Awakening endogenous repair: salidroside boosts mitophagy in NPMSCs via SIRT1/FOXO3 to combat intervertebral disc degeneration

Abstract

Background

Intervertebral disc degeneration (IVDD) is a major cause of low back pain, characterized by mitochondrial dysfunction and impaired mitophagy in nucleus pulposus-derived mesenchymal stem cells (NPMSCs). Sirtuin 1 (SIRT1), a key regulator of mitochondrial quality control, is downregulated in degenerated discs. Salidroside (Sal), a natural compound from Rhodiola rosea, has shown potential in enhancing mitophagy, but its mechanism in IVDD remains unclear.

Methods

Using network pharmacology, molecular docking, and dynamics simulations, we identified SIRT1 as a key target of Sal. Human and rat NPMSCs were isolated and treated with tert-butyl hydroperoxide (TBHP) to induce degeneration. In vitro assays included CCK-8, EdU, SA-β-Gal, JC-1, Western blot, immunofluorescence, and transmission electron microscope (TEM). An in vivo rat IVDD model was established via needle puncture and treated with Sal and/or the autophagy inhibitor 3-MA. Histological, immunohistochemical, and imaging analyses were performed to evaluate IVDD.

Results

Sal bound stably to SIRT1 and activated SIRT1/FOXO3 pathway, promoting mitophagic flux, reducing reactive oxygen species accumulation, and suppressing apoptosis in NPMSCs. SIRT1 knockdown or 3-MA treatment abolished Sal’s protective effects. In vivo, Sal treatment preserved disc height, reduced apoptosis, and enhanced mitophagy, while 3-MA exacerbated degeneration.

Conclusions

Sal attenuates IVDD by activating SIRT1/FOXO3-mediated mitophagy, restoring mitochondrial homeostasis, and reducing NPMSCs apoptosis. These results suggest that the activation of the SIRT1/FOXO3-mitophagy axis may represent a potential therapeutic strategy for mitigating IVDD.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13287-026-05051-z.

Background

Intervertebral disc degeneration (IVDD) represents a primary underlying etiology of low back pain (LBP), a pervasive global health concern imposing substantial socioeconomic burdens through diminished productivity and escalating healthcare expenditures [1]. This progressive, multifactorial condition is characterized by a complex cascade of pathological alterations within the disc microenvironment. Key features include the progressive degradation of the extracellular matrix (ECM), notably loss of proteoglycans like aggrecan and the disorganization of collagen fibrils, coupled with aberrant cellular responses such as senescence, apoptosis, and a shift towards a pro-inflammatory phenotype [2]. Despite extensive research, the precise molecular mechanisms orchestrating the initiation and perpetuation of IVDD remain incompletely elucidated, particularly concerning the intricate interplay between mechanical stress, inflammatory cascades (e.g., Tumor Necrosis Factor-α (TNF-α) and Interleukin-1 β (IL-1β)), and abnormal autophagy process. Therefore, delving deeper into the fundamental pathobiological pathways driving IVDD is imperative for identifying novel therapeutic targets aimed at mitigating degeneration or promoting functional regeneration.

Mesenchymal stem cells (MSCs) can be harvested and expanded from various adult and perinatal tissues, including adipose tissue, bone marrow (BM), dental pulp, and umbilical cord (UC), exhibiting diverse pharmacological properties [3–5]. Currently, MSCs are widely investigated for therapeutic applications in IVDD; however, endogenous nucleus pulposus-derived mesenchymal stem cells (NPMSCs) offer distinct inherent advantages. Compared to exogenous MSCs, NPMSCs demonstrate superior adaptation to the unique intradiscal microenvironment, positioning them as critical targets for IVDD therapy [6, 7]. Consistent with our prior findings, NPMSCs reside within the nucleus pulposus (NP) tissue and possess intrinsic capacities for self-renewal and differentiation, thereby contributing to the mitigation of disc degeneration [8]. Notably, endogenous NPMSCs exhibit significant regenerative potential, primarily through differentiation into functional NP-like cells and inhibition of apoptosis [9]. Nevertheless, this functional capacity deteriorates with advancing age and IVDD progression, highlighting the necessity to preserve NPMSCs viability for effective endogenous repair strategies [10]. Our prior research confirmed the presence of endogenous NPMSCs within the NP tissue. Furthermore, we demonstrated that mitochondrial dysfunction is a hallmark of senescent NPMSCs, and crucially, that enhancing mitochondrial function attenuates oxidative stress-induced NPMSCs senescence. This mitochondrial restoration facilitates endogenous repair mechanisms, thereby mitigating IVDD [8, 11]. However, the precise molecular mechanisms underlying mitochondrial impairment in degenerated NPMSCs remain incompletely understood. In the present study, we therefore elucidate the specific mechanisms governing mitochondrial damage within degenerated NPMSCs, aiming to provide novel therapeutic strategies for delaying IVDD progression.

Macroautophagy/autophagy is a catabolic pathway that mediates the degradation and recycling of intracellular components. This process is essential for maintaining cellular homeostasis and promoting cell survival and function [12, 13]. Accumulating evidence indicates that dysregulation of the autophagic pathway contributes to IVDD [1, 14]. Autophagy is a dynamic process involving structural changes in subcellular membranes, which encapsulate cytoplasmic constituents to form autophagosomes. These autophagosomes subsequently fuse with lysosomes to form autolysosomes, where the encapsulated contents are degraded [13, 15, 16]. Autophagic flux, defined as the entire autophagic process, serves as a reliable indicator of genuine autophagic activity. Accumulating evidence substantiates a close correlation between autophagic flux and IVDD pathogenesis [17, 18].

Emerging evidence underscores the pivotal role of mitochondrial dysfunction in the pathogenesis of IVDD [19]. NPMSCs cells residing in an inherently hypoxic and nutrient-poor microenvironment, exhibit heightened susceptibility to mitochondrial impairment. This manifests as compromised oxidative phosphorylation (OXPHOS), elevated reactive oxygen species (ROS) production, and diminished ATP synthesis, collectively driving cellular senescence, apoptosis, and inflammatory activation [20]. Apoptosis, aging, excessive mechanical stress, and inflammatory cytokines can disrupt mitophagic flux, leading to the pathological accumulation of damaged mitochondria [21]. During this process, NPMSCs cells mediate the selective removal of defective or superfluous mitochondria via lysosomal degradation, a mechanism indispensable for sustaining mitochondrial quality [22]. Failure to promptly clear damaged mitochondria lead to the compromise of healthy organelles via ROS-induced ROS Release (RIRR). This self-perpetuating vicious cycle amplifies ROS signaling, culminating in irreversible cellular damage. Excess ROS further impairs cell proliferation and triggers lipid peroxidation, DNA damage, and apoptosis [23, 24]. Furthermore, the functional capacity of mitochondria is inextricably tied to their ultrastructural organization. Deviations in mitochondrial abundance, size, or configuration invariably result in functional compromise and the onset of disease phenotypes. This is exemplified by disturbances in the molecular machinery governing mitochondrial fusion and fission, which provoke pathological elongation of mitochondria and the disintegration of their cristae [25, 26]. Consequently, impaired mitophagy is now regarded not merely as a consequence but as a critical contributor to the progression of IVDD.

Lysosomes, as membrane-bound organelles, harbor diverse acidic hydrolases that ensure the efficiency of autophagic degradation processes [1, 27, 28]. The significance of lysosomal function and quality control is increasingly recognized in research exploring pathogenetic mechanisms and potential therapeutic strategies for various diseases. As a selective branch of macroautophagy, mitophagy executes targeted clearance of defective mitochondria via the autophagosome-lysosome machinery [29, 30]. Our investigation revealed that IVDD induces impaired autophagic flux and lysosomal dysfunction in NPMSCs both in vitro and in vivo. Consequently, restoration of mitophagic flux represents a novel therapeutic approach for IVDD.

Sirtuin 1 (SIRT1) is an NAD-dependent deacetylase and a key regulatory factor in cellular energy perception and stress response [12]. It plays a crucial role in maintaining mitochondrial function, promoting mitochondrial biosynthesis, and regulating the autophagy-lysosomal pathway [31]. As a member of the Forkhead box O (FOXO) transcription factor family, FOXO3 critically orchestrates cellular processes including apoptosis regulation, oxidative stress response, and cell cycle control [31, 32]. It is recognized that there are two classical molecular pathways of mitophagy: the non-receptor-mediated, canonical PINK-PRKN/PARKIN pathway and the receptor-mediated noncanonical pathway [12, 25, 33]. In addition, it has been reported that AMP-activated protein kinase (AMPK), an evolutionarily conserved metabolic energy sensor in eukaryotes, enhances SIRT1 activity by elevating cellular NAD+ levels [12]. Furthermore, AMPK can also coordinate the activation of downstream SIRT1 targets (including peroxisome proliferator-activated receptor -γ coactivator 1α and forkhead transcription factor FOXO3a) by stimulating their deacetylation, thereby crucially regulating mitochondrial homeostasis [34–36]. Our prior research demonstrated that circSPG21 can alleviates IVDD primarily through the miR-217/SIRT1 axis, which mitigates NPMSCs senescence and potentiates mitophagy [8]. Therefore, the integrity of mitophagic flux in degenerative NPMSCs cells can effectively remove the damaged mitochondria and alleviate IVDD, and the mechanism of mitophagic flux needs to be further elucidated.

Salidroside (Sal), a naturally occurring phenylethanoid glycoside isolated from the traditional medicinal plant Rhodiola rosea, demonstrates significant potential in conferring antioxidant protection, inhibiting apoptosis, attenuating cellular senescence, reducing hyperlipidemia, and modulating immune responses [37, 38]. Emerging evidence from Sal research highlights its promise in ameliorating IVDD, cardiomyopathy, diabetic nephropathy, and Alzheimer’s disease. Crucially, the underlying therapeutic action of Sal in these disparate disease contexts is attributed to its potent influence on mitophagy [37, 39–41]. However, its potential role in attenuating NPMSCs apoptosis and IVDD has not yet been investigated.

To elucidate the mechanism of Sal in IVDD, this study integrated network pharmacology, molecular dynamics simulations, and bioinformatic analyses for target prediction and pathway mapping, followed by systematic in vitro and in vivo validation in rat models. We demonstrated that Sal significantly ameliorates mitochondrial dysfunction and suppresses apoptosis in NPMSCs cells by activating the SIRT1-mediated mitophagic flux. Crucially, Sal exerted pronounced therapeutic effects in IVDD rats via the SIRT1/FOXO3 axis. These findings not only directly implicate SIRT1 in IVDD pathogenesis but also establish its potential as a novel druggable target for IVDD intervention.

Materials and methods

Data source

The drug-related targets of Sal were gathered from multiple databases, including the Comparative Toxicogenomics Database (CTD, https://ctdbase.org/), the Encyclopedia of Traditional Chinese Medicine (ETCM, https://www.tcmip.cn/ETCM/index.php/Home/), HERB (https://herb.ac.cn/), and SwissTargetPrediction (https://swisstargetprediction.ch/). The disease-related targets of IVDD were acquired from the CTD (accessed on February 21, 2025) and GeneCards (https://www.genecards.org/) (accessed on February 18, 2025) databases. Additionally, the transcriptomic dataset GSE34095, comprising gene expression profiles of human degenerative and non-degenerative intervertebral disc tissues, was retrieved from the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo).

Construction of protein–protein interaction (PPI) network

The protein-protein interaction (PPI) network was constructed using the STRING database (https://string-db.org). Common targets of Sal in IVDD were uploaded to STRING, with the species limited to Homo sapiens. The resulting TSV and PNG files were imported into Cytoscape (version 3.10.3) to generate a co-expression network of the common targets. Multiple topological attributes were calculated using the CytoNCA plugin, and nodes were ranked accordingly. The top 44 nodes were subsequently identified and visualized.

Gene ontology (GO), kyoto encyclopedia of genes and genomes (KEGG) functional analyses

KEGG pathway and GO enrichment analyses were performed using the DAVID 6.7 database (https://david.ncifcrf.gov/) with a significance threshold of p < 0.05 applied to both. For KEGG enrichment, the top 10 pathways were selected based on Gene Ratio in descending order. Similarly, the top 10 significantly enriched terms from each GO category, including Biological Process (BP), Cellular Component (CC), and Molecular Function (MF), were extracted for further analysis. Results were visualized as bubble plots and bar charts using a bioinformatics platform (http://www.bioinformatics.com.cn/).

Human NP tissue collection

Clinical samples of human NP tissue were obtained from patients undergoing surgery for lumbar spinal stenosis or lumbar disc herniation. Magnetic resonance imaging (MRI) scans were retrieved from patient records. IVDD severity was systematically evaluated for each patient based on MRI using the Pfirrmann grading system, and patients were subsequently categorized into four groups according to their IVDD grade. Human NPMSCs were then isolated from NP tissues exhibiting varying degenerative grades. The expression levels of SIRT1 and FOXO3 in these NPMSCs were compared using Western blotting analysis. Furthermore, quantitative reverse transcription polymerase chain reaction (RT-qPCR) was employed to analyze the correlation between SIRT1 and FOXO3 expression levels, respectively, and the degree of IVDD (Table 1). Ethical approval for this study was obtained from the Ethics Committee of Northern Jiangsu People’s Hospital. Written informed consent was acquired from all participants prior to their inclusion.

Table 1.

List of primers used in quantitative real-time polymerase chain reaction (RT-qPCR)

Gene	Primer sequence-forward	Primer sequence-reverse
SIRT1	TAGCCTTGTCAGATAAGGAAGGA	ACAGCTTCACAGTCAACTTTGT
FOXO3	CTGGGGGAACCTGTCCTATG	TCATTCTGAACGCGCATGAAG
GAPDH	GGAGCGAGATCCCTCCAAAAT	GGCTGTTGTCATACTTCTCATGG
BNIP3	TTCTCACTGTGACAGCCCAC	TCTTCCTCAGACAGAGTGCT
NIX	TGCAGTTGTTTCTGCTCCCA	CCACCCAGGAACCTTGTGAA
FUNDC1	ATCATGGCATCCCGGAACC	AATCCTGCACACCAGCCAG

Human NPMSCs cell isolation and culture

NP tissues were collected from patients representing different Pfirrmann grades and stored in Hank’s Balanced Salt Solution (HBSS) for transport. The tissues were then minced into fragments and used to seed and expand NPMSCs cells through culture in complete growth medium at 37 °C under 5% CO₂. Cells at passage 2 (P2) were utilized for subsequent experiments [8]. NPMSCs cells derived from Pfirrmann grade II tissues were designated as the healthy control group and employed to establish a degenerative cell model. Conversely, cells originating from Pfirrmann grade IV tissues were classified as the severely degenerative group and used to assess therapeutic outcomes. Patient information was included in Supplementary Table 1 (Table S1).

Rat NPMSCs cell Isolation and culture

Sprague-Dawley (SD) rats (3 months old) were purchased from Nanjing Changjing Biotechnology Co., Ltd. All animal procedures performed in this study were approved by the Animal Ethics Committee of Yangzhou University (Approval No. 202103427). NPMSCs were isolated as previously described. Briefly, NP tissues were aseptically dissected from the caudal intervertebral discs (IVDs) of SD rats. The tissues were then minced into small fragments and digested in complete growth medium supplemented with 0.2% type II collagenase (Gibco, USA) for 12 h at 37 °C in a humidified atmosphere containing 5% CO₂. Following digestion, the resulting cell suspension and any remaining undigested tissue fragments were washed twice with phosphate-buffered saline (PBS), filtered through a 75-µm cell strainer, and centrifuged at 1000 × g for 5 min. The pelleted cells were resuspended in complete MSC growth medium and cultured at 37 °C under 5% CO₂. The culture medium was replaced every 48 h. Upon reaching 80–90% confluence, cells were subcultured at a 1:3 ratio. Cells at passage 3 were cryopreserved for subsequent experiments [8, 42].

Cell viability assay

To establish an oxidative stress-induced senescence model in NPMSCs, the optimal concentration of tert-butyl hydroperoxide (TBHP) was determined by assessing cell viability using the Cell Counting Kit-8 (CCK-8) assay. NPMSCs were seeded in 96-well plates at a density of 2 × 10^4 cells per well and incubated overnight in complete culture medium at 37 °C under 5% CO₂. Subsequently, the cells were treated with varying concentrations of TBHP for the specified duration. Following treatment, CCK-8 reagent was added to each well according to the manufacturer’s instructions, and the plates were incubated for 1 h at 37 °C. Absorbance was then measured at a wavelength of 450 nm using a microplate reader (Bio-Rad, USA). Cell viability was calculated using the following formula: cell viability (of control) = [(Ae-Ab)/(Ac-Ab)]. Ae, Ab, and Ac represent the OD values of the TBHP treatment, blank and control groups, respectively. An autophagic inhibitor 3-methyladenine (3-MA) (Beyotime, China, 5µM, 12 h) was used to regulate the signaling pathway [31].

Cell transfection

Validated shRNA constructs targeting SIRT1 and corresponding control shRNA were procured from GENCEFE Biotech (Jiangsu, China). For transfection, NPMSCs were transfected with SIRT1-specific shRNA using Lipofectamine 3000 reagent (Thermo Fisher Scientific), strictly following the manufacturer’s protocol, when cells reached 50–60% confluency. After 12 h of transfection, the medium was replaced with fresh complete medium, and the NPMSCs were cultured for an additional 24 h before being harvested for further experiments.

Cell proliferation assay

A 5-ethynyl-2’-deoxyuridine (EdU) cell proliferation kit with Alexa Fluor 488 (Beyotime, China) was used to evaluate the cell proliferative capacity of the NPMSCs [43]. NPMSCs (5 × 10⁴ cells/well) were seeded in a 6-well plate and cultured in a 37 °C and 5% CO₂ incubator. When stably adhered to the plate, the cells in the different groups were treated according to the manufacturer’s instructions. After incubation with EdU for 2 h, the NPMSCs were fixed with 4% paraformaldehyde for 15 min at room temperature, washed three times (5 min each time) with PBS, permeabilized with 0.5% Triton X-100 for 15 min, and washed twice (5 min each time) with PBS. The cells were subsequently incubated for 30 min at room temperature in the dark with click additive solution. The cells were then washed three times (5 min each time) with PBS and incubated with Hoechst 33,342 in the dark at room temperature for 10 min. Six random microscopic fields were subsequently observed and paragraphed under a fluorescence microscope and analyzed via ImageJ software.

Senescence-associated β-galactosidase (SA-β-Gal) staining

SA-β-galactosidase (SA-β-Gal) activity in NPMSCs was assessed using a commercial kit (Beyotime, China) [44]. NPMSCs were seeded in 6-well plates (4 × 10⁴ cells/well) and incubated overnight at 37 °C under 5% CO₂. Following treatment, cells were processed according to the manufacturer’s protocol. Briefly, culture medium was aspirated, cells were washed twice with PBS (5 min/wash), fixed with SA-β-Gal fixative (15 min), and washed twice with PBS. SA-β-Gal staining solution was then applied, plates were sealed with paraffin film, and incubated overnight at 37 °C without CO₂. Six random fields per well were imaged using bright-field microscopy and quantitatively analyzed with ImageJ software.

Cell cycle assay

Cell cycle distribution of NPMSCs was quantified by flow cytometry using a Cell Cycle Detection Kit (Keygen Biotech, China) [45]. NPMSCs harvested from 6-well plates in each experimental group were processed according to the manufacturer’s protocol. Briefly, after PBS washing and centrifugation, cell pellets were resuspended at 1 × 10⁶ cells/mL and fixed overnight in 70% cold ethanol at 4 °C. The fixative was washed away with PBS, the staining solution was added, and the samples were incubated in the dark at room temperature for 60 min. Cell cycle analysis was subsequently performed on a BD FACSCanto II flow cytometer (BD Biosciences, USA).

JC‑1 assay for determining the mitochondrial membrane potential (MMP)

The MMP of NPMSCs across experimental groups was assessed using the JC-1 fluorescent probe (5,5’,6,6’-tetrachloro-1,1’,3,3’-tetraethylbenzimidazolcarbocyanine iodide; Beyotime, China) per manufacturer’s protocol [11, 46]. This cationic dye exhibits fluorescence shift characteristics: intact mitochondria in untreated NPMSCs emit red fluorescence (J-aggregates), whereas mitochondrial depolarization induces green monomeric fluorescence. Following medium aspiration, cells were washed with PBS and incubated with JC-1 working solution at 37 °C for 20 min. Subsequent to dual washes with JC-1 staining buffer, six randomly selected fields per well were imaged under fluorescence microscope. The green-to-red fluorescence intensity ratio was quantified using ImageJ software.

Western blot analysis

Western blot analysis was performed to determine protein expression levels across experimental groups [47]. Following experimental treatments, total cellular protein was extracted using the Whole Cell Lysis Assay Kit (Keygen Biotech, China) according to the manufacturer’s protocol and quantified via the Enhanced BCA Protein Assay Kit (Beyotime, China). Equal amounts of protein (30–40 µg per lane) from each group were resolved by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS‒PAGE). Subsequently, proteins were electrophoretically transferred onto polyvinylidene difluoride (PVDF) membranes under ice-cold conditions. Membranes were subsequently blocked with 5% skim milk for 2 h at room temperature. Following blocking, membranes were incubated overnight at 4 °C with primary antibodies targeting: GAPDH (1:5000; Affinity Biosciences, USA); SIRT1 (1:5000; abcam, USA); FOXO3 (1:5000; PROTEINTECH, China); P62 (1:5000; Zenbio, China); LC3B (1:5000; IMMUNOWAY, USA); Beclin-1 (1:5000; Yeasen Biotechnology, Shanghai); Bcl-2 (1:5000; ABclonal, China); Cyto C (1:5000; Zenbio, China); Bax (1:5000; ABclonal, China); Cleaved Caspase-3 (1:5000; IMMUNOWAY, USA); LAMP1 (1:5000; Beyotime, China). After primary antibody incubation, membranes were washed three times for 15 min each with 1×TBST under constant agitation at room temperature. Membranes were then probed for 1 h at room temperature with a horseradish peroxidase (HRP)-conjugated secondary antibody (1:5000; Abcam, UK) under agitation. Following secondary antibody incubation, membranes were washed an additional three times with 1×TBST. Finally, protein bands were visualized using an enhanced chemiluminescence (ECL) system, and relative target protein expression levels were quantified by densitometric analysis with ImageJ software.

Co-immunoprecipitation (Co-IP)

To detect the protein-protein interaction between SIRT1 and FOXO3, a co-immunoprecipitation (Co-IP) assay was performed in NPMSCs. Cells were lysed on ice for 10 min using IP lysis buffer (Beyotime, China) supplemented with protease and phosphatase inhibitors. The lysates were then centrifuged at 12,000 × g for 20 min at 4 °C to collect the supernatant. For each immunoprecipitation, 20 µL of the total lysate was reserved as the input control. The remaining lysate was incubated with 10 µL of Protein G magnetic beads (Thermo Fisher Scientific, USA) and 1 µL of either anti-SIRT1 (Cell Signaling Technology, USA) or anti-FOXO3 (Cell Signaling Technology, USA) antibody. This mixture was gently rotated overnight at 4 °C. Subsequently, the immunocomplexes were washed four times with NETN buffer (20 mM Tris, pH 8.0, 100 mM NaCl, 1 mM EDTA, and 0.5% NP-40). The precipitated proteins were then eluted, separated by SDS-PAGE, and finally analyzed by western blotting using the corresponding antibodies [32].

Flow cytometry

NPMSCs apoptosis was quantified using an Annexin V-FITC/PI apoptosis detection kit (Beyotime, China). Briefly, cells were seeded in six-well plates at a density of 1 × 10⁶ cells per well. Following the designated treatments, cells were harvested, washed, and resuspended in 195 µL of Annexin V-FITC binding buffer. The suspension was then incubated with 5 µL of Annexin V-FITC and 10 µL of propidium iodide (PI) solution for 15 min at room temperature in the dark. Flow cytometry analysis was performed within 20 min using a BD FACS Calibur flow cytometer (BD Biosciences, USA), and the percentages of apoptotic cells were determined [48].

Isolation of mitochondria from NPMSCs

Mitochondria were isolated from cultured cells beginning with cell homogenization: cells were harvested by trypsinization, washed with PBS, and pelleted by centrifugation at 800 × g for 5–10 min at 4 °C. Subsequently, 1–5 × 10⁷ cells were resuspended in 1–2.5 mL of ice-cold Reagent A, gently homogenized, and incubated on ice for 10–15 min with intermittent vortexing to ensure complete lysis. Cell rupture was verified by mixing an aliquot of the lysate 1:1 with trypan blue staining solution, with lysis deemed sufficient when ≥ 80% of cells were stained blue. Thereafter, the homogenate was subjected to differential centrifugation: it was first centrifuged at 1,000 × g for 10 min at 4 °C to pellet nuclei, large membrane debris, and unlysed cells, after which the supernatant was transferred and centrifuged again under the same conditions to remove residual debris. The resulting supernatant was then centrifuged at 12,000 × g for 10 min at 4 °C to pellet mitochondria; the supernatant containing cytosolic components was carefully aspirated, and the mitochondrial pellet was retained. For mitochondrial protein extraction, Reagent C (10 µL per mL) was first added to ice-cold Reagent B to prepare a lysis buffer, which was kept on ice before use. The mitochondrial pellet was resuspended in this buffer at a ratio of 100 µL per 10 µL of packed mitochondrial volume, incubated at 4 °C for 15–30 min with intermittent vortexing, and clarified by centrifugation at 10,000 × g for 15 min at 4 °C. Finally, the supernatant containing solubilized mitochondrial proteins was collected, quantified using a BCA assay, aliquoted, and stored at − 70 °C to avoid repeated freeze-thaw cycles.

Dual-luciferase reporter assay

Using Lipofectamine 2000 transfection reagent, NPMSCs were transfected with the respective plasmids for the reporter assay. Following transfection and treatment, cellular lysates were collected. Firefly and Renilla luciferase activity was evaluated using a Dual-Luciferase Reporter Assay Kit (Genepharma Biotech. Co. Ltd., Shanghai, China) and a microplate reader (Biotek Synergy, USA) [49].

Immunohistochemistry

Following deparaffinization, sections designated for immunohistochemistry were incubated with 3% H₂O₂ for 10 min and washed thrice in PBS. Sections were subsequently treated with 0.1% trypsin for 20 min at 37 °C, followed by three PBS washes. Blocking was performed using 1% (w/v) goat serum albumin at 37 °C for 1 h. Sections were then incubated overnight at 4 °C with primary antibodies against: SIRT1, MKI67 (Ki-67 proliferation marker), Bax, LC3II, P62, Cleaved Caspase-3 apoptosis. Negative control sections received isotype-matched non-specific IgG. After three additional PBS washes, sections were incubated with HRP-conjugated secondary antibodies at 37 °C for 1 h. For quantitative analysis, a minimum of three sections per specimen were evaluated. Positive cell ratios were determined by blinded observers unaware of experimental groupings, using standardized counting protocols.

Immunofluorescence

Cells grown on coverslips were washed thrice with phosphate-buffered saline (PBS), fixed with 4% paraformaldehyde for 15 min, and permeabilized using 0.5% (v/v) Triton X-100 for 20 min. Non-specific binding sites were blocked with 1% (w/v) goat serum albumin at 37°C for 1 h. Samples were subsequently incubated overnight at 4°C with primary antibodies targeting: SIRT1, Parkin, LC3-II, Tom20, Bax, LAMP1(lysosomal-associated membrane protein 1). After PBS washes, samples were incubated with fluorophore-conjugated secondary antibodies for 1 h at room temperature. Nuclei were counterstained with 0.1 µg/mL 4’,6-diamidino-2-phenylindole (DAPI) for 5 min, followed by final washes. Fluorescence images were acquired using an inverted microscope. Protein expression levels were quantified by measuring integrated optical density (IOD) via ImageJ software with standardized threshold settings.

TUNEL assay

The terminal deoxynucleotidyl transferase (TdT) dUTP nick-end labeling (TUNEL) assay was employed to quantify apoptotic DNA fragmentation [50]. NPMSCs cells were fixed with freshly prepared 4% paraformaldehyde for 1 h, followed by incubation with 3% H₂O₂ and 0.1% Triton X-100 for 10 min. Three PBS washes were performed between each procedural step. According to the manufacturer’s protocol, cells were stained using an In Situ Cell Death Detection Kit (Beyotime, China) with DAPI nuclear counterstaining. Fluorescence microscopy images were acquired from three randomly selected fields per slide.

Reactive oxygen species assay

Cells were seeded in 6-well plates containing complete growth medium. After 16 h of incubation (reaching 90% confluence), cultures were washed twice with PBS. Subsequently, cells were incubated with the ROS detection probe mixture according to the manufacturer’s protocol (Beyotime, China). Fluorescence images were acquired from three randomly selected fields per well using an inverted fluorescence microscope.

Measurement of mitochondrial ROS

Mitochondrial reactive oxygen species (mtROS) were quantified using MitoSOX™ Red staining (Beyotime, China). NPMSCs were seeded in 6-well plates with complete growth medium. Following treatment with or without Sal under pathological conditions induced by TBHP cells were incubated with 5 µM MitoSOX™ Red at 37 °C for 30 min in the dark. After three washes with PBS, mtROS fluorescence was immediately visualized using an inverted fluorescence microscope with excitation/emission at 550/580 nm.

Mito-tracker red staining

Mitochondria in live cells were labeled using MitoTracker^™ Deep Red FM (Yeasen Biotechnology, Shanghai, China) [51]. NPMSCs were incubated with 50 nM MitoTracker™ probe at 37 °C for 30 min in the dark. Subsequently, nuclei were counterstained with DAPI for 10 min at 37 °C. Following three washes with PBS, fluorescence images were acquired using a confocal microscope. Mitochondrial fluorescence intensity was quantified in ≥ 5 random fields per sample using ImageJ software with background subtraction.

Rat IVDD model induction

A total of 24 adult male Sprague-Dawley (SD) rats (200–250 g) were purchased from Nanjing Changjing Biotechnology Co., Ltd. (Nanjing, China). These rats were utilized for cell extraction and subsequent animal experiments. Animals were housed under standard conditions with controlled temperature, humidity, and a 12-hour light/dark cycle. Rats were randomly divided into five groups (n = 6 per group): Control group, IVDD group, Salidroside group, IVDD + Salidroside group, IVDD + Salidroside + 3-MA group. Briefly, rats in all groups were anesthetized by intraperitoneal injection of pentobarbital sodium (0.1 mg/kg) and positioned prone. The target intervertebral disc (Co6-7) was identified, marked, and disinfected. Under aseptic conditions, percutaneous puncture of the target disc was performed using a 21-gauge needle (puncture depth: 5 mm; needle rotated 180° after insertion and maintained in situ for 5 s) [52]. Two weeks post-puncture, rats in the IVDD + Salidroside group received immediate intraperitoneal injection of salidroside (20 mg/kg), and rats in the Salidroside group were administered salidroside (20 mg/kg, intraperitoneal injection). Rats in the IVDD + Salidroside + 3-MA group were administered salidroside (20 mg/kg, intraperitoneal) and 3-MA (20 mg/kg, oral gavage). The remaining rats received equal volumes of physiological saline via intraperitoneal injection and oral gavage every two days until sacrifice. Treatments were administered at 48-hour intervals. All animals were monitored daily and maintained under specific pathogen-free (SPF) conditions with unrestricted weight-bearing and activity.

Imaging evaluation and histological analysis

At 6- and 10-weeks post-modeling success, rats were anesthetized with pentobarbital sodium followed by radiographic (X-ray) and MRI of the caudal spine. The disc height index (DHI) and Pfirrmann grading of IVDD were calculated according to previously established methods [53]. After the final imaging session, rats were euthanized with an overdose of pentobarbital sodium, and caudal segments were harvested. Specimens were fixed in 10% neutral buffered formalin for 48 h, decalcified in EDTA-based solution for 4 w, dehydrated through a graded ethanol series, and embedded in paraffin. Sections of 5 μm thickness were prepared from each disc. Serial slides were stained with Safranin O-Fast Green (S-O) and hematoxylin and eosin (H&E). Histological examination of cellular architecture and morphology in the NPMSCs and annulus fibrosus (AF) was performed independently by a panel of experienced researchers using light microscopy under blinded conditions. Tissue sections were evaluated using a semi-quantitative grading scale, with histological scores determined based on characteristic structural alterations in the NPMSCs and AF regions.

Statistical analysis

Data are presented as mean ± standard deviation (SD). Statistical analyses were performed using GraphPad Prism 8 (GraphPad Software, La Jolla, CA, USA). Differences between two groups were assessed by Student’s t-test, while comparisons among multiple independent groups used one-way analysis of variance (ANOVA). The Kruskal-Wallis H test was applied to analyze histological scores. A threshold of P < 0.05 defined statistical significance.

The work has been reported in line with the ARRIVE guidelines 2.0.

Results

Collection of sal and IVDD associated targets

To identify relevant genetic targets associated with IVDD, we utilized the CTD (score ≥ 15) and the GeneCards database (score ≥ 20). Following initial screening, 13,716 potential target genes were retrieved, comprising 12,763 from CTD and 953 from GeneCards. Subsequent integration of these datasets and removal of duplicates yielded 13,061 unique IVDD-associated target genes. To enhance the robustness of our findings, we employed bioinformatics analysis to identify differentially expressed genes (DEGs) between normal and degenerated NP tissues using the microarray dataset GSE34095 from the GEO database. Comparative analysis of GSE34095 with the IVDD-related target genes revealed 7,803 overlapping genes (Fig. 1A). Differential expression gene (DEG) analysis of these 7,803 overlapping genes identified 563 significantly down-regulated and 547 significantly up-regulated DEGs, while the remaining 6,693 genes showed no statistically significant changes (Fig. 1B). Based on the GSE34095 dataset, the top 50 ranked genes were further analyzed. Heatmap visualization demonstrated strong correlations between these 50 target genes and IVDD pathogenesis (Fig. 1C). To identify potential targets of Sal, we conducted systematic searches across four distinct databases: the CTD (147 targets), the ETCM (30 targets), the HERB database (182 targets), and SwissTargetPrediction (100 targets) (p > 0). Following the removal of duplicates, a total of 376 unique Sal-associated targets were compiled. By intersecting the consolidated pool of Sal-associated targets, IVDD-related genes, and DEGs, 86 shared targets were ultimately identified (Fig. 1D).

Fig. 1.

Collection of Sal and IVDD Associated Targets. A Targets from disease-associated databases (CTD and GeneCards) overlap with significantly DEGs identified in the GSE34095 dataset. B Volcano plot and C heatmap of DEGs in IVDD samples. D The shared genes of Sal, IVDD, and DEGs. E, F A PPI network based on the STRING database. G Molecular docking analysis of Sal with the six core candidate targets

Subsequently, the 86 overlapping targets were imported into the STRING database to generate a PPI network (Fig. 1E), which was further analyzed and visualized using Cytoscape (Fig. 1F). Visualization of the PPI network for the top 44 DEGs in Cytoscape led to the identification of six core candidate genes demonstrating the highest significance: TNF, SIRT1, IL1B, FN1, IL6, and AKT1 (Fig. 1F). Molecular docking simulations were then performed between Sal and each of these six core candidates. Among them, Sal exhibited the strongest binding potential to SIRT1, with a binding energy (BE) of -7.0 kcal/mol (Fig. 1G). The superior binding affinity of Sal for SIRT1 was further corroborated by molecular dynamics simulations, reinforcing the reliability of this finding. This aligned with and extends our prior study, which highlighted the prominent role of SIRT1 in IVDD and its critical involvement in the pathogenesis and progression of this condition.

Molecular dynamics simulations confirm the stable binding of sal to SIRT1

Subsequently, molecular dynamics (MD) simulations were performed to validate the binding interaction between Sal and SIRT1. The complex was subjected to a 100 ns MD simulation using Gromacs 2022. The protein was described with the CHARMM36 force field [54], while the ligand topology was generated using parameters from the GAFF2 force field. Periodic boundary conditions were applied by placing the protein-ligand complex in a cubic box, which was solvated with TIP3P water molecules and maintained with a minimum distance of 1.2 nm from the box boundaries [55]. Electrostatic interactions were treated using the Particle Mesh Ewald (PME) method, and the Verlet al.gorithm was employed for integration. The system was energy-minimized and equilibrated under NVT and NPT ensembles for 100 ps each, with a coupling constant of 0.1 ps. Van der Waals and Coulomb interactions were calculated with a 1.0 nm cutoff. Finally, a production MD simulation was conducted for 100 ns under constant temperature (310 K) and pressure (1 bar) using Gromacs 2022. The root mean square deviation (RMSD) was used to evaluate the conformational stability of the protein and ligand by measuring the deviation of atomic positions from their initial coordinates. Lower RMSD values indicate greater structural stability. As shown in Fig. 2A, the RMSD of the complex fluctuated stably between 5 and 55 ns and exhibited a slight increase after 55 ns, yet remained below 4.2 Å throughout, indicating high stability of the ligand-protein complex. Further analysis revealed mild fluctuations in the radius of gyration (Rg) and solvent accessible surface area (SASA), suggesting conformational adjustments during the simulation. Hydrogen bonds play a crucial role in ligand-protein binding. As depicted in Fig. 2A, the number of hydrogen bonds between Sal and SIRT1 ranged from 0 to 6, with an average of approximately 3, indicating consistent and favorable hydrogen bonding interactions. The root mean square fluctuation (RMSF) was used to assess residue flexibility. Most residues showed low RMSF values (below 2.7 Å), reflecting limited flexibility and high stability of the complex (Fig. 2A). The free energy landscape (FEL) was constructed to visualize conformational stability and ligand-protein interactions. Energy minima correspond to stable states, while maxima represent energy barriers for conformational transitions. This approach helps predict ligand binding affinity and elucidate molecular recognition mechanisms. In the lowest energy conformation of the Sal-SIRT1 complex (Fig. 2B, C), van der Waals interactions were formed with PRO447, TYR343, PHE422, LEU418, ILE360, ALA425, VAL412, HIS262, GLY364, ARG341, and ARG424. A carbon-hydrogen bond was formed with GLN361, and a conventional hydrogen bond was observed with CYS362.

Fig. 2.

Evaluation of the binding stability and interactions of the Sal-SIRT1 complex via MD simulations. A Time-dependent evolution of RMSD, Rg, SASA, hydrogen bond number, and RMSF values throughout the 100 ns production run. B, C FEL and the lowest-energy binding mode identifying key residual interactions. The simulation confirms the formation of a stable complex supported by consistent hydrogen bonding

In summary, the Sal-SIRT1 complex exhibited stable binding, supported by consistent hydrogen bonding and low conformational fluctuation, indicating strong ligand-protein interactions.

Functional enrichment analysis results

KEGG enrichment analysis identified 148 significantly enriched signaling pathways (P < 0.05). The top 10 pathways ranked by P-value were visualized in a bubble plot. Among these, Mitophagy - animal, MAPK signaling pathway, FOXO signaling pathway, and Th17 cell differentiation demonstrated the strongest relevance to IVDD pathogenesis (Fig. 3A, B).

Fig. 3.

Functional enrichment analysis of the 86 overlapping target genes. A, B KEGG pathway enrichment analysis. The top 10 significantly enriched pathways are shown. C–E GO enrichment analysis across BP, CC, and MF categories. The top 10 terms for each category are displayed

GO and KEGG pathway enrichment analyses were performed on the 86 target genes using the bioinformatics platform (https://bioinformatics.com.cn/). GO enrichment analysis revealed the following significantly enriched terms (P < 0.05): BP: 385 terms, predominantly involving positive/negative regulation of apoptotic process, apoptotic process, and signal transduction. CC: 37 terms, primarily encompassing glutamatergic synapse, mitochondrion, cytoplasm, and protein-containing complex. MF: 76 terms, chiefly including identical protein binding, protein kinase binding, and cytokine activity. The top 10 most significantly enriched terms (ranked by P-value) within each GO category (BP, CC, MF) were selected for visualization using bubble plots (Fig. 3C–E). Integrating the GO and KEGG enrichment findings, we postulate that Sal may mitigate the progression of IVDD by activating SIRT1-mediated mitophagy.

Downregulation of SIRT1 in Degenerative Human NP tissue and TBHP stimulated NPMSCs

Mitochondrial impairment represents a key pathogenic factor triggering apoptosis and senescence in NPMSCs, thereby contributing to IVDD. To elucidate the association between SIRT1 and IVDD, human NP tissues were surgically collected and stratified according to Pfirrmann grade. NPMSCs were subsequently isolated and cultured from these tissues (Fig. 4A). Quantitative assessment of SIRT1 and FOXO3 protein expression in human NP tissues via Western blotting revealed a significant downregulation of SIRT1 concomitant with advancing degenerative stages (Figs. S1,  4B, C). We next evaluated markers of cellular proliferation and apoptosis within corresponding NP tissues. Immunohistochemical (IHC) analysis demonstrated significantly reduced expression of the proliferation marker MKI67 (Ki-67) alongside elevated levels of the apoptotic protein Bax in degenerative NP tissues (Fig. 4D). Consistent with Western blot findings, IHC further confirmed diminished SIRT1 protein expression and positive staining intensity in degenerative specimens (Fig. 4D). Critically, linear regression analysis of immunostaining optical density revealed a significant negative correlation between SIRT1 and Bax levels, whereas a positive correlation was observed between SIRT1 and MKI67 expression (Fig. 4E). Immunofluorescence analysis of NPMSCs isolated across Pfirrmann grades (II–V) consistently showed a marked reduction in SIRT1 expression and a concomitant increase in cellular senescence, as indicated by elevated SA‑β‑gal‑positive cells, in samples from more severely degenerated discs (Fig. 4F–G). Collectively, these results indicated that SIRT1 is aberrantly downregulated in degenerated human NP tissues. This suppression is robustly associated with enhanced NPMSCs apoptosis, accelerated senescence, and progressive disc degeneration. Subsequently, NPMSCs isolated from healthy human NP tissues were cultured and subjected to an in vitro IVDD model by treatment with TBHP (100 µM). Western blot analysis revealed a significant downregulation in the protein expression levels of both SIRT1 and FOXO3 in TBHP-treated NPMSCs (Fig. 4H, I). The observed reduction in SIRT1 and FOXO3 protein expression in TBHP-treated cells (100 µM) phenocopied the profile seen in naturally degenerated NPMSCs, providing preliminary evidence to support the successful establishment of an in vitro degeneration model. Furthermore, RT-qPCR analysis of SIRT1 and FOXO3 mRNA levels in both healthy control NPMSCs and IVDD-derived NPMSCs confirmed significantly decreased transcript levels of both genes in the IVDD group compared to controls (Fig. 4J). Collectively, these findings established that SIRT1 deficiency as a defining pathological hallmark of IVDD.

Fig. 4.

Downregulation of SIRT1 in Degenerative Human NP Tissue and TBHP Stimulated NPMSCs. A Representative MRI images at T2 weight sequence were evaluated by Pfirrmann grading system. II: grade II, III: grade III, IV: grade IV, V: grade V (scale bar, 100 μm). B–D Western blot and immunohistochemical analyses showing downregulation of SIRT1 and FOXO3, decreased proliferation (MKI67), and increased apoptosis (Bax) in human NP tissues with advancing degeneration (scale bar, 50 μm). E Correlation analysis between SIRT1 expression and apoptotic/proliferative markers. F, G Immunofluorescence and SA-β-gal staining confirmed progressively reduced SIRT1 expression and increased cellular senescence in NPMSCs across advancing Pfirrmann grades (scale bar, 100 μm). H–J SIRT1 and FOXO3 are downregulated at the protein level in the TBHP-induced IVDD model and at the mRNA level in severely degenerated human NPMSCs. Full-length blots/gels are presented in supporting materials. The data are expressed as the mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 indicates statistical significance, n = 4

TBHP induces mitochondrial ROS production, apoptosis, and autophagic activation in NPMSCs

The CCK-8 assay results (Fig. 5A) revealed a dose- and time-dependent decline in cell viability with increasing concentrations of TBHP, a finding further validated by multi-timepoint flow cytometry analysis of apoptosis (Fig. S2A, B). This kinetic assessment demonstrated a progressive increase in apoptotic cells following TBHP exposure, peaking at 24 h, and fitting of the apoptosis-progression curve confirmed that Sal significantly suppressed early apoptosis and improved cell survival, with a calculated inhibition rate constant (k) underscoring its potent cytoprotective effect (Fig. S2C). A significant reduction in cell viability was observed at 24 h with the 100 µM concentration, which was thus selected as the optimal concentration for inducing cellular injury in subsequent experiments. Results showed that TBHP stimulation of NPMSCs significantly increased intracellular mitochondrial ROS (Fig. 5B, I). TUNEL assay further confirmed that TBHP promoted apoptosis and reduced the proliferative capacity of NPMSCs (Fig. 5C, G). Following TBHP treatment, the percentage of EdU-positive cells was markedly decreased, reflecting diminished cell proliferation in the apoptosis model (Fig. 5E, J). Concomitantly, a significant increase in SA-β-Gal-positive cells was observed, corroborating the model’s capacity to induce cellular senescence (Fig. 5E, K). Transmission electron microscopy (TEM) consistently revealed elongated tubular mitochondrial networks with densely packed cristae in the control group, whereas TBHP-treated NPMSCs exhibited numerous autophagic vacuoles, mitochondrial swelling with rounded morphology, and disrupted cristae integrity (Fig. 5F). Western blot analysis demonstrated a significant upregulation of pro-apoptotic markers (c-caspase 3 and Bax) and downregulation of the anti-apoptotic protein Bcl-2 in NPMSCs after TBHP stimulation (Fig. 5N, O). Cytochrome C (Cyto C), a hemoglobin primarily localized in mitochondria, was significantly increased in the cytoplasm of TBHP-stimulated NPMSCs, indicating disruption of mitochondrial structure (Fig. 5N, O) [56]. Combined Western blot and immunofluorescence analyses revealed that TBHP not only promoted apoptosis but also activated autophagy in NPMSCs (Fig. 5D, H, L, M). Notably, elevated expression of LC3-II and Beclin-1 was accompanied by increased p62 levels following TBHP stimulation, suggesting that while mitochondrial autophagy was initiated, autophagic flux was impaired (Fig. 5L, M). Collectively, these findings validated the successful establishment of a TBHP-induced apoptotic model. NPMSCs subjected to TBHP stimulation exhibited mitochondrial dysfunction, elevated ROS production, enhanced apoptosis, and activated autophagy.

Fig. 5.

TBHP induces oxidative stress, mitochondrial dysfunction, and apoptosis in NPMSCs. A CCK-8 assay showing the dose- and time-dependent cytotoxic effect of TBHP. TBHP stimulation increased Mito-SOX (scale bar, 100 μm) and mitochondrial ROS (scale bar, 250 μm) production (B, I), apoptosis (TUNEL) (scale bar, 100 μm) (C, G), and senescence (SA-β-gal), while inhibiting proliferation (EdU) (E, J, K) (scale bar, 250 μm). F TEM images revealing ultrastructural mitochondrial damage and autophagic vacuoles (scale bar, 500 nm). Western blot (L–O) and immunofluorescence (D, H) (scale bar, 100 μm) analyses confirmed the activation of apoptosis and autophagy, alongside cytochrome c release. Full-length blots/gels are presented in supporting materials. The data are expressed as the mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 indicates statistical significance, n = 4

SIRT1 knockout potentiates TBHP-induced apoptosis in rat NPMSCs

Given the observed downregulation of SIRT1 expression in TBHP-stimulated NPMSCs and degenerated NP tissue, we sought to determine whether this reduction correlates with increased apoptosis and elevated ROS levels in TBHP-stimulated NPMSCs. As illustrated in Fig. 6A, SIRT1 expression was successfully inhibited. Western blot analysis revealed a significant increase in the expression of apoptosis-related proteins in the TBHP group compared to control group, concomitant with reduced expression of FOXO3 and the anti-apoptotic protein Bcl-2. Furthermore, upon SIRT1 knockdown using Sh-RNA, the expression of these apoptosis-related proteins in TBHP-stimulated cells was further augmented (Fig. 6B, D). Concurrently, Mito-Tracker Red fluorescence intensity exhibited a significant decrease in the TBHP group relative to controls. This reduction was further exacerbated by Sh-RNA treatment (Fig. 6C, E), suggesting that SIRT1 knockdown may potentiate TBHP-induced mitochondrial dysfunction. Assessment of ROS generation demonstrated that TBHP-treated NPMSCs exhibited increased ROS production, which was further intensified in the presence of Sh-SIRT1 under TBHP stimulation (Fig. 6C, F). MMP assays corroborated the critical role of SIRT1 in maintaining mitochondrial polarization. Both the TBHP group and the TBHP + Sh-SIRT1 group displayed progressively exacerbated mitochondrial depolarization compared to controls (Fig. 6G, H). Collectively, these findings indicate that SIRT1 likely confers a protective effect on NPMSCs survival, potentially by alleviating TBHP-induced mitochondrial dysfunction.

Fig. 6.

SIRT1 knockdown exacerbates TBHP-induced mitochondrial dysfunction and apoptosis in NPMSCs. A Validation of SIRT1 knockdown efficiency. B, D Western blot analysis showing that SIRT1 deficiency further increases the expression of apoptosis-related proteins and decreases FOXO3 and Bcl-2 under TBHP stimulation. C, E–H Functional assays demonstrate that SIRT1 knockdown potentiates TBHP-induced mitochondrial damage, as evidenced by loss of mitochondrial mass (MitoTracker Red) (scale bar, 100 μm), increased ROS production (scale bar, 250 μm), and loss of MMP (scale bar, 250 μm). I Isolation and grouping of rat NPMSCs. Full-length blots/gels are presented in supporting materials (Created in BioRender). The data are expressed as the mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 indicates statistical significance, n = 4

Sal upregulates SIRT1 expression and inhibits apoptosis via enhanced mitophagy in TBHP-stimulated NPMSCs

Previous studies have demonstrated that Sal can modulate microglial polarization via autophagy regulation in spinal cord injury [57]. Therefore, we sought to investigate whether Sal could enhance the survival of NPMSCs by eliminating damaged mitochondria through SIRT1-induced mitophagy. To determine the optimal concentration and treatment duration of Sal for NPMSCs, cell viability was assessed using the CCK-8 assay. The results indicated a gradual decline in viability starting at 12, 24, and 48 h post-treatment at a concentration of 100 µM. Considering overall experimental time efficiency and cost-effectiveness, a concentration of 75 µM with a 12-h treatment duration was selected as optimal for subsequent experiments (Fig. 7A). The aforementioned results demonstrated the critical role of SIRT1 in mitigating mitochondrial dysfunction and NPMSCs apoptosis (Fig. 6). However, whether Sal regulates mitophagy via SIRT1 remained undetermined. To elucidate the mechanism underlying salidroside’s effects on mitophagy, SIRT1 expression was knocked down using ShRNA. Following Sh-SIRT1 pretreatment, levels of LC3-I (an essential component for autophagosome initiation in NPMSCs) were significantly reduced. Concurrently, expression of LC3-II, an indicator of autophagosome formation, markedly decreased. SIRT1 inhibition also suppressed Beclin-1 expression in NPMSCs. Furthermore, the Sh-SIRT1 group exhibited substantially increased p62 accumulation, indicating impaired autophagic flux. Simultaneously, FOXO3 expression was significantly downregulated, attenuating its deacetylation and substantially impairing its transcriptional activity (Fig. 7B, D). TEM revealed dilated mitochondrial matrix compartments, disorganized and sparse cristae progressing to complete disintegration, and vacuole formation in severely damaged mitochondria within TBHP-stimulated NPMSCs. Notably, substantial accumulation of impaired mitochondria was observed. In contrast, TEM analysis of Sal-treated NPMSCs demonstrated structurally intact mitochondria exhibiting elliptical or rod-shaped morphologies with well-defined cristae (Fig. 7C). Integrated with immunoblotting data, these findings demonstrate that Sal effectively enhances autophagy and promotes degradation of damaged mitochondria in TBHP-stimulated NPMSCs. This protective mechanism, however, was significantly attenuated when SIRT1 is inhibited (Fig. 7C). The translocation of cytosolic Parkin to defective mitochondria represents a pivotal event in mitophagy initiation [58, 59]. Consistent with this, we observed Parkin accumulation on damaged mitochondria within TBHP stimulated NPMSCs. Notably, Sal treated NPMSCs exhibited markedly elevated Parkin aggregation (Fig. 7E). Following SIRT1 inhibition, TBHP stimulated NPMSCs displayed a rapid reduction in mitochondrial Parkin accumulation (Fig. 7E). To determine whether Sal attenuates TBHP induced apoptosis in NPMSCs through SIRT1 upregulation, we performed immunoblot analysis. Notably, Sh-SIRT1 transfection in Sal treated NPMSCs substantially decreased the anti-apoptotic protein (Bcl-2) while elevating pro-apoptotic proteins (Bax, Cyto c, and cleaved caspase-3) (Fig. 7F, G). Consequently, Sh- SIRT1 effectively reversed the Sal mediated reduction of apoptotic cells in TBHP stimulated NPMSCs (Fig. 7F, G). In summary, these findings established SIRT1 as a master regulator in NPMSCs that orchestrates Sal triggered clearance of damaged mitochondria through mitophagy while concurrently suppressing cellular apoptosis.

Fig. 7.

Sal activates SIRT1-dependent mitophagy to clear damaged mitochondria and attenuate apoptosis. A CCK-8 assay determining the optimal concentration and treatment duration of Sal for NPMSCs. B, D Immunoblot analysis showing that SIRT1 knockdown abrogates Sal-induced mitophagy activation and FOXO3 deacetylation. C TEM images demonstrating that Sal rescues TBHP-induced mitochondrial damage, an effect blocked by SIRT1 inhibition (scale bar, 500 nm). E Immunofluorescence confirming that Sal promotes Parkin translocation to mitochondria in a SIRT1-dependent manner (scale bar, 25 μm). F, G SIRT1 knockdown reverses the anti-apoptotic effects of Sal, as evidenced by altered levels of Bcl-2, Bax, Cyto c, and CC3. Full-length blots/gels are presented in supporting materials. The data are expressed as the mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 indicates statistical significance, n = 4

SIRT1-specific inhibition and FOXO 3 transcriptional regulation validate the linear pathway

To ascertain the necessity of SIRT1 enzymatic activity for the observed effects, we employed the specific SIRT1 inhibitor EX-527. Immunofluorescence analysis indicated that Sal significantly promoted TBHP-induced nuclear translocation of FOXO3. Notably, this promotive effect was effectively reversed by co-treatment with the SIRT1-specific inhibitor EX-527 (Fig. S3A, B). Furthermore, co-immunoprecipitation assays demonstrated that Sal promoted the physical interaction between SIRT1 and FOXO 3, leading to a marked decrease in FOXO 3 acetylation. Conversely, pharmacological inhibition of SIRT1 by EX-527 abolished this interaction and prevented FOXO 3 deacetylation (Fig. S3C, D). These data collectively indicate that Sal facilitates the SIRT1-mediated deacetylation and subsequent nuclear activation of FOXO 3, establishing a direct functional link upstream of mitophagic regulation. To evaluate the regulation of mitophagic flux by Sal, we monitored the degradation of the selective autophagy substrate p62. TBHP stimulation induced mitophagic initiation but resulted in pronounced p62 accumulation, indicating impaired autophagic flux and insufficient clearance of autophagic cargo. Treatment with Sal not only enhanced mitophagy but also restored flux efficiency, as shown by a significant reduction in p62 levels relative to the TBHP group. Notably, this restorative effect on autophagic flux was abolished by co‑treatment with the SIRT1‑specific inhibitor EX‑527 (Fig. S3E, F). These findings demonstrate that Sal promotes complete mitophagic progression in a SIRT1‑dependent manner, thereby ensuring the effective removal of damaged mitochondria. To further validate the transcriptional role of FOXO 3 in mitophagic initiation, luciferase reporter assays were conducted using constructs containing the promoter regions of canonical mitophagy receptor genes, including BNIP3, NIX, and FUNDC1. These assays confirmed that FOXO 3 transactivates these promoters, demonstrating its direct role in initiating the transcriptional program required for mitophagy (Fig. S4A, B). Furthermore, the ChIP assay validated the targeted regulation relationship between FOXO3 and the promoters of BNIP3, NIX, and FUNDC1 (Fig. S4C).

To further substantiate the functional consequences of FOXO3 transcriptional regulation and to provide direct evidence that Sal promotes selective mitophagy, we isolated purified mitochondrial fractions from NPMSCs under key experimental conditions. Immunoblot analysis of these mitochondrial extracts revealed a highly dynamic and well-defined process. In TBHP-stimulated cells, elevated levels of both mitochondrial-associated LC3-II and p62 were observed, indicative of concurrent mitophagic initiation coupled with subsequent flux impairment. Notably, Sal co-treatment significantly enhanced the co-accumulation of LC3-II and p62 on mitochondria, which serves as a clear signature of enhanced cargo recognition and specific autophagosomal targeting to the mitochondrial compartment. This Sal-induced enhancement was entirely dependent on SIRT1 activity and functional autophagic flux, as both Sh-SIRT1 and autophagy inhibition with 3-MA markedly reduced mitochondrial levels of LC3-II and p62, thereby completely abolishing the effect. Crucially, the purity of the mitochondrial fraction was verified by the absence of cytosolic (GAPDH) and lysosomal (LAMP1) markers, while consistent VDAC1 levels confirmed equal loading. These results provide direct biochemical evidence that Sal specifically amplifies the mitochondrial targeting of mitophagy via the SIRT1 pathway. This finding consolidates and mechanistically reinforces our earlier observations from whole-cell assays, which demonstrated enhanced LC3–mitochondria co‑localization and restored autophagic flux (Fig. S5). To further substantiate the linearity of the SIRT1 and FOXO3 axis, we performed a gain of function analysis coupled with genetic ablation of FOXO3. Overexpression of SIRT1 in wild type (WT) NPMSCs robustly enhanced the mRNA expression of the canonical mitophagy receptor genes BNIP3, NIX, and FUNDC1, which are established transcriptional targets of FOXO3 (Fig. S6C). In stark contrast, this SIRT1 driven transcriptional upregulation was completely abolished in FOXO3 knockout cells (Fig. S6C). At the protein level, SIRT1 overexpression in wild type cells potently activated mitophagic flux, evidenced by a marked increase in LC3 II levels coupled with a significant decrease in p62 accumulation (Fig. S6A, B). However, in the absence of FOXO3, SIRT1 overexpression failed to elicit these protective effects. Instead, FOXO3 knockout cells exhibited a substantial increase in apoptosis, indicated by elevated levels of CC3, irrespective of SIRT1 status (Fig. S6A, B). Collectively, these genetic loss of function experiments demonstrate that the ability of SIRT1 to promote mitophagy and confer cytoprotection is strictly dependent on FOXO3. This finding provides direct mechanistic evidence that the pathway involving Sal, SIRT1, and FOXO3 operates as a requisite linear signaling cascade rather than a bifurcating or parallel network.

Sal activates SIRT1-dependent mitophagic flux to promote mitochondrial homeostasis in NPMSCs

Clearance of depolarized or dysfunctional mitochondria mediated by mitophagy in TBHP stimulated NPMSCs was assessed through autophagosome co-localization analysis. Autophagosomes were labeled using LC3 immunofluorescence, while mitochondria were identified with TOM20, an outer mitochondrial membrane marker. Results revealed increased formation of LC3 positive autophagosomes (green) demonstrating enhanced co-localization with mitochondria (red) in Sal treated samples (Fig. 8A). Conversely, SIRT1 knockdown via ShRNA substantially reduced LC3 co-localization with mitochondria (Fig. 8A). Furthermore, immunofluorescence staining revealed that Sal treatment mitigated TBHP stimulated upregulation of Bax protein levels. Concomitantly, the protective effect of Sal against NPMSCs apoptosis was significantly attenuated upon reduced SIRT1 expression (Fig. 8B, C). Furthermore, mitochondrial homeostasis was evaluated using MitoTracker Red staining. Sal treatment enhanced MitoTracker Red fluorescence intensity in NPMSCs, whereas SIRT1 knockdown via ShRNA attenuated this enhancement (Fig. 8D, E). Concurrent MMP analysis revealed significant depolarization in both TBHP and TBHP + Sal+Sh- SIRT1 groups versus controls, whereas Sal treatment progressively restored mitochondrial membrane potential in TBHP stimulated NPMSCs (Fig. 8F, G).

Fig. 8.

Sal Activates SIRT1-Dependent Mitophagic Flux to Promote Mitochondrial Homeostasis in NPMSCs. A Representative images and quantification of LC3 (autophagosomes) and Tom20 (mitochondria) co-localization (scale bar, 25 μm). B, C Immunofluorescence analysis of Bax expression (scale bar, 100 μm). D, E MitoTracker Red staining assessing mitochondrial content (scale bar, 250 μm). F, G JC-1 assay measuring MMP. SIRT1 knockdown abolishes the beneficial effects of Sal (scale bar, 250 μm). The data are expressed as the mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 indicates statistical significance, n = 4

SIRT1-dependent mitophagic flux governs apoptosis resistance and cell cycle progression in NPMSCs

NPMSCs were cultured and divided into six experimental groups: Control, Sal+TBHP, Sal+TBHP + 3-MA + sh-SIRT1, Sal + sh-SIRT1, Sal+TBHP + sh-SIRT1, Sal+TBHP + 3-MA. All groups received standardized treatments: Sal (75 µM, 12 h), 3-MA (5 µM, 12 h), and TBHP (100 µM, 24 h). Immunoblot analysis revealed that while Sal treatment promoted mitophagy in TBHP stimulated NPMSCs, progressive P62 accumulation indicated severely impaired mitophagic flux (Fig. 9A, C). Relative to the Sal+TBHP + sh-SIRT1 group, 3-MA co-treatment substantially attenuated mitophagy with concomitant P62 accumulation indicating severe mitophagic flux blockade, yet demonstrated comparable apoptosis levels (Fig. 9A, C). Comparative analysis with the Sal+TBHP group showed that both SIRT1 knockdown and 3-MA administration suppressed mitophagy, exacerbated apoptosis, and upregulated P62 expression-indicating profound blockade of mitophagic flux (Fig. 9A, C). Notably, mitophagic activity showed no significant difference between Sal + sh-SIRT1 and Sal+TBHP + 3-MA + sh-SIRT1 groups, suggesting that TBHP-induced mitophagic activation and 3-MA-mediated inhibition reach comparable magnitudes under SIRT1-deficient conditions. Cell cycle analysis demonstrated pronounced G2/M phase arrest in both Sal+TBHP + 3-MA + sh-SIRT1 and Sal+TBHP + 3-MA groups following 3-MA treatment. SIRT1 knockdown similarly induced G2/M arrest in Sal + sh- SIRT1 and Sal+TBHP + sh-SIRT1 groups relative to Sal+TBHP controls, underscoring SIRT1-mediated mitophagy’s regulatory role in cell cycle progression (Fig. 9B, D).

Fig. 9.

Impaired mitophagic flux, induced by SIRT1 knockdown or 3-MA treatment, triggers cell cycle arrest at G2/M phase. A, C Immunoblot analysis of mitophagy markers across six experimental groups. Both SIRT1 deficiency and autophagic inhibition (3-MA) block mitophagic flux and exacerbate apoptosis. B, D Cell cycle analysis reveals that disruption of mitophagic flux, regardless of the method (Sh-SIRT1 or 3-MA), induces a pronounced G2/M phase arrest, highlighting the role of functional mitophagy in cell cycle progression. Full-length blots/gels are presented in supporting materials.The data are expressed as the mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 indicates statistical significance, n = 4

SIRT1 orchestrates lysosomal reprogramming to restore mitophagic flux in degenerative NPMSCs

As a transmembrane glycoprotein localized to lysosomal membranes, LAMP1 serves as a reliable indicator of lysosomal functional integrity [60]. The expression level of LAMP1 consequently reflects the efficiency of the lysosomal degradation phase during mitophagy. Initially, the results (Fig. 7) revealed that while TBHP stimulation initiated mitophagy, interestingly, concurrent Bax upregulation was observed. Subsequently, Sal treatment enhanced this TBHP induced mitophagic initiation while significantly suppressing Bax expression. Concurrently, TBHP stimulated NPMSCs exhibited P62 accumulation indicating impaired autophagic flux-a phenomenon alleviated by Sal. Notably however, SIRT1 knockdown via ShRNA markedly exacerbated this autophagic flux impairment. Given these observations, we postulated TBHP disruption of specific mitophagic stages caused P62 accumulation. To probe this, lysosome autophagosome fusion was examined through LC3 LAMP1 co-localization. Strikingly, TBHP stimulated NPMSCs showed reduced co-localization between LAMP1 positive lysosomes and LC3 positive autophagosomes. Sal treatment rescued this defect, whereas SIRT1 knockdown abolished Sal’s restorative effect (Fig. 10B). Further analysis demonstrated TBHP stimulation downregulated LAMP1 expression, indicating compromised lysosomal function. While Sal reversed this suppression, alarmingly, its protection was abrogated by either SIRT1 knockdown or chloroquine (CQ) - an autophagic flux inhibitor (Fig. 10A, C). This establishes Sal’s lysosomal activation via LAMP1 upregulation as SIRT1 dependent. Furthermore, CQ not only exacerbated lysosomal dysfunction but also blocked autophagic flux, consequently intensifying cellular apoptosis (Fig. 10A, C). Collectively, these results demonstrate that TBHP induced mitophagic initiation coincides with autophagic flux impairment attributable to diminished autophagosome lysosome fusion and compromised lysosomal degradation. Importantly, Sal activates SIRT1 to enhance lysosomal activity, ultimately facilitating autophagosome lysosome fusion and damaged mitochondrial degradation, thereby promoting NPMSCs survival. We concluded that mitophagic flux impairment stems from defective autophagosome lysosome fusion coupled with disrupted lysosomal degradation.

Fig. 10.

Sal restores lysosomal function and autophagosome-lysosome fusion in a SIRT1-dependent manner. A, C Immunoblot analysis showing that Sal upregulates LAMP1 expression and alleviates p62 accumulation, effects blocked by SIRT1 knockdown or CQ. B Representative images and quantification of LC3 and LAMP1 co-localization, demonstrating that Sal rescues TBHP-impaired autophagosome-lysosome fusion (scale bar, 25 μm). Full-length blots/gels are presented in supporting materials. The data are expressed as the mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 indicates statistical significance, n = 4

Sal attenuates IVDD progression via mitophagy activation in vivo

Based upon in vitro findings, we further validated the in vivo protective effects of Sal-activated mitophagy using a needle puncture-induced rat IVDD model, with comprehensive imaging and histological assessments (Fig. 11A). Five experimental groups were evaluated: Control (sham), IVDD, Sal alone (no puncture), IVDD + Sal, and IVDD + Sal+3-MA. Serial radiographic (X-ray) and MRI were performed at 0, 6, and 10 weeks post intervention, with subsequent histological evaluation of intervertebral disc at week 10 (Fig. 11A). As MRI represents the gold standard for IVDD assessment, decreased signal intensity in disc regions indicates degenerative progression [61]. Serial radiographic and MRI analyses revealed that the IVDD group exhibited characteristic progressive decreases in T2-weighted signal intensity and disc height index (DHI), whereas the Sal group and IVDD + Sal group showed significant preservation of both structural and hydrative integrity. Co-administration of the autophagy inhibitor 3-MA abrogated these protective effects (Fig. 11C–E). Sal treatment in healthy rats (Sal alone group) exhibited no intrinsic effect on normal disc physiology, confirming that its therapeutic actions are specifically elicited within a degenerative context. Collectively, these in vivo results confirmed that Sal specifically mitigates IVDD progression through mitophagy activation and matrix remodeling without altering normal disc homeostasis.

Fig. 11.

Sal alleviates intervertebral disc degeneration in vivo via activating mitophagy. A Experimental timeline of the needle puncture-induced rat IVDD model and therapeutic interventions (Created in BioRender). B, C, E Representative radiographic images (X-ray) and quantitative analysis of DHI (Created in BioRender) showing progressive disc space narrowing in IVDD and IVDD + Sal+3-MA groups, which was significantly mitigated by Sal treatment. D, E T2-weighted MRI images and signal intensity quantification demonstrating severe disc dehydration in the IVDD group, while Sal group and IVDD + Sal group preserved hydration and structure. 3-MA co-treatment abolished Sal’s protective effect. F, G, H (Fig. 12A) Histological evaluation (H&E, Safranin O staining, Masson) and semi-quantitative scoring revealing severe tissue disorganization, fibrosis, and proteoglycan loss in IVDD discs (scale bar, 1 mm). Treatment with Sal preserved histological organization and matrix integrity near normal levels in IVDD, while 3-MA exacerbated the degenerative process. The data are expressed as the mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 indicates statistical significance, n = 5

Histological assessment via hematoxylin and eosin staining and Masson trichrome staining revealed well preserved NP tissue occupying most disc areas in controls, with dispersed NP cells and well organized AF. Following puncture, NP cellularity progressively diminished with fibroblastic replacement. Sal treatment mitigated these degenerative changes compared to the Control group. The 3-MA treated group exhibited minimal cellular remnants within the NP region, with extensive replacement by disorganized fibroblasts, demonstrating that 3-MA compromises the protective efficacy of Sal on IVDD (Figs. 11F and 12A). Safranin O staining demonstrated substantial proteoglycan and glycosaminoglycan loss in degenerated NP tissue versus Control group, accompanied by severe lamellar disorganization or fragmentation. Sal attenuated these pathological alterations, while 3-MA administration recapitulated IVDD group histology (Fig. 11G). Notably, the Sal alone group displayed histological integrity indistinguishable from the Control group, confirming the absence of adverse effects on normal disc homeostasis. Quantitative histological scoring confirmed that while the IVDD + Sal group exhibited significantly lower degeneration grades compared to both the IVDD and IVDD + Sal+3-MA cohorts, the Sal alone group itself displayed no discernible degeneration (Fig. 11H). Morphological evaluation of NP architecture, cellularity, NP-AF border integrity, AF organization, and cellular structure was visualized through hierarchical clustering in the heatmap (Fig. 12D). Collectively, Sal demonstrated therapeutic potential for mitigating IVDD progression.

Fig. 12.

Sal enhances mitophagy and reduces apoptosis in punctured discs in a mitophagy-dependent manner. A (Fig. 11F–H) Histological evaluation (H&E, Safranin O staining, Masson) and semi-quantitative scoring revealing severe tissue disorganization, fibrosis, and proteoglycan loss in IVDD discs (scale bar, 1 mm). B, C Expression of CC3, P62, LC3-II, Aggrecan and MMP13 was detected by IHC (scale bar, 100 μm). D Hierarchical clustering heatmap summarizing key morphological and molecular parameters across experimental groups, confirming the protective profile of the Sal-treated group. The data are expressed as the mean ± SD; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 indicates statistical significance, n = 5

Immunohistochemical analysis demonstrated a significant increase in apoptosis of NPMSCs, indicated by elevated Cleaved-caspase 3 expression, in the IVDD group compared to controls at week 10. This pathological elevation in apoptosis was effectively reversed by Sal administration. At the mechanistic level, Sal treatment activated mitophagy in vivo, as shown by increased LC3-II expression alongside a concurrent reduction in P62 accumulation within NPMSCs of the punctured discs. Both the anti-apoptotic effect and the induction of mitophagy were abolished upon co-administration of the autophagy inhibitor 3-MA. Furthermore, Sal exerted a dual restorative influence on the extracellular matrix by simultaneously suppressing catabolic activity (downregulation of MMP13) and promoting anabolic processes (enhanced aggrecan deposition) (Fig. 12B, C). Importantly, administration of Sal alone did not induce any discernible degenerative changes or toxicity in healthy disc tissue, confirming that its therapeutic actions are specifically contingent upon a degenerative pathological context.

Discussion

Currently, IVDD is recognized as a primary cause of LBP, a condition affecting approximately 80% of individuals during their lifetime [62]. The progression of IVDD is closely linked to the functional decline of NPMSCs. These endogenous repair cells promote disc regeneration through their intrinsic regenerative capabilities [8, 63]. Studies have established that NPMSCs impairment arises from multiple factors, including biomechanical alterations, genetic predisposition, trauma, oxidative stress, acidic pH stress, and inflammation [31, 64]. Mitochondrial dysfunction plays a pivotal role in both the pathogenesis and progression of IVDD [51, 65]. Within the NP tissue, NPMSCs serve as endogenous repair cells. They mitigate NPMSCs apoptosis by initiating mitophagy to degrade damaged mitochondria; however, the precise mechanisms underlying this process remain incompletely elucidated. Excessive generation of ROS within mitochondria can trigger an imbalance in mitochondrial homeostasis or exacerbate impairments in mitochondrial quality control, ultimately inducing mitochondrion mediated apoptosis in NPMSCs [10]. Consequently, this study aims to investigate the significant therapeutic potential of pharmacologically targeting the SIRT1/FOXO3 pathway for disc rejuvenation.

SIRT1, a critical NAD-dependent deacetylase, plays multifaceted roles in maintaining mitochondrial function, promoting mitochondrial biogenesis, regulating autophagosome-lysosome processes, and participating in DNA repair and apoptosis modulation [12, 32]. FOXO3, a member of the forkhead box (FOX) transcription factor family, belongs to an evolutionarily conserved group of transcriptional regulators characterized by a distinctive forkhead DNA-binding domain. It plays a pivotal role in modulating apoptosis, oxidative stress responses, and cell cycle progression [31, 66, 67]. In this study, we observed that SIRT1 expression was downregulated not only in degenerated human NP tissues but also in TBHP-stimulated rat NPMSCs. Although our research group previously identified a strong correlation between SIRT1 and IVDD [8, 45, 62], the novel discovery herein is that the SIRT1/FOXO3 pathway can mediate mitophagy to alleviate IVDD. Specifically, we demonstrated that Sal activates SIRT1, leading to FOXO3 deacetylation and nuclear translocation. Crucially, dual-luciferase reporter and ChIP assays confirmed that FOXO3 directly binds to and transactivates the promoters of key mitophagy receptor genes—BNIP3, NIX, and FUNDC1—thereby initiating receptor-mediated mitophagy. Notably, while TBHP triggered mitophagic initiation, it concurrently impaired autophagic flux, as indicated by p62 accumulation. Sal not only promoted mitophagy initiation but also restored flux completion by enhancing lysosomal function and autophagosome–lysosome fusion, an effect dependent on SIRT1. This study advances prior work by examining not only mitophagic initiation but also the lysosomal degradation phase, thereby delineating the complete autophagic flux—a perspective often overlooked in previous IVDD research. Importantly, our findings suggest that therapeutic enhancement of mitophagy must consider flux integrity. Clinical decision-making should instead be guided by comprehensive molecular characterization, particularly with regard to SIRT1 expression status. The integrated implementation of network pharmacology, in silico virtual drug screening, molecular docking, and molecular dynamics simulations has substantially facilitated our exploration of SIRT1 and FOXO3 functions in NPMSCs. However, we must acknowledge the inherent limitations of these computational approaches. While generating extensive datasets, these methodologies necessitate sophisticated analytical frameworks for proper interpretation.

In this study, we observed that SIRT1 expression was downregulated in NPMSCs upon TBHP stimulation, accompanied by increased ROS production, activation of autophagy, and induction of apoptosis. Knockdown of SIRT1 via ShRNA exacerbated TBHP-induced ROS accumulation and apoptosis in NPMSCs, suggesting that SIRT1 may play a critical role in clearing damaged mitochondria and regulating ROS generation and cell survival. Notably, Sal treatment upregulated SIRT1 expression, activated mitophagy, and attenuated both apoptosis and ROS production in NPMSCs. To elucidate the protective role of SIRT1 in TBHP stimulated NPMSCs and the mechanism of Sal action, we modulated SIRT1 expression using Sal and Sh-SIRT1. Our findings demonstrated that SIRT1 suppression reduced ROS levels and enhanced NPMSCs viability; however, the cytoprotective effects of Sal were compromised by 3-MA. Consistent with our results, prior studies have shown that SIRT1 mediated mitophagy eliminates dysfunctional mitochondria and diminishes ROS production in endothelial and cardiomyocytes [68, 69]. Our mechanistic investigation further substantiated the linear SIRT1/FOXO3 signaling axis underpinning salidroside action. Pharmacological inhibition of SIRT1 with EX-527 abrogated Sal-induced FOXO3 nuclear translocation and deacetylation, confirming the necessity of SIRT1 enzymatic activity for FOXO3 activation. Co-immunoprecipitation assays directly demonstrated that Sal enhances the physical interaction between SIRT1 and FOXO3, facilitating FOXO3 deacetylation. Crucially, dual-luciferase reporter assays established FOXO3 as a direct transcriptional activator of key mitophagy receptor genes, including BNIP3, NIX, and FUNDC1. Furthermore, while TBHP stimulation triggered mitophagic initiation, it concurrently impaired autophagic flux, evidenced by p62 accumulation. Sal treatment not only promoted mitophagic initiation but also restored flux completion, as indicated by reduced p62 levels, an effect dependent on SIRT1. This restoration of functional flux was visually confirmed by enhanced LC3-LAMP1 co-localization, indicating improved autophagosome-lysosome fusion. Most definitively, immunoblot analysis of isolated mitochondrial fractions provided direct biochemical evidence that Sal specifically enriches LC3-II and p62 on mitochondria, a hallmark of targeted mitophagic engagement, in a manner entirely dependent on SIRT1 activity and intact autophagic flux. These data collectively delineate a coherent pathway whereby Sal activates SIRT1, leading to FOXO3 deacetylation and transcriptional upregulation of mitophagy receptors, thereby restoring functional mitophagic flux for the selective clearance of damaged mitochondria.

In vivo experiments further confirmed that Sal induced mitophagy exerted a protective effect in a rat model of IVDD, which was reversible upon blockade of the mitophagic flux. Furthermore, it is noteworthy that administration of Sal alone to healthy rats elicited no detectable degenerative changes or toxicity in intervertebral disc tissue. This group demonstrated histological, imaging, and molecular profiles indistinguishable from those of the sham control. This critical finding confirmed that the therapeutic actions of Sal are not constitutive but are specifically elicited within a degenerative pathological context. It underscores the compound’s favorable safety profile and its potential as a targeted therapeutic agent that modulates pathogenic pathways without disrupting normal tissue homeostasis. Intriguingly, although mitophagy was activated in TBHP stimulated NPMSCs, elevated P62 levels indicated impaired mitophagic flux. Subsequent experiments revealed that Sal not only triggered mitophagy initiation but also restored flux integrity, as evidenced by reduced P62 expression compared to TBHP stimulated NPMSCs. To investigate this phenomenon, we performed co-localization analysis of LC3 and LAMP1 via immunofluorescence. While TBHP stimulation initiated mitophagy, the interaction between damaged mitochondria and lysosomes was markedly attenuated. In contrast, Sal treatment robustly enhanced this association, facilitating lysosomal degradation of impaired mitochondria and ensuring mitophagic flux completion. These results align with reports by Liu et al., wherein lysosomal function restoration mitigated autophagic flux impairment in nucleus pulposus cells and alleviated mechanically induced intervertebral disc degeneration [1]. To further delineate the protective role of SIRT1 in IVDD pathogenesis, future studies should explore the mechanisms underlying lysosome-autophagosome fusion during the terminal phase of mitophagic flux.

This study has several limitations. First, while we demonstrate that Sal restores mitophagic flux, the precise mechanisms by which it facilitates the late stages of autophagy, particularly the enhancement of autophagosome-lysosome fusion, require further molecular delineation. Second, while the rat IVDD model established via caudal needle puncture is a well-established research tool, it fails to fully recapitulate the natural, multifactorial progression of human intervertebral disc degeneration. Future investigations should employ more physiologically relevant models, such as aged rat models or those incorporating abnormal mechanical loading, to better mimic the human disease phenotype. Third, our in vitro findings are based on a specific oxidative stressor (TBHP) and a defined concentration/time window for salidroside. The therapeutic window and efficacy of Sal across a broader range of degenerative stimuli and doses, particularly in a complex in vivo milieu, need further systematic investigation. Finally, although we established a linear SIRT1/FOXO3 axis, the potential crosstalk with other related pathways, such as AMPK or PINK1/Parkin, in the context of Sal’s action remains to be explored (Fig. 13).

Fig. 13.

Graphical abstract of the proposed mechanism by which Sal mitigates apoptosis in NPMSCs via the SIRT1/FOXO3-mitophagy pathway (Created in BioRender)

Conclusions

In conclusion, our data demonstrate that Sal alleviates mitochondrial dysfunction and apoptosis in NPMSCs under oxidative stress, and attenuates disc degeneration in a rat model. These effects are closely associated with the activation of SIRT1/FOXO3 signaling and the restoration of functional mitophagic flux. Our work highlights the SIRT1/FOXO3-mitophagy axis as a promising target for further investigation in the development of therapeutic strategies for IVDD.

Abbreviations

IVDD

Intervertebral disc degeneration

NPMSCs

Nucleus pulposus-derived mesenchymal stem cells

Sal

Salidroside

SIRT1

Sirtuin 1

LBP

Low back pain

ECM

Extracellular matrix

TNF-α

Tumor Necrosis Factor-α

IL-1β

Interleukin-1β

MSCs

Mesenchymal stem cells

BM

Bone marrow

UC

Umbilical cord

OXPHOS

Oxidative phosphorylation

ROS

Reactive oxygen species

FOXO

Forkhead box O

AMPK

AMP-activated protein kinase

HBSS

Hank’s Balanced Salt Solution

CTD

Comparative Toxicogenomics Database

DEGs

Differentially expressed genes

GEO

Gene Expression Omnibus

TEM

Transmission electron microscopy

Author contributions

**Zhengguang Li/Yiming Wu: ** Methodology, Software, Supervision, Writing-Original Draft; **Benkui Hua/Yongbo Zhang: ** Investigation; **Hua Sun/Chen Liu: ** Resources, Visualization; **Huofeng Wu/Shuangjia Zai/ Yufeng Huang: ** Data curation, Formal analysis; **Zhaoyu Li/Xuan You: ** Validation; **Xuhua Lu/Guoyong Yin/Liang Zhang: ** Funding acquisition, Project administration, Conceptualization.

Funding

This study is supported by the National Natural Science Foundation of China (82172462), Key Projects of Social development of Yangzhou City (YZ2022091/YZ2021083), The Sixth “333 project” Young Talent Project of Jiangsu Province [11], The Medical Discipline Construction Program of Shanghai Pudong New Area Health Commission (the Key Disciplines Program) (PWZxk2022-21), Postgraduate Research and Practice Innovation Program of Jiangsu Province (SJCX24_1563) and Postgraduate Research and Practice Innovation Program of Jiangsu Province (SJCX24_2340).

Data availability

All data generated or analyzed during this study are included in this article. The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study “Awakening endogenous repair: Salidroside boosts mitophagy in NPMSCs via SIRT1/FOXO3 to combat intervertebral disc degeneration” was approved by the Animal Ethics Committee of Yangzhou University (Approval No. 2021ky050) on February 22, 2021. Additionally, the study protocol was approved by the Ethics Committee of the Northern Jiangsu People’s Hospital under the project entitled “Clinical acquisition of human intervertebral disc degeneration tissues and associated mechanistic exploration” (Approval No. 202103427) on March 2, 2021. Informed consent was obtained from all individual participants included in the study.

Consent for publication

Our manuscript does not contain any individual person’s data in any form (including any individual details, images or videos).

Competing interests

The authors declare no competing interests.