Written informed consent was obtained from patients and their relatives before obtaining the nucleus pulposus tissue during surgery. This study was approved by the Ethics Committee of Shanghai Changzheng Hospital (ID number: 2017SL040). The experiments in this study were carried out in biosafety laboratory according to the Laboratory Biosafety Manual in our laboratory. Human normal disk tissue was taken from surgical patients with or lumbar trauma (Pfirrmann grade I, n = 9, age 27–53 years, mean 42 years). Degenerated disk tissue was collected from patients with lumbar disk herniation (Pfirrmann grade II–III, n = 6, age 28–70 years, mean 51 years; and Pfirrmann grade IV–V, n = 6, age 38–75 years, mean 53 years). The degree of IDD was evaluated by the Pfirrmann grading system. Three individual samples in each group were used for immunohistochemistry staining. Tissue protein was extracted from three samples in each group. NPCs were isolated from nucleus pulposus tissue in three patients with congenital scoliosis or lumbar trauma but not with IDD, because IDD is characterized by the loss of NPCs and phenotypic abnormalities (Zheng et al., 2020).

Nucleus pulposus cells were isolated from different donors and these cells were used for the cell-based in vitro experiments in this study. The isolation process of NPCs was performed as described previously (Zhang et al., 2018). Nucleus pulposus tissue samples were washed three times with PBS, then cut and digested using 0.25% trypsin (Servicebio, Wuhan, China) and 0.01% EDTA at 37°C for approximately 45 min. Next, tissue pieces were digested by type II collagenase (Servicebio) for 4 h at 37°C. The cell suspension was sieved through a cell strainer (70 μm; Beyotime, Shanghai, China) and centrifuged at 1,000 × g for 5 min. NPCs were resuspended in DMEM/F12 with 10% fetal bovine serum (FBS, Gibco) and 1% penicillin/streptomycin (Beyotime) and cultured in a humidified atmosphere of 5% CO₂ at 37°C, changing medium every 3 days. Cells at second passage were used for subsequent experiments. To establish the degeneration model of NPCs, cells were incubated for 24 h then treated with Recombinant Human Interleukin 1β (IL-1β) (10 ng/mL, Beyotime) for 24 or 48 h. Total RNA or protein were then extracted (Bai et al., 2019; Zhang et al., 2019).

Total RNA was extracted from NPCs using the HiPure Total RNA Mini Kit (Magen, China) according to the manufacturer’s instructions. The HiScript^® III RT SuperMix for qPCR Kit (R323-01, Vazyme, Nanjing, China) was used to conduct reverse transcription with a 20 μl final reaction mixture according to the manufacturer’s instructions. Real-time PCR was performed using ChamQ^TM Universal SYBR qPCR Master Mix on an ABI 7500 Real-Time PCR system (Applied Biosystems, Foster City, CA, United States). The relative fold changes in mRNA expression were calculated using the 2^–ΔΔCt method. The sequences of primers used in this study are shown in Table 1.

Total cellular protein was extracted using radioimmunoprecipitation assay (RIPA) lysis buffer (Beyotime) and quantified using the bicinchoninic acid (BCA) protein assay kit (Beyotime). Protein extraction was performed in a similar manner for both tissues and cells in culture. Equal quantities of total protein were separated by 10% sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE), and then transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, Billerica, MA, United States) by electroblotting. The membranes were blocked using tris-buffered saline tween-20 (TBST) containing 5% skimmed milk at 25°C for 2 h, then incubated with primary antibody diluent of the target protein overnight at 4°C. Immunolabeling was detected using a Tanon Imaging System (version 5200, Tanon Science & Technology Co., Ltd., Shanghai, China). The following primary antibodies were used: ARG2 (CST, #55003), TNF-α (CST, #3707), IL-6 (CST, #3732), COX2 (CST, #12282), iNOS (CST, #20609), Iκ-Bα (CST, # 9247), p- Iκ-Bα (CST, # 2859), MMP3 (Zenbio, Chengdu, China, 380816), MMP13 (Zenbio, 820098), type II collagen (Abcam, ab34712), aggrecan (Abcam, ab36861), Bcl-2 (Zenbio, 381702), Bax (Zenbio, 200958), cleaved caspase3 (CST, #9664), P16INK4a (CST, #80772), and P21 (Abcam, ab109520). GAPDH (CST, #5174) was used as the protein loading control. The protein bands were quantified using ImageJ and normalized to GAPDH.

We silenced the expression of ARG2 in NPCs using small interfering RNA (siRNA) technology. siRNAs were purchased from GenePharma Co., (Shanghai, China). The ARG2 siRNA sequences were: GCUGAGGAAAUACACAAUATT (sense 5′-3′) and UAUUGUGUAUUUCCUCAGCTT (antisense 5′-3′). Briefly, a total of 5 × 10⁵ cells/well was seeded in 12-well plates. After 24 h, cells were transfected with ARG2 siRNA or control siRNA duplexes according to the manufacturer’s instructions. After 4 h, the medium was replaced with complete medium and IL-1β was added to the medium at a final concentration of 50 ng/ml. After 24 h incubation, RNA was extracted.

The plasmid containing full-length ARG2 and a negative control plasmid were designed by GenePharma (Shanghai, China). Packaging of viral vectors was performed as described previously (Kong et al., 2020), and a titers test was performed. NPCs were plated at a density of 3 × 10⁵ cells/well in 24-well plates. After 24 h, the cells were co-transfected with the recombinant plasmid using Lipofectamine 2000 transfection reagent. The medium was replaced 4 h later with fresh medium. The overexpression efficacy of ARG2 was verified by qPCR and western blotting.

Nucleus pulposus tissues from patients with lumbar trauma and patients with moderate and severe IDD were collected and fixed in 4% paraformaldehyde. Immunohistochemical staining was performed to locate ARG2 in human nucleus pulposus tissues as described previously (Tang et al., 2019). Sections were incubated with 0.1% trypsin for antigen retrieval for 30 min at 37°C, followed by blocking with bovine serum albumin (BSA) at room temperature for 15 min. Then, sections were incubated with a primary antibody against ARG2 (CST, #55003, 1:1,000) at 4°C overnight, followed by the secondary antibody (Servicebio, GB23303) and counterstaining with hematoxylin. Images were visualized by light microscopy (Olympus, Japan). Quantification analysis was performed using ImageJ. Immunostaining was independently assessed by three histopathologists who were blinded to the data and outcomes of the patients.

For type II collagen and MMP13 immunofluorescence staining, NPCs were fixed in 4% paraformaldehyde for 15–20 min, then permeabilized using 0.1% v/v Triton X-100 for 5 min. Samples were incubated with primary antibodies against type II collagen (1:100 – 1:500, Solarbio, K009364P), MMP13 (1:400 – 1:1,600, Servicebio, GB11247), and NF-κB p65 (1:1,000, Cell Signaling Technology, Inc., Danvers, MA, United States, D14E12, #8242) diluted in 0.2% w/v BSA-TBS for 1 h. After washing with PBS, samples were incubated with DAPI solution (Servicebio, G1012) for visualization of the nuclei followed by incubation with FITC-conjugated goat anti-rabbit IgG (Servicebio, GB22303) and Cy3-conjugated goat anti-rabbit IgG (Servicebio, GB21301) for 30 min in the dark. Fluorescence detection was performed by fluorescence microscope (Olympus, Japan). Immunofluorescence quantification was performed using ImageJ, following the ImageJ User Guide.

JC-1 is a fluorescent probe commonly used to measure mitochondrial membrane potential (MIMP). NPCs were seeded in 6-well plates for 24 h. JC-1 (200X) was added to ultrapure water and JC-1 staining buffer (5X) to make the JC-1 working solution. The JC-1 working solution was added into the wells and incubated at 37°C for 20 min. Samples were then washed twice with JC-1 staining buffer (1X) and visualized by confocal microscopy (Carl Zeiss LSM 700). Quantification was performed using ImageJ, following the ImageJ User Guide.

Apoptosis was detected by the terminal deoxynucleotidyl transferase (TdT) dUTP nick end labeling (TUNEL) method. NPCs were cultured in six-well plates. After fixing in 4% paraformaldehyde for 1 h, samples were permeabilized with 0.3% Triton X-100 for 10 min. Samples were washed twice with PBS, then incubated with the TUNEL detection solution for 60 min at 37°C. Next, samples were washed three times with PBS, followed by incubation with DAPI solution. Finally, TUNEL-positive cells were detected by fluorescence microscopy (570 nm, Olympus, Japan). The number of TUNEL-positive cells was counted in three randomly selected fields. Quantification was performed using ImageJ, following the ImageJ User Guide.

Cells were fixed with 0.2% glutaraldehyde at room temperature for 15 min. After washing twice with PBS, samples were incubated with SA-β-gal working solution at 37°C overnight. Images were captured with a light microscope (Olympus, Japan). Quantification was performed using ImageJ, following the ImageJ User Guide.

Reactive oxygen species was detected using a ROS Assay Kit (Beyotime). Briefly, cells were incubated with DCFH-DA (diluted to 1:1,000 in serum-free medium) at 37°C for 20 min, then washed three times in serum-free cell culture medium to remove excess fluorescent probes. Samples were visualized by confocal microscopy (Carl Zeiss LSM 700). Quantification was performed using ImageJ, following the ImageJ User Guide.

The concentration of TNFα and IL-1β in cell culture supernatants was measured using an ELISA kit (Abcam, ab181421, ab214025) according to the manufacturer’s instructions.

Experiments were carried out more than three times. Quantification of immunohistochemical analyses, immunofluorescence, and western blotting data were performed using ImageJ software. All quantitative data are presented as the mean ± standard deviation. Two-group comparisons were analyzed by Student’s t-test, and three or more group comparisons were performed by one-way analysis of variance. Immunohistochemical data were analyzed by unpaired Students t-test. Non-parametric data (Pfirrmann grading) were analyzed by the Kruskal–Wallis H test. Statistical analyses were performed using SPSS version 18.0 (IBM, Armonk, NY, United States). ^∗P < 0.05, ^∗∗P < 0.01, and ^∗∗∗P < 0.001 were considered to be statistically significant.

The flow diagram of this study is shown in Figure 1A. The expression of ARG2 during the development of IDD was examined by immunohistochemical staining, quantitative reverse transcription PCR (qRT-PCR), and western blotting. Normal nucleus pulposus tissues were obtained from patients diagnosed with lumbar trauma and the degenerated nucleus pulposus tissues were acquired from patients with IDD. The intervertebral disk samples were divided into three groups [grade I: normal (the nucleus pulposus tissues were all from patients with lumbar trauma), n = 3; grade II–III: moderate degeneration, n = 3; and grade IV–V: severe degeneration, n = 3] according to the modified Pfirrmann grading system. Immunohistochemistry indicated that ARG2 expression was significantly higher in degenerated intervertebral disk than normal intervertebral disk. Furthermore, ARG2 expression was higher in severe IDD than moderate IDD (Figures 1B,C). The expression of ARG2 was also examined in nucleus pulposus tissues from patients with lumbar vertebral fracture or IDD. Compared with normal nucleus pulposus tissues, the mRNA and protein levels of ARG2 were significantly higher in both moderately and severely degenerative nucleus pulposus tissues (Figures 1D–F), suggesting that ARG2 was involved in the degeneration of human intervertebral disk.

The effects of IL-1β on ARG2 mRNA and protein expression were examined in NPCs that had been treated with IL-1β (10 ng/ml) for 24 h. ARG2 protein levels were significantly enhanced by IL-1β treatment (Figures 2A,B). Similarly, qRT-PCR analysis revealed that IL-1β promoted ARG2 mRNA expression (Figure 2C). Taken together, our results suggest that IL-1β could promote the expression of ARG2 in NPCs.

Three specific siRNAs for ARG2 and a control siRNA were used in this study, and the siRNA with the highest silencing efficiency was selected for subsequent experiments (Figure 3A). ARG2 mRNA levels were significantly reduced by siRNA treatment, demonstrating that ARG2 expression was successfully silenced in human NPCs. The efficacy of siRNA-ARG2 in NPCs treated with IL-1β was assessed by measuring ARG2 mRNA levels. We found that siRNA-ARG2 could also significantly reduce ARG2 levels under these conditions (Figure 3B). Degradation of the ECM is a primary factor contributing to the pathogenesis of IDD. To investigate the effect of ARG2 downregulation on ECM degradation, we examined MMP3 and aggrecan mRNA levels in NPCs transfected with siRNA−NC or siRNA−ARG2 and treated with IL-1β. We found that ARG2 silencing reduced MMP3 mRNA expression while enhancing aggrecan mRNA levels in the NPCs treated with IL-1β (Figures 3C,D). Immunofluorescence staining revealed that siRNA-ARG2 significantly reduced MMP13 expression and increased type II collagen levels in NPCs induced by IL-1β (Figure 3E). Moreover, the effects of ARG2 silencing on ARG2, type II collagen, aggrecan, MMP3, and MMP13 proteins expression were consistent with these findings. A significant increase in type II collagen and aggrecan protein expression, and significant reduction in MMP3 and MMP13 levels were observed in the ARG2 silencing NPCs treated with IL-1β (Figures 3F,H–K). Taken together, our data suggest that downregulation of ARG2 ameliorates NPC ECM degradation.

The pathogenesis of IDD is closely associated with cellular senescence and apoptosis of NPCs. To quantify the number of senescent cells, we performed SA-β-gal staining and calculated the ratio of SA-β-gal-positive cells of the IL-1β-treated NPCs (Figures 4A,B). We found that siRNA-ARG2 significantly reduced the number of SA-β-gal-positive senescent NPCs. Consistent with these findings, we showed that protein expression levels of the senescence markers, P16INK4a and p21, were significantly reduced by siRNA-ARG2 (Figures 4C–E). Thus, our data indicated that silencing ARG2 alleviated IL-1β-induced cellular senescence in NPCs. Next, we examined the effects of ARG2 silencing on apoptosis. Western blot analysis revealed that siRNA-ARG2 decreased the levels of apoptotic cleaved-caspase 3 and Bax, while increasing expression of the anti-apoptotic Bcl2 compared with the IL-1β treatment group (Figures 4C,F,G). Changes in the MIMP reflect mitochondrial dysfunction during the early stages of apoptosis. Thus, we used JC-1 staining to detect MIMP and measure early apoptosis. Monomeric JC-1 (green fluorescent) represents high-MIMP, which reversibly transforms into aggregated JC-1 (red fluorescent) with loss of MIMP. Our data showed that siRNA-ARG2 significantly improved the MIMP inhibited by IL-1β (Figures 4H,I). Finally, we assessed the rate of apoptosis with TUNEL staining. We found that ARG2 silencing markedly reduced the number of TUNEL-positive cells induced by IL-1β (Figures 4J,K). Taken together, our data indicate that siRNA-ARG2 ameliorated cellular IL-1β-induced senescence and apoptosis in NPCs.

Increased oxidative stress results in cellular apoptosis. ROS levels frequently represent intracellular oxidative conditions. To investigate the effects of ARG2 silencing on oxidative stress, we examined ROS levels in IL-1β-treated NPCs after treatment with siRNA-ARG2. IL-1β-induced ROS levels were significantly reduced by ARG2 silencing (Figures 5A,B), suggesting that the oxidative stress in NPCs was ameliorated via the downregulation of ARG2.

The inflammatory response that promotes both cellular senescence and apoptosis plays a vital role in the pathogenesis of IDD. In order to further examine the association between ARG2 and NPC inflammation, we examined the effect of ARG2 downregulation on IL-1β-induced proinflammatory factors. Our western blot data showed that IL-1β significantly enhanced the expression of COX2, iNOS, IL-6, and TNF-α. However, downregulation of ARG2 dramatically decreased COX2, iNOS, IL-6, and TNF-α levels compared with IL-1β treatment alone (Figures 5C–G). Consistent with these findings, our ELISA data also indicated that IL-1β treatment significantly increased IL-6 and TNF-α levels, while siRNA-ARG2 reversed these effects (Figures 5H,I). Taken together, our findings indicate that inhibition of ARG2 reduces inflammation in NPCs.

To further examine the effects of ARG2 upregulation on NPCs, we constructed an ARG2 overexpression plasmid (OE) and examined the effects of ARG2 on ECM metabolism, apoptosis and the inflammatory response. The efficiency of ARG2 overexpression was quantified by qPCR and western blotting. ARG2 expression levels were significantly higher in the OE group than in the vec group (Figures 6A–C). MMP13 and aggrecan are markers of ECM metabolism in NPCs. Overexpression of ARG2 led to a significant increase in IL-1β-induced MMP13 expression levels in NPCs (Figures 6A,D). Conversely, compared with the vec + IL-1β group, the level of aggrecan was markedly decreased in the OE + IL-1β group (Figures 6A,E). Compared with the vec + IL-1β group, overexpression of ARG2 in the OE + IL-1β group resulted in a significant decrease in the anti-apoptotic marker Bcl2, while the apoptotic marker Bax was markedly increased, suggesting that apoptosis was exacerbated in NPCs (Figures 6A,F,G). COX2 protein levels were also markedly higher in the OE + IL-1β group than the vec + IL-1β group (Figures 6A,H). The senescence-associated gene p21 was also found to be significantly upregulated in the OE + IL-1β group compared with the vec + IL-1β group (Figures 6A,I). To further elucidate the effect of ARG2 expression on the inflammatory response, TNFα and IL-1β levels in the cell culture supernatants were measured by ELISA. Both TNFα and IL-6 levels were significantly increased in IL-1β-treated cells overexpressing ARG2 (Figures 6J,K). ROS generation was also increased in the OE + IL-1β group (Figures 6L,M). Our findings indicate that overexpression of ARG2 increased ECM degradation, apoptosis, and inflammation in IL-1β-treated NPCs.

The NF-κB pathway plays a critical role in oxidative stress and apoptosis in NPCs (Kang et al., 2017; Zhu et al., 2019). Furthermore, some studies have indicated that ARG2 is involved in activation of the NF−κB pathway (Grandvaux et al., 2005). Thus, we sought to determine whether ARG2 regulates the NF−κB pathway in NPCs following treatment with IL-1β. Normally, NF-κB combines with IκBα, thereby inhibiting the nuclear translocation of NF-κB p65. Here, ARG2 protein levels were significantly elevated after overexpression of ARG2 (Figures 7A,B). In the absence of IL-1β, ARG2 did not affect the phosphorylation and degradation of IκBα (an endogenous NF-kB inhibitor) (Figures 7A,C). However, as shown in Figures 7D,F,G, in the presence of IL-1β, silencing ARG2 expression led to a decrease in phosphorylated IκBα, while ARG2 overexpression enhanced phosphorylation of IκBα (Figures 7D,F,G). Immunofluorescence staining was used to determine the intracellular localization of p65 (a major component of NF-κB) (Figure 7J). After IL-1β treatment, reduced p65 nuclear localization was observed in ARG2-silenced NPCs, whereas increased p65 translocation into the nucleus was found in ARG2-overexpressing NPCs relative to control NPCs. To further validate the involvement of NF-κB signaling in the ARG2-mediated inflammatory, oxidative stress, and cellular apoptosis responses, JSH-23 (a small molecule inhibitor targeting NF-κB pathway) was used. As shown in Figures 7D,E, JSH-23 treatment did not affect ARG2 protein expression (Figures 7D,E), but did result in a significant decrease in the amount of nuclear p65 (Figure 7J). Overexpression of ARG2 inhibited Bcl2 expression and promoted Bax expression (Figures 7D,H). However, inhibition of the NF-κB pathway with JSH-23 reversed the effect of ARG2 overexpression on Bcl2 and Bax levels (Figures 7D,H). IL-1β facilitates the release of inflammatory mediators. Using ELISA to detect levels of inflammatory factors, we found that IL-6 levels were significantly increased in ARG2-overexpressing cells treated with IL-1β (Figure 7I). However, inhibition of the phosphorylation of IκBα led to a significant reduction in IL-6 levels. We also found that JSH-23 significantly decreased ROS levels (Figure 7K). Taken together, our data suggest that ARG2 promoted IL-1β-induced oxidative stress, inflammatory response, and apoptosis through the NF-κB pathway.