Peritubular myoid cells participate in male mouse spermatogonial stem cell maintenance

Abstract

Peritubular myoid (PM) cells surround the seminiferous tubule and together with Sertoli cells form the cellular boundary of the spermatogonial stem cell (SSC) niche. However, it remains unclear what role PM cells have in determining the microenvironment in the niche required for maintenance of the ability of SSCs to undergo self-renewal and differentiation into spermatogonia. Mice with a targeted disruption of the androgen receptor gene (Ar) in PM cells experienced a progressive loss of spermatogonia, suggesting that PM cells require testosterone (T) action to produce factors influencing SSC maintenance in the niche. Other studies showed that glial cell line-derived neurotrophic factor (GDNF) is required for SSC self-renewal and differentiation of SSCs in vitro and in vivo. This led us to hypothesize that T-regulated GDNF expression by PM cells contributes to the maintenance of SSCs. This hypothesis was tested using an adult mouse PM cell primary culture system and germ cell transplantation. We found that T induced GDNF expression at the mRNA and protein levels in PM cells. Furthermore, when thymus cell antigen 1-positive spermatogonia isolated from neonatal mice were cocultured with PM cells with or without T and transplanted to the testes of germ cell-depleted mice, the number and length of transplant-derived colonies was increased considerably by in vitro T treatment. These results support the novel hypothesis that T-dependent regulation of GDNF expression in PM cells has a significant influence on the microenvironment of the niche and SSC maintenance.

Figure 1.

GDNF detection by immunofluorescence and confocal microscopy in seminiferous tubules and in PM cells cultured for 7 days. A, GDNF (red) and ACTA2 (green) are localized in PM cells in a 1-μm-thick optical section at the surface of a whole-mount seminiferous tubule. An individual PM cell is enclosed in the square in the upper images and enlarged in the lower images. B, Cross-sections of seminiferous tubules immunostained for GDNF (red) and ACTA2 (green) and viewed by confocal microscopy. GDNF is detected in PM and Sertoli cells, and ACTA2 is detected in PM cells. A PM cell (indicated by the arrow) is enclosed by a square in the upper images and enlarged in the lower images. Nuclei are stained with 4′,6-diamidino-2-phenylindole (blue). Scale bar, 50 μm.

Figure 2.

GDNF (red) and ACTA2 (green) localization in cultured PM cells with or without T treatment. A–D, GDNF was present in a reticular pattern and in vesicles in T-treated PM cells (arrows). E–H, GDNF was not detected in PM cells cultured in medium without T. Cell nuclei are stained with 4′,6-diamidino-2-phenylindole (DAPI). Scale bar, 20 μm.

Figure 3.

PCR assays for Leydig cells (Cyp17a1) and Sertoli cells (Sox9) in PM cell cultures. A, Cyp17a1, Sox9, and Gapdh levels were determined by PCR in PM cells, mixed peritubular, and Sertoli cell cultures (P+S), testis (T), and brain (B). Testis and brain were used as positive controls. B and C, The Cyp17a1 (B) and Sox9 (C) transcript levels were detected by PCR determined by densitometry and normalized to the Gapdh transcript levels. *, P < .001.

Figure 4.

Effects of dose and time on Gdnf mRNA and GDNF protein expression by PM cells in vitro. A, Selected concentrations of T (black bars) were added to individual culture wells for 6 hours, and the Gdnf mRNA levels were determined by qPCR. Control cultures did not receive T (open bars). The results shown are fold changes in Gdnf levels relative to Gapdh levels. B, Cultures were treated with T for up to 120 hours, and the Gdnf mRNA levels were determined by qPCR at different time points. Control cultures did not receive T. The results shown are fold changes in Gdnf levels compared with Gapdh levels. C, PM cells were treated with T (solid line) for 24 to 120 hours, and the concentration of GDNF in the culture medium was determined by ELISA at the times indicated. Control cultures did not receive T (dashed line). Each data point represents mean ± SEM. *, P < .05

Figure 5.

THY1-positive spermatogonia (10⁴) cocultured with 10⁶ PM cells treated with T or untreated (control). A, Cultured cells were analyzed by flow cytometry, and based on the cell size and density using gates reported by previous investigators (54), most the SSCs were located in the gray area. Cells from the gated area were further analyzed by flow cytometry after labeling with antibodies to GFRA1 or KIT. B, Most spermatogonia cocultured with T-treated PM cells were GFRA1⁺, whereas spermatogonia cocultured with PM cells not treated with T were GFRA1⁻. C, Few spermatogonia cocultured with PM treated or untreated with T were KIT⁺. D, Representative quadrant gated cells treated or untreated with T.

Figure 6.

SSC colonization of germ cell-depleted testes. A, Testes from busulfan-treated mice 11 weeks after receiving spermatogonia from ROSA26 mice cocultured with T-treated or untreated PM cells. The blue staining identifies β-Gal–expressing colonies of spermatogenic cells derived from transplanted SSCs. The testis on the left received spermatogonia cocultured with PM cells treated with T, whereas the testis on the right received spermatogonia cocultured with PM cells untreated with T. B and C, The recipient testes were evaluated for number of colonies (B) and the length of those colonies (C) derived from transplanted spermatogonia cocultured with T-treated or untreated PM cells. Scale bar, 2 mm. *. P < .001.

Figure 7.

Cultures containing a mixture of PM and Sertoli cells were untreated (control, open bars), treated with T (black bars), or treated with 25 ng/mL FSH (hatched bars), and the Gdnf mRNA levels were determined by qPCR. No significant differences in Gdnf mRNA levels were observed between controls, T-treated, and FSH-treated cultures.