Mechanical signal-mediated mitochondria–endoplasmic reticulum contacts modulate Leydig cell testosterone biosynthesis during testis development

Abstract

In males, 95% of testosterone is synthesized by Leydig cells, and a deficiency in this synthesis will cause metabolic disorders and multiple organ dysfunction. Testosterone deficiency is not only affected by aged or diseased Leydig cells, which have been studied extensively, but is also closely related to the development of the testis. At present, the focus on the mechanism of testis development includes epigenetic and hormone regulation. However, testicular development is constrained by the external tough tunica albuginea, suggesting that mechanical signals may also play an important role in the regulation of testis development; however, this is not yet well understood. In this in vitro study, we found that a gradual increase in extracellular substrate stiffness for testis development leads to the activation of mechanical signals to promote cytoskeleton remodeling. Eventually, the mechanical signal mediates changes in the mitochondrial–endoplasmic reticulum and affects the synthesis of testosterone in Leydig cells. Through organoid and animal experiments, we found that targeting mechanical signaling pathways that regulate testosterone biosynthesis is feasible. This provides a new angle for further exploration of testis development and new insights into how substrate stiffness affects the testis, raising new clues for clinical applications.

Introduction

Extracellular matrix (ECM), as a highly dynamic structural network, provides a microenvironment for tissue formation and repair (Liu et al., 2022). The main components of ECM include collagen, elastin, fibronectin, laminin, and proteoglycans, all of which are fibrillar proteins (Theocharis et al., 2016). These proteins exist in varying proportions and combinations in different tissues and organs, which can be divided into two types: interstitial substrate and basement membrane (Bonnans et al., 2014), endowing ECM with unique physical and chemical properties. Substrate stiffness is an important mechanical characteristic of ECM, determined by the content and density of cross-linked proteins, including collagen, fibronectin, and laminin (Deng et al., 2022). These proteins form the skeletal structure of ECM, thereby determining the response of ECM to mechanical deformation. Cells can perceive substrate stiffness and convert it into biological signals, affecting their survival (Theocharis et al., 2016), proliferation (Hadjipanayi et al., 2009), differentiation (Schwartz et al., 1995; Pang et al., 2023), and migration (Pelham and Wang, 1997; Guo et al., 2006). This process involves multiple cellular signaling pathways, including integrin–cytoskeleton-mediated signal pathways (Humphrey et al., 2014) and the activity of transcription factors (Saraswathibhatla et al., 2023).

In recent years, an increasing number of studies have revealed the key role of substrate stiffness in physiological and pathological states. Under physiological conditions, substrate stiffness is involved in the growth, development, structural remodeling, and functional maintenance of organs such as the heart (Pesce et al., 2023), brain (Javier-Torrent et al., 2021), and bones (Arvind and Huang, 2017). In pathological states, substrate stiffness is associated with the development of various diseases, including atherosclerosis (Wang et al., 2022), pulmonary fibrosis (Guo et al., 2022), urethral stricture (Li et al., 2023), and malignant tumors (Yuan et al., 2023), where it plays a promoting role. The testis, an organ enclosed by a dense layer of fibrous tissue, includes seminiferous tubules and surrounding interstitial tissue. During its growth and development, the mechanical microenvironment in the testis undergoes changes. However, there are no reports on the role of substrate stiffness in the male reproductive system, especially in the growth and development of the testis and the regulation of Leydig cells (LCs) function.

LCs are mainly distributed in the loose connective tissue between seminiferous tubules and are responsible for synthesizing and secreting testosterone. LCs account for about 2–4% of the total testicular cell number, but they synthesize and secrete more than 95% of the body’s total testosterone. From a developmental perspective, LCs can be divided into fetal and adult types. Fetal LCs originate from the embryonic period and their numbers gradually decrease after birth. The testosterone they secrete is essential for the development of the male reproductive tract and testicular descent. Adult LCs appear during puberty and are derived from the differentiation and development of testicular stem cells. Once formed, they rarely renew or die, and the testosterone they secrete is responsible for maintaining spermatogenesis and secondary sexual characteristics (Ye et al., 2011).

The synthesis of testosterone is regulated by the hypothalamic–pituitary–testis axis. Luteinizing hormone combines with specific receptors on the surface of LCs (Li et al., 2019), activating a series of downstream reactions. The interaction between mitochondria and the endoplasmic reticulum plays a crucial role at the organelle level in testosterone synthesis (Hall et al., 1969; Payne and Youngblood, 1995; Zirkin and Papadopoulos, 2018). In recent years, maintaining mitochondrial structural integrity (Lu et al., 2022) has been found to limit mitochondrial oxidative damage, improve mitochondrial dynamics (Wang et al., 2024), and reduce endoplasmic reticulum stress (Wu et al., 2024), in order to promote testosterone synthesis.

Mitochondria are the energy centers of the cell, responsible for producing the majority of ATP and a variety of biosynthetic intermediates during metabolism. Additionally, mitochondria are involved in the stress response of cells to various internal and external stressors, causing the induction of apoptosis and autophagy (Nunnari and Suomalainen, 2012). The ER is involved in protein synthesis and modification in cells and is the primary storage site for Ca²⁺ within the cell. The ER also contains a large number of biosynthetic enzymes involved in cellular lipid synthesis. With the development of high-resolution microscopy technology, researchers have observed the close connection between mitochondria and the ER. These two organelles are interconnected through proteins located on adjacent membranes, forming contact sites 10–30 nm wide, known as mitochondria–endoplasmic reticulum contacts (MERCs) (Rowland and Voeltz, 2012).

In recent years, research on the molecular composition and function of MERCs in physiological and pathological processes has increased. For example, some proteins located at these contact sites contain synaptotagmin-like mitochondrial lipid-binding protein (SMP) domains, and structural analysis and mass spectrometry have demonstrated that these SMP domains are involved in lipid transport (Schauder et al., 2014); inositol-1,4,5-adenosine triphosphate receptor (IP3R) on the ER interacts with voltage-dependent anion-selective channel protein 1 (VDAC1) on the outer membrane of mitochondria, allowing Ca²⁺ transfer (Rizzuto et al., 1998); and dynamin-related protein Dnm1/Drp1 forms punctate structures under homeostasis, appearing at mitochondria–ER contact sites and is associated with subsequent mitochondrial fission (Friedman et al., 2011). These processes are closely intertwined, jointly maintaining the homeostasis of the intracellular environment. MERCs dysfunction is closely related to systemic diseases such as obesity and type 2 diabetes(Beaulant et al., 2022), as well as central nervous system diseases such as Parkinson’s disease (Grossmann et al., 2019). In the male reproductive system, MERCs also play an important role. For example, cytoskeleton-dependent MERCs detachment during male reproductive system aging leads to reduced differentiation capacity of testicular mesenchymal stem cells (Yao et al., 2022); in vitro exposure to polystyrene microplastics leads to MERCs dysfunction in LCs, subsequently causing testosterone synthesis disorders (Grillo et al., 2024). These studies provide us with a new perspective on the involvement of MERCs in the regulation of testosterone synthesis in physiological processes.

In this study, we investigate the role of the mechanical microenvironment in testis development. We are the first to report an increase in substrate stiffness during testis development and identify mechanically regulated MERCs as a critical factor in modulating the testosterone synthesis capacity of LCs. Our findings demonstrate that changes in mechanical environment induce cytoskeletal remodeling, particularly through Actin polymerization, and increase MERCs, which enhances LCs metabolic activity and promotes the expression of key enzymes such as STAR and CYP11A1, ultimately elevating testosterone levels. These results highlight the importance of mechanical signaling in testis development and provide new insights into the interplay between the mechanical environment, organelle interactions, and hormonal regulation. This study opens new avenues for exploring the mechanical regulation of testis development and offers a deeper understanding of the processes governing this complex biological system.

Materials and methods

Ethics approval

All procedures were approved by the ethics committee of the Shenzhen TOPBIOTECH Biotechnology Co., Ltd (Shenzhen, China), Ethical approval No. TOPGM-IACUC-2024-0021.

Animals

C57bl/6 mice at 1 week (at birth), 4 weeks (at puberty), 5 weeks, and 8 weeks (at adulthood) of age were obtained from Zhuhai BESTEST Biotechnology Co., Ltd (Zhuhai, China). All mice were maintained under controlled temperature (24 ± 1°C) and relative humidity (50–60%) with an alternating 12-h light/12-h dark cycle and were given free access to a standard rodent diet and drinking water.

Tissue collection and sectioning

C57bl/6 mice at 1, 4, 5, and 8 weeks of age were euthanized by cervical dislocation, and their testicular tissues were harvested for tissue mechanical property determination (atomic force microscopy, AFM). The fresh testicular tissues for AFM were embedded in OCT and prepared as frozen sections; testicular tissues for hematoxylin and eosin (HE) staining were fixed, dehydrated, cleared, and embedded in paraffin, then prepared as paraffin sections.

Preparation of substrate gels of different stiffness

To construct two-dimensional substrate gels with varying stiffness, we mixed 40% w/v acrylamide, 2% w/v polyacrylamide, 1M HEPES (GIBCO, Grand Island, NY, USA), and ddH₂O according to the ratio reported in other articles (Tian et al., 2019a,b; Tse and Engler, 2010). 10 kPa: 40% w/v acrylamide, 2% w/v diacrylamide, 1M HEPES, ddH₂O in a ratio of 50:10:2:138; 100 kPa: 40% w/v acrylamide, 2% w/v diacrylamide, 1M HEPES, ddH₂O in a ratio of 98.4:80.2:2:19.4. The specific dosage used can be found in Supplementary Table S1.

Then we added 10% ammonium persulfate and TEMED to promote coagulation. The mixture was injected into 100 µm thick WB plates to stand for 20 min. The gels were then soaked in PBS solution, cut into pieces, and laid flat in the wells of a plate. After washing with 50 mM HEPES and PBS, Sulfo-SANPH (Macklin, Shanghai, China) violet crosslinking agent was then added to the gels and exposed to ultraviolet light for 1 h, followed by PBS washing. Mouse tail type I collagen was then added and incubated overnight. After PBS washing, the gels were ready for cell inoculation. The reagents and kits used can be found in Supplementary Table S2.

Cell culture

TM3 LCs were cultured in DMEM/F12 medium containing 2.5% fetal bovine serum, 5% horse serum and 1% penicillin/streptomycin (GIBCO, Grand Island, NY, USA) in a cell incubator at 37°C and 5% CO₂. We changed the DMEM/F12 medium every 2 days. When the cell density reached 80% ∼ 90%, the cells were digested with 0.25% w/v trypsin for cell passage. TM3 was randomly divided into 10 and 100 kPa groups and inoculated on hydrogels in six-well plates at a density of 3∼4 × 10⁵ per well. Follow-up experiments were carried out when the cell density reached 80% ∼ 90%. To activate or inhibit ROCK kinase in vitro, 10 kPa group medium was pretreated with CN01 (MCE, Monmouth Junction, NJ, USA) or 100 kPa group medium was pretreated with Y-27632 (MCE) for 1 day, depending on the final fluid volume of cultured cells.

Testing of tissue for mechanical properties

We used an AFM (BRUKER, Billerica, MA, USA) to detect the spatial variation of mechanical properties of testicular tissue samples at different developmental stages. The fresh testicular tissues for AFM were embedded in OCT and prepared as frozen sections. Tissue sections were mounted on a cantilever holder and platform, immersed in PBS at 25°C. The interaction between the microcantilever tip and the sample caused a change in the position of the reflected light, leading to a deflection. The spring constant (kc) of each cantilever was measured using the thermal fluctuation method (the final value of kc ranged from 0.112 to 0.148 N/m, with an uncertainty of ∼0.002 N/m). Force spectroscopy analysis was performed at each point in a 64 × 16 grid over an area of 5 μm × 1.25 μm (∼80 nm pixel size). A Young’s modulus map was generated by the built-in software Nanoscope Analysis 2.0 (BRUKER), using the Oliver–Pharr contact model.

ELISA detection of mouse supernatant testosterone concentration

A testosterone ELISA kit (Solarbio, Beijing, China) was used. Cell culture supernatant was collected and centrifuged at 1000 g for 10 min at 4°C. Standard samples were diluted in a gradient, added to corresponding wells, and mixed with reconstituted enzyme conjugates. After incubation at room temperature in the dark for 120 min, the liquid in the wells was decanted, and the wells were washed with washing solution. The substrate was added for color development, and after incubation at room temperature in the dark for 20 min, the stop solution was added. We measured the absorbance using a microplate reader (Thermo Fisher, Waltham, MA, USA), and the corresponding concentration was calculated from a standard curve.

Immunohistochemistry and immunofluorescent staining

For immunohistochemistry staining, mouse testis tissues were fixed with 4% neutral-buffered paraformaldehyde. Paraffin-embedded tissues were sectioned by 5 μm on a rotary microtome (Leica, Wetzlar, Germany), which was subjected to immunostaining. After dewaxing, the samples were alternately placed in heated and cold Tris–EDTA antigen repair solution and repeated three times. Then, the samples were incubated overnight at 4°C with corresponding antibodies and DAPI (Sigma-Aldrich, St Louis, MO, USA).

Cells cultured in confocal dishes (density 60–80%) were washed with PBS. Then we used Mito-Tracker-Red (CST, Boston, MA, USA) at a dilution of 1:500 in serum-free DMEM (GIBCO) for staining for 40 min. After washing with PBS, ER-Tracker (CST) was used at a dilution of 1:400 in a special diluent for staining for 40 min. After washing with PBS, DAPI was used for nuclear staining, and the confocal dishes were sealed and imaged using a laser confocal microscope. Colocalization analysis was performed using Image J software (NIH, Bethesda, MD, USA) (Zhang et al., 2023).

For immunofluorescence, cells cultured in confocal dishes (density 60–80%) were washed with PBS, fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, and blocked with 5% BSA. The appropriate primary antibodies were added at the recommended dilution, and the confocal dishes were incubated at 4°C overnight. After washing with PBS, the secondary antibodies were added and incubated at room temperature in the dark for 1 h. The primary and secondary antibodies used can be found in Supplementary Table S3. After washing with PBS, DAPI was used for nuclear staining, and the confocal dishes were sealed and imaged using a laser confocal microscope. Image J software was used for analysis.

Western blots

TM3 cells cultured on substrate gels were lysed in RIPA lysis buffer (RIPA 50 μl, PMSF 100 × 1μl, protease/phosphatase inhibitors 50 × 2μl). The cells were scraped, collected, sonicated, and centrifuged, and the supernatant was taken. We determined protein concentrations by using a BCA protein assay kit (Beyotime, Shanghai, China). Then, proteins were denatured. Samples were loaded onto a WB gel, electrophoresed, transferred to a membrane, blocked, and incubated with diluted primary antibodies at 4°C overnight. The membrane was washed with TBST and incubated with secondary antibodies at room temperature for 1 h. The primary and secondary antibodies used can be found in Supplementary Table S3. After washing, the membrane was placed in a gel imager, developed with developing solution and exposed. Then the protein concentrations were analyzed using Image J. Raw data for all WB images are available in Supplementary Figs S6–S21.

RNA isolation and real-time quantitative PCR

We extracted RNA from adherent cells by using Trizol solution (Thermo Fisher). After chloroform, isopropanol, and ethanol layering, centrifugation was performed to obtain RNA precipitation. HiFiScript gDNA Removal RT MasterMix reagent kit (Thermo Fisher) was used to reverse transcribe 1 μg mRNA into cDNA. PCR reaction mixtures (cDNA, primers, SYBR Green) were prepared and centrifuged quickly before amplification on a PCR machine. The reaction conditions were: 95°C pre-denaturation for 30 s, 95°C denaturation for 5 min, 60°C annealing, extension for 37 s, for 40 cycles. We read fluorescent quantitative CT values by using software, and performed relative quantification analysis, with gene transcription levels normalized to the expression level of Gapdh transcripts. The primers designed and used for qPCR are described in Supplementary Table S4.

ATP content assay

An ATP content detection kit (Beyotime, Shanghai, China) was used. Cell pellets were collected by centrifugation, and samples were treated with extraction solution under ice-bath conditions (cell number (10⁴): extraction solution volume (ml)=10³:1), sonicated, and centrifuged at 4°C at 10 000 g for 10 min. We removed the supernatant and added chloroform. After centrifugation at 4°C at 10 000 g for 5 min, the supernatant was taken for testing. We added the test sample or standard dilution, reagent one, and detection working solution in a 1 ml quartz cuvette. Then, we mixed it well and measured the absorbance at 340 nm after 10 s, recorded as A1 test/A1 standard. After 190 s, we measured the absorbance at 340 nm, recorded as A2 test/A2 standard, and we calculated ΔA=2−A1. A standard curve was established, and ΔA was substituted to obtain x (μmol/ml). ATP content (μmol/10⁴ cells)=x/cell number (10⁴).

Reactive oxygen species generation and flow cytometric assay

Reactive oxygen species (ROS) detection was performed using CellROX™ Green reagent (Thermo Fisher): TM3 cells in 6-well plates (density 50–70%) were digested with 0.25% trypsin, terminated with twice the volume of serum-containing DMEM, transferred to EP tubes, washed with PBS, and resuspended with CellROX dye. After being washed with PBS, the cells were transferred to flow cytometry tubes. Then we collected the data and analyzed it through a flow cytometer. Data were analyzed by Flow Jo Software.

HE staining of testicular tissue

We prepared paraffin-embedded tissue sections as mentioned above. Then, we deparaffinized the sections in xylene I and II for 3 min each, followed by treatment in absolute ethanol for 3 min, 95% ethanol for 3 min, 80% ethanol for 1 min, and then in distilled water for 1 min. We stained the sections with Harris hematoxylin solution for 10 min, rinsed in distilled water for 3 s, in 1% hydrochloric acid ethanol solution for 3 s, and then rinsed again in distilled water for 10 s, treating them with a bluing reagent for 6 s. We rinsed the sections under running water for 1 min and then stained them with 0.5% eosin solution for 3 min, followed by a quick rinse in distilled water for 2 s, 80% ethanol for 2 s, and 95% ethanol for 2 s. We dehydrated the sections through a graded ethanol series: absolute ethanol I, II, and III for 5, 10, and 30 s, respectively, and then in xylene I and II for 3 min each. Finally, we mounted the sections with a sealing agent and observed them under a microscope.

Testicular organoids with different substrate stiffness

The testes from mature male mice were minced, and the primary cells of the testis (comprising both somatic and germ cells) were extracted through an enzymatic digestion process. These primary testicular cells were then suspended in matrigel with different concentrations of type I collagen (low stiffness substrate: 0.25 mg/ml Col I; high stiffness substrate: 1.25 mg/ml Col I) and were subsequently cultivated within a three-layer gradient system (3-LGS), following the methodology outlined previously (Alves-Lopes et al., 2018). The testicular organoids were cultured in a medium composed of MEM-α (GIBCO), supplemented with 10% KSR (KnockOut serum replacement, GIBCO) and 1% penicillin/streptomycin (GIBCO), for 21 days at 35°C, with the medium being refreshed every 48 h.

Data collection and statistical analysis

Three biological replicates were used for each group. All data are mean±SD. Intergroup comparisons are made using unpaired t-tests, one-way ANOVA, and two-way ANOVA for statistical analysis. Graphs were generated using GraphPad Prism 10 (GraphPad Software, Boston, MA, USA), and significance levels were set at *P < 0.05, **P < 0.01, ***P < 0.001, and ns, P > 0.05.

Results

Substrate stiffness affects testosterone biosynthesis in LCs

In order to explore the development of testis and the changes in the mechanical microenvironment, we used paraffin section technology combined with H&E staining to observe the changes of internal structure of the testis. It was found that the proportion of interstitial cells to the area of the stroma in the testis of 1-, 4-, and 8-week-old mice increased gradually, indicating that the testis developed gradually (Fig. 1A and B). AFM was used to detect tissue mechanical properties, and it was found that the Young’s modulus of mouse testicular tissue increased gradually at week 1, week 4, and week 8, indicating that the substrate stiffness in the testis gradually increased from birth to maturity (Fig. 1C). We performed immunofluorescence labeling of key testosterone synthesis enzymes STAR and CYP11A1, and the results showed that the content of key testosterone synthesis enzymes in testicular interstitial cells was significantly increased during testis development (Fig. 1D, Supplementary Fig. S1A). In addition, we detected serum testosterone concentrations in mice at 1-, 4-, and 8-weeks old, and found that serum testosterone concentrations gradually increased (Fig. 1E). As mentioned above, testosterone is mainly secreted by LCs, so we explored in vitro whether changes in substrate stiffness affects the testosterone synthesis function of LCs. We used PA hydrogels to construct the substrate. Acrylamide, polyacrylamide, HEPES, and ddH₂O were mixed according to previously reported ratios to obtain hydrogels that exhibit varying stiffnesses (Tian et al., 2019a; Tse and Engler, 2010). LCs were cultured on the substrates exhibiting either 10, 40, 70, 100, or 130 kPa stiffness and western blots were performed to compare the content of STAR and CYP11A1, the key enzymes of testosterone synthesis in LCs. It was found that the expression of STAR and CYP11A1 in LCs at 10–100 kPa gradually increased. However, there was no significant difference between 130 and 100 kPa, so we chose to use 10 and 100 kPa as experimental groups in vitro (Supplementary Fig. S1B and C). We found that through timed quantitative PCR and protein immunoblotting, Cilengitide, a selective α_vβ₃ and α_vβ₅ receptor integrin inhibitor, can inhibit the expression of STAR and CYP11A1, the key enzymes of testosterone synthesis, and the same conclusions were obtained in vivo (Fig. 1F and G, Supplementary Fig. S1D). The supernatant of cell culture dishes was extracted for the testosterone enzyme-linked immunoassay, and it was found that testosterone synthesis showed the same trend (Fig. 1H). These results indicate that during development, along with the increase of substrate stiffness in testis, the levels of key enzymes for testosterone synthesis in LCs increased, leading to the increase in testosterone synthesis in LCs.

Figure 1.

Substrate stiffness impacts testosterone synthesis in Leydig cells (LCs). (A) H&E staining and atomic force microscopy (AFM) detection of substrate stiffness in the testes of C57bl/6 mice at 1 week, 4 weeks, and 8 weeks of age. Scale bar (upper), 50 μm, Scale bar (lower), 1 μm. (B) Statistical analysis of the proportion of interstitial cells to the area of the stroma. n = 5. ***P < 0.001, one-way ANOVA. (C) Statistical analysis of Young’s modulus characterized using AFM detection. n = 5. ***P < 0.001, one-way ANOVA. (D) Immunofluorescence images of key enzymes STAR and CYP11A1 for testosterone synthesis in C57bl/6 mice at 1 week, 4 weeks, and 8 weeks of age. n = 5. (E) Statistical analysis of serum testosterone concentration in C57bl/6 mice at 1 week, 4 weeks, and 8 weeks of age. n = 5. ***P < 0.001, one-way ANOVA. (F) Western blots of STAR and CYP11A1 in LCs cultured on substrates exhibiting different substrate stiffness and treatment with Cilengitide. (G) Relative statistical analysis of (F). ***P < 0.001, two-way ANOVA. (H) Statistical analysis of supernatant testosterone concentration in LCs cultured on substrates exhibiting different substrate stiffness and treatment with Cilengitide. Groups and colors are same as (G). ***P < 0.001, one-way ANOVA. All data are mean±SD. Error bars represent SDs.

Substrate stiffness influences cytoskeleton-dependent mitochondria-endoplasmic reticulum contacts

MERCs maintain intracellular environmental homeostasis by regulating cellular functions such as lipid synthesis, calcium homeostasis, and mitochondrial dynamics. Considering that lipid metabolism is closely related to mitochondria and endoplasmic reticulum organelles, we hypothesized that substrate stiffness affects testosterone synthesis by regulating MERCs. Therefore, we investigated whether the increase in testosterone synthesis capacity of LCs cultured on high substrate stiffness was due to the close proximity of mitochondria to the ER. Colocalization analysis showed that contacts between mitochondria (labeled MitoTracker Green FM) and endoplasmic reticulum (labeled ER tracker RED FM) increased as substrate stiffness increased, and the green and red fluorescence were visually overlapping. After quantitative analysis, it was found that Pearson’s coefficient and Manders’ coefficient were both close to 1, indicating a high degree of co-localization between the two fluoresces (Fig. 2A–C). Previous literature defined MERCs as a contact site with a width of 10–30 nm. According to this, we imaged the organelles structure by transmission electron microscopy and found that the distance between mitochondria and endoplasmic reticulum was shortened in cells cultured on a substrate exhibiting high substrate stiffness (more concentrated in the range of 10–30 nm), and the close contacts between mitochondrial outer membrane and endoplasmic reticulum was increased. Signs of membrane fusion were intensified (Fig. 2D and E). We also detected the levels of ITPR1 and MFN2, two mitochondria-associated endoplasmic reticulum connexins, which play an important role (Hayashi and Su, 2007; Hu et al., 2021) in mitochondrial dysfunction and endoplasmic reticulum stress. We found that both proteins were significantly upregulated in the LCs cultured on high-stiffness substrates relative to low-stiffness substrates, suggesting that they may be a downstream effect of increased MERCs (Fig. 2F–H). Together, these results suggest that elevated substrate stiffness may influence LCs by promoting mitochondrial and endoplasmic reticulum contacts.

Figure 2.

Substrate stiffness affects cytoskeleton-dependent mitochondria–endoplasmic reticulum contacts. (A) Immunofluorescence images of mitochondria–endoplasmic reticulum colocalization in different substrate stiffness. Scale bar (upper), 4 μm, Scale bar (lower), 10 μm. (B, C) Statistical analysis of (A). **P < 0.01, Unpaired t-test. (D) Representative electron microscopy images of distance between mitochondria and endoplasmic reticulum in different substrate stiffness. Scale bar, 500 nm for original pictures and 200 nm for enlarged pictures. (E) Statistical analysis of (D). *P < 0.05, Unpaired t-test. (F) Quantitative PCR analysis of mRNA expression of ITPR1 and MFN2 in different substrate stiffness. ***P < 0.001, two-way ANOVA. (G) Western blots of endoplasmic reticulum and mitochondria resident proteins ITPR1 and MFN2 in different substrate stiffness. (H) Relative statistical analysis of (G). ***P < 0.001, two-way ANOVA. All data are mean±SD. Error bars represent SDs.

Increased mitochondria–ER contacts maintain mitochondrial and ER function

Material exchange and interaction between organelles mainly occur at the direct contact sites between organelles. MERCs are crucial for mitochondria and ER to perform various functions. We therefore sought to understand how substrate stiffness further affects mitochondrial and ER functions. We first studied the changes in mitochondrial function. According to Mito-tracker staining fluorescence, mitochondrial fluorescence intensity was enhanced in the 100 kPa group, indicating that increased substrate stiffness promoted mitochondrial genesis (P < 0.001) (Fig. 3A and B). The main function of mitochondria is to provide energy for cells by producing ATP through aerobic respiration, and the main source of ATP production in cells is mitochondria (Prinz et al., 2020; Harrington et al., 2023). Therefore, we measured the ATP content of cells and found that the ATP content in the high substrate stiffness group was significantly increased (P < 0.001) (Fig. 3C). Real-time fluorescent quantitative PCR was used to determine the MtDNA copy number of cells, and it was found that the MtDNA copy number of the 100 kPa stiffness group was significantly higher relative to that of the 10 kPa stiffness group (P < 0.001) (Fig. 3D). These findings suggest that increased substrate stiffness enhanced mitochondrial function. Cellular homeostasis largely depends on the normal function of mitochondria and endoplasmic reticulum. Disruption of MERCs often leads to impaired cellular homeostasis, which is mainly characterized by unbalanced oxidation state and increased ER stress (Senft and Ronai, 2015). ROS flow cytometry showed that the CellROX concentration was higher in the 10 kPa stiffness group (P < 0.001), and the cellular antioxidant capacity was decreased (Fig. 3E and F). Western blotting showed that the expression of Bip and Chop, the phosphorylation level of IRE1α and endoplasmic reticulum stress levels were down-regulated in high 100 kPa stiffness substrate relative to the 10 kPa stiffness substrate (P < 0.001) (Fig. 3G and H). These results suggest that high substrate stiffness increases mitochondria–ER contacts and maintains the functions of mitochondria and ER.

Figure 3.

Increased mitochondria–endoplasmic reticulum contacts maintain mitochondrial and endoplasmic reticulum functions. (A) Fluorescence images of mitochondrial Mito-tracker staining in different substrate stiffness. Scale bar, 20 μm. NC, negative control, no Mito-Tracker. (B) Relative statistical analysis of (A). ***P < 0.001, Unpaired t-test. (C) Relative statistical analysis of ATP concentrations in different substrate stiffness. ***P < 0.001, Unpaired t-test. (D) Relative statistical analysis of mtDNA copy numbers in different substrate stiffness. ***P < 0.001, Unpaired t-test. (E) Flow cytometry detection of CellROX curves in different substrate stiffness. (F) Relative statistical analysis of (E). ***P < 0.001, Unpaired t-test. (G) Western blots of endoplasmic reticulum stress proteins BIP, CHOP, p-IRE1α, and IRE1α in different substrate stiffness. (H) Relative statistical analysis of (G). ***P < 0.001, *P < 0.05, two-way ANOVA. All data are mean±SD. Error bars represent SDs.

Substrate stiffness regulates the Rho/ROCK pathway and influences cytoskeletal remodeling

The activity of organelles is tightly linked to the mechanistically mediated cytoskeleton (Akhmanova and Kapitein, 2022; Kalukula et al., 2022; Yao et al., 2022). Mechanical signals can activate the Rho pathway, induce downstream molecules mDia1 and ROCK to promote G-Actin polymerization to F-Actin, induce the crosslinking and contraction of Actin bundles, and regulate cytoskeletal dynamics (Narumiya et al., 2009; Higashida et al., 2013). Therefore, three types of cytoskeleton proteins were selected: Tubulin, Actin, and Nestin, and these proteins were stained by immunofluorescence in testicular interstitial cells cultured on hydrogel substrates exhibiting either 100 or 10 kPa substrate stiffness. It was found that the content of F-Actin in the 100 kPa stiffness group was significantly increased (P < 0.001), while Tubulin and Nestin contents were not significantly changed in the 100 kPa stiffness group (Fig. 4A and B). Further, western-blot analysis showed that high substrate stiffness significantly up-regulated the expression of F-Actin (P < 0.001) and had a weak effect on the expression of Tubulin and Nestin (Fig. 4C and D). Timed fluorescent quantitative PCR was used to detect the mRNA expression levels of the three proteins, and it was found that the mRNA expression levels of F-Actin, Tubulin, and Nestin were not significantly changed in the high substrate stiffness group (Fig. 4E). These results suggest that high substrate stiffness could promote F-Actin polymerization without changing the total amount of Actin. Western-blot analysis was performed on the downstream related proteins of the Rho/ROCK pathway, and it was found that high substrate stiffness up-regulated the phosphorylation levels of downstream LIMK and Cofilin (P < 0.001) and activated the Rho/ROCK pathway (Fig. 4F and G, Supplementary Fig. S2A and B). To further verify the effect of the Rho/ROCK pathway on cytoskeletal proteins, we applied the Rho inhibitor Y-27632 to the high substrate stiffness group and found that the increase of F-Actin was significantly decreased (P < 0.001), while the mRNA expression levels of Actin had no significant changes (Fig. 4H and I, Supplementary Fig. S2C). The distance between mitochondria and endoplasmic reticulum in LCs cultured with Y-27632 on 100 kPa stiffness substrates was larger than that on 100 kPa stiffness substrates alone, suggesting that the close contacts between mitochondrial outer membrane and endoplasmic reticulum were reduced, which was also confirmed by the detection of testosterone levels in the supernatant (Supplementary Fig. S2D–F). In addition, the use of Rho/ROCK activator CN01-A also promoted F-Actin expression without affecting Tubulin and Nestin expression, further promoting testosterone production (Supplementary Figs S2G and H and S3A). To further examine the upstream and downstream relationship between mechanical signaling and the cytoskeleton mediated by the Rho/ROCK pathway, LCs cultured on 100 kPa substrate stiffness for more than 72 h were transferred to 10 kPa stiffness substrates for detection. It was found that the expression of F-Actin, p-LIMK, STAR, and CYP11A1 in the 100–10 kPa group was similar to that in the 10 kPa group and was significantly lower than that in the 100 kPa group (Supplementary Fig. S3B–E). In summary, substrate stiffness activates the Rho/ROCK pathway, affects the crosslinking and assembly of Actin bundles, and induces cytoskeletal remodeling.

Figure 4.

Substrate stiffness regulates the Rho/ROCK pathway affecting cytoskeletal remodeling. (A) Immunofluorescence images of Tubulin, F-Actin, and Nestin in Leydig cells cultured in different substrate stiffness. Scale bar, 10 μm. (B) Relative statistical analysis of (A). ***P < 0.001, ns, P > 0.05; two-way ANOVA. (C) Western blots of Tubulin, F-Actin, and Nestin in different substrate stiffness. (D) Relative statistical analysis of (C). ***P < 0.001, ns, P > 0.05; two-way ANOVA. (E) Quantitative PCR analysis of mRNA expression of Tubulin, F-Actin, and Nestin in different substrate stiffness. ***P < 0.001, ns, P > 0.05; two-way ANOVA. (F) Western blots of Rho/ROCK downstream proteins p-LIMK, LIMK, p-Cofilin, and Cofilin in different substrate stiffness. (G) Relative statistical analysis of (F). ***P < 0.001, ns, P > 0.05; two-way ANOVA. (H) Western blots of F-Actin in different substrate stiffness and treatment with Y-27632. (I) Statistical analysis of (H). ***P < 0.001, one-way ANOVA. All data are mean±SD. Error bars represent SDs.

Obstructing mechanical signaling inhibits testicular development in mice

In order to further explore the effects of mechanical cues on testicular development in mice, Y-27632, an inhibitor of the mechanical signal transduction pathway, was injected intraperitoneally into mice immediately after birth. It was found that there was no difference in testicular tissue stiffness between the Y-27632 groups and non-Y-27632 groups during development by AFM (Supplementary Fig. S4A and B). Mice injected with Y-27632 had a sparse interstitial distribution and a low number of interstitial cells in the testis (Fig. 5A, Supplementary Fig. S4C), accompanied by a decrease in serum testosterone levels (Fig. 5B). This suggests that mechanical signals can promote testicular development. We found that Y-27632 inhibited F-Actin synthesis during development (Fig. 5C and D, Supplementary Fig. S4D) and induced a decrease in ATP synthesis (Supplementary Fig. S4E). In addition, the detection of endoplasmic reticulum stress proteins by protein western blotting found that the contents of Bip, Chop, and p-IRE1a increased in the Y-27632 group (Fig. 5E and F, Supplementary Fig. S4F), suggesting that the inhibition of mechanical signal transduction pathway led to a decrease in energy metabolism. In summary, the mechanical signaling pathway enhances energy metabolism and promotes the number and function of testicular interstitial cells by reshaping the cytoskeleton during the development of testes in mice, thus promoting testicular development.

Figure 5.

Inhibition of biomechanical signaling suppresses mouse testis development. (A) Testis H&E staining of C57bl/6 mice after injection of Y-27632 at 4 and 8 weeks of age. Scale bar, 1 mm (left), 50 μm (right). (B) Statistical analysis of serum testosterone concentration in C57bl/6 mice after injection of Y-27632 at 4 and 8 weeks of age. n = 5. ***P < 0.001, two-way ANOVA. (C) Western blots of F-Actin in Leydig cells from mice after injection of Y-27632 for 4 and 8 weeks of age. n = 5. (D) Relative statistical analysis of (C). n = 5. ***P < 0.001, two-way ANOVA. (E) Western blots of endoplasmic reticulum stress proteins Bip, Chop, p-IRE1α, and IRE1α after injection of Y-27632 at 4 and 8 weeks of age. n = 5. (F) Relative statistical analysis of (E). n = 5. ***P < 0.001, *P < 0.05, ns, P > 0.05; two-way ANOVA. All data are mean±SD. Error bars represent SDs.

Organoids demonstrate that substrate stiffness regulates testosterone biosynthesis

In order to simulate the effects of mechanical signals on testicular development (Supplementary Fig. S5A), we used organoid techniques to grow testicular organoids on gels of different substrate stiffness to verify the effect of high substrate stiffness on testicular development (Supplementary Fig. S4E). We used a 3-LGS to construct testicular organoids, with different ratios of collagen to achieve high and low stiffness (Alves-Lopes et al., 2017, 2018). We found that organoids with high substrate stiffness had larger mean diameters at both 7 and 21 days than those with low substrate stiffness (Fig. 6A, Supplementary Fig. S5B). Further, through protein immunoblotting, it was found that the contents of key enzymes STAR and CYP11A1 of testosterone synthesis in the 100 kPa group were higher than those in the 10 kPa group (Fig. 6B and C), suggesting the promoting effect of high substrate stiffness on testosterone synthesis. To further verify the mechanism by which mechanical signals affect testicular development and testosterone synthesis, we used protein western blotting technology and found that the expression of F-Actin in the high-stiffness group was higher than that in the low-stiffness group (Fig. 6D and E). Similarly, the phosphorylation contents of LIMK and Cofilin downstream of the mechanical signal transduction pathway in the high substrate stiffness group were higher than those in the low substrate stiffness group (Fig. 6F and G), suggesting that high substrate stiffness activated the Rho/ROCK pathway and led to cytoskeletal remodeling, resulting in increased ATP synthesis (Fig. 6H) and increased testosterone synthesis (Fig. 6I).

Figure 6.

Organoids validate substrate stiffness regulation of testosterone biosynthesis. (A) Representative images of testicular organoids constructed in different substrate stiffness. Scale bar, 100 μm. (B) Western blots of STAR and CYP11A1 in different substrate stiffness. (C) Relative statistical analysis of (B). ***P < 0.001, two-way ANOVA. (D) Western blots of F-Actin in different substrate stiffness. (E) Relative statistical analysis of (D). ***P < 0.001, Unpaired t-test. (F) Western blots of Rho/ROCK downstream proteins p-LIMK, LIMK, p-Cofilin, and Cofilin in different substrate stiffness. (G) Relative statistical analysis of (F). ***P < 0.001, ns, P > 0.05; two-way ANOVA. (H) Relative statistical analysis of ATP concentration in different substrate stiffness. ***P < 0.001, Unpaired t-test. (I) Statistical analysis of supernatant testosterone concentration in different substrate stiffness. ***P < 0.001, Unpaired t-test. All data are mean±SD. Error bars represent SDs.

Discussion

In this study, we characterized the mechanical microenvironment during testis development and discovered an increase in substrate stiffness during this process for the first time. Further, we identified mechanically regulated MERCs as an important factor in modulating the testosterone synthesis capacity of LCs, enhancing our understanding of the regulatory mechanisms of LCs during testis development.

The development of testis is a multi-stage, multi-factorial process involving gene expression regulation (Ungewitter and Yao, 2013; Guo et al., 2020), intercellular signaling (Kroll et al., 2011), epigenetic regulation (Tsuji-Hosokawa et al., 2018), and dynamic adjustment of hormone levels. Testosterone can activate various signaling molecules such as MAPK and CREB, which are involved in the proliferation, differentiation, and functional regulation of testis tissue (Shupe et al., 2011). In this study, we first show that mechanical signals are involved in the regulation of testicular development. Since the mechanical properties of the ECM are difficult to control in vivo, we used hydrogels with different stiffness to mimic the in vivo ECM and explore the effect of mechanical signals on LCs in vitro. A substrate stiffness of 10 kPa can represent the behavior of cells in newborn testicular tissues under normal physiological conditions, and may also simulate some early pathological conditions (Trottmann et al., 2016; Pedersen et al., 2017; Roy et al., 2020) (such as the early stage of tumors). While 100 kPa can simulate the mechanical environment of mature testicular tissue, it is also related to high-stiffness pathological conditions such as fibrotic diseases or the tumor microenvironment (Trottmann et al., 2014; Dikici et al., 2016). By comparing LC behavior at 10 and 100 kPa substrate stiffness, we found that LCs can sense mechanical cues leading to cytoskeletal remodeling and increased MERCs with increasing substrate stiffness, which further promotes LCs metabolic activity. It also directly leads to elevated levels of testosterone and indirectly regulates the developmental process of testicular tissue. These results reveal the critical role of mechanical signaling in testicular development and provide a new perspective for understanding the mechanical response of testicular tissue under different pathological states.

MERCs play a key role in cell signaling, calcium ion, and ROS transmission (Stacchiotti et al., 2018), which is of great importance for maintaining the function and homeostasis of these two organelles (Bravo et al., 2012; Brzoskwinia et al., 2021). In the testis, mitochondrial dysfunction affects the transport of cholesterol to LCs and the energy supply for lipid metabolism (Wang et al., 2019; Arab et al., 2022), while endoplasmic reticulum stress inhibits the expression of testosterone biosynthesis genes such as STAR and 3β-HSD in mice testes (Kirat et al., 2023). We found that an increase in MERCs promotes the expression of key enzymes for testosterone synthesis such as STAR and CYP11A1 and leads to an increase in testosterone levels both in vitro and during testis development. These studies indicate that MERCs, as interactions between key organelles, play a central role in regulating testosterone synthesis.

The cytoskeleton is divided into microtubules, microfilaments, and intermediate filaments (Petzold and Gentleman, 2021), among which microfilaments are the main proteins to sense mechanical signals. Mechanical signals activate the Rho family and its downstream targets, ROCK (Rho-associated coiled-coil forming kinase), which can phosphorylate LIM-kinase, further phosphorylating Cofilin to inhibit the depolymerization of F-Actin, while mDia (mammalian homolog of Drosophila diaphanous) can promote the polymerization of G-Actin, ultimately leading to the rearrangement of microfilament proteins (Narumiya et al., 2009; Liu et al., 2018). We found that increased substrate stiffness promoted F-Actin rearrangement and Rho-GTP levels, indicating the important role of Actin rearrangement in sensing mechanical cues. Furthermore, by transferring LCs cultured on the 100 kPa substrate stiffness gels back to the 10 kPa gels, we found that Rho activation was down-regulated, further demonstrating that substrate stiffness is an important factor driving these changes. Korobova et al. (2013) pointed out that Actin filaments are involved in mitochondrial fission in MERCs. We also found that Actin polymerization can promote mitochondrial biogenesis, which deserves further investigation.

In summary, we have identified mechanically regulated MERCs as an important factor in modulating the testosterone synthesis capacity of LCs. It opens up a new pathway for further exploration of testis development and provides new insights into how substrate stiffness affects the testis.

Supplementary data

Supplementary data are available at Molecular Human Reproduction online.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Authors’ roles

J.Wu.: substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data; R.H.: substantial contributions to study conception and design, acquisition of data, and analysis and interpretation of data; Z.X.: substantial contributions to conception and design, acquisition of data, and analysis and interpretation of data; H.Y.: responsible for the experiment and text revision in the process of revision; Y.G.: responsible for the experiment and text revision in the process of revision; L.S.: drafting the article and revising it critically for important intellectual content; W.L.: drafting the article and revising it critically for important intellectual content; J.H., responsible for organoid experiments and drafting the article and revising it critically for important intellectual content; and J.Wa. and all authors gave final approval of the version to be published and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Funding

This work was supported by grants from the National Natural Science Foundation of China (32170799, 82371608), Guangdong Basic and Applied Basic Research Foundation (2023B1515020016), and Shenzhen Fundamental Research Program (JCYJ20240813150417024).

Conflict of interest

None declared.