NEUROG3 is a critical downstream effector for STAT3-regulated differentiation of mammalian stem and progenitor spermatogonia

Abstract

Spermatogenesis relies on coordinated differentiation of stem and progenitor spermatogonia, and the transcription factor STAT3 is essential for this process in mammals. Here we studied the THY1+ spermatogonial population in mouse testes, which contains spermatogonial stem cells (SSC) and non-stem cell progenitor spermatogonia, to further define the downstream mechanism regulating differentiation. Transcript abundance for the bHLH transcription factor Neurog3 was found to be significantly reduced upon transient inhibition of STAT3 signaling in these cells and exposure to GDNF, a key growth factor regulating self-renewal of SSCs, suppressed activation of STAT3 and in accordance Neurog3 gene expression. Moreover, STAT3 was found to bind the distal Neurog3 promoter/enhancer region in THY1+ spermatogonia and regulate transcription. Transient inhibition of Neurog3 expression in cultures of proliferating THY1+ spermatogonia increased stem cell content after several self-renewal cycles without effecting overall proliferation of the cells, indicating impaired differentiation of SSCs to produce progenitor spermatogonia. Furthermore, cultured THY1+ spermatogonia with induced deficiency of Neurog3 were found to be incapable of differentiation in vivo following transplantation into testes of recipient mice. Collectively, these results establish a mechanism by which activation of STAT3 regulates the expression of NEUROG3 to subsequently drive differentiation of SSC and progenitor spermatogonia in the mammalian germline.

FIG. 1.

Expression of Stat3 and Neurog3 is linked in SSC and progenitor spermatogonia. A) Representative image of germ cell clumps (arrows) that form in cultures of THY1+ spermatogonia maintained under serum-free conditions with the supplementation of the growth factors GDNF and FGF2. These clumps consist of SSC and non-stem cell progenitor spermatogonia and can be maintained as a proliferative population for long-periods of time providing an in vitro model to study mechanisms regulating self-renewal and differentiation. Bar  =  50 μm. B) Representative image of a recipient mouse testis transplanted with a single cell suspension of cultured THY1+ germ cell clumps from a LacZ expressing Rosa donor mouse. Each blue segment represents a reestablished colony of spermatogenesis derived from an injected donor SSC, thereby allowing for quantitative assessment of stem cell number in the cultured germ cell populations. Also, qualitative assessment of in vivo biological function for the cultured SSC and progenitor spermatogonia can be made based on the extent of reestablished spermatogenesis. Bar  =  2 mm. C) Representative images of immunocytochemical staining for NEUROG3 (arrows) and p-STAT3^Tyr705 (arrows and arrowheads) expression in single-cell suspensions generated from cultured THY1+ germ cell clumps. DAPI was used to stain the nuclei of all cells, and an overlay of the images revealed that ∼30% of the cells express NEUROG3, whereas almost all (>95%) of the cells contain p-STAT3^Tyr705. Also, it was observed that all NEUROG3-expressing cells coexpress p-STAT3^Tyr705 (arrows), whereas some cells express p-STAT3^Tyr705 only (arrowheads). Bar  =  100 μm. D) qPCR analyses for expression of Neurog3 and Bcl6b in cultured THY1+ germ cells treated with a STAT3-specific inhibitor peptide or control peptide. Data are presented as fold-differences for STAT3 inhibitor-treated cells compared to control cells and are means ± SEM for three replicate experiments with different cultures. *Significantly different at P ≤ 0.05. After 24 h of treatment Neurog3 expression was significantly reduced in STAT3 inhibitor treated cells compared to control cells and reduced even further after 72 h of treatment. However, expression of Bcl6b, a gene involved in the self-renewal of SSC, was not different between STAT3 inhibitor treated and control cells at either 24 or 72 h of treatment.

FIG. 2.

Signaling from the self-renewal growth factor GDNF suppresses STAT3 activation and Neurog3 gene expression in SSC and progenitor spermatogonia. A) Representative images of Western blot analyses for expression of phosphorylated STAT3 (p-STAT3^Tyr705) and total STAT3 protein in cultures of THY1+ spermatogonia continually supplemented with GDNF (+GDNF), 18 h after withdrawal of GDNF supplementation (-GDNF 18 h), and 4 h replacement of GDNF after 18 h withdrawal (−GDNF 18 h/+GDNF 4 h). B) Quantitative analyses of Western blot images for p-STAT3^Tyr705 expression. Data are presented as the fold-difference compared to cells continually supplemented with GDNF (+GDNF) for the ratio of p-STAT3^Tyr705-to-total STAT3 and are means ± SEM for three independent experiments with different cultures. *Significantly different at P ≤ 0.05. After 18 h withdrawal of GDNF, the level of p-STAT3^Tyr705 increased significantly and was subsequently suppressed after 4 h of GDNF replacement. C) qPCR analyses for Neurog3 transcript abundance in cells continually exposed to GDNF (+GDNF), subjected to 18 h of withdrawal of GDNF (−GDNF 18 h), and after 4 h of GDNF replacement following 18 h of withdrawal (−GDNF 18 h/+GDNF 4 h). Data are presented as fold-difference compared to cells continually supplemented with GDNF and are mean ± SEM for three independent experiments with different cultures. *Significantly different at P ≤ 0.05. After 18 h withdrawal of GDNF supplementation, expression of Neurog3 was increased significantly and subsequently suppressed following 4 h of GDNF replacement.

FIG. 3.

STAT3 physically interacts with the distal Neurog3 promoter/enhancer region in SSC and progenitor spermatogonia and is capable of stimulating Neurog3 promoter activity. A) Nucleotide sequence of the distal Neurog3 promoter/enhancer region in the mouse genome (NCBI RefSeq accession no. NC_000076.5), which contains a canonical STAT3 binding site (bolded and underlined sequence). B) Representative image of PCR analyses for the Neurog3 promoter/enhancer sequence with DNA template that was immunoprecipitated from THY1+ germ cells using a STAT3-specific antibody. Primer sequences for the Neurog3 promoter/enhancer are identified by bold in A. PCR analyses with primers for the Myc promoter which is a known target of STAT3 and input DNA prior to immunoprecipitation with primers for the Neurog3 promoter/enhancer were used as positive controls. Use of normal IgG instead of STAT3-specific antibody for immunoprecipitation served as a negative control. STAT3 was found to bind both the Neurog3 promoter/enhancer and Myc promoter sequences within THY1+ spermatogonia. C) Quantitative ChIP analyses for STAT3 occupancy of the Neurog3 promoter/enhancer region in cultured THY1+ spermatogonia following treatments that alter Neurog3 transcript abundance. Cells were continually cultured with GDNF (+GDNF), subjected to withdrawal of GDNF for 18 h (−GDNF 18 h), or subjected to replacement of GDNF for 4 h after 18 h withdrawal (−GDNF 18 h/+GDNF 4 h). Also, cells were treated with control inhibiting peptide or STAT3-specific inhibiting peptide for 24 h. Data are presented as mean ± SEM of three independent experiments with different primary cultures. *Significantly different at P ≤ 0.05. D) Luciferase activity in HeLa cells stimulated with INFA to activate STAT3 following transfection with an empty luciferase reporter vector (Empty Vector) or reporter vector in which the Neurog3 promoter sequence was cloned upstream of the firefly luciferase gene and treated with a control or STAT3-specific inhibiting peptide. Data are presented as RLU, which was derived from normalization to a Renilla luciferase internal control vector and are means ± SEM for four independent experiments. *Significantly different at P ≤ 0.05. NT, cells receiving no treatment; a representation of background luminescent signal. Addition of the Neurog3 promoter construct into IFNA stimulated HeLa cells treated with a control peptide resulted in a significant increase of luciferase activity, whereas inclusion of a STAT3-specific inhibitor peptide suppressed the promoter activity.

FIG. 4.

Inhibition of Neurog3 enhances the self-renewal of SSCs in vitro. A) Experimental strategy for examining the impact of Neurog3 deficiency on self-renewal of SSCs in vitro. Cultured THY1+ germ cells were treated with Neurog3-specific or control siRNA for 24 h and after another 6 days an aliquot of the cell population was transplanted into recipient testes to measure stem cell content. The remaining cells were again treated with Neurog3-specific or control siRNA for 24 h and cultured another 6 days followed by transplantation to measure stem cell content. The proliferation rate of SSCs within cultures of THY1+ germ cells is ∼6 days. Thus, these analyses allowed for determining stem cell content after >2 self-renewal cycles. B) qPCR analyses for Neurog3 gene expression in cultured THY1+ germ cells 24 h and 7 days after treatment with Neurog3-specific or control siRNA. Data are presented as fold-difference compared to control siRNA treatment and are mean ± SEM for three independent experiments with different cultures. *Significantly different at P ≤ 0.05. Neurog3 transcript abundance was significantly reduced 24 h after treatment and was not different after 7 days, indicating transient reduction by siRNA treatment. Furthermore, there was no difference in expression at the time of transplantation on Days 7 and 14 of culture. C) SSC content in cultures of THY1+ germ cells treated with Neurog3-specific or control siRNA at 0, 7, and 14 days of culture. SSC concentrations were determined based on the number of colonies of rederived spermatogenesis in recipient testes per 1 × 10⁵ cells transplanted. Data are presented as fold-difference compared to control siRNA treated cells and are mean ± SEM for three independent experiments with different cultures. *Significantly different at P ≤ 0.05. At Day 14, which is greater than two SSC self-renewal cycles in vitro, SSC concentration was significantly increased in cultures treated with Neurog3-specific siRNA. D) Total number of cells in cultures of THY1+ germ cells treated with Neurog3-specific or control siRNA at 0, 7, and 14 days of culture. Data are presented as fold-difference compared to control siRNA treatment and are mean ± SEM for the same three independent experiments with the same cultures used for transplantation analyses. *Significantly different at P ≤ 0.05. No difference in total germ cell content was found between Neurog3-specific and control siRNA treated cultures at any of the time points examined.

FIG. 5.

SSC and progenitor spermatogonia deficient for Neurog3 expression are incapable of differentiation in vivo. A) qPCR analyses for expression of Neurog3 in cultured THY1+ germ cells after treatment with non-targeting control or Neurog3-specific shRNA lentiviral vectors. Treatment with Neurog3-specific shRNA lentivirus resulted in a significant reduction of Neurog3 transcript abundance by ∼86% compared to control. Data are presented as fold-difference from control and are mean ± SEM for three independent experiments with different cultures. *Significantly different at P ≤ 0.05. B) Representative image of a recipient mouse seminiferous tubule 2 mo after transplantation with cultured THY1+ Rosa germ cells that were transduced with a non-targeting control shRNA lentiviral vector and subjected to puromycin treatment to select cells with stable integration of the transgene. Dense blue staining of the colony indicates complete spermatogenesis derived from a single transplanted SSC. Bar  =  2 mm. C and D) Representative images of recipient mouse seminiferous tubules 2 mo after transplantation with cultured THY1+ Rosa germ cells that were transduced with a Neurog3-specific shRNA lentiviral vector and subjected to puromycin treatment to select cells with stable integration of the transgene. Only single (arrows in D) and short chains (arrows in C) of cells were observed, indicating a block in differentiation of both SSC and progenitor spermatogonia. Bars  =  1 mm.