Mitigating nucleus pulposus cell senescence in intervertebral disk degeneration: the role of nodakenin in SIRT6 and PINK1/Parkin-associated mitophagy signaling

Abstract

Intervertebral disc degeneration (IVDD) is a common and challenging chronic condition in orthopedics, primarily characterized by the aging of nucleus pulposus cells (NPC). Current treatment strategies for IVDD, particularly those targeting NPC senescence, remain underdeveloped. Research has shown that NPC senescence is closely associated with mitochondrial damage, leading to the accumulation of cytoplasmic reactive oxygen species (ROS) and mitochondrial DNA (mtDNA). Mitochondrial autophagy, as a key mechanism of mitochondrial quality control, regulates ROS and mtDNA levels by eliminating dysfunctional and damaged mitochondria, thereby delaying cellular aging. Notably, mitophagy signaling associated with Sirtuin 6 (SIRT6) and PTEN-induced kinase 1 (PINK1)/Parkin has been implicated in this process. Traditional Chinese Medicine (TCM), with its holistic approach and unique theoretical system of syndrome differentiation and treatment, offers significant advantages in preventing and treating degenerative diseases such as IVDD. However, research into TCM formulations aimed at NPC senescence remains limited. In preliminary studies, we observed that Nodakenin, the primary active compound from the TCM formula Duhuo Jisheng decoction (DHJSD), exerts protective effects in a SIRT6-associated manner and appears to be associated with activation of PINK1/Parkin-related mitophagy signaling. In this study, we used single-cell analysis to construct human NPC senescence and rat IVDD models, assessing mitochondrial morphology, mitochondrial membrane potential, cell senescence, and autophagy-related gene expression. We further evaluated the effects of Nodakenin (Nod) on H₂O₂-induced NPC senescence and determined the optimal intervention concentration, followed by assessment of mitochondrial phenotypes and SIRT6- and PINK1/Parkin-associated markers. These findings provide new theoretical support for the clinical application of TCM in the treatment of IVDD.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s11658-026-00968-y.

Introduction

Intervertebral disc degeneration (IVDD) is commonly observed in middle-aged and elderly individuals and is closely linked to cellular aging, immune decline, and degeneration of the intervertebral disc structure. It is a chronic and refractory condition in orthopedics [1, 2]. Low back pain (LBP) caused by IVDD affects approximately 40% of the global population, posing a significant public health issue [3]. In China, up to 90% of individuals over the age of 60 years suffer from IVDD, and the disease is becoming more prevalent among younger populations. This condition severely impacts patients’ quality of life and mental health, while also imposing a substantial economic burden on society [4, 5]. Currently, the primary treatments for IVDD include pharmacological and physical therapies, which can alleviate symptoms to some extent but fail to address the long-term suffering of patients [6]. Furthermore, surgical removal of protruding nucleus pulposus tissue can provide temporary relief from LBP, but many patients experience recurrent pain post-surgery [7]. Therefore, there is an urgent need for novel therapeutic approaches and strategies to improve the clinical management of IVDD [8].

The delayed research on the etiology and pathological mechanisms of IVDD has hindered the development of effective clinical treatments. Basic studies have indicated that the pathogenesis of IVDD is closely related to cellular aging, immune dysregulation, genetic factors, apoptosis, abnormal mechanical stress, and increased secretion of inflammatory cytokines [9, 10]. Therefore, a more detailed understanding of the pathogenesis of IVDD is crucial for improving existing clinical treatments. As a quintessential degenerative disease, the onset and progression of IVDD are tightly linked to aging, which is a critical factor in its development [11, 12]. At the cellular level, aging typically refers to a decline in cell proliferation, differentiation capacity, and physiological function, resulting in a stable cell cycle arrest [13, 14]. Current theories on aging mechanisms include the biological clock theory, reactive oxygen species (ROS) theory, and DNA damage accumulation theory [15]. External stimuli-induced ROS, DNA damage, and mitochondrial dysfunction play significant roles in the pathogenesis of degenerative diseases such as IVDD [16]. The oxidative stress caused by ROS and the subsequent mitochondrial dysfunction are closely associated with the development of aging-related diseases.

Mitochondrial damage is a hallmark of cellular aging and one of the potential targets in anti-aging research. Dysfunctional mitochondria exhibit increased outer membrane permeability, reduced membrane potential, and the entry of small molecules into the inner mitochondrial membrane, leading to mitochondrial swelling, the disappearance of cristae, and impaired respiratory chain function [17]. This exacerbates ROS production and results in the leakage of mitochondrial DNA (mtDNA) into the cytoplasm, activating the cyclic GMP-AMP synthase (cGAS)–Stimulator of Interferon Genes (STING) pathway and inducing the senescence-associated secretory phenotype (SASP) [18]. Recent studies have shown that, under nonlethal stress conditions, only a small portion of mitochondria undergo outer membrane permeabilization (minority mitochondrial outer membrane permeabilization, miMOMP) [19].Thus, selectively and efficiently removing miMOMP-affected mitochondria may serve as an important theoretical basis for reversing nucleus pulposus cell aging. Mitochondrial autophagy (mitophagy) is one of the most efficient mechanisms for eliminating dysfunctional mitochondria in aging cells. Mitophagy, as a primary mitochondrial quality control mechanism, effectively removes damaged and dysfunctional mitochondria [20]. The ubiquitin-dependent pathway is a crucial mechanism in this process. Ubiquitination of mitochondrial surface proteins promotes mitophagy. In this pathway, the PINK1/Parkin pathway plays a pivotal role. PTEN-induced kinase 1 (PINK1) localizes to damaged mitochondria, while Parkin, an E3 ubiquitin ligase, induces further ubiquitination and autophagy. When the membrane potential of some miMOMP-affected mitochondria is compromised, PINK1 is unable to enter the mitochondrial inner membrane, leading to its accumulation on the cytoplasmic face of the mitochondrial outer membrane, thereby facilitating the localization of damaged mitochondria. PINK1 recruits and activates Parkin, causing a conformational change in Parkin to its active E3 ubiquitin ligase form, which triggers the ubiquitination of mitochondrial outer membrane proteins. Phosphorylated ubiquitin (pSer65-Ub) then recruits autophagy receptors, such as Optineurin (OPTN) and Nuclear Dot Protein 52 (NDP52), to initiate the mitophagy process. Together, PINK1 and Parkin regulate mitochondrial autophagy to maintain mitochondrial quality [21].

SIRT6, a member of the sirtuin family of histone deacetylases, plays a crucial role in various cellular processes, including inflammation, apoptosis, aging, metabolism, and stress response. Loss of SIRT6 leads to cellular senescence and apoptosis [22]. Recent studies have shown that transgenic mice overexpressing SIRT6 exhibits a significant extension of lifespan, and its anti-aging effects are closely linked to mitochondrial function [23]. Research indicates that SIRT6 plays a key role in nucleus pulposus cells (NPCs) senescence, promoting autophagosome formation by regulating the mechanistic target of rapamycin (mTOR) signaling pathway. Overexpression of SIRT6 reduces the expression of senescence-associated proteins in NPCs and significantly improves the Pfirrmann scores in IVDD models [22]. Furthermore, SIRT6 agonists can enhance mitochondrial function in mouse cardiomyocytes by activating the PINK1/Parkin pathway [24]. However, the specific mechanisms by which SIRT6 delays NPC senescence remain unclear. It is hypothesized that SIRT6 may activate mitochondrial autophagy by regulating the PINK1/Parkin pathway, thereby delaying NPC senescence.

IVDD is classified in Traditional Chinese Medicine (TCM) under the categories of “Bi syndrome” and “low back pain.” The prevailing understanding of IVDD is that it primarily involves a pattern of deficiency with excessive conditions. The Duhuo Jisheng decoction (DHJSD), recorded in the “Bei ji Qian jin Yao fang” by the renowned Tang Dynasty physician Simiao Sun, consists of 15 herbal ingredients and has been in use for over a 1000 years. This formulation is known for its ability to expel wind and dampness, relieve pain, tonify the liver and kidneys, and invigorate blood circulation, making it an effective remedy for treating low back pain. Our research team has applied DHJSD in the clinical treatment of IVDD for over a decade, and studies have demonstrated its significant clinical efficacy [25]. Nodakenin (Nod), the principal active component of the key herb Duhuo in DHJSD, has been shown to regulate mitochondrial autophagy and maintain mitochondrial function in models of cartilage degeneration and knee osteoarthritis. It modulates reactive oxygen species (ROS) and the downstream formation of the NOD-like receptor family pyrin domain containing 3 (NLRP3) inflammasome, exhibiting protective effects against extracellular matrix degradation, anti-cell death, and antioxidant properties [26]. Previous studies have suggested that Nodakenin may modulate mitochondrial homeostasis and autophagy-related processes in degenerative disease models. In our preliminary analyses, molecular docking indicated a potential interaction between Nodakenin and SIRT6, and experimental results were consistent with SIRT6-related changes after Nodakenin treatment. Therefore, we investigated whether Nodakenin could alleviate oxidative stress–induced NPC senescence and disc degeneration, and whether these effects are accompanied by changes in mitophagy-related signaling, including SIRT6 and PINK1/Parkin-associated markers.

Materials and methods

Single-cell RNA sequencing data processing and analysis

Single-cell RNA-seq data of nucleus pulposus tissue were obtained from the Gene Expression Omnibus (GEO) dataset GSE244889, including seven patients with intervertebral disc degeneration (IVDD) (four mild and three severe cases). Quality control was performed using Seurat (version 4.4.0). Cells were removed if they expressed fewer than 200 or more than 8000 genes, if mitochondrial gene counts exceeded 20% of total unique molecular identifier (UMI) counts, or if hemoglobin gene counts exceeded 3%. Genes expressed in fewer than three cells were also discarded.

Data were normalized using the NormalizeData function with the “LogNormalize” method, followed by scaling with ScaleData to center and scale gene expression and to regress out potential technical confounders. Principal component analysis (PCA; RunPCA) was performed on the top 2000 highly variable genes, and the first 20 principal components were used for downstream analyses. Batch effects across samples were corrected using the Harmony algorithm. A shared nearest neighbor graph was constructed (FindNeighbors, dims = 1:20), and unsupervised clustering was performed with a graph-based algorithm (FindClusters, resolution = 0.3). Uniform manifold approximation and projection (UMAP; RunUMAP, dims = 1:20) was used for visualization.

Cluster-specific marker genes were identified using the FindAllMarkers function with the Wilcoxon rank-sum test. p-values were adjusted for multiple testing using the Benjamini–Hochberg method, and genes with log₂ fold change (log₂FC) > 0.25, adjusted p value (FDR) < 0.05 and min.pct > 0.2 were considered significant markers. To refine cell-type annotation, subclusters belonging to the same cell type were reanalyzed by repeating UMAP, graph-based clustering, and marker detection to resolve intratype heterogeneity.

Differentially expressed genes (DEGs) between groups (e.g., mild versus severe IVDD or between selected cell subsets) were identified using the FindMarkers function (Wilcoxon rank-sum test), with thresholds of log₂FC > 0.5, adjusted p value (FDR) < 0.05, and min.pct > 0.2. Gene symbols of DEGs were converted to Entrez IDs using the bitr function in the clusterProfiler package, and cluster-wise gene lists were constructed for downstream functional enrichment. Gene Ontology (GO) enrichment of biological process (BP) terms across clusters was performed using the compareCluster function in clusterProfiler, with FDR < 0.05 regarded as significant.

Gene set variation analysis (GSVA) was carried out with the GSVA package to calculate enrichment scores of selected gene sets in each cell cluster. Clusters were annotated based on the expression of canonical cell-type-specific marker genes. To assess differentiation status, CytoTRACE was used to infer differentiation potentials, and the cluster with the highest predicted stemness was chosen as the root for trajectory inference. Lineage trajectories and gene expression dynamics along pseudotime were analyzed using the slingshot package and visualized accordingly.

Reagents and materials

The immortalized human nucleus pulposus cells (HUM-iCELL-s012) used in this study were provided by the Chinese Academy of Sciences Cell Bank. These cells were derived from intervertebral disc tissue and were transduced with a lentivirus carrying the SV40 gene. Hydrogen peroxide (H₂O₂) (CAS no. 7722–84-1) was purchased from Sigma-Aldrich, and Nodakenin (CAS no. 495–31-8) was purchased from Aladdin Information Technology. Fetal bovine serum (FBS) (cat. no. F801-500) was purchased from Epizyme Biomedical, and Dulbecco’s modified Eagle medium (DMEM)/F12 medium (cat. no.. SP032030500, SP03103-0500) was purchased from SPERIKON. Antibodies for p62 (YM8025), Beclin1 (YM1326), LC3II/I (YT7936), p53 (YM4853), p21 (YT3497), p16 (YT5664), IL-1β (YT5201), TNF-α (YT4689), MMP13 (YT2796), IL-6 (YT5348), and GAPDH (YN5585) were obtained from Immunoway. The CCK-8 assay kit (cat. no. C0037) and Lipofectamine 3000 transfection reagent (cat. no. L3000015) were purchased from Thermo Fisher. JC-1 staining kit (cat. no. C2006), SA-β-gal staining kit (cat. no. C0603), and MitoSOX Red dye (cat. no. S0061S) were obtained from Beyotime. The ROS detection kit (cat. no. G1706-100) was purchased from Servicebio, penicillin–streptomycin solution (cat. no. DW0328) was from Dowobio, and the qRT-PCR kit (cat. no. Q222-01) was from Vazyme. The siSIRT6 reagent was purchased from TsingkeBiotech Co. Additional reagents used in this study included anti-acetyl-histone H3 (Lys9) rabbit pAb (cat. no. P013698, EpiZyme Biomedical, Shanghai, China), anti-acetyl-histone H3 (Lys56) rabbit mAb (cat. no. R015379, EpiZyme Biomedical), anti-LC3B rabbit mAb (cat. no. R013940, EpiZyme Biomedical), the SIRT6 agonist MDL-800 (cat. no. T64529, TargetMol, Shanghai, China), bafilomycin A1 (cat. no. T6740, TargetMol), proteinase K (cat. no. T8936, Solarbio, Beijing, China), and a non-denaturing tissue/cell lysis buffer (cat. no. R0030, Solarbio). Anti-phospho-Ubiquitin (Ser65) rabbit pAb (cat. no. AF3888, Affinity Biosciences, China), mt-Keima mitophagy reporter was purchased from GeneChem (contract no. 10132529; Shanghai, China).

Culture of nucleus pulposus cells

NPCs were cultured in DMEM/F12 medium supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin. The cells were maintained at 37 °C in a 5% CO₂ incubator. The culture medium was changed every 1–2 days, and cells were passaged when they reached 80–90% confluence. For passage, the cells were digested with 0.25% trypsin–ethylenediaminetetraacetic acid (EDTA) solution, collected, and then reseeded into new culture flasks.

Cell viability assay (CCK8)

Cell viability was assessed using a CCK8 kit. NPCs were seeded into a 96-well plate at a density of 1 × 10⁴ cells per well. After 24 h of incubation, 10 μL of CCK8 reagent was added to each well, and the plate was incubated at 37 °C for 1–4 h. The absorbance was measured at 450 nm using a microplate reader. Each group had three replicates.

Cell intervention

When NPCs reached 80% confluence, the following intervention steps were performed: Different concentrations (0 μmol/mL, 10 μmol/mL, 20 μmol/mL, 50 μmol/mL, and 100 μmol/mL) of H₂O₂ were added for a 24-h incubation. Cell viability was assessed using CCK8, and the expression levels of senescence markers (p53, p21, and p16) were measured by Western blotting to determine the optimal concentration for the intervention. On the basis of the determined optimal H₂O₂ concentration, various concentrations (0 μmol/L, 10 μmol/L, 20 μmol/L, 50 μmol/L, 75 μmol/L, and 100 μmol/L) of Nod were added to the cells for a 24-h treatment. The effects of Nod on NPC viability were evaluated using CCK8, and the low-, medium-, and high-dose groups with the least impact on cell viability were selected. Subsequently, the expression of autophagy and senescence markers was analyzed by Western blot to determine the optimal dose for subsequent experiments.

Cell transfection

Nucleus pulposus cells (NPCs) were cultured to 80% confluence, and 24 h prior to transfection, approximately 5 × 10⁴ cells were seeded into each well of a six-well plate. Lipofectamine 3000 transfection reagent was used for transfection. A mixture was prepared by combining 5 μL (20 μM) siRNA with 125 μL Opti-MEM medium in one tube, and 2.5 μL Lipofectamine 3000 with 125 μL Opti-MEM in another tube. The two solutions were gently mixed and incubated at room temperature for 10 min. The transfection mixture was then slowly added to the cell culture wells and gently mixed. The cells were incubated in a 37 °C, 5% CO₂ incubator for 6 h. After 6 h, the medium was replaced with fresh culture medium, and cells were incubated for an additional 24–48 h. Transfection efficiency was verified by qRT-PCR and Western blot analysis.

Autophagic flux assay with bafilomycin A1

To evaluate autophagic flux, NPCs under the indicated conditions (control, H₂O₂, H₂O₂ + Nod, and H₂O₂ + Nod + siSIRT6) were incubated with the lysosomal V-ATPase inhibitor bafilomycin A1 (BafA1) or vehicle for the last 4 h before lysis. LC3-II and p62 levels were then examined by Western blotting. The increase in LC3-II and p62 in the presence versus absence of BafA1 was used as an index of autophagic flux.

IL-1β-induced inflammatory model and SIRT6 modulation

To mimic inflammatory conditions associated with chronic disc degeneration, NPCs were stimulated with IL-1β (10 ng/mL, 24 h) [27]. Cells were assigned to control, IL-1β, IL-1β + Nod, IL-1β + siSIRT6, and IL-1β + MDL-800 groups. After treatment, protein expression of H3K9ac, H3K56ac, p62, PINK1, and SIRT6 was detected by Western blotting to assess whether SIRT6 activation also mediates Nod-induced mitophagy under inflammatory stress.

Drug affinity responsive target stability (DARTS) assay for SIRT6

The direct interaction between Nod and SIRT6 was further evaluated using a DARTS assay. Briefly, total protein lysates from NPCs were incubated with Nod at 20, 50, or 75 μM, or with DMSO (vehicle) for 1 h at 4 °C. Samples were then subjected to limited proteolysis with a protease mixture for a fixed time, whereas an aliquot without protease served as a “no-protease” control. Proteolysis was stopped by adding sodium dodecyl sulfate (SDS) loading buffer and boiling, and residual SIRT6 protein was quantified by Western blot. Protection of SIRT6 from proteolytic digestion by Nod was interpreted as evidence of drug–target binding.

Mitochondrial membrane potential measurement

Mitochondrial membrane potential was assessed using the JC-1 staining kit. After the cell intervention, cells were washed three times with phosphate-buffered saline (PBS) and then incubated with JC-1 staining working solution at 37 °C for 20 min. Following incubation, cells were washed three times with JC-1 staining buffer. Fluorescence microscopy was used to observe the cells. JC-1 emits red fluorescence at high membrane potential and green fluorescence at low membrane potential. The red/green fluorescence ratio was calculated to evaluate changes in mitochondrial membrane potential.

Measurement of intracellular ROS levels

Intracellular ROS levels were measured using the DCF-DA probe. Briefly, NPCs were seeded in six-well plates and treated at the designated time points. Then, 1 mL of DMEM/F12 containing 10 μM DCFH-DA was added to the cells, and the cells were incubated in the dark at 37 °C for 30 min. After incubation, cells were washed three times with PBS. Fluorescence microscopy was used to observe the cells, and the green fluorescence intensity was recorded to assess ROS levels.

Measurement of mitochondrial ROS levels

Mitochondrial ROS levels were detected using MitoSOX Red dye. The MitoSOX probe selectively enters mitochondria and gets oxidized by superoxide to produce red fluorescence. Briefly, NPCs were seeded in six-well plates, and 1 mL of DMEM/F12 containing 5 μM MitoSOX dye was added to each well. The cells were then incubated in the dark at 37 °C for 10 min. Fluorescence microscopy was used to observe the staining results, and the red fluorescence intensity was quantified for statistical analysis of mitochondrial ROS levels.

SA-β-gal staining

SA-β-gal staining was performed to assess cellular senescence. NPCs were seeded in six-well plates, and after the treatment, cells were washed three times with PBS and fixed at room temperature for 5 min using fixation solution (2% formaldehyde, 0.2% glutaraldehyde). After three washes with PBS, 1 mL of freshly prepared SA-β-gal staining solution was added to each well, and the cells were incubated at 37 °C (without CO₂) for 12–16 h. The blue precipitates in the cytoplasm were observed. Six random fields from each sample were selected for counting, and the percentage of positive cells was calculated.

Cell protein extraction and western blotting

Cells were lysed using radioimmunoprecipitation assay (RIPA) lysis buffer on ice for 30 min. After centrifugation at 12,000 rpm for 10 min, the supernatant was collected. Protein concentration was quantified using the bicinchoninic acid (BCA) assay. Nucleus pulposus tissues were collected and homogenized in RIPA buffer with protease inhibitors. For SDS-polyacrylamide gel electrophoresis (PAGE), 30 μg of protein was loaded onto the gel and separated by electrophoresis. The proteins were then transferred to a polyvinylidene fluoride (PVDF) membrane, which was blocked with 5% nonfat milk for 1 h. The membrane was incubated with primary antibodies overnight at 4 °C, followed by incubation with secondary antibodies for 1 h at room temperature. Chemiluminescence detection was used to visualize and measure the expression of target proteins.

RNA extraction and qRT-PCR

Total RNA was extracted from cells using an RNA extraction kit. After the cell intervention, cells were gently washed twice with PBS, and 1 mL of TRIzol reagent was added to each well of the six-well plate. Cells were incubated at room temperature for 5 min to fully lyse. To the lysate, 200 μL of chloroform was added and the mixture was vortexed vigorously for 15 s, followed by incubation at room temperature for 3 min. The sample was then centrifuged at 12,000 × g for 15 min (4 °C), and the supernatant was transferred to a new RNase-free tube. To the supernatant, an equal volume of isopropanol was added and mixed gently. The sample was incubated at room temperature for 10 min, then centrifuged at 12,000 × g for 10 min (4 °C). The supernatant was discarded, and the RNA pellet was retained. The pellet was washed with 1 mL of 75% ethanol and centrifuged at 7500 × g for 5 min (4 °C). This washing step was repeated once. After air drying for 5–10 min, the RNA was dissolved in 30–50 μL of RNase-free water. RNA concentration and purity were determined using NanoDrop 2000, with an A₂₆₀/A₂₈₀ ratio between 1.8 and 2.0. RNA was then reverse-transcribed into complementary DNA (cDNA) using a reverse transcription kit. Quantitative PCR was performed using a qRT-PCR kit to detect the expression of target genes, with reduced glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as the internal control. PCR conditions were as follows: initial denaturation at 95 °C for 3 min, followed by 40 cycles of denaturation at 95 °C for 30 s, annealing at 60 °C for 30 s, and extension at 72 °C for 30 s. The relative quantification method used was 2−ΔΔCt. The primer sequences were as follows:

Gene	Species	Primer sequence (5′ to 3′)
P53	Human	Forward: TGTCAAGCTCATTTCCTGGTATG
		Reverse: TCTCTCTTCCTCTTGTGCTCTTG
P21	Human	Forward: TCAGGGTCGAAAACGGCG
		Reverse: CTTCCTGTGGGCGGATTAGG
P16	Human	Forward: GCCCAACGCACCGAATAGTT
		Reverse: ATGGTTACTGCCTCTGGTGC
BECLIN1	Human	Forward: TCAGGAGGAAGCTCAGTATCAGA
		Reverse: GCGACCCAGCCTGAAGTTATT
LC3	Human	Forward: GAGCGCTACAAGGGTGAGAAG Reverse: GGCGTAGACCATATAGAGGAAGC
PINK1	Human	Forward: GGGAGTATGGAGCAGTCACTTAC Reverse: GCAGGGTACAGGGATAGTTCTTC
PARKIN	Human	Forward: TTGCGTGTGATTTTCGCAGG Reverse: TCCACGGTCTCTGCACAATG
IL-1β	Human	Forward: ATGATGGCTTATTACAGTGGCAA Reverse: GTCGGAGATTCGTAGCTGGA
TNF-α	Human	Forward: GCAACAAGACCACCACTTCG Reverse: GATCAAAGCTGTAGGCCCCA
IL-6	Human	Forward: TGACAACTCATCTCATTCTGCG Reverse: AGAAGAAGGAATGCCCATTAACA
MMP-13	Human	Forward: GGCTTAGAGGTGACTGGCA Reverse: CATCAGGAACCCCGCATCTT
GAPDH	Human	Forward: CCAGCAAGAGCACAAGAGGA Reverse: TGAGGAGGGGAGATTCAGTGT

Transmission electron microscopy (TEM) observation

After cell intervention, transmission electron microscopy (TEM) was performed to observe mitochondrial morphology. Briefly, cells were fixed with 2.5% glutaraldehyde at 4 °C for 2 h, followed by three washes with PBS, each for 10 min. The cells were then fixed with 1% osmium tetroxide for 2 h. For dehydration, cells were subjected to an ethanol gradient, sequentially using 50%, 70%, 90%, and 100% ethanol, each for 10 min. The samples were then infiltrated with 100% acetone for 10 min, three times. Cells were embedded at 37 °C for 12 h, sectioned, and stained. Finally, mitochondrial morphology was observed under a transmission electron microscope.

Molecular docking experiment

To investigate the interaction between Nod and SIRT6, we conducted molecular docking experiments. First, the three-dimensional structure of SIRT6 was obtained from the Protein Data Bank (PDB) (PDB ID: 6HOY). The two-dimensional structure of Nod was drawn using ChemOffice, and its three-dimensional structure was generated by performing energy minimization using Chem3D.

Next, molecular docking was performed using AutoDock Vina software. The three-dimensional structures of SIRT6 and Nod were imported into AutoDock tools, and necessary preprocessing steps were carried out, including the addition of hydrogen atoms, calculation of charges, and definition of the docking grid. Docking parameters were set, and molecular docking calculations were performed. After docking, the results were analyzed, and the conformation with the lowest binding energy was selected for further analysis.

mt-Keima mitophagy reporter assay

NPCs were transduced with the mt-Keima lentiviral reporter at a multiplicity of infection (MOI) of 20 for 24 h, followed by replacement with fresh complete medium. After 48 h of expression, cells were assigned to the following groups: Control, H₂O₂, H₂O₂ + Nod, H₂O₂ + Nod + siSirt6, and H₂O₂ + MDL-800. Treatments were performed as described above. Fluorescence images were acquired under identical microscope settings for each channel across all groups. Mitophagy activity was quantified as the mt-Keima red/green fluorescence ratio using ImageJ. For each independent experiment, at least five randomly selected fields were analyzed, and data were pooled from n = 3 independent experiments for statistical analysis.

In vivo validation of IVDD in rats

Animal grouping and treatment

Healthy male SD rats (weight 250–300 g) were obtained from the Laboratory Animal Center of Southwest Medical University, with ethical approval granted by the Ethics Committee [20240508–008]. The rats were randomly divided into three groups (N = 10 per group): sham surgery group, IVDD model group, and Nod treatment group. The sham surgery group underwent skin puncture without penetrating the intervertebral disc. The IVDD model group and Nod treatment group were first subjected to an IVDD model using acupuncture. Rats were anesthetized intraperitoneally with 50 mg/kg sodium pentobarbital. Once adequately anesthetized, a 16G needle was vertically inserted into the annulus fibrosus of the intervertebral disc, with a penetration depth of 4 mm, and left for 10 s before removal. The L3, L4, and L5 lumbar vertebrae were selected for this procedure. Following the modeling, rats in the treatment group were treated with Nod via gavage for 30 days at a concentration of 10 mg/mL. In addition to the above groups, an additional Model + Nodakenin + 3-MA group was included. To pharmacologically inhibit autophagy in vivo, rats in the Model + Nodakenin + 3-MA group received intraperitoneal injections of 3-MA at a dose of 15 mg/kg body weight, prepared in sterile PBS. We administered 3-MA once daily starting from the first day of Nodakenin treatment. All injection procedures were performed under aseptic conditions.

DR detection of intervertebral disc height index (DHI)

After gavage treatment, the intervertebral disc height of rats was assessed using the DR imaging system. The disc height index (DHI) was calculated by measuring the ratio of the intervertebral disc height to the height of the adjacent vertebral bodies. Comparisons were made between the sham surgery, IVDD model, and Nod treatment groups.

HE, Masson, and Safranin O–Fast Green staining

Rats were euthanized after anesthesia, and their intervertebral disc tissues were collected and fixed in a 10% neutral formalin solution. The tissues were embedded in paraffin and sectioned to 4 μm thickness. After dewaxing in xylene and hydration through graded alcohol, the sections were stained with hematoxylin and eosin (HE) for observation of cellular morphology and tissue structure changes under a light microscope.

For Masson staining, paraffin sections were dewaxed to water and stained following the instructions of the Masson trichrome staining kit. The distribution of collagen fibers was observed and captured for analysis.

For Safranin O–Fast Green staining, paraffin sections were deparaffinized and rehydrated, and staining was performed following the manufacturer’s protocol. Proteoglycan content and cartilage matrix integrity were evaluated under a light microscope, and representative images were captured for subsequent analysis.

Immunohistochemistry

Paraffin sections were dewaxed to water, and antigen retrieval was performed by high-temperature and high-pressure treatment in citrate buffer (pH 6.0). The sections were incubated at room temperature with 3% hydrogen peroxide for 30 min to block endogenous peroxidase activity, followed by blocking with 5% bovine serum albumin (BSA) for 1 h. Primary antibodies (p53, LC3, SIRT6, p16, and p21) were applied at a dilution of 1:200 and incubated overnight at 4 °C. The following day, sections were washed with PBS, and secondary antibody was added for 1 h at room temperature. We used 3,3′-diaminobenzidine (DAB) for color development, and hematoxylin counterstaining was performed. The sections were observed and photographed under a light microscope.

Data statistics

All experimental data were statistically analyzed using GraphPad Prism 8 software. For in vitro experiments, data are presented as mean ± standard deviation (mean ± SD) from at least three independent experiments (n = 3). For in vivo studies, n = 10 rats per group unless otherwise stated. Statistical comparisons between groups were made using Student’s t-test or one-way analysis of variance (ANOVA), with p < 0.05 considered statistically significant.

Results

Single-cell sequencing analysis reveals the relationship between nucleus pulposus cell senescence and intervertebral disc degeneration

Intervertebral disc degeneration is closely related to cellular senescence [28]. To explore this relationship, we performed further analysis using single-cell sequencing data from the GSE244889 dataset in the GEO database, which includes nucleus pulposus tissue from seven patients with intervertebral disc degeneration (three males and four females, with four cases of mild degeneration and three cases of severe degeneration). After filtering low-quality genes and cells, we obtained 52,294 cells that clustered into 14 distinct cell populations (Fig. 1A). We then identified nine cell types based on the specific gene markers of each cluster: nucleus pulposus cells, T cells, macrophages, neutrophils, smooth muscle cells, red blood cells, B cells, endothelial cells, and plasma cells (Fig. 1C, E).

Fig. 1.

Single-cell sequencing analysis of nucleus pulposus tissue in intervertebral disc degeneration. A UMAP plot of cell clustering: A total of 52,294 cells after filtering from seven samples. B Heatmap of marker gene expression for each cell type (yellow: high expression; blue/black: low expression). C UMAP plot displaying identified cell types: nucleus pulposus cells, T cells, macrophages, neutrophils, smooth muscle cells, red blood cells, B cells, endothelial cells, and plasma cells. D Violin plots of marker gene expression for each cell type. E UMAP plot showing cell distribution between mild degeneration (MDD) and severe degeneration (SDD) groups. F Proportion of cell types and cell cycle phases (G1, S, G2/M) in MDD and SDD groups. G Correlation heatmap among cell types. H Heatmap of senescence-associated gene expression across different groups. I Heatmap of Hallmark gene set enrichment between groups (red: high enrichment; blue: low enrichment)

Nucleus pulposus cells: SOX9 (SRY-box transcription factor 9), ACAN (aggrecan). T Cells: CD3 (CD3e molecule), CD3G (CD3γ molecule). Macrophages: FCGR2A (Fc fragment of IgG receptor IIa), CD163 (CD163 molecule). Neutrophils: MPO (myeloperoxidase), LTF (lactoferrin). Smooth muscle cells: PLN (phospholamban), ACTA2 (smooth muscle alpha-actin 2). Red blood cells: HBA1 (hemoglobin alpha-1 chain), HBA2 (hemoglobin alpha-2 chain). B cells: CD79A (CD79a molecule), CD79B (CD79b molecule). Plasma cells: PECAM1 (platelet/endothelial cell adhesion molecule 1), RAMP2 (receptor activity-modifying protein 2) (Fig. 1B, D). As degeneration worsened, the proportion of cells in the G1 phase (assessed on the basis of the expression of classical marker genes for the G2/M and S phases, where cells not expressing these markers are considered in G1 phase) was elevated in the SDD (severe degeneration) group [29]. Additionally, the expression of the senescence-associated secretory phenotype (SASP) markers, such as p21, p16, TNF, and MMP13, was significantly higher in the SDD group, and similar trends were observed for senescence and oxidative stress-related pathways (Fig. 1F–I). On the basis of these findings, we hypothesize that excessive senescence of nucleus pulposus cells exacerbates intervertebral disc degeneration.

Single-cell sequencing analysis of nucleus pulposus cell subpopulations and their role in aging-related disc degeneration

To further explore whether aging-related subpopulations in nucleus pulposus (NP) cells contribute to aging, we isolated and reclustered the NP cells, obtaining six cell populations from a total of 43,124 cells. Subpopulation annotation was based on specific marker genes and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis of each population (Fig. 2A–C):

Fig. 2.

Subpopulation analysis reveals nucleus pulposus cell senescence-associated subpopulations and their role in intervertebral disc degeneration. A Reclustering of NPCs, followed by GO BP enrichment analysis of differentially expressed genes between the six clusters, with the specific biological processes selected highlighted in red boxes. B NPCs subpopulations identified through specific gene or biological process annotation. C Heatmap of marker gene expression for each cell type (yellow represents high expression, blue and black represent low expression). D Proportion of nucleus pulposus cell subpopulations across different samples. E Correlation heatmap of cell types, with red indicating high correlation and blue indicating low correlation. F Cell differentiation potential analysis: blue indicates low differentiation potential; red indicates high differentiation potential. G Pseudo-time trajectory analysis of nucleus pulposus cell subpopulations, with arrows indicating differentiation direction; the right panel shows the differentiation time course on the x-axis and cell/sample density on the y-axis. H Stem cell gene scoring in nucleus pulposus cells across different samples, with higher scores in the MDD group. I Heatmap of Hallmark gene set enrichment across different groups

AS-NP (aging NP): associated with “response to transforming growth factor beta” and “cellular response to transforming growth factor beta stimulus” [30].

OS-NP (oxidative stress NP): Associated with “reactive oxygen species metabolic process” and “response to oxidative stress” [31].

MS-NP (matrix synthesizing NP): related to “extracellular matrix organization,” “extracellular structure organization,” “external encapsulating structure organization,” “cartilage development,” “connective tissue development,” and “glycoprotein metabolic process.”

Nerve-NP (neuroregenerative NP): associated with “positive regulation of nervous system development” and “positive regulation of neurogenesis” [32].

IR-NP (immune response NP): related to “leukocyte migration,” “cell–substrate adhesion,” and “regulation of leukocyte migration” [33].

In SDD, a notable increase in the proportion of aging NPCs, specifically the AS-NP subpopulation, was observed. The violin plots of marker gene expression for each cell type illustrate distinct expression profiles, highlighting aging-related markers prominently in the AS-NP cells compared with other subpopulations. The UMAP plot shows the distribution of cells in MDD and SDD groups, with a clear shift toward a higher prevalence of AS-NP cells in the SDD group, reflecting advanced cellular aging and degeneration (Fig. 2D, E).

CytoTRACE analysis revealed that NP cell differentiation originates from the Nerve-NP subpopulation, while aging-related AS-NP is in the terminal stage of differentiation and is associated with increased secretion of p21. Additionally, the proportion of AS-NP is significantly higher in the SDD (severe degeneration) group. Clearly, as disc degeneration progresses, the degree of NP cell aging also increases significantly (Fig. 2F–I). These results strongly suggest that aging of NP cells exacerbates disc degeneration.

Hydrogen-peroxide-induced NPC senescence

On the basis of the results from the previous single-cell analysis, we confirmed that NPC senescence plays a crucial role in IVDD. We then established an NPC senescence model using hydrogen peroxide (H₂O₂) at different concentrations to induce senescence, based on previous studies. The results showed that treatment with 100 μmol/L H₂O₂ significantly reduced NPCs cell viability. CCK-8 assays indicated a dose-dependent decrease in cell viability with increasing H₂O₂ concentrations. The data showed that the cell viability in the 100 μmol/L H₂O₂ treatment group was significantly lower than that in the control group (p < 0.05) (Fig. 3A). Therefore, this concentration was selected as the optimal intervention concentration for subsequent experiments. SA-β-gal staining results showed that H₂O₂ treatment significantly increased the proportion of SA-β-gal-positive cells in NPCs. The staining demonstrated that, after 24 h of treatment, the H₂O₂ treatment group exhibited significantly more SA-β-gal-positive cells compared with the control group, indicating a pronounced senescence effect induced by H₂O₂ (p < 0.05) (Fig. 3B).

Fig. 3.

Nodakenin alleviates H₂O₂-induced senescence and enhances autophagy in NPCs. A CCK-8 assay demonstrating dose-dependent reduction in NPC viability after H₂O₂ treatment, with 100 μmol/L selected as the optimal concentration for inducing senescence (p < 0.05). B SA-β-gal staining showing increased senescent (SA-β-gal positive) NPCs in the H₂O₂ group compared with control (p < 0.05). C, D Western blot analysis of senescence markers p53, p21, and p16. E CCK-8 assay revealing increased viability of H₂O₂-treated NPCs with 20 μmol/L, 50 μmol/L, and 75 μmol/L Nod, identifying 75 μmol/L as the optimal dose (p < 0.05). F, G Western blot analysis of autophagy markers Beclin1, p62, showing increased Beclin1 and reduced p62 in Nod-treated groups, particularly at 75 μmol/L (p < 0.05). H Quantification of mitochondrial membrane potential expressed as the JC-1 red/green fluorescence ratio in each group. I Representative JC-1 staining images showing changes in mitochondrial membrane potential (red fluorescence, high ΔΨm; green fluorescence, low ΔΨm). Data are presented as mean ± SD (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Additionally, Western blot analysis revealed that, after H₂O₂ treatment, the expression levels of senescence-related proteins, such as p53, p21, and p16, were significantly increased (Fig. 3C, D).

Nodakenin optimal intervention concentration improves nucleus pulposus cell senescence

To evaluate the protective effect of Nod on H₂O₂-induced NPC senescence and determine the optimal intervention concentration, we first assessed cell viability by CCK-8. Treatment with Nod at10, 20, 50, and 75 μmol/L for 24 h significantly improved viability compared with the H₂O₂ model group (Fig. 3E); therefore, these doses were selected as the low-, medium-, and high-dose groups. Western blotting showed that H₂O₂ markedly increased the expression of the senescence-associated proteins p53, p21, and p16, whereas Nod treatment dose-dependently reduced their levels, with the most pronounced effect at 75 μmol/L (Fig. 3F, G). Nod also appeared to promote autophagy, as evidenced by increased Beclin1 and decreased p62 expression, particularly at 75 μmol/L. Consistently, JC-1 staining showed that Nod restored the mitochondrial membrane potential of senescent NPCs in a dose-dependent manner, with the high-dose group (75 μmol/L) exhibiting the greatest recovery in the JC-1 red/green fluorescence ratio (Fig. 3H, I).

Nodakenin improves mitochondrial stress injury and NPC senescence, accompanied by changes in autophagy-related markers

To investigate the mechanism by which Nod improves mitochondrial function and alleviates NPC senescence, we divided NPCs into four groups: normal, model (H₂O₂), model + Nod, and model + Nod + 3-MA (autophagy inhibitor). First, we performed Western blot (WB) analysis to examine the expression of mitochondrial autophagy-related proteins. The results showed that, in the model group, the expression levels of autophagy-related proteins PINK1, Parkin, Beclin1, and LC3-II/I showed a slight increase, while p62 levels decreased slightly, indicating that autophagy processes were still present but insufficient to improve the oxidative stress-induced senescence. In the Nod-treated group, the expression levels of PINK1, Parkin, Beclin1, and LC3-II/I were significantly higher, and p62 levels were lower, suggesting that Nod effectively activated the autophagy pathway. Moreover, in the model group, the expression levels of senescence-related proteins p53, p21, and p16 were elevated, along with an increase in the expression of senescence-associated secretory phenotype (SASP) markers (IL-1β, TNF-α, IL-6, and MMP13). In contrast, in the Nod-treated group, the expression levels of these proteins were significantly reduced, indicating that Nod inhibited cellular senescence. Furthermore, when the 3-MA autophagy inhibitor was added after Nod treatment, the expression levels of PINK1, Parkin, Beclin1, and LC3-II/I decreased, and p62 levels increased, indicating that the autophagy activation effect of Nod was reversed by the autophagy inhibitor. The expression levels of IL-1β, TNF-α, IL-6, MMP13, p53, p21, and p16 were again elevated (Fig. 4A, B). Subsequently, qRT-PCR analysis of autophagy and senescence-related genes confirmed these findings, showing consistent results with the WB analysis (Fig. 4C). These results suggest that, when autophagy function is inhibited, the anti-senescence effects of Nod are lost, emphasizing the crucial role of autophagy in Nod’s therapeutic action.

Fig. 4.

Nodakenin activates mitochondrial autophagy to alleviate senescence and inflammation in NPCs. A Western blot analysis of autophagy-related proteins (PINK1, Parkin, Beclin1, LC3-II/I, and p62) and senescence-associated proteins (p53, p21, p16, IL-1β, TNF-α, IL-6, and MMP13) in different treatment groups. B Quantification of Western blot results. C qRT-PCR analysis of autophagy- and senescence-related gene expression. D JC-1 staining of mitochondrial membrane potential. Data represent mean ± SD from three independent experiments (n = 3); *p < 0.05 compared with the model group (one-way ANOVA)

Additionally, mitochondrial membrane potential was assessed using JC-1 staining. The results showed that the mitochondrial membrane potential was decreased in the model group, while the Nod-treated group exhibited significant recovery of the membrane potential. However, in the H₂O₂ + Nod + 3-MA autophagy inhibitor group, the membrane potential was reversed, reflecting the inhibitory effect of 3-MA on Nod-induced mitochondrial membrane potential restoration (Fig. 4D). ROS staining and MitoSOX staining were performed to evaluate the levels of reactive oxygen species (ROS). The results demonstrated that both intracellular and mitochondrial ROS levels were elevated in the model group. In contrast, the Nod-treated group showed a significant reduction in both intracellular and mitochondrial ROS levels. Furthermore, the H₂O₂ + Nod + 3-MA autophagy inhibitor group reversed the improvements in ROS levels observed in the Nod-treated group (Fig. 5A, B).

Fig. 5.

Nodakenin reduces ROS levels and mitochondrial damage via mitochondrial autophagy activation. A ROS staining in NPCs, showing intracellular ROS levels across treatment groups. B MitoSOX staining indicating mitochondrial ROS levels. C SA-β-gal staining for senescence analysis. D TEM images showing mitochondrial morphology and autophagosome formation. Data represent mean ± SD from three independent experiments (n = 3); *p < 0.05 compared with the model group (one-way ANOVA)

Cellular senescence was analyzed through SA-β-gal staining. The results showed an increased proportion of SA-β-gal-positive cells in the model group, while the proportion of SA-β-gal-positive cells was significantly reduced in the Nod-treated group, indicating that Nod effectively inhibits cellular senescence. However, the addition of the 3-MA autophagy inhibitor to the Nod-treated group resulted in a marked increase in the proportion of SA-β-gal-positive cells, suggesting that the 3-MA inhibitor reversed the anti-senescence effect of Nod (Fig. 5C).

Finally, transmission electron microscopy (TEM) was used to observe the ultrastructure of the cells. The results showed that the model group exhibited altered mitochondrial morphology, characterized by swelling, cristae rupture, or disappearance, indicating mitochondrial damage. In contrast, the Nod treatment group displayed a significant reduction in damage features, with a notable increase in the number of autophagosomes and lysosomes, suggesting that Nod treatment activated mitochondrial autophagy. However, in the Nod + 3-MA inhibitor group, mitochondrial morphology was once again damaged, indicating that the 3-MA inhibitor reversed the protective effects of Nod on mitochondria (Fig. 5D).

In summary, these results suggest that Nod improves mitochondrial function and reduces ROS levels both in the cells and mitochondria by activating mitochondrial autophagy, thereby slowing down the process of nucleus pulposus cell senescence.

Nodakenin attenuates NPC senescence in a SIRT6-associated manner

To verify that Nod improves mitochondrial dysfunction and nucleus pulposus cell senescence through regulating SIRT6 and activating mitochondrial autophagy, we divided the nucleus pulposus cells into four groups: normal group, model group (H₂O₂ treatment), model + Nod group, and model + Nod + siSIRT6 group. In the siSIRT6 group, siRNA was used to silence the expression of SIRT6. First, Western blot (WB) analysis was used to confirm the expression of SIRT6. The results showed that SIRT6 expression was significantly higher in the model + Nod treatment group compared to the model group, while expression of SIRT6 was reduced in the siSIRT6 group (Fig. 6A). Additionally, molecular docking results showed that the binding energy between Nod and SIRT6 was −9.0 kcal/mol (Fig. 6B), indicating a strong binding affinity, further supporting the mechanism by which Nod regulates SIRT6 to exert its effects.

Fig. 6.

Nodakenin regulates Sirt6 to activate mitochondrial autophagy, reducing NPC senescence. A Western blot analysis of Sirt6 expression across treatment groups. B Molecular docking of Nodakenin with Sirt6, showing a binding energy of −9.0 kcal/mol. C Correlation analysis of SIRT6 with PINK1, Parkin, and senescence-related markers in the SDD group. D Western blot analysis of mitochondrial autophagy proteins (PINK1, Parkin, LC3, Beclin1, and p62) under different treatments. E Box plot showing the expression levels of SIRT6 in the MDD and SDD groups. F qRT-PCR analysis of mitochondrial autophagy-related gene expression. G JC-1 staining of mitochondrial membrane potential in treated NPCs. H SA-β-gal staining indicating senescence levels across groups. Data represent mean ± SD from three independent experiments (n = 3); *p < 0.05 compared with the model group (one-way ANOVA)

Next, WB analysis was performed to assess the expression of mitochondrial autophagy and senescence-related proteins. The results showed that, compared with the model group, the model + Nod treatment group exhibited significantly increased levels of mitochondrial autophagy-related proteins PINK1, Parkin, LC3, and Beclin1, while the expression of p62 decreased, indicating that the autophagic process was activated. In the model + Nod + siSIRT6 group, however, the expression of PINK1 and Beclin1 decreased, and the expression of p62 increased, suggesting that siSIRT6 reversed the protective effects of Nod (Fig. 6D).

RNA-Seq analysis also revealed that in the SDD group, SIRT6 expression was downregulated and positively correlated with PINK1 and Parkin, while it showed a negative correlation with cellular senescence markers (Fig. 6C, E).

The qRT-PCR results were consistent with the WB findings. In the model + Nod treatment group, the messenger RNA (mRNA) levels of PINK1, Parkin, LC3, and Beclin1 significantly increased, while the mRNA level of p62 decreased. In the model + Nod + siSIRT6 group, the mRNA level changes of these markers were consistent with the WB results, further confirming the key role of SIRT6 in the mechanism of action of Nod (Fig. 6F).

The membrane potential staining (JC-1 staining) results showed that the mitochondrial membrane potential significantly decreased in the model group, whereas the mitochondrial membrane potential was better restored in the Nod treatment group. In the Nod + siSIRT6 group, the mitochondrial membrane potential decreased again, indicating that the silencing of SIRT6 could reverse the protective effect of Nod on mitochondrial function (Fig. 6G).

The SA-β-gal staining results showed that the proportion of SA-β-gal-positive cells was significantly increased in the model group, while the proportion of SA-β-gal-positive cells was significantly reduced in the Nod treatment group. In the Nod + siSIRT6 group, the proportion of SA-β-gal-positive cells increased again, indicating that silencing SIRT6 could reverse the anti-aging effect of Nod (Fig. 6H).

In conclusion, Nod regulates SIRT6 to activate mitochondrial autophagy, improving mitochondrial function, reducing intracellular ROS levels, and inhibiting cellular senescence, thereby slowing down the aging process of nucleus pulposus cells. However, silencing SIRT6 can reverse these protective effects of Nod, further confirming the pivotal role of SIRT6 in the anti-aging effects mediated by Nod.

Nodakenin enhances PINK1/Parkin-associated mitophagy signaling in NPCs

To further delineate the role of SIRT6 in Nodakenin-mediated mitophagy, we combined SIRT6 gain- and loss-of-function approaches, autophagic flux analysis, and a DARTS binding assay. In H₂O₂-treated NPCs (Fig. 7A), H3K9ac/H3K56ac, p62, p16, p21, and p53 were markedly increased, whereas LC3-II/I and PINK1/Parkin were reduced compared with control cells, indicating impaired SIRT6 activity and mitophagy; Nodakenin largely reversed these changes, SIRT6 silencing markedly blunted them, and the SIRT6 agonist MDL-800 mimicked or partially rescued the effects of Nod. In the presence of bafilomycin A1 (Fig. 7B), Nod produced the greatest accumulation of LC3-II and p62, and this increase in autophagic flux was attenuated by SIRT6 knockdown, supporting a SIRT6-associated enhancement of mitophagy rather than passive LC3-II accumulation. Under IL-1β-induced inflammatory stress (Fig. 7C), Nod similarly reduced histone acetylation, restored SIRT6 and PINK1 expression and decreased p62, whereas these effects were abolished by siSIRT6 but reproduced by MDL-800, extending this mechanism to an inflammatory model. Finally, DARTS experiments (Fig. 7D) showed that Nod dose-dependently protected SIRT6 from proteolytic degradation, indicating direct drug–target engagement consistent with the molecular docking results. Together, these data support that Nodakenin enhances mitophagy, accompanied by changes in SIRT6 and PINK1/Parkin-associated markers, and attenuates NPC senescence under both oxidative and inflammatory insults.

Fig. 7.

Nodakenin enhances mitophagy with changes in SIRT6 and PINK1/Parkin-associated markers in NPCs. A Western blots and quantification of H3K9ac, H3K56ac, LC3-I/II, p62, PINK1, Parkin, p16, p21, and p53 in NPCs under the indicated treatments (control, H₂O₂, H₂O₂ + Nod, H₂O₂ + Nod + siSIRT6, H₂O₂ + MDL-800, H₂O₂ + siSIRT6, H₂O₂ + siSIRT6 + MDL-800). B Autophagic flux assay with or without bafilomycin A1 (BafA1) showing LC3-II and p62 accumulation in each group. C Effects of Nod, siSIRT6, and MDL-800 on H3K9ac/H3K56ac, p62, PINK1, and SIRT6 expression in IL-1β–treated NPCs. D DARTS assay demonstrating dose-dependent protection of SIRT6 from proteolytic degradation by Nodakenin. Data are presented as mean ± SD (n = 3); *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

In vivo validation of nodakenin in rat IVDD model

To further verify the therapeutic effect of Nodakenin in vivo, rats were divided into sham, IVDD model, and Nod-treated groups. X-ray imaging showed that the disc height index (DHI) remained high in the sham group but was significantly reduced in the model group, whereas Nod treatment partially restored DHI compared with the model group (Fig. 8A). HE and Masson staining revealed an intact disc architecture with orderly NP cells and well-preserved matrix/collagen fibers in the sham group; in contrast, model discs exhibited marked degeneration, including disorganized NP cells, loss of extracellular matrix, and disrupted collagen structure. These structural abnormalities were alleviated in the Nod group, which showed improved cellular organization and collagen preservation (Fig. 8B).

Fig. 8.

In vivo effects of nodakenin on IVDD in a rat model. A Representative X-ray images and disc height index (DHI) of lumbar discs in sham, IVDD model, and Nod-treated rats. B HE and Masson’s trichrome staining showing disc morphology and collagen organization in the three groups, with quantitative histological scores. C Western blots and quantification of p62, PINK1, and Parkin in nucleus pulposus tissues. D Immunohistochemical staining and semiquantitative analysis of p53, LC3, SIRT6, p16, and p21 in rat discs. Data are presented as mean ± SD (n = 3 for WB; n = 5–6 sections from 5–6 rats per group for IHC); *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

At the molecular level, Western blot analysis of NP tissues demonstrated decreased PINK1 and Parkin and increased p62 in the model group relative to sham, consistent with impaired PINK1/Parkin-mediated mitophagy, while Nod treatment significantly upregulated PINK1 and Parkin and reduced p62 expression (Fig. 8C). Immunohistochemistry further showed that p53, p16, and p21 staining was markedly enhanced in model discs, whereas LC3 and SIRT6 signals were reduced; Nod administration decreased p53, p16, and p21 and restored LC3 and SIRT6 levels (Fig. 8D). Together, these in vivo findings support that Nodakenin attenuates disc degeneration, accompanied by increased SIRT6 and LC3 signals and changes in the levels of PINK1/Parkin and p62.

Nodakenin activates mitophagy in NPCs, and its in vivo protection is reduced by autophagy inhibition

We performed a mitophagy-specific reporter assay and additional pathway activation analyses. As shown in Fig. 9A, pS65-Ub, a canonical readout of PINK1 kinase activity, was altered after H₂O₂ treatment, while Nodakenin significantly increased pS65-Ub levels; this increase was partially attenuated by siSIRT6, and the SIRT6 agonist MDL-800 induced a consistent elevation. In parallel, the mt-Keima reporter assay showed that Nodakenin markedly increased the mt-Keima red/green ratio under H₂O₂-induced stress, whereas siSIRT6 partially reversed this effect; MDL-800 also increased the mt-Keima ratio (Fig. 9B), suggesting enhanced mitochondrial delivery to lysosomes.

Fig. 9.

Mitophagy-related effects of Nodakenin in vitro and in vivo. A Immunoblotting and densitometric analysis of pS65-Ub in NPCs under the indicated treatments. B mt-Keima fluorescence images (green, red, and merged) and quantification of the mt-Keima red/green fluorescence ratio in NPCs across the indicated groups. C Immunoblotting and densitometric analysis of p62 (SQSTM1) in intervertebral disc tissues from Sham, Model, Model + Nodakenin, and Model + Nodakenin + 3-MA groups. D HE staining and Safranin O–Fast Green staining of intervertebral disc tissues from the indicated groups, with corresponding quantitative analyses. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

In vivo, Nodakenin alleviated disc degeneration as assessed by HE staining and Safranin O–Fast Green staining, whereas co-treatment with 3-MA partially reversed these histological improvements (Fig. 9D). Consistently, p62 (SQSTM1) protein levels were examined in the same in vivo samples as an auxiliary indicator of autophagy inhibition (Fig. 9C). Together, these data provide mitophagy-specific functional evidence and additional activation readouts supporting Nodakenin-enhanced PINK1/Parkin-associated mitophagy and reduced protection under autophagy inhibition.

Discussion

IVDD is a leading cause of low back pain, with a complex and not yet fully understood pathogenesis [34]. However, the aging of NPCs is considered a key factor in the onset of IVDD [35]. Recent studies have increasingly suggested that mitochondrial dysfunction leads to an accumulation of reactive oxygen species (ROS) and mitochondrial DNA (mtDNA), which in turn triggers NPCs’ aging and cell death [36]. In this context, the activation of mitochondrial autophagy plays a crucial role by selectively removing damaged mitochondria, ensuring proper mitochondrial function, and maintaining cellular homeostasis [37]. PTEN-induced kinase 1 (PINK1) and Parkin (RBR E3 ubiquitin ligase) are central to regulating mitochondrial autophagy and maintaining mitochondrial integrity. Upon mitochondrial damage or dysfunction, the mitochondrial membrane potential (ΔΨm) decreases, leading to the accumulation of PINK1 on the outer mitochondrial membrane (OMM). This process further recruits Parkin from the cytoplasm to the OMM, where its E3 ligase activity promotes the ubiquitination of mitochondrial proteins, thereby stimulating mitochondrial autophagy and facilitating the degradation of damaged mitochondria. This selective degradation mechanism ensures the removal of dysfunctional mitochondria, thus maintaining cellular health and homeostasis[38, 39]. Despite the pivotal role of mitochondrial autophagy in cellular health, there have been limited studies focusing on its regulatory mechanisms during IVDD. Given the crucial involvement of SIRT6 in both cellular aging and mitochondrial autophagy, we hypothesize that SIRT6 may mediate the PINK1/Parkin pathway, regulate mitochondrial autophagy and thereby delaying NPC senescence in the context of IVDD.

Therefore, our findings support a working model in which Nod may regulate SIRT6, accompanied by changes in PINK1/Parkin-related markers and mitophagy activity, thereby improving mitochondrial function and cellular homeostasis to delay senescence.In our study, we first established an NPC senescence model induced by H₂O₂ to simulate the cellular aging caused by oxidative stress. To explore potential therapeutic strategies to mitigate this phenomenon, we used Nod for intervention. The results showed that treatment with Nod significantly restored autophagic flux and reduced the expression of cellular senescence markers, including p53, p21, and p16, indicating its anti-aging effects. Further investigation revealed that Nod upregulated SIRT6 expression, accompanied by changes in PINK1/Parkin-associated mitophagy markers, reduced mitochondrial ROS production, restored mitochondrial membrane potential, and improved mitochondrial function.To validate the central role of SIRT6 in the anti-aging effects of Nod, we performed siRNA-mediated silencing of SIRT6 gene expression. The results showed that silencing SIRT6 significantly weakened the protective effects of Nod on NPCs, further supporting the critical role of SIRT6 in this process. Additionally, molecular docking studies showed that Nod exhibited a strong predicted binding affinity for SIRT6, providing a putative molecular basis for a direct interaction between Nod and SIRT6 that may underpin the biological effects of Nod. We have added mitophagy-specific functional evidence using the mt-Keima reporter assay and further assessed pS65-Ub as a canonical activation readout of PINK1 kinase activity. Nevertheless, additional approaches (e.g., mitochondrial fractionation to examine Parkin translocation or alternative mitophagy reporters) may further strengthen the mechanistic interpretation in future studies.

In the in vivo experiment, we established a rat IVDD model and administered Nod via gavage for 30 days to observe its effects on intervertebral disc degeneration. The results showed that the Nod treatment group exhibited significantly better disc structure compared with the model group, with a significantly higher DHI index, indicating improvement in both disc height and overall health. Histological analysis (HE staining and Masson staining) and immunohistochemical results (SIRT6, p53, p21, p16, and LC3) further supported the conclusion that Nod ameliorates IVDD progression, accompanied by changes in SIRT6 expression and PINK1/Parkin-associated mitophagy markers, thereby mitigating cellular aging during the IVDD process. In our rat IVDD model, Nodakenin was administered orally. Because the nucleus pulposus is essentially avascular, nutrients and small solutes are believed to reach this tissue mainly by diffusion from capillaries in the cartilaginous endplate and outer annulus, rather than via a direct vascular supply [40–42]. Given that Nodakenin is a relatively small coumarin glycoside (~ 438 Da), we speculate that, after systemic absorption, it may enter these vascularized regions and then diffuse into the NP matrix where it could interact with SIRT6. We did not, however, directly quantify Nodakenin levels in disc tissues, so the present work provides functional and histological evidence of protection rather than a full pharmacokinetic profile. Future studies, for example, ultrahigh-performance liquid chromatography coupled with tandem mass spectrometry (UHPLC–MS/MS)-based quantification comparing systemic versus local delivery, are needed to define its distribution within the disc.

In this study, several lines of evidence support a SIRT6-associated regulation of PINK1/Parkin-related mitophagy markers by Nodakenin in NP cells, including SIRT6 agonist and knockdown experiments, autophagic flux analysis with bafilomycin A1, and DARTS-based drug–target engagement of SIRT6. These findings suggest that SIRT6 functions upstream of PINK1/Parkin in the context of oxidative stress-induced NPC senescence. Nevertheless, we did not perform isothermal titration calorimetry (ITC)/surface plasmon resonance (SPR) or chromatin immunoprecipitation (ChIP)/luciferase reporter assays, and the precise transcriptional regulation of PINK1/Parkin by SIRT6 at the promoter level remains to be elucidated in future work. Finally, we did not perform genetic inhibition of PINK1 or Parkin in vitro or in vivo, so the causal contribution of the PINK1/Parkin arm downstream of SIRT6 to Nodakenin’s protective effects remains to be indicated in future studies using PINK1/Parkin knockdown or knockout models. Finally, we did not perform genetic inhibition of PINK1 or Parkin in vitro or in vivo; therefore, the causal contribution of the PINK1/Parkin arm downstream of SIRT6 to Nodakenin’s protective effects remains to be confirmed. Future studies incorporating targeted inhibition of SIRT6, PINK1, or Parkin, particularly in vivo, will help determine whether the anti-senescent and anti-degenerative effects of Nodakenin are causally dependent on this mitophagy-related axis.

Conclusions

This study investigated the potential mechanisms by which Nodakenin ameliorates intervertebral disc degeneration (IVDD), with a particular focus on its effects on nucleus pulposus cell (NPC) senescence. Our data indicate that Nodakenin attenuates oxidative stress-induced NPC senescence and improves mitochondrial function in a SIRT6-associated manner, accompanied by changes in PINK1/Parkin-associated mitophagy markers. Through both vitro and in vivo experiments, we provide experimental evidence that Nodakenin exerts anti-senescent effects on NPCs and may slow the progression of IVDD. These findings suggest Nodakenin as a promising candidate for the treatment of degenerative disc disease, while further studies are needed to clarify its precise upstream and downstream signaling and to evaluate its long-term efficacy and safety.

Abbreviations

IVDD

Intervertebral disc degeneration

NPCs

Nucleus pulposus cells

ROS

Reactive oxygen species

mtDNA

Mitochondrial DNA

SASP

Senescence-associated secretory phenotype

TCM

Traditional Chinese Medicine

DHI

Disc height index

HE

Hematoxylin–eosin

TEM

Transmission electron microscopy

WB

Western blot

JC-1

5,5′,6,6′-Tetrachloro-1,1′,3,3′-tetraethylbenzimidazolylcarbocyanine iodide (mitochondrial membrane potential dye)

FBS

Fetal bovine serum

RT-PCR

Reverse transcription polymerase chain reaction

siRNA

Small interfering RNA

3-MA

3-Methyladenine (autophagy inhibitor)

LC3

Microtubule-associated protein light chain 3

PINK1

PTEN-induced kinase 1

qRT-PCR

Quantitative reverse transcription polymerase chain reaction

SA-β-gal

Senescence-associated β-galactosidase

BafA1

Bafilomycin A1

MDL-800

SIRT6-specific agonist

DARTS

Drug affinity responsive target stability

DCFH-DA

2′,7′-Dichlorodihydrofluorescein diacetate

MitoSOX

MitoSOX Red mitochondrial superoxide indicator

IL-1β

Interleukin-1β

TNF-α

Tumor necrosis factor-α

FDR

False discovery rate

PCA

Principal component analysis

UMAP

Uniform manifold approximation and projection

GSVA

Gene set variation analysis

pS65-Ub

Ser65-phospho-ubiquitin; mt-Keima (mitophagy reporter)

Author contributions

Conceptualization: ZL; data curation: FY, TW; formal analysis: JL, YM; methodology: LW, TL; visualization: YZ; writing—original draft: DZ; review and editing: CS.

Funding

2025 Traditional Chinese Medicine (TCM) Scientific Research Special Project of Sichuan Provincial Administration of Traditional Chinese Medicine(25ZDIZX029),Science and Technology Strategic Cooperation between Xuyong County People's Hospital and Southwest Medical University(2024) (No. 2024XYXNYD06).

Data availability

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

All animal procedures were approved by the Experimental Animal Ethics Committee of Southwest Medical University (approval no. 20240922–008, 22 September 2024) and were conducted in accordance with the ethical guidelines of the International Council for Laboratory Animal Science (ICLAS).

Competing interests

The authors declare no competing interests.