation heatmap was created
   using the R software package with the ''ggplot2'' package. GSEA was
   performed using GSEA v4.3 software program. Genes were ranked before
   analysis based on their fold change in gene expression. This
   application assigns scores to a sorted list of genes regarding their
   enrichment in selected functional categories (Kyoto Encyclopedia of
   Genes and Genomes and GO). The significance of the enrichment score was
   evaluated through 1000 permutations. Benjamini and Hochberg's FDR was
   computed to account for multiple testing adjustments. A q-value of <
   0.05 was considered statistically significant.

scRNA-seq library preparation

   scRNA-seq libraries were prepared using Chromium Next GEM Single Cell
   3′ Reagent Kits v3.1 (10× Genomics) according to the manufacturer's
   instructions. Briefly, single cells obtained from digested testes were
   mixed with a suspension containing barcoded beads and unique molecular
   identifier (UMI) elements that allow specific tagging of messenger RNA.
   After partitioning thousands of cells into nanoliter-scale Gel
   Bead-In-EMulsions (GEMs) and barcoding, full-length barcoded cDNA was
   then amplified by PCR to generate sufficient mass for library
   construction. Libraries were constructed by fragmentation, end repair,
   A-tailing, adaptor ligation, and index PCR. After ensuring adequate
   quality of cDNA libraries, the samples were sequenced on an Illumina
   NovaSeq 6000 platform at Guangzhou Gene Denovo Biotechnology Co., Ltd.

scRNA-seq data processing

   Raw sequencing reads were carried out using the Cell Ranger Single-Cell
   Software Suite (10× Genomics) for tasks such as alignment, filtering,
   UMI quantification, and generation of the gene-barcode matrix. The raw
   gene expression matrices produced per sample using CellRanger were
   imported into R (version 4.3.0) and converted to a Seurat object
   utilizing the Seurat R package (version 4.3.0.1) [209]^43. The
   filtration process eliminated deceased cells, doublets, and low-quality
   cells by considering total UMIs, total genes, and the percentage of
   mitochondrial UMIs. Subsequently, quality filters were set to retain
   cells that had 1) ≥ 200 and ≤ 8000 uniquely expressed genes; 2) ≥ 1000
   UMIs; 3) ≤ 20% of reads mapped to the mitochondrial genome; and 4) ≤ 5%
   of reads mapped to hemoglobin genes, as these cells likely depict
   contaminating erythrocytes.

Data integration and major cell type determination

   Cells from control and Dis3l2 cKO samples were integrated, and batch
   effects were corrected using Harmony from Seurat [210]^44.
   Normalization was applied to each sample, and cell cycle scoring was
   conducted using Seurat. Highly variable genes (HVGs) were identified
   using SCTransform, with mitochondrial genes being excluded.
   Subsequently, scaled PCA was performed on the filtered HVGs, and UMAP
   was used for dimensionality reduction. A shared nearest neighbor graph
   was constructed, followed by clustering at a resolution of 1.6 in
   Seurat. The canonical marker genes were applied to annotated cell
   clusters to identify biological cell types.

Identification of marker genes and differentially expressed genes

   Marker genes for cell types were identified by comparing expression
   levels in the cluster of interest against other clusters using the
   Seurat FindMarkers function (MAST, version 1.26.0) [211]^45. Marker
   genes were defined based on the following criteria: (1) the average
   expression value in the cluster of interest was at least 2.5-fold
   higher than that in the rest of the clusters; (2) marker genes should
   be detectable in at least 10% of the cells in the cluster of interest;
   and (3) marker genes should have the highest mean expression in the
   cluster of interest compared with the rest of the clusters.
   Differentially expressed genes between the two samples were computed
   using the Seurat FindMarkers function with parameters set as min.pct =
   0.01 and logfc.threshold = 0.01. GO analysis was conducted using
   Metascape [212]^46 to determine the biological processes and pathways.

Pseudotime analysis

   Pseudotime analysis was performed using monocle 3 (Version 1.0.0)
   [213]^47. Seurat object for cell types was converted to Monocle object
   and metadata and UMAP embedding were passed to Monocle object as well.
   Data were processed using the “preprocess_cds” function. Cells were
   clustered using the “cluster_cells” function with cluster_method =
   “leiden”. The trajectory graph was learned with the “learn_graph”
   function and cells were ordered in pseudotime with “order_cells”
   function.

Statistical analysis

   Data are presented as the mean ± s.d. Statistical analysis was
   conducted using GraphPad Prism. The variances of the two groups were
   compared using the two-tailed Student's t-test, and significance was
   defined as ns (no significance), *P < 0.05, **P < 0.01, ***P < 0.001,
   and ****P < 0.0001.

Supplementary Material

   Supplementary figures and tables.
   [214]thnov14p5621s1.pdf^ (3.6MB, pdf)

Acknowledgments