hnRNPU in Sertoli cells cooperates with WT1 and is essential for testicular development by modulating transcriptional factors Sox8/9

Abstract

Background: Sertoli cells are essential regulators of testicular fate in the differentiating gonad; however, its role and underlying molecular mechanism of regulating testicular development in prepubertal testes are poorly understood. Although several critical regulatory factors of Sertoli cell development and function have been identified, identifying extrinsic factors that regulate gonocyte proliferation and migration processes during neonatal testis development remains largely unknown. Methods: We used the Sertoli cell-specific conditional knockout strategy (Cre/Loxp) in mice and molecular biological analyses (Luciferase assay, ChIP-qPCR, RNA-Seq, etc.) in vitro and in vivo to study the physiological roles of hnRNPU in Sertoli cells on regulating testicular development in prepubertal testes. Results: We identified a co-transcription factor, hnRNPU, which is highly expressed in mouse and human Sertoli cells and required for neonatal Sertoli cell and pre-pubertal testicular development. Conditional knockout of hnRNPU in murine Sertoli cells leads to severe testicular atrophy and male sterility, characterized by rapid depletion of both Sertoli cells and germ cells and failure of spermatogonia proliferation and migration during pre-pubertal testicular development. At molecular levels, we found that hnRNPU interacts with two Sertoli cell markers WT1 and SOX9, and enhances the expression of two transcriptional factors, Sox8 and Sox9, in Sertoli cells by directly binding to their promoter regions. Further RNA-Seq and bioinformatics analyses revealed the transcriptome-wide of key genes essential for Sertoli cell and germ cell fate control, such as biological adhesion, proliferation and migration, were deregulated in Sertoli cell-specific hnRNPU mutant testes. Conclusion: Our findings demonstrate an essential role of hnRNPU in Sertoli cells for prepubertal testicular development and testis microenvironment maintenance and define a new insight for our understanding of male infertility therapy.

Keywords: Mice; Sertoli cells; Testicular development; Transcription factor; hnRNPU.

Figure 1

hnRNPU localizes in mouse and human Sertoli cells and interacts with WT1 in mouse testes. (A) Double immunofluorescent staining with hnRNPU (green) and WT1 (red) on adult wild-type (WT) mouse testis sections are shown. The nuclei were stained with DAPI (blue). Spz, spermatozoa; Spc, spermatocyte; SC, Sertoli cell; RS, round spermatid. Scale bars = 100 µm. (B) Co-immunoprecipitation assays analyze the interaction between hnRNPU and WT1 in mouse testes of postnatal day 3 (P3). Lysates of testes were immunoprecipitated with anti-hnRNPU and anti-WT1 antibodies, respectively. (C) Co-immunostainings of hnRNPU (red) with WT1 (green) was performed on various ages of WT mouse testis sections at embryonic day 17.5 (E17.5), P0, P3, P7, P21, and P60 are shown. The nuclei were stained with DAPI (blue). Scale bars = 50 µm. (D) Double immunofluorescent staining of hnRNPU (red) with WT1 (green) on isolated WT mouse Sertoli cells are shown. The nuclei were stained with DAPI (blue). Scale bars = 50 µm. (E) Double immunofluorescent staining with hnRNPU (red) and WT1 (green) on fetal (left) and adult (right) human testis sections are shown. The nuclei were stained with DAPI (blue). Scale bars = 50 µm.

Figure 2

Conditional inactivation of hnRNPU in Sertoli cells results in severe obstruction of testicular development and male infertility in mice. (A) Schematic diagram showing the targeting strategy of generating a floxed Hnrnpu allele through homologous recombination in the murine embryonic stem cells. Red triangles represent loxP cassettes. Exons 4 to 14 will be deleted after Amh-Cre mediated recombination. (B) Representative PCR genotyping results showing the floxed (lox) and the WT (+) alleles can be detected at 412 bp and 316 bp bands, respectively. NC, non-template control. (C) RT-qPCR analyses showing Hnrnpu mRNA level was nearly undetectable in isolated Amh-Cre-cKO Sertoli cells. Data are presented as mean ± SEM, n = 3. (D) Western blot analyses of hnRNPU protein level in Sertoli cells (SCs) isolated from control and Amh-Cre-cKO testes. β-ACTIN served as a loading control. (E) Representative co-immunofluorescent images of hnRNPU (red) and WT1 (green) in control and Amh-Cre-cKO testis sections at P3 are shown. Scale bars = 50 µm. (F) The histogram shows the average number of pups per litter produced from control and Amh-Cre-cKO male mice. Data are presented as mean ± SEM, n = 5. (G) Gross morphology of the testis, the epididymis, and the seminal vesicle from adult control and Amh-Cre-cKO male mice. (H) Testis growth curve shows the Amh-Cre-cKO mouse testes were significantly decreased from P5. Data are presented as mean ± SEM, n = 3. *P < 0.05, **P < 0.01, and ***P < 0.001 by Student's t-test. (I) Periodic Acid-Schiff (PAS) staining shows the histology of testis and epididymis sections from adult control and Amh-Cre-cKO male mice. Left panels indicate testis histology of control and Amh-Cre-cKO mice. Many degenerated tubules containing aberrant Sertoli cell-like or germ cell-like cells were seen in the seminiferous tubules from the Amh-Cre-cKO mice. Right panels show cauda epididymis was completely lacking spermatozoa in Amh-Cre-cKO mice. Scale bars = 100 µm. (J) Immunofluorescence staining with Laminin (Red) in control and Amh-Cre-cKO testes at P7 and P42 are shown, respectively. The nuclei were stained with DAPI (blue). Scale bars = 100 µm.

Figure 3

Loss of hnRNPU in Sertoli cells impairs germ cell and Sertoli cells development and proliferation. (A) Representative PAS-stained paraffin sections of developing testes at P0, P3, P5, P7, and P35 from control and Amh-Cre-cKO mice are shown. Scale bars = 50 µm. (B) Co-immunofluorescent staining of WT1 (Sertoli cell marker, red) with DDX4 (germ cell marker, green) in testicular sections at P0, P3, P5, and P7 from control and Amh-Cre-cKO mice are shown. Nuclei were stained with DAPI. Scale bars = 50 µm. (C) The histograms showing the quantifications of DDX4⁺ cells per tubule in (B). Data are presented as mean ± SEM, n=3. **P < 0.01 and ***P < 0.001 by Student's t-test. (D) Co-immunofluorescent staining with WT1 (red) and the Ki67 (green) on testis sections from control and Amh-Cre-cKO mice at P3 are shown. Nuclei were stained with DAPI. Scale bars = 50 µm. (E) The histograms showing the quantification of the ratio for Ki67⁺ and WT1⁺ cells to WT1⁺ cells in (D). Data are presented as mean ± SEM, n = 5. ***P < 0.001 by Student's t-test. (F) Transmission electron microscopy images of control (left panel) and Amh-Cre-cKO (right panel) testis ultra-sections at P3. Lower panels are the zoom-in from the rectangle region of upper panels. Red arrows indicate lipid accumulation. Scale bars = 10 µm in upper panels and 2 µm in lower panels. (G) Co-immunofluorescent staining with WT1 (green) and the P27^KIP1 (red) on control and Amh-Cre-cKO mouse testes at P21 is shown. Nuclei were stained with DAPI. Scale bars = 50 µm.

Figure 4

Ablation of hnRNPU in Sertoli cells causes aberrant spermatogonia development and migration. (A) Co-immunofluorescent staining of PLZF (an undifferentiated spermatogonia marker, red) with DDX4 (green) on testis sections from control and Amh-Cre-cKO mice at P3, P5, and P7 are shown. Nuclei were stained with DAPI. Scale bars = 50 µm. White circles indicate each seminiferous tubules. (B) The histograms showing the quantifications of PLZF⁺ cells per tubule in (A). Data are presented as mean ± SEM, n = 3. *P < 0.05, **P < 0.01, ***P < 0.001 by Student's t-test. (C) Co-immunofluorescent staining of TRA98 (red) with Ki67 (green) in control and Amh-Cre-cKO testis sections at P3 are shown. Nuclei were stained with DAPI. Scale bars = 50 µm. White circles indicate each seminiferous tubules. (D) The histograms showing the quantifications of the ratio for TRA98⁺ and Ki67⁺ cells to TRA98⁺ cells at P3 in (C). Data are presented as mean ± SEM, n = 3. *P < 0.05 by Student's t-test. (E) Immunofluorescent staining of STRA8 (a differentiated spermatogonia marker, red) on testis sections from control and Amh-Cre-cKO mice at P3, P5, and P7 are shown. Nuclei were stained with DAPI. Scale bars = 50 µm. White circles indicate each seminiferous tubules. (F) The histograms showing the quantifications of STRA8⁺ cells per tubule in (E). Data are presented as mean ± SEM, n = 3. *P < 0.05, ***P < 0.001 by Student's t-test. (G) Co-immunofluorescent staining of TRA98 (a germ cell marker, red) with WT1 (a Sertoli cell marker, green) in control and Amh-Cre-cKO testis sections at P3, P5, and P7 are shown. Nuclei were stained with DAPI. Scale bars = 20 µm. (H) The quantification of the percentage of the germ cells located at the basal region of the seminiferous tubules in (G). Data are presented as means ± SEM. n = 3. ***P < 0.001 by Student's t-test. (I) Co-immunostaining of N-cadherin (red) with WT1 (green) on testis sections from control and Amh-Cre-cKO mice at P3 are shown. Nuclei were stained with DAPI. Scale bars = 50 µm. White circles indicate each seminiferous tubules. (J) Immunostaining of Vimentin (red) on control and Amh-Cre-cKO mouse testis sections at P0 and P3 are shown. Nuclei were stained with DAPI. Scale bar = 100 µm. (K) Histograms showing RT-qPCR analyses of expression levels of Fshr, Gdnf, and Gfrα1 mRNAs in control and Amh-Cre-cKO testes at P3. Data are presented as mean ± SEM, n = 3. ***P < 0.001 by Student's t-test.

Figure 5

hnRNPU directly binds to Sox8/9 promoter regions and regulates its expression levels. (A) Western blot analyses of hnRNPU, SOX8, and SOX9 in Sertoli cells purified from control and Amh-Cre-cKO mouse testes are shown. GAPDH served as a loading control. (B) The histograms showing the quantifications of SOX8 and SOX9 protein levels in (A). Data are presented as mean ± SEM, n = 3. ***P < 0.001 by Student's t-test. (C-D) Luciferase-based reporter assays showing the luciferase activity of the Sox8 (C) and Sox9 (D) promoter region in Sertoli cells significantly increase when hnRNPU overexpression. (E) ChIP-qPCR shows the hnRNPU enrichments at the different promoter regions of Sox8 and Sox9 genes. IgG was used as a negative control for Sox8/9 binding. The upper schematic image illustrates the position of the forward and reverse primers in two different promoter regions used for the ChIP-qPCR assays. Quantitative data are expressed as the ratio of the ChIP (Bound) to the input DNA. Gapdh promoter was used as a negative control for hnRNPU enrichment. *P < 0.05, **P < 0.01, ***P < 0.001 by Student's t-test. (F) Immunoprecipitation of hnRNPU in isolated Sertoli cells at P3 followed by western blot detection of Pol II, hnRNPU, SOX9, WT1, and SOX8 are shown.

Figure 6

hnRNPU deficiency in Sertoli cells induces the genome-wide transcriptome profile alterations in testes at P3. (A) Volcano plot showing deregulated genes in Amh-Cre-cKO testes at P3. Significantly regulated genes have an FDR ≤ 0.01 and fold change ≥ 2. (B-C) Gene ontology term analyses of the 268 upregulated genes (B) and 221 downregulated genes (C) in Amh-Cre-cKO testes at P3 are shown. The 12 enriched GO pathways in the upregulated genes and 10 enriched GO pathways in the downregulated genes are illustrated by gene counts and P-values. (D) Heat-map shows the differences in transcriptional activity of transcription factors between control and Amh-Cre-cKO testes analyzed by CoRegNet analysis. (E-F) RT-qPCR validates the selected upregulated (E) and downregulated (F) genes in Amh-Cre-cKO testes from RNA-seq data. Data are presented as mean ± SEM, n = 3.