Identification and Experimental Validation of PANoptosis Key Genes for Constructing a PANoptosis Risk Diagnostic Model in Intervertebral Disc Degeneration

Kui Deng; Xia Lan; WenZhao Chen; LiuXue Du; JiangWei Chen

doi:10.1002/jgm.70115

. 2026 Sep 28;28(10):e70115. doi: 10.1002/jgm.70115

Identification and Experimental Validation of PANoptosis Key Genes for Constructing a PANoptosis Risk Diagnostic Model in Intervertebral Disc Degeneration

Kui Deng ¹, Xia Lan ¹, WenZhao Chen ¹, LiuXue Du ¹, JiangWei Chen ^1,^✉

PMCID: PMC13617583 PMID: 42803180

ABSTRACT

Background

Intervertebral disc degeneration (IDD) represents a significant health concern globally. This study aimed to identify the PANoptosis key genes (PKGs) associated with IDD and construct a risk diagnosis model

Methods

Single‐cell sequencing, machine learning algorithms, and LASSO regression were employed to identify PKGs and develop a risk model. The expression of identified PKGs was validated in human IDD clinical tissues using qRT‐PCR, Western blot, and immunohistochemistry. The therapeutic effect of ACSBG1 knockdown was further evaluated in a rat IDD model by histological analysis.

Results

The seven most representative PKGs of IDD (ACSBG1, FXYD1, APCS, CEACAM1, ERAP2, ABL1, and FZD3) were identified. Utilizing these seven PKGs, an IDD PANoptosis risk diagnosis model was constructed and validated. Immune infiltration analysis revealed a significant association between the PANoptosis risk of IDD and immune cells, particularly neutrophils. Clinical validation confirmed that ACSBG1, FXYD1, APCS, and CEACAM1 were upregulated, while ERAP2, ABL1, and FZD3 were downregulated in nucleus pulposus tissues of IDD patients. In a rat model, ACSBG1 knockdown markedly attenuated disc degeneration, preserving disc structure and increasing proteoglycan content.

Conclusion

This study established an IDD PANoptosis risk diagnosis model and demonstrated that targeting ACSBG1 exerted therapeutic effects in vivo, providing new insights into the diagnosis and treatment of IDD.

Keywords: experimental validation, intervertebral disc degeneration, machine learning algorithm, PANoptosis, risk diagnosis model

This study identified seven PANoptosis key genes (ACSBG1, FXYD1, APCS, CEACAM1, ERAP2, ABL1, and FZD3) in intervertebral disc degeneration (IDD) through integrated bioinformatics analysis of GEO datasets. Using machine learning algorithms (RF, SVM, XGB, and GLM), a PANoptosis risk diagnostic model was constructed and validated. Clinical tissue validation confirmed the differential expression of these seven genes in human nucleus pulposus tissues. Furthermore, in vivo experiments demonstrated that ACSBG1 knockdown effectively attenuated IDD progression in a rat model, highlighting ACSBG1 as a potential therapeutic target for IDD.

graphic file with name JGM-28-e70115-g007.webp

1. Introduction

Intervertebral disc degeneration (IDD) is a prevalent degenerative disease [1]. As the primary cause of chronic low back pain and spine‐related disorders, IDD significantly diminishes patients' quality of life and work efficiency while also increasing the burden on the healthcare system [2]. A substantial body of literature indicates that genetic susceptibility, age, lifestyle factors, and nonphysiological mechanical loads are the principal contributors to IDD [3, 4, 5]. Numerous studies have examined IDD from various perspectives, including degeneration and regeneration, mechanics, and biology [6, 7]. However, a comprehensive understanding of the intrinsic cellular changes associated with IDD remains elusive. Despite the availability of treatments that can mitigate degeneration, restoring IDD to a healthy state continues to present significant challenges [8].

The intervertebral disc comprises the central nucleus pulposus (NP), the peripheral annulus fibrosus, and the cartilage endplates [9]. NP cells represent the predominant cell type within NP tissue, responsible for maintaining the height of the intervertebral disc and absorbing various mechanical loads by synthesizing its primary functional components [10]. Furthermore, the programmed cell death of NP cells constitutes one of the pathophysiological processes associated with IDD [11]. Research has indicated that IDD is closely linked to inflammatory responses [6]. PANoptosis, an inflammatory programmed cell death pathway regulated by the PANoptosome, underscores the interplay and coordination among pyroptosis, apoptosis, and necroptosis [12, 13]. Increasing evidence has suggested that oxidative stress may play a crucial role in the onset and progression of IDD [14]. Oxidative stress is able to induce both pyroptosis and necrosis in cells [15, 16]. Additionally, the apoptosis of NP cells is significant in the context of IDD [17]. Consequently, PANoptosis may be implicated in the pathological processes of IDD. However, current research on the biological mechanisms associated with PANoptosis in the diagnosis and treatment of IDD remains limited.

Bioinformatics is centered on the research of biomedical data and has increasingly been applied to investigate the molecular mechanisms of diseases and to develop personalized treatment strategies [18]. The advancement of high‐throughput sequencing and microarray technologies has enabled the application of bioinformatics in analyzing differentially expressed mRNA in specific diseases and in predicting potential therapeutic targets, thereby playing a crucial role in life science research [19]. Moreover, the evolution of bioinformatics has facilitated the development of data processing and analysis techniques, including machine learning algorithms, LASSO regression analysis, and the construction of nomograms [20, 21, 22]. These methodologies offer significant insights into the understanding of disease mechanisms and the identification of biomarkers. However, bioinformatics predictions alone require experimental validation to confirm their clinical relevance and therapeutic potential [23, 24].

In this study, bioinformatics methods and machine learning were employed to analyze the Gene Expression Omnibus (GEO) dataset [25]. We identified PANoptosis key genes (PKGs) in IDD and constructed a diagnostic model to assess the risk of PANoptosis, followed by pathway analysis, immune infiltration analysis, and drug prediction. To validate our findings, the expression of the seven identified PKGs was examined in human IDD clinical tissues. Based on its significant upregulation, ACSBG1 was selected for in vivo functional investigation in a rat IDD model. These integrated analyses provide valuable insights into IDD pathogenesis and highlight potential therapeutic targets.

2. Materials and Methods

2.1. Data Acquisition

The single‐cell data of IDD patients and healthy volunteers came from the GSE205535 dataset, which analyzed two human specimens, including one normal NP tissue and one degenerated NP tissue. The transcriptome data was sourced from the GSE124272 dataset, which included eight patients with lumbar disc herniation and eight healthy volunteers. These datasets were derived from GEO (https://www.ncbi.nlm.nih.gov/geo/).

2.2. Single‐Cell Data Analysis

Single‐cell data (GSE205535) was subjected to quality control, dimensionality reduction, and clustering analysis using the Seurat package in R. First, filtering was performed based on the following criteria: genes expressed in at least 3 cells and cells expressing at least 200 genes; cells with nFeature_RNA > 300, nCount_RNA > 500, nFeature_RNA < 8000, nCount_RNA < 80,000, and Percent_mt < 20 were retained to generate a high‐quality count matrix. The data were normalized and then sample integration was performed using harmony for the two single‐cell samples. The ScaleData function was used to scale all genes, while principal component analysis was performed for dimensionality reduction to identify anchors. Dim = 30 was selected, and cells were clustered using the FindNeighbors and FindClusters functions (resolution = 0.5). TSNE and UMAP dimensionality reduction were performed using the RunTSNE and RunUMAP functions, respectively.

2.3. Annotation of Cell Types

Using the FindAllMarkers function from the Seurat package, we identified genes that were specifically expressed in each cell cluster after clustering. Then, the CellMarker2.0 website was utilized to identify cell types based on marker genes and perform a preliminary classification of cell types. A comprehensive panel of well‐established IVD cell markers was employed for annotation, including COL2A1, KRT18, and TBXT for NP cells; COL1A1 for annulus fibrosus cells; SOX9, ACAN, RUNX2, and CNMD for chondrocytes; CLDN5, PECAM1 for endothelial cells; CD3D for T cells; CD68 for macrophages; and CSFR3 for neutrophils. Of note, all cell type annotations in this study were independently performed by us based on canonical marker gene expression patterns, rather than adopting the original annotations from the GSE205535 dataset. The four NP cell subpopulations were annotated as effector NP cells (EffectorNP, NP_3), homeostatic NP cells (HomNP, NP_1), NP progenitor cells (NPPC, NP_2), and fibrotic NP cells (FibroNP, NP_4) based on their specific marker gene expression patterns.

2.4. Calculation of PANoptosis Score

PANoptosis‐related genes (PRGs) were downloaded from the GeneCard website as a PANoptosis gene set. PANoptosis scores were calculated for each cell using the AUCell_calcAUC function of the AUCell package. Subsequently, the ssGSVA score of the gene set in each cell was calculated using the ssGSVA package, with higher scores indicating that the gene set was more enriched in that class of cells. The UCell, AUCell, and ssGSEA scores were calculated using the escape.matrix function. AddModuleScore was also applied to quantify PANoptosis pathway activity.

2.5. Differential Analysis of Transcriptome Data

By searching the GEO database, a transcriptome dataset of NP tissues from healthy humans and IDD patients (GSE124272) was found. The limma package was employed to filter differentially expressed genes (DEGs) between the healthy and IDD groups. Genes with p value < 0.05 and |logFC| > 1 were considered significantly different. A volcano plot was created to display the screened out DEGs. To further visualize the expression of DEGs in different groups, significantly different DEGs were used to generate a heatmap using the R package ComplexHeatmap.

2.6. Weighted Gene Co‐Expression Network Analysis

To elucidate the correlation of gene expression among the integrated cells, the weighted gene co‐expression network (WGCN) analysis R package was employed to construct a WGCN. WGCN with scale‐free topology was constructed and different soft threshold power β values were evaluated for network topology analysis, selecting the value as 4. Pearson's correlation coefficient and the signed network option were utilized separately to measure the correlation between the expression of each pair of genes, maintaining only positive correlations. The topological overlap measure (TOM) was applied to identify modules, which examined the similarity between gene pairs based on the number of shared neighbors within the constructed co‐expression network. Gene modules within the WGCN were represented in distinct colors. Genes lacking similar co‐expression with other genes in the network were assigned to the grey module. Thus, grey modules had been removed from more analyses. Based on single‐cell sequencing data, the expression profiles of characteristic genes of different types of cells were investigated to identify their biomarker genes. Furthermore, core module genes associated with different cell types were identified through the WGCN analysis. The GSVA and ssGSEA‐derived PANoptosis scores were used as phenotypic traits for module‐trait correlation analysis.

2.7. Machine Learning Model Construction

Machine learning analysis was performed on the feature genes obtained from the initial screening. First, the dataset was randomly divided into a training set and a validation set in a 3:1 ratio. Then, diagnostic models for gene expression data were constructed using support vector machine (SVM), random forest (RF), XGBoost (XGB), and generalized linear model (GLM) algorithms, along with the “kernlab,” “randomForest,” and “xgboost” R packages. The “caret” and “DALEX” R packages were used to optimize the model construction process and respectively to explain the relationships between input and output variables in the model. SVM is an algorithm that classifies data by constructing hyperplanes, which also can eliminate overfitting in the model using regularization terms. RF is a data‐driven learning algorithm that is able to automatically manage nonlinear effects and interactions between variables. The XGB algorithm is an optimized model that combines linear models and boosting tree models. The GLM algorithm is developed to overcome the shortcomings of linear regression models and is a generalization of linear regression models. The intersection of the best genes obtained from the above four machine learning methods resulted in 15 key genes. Least absolute shrinkage and selection operator (LASSO) analysis was performed on the key genes to select modeling features, which was implemented using the glmnet package in R. Ultimately, seven genes were obtained and defined as IDD marker genes. Then, the area under curve (AUC) and accuracy of receiver operating characteristic curve (ROC) were calculated to assess the classification ability of each model. Finally, the algorithm with the highest AUC and the least residuals was selected to construct the final model. The residual value was defined as the difference between the actual observed values and the model estimates. A smaller residual value indicated that the estimated values generated by the model closely matched the actual data.

2.8. PANoptosis Risk Score Calculation

The PANoptosis risk score was calculated using the LASSO regression coefficients. The risk score was computed based on the following formula: Risk Score = FZD3 × (−0.890797288) + ABL1 × (−0.301294966) + ERAP2 × (−0.012737117) + CEACAM1 × 0.065123375 + APCS × 0.16874763 + FXYD1 × 0.18920502 + ACSBG1 × 0.729271144. The transcriptome samples of NP tissue were divided into high‐risk and low‐risk groups based on the median of the risk score.

2.9. Gene Set Enrichment Analysis

Gene set enrichment analysis (GSEA) is a method used for analyzing genome‐wide expression profile microarray data. It identifies functional enrichment by comparing genes with predefined gene sets. A gene set is a group of genes that share a location, pathway, function, or other characteristics. GSEA was performed using the clusterProfiler software package. The gene expression fold changes between the high‐risk and low‐risk groups were calculated, and then a gene list was generated based on the changes in |log2FC|. Finally, GO enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis based on GSEA were conducted.

2.10. Construction of the Nomogram and Evaluation of Diagnostic Effectiveness

To predict the incidence of IDD, logistic regression analysis was utilized to construct a diagnostic model, which was then visualized using a nomogram. Subsequently, the ROC curve was plotted, followed by calculating the AUC value using the “pROC” R package. The diagnostic effectiveness was evaluated using AUC, calibration curves, and decision curve analysis (DCA). Finally, the differential expression of biomarkers and predictive reliability were further confirmed using a validation set.

2.11. Immune Infiltration Analysis

Using the CIBERSORT algorithm, the immune infiltration status of the NP tissue samples was calculated and the expression of 22 immune cell subpopulations between the IDD group and the control group samples was specifically compared. Furthermore, the correlation between potential biomarkers and immune cells was analyzed.

2.12. Clinical Tissue Specimens

Human NP tissues were obtained from IDD patients at The First Affiliated Hospital of Nanchang University. To ensure tissue specificity, the annulus fibrosus was first incised with a scalpel, and then the central deep NP tissue was carefully extracted using NP forceps. For patients with sequestered disc herniation, only the extruded NP tissue was removed, confirming its NP origin. Normal human NP tissues were obtained from adolescent patients with lumbar vertebral fractures who underwent surgical excision of intervertebral discs for interbody fusion. A total of six NP tissue samples were collected from each group. Written informed consent was obtained from all patients or their relatives, and the institutional ethical committee of The First Affiliated Hospital of Nanchang University hospital (2024)CDYFYYLK(08‐085) approved the study in accordance with the ethical guidelines from the Declaration of Helsinki.

2.13. Animals

All animal experiments were approved by the Animal Ethics Committee of The First Affiliated Hospital of Nanchang University (CDYFYIACUC‐202308QR035). Male Sprague–Dawley rats (8‐week‐old, weighing 200–250 g) were purchased from SiPeiFu Biotechnology (Beijing, China). All animals were housed under specific pathogen‐free conditions with a 12‐h light/dark cycle and free access to food and water.

2.14. IDD Model Establishment

The rat IDD model was established as previously described [26, 27]. Briefly, following isoflurane anesthesia, a midline abdominal incision was made to expose the lumbar spine. IDD was induced by inserting a 21‐gauge needle into the L4‐L5 intervertebral disc at approximately 5 mm depth for 1 min. Rats were randomly divided into two groups (n = 6 per group): IDD+sh‐NC and IDD+sh‐ACSBG1. Immediately after IDD induction, rats received intradiscal injections of empty lentivirus or lentivirus carrying sh‐ACSBG1 (4 μL, 2 × 10⁷ infectious units) at four sites around the injury site. Injections were performed at 1 μL/min using a microsyringe, with the needle retained for 2 min to prevent backflow. All rats were euthanized 12 weeks after surgery, and intervertebral disc tissues were harvested for analysis.

2.15. Histological Analysis

For histological evaluation, rat intervertebral disc tissues were fixed in 4% paraformaldehyde (#G1101, Servicebio, Wuhan, Hubei, China), decalcified in EDTA decalcification solution (#G1105‐500ML, Servicebio) at a volume 15–20 times that of the tissue, with the solution refreshed every 2–3 days until the tissue became soft and could be easily punctured with a pin, dehydrated through graded ethanol, and embedded in paraffin. Serial sections (4 μm) were prepared. For hematoxylin and eosin (H&E) staining, sections were deparaffinized using dewaxing solution, rehydrated through graded ethanol, and stained with hematoxylin (#G1004, Servicebio) for 5 min and eosin (#G1001, Servicebio) for 2 min. Sections were then dehydrated, cleared, and mounted (#WG10004160, Servicebio). Safranin O‐fast green staining was performed using a commercial kit (#G1053‐100ML, Servicebio). Briefly, deparaffinized sections were stained with fast green for 1–5 min, differentiated in 1% hydrochloric acid‐alcohol, and then stained with safranin O for 2–5 min. After rapid dehydration and clearing, sections were mounted with neutral balsam. All stained sections were examined under a light microscope (Leica, Wetzlar, Germany).

2.16. Immunohistochemistry

Immunohistochemistry (IHC) staining was performed to detect ACSBG1, FZD3, and ABL1 protein expression in human NP tissues. Paraffin‐embedded sections were deparaffinized (#G1128‐100ML, Servicebio), rehydrated through graded ethanol, and subjected to antigen retrieval (#G1203, Solarbio, Beijing, China) in a microwave for 10 min. After washing with PBS (#G0002‐2L, Servicebio), endogenous peroxidase was blocked with 3% H₂O₂ for 10 min. Sections were incubated overnight at 4°C with primary antibodies: ACSBG1 (#16077‐1‐AP, 1:100; Proteintech, Wuhan, Hubei, China), FZD3 (#13865‐1‐AP, 1:100; Proteintech), and ABL1 (#M00133‐2, 1:100; Boster, Wuhan, Hubei, China). Following PBS washes, sections were incubated with secondary antibody (#PV‐6000, ZSGB‐Bio, Beijing, China) for 20 min at 37°C. Immunoreactivity was visualized using a DAB substrate kit (#P0202, Beyotime, Shanghai, China) under microscopic monitoring. Sections were counterstained with hematoxylin, differentiated, dehydrated, cleared, and mounted (#WG10004160, Servicebio). Images were captured using a light microscope.

2.17. Quantitative Real‐Time Reverse Transcription PCR

Total RNA was extracted from tissues using RNA extraction reagent (#G3013, Servicebio). RNA concentration and purity were measured using a microspectrophotometer. Reverse transcription was performed using the PrimeScript RT reagent Kit (#RR037Q, Takara, Shiga, Japan) according to the manufacturer's protocol. The reaction conditions were 50°C for 15 min and 85°C for 5 s. qPCR was conducted using TB Green Premix Ex Taq II (#RR820Q, Takara) on a SLAN‐96S system (Hongshi, Shanghai, China). The 20 μL reaction mixture included TB Green Premix, primers, cDNA, and ddH₂O. Thermal cycling conditions were as follows: 95°C for 3 min, followed by 40 cycles of 95°C for 5 s and 60°C for 20 s. GAPDH served as an internal control. Relative mRNA expression was calculated using the 2^‐ΔΔCT method. Primer sequences are listed in Table 1.

TABLE 1.

Primers of qRT‐PCR.

Gene	Forward sequence (5′‐3′)	Reverse sequence (5′‐3′)
ACSBG1 Mouse	CCTCAACATCCGCCTGTATGCA	CTGCATCCTGATTCACCAGCTTC
ACSBG1 Human	AAGGGCGTGATGCTGAGTCAAG	ACTACCACCTCCTGCTGGACTT
FXYD1 Human	AAGTCCAAAGGAACACGACCCG	TGCTGCTGGTTGAACTTGCACC
APCS Human	AGGCTGCATGAGAGCACTTGTG	CACACTTCCAACTTCTCGCAACG
CEACAM1 Human	CACGCCAATAACTCAGTCACTGG	TTGTGGAGCAGGTCAGGTTCAC
ERAP2 Human	CTGTGACCTGAACCATGCTCCT	TCCATCCTGCTGTTGTCTGAGC
ABL1 Human	CCAGGTGTATGAGCTGCTAGAG	GTCAGAGGGATTCCACTGCCAA
FZD3 Human	GGCTCTCATAGTTGGCATTCCC	TGGAGTACCTGTCGGCTCTCAT
GAPDH Mouse	CATCACTGCCACCCAGAAGACTG	ATGCCAGTGAGCTTCCCGTTCAG
GAPDH Human	GTCTCCTCTGACTTCAACAGCG	ACCACCCTGTTGCTGTAGCCAA

Open in a new tab

2.18. Western Blot Analysis

Total protein was extracted from tissues using RIPA lysis buffer (#BL504A, Bio sharp, Hefei, Anhui, China) containing protease inhibitors (#A32955, Thermo Fisher Scientific, Waltham, MA, USA). Protein concentrations were determined using a BCA protein assay kit (#BL521A, Bio sharp). Samples were mixed with 5× SDS loading buffer (#P0015L, Beyotime) and denatured at 95°C for 5 min. Equal protein amounts were separated by SDS‐PAGE and transferred onto PVDF membranes (#IPVH00010, Millipore, Burlington, MA, USA). Membranes were blocked with 5% skim milk for 2 h and incubated overnight at 4°C with primary antibodies: ACSBG1 (#GTX84961, 1:1000; GeneTex, Irvine, CA, USA), FZD3 (#ab217032, 1:1000; Abcam, Cambridge, UK), ABL1 (#68254‐1‐Ig, 1:1000; Proteintech), and GAPDH (#ab8245, 1:1000; Abcam). After washing, membranes were incubated with HRP‐conjugated secondary antibodies (#ab6789 or #ab6721, 1:5000; Abcam) for 2 h. Protein bands were visualized using chemiluminescent substrate (#WBKLS0100, Millipore) and imaged with a MiniChemi 910 system (#SAGECREATION, Beijing, China).

2.19. Statistical Analysis

Single‐cell data were processed using R packages as described above, with quality control filters of nFeature_RNA > 300, nCount_RNA > 500, nFeature_RNA < 8000, nCount_RNA < 80,000, and Percent_mt < 20, followed by normalization and batch correction. WGCNA utilized GSVA/ssGSEA‐derived PANoptosis scores as phenotypic traits. For quantitative data, outliers were identified by Grubbs test (α = 0.05) and boxplot (1.5 × IQR) criteria (none detected). Quantitative real‐time reverse transcription PCR (qRT‐PCR) data were normalized by the 2^‐ΔΔCt method with GAPDH, and Western blot intensities by internal control proteins. Data normality and variance homogeneity were verified by Shapiro–Wilk and Levene's tests, respectively; nonparametric tests or log‐transformation were applied when assumptions were violated. Results are presented as mean ± SD. Sample sizes: n = 3 per group for qRT‐PCR and Western blot, n = 6 per group for histological analyses. All tests were two‐sided with α = 0.05. Two‐group comparisons used Student's t‐test; multigroup comparisons used one‐way ANOVA with Dunnett's post hoc correction. All statistical analyses and graphical presentations were performed using GraphPad Prism 10.5 (La Jolla, CA, USA) and R software (Version 4.2.0). p < 0.05 was considered as a significant difference.

3. Results

3.1. Classification, Function and Differentiation Direction of NP Tissue Cells Were Determined

Single‐cell data for NP tissue samples from healthy volunteers and IDD patients were obtained from the GEO database (GSE205535). Raw single‐cell expression data were filtered via the Seurat R package with strict thresholds: \(300 < nFeature\_RNA < 8000\), \(500 < nCount\_RNA < 80,000\), and \(Percent\_mt < 20\). After normalization, t‐SNE and UMAP dimensionality reduction were performed to visualize all cell clusters (Figure 1A). Canonical marker genes were applied to annotate all cell populations, including NP cells, chondrocytes, AF cells, endothelial cells, and immune cells (T cells, macrophages, neutrophils) (Figure 1B). The UMAP feature plot further exhibited the spatial distribution of all annotated cell subtypes in NP tissues (Figure 1C). Distribution analysis between normal NP and degenerative DP groups showed that HomNP (NP_1) and EffectorNP (NP_3) were predominantly enriched in the DP group, while NPPC (NP_2) and FibroNP (NP_4) were more abundant in the normal NP group (Figure 1D). We further subdivided NP cells into four transcriptionally distinct subclusters NP_1, NP_2, NP_3, and NP_4. Based on representative marker gene expression profiles, we functionally annotated these four NP subpopulations: NP_1 was defined as homeostatic NP cells (HomNP), NP_2 as NP progenitor cells (NPPC), NP_3 as effector NP cells (EffectorNP), and NP_4 as fibrotic NP cells (FibroNP). A dot plot summarized the signature marker genes across the four NP subclusters to validate subpopulation classification (Figure 1E).

Cell type classification and annotation of human NP tissues based on single‐cell sequencing data. (A) UMAP plot showing dimensionality reduction and clustering of all cells from normal and IDD NP tissues after quality control and normalization. (B) Dot plot illustrating the expression of canonical marker genes used for annotating major cell types, including NP cells, chondrocytes, annulus fibrosus cells, endothelial cells, and immune cells (T cells, macrophages, and neutrophils). (C) UMAP feature plot displaying the spatial distribution of all annotated cell subtypes in NP tissues. (D) Bar chart showing the quantitative proportions of each cell type in normal NP and degenerative DP groups. (E) Dot plot summarizing the signature marker genes across four NP subpopulations (HomNP/NP_1, NPPC/NP_2, EffectorNP/NP_3, and FibroNP/NP_4). The single‐cell sequencing data were derived from one normal and one degenerated NP tissue sample (n = 1 per group).

3.2. EffectorNP (NP_3) Cell Was Identified as PANoptosis Related Cell Type in IDD

We quantified panoptosis pathway activity at single‐cell resolution using four independent scoring algorithms: AddModuleScore, AUCell, UCell, and ssGSEA. The UMAP feature plot showed heterogeneous panoptosis signal distribution across all single cells (Figure 2A). A heatmap integrating all four scoring metrics consistently demonstrated that NP_3 (EffectorNP) harbored the highest panoptosis signature among all NP subpopulations (Figure 2B). AUCell‐derived panoptosis scores were compared between normal NP and degenerative DP groups. Boxplot analysis of all captured cells confirmed significantly elevated panoptosis levels in IDD samples (Figure 2C). Subcluster‐specific comparison further verified that panoptosis activity was markedly upregulated in the NP_3 effector NP subpopulation of degenerative tissues (Figure 2D). Bulk transcriptomic differential expression analysis between control and IDD groups identified thousands of dysregulated genes, which were visualized via a centered‐scaled DEG heatmap (Figure 2E). Consistently, both GSVA and ssGSEA scoring pipelines verified robust panoptosis signature enrichment in the DP/IDD group (Figure 2F). These findings suggested a close relationship between the occurrence of IDD and PANoptosis. Based on these results, NP_3 (EffectorNP) was identified as the PANoptosis‐related cell type in IDD, and subsequent studies were conducted utilizing this cell type.

Identification of EffectorNP (NP_3) as the PANoptosis‐related cell type in IDD. (A) UMAP feature plot showing heterogeneous PANoptosis signal distribution across all single cells, with cells from degenerative DP samples exhibiting stronger PANoptosis signals. (B) Heatmap integrating four independent scoring algorithms (AddModuleScore, AUCell, UCell, and ssGSEA) demonstrating that NP_3 (EffectorNP) harbored the highest PANoptosis signature among all NP subpopulations. (C) Box plot of AUCell‐derived PANoptosis scores comparing all captured cells between normal NP and degenerative DP groups. (D) Box plot of AUCell‐derived PANoptosis scores specifically in NP_3 effector NP cells between normal and degenerative groups. (E) Centered‐scaled heatmap of differentially expressed genes between control and IDD groups from bulk transcriptomic analysis. (F) Box plots of GSVA and ssGSEA scores confirming robust PANoptosis signature enrichment in the DP/IDD group. The single‐cell sequencing data were derived from one normal and one degenerated NP tissue sample (n = 1 per group).

3.3. Identification of PKGs for IDD

Using WGCNA analysis to construct a scale‐free network, the optimal soft threshold was identified as 4, followed by illustrating the gene modules that exhibited a strong association with PANoptosis in the transcriptome data (Figure 3A). Furthermore, correlation analysis and TOM evaluations revealed 24 gene modules significantly associated with PANoptosis, with varying correlations among these modules and different cell types in NP tissue (Figure 3B,C). We performed module‐trait correlation analysis by matching each module eigengene with panoptosis phenotypic scores calculated by GSVA and ssGSEA (Figure 3D). Among all identified co‐expression modules, the turquoise module exhibited the strongest positive correlation with panoptosis signatures, so we selected the turquoise module as the core gene set associated with panoptosis in IDD. Differential expression analysis revealed 149 down‐regulated DEGs and 128 up‐regulated DEGs in IDD (Figure 3E). By intersecting the DEGs associated with IDD and the turquoise gene module, 51 IDD‐related PKGs were identified as subjects for subsequent machine learning investigations (Figure 3F).

Identification of PANoptosis key genes (PKGs) for IDD through WGCNA and Venn diagram analysis. (A) The optimal soft threshold parameter was determined based on WGCN analysis of the GSE124272 dataset. (B) Gene modules closely associated with PANoptosis were screened through correlation analysis, with different colors representing distinct gene modules. (C) A TOM heatmap was utilized to depict the correlation between various cell types in NP tissue and PANoptosis‐related gene modules. (D) Module‐trait correlation analysis showing the relationship between each module eigengene and PANoptosis phenotypic scores calculated by GSVA and ssGSEA; the turquoise module exhibited the strongest positive correlation with PANoptosis. (E) A volcano plot was presented to display the results of differential expression analysis of IDD based on the transcriptome dataset GSE124272. (F) The intersection of IDD DEGs and the module genes most closely associated with HTNP cells was illustrated through a Venn diagram. The single‐cell sequencing data (GSE205535) were derived from one normal and one degenerated NP tissue sample (n = 1 per group); transcriptome data (GSE124272) included eight healthy controls and eight IDD patients.

3.4. Construction of IDD PANoptosis Risk Model and Calculation of Risk Score

Based on the PKGs screened above, IDD patients were distinguished from healthy volunteers utilizing the following four machine learning methods: RF, SVM, GLM, and XGB. The results from the residual boxplot indicated that the RF machine learning method exhibited the lowest residual value (Figure 4A). Based on importance, various machine learning models were employed to identify the top 8 PKGs associated with IDD (Figure 4B). Moreover, five rounds of cross‐validation confirmed that the AUC values of the ROC curves for all four machine learning models were equal to 1, demonstrating that these models possessed strong predictive capabilities (Figure 4C). Subsequently, the best gene sets selected by the four machine learning methods to distinguish IDD from healthy individuals were intersected, resulting in the identification of 15 PKGs associated with IDD (Figure 4D). LASSO regression analysis revealed the presence of seven key IDD PKGs, constructing IDD PANoptosis risk model (Figure 4E). Among these, the genes ACSBG1, FXYD1, APCS, and CEACAM1 were found to be up‐regulated in IDD patients, while ERAP2, ABL1, and FZD3 genes were down‐regulated. Finally, a histogram of the coefficient values for the seven PKGs was generated to calculate the IDD PANoptosis risk score (Figure 4F).

Identification of key PKGs for IDD based on machine learning algorithms. (A) A box plot of residuals for four machine learning methods was presented based on PKGs. (B) Characteristic genes were ranked according to their importance as selected by the four machine learning models, utilizing root mean square error. (C) Five‐fold cross‐validation was employed to calculate the AUC values for various machine learning models. (D) A Venn diagram illustrated the intersection of the best gene sets selected by the four machine learning methods to distinguish between individuals with IDD and healthy controls. (E) Lasso regression analysis was applied to identify key PKGs associated with IDD. (F) The IDD PANoptosis risk score was calculated using a histogram of the coefficients for the seven identified PKGs. The single‐cell sequencing data (GSE205535) were derived from one normal and one degenerated NP tissue sample (n = 1 per group); transcriptome data (GSE124272) included eight healthy controls and eight IDD patients.

3.5. The Construction and Validation of IDD PANoptosis Risk Diagnostic Model

To validate the IDD PANoptosis risk model developed using four machine learning methods, ROC analysis was conducted on two additional datasets (GSE150408 and GSE34095) and the GSE124272 dataset from GEO database. The analysis results indicated that the AUC values for the aforementioned IDD PANoptosis risk model exceeded 0.7 when applied to these three datasets, demonstrating a strong classification performance of this risk model (Figure 5A–C). Besides, a nomogram based on seven key IDD‐related PKGs was successfully developed to predict the PANoptosis risk of IDD (Figure 5D). Specifically, each influencing factor was assigned a score based on the magnitude of the regression coefficient for each gene in the IDD PANoptosis risk diagnostic model. The individual scores were then summed to derive an overall score. Ultimately, the predicted PANoptosis risk value for IDD was calculated through a functional conversion relationship between the total score and the probability of outcome events. Calibration curve results indicated that the predicted values obtained from the nomogram closely aligned with the actual values, thus verifying the accuracy of the IDD PANoptosis risk diagnostic model (Figure 5E). Additionally, DCA revealed that the risk diagnostic model curve was significantly distanced from the two extreme curves (Figure 5F). This suggested a relatively wide optional probability threshold range for the model curve, indicating that the IDD PANoptosis risk diagnostic model was relatively safe. Collectively, these findings affirmed the high accuracy of the constructed IDD machine learning model.

Construction and validation of nomogram of PKGs based on 7 key IDDs. (A–C) The ROC curves of the GSE150408, GSE34095, and GSE124272 datasets applied to the IDD PANoptosis risk model, respectively. (D) A nomogram was presented for diagnosing the PANoptosis risk of IDD. (E) The accuracy of the nomogram in predicting IDD PANoptosis risk was assessed by a calibration curve. (F) A DCA plot was utilized to verify the safety and accuracy of the IDD PANoptosis risk diagnostic model. The single‐cell sequencing data (GSE205535) were derived from one normal and one degenerated NP tissue sample (n = 1 per group).

3.6. IDD‐Related Pathway Analysis and Drug Prediction Based on IDD PANoptosis Risk Model

The risk coefficient was calculated using the genes and their coefficients obtained from the IDD PANoptosis risk model. Based on the median risk coefficient, the samples were categorized into high‐risk and low‐risk groups. Differential expression analysis was then performed to identify DEGs in both groups (Figure 6A). Furthermore, the results of the GSEA analysis of the DEGs indicated that the upregulated genes in the high‐risk group were primarily enriched in signaling pathways associated with cell proliferation, survival, cellular immune response, and inflammasome activity (Figure 6B). These pathways are closely linked to the pathogenesis of IDD and cell PANoptosis. Notably, KEGG_MEDICUS_REFERENCE_NLRP3_INFLAMMASOME_SIGNALING_PATHWAY was specifically related to the NLRP3 inflammasome. Additionally, potential therapeutic agents for IDD were predicted using the CMap database based on upregulated genes in the high‐risk group (Figure 6C). A lower drug score indicated a stronger antagonistic effect on the upregulated genes, offering potential clues for future drug discovery.

GSEA analysis and drug prediction based on IDD PANoptosis high‐risk DEGs. (A) The results of differential expression analysis of PANoptosis high‐risk and low‐risk IDD groups were displayed through a heatmap of DEGs. (B) GSEA analysis, based on the identified DEGs, was conducted to observe pathways related to DEGs in the PANoptosis high‐risk IDD group. (C) Drug prediction for IDD was based on up‐regulated genes in the PANoptosis high‐risk group. The single‐cell sequencing data (GSE205535) were derived from one normal and one degenerated NP tissue sample (n = 1 per group).

3.7. The PANoptosis Risk of IDD Was Related to Immune Cell Infiltration

The results of the immune infiltration analysis indicated that neutrophils were the predominant component of immune cells in both the IDD high‐risk and low‐risk groups (Figure 7A). Moreover, the proportion of neutrophils in the IDD high‐risk group was significantly higher than in the IDD low‐risk group, suggesting that the immune microenvironment of IDD was primarily characterized by neutrophil infiltration (Figure 7B). Subsequently, the expression levels of immune suppression‐related genes, immune activation‐related genes, major histocompatibility complex (MHC)–related genes, and immune checkpoint‐related genes were analyzed in both the IDD high‐risk and low‐risk groups (Figure 7C–F). Notably, the IDD high‐risk group exhibited a significant increase in three immune activation‐related genes (TLR4, TNF, and ENTPD1), a significant decrease in two immune suppression‐related genes (SLAMF7 and VEGFB), a significant increase in one MHC‐related gene (MICA), and no significant changes in immune checkpoint‐related genes. These findings underscored the close relationship between the PANoptosis risk of IDD and immune cells, particularly neutrophils.

Immune infiltration analysis of IDD PANoptosis high‐risk and low‐risk samples. (A) A bar chart was utilized to display the types and proportions of immune cells in the PANoptosis high‐risk and low‐risk IDD groups. (B) A box plot illustrated the immune cell proportions, highlighting significant differences between the PANoptosis high‐risk and low‐risk groups. (C–F) Box plots were employed to evaluate the expression of immune suppression‐related genes, immune activation‐related genes, MHC‐related genes, and immune checkpoint‐related genes across the PANoptosis high‐risk and low‐risk groups of IDD. The single‐cell sequencing data (GSE205535) were derived from one normal and one degenerated NP tissue sample (n = 1 per group).

3.8. Validation of the Seven PKGs in Human IDD Clinical Tissues

To validate the expression patterns of the seven identified PKGs, we examined their levels in human NP tissues from Normal and IDD groups. First, radiographic assessment using DR and CT imaging confirmed the successful grouping, as the IDD group exhibited endplate sclerosis and reduced disc height compared to the Normal group (Figure 8A,B). Quantitative analysis using the Pfirrmann grading system and disc height index (DHI) further confirmed that the IDD group had significantly higher Pfirrmann grades and lower DHI values compared to the Normal group (Figure 8C). qRT‐PCR analysis revealed that the mRNA expression levels of ACSBG1, FXYD1, APCS, and CEACAM1 were significantly upregulated in the IDD group relative to the normal group (Figure 8D–G). Conversely, the mRNA levels of ERAP2, ABL1, and FZD3 were significantly downregulated in the IDD group (Figure 8H–J). These trends were consistent with the bioinformatics predictions. At the protein level, Western blot analysis confirmed that ACSBG1 protein expression was markedly increased, while FZD3 and ABL1 protein levels were significantly decreased in NP tissues from IDD group, consistent with their mRNA expression patterns (Figure 8K–M). Furthermore, IHC staining corroborated these findings, demonstrating stronger positive staining for ACSBG1 and weaker staining for FZD3 and ABL1 in the IDD group compared to the Normal group (Figure 8N). Collectively, these results validated the differential expression of the seven PKGs in IDD, implicating them in the pathological progression of the disease.

The expression of PKGs was altered in NP tissues from human IDD patients. Human NP tissues were collected from Normal and IDD groups. (A,B) Representative DR and CT images of cartilage endplates showing degenerative changes in the IDD group (n = 6 per group). (C) Pfirrmann grading system was used to assess the degree of cartilage endplate degeneration, and DHI was measured to quantify disc height loss (n = 6 per group). (D–J) qRT‐PCR analysis of ACSBG1, FXYD1, APCS, CEACAM1, ERAP2, ABL1, and FZD3 mRNA expression levels in NP tissues (n = 3 per group). (K,M) Western blot analysis and quantification of ACSBG1, FZD3, and ABL1 protein expression (n = 3 per group). (N) IHC staining of ACSBG1, FZD3, and ABL1 in NP tissues, with PBS replacing the primary antibody as a negative control (n = 6 per group). Data are presented as mean ± SD Comparisons between two groups were performed using two‐tailed independent samples Student's t‐test. *p < 0.05, **p < 0.01, ***p < 0.001 versus normal.

3.9. Knockdown of ACSBG1 Alleviated IDD Progression in a Rat Model

Based on its significant upregulation in human degenerative tissues, ACSBG1 was selected for further in vivo functional investigation using a rat IDD model. The knockdown efficiency of ACSBG1‐targeting shRNAs was validated, showing that sh‐ACSBG1‐1 and sh‐ACSBG1‐2 significantly reduced ACSBG1 mRNA and protein levels in rat intervertebral disc tissue compared to the sh‐NC group (Figure 9A,B). Western blot analysis confirmed that ACSBG1 knockdown significantly reduced ACSBG1 protein expression in intervertebral disc tissue of the IDD rat model (Figure 9C). Histological analysis using H&E staining revealed that compared with the IDD+sh‐NC group, the IDD+sh‐ACSBG1 group exhibited preserved disc structure with normal thickness, a plump NP rich in extracellular matrix, and uniformly distributed chondrocyte‐like cells, with significantly lower histological degeneration grading scores (Figure 9D,E). Safranin O‐fast green staining confirmed that ACSBG1 knockdown preserved NP proteoglycans, with the IDD+sh‐ACSBG1 group displaying intense orange‐red staining and a distinct NP‐annulus fibrosus border, in contrast to the IDD+sh‐NC group (Figure 9F). These findings collectively demonstrated that silencing ACSBG1 knockdown effectively ameliorated IDD progression in vivo, highlighting its potential as a therapeutic target for IDD.

ACSBG1 knockdown attenuated IDD in a rat model. Male SD rats were subjected to IDD induction and treated with lentivirus carrying sh‐NC or sh‐ACSBG1. (A,B) Validation of ACSBG1 knockdown efficiency (n = 3 per group). (C) Western blot analysis and quantification of ACSBG1 protein expression in IDD+sh‐NC and IDD+sh‐ACSBG1 groups (n = 3 per group). (D,E) Representative H&E staining images and histological scores showing histological morphology of rat intervertebral disc tissues (n = 6 per group). (F) Representative Safranin O‐fast green staining images showing proteoglycan content in rat intervertebral disc tissues (n = 6 per group). Data are presented as mean ± SD. Comparisons among multiple groups were performed using one‐way ANOVA with Dunnett's post hoc test; comparisons between two groups were performed using two‐tailed Student's t‐test. *p < 0.05, **p < 0.01, ***p < 0.001 versus sh‐NC/IDD+sh‐NC group.

4. Discussion

Low back pain resulting from IDD is a prevalent health issue globally [28]. The pathophysiological mechanisms underlying IDD are complex, with genetic factors contributing to approximately 70% of its incidence [29]. Additionally, the development of IDD is linked to the PANoptosis of NP cells [30, 31]. Therefore, investigating PKGs and their pathogenic mechanisms in IDD is essential for a comprehensive understanding of the disorder's underlying pathology.

In this study, HTNP cells within NP tissue were identified as PANoptosis‐related cell types in IDD. The PKGs associated with IDD were identified through WGCN analysis of gene modules most strongly correlated with HTNP cell types. Subsequently, four machine learning algorithms along with LASSO regression analysis identified seven PKGs of IDD (ACSBG1, FXYD1, APCS, CEACAM1, ERAP2, ABL1, and FZD3) to construct a nomogram to predict the risk of PANoptosis in IDD. The ROC curve and calibration curve demonstrated that the nomogram exhibited strong predictive capability for IDD. Notably, no prior studies have investigated the roles of these seven PKGs in IDD. Our findings indicated the seven PKGs' potential as molecular targets for IDD, along with upregulated ACSBG1, FXYD1, APCS, and CEACAM1 in the IDD group. Previous research has shown that FXYD1 provides endogenous protection to the heart against oxidative and inflammatory damage [32]. CEACAM1 has been implicated in mouse hepatitis virus‐induced PANoptosis of cells [33]. However, there is a dearth of relevant studies examining the effects of ACSBG1 and APCS on inflammatory damage or cellular PANoptosis. Furthermore, ERAP2, ABL1, and FZD3 were found to be downregulated in patients with IDD in this study. Previous investigations have confirmed that ERAP2 induces CD4 T cell pyroptosis by upregulating NLRP3 and Caspase‐1 [34]. ABL1 is known to enhance the inflammatory response in pneumonia by activating the NF‐κB pathway [35]. Moreover, upregulation of FZD3 can inhibit the activation of the Wnt signaling pathway, thereby reducing oxidative stress [36]. Further clinical samples and cellular experiments are necessary to validate the expression of these genes and their effects on inflammation and PANoptosis in IDD.

For confirming the clinical relevance of our bioinformatics findings, the expression of the seven PKGs was analyzed in human NP tissues. The results confirmed the differential expression patterns predicted by our bioinformatics analysis, with ACSBG1, FXYD1, APCS, and CEACAM1 upregulated, and ERAP2, ABL1, and FZD3 downregulated in degenerative NP tissues, thereby validating the reliability of our computational predictions [37]. ACSBG1 has been primarily studied in lipid metabolism and immune regulation [38, 39], but its role in IDD has not been previously reported. In this study, ACSBG1 was selected for further investigation due to its significant upregulation in degenerative NP tissues and its unknown involvement in IDD pathogenesis and PANoptosis. Using a rat IDD model, we demonstrated that ACSBG1 knockdown markedly attenuated IDD progression. Histological analysis revealed that ACSBG1 knockdown preserved disc structure, maintained NP cellularity, and prevented proteoglycan depletion. These findings provide the first in vivo evidence that ACSBG1 contributes to IDD progression, identifying it as a potential therapeutic target. This is consistent with emerging evidence implicating ACSBG1 as a key regulator of inflammatory responses [39, 40], as it modulates CD4⁺ T cell differentiation and maintains immune homeostasis [40], particularly through regulating ST2⁺ regulatory T cell function [39]. Importantly, excessive inflammation and immune cell infiltration have been shown to disrupt collagen network integrity and accelerate ECM degradation in the IVD [41, 42], suggesting a potential mechanistic link between ACSBG1‐mediated immune regulation and the maintenance of disc structural integrity. However, we acknowledge that a single knockdown study is insufficient to definitively establish ACSBG1 as a therapeutic target, and further investigations—including overexpression and rescue experiments—are warranted to elucidate its downstream signaling pathways and mechanistic role in IDD.

To investigate the pathway differences between patients with varying risks and to differentiate IDD based on the IDD PANoptosis risk model, we conducted GSEA on the DEGs from the high‐risk and low‐risk groups. The findings indicated that the signaling pathways exhibiting significant differences between the two patient groups primarily involved cell proliferation and survival, cellular immune response, and inflammasome activation. These pathways are mechanistically interconnected: NF‐κB activation serves as a transcriptional primer for NLRP3 inflammasome components [43], while PI3K/AKT facilitates NF‐κB phosphorylation and nuclear translocation [44, 45], collectively amplifying inflammatory signaling and PANoptotic cell death in NP cells. The NLRP3 inflammasome, an intracellular supramolecular complex, plays a crucial role in regulating pyroptosis, apoptosis, and PANoptosis [46], functioning as a central hub that releases IL‐1β and IL‐18 to further propagate inflammation and accelerate extracellular matrix degradation [47, 48, 49]. Besides, recent studies have demonstrated a close association between the activation of the NLRP3 inflammasome and the pathological processes of IDD [50, 51]. ZBP1 has been shown to initiate cellular PANoptosis by activating the NF‐κB pathway [52]. Zhang et al. have reported that inhibiting the PI3K/AKT signaling pathway diminishes PANoptosis following cerebral ischemia/reperfusion and safeguarded the integrity of the blood–brain barrier [53]. Furthermore, several studies have indicated that inhibiting the PI3K/AKT/NF‐κB pathway may ameliorate IDD [44, 54, 55]. Consistently, elevated PANoptosis risk scores in IDD patients reflect heightened activity of this pro‐inflammatory cell death network, which may serve as a prognostic indicator for more rapid degenerative progression. Patients with high‐risk scores exhibited increased neutrophil infiltration and upregulation of immune activation markers (TLR4, TNF, and ENTPD1), suggesting that the model captures both intrinsic cell death activity and the systemic immune‐inflammatory status driving IDD pathogenesis. Clinically, this risk stratification approach may enable early identification of patients at high risk for rapid degeneration, facilitating timely therapeutic intervention targeting PANoptosis‐related pathways. However, additional research is warranted to ascertain whether targeting these signaling pathways influences the onset and progression of IDD by modulating the PANoptosis of NP cells.

Several studies have demonstrated a close relationship between immune cell infiltration and the development of IDD [56, 57]. Additionally, there are reports addressing the immune microenvironment associated with IDD [58, 59]. However, the specific immune cells that are crucial to IDD and their relationship with relevant genes remain unclear. Our findings revealed that neutrophils constituted the predominant immune cell type in patients with IDD and were significantly elevated in the IDD PANoptosis high‐risk group. Neutrophils are typically the first immune cells to respond to inflammation and infection [60]. In addition, they have the capacity to upregulate PANoptosis, thereby fostering an immunosuppressive microenvironment [61]. Key genes such as TLR4, TNF, and ENTPD1 have been identified as important mediators of immune activation [62, 63, 64], while SLAMF7 and VEGFB serve as negative regulators of immune responses [65, 66]. In this study, significant differences in these key immune‐regulating genes were observed across IDD PANoptosis risk groups. These results suggested that the regulation of the immune microenvironment was critical to the progression of IDD. Moreover, neutrophils may influence the immune microenvironment in IDD through the modulation of PANoptosis, although further foundational experiments are necessary to confirm this hypothesis.

Beyond mechanistic insights, the PANoptosis risk diagnostic model developed in this study offers potential clinical utility. By integrating the expression profiles of seven PKGs, the model provides a quantitative risk score that distinguishes IDD patients from healthy individuals with high accuracy (AUC > 0.7 across multiple validation datasets). This noninvasive approach, potentially applicable to peripheral blood or minimally invasive NP tissue sampling, could complement radiographic assessment in early disease detection and risk stratification. Future prospective studies are warranted to evaluate whether the risk score correlates with clinical outcomes, such as pain progression or surgical intervention rates, and to determine the optimal threshold for guiding therapeutic decisions.

This study has several limitations. Although experimental validation was performed using clinical tissues and an animal model, the sample size for clinical validation was relatively modest. Additionally, the precise molecular mechanisms by which ACSBG1 regulates PANoptosis and IDD progression remain to be fully elucidated. Future studies employing ACSBG1 overexpression or knockout models are needed to elucidate its downstream signaling pathways and mechanistic role in IDD. Nevertheless, the IDD PANoptosis risk diagnostic model developed in this research demonstrated the capability to differentiate IDD patients from healthy subjects, thereby affirming the model's credibility. Furthermore, our in vivo findings provide functional evidence supporting ACSBG1 as a promising therapeutic target for IDD.

5. Conclusions

In this study, we identified seven PANoptosis key genes associated with IDD and constructed a risk diagnostic model that enables accurate differentiation of IDD patients from healthy individuals. Beyond its diagnostic value, the model provides insights into the immune‐inflammatory landscape of IDD, linking PANoptosis risk to neutrophil infiltration and immune activation. Clinical tissue validation confirmed the differential expression of these seven genes, and in vivo experiments demonstrated that ACSBG1 knockdown effectively attenuated IDD progression. While ACSBG1 emerges as a promising therapeutic candidate, its precise mechanistic role warrants further investigation. Collectively, our findings establish a foundation for PANoptosis‐based risk stratification and highlight new avenues for early intervention in IDD.

Author Contributions

Kui Deng: conceptualization, data curation, software, formal analysis, investigation, project administration, writing‐original draft, writing – review and editing. Xia Lan: conceptualization, data curation, formal analysis, investigation, project administration, writing‐original draft, writing – review and editing. WenZhao Chen: conceptualization, investigation, software, formal analysis, writing – review and editing. LiuXue Du: software, formal analysis, writing – review and editing. JiangWei Chen: conceptualization, project administration, funding acquisition, supervision, writing – review and editing.

Funding

This work was supported by the National Natural Science Foundation of China (82260441) and the Natural Science Funds of Jiangxi Province (20202BAB206020).

Ethics Statement

Written informed consent was obtained from all patients or their relatives, and the institutional ethical committee of The First Affiliated Hospital of Nanchang University hospital (2024)CDYFYYLK(08–085) approved the study in accordance with the ethical guidelines from the Declaration of Helsinki. All animal experiments were approved by the Animal Ethics Committee of The First Affiliated Hospital of Nanchang University (CDYFYIACUC‐202308QR035).

Conflicts of Interest

The authors declare no conflicts of interest.

Acknowledgements

We thank all colleagues in the laboratory who have contributed to this research.

Data Availability Statement

The raw data for this bioinformatics analysis all came from online website database, RNAseq datasets (GSE205535 and GSE124272) were obtained from Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo/), and PANoptosis‐related genes (PRGs) were retrieved from GeneCards database (https://www.genecards.org/). The data generated in this study are available from the corresponding author on reasonable request.

References

1. Buckwalter J. A., “Aging and Degeneration of the Human Intervertebral Disc,” Spine (Phila Pa 1976) 20, no. 11 (1995): 1307–1314. [DOI] [PubMed] [Google Scholar]
2. Kalichman L. and Hunter D. J., “The Genetics of Intervertebral Disc Degeneration. Familial Predisposition and Heritability Estimation,” Joint Bone Spine 75, no. 4 (2008): 383–387. [DOI] [PubMed] [Google Scholar]
3. Clouet J., Vinatier C., Merceron C., et al., “The Intervertebral Disc: From Pathophysiology to Tissue Engineering,” Joint Bone Spine 76, no. 6 (2009): 614–618. [DOI] [PubMed] [Google Scholar]
4. Urban J. P. and Roberts S., “Degeneration of the Intervertebral Disc,” Arthritis Research & Therapy 5, no. 3 (2003): 120–130. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Hadjipavlou A. G., Tzermiadianos M. N., Bogduk N., and Zindrick M. R., “The Pathophysiology of Disc Degeneration: A Critical Review,” Journal of Bone and Joint Surgery 90, no. 10 (2008): 1261–1270. [DOI] [PubMed] [Google Scholar]
6. Molinos M., Almeida C. R., Caldeira J., Cunha C., Gonçalves R. M., and Barbosa M. A., “Inflammation in Intervertebral Disc Degeneration and Regeneration,” Journal of the Royal Society Interface 12, no. 104 (2015): 20141191. [DOI] [PMC free article] [PubMed] [Google Scholar]
7. Vergroesen P. P., Kingma I., Emanuel K. S., et al., “Mechanics and Biology in Intervertebral Disc Degeneration: A Vicious Circle,” Osteoarthritis and Cartilage 23, no. 7 (2015): 1057–1070. [DOI] [PubMed] [Google Scholar]
8. Hughes S. P., Freemont A. J., Hukins D. W., McGregor A. H., and Roberts S., “The Pathogenesis of Degeneration of the Intervertebral Disc and Emerging Therapies in the Management of Back Pain,” Journal of Bone and Joint Surgery 94, no. 10 (2012): 1298–1304. [DOI] [PubMed] [Google Scholar]
9. Dudek M., Yang N., Ruckshanthi J. P., et al., “The Intervertebral Disc Contains Intrinsic Circadian Clocks That Are Regulated by Age and Cytokines and Linked to Degeneration,” Annals of the Rheumatic Diseases 76, no. 3 (2017): 576–584. [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Wang J., Markova D., Anderson D. G., Zheng Z., Shapiro I. M., and Risbud M. V., “TNF‐α and IL‐1β Promote a Disintegrin‐Like and Metalloprotease With Thrombospondin Type I Motif‐5‐Mediated Aggrecan Degradation Through Syndecan‐4 in Intervertebral Disc,” Journal of Biological Chemistry 286, no. 46 (2011): 39738–39749. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Zhao C. Q., Wang L. M., Jiang L. S., and Dai L. Y., “The Cell Biology of Intervertebral Disc Aging and Degeneration,” Ageing Research Reviews 6, no. 3 (2007): 247–261. [DOI] [PubMed] [Google Scholar]
12. Samir P., Malireddi R. K. S., and Kanneganti T. D., “The PANoptosome: A Deadly Protein Complex Driving Pyroptosis, Apoptosis, and Necroptosis (PANoptosis),” Frontiers in Cellular and Infection Microbiology 10 (2020): 238. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Malireddi R. K. S., Kesavardhana S., and Kanneganti T. D., “ZBP1 and TAK1: Master Regulators of NLRP3 Inflammasome/Pyroptosis, Apoptosis, and Necroptosis (PAN‐optosis),” Frontiers in Cellular and Infection Microbiology 9 (2019): 406. [DOI] [PMC free article] [PubMed] [Google Scholar]
14. Qin R., Dai S., Zhang X., et al., “Danshen Attenuates Intervertebral Disc Degeneration via Antioxidation in SD Rats,” Oxidative Medicine and Cellular Longevity 2020 (2020): 1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Tan S., Schubert D., and Maher P., “Oxytosis: A Novel Form of Programmed Cell Death,” Current Topics in Medicinal Chemistry 1, no. 6 (2001): 497–506. [DOI] [PubMed] [Google Scholar]
16. Kepp O., Galluzzi L., Zitvogel L., and Kroemer G., “Pyroptosis ‐ A Cell Death Modality of Its Kind?,” European Journal of Immunology 40, no. 3 (2010): 627–630. [DOI] [PubMed] [Google Scholar]
17. Zhu C., Jiang W., Cheng Q., Hu Z., and Hao J., “Hemeoxygenase‐1 Suppresses IL‐1β‐Induced Apoptosis Through the NF‐κB Pathway in Human Degenerative Nucleus Pulposus Cells,” Cellular Physiology and Biochemistry 46, no. 2 (2018): 644–653. [DOI] [PubMed] [Google Scholar]
18. Peng S., Zhang T., Zhang S., Tang Q., Yan Y., and Feng H., “Integrated Bioinformatics and Validation Reveal IL1B and Its Related Molecules as Potential Biomarkers in Chronic Spontaneous Urticaria,” Frontiers in Immunology 13 (2022): 850993. [DOI] [PMC free article] [PubMed] [Google Scholar]
19. Crichton D. J., Mattmann C. A., Thornquist M., Anton K., and Hughes J. S., “Bioinformatics: Biomarkers of Early Detection,” Cancer Biomarkers 9, no. 1–6 (2010): 511–530. [DOI] [PMC free article] [PubMed] [Google Scholar]
20. Greener J. G., Kandathil S. M., Moffat L., and Jones D. T., “A Guide to Machine Learning for Biologists,” Nature Reviews. Molecular Cell Biology 23, no. 1 (2022): 40–55. [DOI] [PubMed] [Google Scholar]
21. Strunz S., Wolkenhauer O., and de la Fuente A., “Network‐Assisted Disease Classification and Biomarker Discovery,” Methods in Molecular Biology 1386 (2016): 353–374. [DOI] [PubMed] [Google Scholar]
22. Li F., Sun X., Wang Y., Gao L., Shi J., and Sun K., “Development and Validation of a Novel Nomogram to Predict the Risk of Intervertebral Disc Degeneration,” Mediators of Inflammation 2022 (2022): 3665934. [DOI] [PMC free article] [PubMed] [Google Scholar]
23. Wang T., Zhang H., Wang K., et al., “Development of a Diagnostic Model for MASLD and Identification of Daidzein as the Potential Drug Using Bioinformatics Analysis and Experiments,” Frontiers in Immunology 16 (2025): 1698740. [DOI] [PMC free article] [PubMed] [Google Scholar]
24. García‐Illarramendi J. M., Matos‐Filipe P., Mas J. M., Farrés J., and Daura X., “Digital Patient Modeling Identifies Predictive Biomarkers of Regorafenib Response in Elderly Metastatic Colorectal Cancer,” Frontiers in Systems Biology 5 (2025): 1648559. [DOI] [PMC free article] [PubMed] [Google Scholar]
25. Clough E. and Barrett T., “The Gene Expression Omnibus Database,” Methods in Molecular Biology 1418 (2016): 93–110. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Yu H., Zhang Z., Wei F., et al., “Hydroxytyrosol Ameliorates Intervertebral Disc Degeneration and Neuropathic Pain by Reducing Oxidative Stress and Inflammation,” Oxidative Medicine and Cellular Longevity 2022 (2022): 2240894. [DOI] [PMC free article] [PubMed] [Google Scholar]
27. Yang G., Chen L., Gao Z., and Wang Y., “Implication of Microglia Activation and CSF‐1/CSF‐1R Pathway in Lumbar Disc Degeneration‐Related Back Pain,” Molecular Pain 14 (2018): 1744806918811238. [DOI] [PMC free article] [PubMed] [Google Scholar]
28. Cheng X., Zhang L., Zhang K., et al., “Circular RNA VMA21 Protects Against Intervertebral Disc Degeneration Through Targeting miR‐200c and X Linked Inhibitor‐of‐Apoptosis Protein,” Annals of the Rheumatic Diseases 77, no. 5 (2018): 770–779. [DOI] [PMC free article] [PubMed] [Google Scholar]
29. Guo D., Zeng M., Yu M., et al., “SSR1 and CKAP4 as Potential Biomarkers for Intervertebral Disc Degeneration Based on Integrated Bioinformatics Analysis,” JOR Spine 7, no. 1 (2024): e1309. [DOI] [PMC free article] [PubMed] [Google Scholar]
30. Qin T., Shi M., Zhang C., et al., “The Muscle‐Intervertebral Disc Interaction Mediated by L‐BAIBA Modulates Extracellular Matrix Homeostasis and PANoptosis in Nucleus Pulposus Cells,” Experimental & Molecular Medicine 56 (2024): 2503–2518. [DOI] [PMC free article] [PubMed] [Google Scholar]
31. Zhou D., Mei Y., Song C., et al., “Exploration of the Mode of Death and Potential Death Mechanisms of Nucleus Pulposus Cells,” European Journal of Clinical Investigation 54, no. 9 (2024): e14226. [DOI] [PubMed] [Google Scholar]
32. Hansen T. S., Karimi Galougahi K., Tang O., et al., “The FXYD1 Protein Plays a Protective Role Against Pulmonary Hypertension and Arterial Remodeling via Redox and Inflammatory mechanisms,” American Journal of Physiology. Heart and Circulatory Physiology 326, no. 3 (2024): H623–h635. [DOI] [PubMed] [Google Scholar]
33. Malireddi R. K. S., Bynigeri R. R., Mall R., Connelly J. P., Pruett‐Miller S. M., and Kanneganti T. D., “Inflammatory Cell Death, PANoptosis, Screen Identifies Host Factors in Coronavirus Innate Immune Response as Therapeutic Targets,” Communications Biology 6, no. 1 (2023): 1071. [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Zhang J., Cai H., Sun W., et al., “Endoplasmic Reticulum Aminopeptidase 2 Regulates CD4(+) T Cells Pyroptosis in Rheumatoid Arthritis,” Arthritis Research & Therapy 26, no. 1 (2024): 36. [DOI] [PMC free article] [PubMed] [Google Scholar]
35. Xu F., Yao W., Xue Y., Sun Q., and Cao C., “The Oncogene ABL1 Regulates the Inflammatory Response of Innate Immunity via Mediating TRAF6 Ubiquitination,” Immunobiology 227, no. 5 (2022): 152262. [DOI] [PubMed] [Google Scholar]
36. Yao W., Huang J., and He H., “Corrigendum to 'Over‐Expressed LOC101927196 Suppressed Oxidative Stress Levels and Neuron Cell Proliferation in a Rat Model of Autism Through Disrupting the Wnt Signaling Pathway by Targeting FZD3',” Cellular Signalling 64 (2019): 109396. [DOI] [PubMed] [Google Scholar]
37. Shi X., Yan Z., Ding R., et al., “Integrated Bioinformatics and Experiment Validation Reveal Cuproptosis‐Related Biomarkers and Therapeutic Targets in Sepsis‐Induced Myocardial Dysfunction,” BMC Infectious Diseases 25, no. 1 (2025): 445. [DOI] [PMC free article] [PubMed] [Google Scholar]
38. Ye X., Li Y., González‐Lamuño D., et al., “Role of ACSBG1 in Brain Lipid Metabolism and X‐Linked Adrenoleukodystrophy Pathogenesis: Insights From a Knockout Mouse Model,” Cells 13, no. 20 (2024): 1687. [DOI] [PMC free article] [PubMed] [Google Scholar]
39. Palatella M., Kruse F., Ji H., et al., “Acsbg1 Maintains Intestinal Immune Homeostasis and Controls Inflammation by Regulating ST2(+) Tregs,” Mucosal Immunology 19, no. 1 (2026): 1526–1537. [DOI] [PubMed] [Google Scholar]
40. Palatella M., Kruse F., Glage S., Bleich A., Greweling‐Pils M., and Huehn J., “Acsbg1 Regulates Differentiation and Inflammatory Properties of CD4+ T Cells,” European Journal of Microbiology and Immunology 15, no. 1 (2025): 21–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Duance V. C., Crean J. K., Sims T. J., et al., “Changes in Collagen Cross‐Linking in Degenerative Disc Disease and Scoliosis,” Spine 23, no. 23 (1998): 2545–2551. [DOI] [PubMed] [Google Scholar]
42. Feng H., Danfelter M., Strömqvist B., and Heinegård D., “Extracellular Matrix in Disc Degeneration,” Journal of Bone and Joint Surgery. American Volume 88, no. Suppl 2 (2006): 25–29. [DOI] [PubMed] [Google Scholar]
43. Liu T., Zhang L., Joo D., and Sun S. C., “NF‐κB Signaling in Inflammation,” Signal Transduction and Targeted Therapy 2 (2017): 17023. [DOI] [PMC free article] [PubMed] [Google Scholar]
44. Gong Y., Qiu J., Jiang T., et al., “Maltol Ameliorates Intervertebral Disc Degeneration Through Inhibiting PI3K/AKT/NF‐κB Pathway and Regulating NLRP3 Inflammasome‐Mediated Pyroptosis,” Inflammopharmacology 31, no. 1 (2023): 369–384. [DOI] [PMC free article] [PubMed] [Google Scholar]
45. Xuan J., Gao J., Wang X., Chen C., and Zhang Z., “Forsythiaside A Attenuates Intervertebral Disc Degeneration by Suppressing PI3K/AKT/NF‐κB‐Mediated NLRP3 Inflammasome Activation and Pyroptosis,” Journal of Nutritional Biochemistry 157 (2026): 110407. [DOI] [PubMed] [Google Scholar]
46. Zheng M. and Kanneganti T. D., “The Regulation of the ZBP1‐NLRP3 Inflammasome and Its Implications in Pyroptosis, Apoptosis, and Necroptosis (PANoptosis),” Immunological Reviews 297, no. 1 (2020): 26–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
47. Jiang Y., Qiang Z., Liu Y., et al., “Diverse Functions of NLRP3 Inflammasome in PANoptosis and Diseases,” Cell Death Discov 11, no. 1 (2025): 389. [DOI] [PMC free article] [PubMed] [Google Scholar]
48. Hai B., Mao T., Du C., et al., “USP14 Promotes Pyroptosis of Human Annulus Fibrosus Cells Derived From Patients With Intervertebral Disc Degeneration Through Deubiquitination of NLRP3,” Acta Biochimica et Biophysica Sinica Shanghai 54, no. 11 (2022): 1720–1730. [DOI] [PMC free article] [PubMed] [Google Scholar]
49. Chen Y., Guo C., Lan W., et al., “MicroRNAs Regulating Apoptosis and Delivery Strategies for Treating Intervertebral Disc Degeneration,” Frontiers in Cell and Development Biology 13 (2025): 1683247. [DOI] [PMC free article] [PubMed] [Google Scholar]
50. Chao‐Yang G., Peng C., and Hai‐Hong Z., “Roles of NLRP3 Inflammasome in Intervertebral Disc Degeneration,” Osteoarthritis and Cartilage 29, no. 6 (2021): 793–801. [DOI] [PubMed] [Google Scholar]
51. Zhao Y., Qiu C., Wang W., et al., “Cortistatin Protects Against Intervertebral Disc Degeneration Through Targeting Mitochondrial ROS‐Dependent NLRP3 Inflammasome Activation,” Theranostics 10, no. 15 (2020): 7015–7033. [DOI] [PMC free article] [PubMed] [Google Scholar]
52. Song Q., Fan Y., Zhang H., and Wang N., “Z‐DNA Binding Protein 1 Orchestrates Innate Immunity and Inflammatory Cell Death,” Cytokine & Growth Factor Reviews 77 (2024): 15–29. [DOI] [PubMed] [Google Scholar]
53. Zhang K., Wang Z. C., Sun H., Long H., and Wang Y., “Esculentoside H reduces the PANoptosis and Protects the Blood‐Brain Barrier After Cerebral Ischemia/Reperfusion Through the TLE1/PI3K/AKT Signaling Pathway,” Experimental Neurology 379 (2024): 114850. [DOI] [PubMed] [Google Scholar]
54. Elmounedi N., Bahloul W., Kharrat A., et al., “Ozone Therapy (O(2)‐O(3)) Alleviates the Progression of Early Intervertebral Disc Degeneration via the Inhibition of Oxidative Stress and the Interception of the PI3K/Akt/NF‐κB Signaling Pathway,” International Immunopharmacology 129 (2024): 111596. [DOI] [PubMed] [Google Scholar]
55. He S., Fu Y., Yan B., et al., “Curcumol Alleviates the Inflammation of Nucleus Pulposus Cells via the PI3K/Akt/NF‐κB Signaling Pathway and Delays Intervertebral Disk Degeneration,” World Neurosurgery 155 (2021): e402–e411. [DOI] [PubMed] [Google Scholar]
56. Risbud M. V. and Shapiro I. M., “Role of Cytokines in Intervertebral Disc Degeneration: Pain and Disc Content,” Nature Reviews Rheumatology 10, no. 1 (2014): 44–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
57. Liu X. W., Xu H. W., Yi Y. Y., Zhang S. B., and Wang S. J., “Role of Ferroptosis and Immune Infiltration in Intervertebral Disc Degeneration: Novel Insights From Bioinformatics Analyses,” Frontiers in Cell and Developmental Biology 11 (2023): 1170758. [DOI] [PMC free article] [PubMed] [Google Scholar]
58. Saberi M., Zhang X., and Mobasheri A., “Targeting Mitochondrial Dysfunction With Small Molecules in Intervertebral Disc Aging and Degeneration,” Geroscience 43, no. 2 (2021): 517–537. [DOI] [PMC free article] [PubMed] [Google Scholar]
59. Wang Z., Zhang S., Zhao Y., et al., “MicroRNA‐140‐3p Alleviates Intervertebral Disc Degeneration via KLF5/N‐Cadherin/MDM2/Slug Axis,” RNA Biology 18, no. 12 (2021): 2247–2260. [DOI] [PMC free article] [PubMed] [Google Scholar]
60. Liew P. X. and Kubes P., “The Neutrophil's Role During Health and Disease,” Physiological Reviews 99, no. 2 (2019): 1223–1248. [DOI] [PubMed] [Google Scholar]
61. Hu Q., Wang R., Zhang J., Xue Q., and Ding B., “Tumor‐Associated Neutrophils Upregulate PANoptosis to Foster an Immunosuppressive Microenvironment of Non‐Small Cell Lung Cancer,” Cancer Immunology, Immunotherapy 72, no. 12 (2023): 4293–4308. [DOI] [PMC free article] [PubMed] [Google Scholar]
62. Mezzasoma L., Schmidt‐Weber C. B., and Fallarino F., “In Vitro Study of TLR4‐NLRP3‐Inflammasome Activation in Innate Immune Response,” Methods in Molecular Biology 2700 (2023): 163–176. [DOI] [PubMed] [Google Scholar]
63. Juhász K., Buzás K., and Duda E., “Importance of Reverse Signaling of the TNF Superfamily in Immune Regulation,” Expert Review of Clinical Immunology 9, no. 4 (2013): 335–348. [DOI] [PubMed] [Google Scholar]
64. Li C., Zhang L., Jin Q., Jiang H., and Wu C., “CD39 (ENTPD1) in Tumors: A Potential Therapeutic Target and Prognostic Biomarker,” Biomarkers in Medicine 17, no. 12 (2023): 563–576. [DOI] [PubMed] [Google Scholar]
65. O'Connell P., Pepelyayeva Y., Blake M. K., et al., “SLAMF7 Is a Critical Negative Regulator of IFN‐α‐Mediated CXCL10 Production in Chronic HIV Infection,” Journal of Immunology 202, no. 1 (2019): 228–238. [DOI] [PMC free article] [PubMed] [Google Scholar]
66. Zou Y., Chen Q., Ye Z., Li X., and Ju R., “VEGFR1 Signaling Regulates IL‐4‐Mediated Arginase 1 Expression in Macrophages,” Current Molecular Medicine 17, no. 4 (2017): 304–311. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

[jgm70115-bib-0001] 1. Buckwalter J. A., “Aging and Degeneration of the Human Intervertebral Disc,” Spine (Phila Pa 1976) 20, no. 11 (1995): 1307–1314. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0002] 2. Kalichman L. and Hunter D. J., “The Genetics of Intervertebral Disc Degeneration. Familial Predisposition and Heritability Estimation,” Joint Bone Spine 75, no. 4 (2008): 383–387. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0003] 3. Clouet J., Vinatier C., Merceron C., et al., “The Intervertebral Disc: From Pathophysiology to Tissue Engineering,” Joint Bone Spine 76, no. 6 (2009): 614–618. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0004] 4. Urban J. P. and Roberts S., “Degeneration of the Intervertebral Disc,” Arthritis Research & Therapy 5, no. 3 (2003): 120–130. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0005] 5. Hadjipavlou A. G., Tzermiadianos M. N., Bogduk N., and Zindrick M. R., “The Pathophysiology of Disc Degeneration: A Critical Review,” Journal of Bone and Joint Surgery 90, no. 10 (2008): 1261–1270. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0006] 6. Molinos M., Almeida C. R., Caldeira J., Cunha C., Gonçalves R. M., and Barbosa M. A., “Inflammation in Intervertebral Disc Degeneration and Regeneration,” Journal of the Royal Society Interface 12, no. 104 (2015): 20141191. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0007] 7. Vergroesen P. P., Kingma I., Emanuel K. S., et al., “Mechanics and Biology in Intervertebral Disc Degeneration: A Vicious Circle,” Osteoarthritis and Cartilage 23, no. 7 (2015): 1057–1070. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0008] 8. Hughes S. P., Freemont A. J., Hukins D. W., McGregor A. H., and Roberts S., “The Pathogenesis of Degeneration of the Intervertebral Disc and Emerging Therapies in the Management of Back Pain,” Journal of Bone and Joint Surgery 94, no. 10 (2012): 1298–1304. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0009] 9. Dudek M., Yang N., Ruckshanthi J. P., et al., “The Intervertebral Disc Contains Intrinsic Circadian Clocks That Are Regulated by Age and Cytokines and Linked to Degeneration,” Annals of the Rheumatic Diseases 76, no. 3 (2017): 576–584. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0010] 10. Wang J., Markova D., Anderson D. G., Zheng Z., Shapiro I. M., and Risbud M. V., “TNF‐α and IL‐1β Promote a Disintegrin‐Like and Metalloprotease With Thrombospondin Type I Motif‐5‐Mediated Aggrecan Degradation Through Syndecan‐4 in Intervertebral Disc,” Journal of Biological Chemistry 286, no. 46 (2011): 39738–39749. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0011] 11. Zhao C. Q., Wang L. M., Jiang L. S., and Dai L. Y., “The Cell Biology of Intervertebral Disc Aging and Degeneration,” Ageing Research Reviews 6, no. 3 (2007): 247–261. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0012] 12. Samir P., Malireddi R. K. S., and Kanneganti T. D., “The PANoptosome: A Deadly Protein Complex Driving Pyroptosis, Apoptosis, and Necroptosis (PANoptosis),” Frontiers in Cellular and Infection Microbiology 10 (2020): 238. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0013] 13. Malireddi R. K. S., Kesavardhana S., and Kanneganti T. D., “ZBP1 and TAK1: Master Regulators of NLRP3 Inflammasome/Pyroptosis, Apoptosis, and Necroptosis (PAN‐optosis),” Frontiers in Cellular and Infection Microbiology 9 (2019): 406. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0014] 14. Qin R., Dai S., Zhang X., et al., “Danshen Attenuates Intervertebral Disc Degeneration via Antioxidation in SD Rats,” Oxidative Medicine and Cellular Longevity 2020 (2020): 1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0015] 15. Tan S., Schubert D., and Maher P., “Oxytosis: A Novel Form of Programmed Cell Death,” Current Topics in Medicinal Chemistry 1, no. 6 (2001): 497–506. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0016] 16. Kepp O., Galluzzi L., Zitvogel L., and Kroemer G., “Pyroptosis ‐ A Cell Death Modality of Its Kind?,” European Journal of Immunology 40, no. 3 (2010): 627–630. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0017] 17. Zhu C., Jiang W., Cheng Q., Hu Z., and Hao J., “Hemeoxygenase‐1 Suppresses IL‐1β‐Induced Apoptosis Through the NF‐κB Pathway in Human Degenerative Nucleus Pulposus Cells,” Cellular Physiology and Biochemistry 46, no. 2 (2018): 644–653. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0018] 18. Peng S., Zhang T., Zhang S., Tang Q., Yan Y., and Feng H., “Integrated Bioinformatics and Validation Reveal IL1B and Its Related Molecules as Potential Biomarkers in Chronic Spontaneous Urticaria,” Frontiers in Immunology 13 (2022): 850993. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0019] 19. Crichton D. J., Mattmann C. A., Thornquist M., Anton K., and Hughes J. S., “Bioinformatics: Biomarkers of Early Detection,” Cancer Biomarkers 9, no. 1–6 (2010): 511–530. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0020] 20. Greener J. G., Kandathil S. M., Moffat L., and Jones D. T., “A Guide to Machine Learning for Biologists,” Nature Reviews. Molecular Cell Biology 23, no. 1 (2022): 40–55. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0021] 21. Strunz S., Wolkenhauer O., and de la Fuente A., “Network‐Assisted Disease Classification and Biomarker Discovery,” Methods in Molecular Biology 1386 (2016): 353–374. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0022] 22. Li F., Sun X., Wang Y., Gao L., Shi J., and Sun K., “Development and Validation of a Novel Nomogram to Predict the Risk of Intervertebral Disc Degeneration,” Mediators of Inflammation 2022 (2022): 3665934. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0023] 23. Wang T., Zhang H., Wang K., et al., “Development of a Diagnostic Model for MASLD and Identification of Daidzein as the Potential Drug Using Bioinformatics Analysis and Experiments,” Frontiers in Immunology 16 (2025): 1698740. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0024] 24. García‐Illarramendi J. M., Matos‐Filipe P., Mas J. M., Farrés J., and Daura X., “Digital Patient Modeling Identifies Predictive Biomarkers of Regorafenib Response in Elderly Metastatic Colorectal Cancer,” Frontiers in Systems Biology 5 (2025): 1648559. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0025] 25. Clough E. and Barrett T., “The Gene Expression Omnibus Database,” Methods in Molecular Biology 1418 (2016): 93–110. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0026] 26. Yu H., Zhang Z., Wei F., et al., “Hydroxytyrosol Ameliorates Intervertebral Disc Degeneration and Neuropathic Pain by Reducing Oxidative Stress and Inflammation,” Oxidative Medicine and Cellular Longevity 2022 (2022): 2240894. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0027] 27. Yang G., Chen L., Gao Z., and Wang Y., “Implication of Microglia Activation and CSF‐1/CSF‐1R Pathway in Lumbar Disc Degeneration‐Related Back Pain,” Molecular Pain 14 (2018): 1744806918811238. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0028] 28. Cheng X., Zhang L., Zhang K., et al., “Circular RNA VMA21 Protects Against Intervertebral Disc Degeneration Through Targeting miR‐200c and X Linked Inhibitor‐of‐Apoptosis Protein,” Annals of the Rheumatic Diseases 77, no. 5 (2018): 770–779. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0029] 29. Guo D., Zeng M., Yu M., et al., “SSR1 and CKAP4 as Potential Biomarkers for Intervertebral Disc Degeneration Based on Integrated Bioinformatics Analysis,” JOR Spine 7, no. 1 (2024): e1309. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0030] 30. Qin T., Shi M., Zhang C., et al., “The Muscle‐Intervertebral Disc Interaction Mediated by L‐BAIBA Modulates Extracellular Matrix Homeostasis and PANoptosis in Nucleus Pulposus Cells,” Experimental & Molecular Medicine 56 (2024): 2503–2518. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0031] 31. Zhou D., Mei Y., Song C., et al., “Exploration of the Mode of Death and Potential Death Mechanisms of Nucleus Pulposus Cells,” European Journal of Clinical Investigation 54, no. 9 (2024): e14226. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0032] 32. Hansen T. S., Karimi Galougahi K., Tang O., et al., “The FXYD1 Protein Plays a Protective Role Against Pulmonary Hypertension and Arterial Remodeling via Redox and Inflammatory mechanisms,” American Journal of Physiology. Heart and Circulatory Physiology 326, no. 3 (2024): H623–h635. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0033] 33. Malireddi R. K. S., Bynigeri R. R., Mall R., Connelly J. P., Pruett‐Miller S. M., and Kanneganti T. D., “Inflammatory Cell Death, PANoptosis, Screen Identifies Host Factors in Coronavirus Innate Immune Response as Therapeutic Targets,” Communications Biology 6, no. 1 (2023): 1071. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0034] 34. Zhang J., Cai H., Sun W., et al., “Endoplasmic Reticulum Aminopeptidase 2 Regulates CD4(+) T Cells Pyroptosis in Rheumatoid Arthritis,” Arthritis Research & Therapy 26, no. 1 (2024): 36. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0035] 35. Xu F., Yao W., Xue Y., Sun Q., and Cao C., “The Oncogene ABL1 Regulates the Inflammatory Response of Innate Immunity via Mediating TRAF6 Ubiquitination,” Immunobiology 227, no. 5 (2022): 152262. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0036] 36. Yao W., Huang J., and He H., “Corrigendum to 'Over‐Expressed LOC101927196 Suppressed Oxidative Stress Levels and Neuron Cell Proliferation in a Rat Model of Autism Through Disrupting the Wnt Signaling Pathway by Targeting FZD3',” Cellular Signalling 64 (2019): 109396. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0037] 37. Shi X., Yan Z., Ding R., et al., “Integrated Bioinformatics and Experiment Validation Reveal Cuproptosis‐Related Biomarkers and Therapeutic Targets in Sepsis‐Induced Myocardial Dysfunction,” BMC Infectious Diseases 25, no. 1 (2025): 445. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0038] 38. Ye X., Li Y., González‐Lamuño D., et al., “Role of ACSBG1 in Brain Lipid Metabolism and X‐Linked Adrenoleukodystrophy Pathogenesis: Insights From a Knockout Mouse Model,” Cells 13, no. 20 (2024): 1687. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0039] 39. Palatella M., Kruse F., Ji H., et al., “Acsbg1 Maintains Intestinal Immune Homeostasis and Controls Inflammation by Regulating ST2(+) Tregs,” Mucosal Immunology 19, no. 1 (2026): 1526–1537. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0040] 40. Palatella M., Kruse F., Glage S., Bleich A., Greweling‐Pils M., and Huehn J., “Acsbg1 Regulates Differentiation and Inflammatory Properties of CD4+ T Cells,” European Journal of Microbiology and Immunology 15, no. 1 (2025): 21–31. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0041] 41. Duance V. C., Crean J. K., Sims T. J., et al., “Changes in Collagen Cross‐Linking in Degenerative Disc Disease and Scoliosis,” Spine 23, no. 23 (1998): 2545–2551. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0042] 42. Feng H., Danfelter M., Strömqvist B., and Heinegård D., “Extracellular Matrix in Disc Degeneration,” Journal of Bone and Joint Surgery. American Volume 88, no. Suppl 2 (2006): 25–29. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0043] 43. Liu T., Zhang L., Joo D., and Sun S. C., “NF‐κB Signaling in Inflammation,” Signal Transduction and Targeted Therapy 2 (2017): 17023. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0044] 44. Gong Y., Qiu J., Jiang T., et al., “Maltol Ameliorates Intervertebral Disc Degeneration Through Inhibiting PI3K/AKT/NF‐κB Pathway and Regulating NLRP3 Inflammasome‐Mediated Pyroptosis,” Inflammopharmacology 31, no. 1 (2023): 369–384. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0045] 45. Xuan J., Gao J., Wang X., Chen C., and Zhang Z., “Forsythiaside A Attenuates Intervertebral Disc Degeneration by Suppressing PI3K/AKT/NF‐κB‐Mediated NLRP3 Inflammasome Activation and Pyroptosis,” Journal of Nutritional Biochemistry 157 (2026): 110407. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0046] 46. Zheng M. and Kanneganti T. D., “The Regulation of the ZBP1‐NLRP3 Inflammasome and Its Implications in Pyroptosis, Apoptosis, and Necroptosis (PANoptosis),” Immunological Reviews 297, no. 1 (2020): 26–38. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0047] 47. Jiang Y., Qiang Z., Liu Y., et al., “Diverse Functions of NLRP3 Inflammasome in PANoptosis and Diseases,” Cell Death Discov 11, no. 1 (2025): 389. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0048] 48. Hai B., Mao T., Du C., et al., “USP14 Promotes Pyroptosis of Human Annulus Fibrosus Cells Derived From Patients With Intervertebral Disc Degeneration Through Deubiquitination of NLRP3,” Acta Biochimica et Biophysica Sinica Shanghai 54, no. 11 (2022): 1720–1730. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0049] 49. Chen Y., Guo C., Lan W., et al., “MicroRNAs Regulating Apoptosis and Delivery Strategies for Treating Intervertebral Disc Degeneration,” Frontiers in Cell and Development Biology 13 (2025): 1683247. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0050] 50. Chao‐Yang G., Peng C., and Hai‐Hong Z., “Roles of NLRP3 Inflammasome in Intervertebral Disc Degeneration,” Osteoarthritis and Cartilage 29, no. 6 (2021): 793–801. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0051] 51. Zhao Y., Qiu C., Wang W., et al., “Cortistatin Protects Against Intervertebral Disc Degeneration Through Targeting Mitochondrial ROS‐Dependent NLRP3 Inflammasome Activation,” Theranostics 10, no. 15 (2020): 7015–7033. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0052] 52. Song Q., Fan Y., Zhang H., and Wang N., “Z‐DNA Binding Protein 1 Orchestrates Innate Immunity and Inflammatory Cell Death,” Cytokine & Growth Factor Reviews 77 (2024): 15–29. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0053] 53. Zhang K., Wang Z. C., Sun H., Long H., and Wang Y., “Esculentoside H reduces the PANoptosis and Protects the Blood‐Brain Barrier After Cerebral Ischemia/Reperfusion Through the TLE1/PI3K/AKT Signaling Pathway,” Experimental Neurology 379 (2024): 114850. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0054] 54. Elmounedi N., Bahloul W., Kharrat A., et al., “Ozone Therapy (O(2)‐O(3)) Alleviates the Progression of Early Intervertebral Disc Degeneration via the Inhibition of Oxidative Stress and the Interception of the PI3K/Akt/NF‐κB Signaling Pathway,” International Immunopharmacology 129 (2024): 111596. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0055] 55. He S., Fu Y., Yan B., et al., “Curcumol Alleviates the Inflammation of Nucleus Pulposus Cells via the PI3K/Akt/NF‐κB Signaling Pathway and Delays Intervertebral Disk Degeneration,” World Neurosurgery 155 (2021): e402–e411. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0056] 56. Risbud M. V. and Shapiro I. M., “Role of Cytokines in Intervertebral Disc Degeneration: Pain and Disc Content,” Nature Reviews Rheumatology 10, no. 1 (2014): 44–56. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0057] 57. Liu X. W., Xu H. W., Yi Y. Y., Zhang S. B., and Wang S. J., “Role of Ferroptosis and Immune Infiltration in Intervertebral Disc Degeneration: Novel Insights From Bioinformatics Analyses,” Frontiers in Cell and Developmental Biology 11 (2023): 1170758. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0058] 58. Saberi M., Zhang X., and Mobasheri A., “Targeting Mitochondrial Dysfunction With Small Molecules in Intervertebral Disc Aging and Degeneration,” Geroscience 43, no. 2 (2021): 517–537. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0059] 59. Wang Z., Zhang S., Zhao Y., et al., “MicroRNA‐140‐3p Alleviates Intervertebral Disc Degeneration via KLF5/N‐Cadherin/MDM2/Slug Axis,” RNA Biology 18, no. 12 (2021): 2247–2260. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0060] 60. Liew P. X. and Kubes P., “The Neutrophil's Role During Health and Disease,” Physiological Reviews 99, no. 2 (2019): 1223–1248. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0061] 61. Hu Q., Wang R., Zhang J., Xue Q., and Ding B., “Tumor‐Associated Neutrophils Upregulate PANoptosis to Foster an Immunosuppressive Microenvironment of Non‐Small Cell Lung Cancer,” Cancer Immunology, Immunotherapy 72, no. 12 (2023): 4293–4308. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0062] 62. Mezzasoma L., Schmidt‐Weber C. B., and Fallarino F., “In Vitro Study of TLR4‐NLRP3‐Inflammasome Activation in Innate Immune Response,” Methods in Molecular Biology 2700 (2023): 163–176. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0063] 63. Juhász K., Buzás K., and Duda E., “Importance of Reverse Signaling of the TNF Superfamily in Immune Regulation,” Expert Review of Clinical Immunology 9, no. 4 (2013): 335–348. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0064] 64. Li C., Zhang L., Jin Q., Jiang H., and Wu C., “CD39 (ENTPD1) in Tumors: A Potential Therapeutic Target and Prognostic Biomarker,” Biomarkers in Medicine 17, no. 12 (2023): 563–576. [DOI] [PubMed] [Google Scholar]

[jgm70115-bib-0065] 65. O'Connell P., Pepelyayeva Y., Blake M. K., et al., “SLAMF7 Is a Critical Negative Regulator of IFN‐α‐Mediated CXCL10 Production in Chronic HIV Infection,” Journal of Immunology 202, no. 1 (2019): 228–238. [DOI] [PMC free article] [PubMed] [Google Scholar]

[jgm70115-bib-0066] 66. Zou Y., Chen Q., Ye Z., Li X., and Ju R., “VEGFR1 Signaling Regulates IL‐4‐Mediated Arginase 1 Expression in Macrophages,” Current Molecular Medicine 17, no. 4 (2017): 304–311. [DOI] [PubMed] [Google Scholar]

PERMALINK

Identification and Experimental Validation of PANoptosis Key Genes for Constructing a PANoptosis Risk Diagnostic Model in Intervertebral Disc Degeneration

Kui Deng

Xia Lan

WenZhao Chen

LiuXue Du

JiangWei Chen

ABSTRACT

Background

Methods

Results

Conclusion

1. Introduction

2. Materials and Methods

2.1. Data Acquisition

2.2. Single‐Cell Data Analysis

2.3. Annotation of Cell Types

2.4. Calculation of PANoptosis Score

2.5. Differential Analysis of Transcriptome Data

2.6. Weighted Gene Co‐Expression Network Analysis

2.7. Machine Learning Model Construction

2.8. PANoptosis Risk Score Calculation

2.9. Gene Set Enrichment Analysis

2.10. Construction of the Nomogram and Evaluation of Diagnostic Effectiveness

2.11. Immune Infiltration Analysis

2.12. Clinical Tissue Specimens

2.13. Animals

2.14. IDD Model Establishment

2.15. Histological Analysis

2.16. Immunohistochemistry

2.17. Quantitative Real‐Time Reverse Transcription PCR

TABLE 1.

2.18. Western Blot Analysis

2.19. Statistical Analysis

3. Results

3.1. Classification, Function and Differentiation Direction of NP Tissue Cells Were Determined

FIGURE 1.

3.2. EffectorNP (NP_3) Cell Was Identified as PANoptosis Related Cell Type in IDD

FIGURE 2.

3.3. Identification of PKGs for IDD

FIGURE 3.

3.4. Construction of IDD PANoptosis Risk Model and Calculation of Risk Score

FIGURE 4.

3.5. The Construction and Validation of IDD PANoptosis Risk Diagnostic Model

FIGURE 5.

3.6. IDD‐Related Pathway Analysis and Drug Prediction Based on IDD PANoptosis Risk Model

FIGURE 6.

3.7. The PANoptosis Risk of IDD Was Related to Immune Cell Infiltration

FIGURE 7.

3.8. Validation of the Seven PKGs in Human IDD Clinical Tissues

FIGURE 8.

3.9. Knockdown of ACSBG1 Alleviated IDD Progression in a Rat Model

FIGURE 9.

4. Discussion

5. Conclusions

Author Contributions

Funding

Ethics Statement

Conflicts of Interest

Acknowledgements

Data Availability Statement

References

Associated Data

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases