All C57BL/6J mice (male, 8–10 weeks old) were purchased from the Animal Center at the Medical Laboratory of Guangdong Province, and all experimental animal processes were approved by the Animals Care and Used Committee of the First Affiliated Hospital of Sun Yat-sen University (Guangzhou, China). Animals had free access to a standard laboratory diet and water before and after surgery and were maintained in a constant environment of 50% humidity and 20°C with a 12-h dark-light cycle.72 mice were randomly divided into three groups: sham-operated control group (n = 24), testicular ischemia-reperfusion + saline group (n = 24) and testicular ischemia-reperfusion + icariside II treated group (n = 24). Mice were anesthetized by 10% chloral hydrate (3μL/1g). No analgesic was administered. Mice were anesthetized by 10% chloral hydrate (3μL/1g). No analgesic was administered. A standard testicular torsion model was established as previously described (22, 23). The abdomen was opened, and the left testis was rotated 360° in a clockwise direction followed by fixation with silk suture maintained for 30 minutes. The rotated testis was reversed and the contralateral testis was removed. ICA II in saline (5 mg/kg/day) was administrated by gavage immediately after detorsion (n=6 mice per group). The control animals (n=6 mice per group) received saline with the same volume. Normal animals were used as the untreated group. On day 1, 3, 7 and 14, 6 mice per group were sacrificed at each time point and testicular tissue samples were harvested for further analysis.

Testicular tissues were harvested from sacrificed mice on day 14. Tissues were fixed in 4% paraformaldehyde, embedded in paraffin and cut into 5-µm sections. The sections were dewaxed, rehydrated, and repaired with ethylenediamine tetra acetic acid. Testicular tissue sections were blocked for 60 min with 5% BSA then incubated with primary antibodies overnight at 4°C. Sections were washed with PBS three times and incubated with secondary antibodies for 60 min at room temperature in the dark. DAPI was used for 5 min to stain nuclei. The following antibodies were used: anti-F4/80 (Abcam, ab100790, 1:400), anti-Ki67 (Abcam, ab15580, 1:400), anti-SOX9 (Abcam, ab185966, 1:100), and anti-VASA (Abcam, ab270534, 1:100), anti-SCP3 (Abcam, ab97672, 1:100), anti-CD31 (Abcam, ab182981, 1:100), anti-α-SMA (Abcam, ab5694, 1:100), Alexa Fluor 488-conjugated goat anti-rabbit (Abcam, ab150077, 1:500), Alexa Fluor 488-conjugated goat anti-mouse (Abcam, ab150113, 1:500), Alexa Fluor 594-conjugated goat anti-rabbit (Abcam, ab150080, 1:500). Alexa Fluor 594-conjugated goat anti-mouse (Abcam, ab150116, 1:500), DAPI (Abcam, ab228549, 1:1000). Images were acquired using a confocal microscope (LSM800, Zeiss). To investigate the testis structure, Sections of 5 μm thickness were stained with haematoxylin and eosin-phloxine solution.

ROS are indicators of the level of oxidative stress and may be used to assess the oxidative environment of the testes. ROS production was assessed using a 2′,7′-dichlorodihydrofluorescein diacetate (Beyotime, S0063, 10 μmol/L) histochemical assay. Nuclei were stained with DAPI, and images were captured under a microscope (LSM800, Zeiss). The fluorescence intensity of the staining was measured using ImageJ.

Total RNA was purified from testicular tissue samples using Nucleozol reagent (Macherey-Nagel, 740404.200) according to the manufacturer’s protocol. Total RNA was reverse transcribed into cDNA using the Prime Script RT kit (TaKaRa, E237). Quantitative PCR (qPCR) was performed using SYBR mix (ChamQ, Q311-02) according to the manufacturer’s instructions. Signals were detected using a Bio–Rad CFX96 detection system (Roche). The primers that were used for this study are outlined in Table 1.

Testicular tissues were lysed in ice-cold RIPA lysis buffer (Elabscience^®, E-BC-R327) and the supernatants were collected. The supernatants protein concentrations were normalized using the BCA assay. Protein levels were analyzed using a LEGEND plex bead-based immunoassay (BioLegend, 740134) according to the manufacturer’s instructions. The mean fluorescence intensity of the cytokines was analyzed using LEGEND plex8.0 software.

Testicular tissues were cut into small pieces using micro scissors followed by the addition of collagenase type IV (Stem cell Technologies, 07909, 1 mg/mL) and digested in water at 37°C for 15 min. The digestion was terminated via the addition of PBS containing 10% BSA. The homogenate was filtered through a 50-μm filter and centrifuged at 1200 g for 3 min at 4°C. Cell precipitates were resuspended in PBS to obtain a mouse testicular cell suspension. For flow cytometry analysis of macrophages, monocytes and neutrophils, cells were incubated with antibodies diluted 1:100 at 4°C for 20 min, and antibodies were used as follows: CD45-PE-Cyanine7 (eBioscience™, 4277984, 1:100), F4/80-Brilliant Violet 421 (BioLegend, 123131, 1:100), CD11b-FITC (Abcam, ab24874, 1:100), Ly6G-APC (Thermo Fisher, 17-9668-82, 1:100) and Ly6C-APC (Abcam, ab93550, 1:100). For flow cytometry analysis of Leydig cell, cells were incubated with primary antibodies diluted 1:100 at 4°C for 20 min, cells were washed with PBS three times and incubated with secondary antibodies diluted 1:100 at 4°C for 20 min in the dark, and antibodies were used as follows: Anti-LH Receptor (Alomone Labs, ALR-010, 1:100) and Alexa Fluor 488-conjugated goat anti-rabbit (BioLegend, 406416, 1:100). For detection of cell apoptosis, PI and Annexin V-APC staining liquid (Elabscience^®, E-CK-A217) was added, and the cells were sorted using flow cytometry after avoiding light for 10 min to obtain the results of flow analysis. Fluorescence-activated cell sorting (FACS) was performed using MoFlo Astrios EQs (Beckman Coulter).

Testicular tissues or cells were lysed in ice-cold RIPA lysis buffer (Elabscience^®, E-BC-R327) containing phosphatase inhibitor cocktail (Abcam, ab201111) with 1 mM PMSF (Thermo Fisher, 36978) for 30 min. Lysates were centrifuged at 12,000 g for 15 min, and the supernatant was collected for protein determination. The concentration of total protein was measured using a BCA protein assay kit (Abcam, ab102536). Protein was prepared in SDS sample buffer, subjected to SDS–PAGE then transferred to a nitrocellulose membrane. The PVDF membranes were blocked in 5% non-fat powdered milk and incubated with primary antibodies diluted 1:1000 in blocking buffer. Bands were detected using the enhanced chemiluminescence (ECL) method (MerckMillipore, WBKLS0050). The following primary antibodies were used: anti-β-tubulin (AAB, A41015-2, 1:1000), anti-caspase3 (Cell Signalling Technology, 9662S, 1:1000), anti-cleaved-caspase3 (Cell Signalling Technology, 9664T, 1:1000).

Blood samples were collected from the tail vein of mice, and the samples were centrifuged (1000 g, 15 min) to obtain the serum. Testosterone levels were measured using an enzyme-linked immunosorbent assay (ELISA) kit (Abcam, ab108666). The absorbance of testosterone was measured at 405 nm.

Statistical analyses were performed using one-way ANOVA in GraphPad Prism version 8. Quantification of fluorescence intensity was performed using ImageJ. Data are expressed as the means ± SD for at least 3 experiments. P<0.05 was considered statistically significant.

To investigate whether ICA II attenuated testicular torsion injury, testes were collected 14 days after ICA II treatment. ICA II treatment significantly restored testis weights (Figure 1A). HE staining showed that the sham-operated animals had intact germinal epithelium. The structure of the seminiferous tubules was disrupted and the number of germ cells decreased significantly in the saline group. ICA II preserved spermatogenic function in testicular torsion-detorsion mouse model of ischaemia/reperfusion injury. Johnsen’s score was significantly higher in ICA II-treated group compared to the saline group (Figure 1B). Sertoli cells is essential for spermatogenesis (24). To verify whether ICA II had a beneficial effect on the maintenance of the spermatogenesis environment, we examined the expression of the Sertoli cell marker SOX9 (25, 26). SOX9⁺ cells virtually disappeared in the saline group compared to the sham group, and the number of SOX9⁺ cells increased significantly in the seminiferous tubules after ICA II treatment compared to the saline group (Figure S1A). VASA is a marker of germ cells, and VASA staining revealed a decrease in the proportions of VASA⁺ cells at day14 in the saline group. The number of VASA⁺ cells was less severely reduced in the seminiferous tubules in the ICA II-treated group. VASA⁺ cells were more abundant in the seminiferous tubules of the ICA II-treated group than the saline group (Figure 1C). To determine whether ICA II protected recipient spermatogenesis, we analysed the expression of Ki67, which is a marker that reflects the proliferation of spermatogonia. Images showed that seminiferous tubules of Ki67⁺ cells virtually decreased in the saline group compared to the sham group, and the number of Ki67⁺ cells increased significantly in the ICA II-treated group compared to the saline group (Figure S1B). We examined the expression of the meiotic marker synaptonemal complex protein 3 (SYCP3). SYCP3⁺ cells nearly disappeared in the saline group compared to the sham-operated group (Figure 1D), and the number of SYCP3⁺ cells increased significantly in the ICA II-treated group compared to the saline group. These results suggest that ICA II contributed to the protection of spermatogenesis.

Reproduction is dependent on the balance of testosterone (27, 28). Testosterone is primarily produced in Leydig cells. To investigate whether ICA II had beneficial effects on the maintenance of testosterone, we examined testosterone secretion in mice 14 days after testicular torsion. Testicular torsion reduced serum testosterone levels on day 14, and ICA II significantly rescued the testicular torsion-induced low testosterone (Figure 2A). ICA II significantly increased the mRNA expression levels of Leydig cell specific markers, Cyp11a1 and Cyp17a1 (Figure 2B). Flow cytometry revealed that testicular torsion reduced the number of Leydig cells on day14 and ICA II significantly rescued the ratio of Leydig cells (Figure 2C). Leydig cells gate selection was presented (Figure S2A). We observed that 3β-HSD⁺ cells were rarely observed in the interstitium in the saline group. However, ICA II treatment dramatically increased the number of Leydig cells (Figure 2D). These results suggest that ICA II promotes Leydig cell regeneration and testosterone synthesis.

Previous studies showed that ICA II improved the activity of antioxidant enzymes in STZ-induced diabetic rats, which produced ROS (29). The present study examined whether ICA II reduced acute oxidative damage. We analysed the level of ROS in testes using flow cytometry on day 1. ROS were excessively produced during ischaemia-reperfusion injury in saline group. However, ROS levels from the ICA II-treated group were significantly lower than the saline group 1 day after testicular torsion (Figure 3A). IF staining showed that ICA II treatment downregulated ROS production (Figure 3B). ROS triggers apoptosis via the intrinsic pathway. Germ cell apoptosis was significantly reduced in the ICA II group compared to the saline group (Figure 3C). Western blot analysis showed that the levels of cleaved caspase 3 in the ICA II group were lower than the saline group (Figure S3A). Taken together, these results indicate that the anti-ROS and anti-apoptotic effects of ICA II on the testes play an important role in protecting against testicular torsion injury.

Testicular torsion leads to germ cell apoptosis exacerbating the inflammatory response. Therefore, we investigated whether ICA II participated in testicular repair by rapidly and effectively suppressing excessive immune responses during the early inflammation stage. Flow cytometry revealed that the accumulation of CD11b⁺ Ly6C⁺ monocytes (Figure 4A) and CD45⁺F4/80⁺ macrophages (Figure S4A) were attenuated in the testes of ICA II-treated mice 3 days after testicular torsion compared to the saline group. IF staining analysis showed that the number of F4/80⁺ macrophages was greatly reduced in the ICA II-treated group 3 days after testicular torsion (Figure 4B). These data suggest that treatment with ICA II significantly reduced the total number of pro-inflammatory immune monocytes/macrophages in the injured testicular tissue after testicular torsion. Flow cytometry revealed that ICA II-treated mice showed diminished accumulation of CD11b⁺ Ly6G⁺ neutrophils in the testes at day 3 compared to the saline group (Figure 4C). Monocytes, macrophages and neutrophils gates selection was presented (Figure S4B). To further assess the effects of ICA II treatment on the testicular torsion environment, we performed qPCR analysis of genes involved in inflammation. The analyses were performed using whole testes tissue collected on day 3 of the ICA II treatment and saline group. The pro-inflammatory cytokine IL-1β was increased in saline group mice, the gene expression (Figure 4D) and protein level (Figure 4E) of IL-1β were suppressed in the ICA II-treated mice. These data indicate that ICA II significantly suppresses the inflammatory response.

ICA II has a protective role in promoting endothelial cell proliferation and alleviating smooth muscle atrophy in diabetic ED (30). We examined whether ICA II alleviated endothelial injury in testicular torsion. Flow cytometry revealed that α-SMA⁺ cells were lower in the testes of the saline group compared to sham group at day7, and the number of α-SMA⁺ cells increased significantly in the ICA II-treated group compared to the sham group (Figure 5A) smooth muscle gate selection was presented (Figure S5A). Flow cytometry found that CD31⁺ cells were lower in the saline group compared to the sham group in the testes at day7, and the number of CD31⁺ cells increased significantly in the ICA II-treated group compared to the sham group (Figure 5B). Endothelial cell gate selection was presented (Figure S5B). Immunofluorescence staining was used to evaluate the effect of ICA II on testis microvasculature. Similarly, the numbers of α-SMA⁺ and CD31⁺ cells were higher in the ICA II-treated group (Figures 5C, D). These results suggest that the therapeutic effects of ICA II promote the processes of vasculogenesis.