Nestin-dependent mitochondria-ER contacts define stem Leydig cell differentiation to attenuate male reproductive ageing

Abstract

Male reproductive system ageing is closely associated with deficiency in testosterone production due to loss of functional Leydig cells, which are differentiated from stem Leydig cells (SLCs). However, the relationship between SLC differentiation and ageing remains unknown. In addition, active lipid metabolism during SLC differentiation in the reproductive system requires transportation and processing of substrates among multiple organelles, e.g., mitochondria and endoplasmic reticulum (ER), highlighting the importance of interorganelle contact. Here, we show that SLC differentiation potential declines with disordered intracellular homeostasis during SLC senescence. Mechanistically, loss of the intermediate filament Nestin results in lower differentiation capacity by separating mitochondria-ER contacts (MERCs) during SLC senescence. Furthermore, pharmacological intervention by melatonin restores Nestin-dependent MERCs, reverses SLC differentiation capacity and alleviates male reproductive system ageing. These findings not only explain SLC senescence from a cytoskeleton-dependent MERCs regulation mechanism, but also suggest a promising therapy targeting SLC differentiation for age-related reproductive system diseases.

Fig. 1. SLCs exhibit lower capacity for differentiation into LCs during ageing.

a Representative H&E staining pictures of testicular mesenchyme from different age groups (2 months, 12 months, and 24 months). Scale bar, 500 μm for original pictures and 50 μm for enlarged pictures. b Serum testosterone level of different age groups. (n = 6 biological repeats for each group; One-way ANOVA). c–e qPCR analysis of relative mRNA expression of LCs markers of testes from different age groups. (n = 6 biological repeats for each group; One-way ANOVA). f Representative immunostaining pictures of testicular mesenchyme from different age groups. LCs are identified as HSD3β + /CYP11A1 + /HSD17β3 + cells. Scale bar, 40 μm for original pictures and 20 μm for enlarged pictures. g–i Quantification of LCs cell numbers with different markers in (f). (n = 6 biological repeats for each group; One-way ANOVA). j Schematic of the experimental animal model for labeling and tracing Nestin+ cells. k Flow cytometry for detecting the percentage of RFP + LHR + cells in total RFP + cells of different age Nestin-Cre^ERT2; Rosa26^RFP mice. l Quantification of percentage of RFP + LHR + cells in total RFP + cells of different age Nestin-Cre^ERT2; Rosa26^RFP mice in (k) (n = 5 biological repeats for each group; One-way ANOVA). Two-sided comparison; All data are mean ± SD; Error bars represent SDs. *p < 0.05, **p < 0.01, ***p < 0.001; Source data are provided as a Source Data file. See also Supplementary Fig. 1.

Fig. 2. SLCs establish fewer mitochondria-endoplasmic reticulum contacts during ageing.

a Representative immunostaining pictures of colocalization between mitochondria and ER in primary Nestin-GFP + SLCs from different age groups. Mitochondria and ER are marked with Mitotracker and ER tracker, respectively. Scale bar, 10 μm. b, c Quantification of the levels of colocalization in (a) (Manders’ (of mitochondria) and Pearson’s coefficients are shown for each condition, n = 49, 55 and 51 cells for 2 m, 12 m, and 24 m SLCs, respectively; One-way ANOVA). d Representative immunostaining pictures and 3D reconstruction of mitochondria and ER in primary Nestin-GFP + SLCs from different age groups. Scale bar, 5 μm. e Representative transmission electron microscope (TEM) images of primary Nestin-GFP + SLCs from different age groups. Scale bar, 250 nm for original pictures and 100 nm for enlarged pictures. f Quantification of the extent of MERCs (<100 nm) from (e) in primary Nestin-GFP + SLCs from different age groups. The relative amount of ER-OMM apposition in percentage of OMM refers to the interaction length of ER tubules within 100 nm of the OMM covered in the percentage of total OMM per mitochondria (n = 26 to 48 mitochondria in 5 fields per condition; One-way ANOVA). g Quantification of the mito-ER proximity from e in primary Nestin-GFP + SLCs from different age groups (n = 26 to 48 mitochondria in 5 fields per condition; One-way ANOVA). h Representative immunostaining pictures, 3D reconstruction and surface-surface contact site area coverage of VDAC (mitochondria) or IP3R1(ER) in primary Nestin-GFP + SLCs from different age groups. Scale bar, 10 μm. i Quantification of the surface-surface contact site area (percentage of mitochondria) in (h) (n = 5 biological repeats for each group; One-way ANOVA). j Quantification of surface-surface contact site area (percentage of ER) in (h) (n = 5 biological repeats for each group; One-way ANOVA). Two-sided comparison; All data are mean ± SD; Error bars represent SDs. **p < 0.01, ***p < 0.001; Source data are provided as a Source Data file. See also Supplementary Fig. 2.

Fig. 3. Fewer MERCs result in mitochondrial dysfunction and ER stress in aged SLCs.

a Flow cytometry of intracellular ROS level stained with DHE of Nestin-GFP + SLCs from different age groups in vivo. b Quantification of mean fluorescence intensity of DHE in (a). (n = 5 biological repeats for each group; One-way ANOVA). c qPCR analysis of relative mRNA expression of anti-oxidative genes of Nestin-GFP + SLCs from different age groups. (n = 5 biological repeats for each group; Multiple t tests). d, e Western Blot analysis and quantification of cellular anti-oxidative protein expression of Nestin-GFP + SLCs from different age groups. (n = 5 biological repeats for each group; Multiple t tests). f Representative immunostaining pictures of mitochondrial morphology of primary Nestin-GFP + SLCs from different age groups in vitro. Mitochondria are marked with Tom20. Scale bar, 20 μm. g A plot of Aspect Ratio (AR) against Form Factor (FF) showed that particles in primary Nestin-GFP + SLCs from different age groups in (f). Lower values of FF and AR, indicating small, circular mitochondria. h Representative TEM images of mitochondrial morphology of primary Nestin-GFP + SLCs from different age groups in vitro. Scale bar, 1 μm for original pictures and 500 nm for enlarged pictures. i Quantitative analysis of the average value of mitochondrial length (micrometers) in (h). (n = 5 biological repeats for each group; n = 126–184 mitochondria; One-way ANOVA). j qPCR analysis of relative mRNA expression of ER stress-related genes of primary Nestin-GFP + SLCs from different age groups in vitro. (n = 3 biological repeats for each group; Multiple t tests). k–o Western Blot analysis and quantification of ER stress-related proteins of primary Nestin-GFP + SLCs from different age groups in vitro. (n = 3 biological repeats for each group; Multiple t tests). Two-sided comparison; All data are mean ± SD; Error bars represent SDs. *p < 0.05, **p < 0.01, ***p < 0.001; Uncropped western blots and source data are provided as a Source Data file. See also Supplementary Figs. 3–5.

Fig. 4. Keap1-mediated Nestin degradation separates MERCs in aged SLCs.

a Representative immunostaining pictures of α-tubulin, F-actin and Nestin in primary SLCs from 2 months and 24 months old mice. Scale bar, 40 μm. b Quantitative analysis of the mean fluorescence intensity of three kinds of the cytoskeleton in (a). (n = 3 biological repeats for each group; Multiple t tests). Arb. units, absorbance units. c Representative detailed immunostaining pictures of α-tubulin, F-actin and Nestin in primary SLCs. Scale bar, 10 μm. d Quantitative analysis of the area covered by α-tubulin in primary SLCs. (n = 3 biological repeats for each group; Unpaired t test). e Quantitative analysis of the number of F-actin fibers in primary SLCs. (n = 3 biological repeats for each group; Unpaired t test). f Standard deviation of the organization of Nestin fibers in primary SLCs were calculated. (n = 3 biological repeats for each group; Unpaired t test). g, h Western Blot analysis and quantification of α-tubulin, F-actin and Nestin in primary SLCs. (n = 3 biological repeats for each group; Multiple t tests). i, j Western Blot analysis and quantification of Nestin in primary SLCs with or without treatment with MG132. (n = 3 biological repeats for each group; Unpaired t test). k Co-IP of Myc-Nestin and ubiquitin in primary SLCs after treatment with MG132. Ubiquitination level are quantified by measuring the gray-scale value of the whole lane of Western Blot band. Relative expression of ubiquitinated Nestin (normalized to GADPH) were normalized to 2 months old group. GADPH was used as loading control. (n = 3 biological repeats for each group; Unpaired t test). l Co-IP of Myc-Nestin and ubiquitin in primary SLCs from 24 months old Keap1-knockdown mice after treatment with MG132. Quantification of ubiquitination level, normalization of ubiquitinated Nestin expression and loading control are indicated in (k). (n = 3 biological repeats for each group; Unpaired t test). m, n Immunoprecipitation analysis and quantification of Keap1–Nestin binding and Keap1-Nrf2 binding degree in primary SLCs. (n = 3 biological repeats for each group; Multiple t tests) Two-sided comparison; All data are mean ± SD; Error bars represent SDs. *p < 0.05, **p < 0.01, ***p < 0.001, ns, p > 0.05; Uncropped western blots and source data are provided as a Source Data file. See also Supplementary Figs. 6 and 7.

Fig. 5. Targeting Nestin degradation restores intracellular homeostasis and differentiation capacity in aged SLCs.

a Western Blot analysis and quantification of Nestin expression in primary SLCs from 2 months old, 24 months old, and melatonin-treated 24 months old mice. (n = 3 biological repeats for each group; Unpaired t test). b Co-IP of Myc-Nestin and ubiquitin in primary SLCs after treatment with MG132. Ubiquitination level are quantified by measuring the gray-scale value of the whole lane of the Western Blot band. Relative expression of ubiquitinated Nestin (normalized to GADPH) were normalized to NTC group and are indicated at the bottom. GADPH was used as loading control. (n = 3 biological repeats for each group; Unpaired t test). c Representative immunostaining pictures of Nestin, mitochondria, and ER in primary SLCs. Mitochondria and ER are marked with Mitotracker and ER tracker, respectively. Scale bar, 5 μm. d Representative immunostaining pictures of intracellular ROS stained with CellRox in primary SLCs. Scale bar, 20 μm. e Quantitative analysis of the mean fluorescence intensity of CellRox in (e). (n = 3 biological repeats for each group; Unpaired t test). f Western Blot analysis of ER stress-related protein expression in primary SLCs. g Flow cytometry analysis of LHR expression in LCs induced from primary SLCs at day 9. h Schematic illustration of the experimental procedure for isolating seminiferous tubules from testes of mice and inducing differentiation into LCs with melatonin treatment. Created with BioRender.com. i Representative immunostaining pictures of seminiferous tubules after being induced to differentiation at day 14. LCs are identified as HSD3β + cells. Scale bar, 100 μm for original pictures and 50 μm for enlarged pictures. j Quantitative analysis of the percentage of HSD3β + cells of total cells in seminiferous tubules after induced differentiation at day 14. (n = 3 biological repeats for each group; Unpaired t test). k Measurement of testosterone concentration in the medium of seminiferous tubules after induced differentiation at day 14. (n = 3 biological repeats for each group; Unpaired t test). Two-sided comparison; All data are mean ± SD; Error bars represent SDs. *p < 0.05, **p < 0.01, ***p < 0.001; Uncropped western blots and source data are provided as a Source Data file. See also Supplementary Figs. 8–13.

Fig. 6. Treatment with Melatonin increases testosterone level and promotes spermatogenesis in old mice.

a, b Western Blot analysis and quantification of Nestin expression in the testes from different age group. (n = 3 biological repeats for each group; Unpaired t test). c, d Analysis of TEM from (e). Percentage of mitochondria coverage covered by ER (n = 115 cross-sections from 5 mice for each group) and mito-ER proximity were quantified (n = 110 cross-sections from 6 mice for each group); Unpaired t test. e Representative TEM images of testicular mesenchyme. (mitochondria, yellow; ER, red; contact sites, blue). Scale bar, 500 nm for original pictures and 250 nm for enlarged pictures. f Measurement of serum testosterone concentration. (n = 6 biological repeats for each group; Unpaired t test). g qPCR analysis of relative mRNA expression of testosterone production-related genes in the testes (n = 6 biological repeats for each group; Unpaired t test). h Representative immunostaining pictures of SYCP3-positive cells and Sox9-positive Sertoli cells in seminiferous tubules of testes. Scale bar, 40 μm. i Quantification of the number of SYCP3-positive cells and Sox9-positive Sertoli cells per seminiferous tubule. (n = 6 biological repeats for each group; Multiple t tests). j Quantification of the ratio of SYCP3-positive cells vs Sox9-positive Sertoli cells per seminiferous tubule. (n = 6 biological repeats for each group; Unpaired t test). k Representative H&E staining pictures of cauda epididymis. Scale bar, 500 μm for original pictures and 50 μm for enlarged pictures. l Schematic of the experimental animal model for conditionally knocking out Nestin in PDGFRα + cells. Nestin KO: Nestin knockout. m Representative immunostaining pictures of Nestin in testicular mesenchyme of Control and Nestin KO mice. Scale bar, 40 μm for original pictures and 20 μm for enlarged pictures. n Representative immunostaining pictures of Nestin in primary SLCs from Control and Nestin KO mice. Scale bar, 20 μm. o Representative immunostaining pictures of testicular mesenchyme from Control and Nestin KO mice, treated by ethanol or melatonin. LCs are identified as HSD3β + cells. Scale bar, 40 μm for original pictures and 40 μm for enlarged pictures. p Quantification of LCs cell numbers in (L). (n = 6 biological repeats for each group; Multiple t tests). q Measurement of serum testosterone level of Control and Nestin KO mice, treated by ethanol or melatonin. (n = 6 biological repeats for each group; Multiple t tests). Two-sided comparison; All data are mean ± SD; Error bars represent SDs. *p < 0.05, ***p < 0.001, ns, p > 0.05; Uncropped western blots and source data are provided as a Source Data file. See also Supplementary Fig. 14.

Fig. 7. AAV-mediated downregulation of Nestin separates MERCs and diminishes Melatonin’s effect in attenuating male reproductive ageing.

a Schematic of the experimental animal model for conditionally knocking out Nestin in testis. b Representative TEM images of testicular mesenchyme. (mitochondria, yellow; ER, red; contact sites, blue). Scale bar, 500 nm for original pictures and 250 nm for enlarged pictures. c Analysis of TEM from (b). Percentage of mitochondria coverage covered by ER (n = 115 cross-sections from 4 mice for each group) and mito-ER proximity were quantified (n = 130 cross-sections from 5 mice for each group; Unpaired t test). d Flow cytometry of the percentage of tdTomato+LHR + cells in vivo. e Quantification of percentage of tdTomato+LHR + cells in total tdTomato+ cells in (d) (n = 3 biological repeats for each group; Unpaired t test). f Serum testosterone level of wild type, AAV8-sgCon, AAV8-sgNes, and AAV8-sgNes+Mel groups. (n = 6 biological repeats for each group; Unpaired t test). g Representative immunostaining pictures of SYCP3-positive cells in seminiferous tubules of testis. Scale bar, 100 μm. h Quantification of SYCP3-positive cells per tubule cross-section. (n = 6 biological repeats for each group; Unpaired t test). i Flow cytometry of intracellular ROS level stained with DHE of tdTomato+ cells in vivo. j Quantification of mean fluorescence intensity of ER stress markers in Supplementary Fig. 15j. (n = 3 biological repeats for each group; Multiple t tests). k Schematic of the chemical-damaged animal model treated with EDS and conditionally knocking out Nestin in testis. l Representative immunostaining pictures of testicular mesenchyme from EDS( + );EDS( + )/Mel(+);EDS( + )/Mel(+)/AAV8-sgCon(+);EDS( + )/Mel(+)/AAV8-sgNes(+) groups 4 weeks after EDS treatment. LCs are identified as HSD3β + /CYP11A1 + /HSD17β3 + cells. Scale bar, 20 μm. m Quantification of LCs cell numbers in (l) with different markers. (n = 3 biological repeats for each group; Multiple t tests). n Measurement of serum testosterone level. (n = 5 biological repeats for each group; Multiple t tests). Two-sided comparison; All data are mean ± SD; Error bars represent SDs. *p < 0.05, **p < 0.01, ***p < 0.001, ns, p > 0.05; Source data are provided as a Source Data file. See also Supplementary Fig. 15.

Fig. 8. Schematic illustration of Nestin-mediated MERCs loss during reproductive ageing and melatonin’s anti-senescence effect.

Loss of intermediate filament Nestin compromises the differentiation capacity of SLCs via separating MERCs during senescence. Pharmacological intervention by melatonin restores Nestin-dependent MERCs, improving SLC differentiation capacity and alleviating male reproductive system ageing.