The collection of human NP tissues and related experiments were approved by the Ethics Committee of the Second Hospital of the Army Medical University.

All surgical, therapeutic and postoperative animal care procedures were carried out in strict compliance with the Laboratory Animal Welfare and Ethics Committee of the Third Military Medical University (AMUWEC20193008).

MitoTEMPO, H₂O₂, CCCP, and collagenase II were obtained from Sigma (St. Louis, MO, United States). Primary antibodies against P16 and LC3B were obtained from Abcam (Cambridge, United Kingdom). Primary antibodies against OPTN, Parkin, and PINK1 were purchased from Santa Cruz Biotechnology (Dallas, TX, United States). Primary antibodies against β-actin, P21, TIMP1, Aggrecan, Collagen II, MMP-1, and P62 were provided by Proteintech (Wuhan, Hubei, China). Cyclosporin A (CsA) was obtained from Selleck (Houston, TX), and 4′,6-diamidino-2-phenylindole (DAPI) was acquired from Beyotime (Shanghai, China).

To study the connection between OPTN and the severity of IVDD, NP tissue was collected from 20 patients with IVDD (3 males and 2 females, aged 18–40 years, grade II; 4 males and 1 female, aged 16–39 years, grade III; 3 males and 2 females, aged 17–40 years, grade IV; 3 males and 2 females, aged 20–40 years, grade V) according to Pfirrmann grading criteria and used for further experiments (Pfirrmann et al., 2001). In our study, the patients from which NP tissue was collected during surgery did not have other diseases associated with IVDD. The collected NP tissues were lysed to extract proteins for Western blotting.

Gelatinous NP tissue was collected from the tails of 1-, 2-, and 20-month-old rats (Sprague–Dawley rats, male) under aseptic conditions. NP tissue was collected from 1- and 20-month-old rats, and protein was extracted for Western blotting. As described in a previous study (Chen et al., 2020; Lin et al., 2020), NP tissues from two-month-old rats were minced under a dissecting microscope, and the minced NP tissues were incubated in 0.2% collagenase II (St. Louis, MO, United States) for 2 h at 37°C. After being washed twice, the tissues were transferred to T25 flasks containing DMEM/F12 (BI, Kibbutz Beit-Haemek, Israel) supplemented with 10% fetal bovine serum (BI, Kibbutz Beit-Haemek, Israel) and antibiotics (1% streptomycin/penicillin) and incubated at 5% CO₂ and 37°C. The medium was changed every 3 days. The cells were digested with 0.25% trypsin-EDTA (Gibco, Invitrogen) when they reached 80% confluence and were then passaged. Third-generation NPCs were used for subsequent experiments.

To study OPTN expression levels in NPCs during oxidative stress, H₂O₂ was used to induce cellular oxidative damage (Lin et al., 2021a; Cheng et al., 2021). NPCs were exposed to different concentrations of H₂O₂ (0, 200, 400, 500, 600, 800 μM) for 2 h. The optimal conditions for inducing cellular senescence were determined by treating NPCs with the same concentration of H₂O₂ (400 μM) for different times (0, 0.5,1, 2, 4, 6 h). To assess mitophagy, NPCs were treated with or without H₂O₂ and then with or without CCCP (30 μM, 1 h). Then the mitophagy inhibitor CsA (10 μM, 1 h) and MitoTEMPO (200 μM, 2 h) were added, and the cells were incubated before subsequent assays.

Lentiviral constructs expressing tandem mCherry-eGFP-FIS1 (GeneChem, China) were used to transfect the NPCs. Briefly, 1 × 10⁵ NPCs were seeded on glass coverslips in 12-well plates and cultured at 37°C until the cells reached 30–50% confluence. After polybrene was added to the medium, lentiviruses expressing mCherry-eGFP-FIS1 were added at a multiplicity of infection (MOI) of 10, 25, or 50 according to the manufacturer’s instructions. After 48 h, the NPCs transfected with the mCherry-eGFP-FIS1 lentiviruses were used for follow-up experiments. Red particles, which represented mitolysosomes, were counted, and the data were statistically analyzed.

An overexpression lentivirus and small hairpin RNAs targeting OPTN were purchased from GeneChem (Shanghai, China). The OPTN-RNAi target sequence was GAG​CTG​CTG​CAG​CAA​ATG​AAA, and the control sequence was TTC​TCC​GAA​CGT​GTC​ACG​T. When the cells reached 30–50% confluence, they were transfected with the lentivirus. After 12–16 h, the medium was replaced. Forty-8 hours later, the cells were screened with medium containing puromycin. When the cells reached 80% confluence, the transfected NPCs were passaged and cultured with medium containing puromycin for subsequent experiments.

Total protein was extracted from NP tissues and cells using precooled radioimmunoprecipitation analysis (RIPA) lysis buffer containing 1 mM phenylmethane sulfonyl fluoride (PMSF). The protein concentrations were measured with a BCA protein assay kit (Beyotime, Shanghai, China). Then, the proteins were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) and transferred to polyvinylidene difluoride (PVDF) membranes (Millicon, United States). The PVDF membranes were blocked with 5% nonfat dry milk for 2 h and then incubated with primary antibodies against OPTN (1:200), P16 (1:2000), LC3B (1:1000), Parkin (1:500), β-actin (1:5000), P21 (1:1000), TIMP1 (1:500), Aggrecan (1:500), Collagen II (1:500), MMP-1 (1:500), and P62 (1:1000) overnight at 4°C. The blots were then washed with Tris-buffered saline/Tween 20 (TBST) three times for 10 min each and incubated with corresponding secondary antibodies for 2 h. After the blots were washed with TBST, the bands were visualized and quantified using an electrochemiluminescence kit (1705060, Bio-Rad, CA, United States), a gel imaging system (Bio-Rad, CA, United States) and Image Lab 3.0 software (Bio-Rad, United States).

Intracellular mtROS levels were measured using MitoSOX (Ali et al., 2021). Briefly, rat NPCs were seeded in a six-well plate and treated as described previously. Then, the NPCs were incubated with 5 μM MitoSOX for 20 min at 37°C, washed twice with PBS, and incubated with Hoechst 33342 solution for 15 min at 37°C. Staining was quantified using a fluorescence microscope (Olympus IX73, Tokyo, Japan).

The MMP was measured using MitoTracker Red CMXRos (C1049B Beyotime, Biotechnology, Shanghai, China), which accumulates in live cells in an MMP-dependent manner to stain mitochondria. NPCs were incubated with 200 nM MitoTracker Red for 30 min at 37°C and then stained with Hoechst 33342 dye at 37°C for 15 min. A fluorescence microscope (Olympus IX73, Tokyo, Japan) was used to observe the staining results, and ImageJ/Fiji (NIH, Bethesda, MD, United States) was used to quantify the results.

Sa-β-gal staining was performed using the Cellular Senescence β-Galactosidase Staining Kit (C0602 Beyotime Biotechnology, Shanghai, China) according to the manufacturer’s instructions. Briefly, NPCs were treated in six-well plates, washed once with PBS, and fixed with β-galactosidase staining solution for 15 min at room temperature. The cells were washed 3 times with PBS for 3 min each. One milliliter of staining working solution was added to each well, and the cells were incubated overnight at 37°C. Images were taken using an Olympus IX71 microscope, and SA-β-gal-positive cells were quantified for statistical analysis.

Twenty-four 8-week-old male Sprague–Dawley rats were randomly divided into 3 groups: the sham group (skin excision but no puncture of the intervertebral space), Lv-ctrl group (puncture of the intervertebral space + injection of empty virus), and Lv-OPTN group (puncture + OPTN overexpression virus injection). A rat model of IVDD was established as described previously (Zhang et al., 2009). After the rats were anesthetized with 2% (w/v) sodium pentobarbital (40 mg/kg), the tail disc (CO4/5) was localized by palpation and labeled. After being sterilized, a puncture needle (27 G) was passed vertically through the back skin into the annulus fibrosus (AF) at a depth of 4 mm, rotated 360° and left in the disc for 1 min. The health of the rats was monitored daily after surgery, and the rats were sacrificed at 8 weeks for follow-up experiments.

Rats that had undergone in vivo experiments were X-rayed at 0 and 8 weeks after surgery using a multimode small animal live imaging system (Xtremell, Bruker, Belgium). Briefly, after being anesthetized by intraperitoneal injection of 2% pentobarbital, the rats were fixed in the prone position on an imaging table, and the disc heigh index (DHI) was measured as previously described (Issy et al., 2013).

MRI was performed with a 7.0 T MRI device (Bruker PharmaScan, Germany) at 0 and 8 weeks postoperatively. Structural and signal differences in the tail discs were assessed with sagittal T2-weighted images. Magnetic resonance parameters were selected according to previous reports (Pfirrmann et al., 2001), and a researcher blinded to the groups evaluated the MRI images in accordance with the Pfirrmann grading criteria.

After MRI and X-ray, the rats were sacrificed by intraperitoneal injection of an overdose of sodium pentobarbital, and the intervertebral discs and adjacent caudal vertebrae of the segments of interest were collected. The specimens were fixed in paraformaldehyde, decalcified, dehydrated embedded in paraffin and cut into sections at a thickness of 5 μm for hematoxylin-eosin and safranin O-fast green staining, and disc degeneration was evaluated. The histological scores were determined in a blinded manner according to histological grading criteria (Tam et al., 2018) (0-5: normal; 6-11: moderate; 12-15: for severe).

Statistical analysis of the raw data was performed using the statistical software SPSS 23.0 (IBM, United States) and GraphPad Prism 8.00 (GraphPad Software, United States), and the results are expressed as the mean ± S.D. Two sets of data were statistically analyzed using Student’s t test. Multiple sets of data were analyzed by one-way analysis of variance (ANOVA) followed by Tukey’s test. Nonparametric data were analyzed by the Kruskal–Wallis H test. A p value <0.05 was considered statistically significant.

To investigate whether OPTN expression is related to IVDD, we collected NP tissues from five patients with Pfirrmann grade II-V IVDD and measured OPTN expression levels. We also collected caudal NP tissues from young and old rats. As shown in Figures 1A–E, the expression level of OPTN decreased as disc degeneration increased in both human and rat NP specimens. We examined OPTN expression in H₂O₂-treated rat NPCs; H₂O₂ was used to mimic oxidative stress-induced NPCs senescence, and the cells were treated with 400 μM H₂O₂ for 2 h, which was selected as the optimal conditions (Supplementary Figures S1A, B). H₂O₂-induced oxidative stress increased OPTN expression in NPCs (Figures 1F,G). As shown in Supplementary Figure S1D, we found that oxidative stress decreased proteoglycan and Collagen II levels, increased MMP-1 expression and decreased TIMP1 expression in NPCs. These results suggest that H₂O₂-induced oxidative stress exacerbates cellular senescence and matrix degeneration in rat NPCs. In summary, our results indicate that OPTN may be involved in the pathological process of IVDD.

Since the change in OPTN expression in human and rat NP tissues was the opposite of that in H₂O₂-treated rat NPCs, we knocked down and overexpressed OPTN to investigate the effect of OPTN on H₂O₂-treated NPCs. First, we overexpressed and knocked down OPTN in rat NPCs (Figures 2A,B,H,I). We then examined the cellular levels of senescence and matrix homeostasis proteins. The Western blot results indicated that OPTN knockdown increased the expression of P16 and P21 and impaired matrix homeostasis in NPCs (Figures 2C, D, E), whereas OPTN overexpression suppressed the expression of P16 and P21 under H₂O₂-induced oxidative stress and reversed the disruption of matrix homeostasis caused by OPTN knockdown (Figures 2J, K, L). SA-β-gal staining was used to examine cellular senescence. Compared with empty vector transfection, OPTN overexpression vector transfection reduced the percentage of SA-β-gal-positive NPCs, and this change was reversed by OPTN knockdown (Figures 2F,G, M, N). These results show that OPTN protects against oxidative stress in NPCs.

Since OPTN acts as a bridging protein in mitophagy, we hypothesized that its role in inhibiting NPCs senescence and maintaining matrix homeostasis might be related to mitophagy. The mitophagy activator CCCP and inhibitor CsA were added to H₂O₂-treated NPCs, and changes in mitophagy were observed. The number of mitolysosomes was considerably increased in the CCCP group compared with the H₂O₂ group, while CsA had the opposite effect (Figures 3A,B). As shown in Figures 3C,D, the Parkin, PINK1, and LC3II/LC3Ⅰ ratios were notably higher in the CCCP group, while the P62 ratio was significantly lower. CsA had the opposite effects. CCCP activated mitophagy in H₂O₂-treated NPCs, while CsA inhibited mitophagy. The levels of the senescence-related proteins P16 and P21 and the percentage of SA-β-gal-positive cells were significantly decreased in the CCCP group compared with the H₂O₂ group, and matrix homeostasis was significantly restored in the CCCP group, indicating that CsA inhibited the protective effect of OPTN, increased the expression of senescence-related proteins, and promoted matrix degradation (Figures 3C,E–H). Taken together, these results suggest that H₂O₂-induced NPCs senescence and matrix degradation can be reduced by activating mitophagy.

H₂O₂ induces oxidative stress in NPCs (Supplementary Figure 1C), and mitochondria are damaged by this process, resulting in the production of a large amount of mtROS. Mitophagy, as the main process by which injured mitochondria are cleared in vivo, plays an important role in scavenging mtROS. We examined the correlation between mitophagy and mtROS levels in NPCs in response to H₂O₂ treatment. The results indicated that the fluorescence intensity of the mtROS indicator MitoSOX was significantly reduced in the presence of CCCP and that the inhibitor had the opposite effect (Figures 4A,B). This finding proves that mitophagy deficiency caused by H₂O₂ is the main cause of mtROS generation. To further verify the relationship between mtROS and NPCs senescence and matrix degradation, we assessed changes in NPCs using the mtROS scavenger mitoTEMPO. The results confirmed that the mtROS scavenger could effectively inhibit H₂O₂-induced cell senescence and matrix degradation. Furthermore, mitoTEMPO mitigated CsA-induced cellular senescence and matrix degradation (Figures 4C–G). These results suggest that mitophagy attenuates the damage to NPCs caused by H₂O₂ by scavenging mtROS.

To demonstrate that OPTN mitigates cellular senescence and matrix degeneration by scavenging mtROS through mitophagy, we performed further experiments. As shown in Figures 5A,B, the number of mitolysosomes was significantly increased in the OPTN overexpression group compared with the H₂O₂ group, indicating that OPTN, as a mitophagy receptor, can significantly enhance mitophagy. As shown in Figures 5C,D, we observed that the fluorescence intensity of MitoSOX in OPTN overexpression vector-transfected NPCs was significantly reduced compared with that in empty vector-transfected NPCs, while the fluorescence intensity of MitoSOX was increased by the addition of CsA, demonstrating that OPTN inhibited the production of mtROS by enhancing mitophagy. Figures 5E–I show that compared with the empty vector-transfected group, the OPTN overexpression vector-transfected group showed a significant reduction in the percentage of Sa-β-gal-positive cells after H₂O₂ treatment; decreased expression of P16, P21, and MMP-1; and enhanced expression of Aggrecan, Collagen II, and TIMP1. These changes were inhibited by the addition of CsA; however, the effect of CsA was reversed by the addition of mitoTEMPO. These results demonstrate the mechanism by which OPTN scavenges MitoSOX by enhancing mitophagy, thereby mitigating cellular senescence and matrix degeneration.

To investigate the effect of OPTN in a rat model of IVDD, we used sagittal T2-weighted magnetic resonance images, and histological scores and measured the intervertebral space height (DHI%) to assess the degree of IVDD in rats at 8 weeks after surgery. OPTN expression in each group was measured by Western blotting. The OPTN expression level was significantly higher in the Lv-OPTN group than in the other two groups (Supplementary Figure S1E). The results indicated that the sagittal T2-weighted MRI signal in the caudal disc was higher and the Pfirrmann grade was lower in the Lv-OPTN group than in the Lv-ctrl group (Figures 6A,B). The DHI% was restored to near normal levels in the Lv-OPTN group compared to the Lv-ctrl group (Figures 6C,D). The H&E staining and safranin O-fast green staining results confirmed the therapeutic effect of OPTN (Figures 6E,F). H&E staining showed that NP tissue in the sham-operated group was large and occupied most of the space in the disc. The surrounding AF was more organized in the sham-operated group than in the Lv-ctrl group. The cells in the NP were stellate and evenly dispersed in a proteoglycan-rich matrix. However, in the Lv-ctrl group, NP tissue was reduced in size and surrounded by a disorganized AF, with cells aggregated in clusters and separated by a streaked proteoglycan matrix, indicating severe degeneration of NPCs. These morphological changes in NP tissue and disintegration and fibrosis of disc structures were significantly alleviated in the Lv-OPTN group compared to the Lv-ctrl group. NP tissue was stained with safranin O in the sham-operated group, indicating the presence of a large amount of proteoglycan matrix. Substantial proteoglycan loss was observed in the Lv-ctrl group compared with the sham-operated group, and the Lv-OPTN group showed less proteoglycan loss. These results confirm our in vitro findings that OPTN upregulation increases the ability of NPCs to resist injury and attenuates the severity of IVDD.