Lower back pain (LBP), a common musculoskeletal problem, is one of the leading causes of disability in the elderly worldwide (1, 2). It is acknowledged by a number of reports that the major contributor to LBP is intervertebral disc degeneration (IDD) (3, 4). Physiologically, the intervertebral disc (IVD) is an avascular structure of the human body in adults. The IVD is composed of three distinct regions, including the internal nucleus pulposus (NP), the peripheral annulus fibrosus (AF), and the inferior and superior cartilaginous endplates (CEP), with characterized cell types in each region respectively. Pathologically, the pathogenesis and progression of IDD is promoted by various and complicated factors, including aging, oxidative stress, inflammation, and mechanical stress, etc. The IDD process is characterized of degeneration of the NP, rapture of the AF, and calcification of the CEP (5, 6). However, the specific molecular mechanisms of disc degeneration remain elusive currently.

Ferroptosis is a newly discovered mode of regulated cell death, which differs from apoptosis, necrosis, and autophagy in morphological, biochemical, and genetic aspects (7). The cells that undergo ferroptosis have distinctive morphological characteristics, including cell membrane disruption and vesiculation, mitochondrial shrinkage with lessened cristae, mitochondrial membrane condensation and outer membrane rupture (8). In mechanism, ferroptosis is typified by intracellular iron-dependent lipid peroxidation and reactive oxygen species (ROS) accumulation to lethal levels (8, 9). Ferroptosis is implicated in various biological activities, including iron homeostasis, lipid peroxidation metabolism, glutathione (GSH) metabolism (10). Ferroptosis is negatively regulated by glutathione peroxidase 4 (GPX4), which is responsible for scavenging intracellular lipid peroxide through GSH (7, 11). It has been reported that the suppression of GPX4 or inhibition of GSH can effectively induce ferroptosis (11). Ferroptosis is also regulated by some other critical pathways, such as p53 signaling, Nrf2 signaling, Hippo signaling as well as mitochondrial signaling pathway (12).

Growing evidence has suggested that ferroptosis is interrelated with multiple pathophysiological contexts, including cancer, degenerative diseases, diabetes, cardiovascular diseases, etc (13, 14). Moreover, it has been demonstrated that ferroptosis might be also associated with some skeletal diseases, including osteoarthritis, osteoporosis, as well as rheumatoid arthritis (15). Osteoarthritis (OA) is a common degenerative joint disorder worldwide, and ferroptosis has been shown to play regulating roles in the pathogenesis and progression of OA. Recently, Miao et al. found that ferroptosis existed in OA, during which the key regulator GPX4 played critical roles in the chondrocyte cell death and extracellular matrix (ECM) degradation (16). Furthermore, by using single cell RNA sequencing analysis, Lv et al. (17) identified an important ferroptosis-associated target named TRPV1 in OA. And TRPV1 activation could protect chondrocytes from ferroptosis and mitigate the development of OA via modulating GPX4. A recent study revealed that some specific ferroptosis-related genes could be promising biomarkers for osteoporosis diagnosis and interventions, including the ER, VDR, IL-6, COL1A1, COL1A2, and PTH (18). Besides, growing evidence has also found the involvement of ferroptosis in the pathogenesis of rheumatoid arthritis, and targeting ferroptosis could be a promising therapeutic strategy for inflammatory arthritis.

Growing evidence has demonstrated that iron overload, closely associated with ferroptosis, is a common phenomenon in the aging process (19, 20). Unexpectedly, recent studies have revealed that this special type of cell death might also be related to IDD, a very common degenerative musculoskeletal disease that progresses with age. In 2021, Zhang et al. (21) established the IDD animal models and found that the levels of iron and Heme Oxygenase 1 (HO-1) were obviously elevated, and the levels of ferritin light chain markedly decreased in IDD compared to control. More recently, a study reported that iron overload could be an independent risk factor for IDD and it promoted endplate degeneration and calcification through oxidative stress and ferroptosis (22). However, the underlying mechanism of ferroptosis in IDD remains elusive and still needs further investigations. In the current research work, we aimed to explore the key roles of ferroptosis in the pathogenesis and progression of IDD comprehensively by using mature and recognized bioinformatic analysis methods, and hope to provide novel diagnostic and therapeutic targets for IDD.

IDD is the one of the leading causes of LBP, which is a very common musculoskeletal disease and has brought a heavy healthcare burden and great socioeconomic cost globally (1–3). Ferroptosis is a newly discovered type of programmed cell death which is distinct from other types of cell death. More and more studies have demonstrated that ferroptosis is closely associated with multiple human diseases, especially degenerative skeletal diseases, including osteoarthritis, osteoporosis, and inflammatory arthritis (15). Interestingly, ferroptosis might also be interrelated with the pathogenesis and progression of IDD. In the present research, we performed comprehensive bioinformatic analysis to uncover the significant roles of ferroptosis in IDD. Firstly, we screened out the DGEs in IDD according to the public datasets, including 847 downregulated genes and 1422 upregulated genes. Then we found that the expression of ferroptosis-related DGEs in IDD group was different from that in control group. We have identified 80 FRDEGs, including 20 downregulated FRDEGs and 60 upregulated FRDEGs.

In this research work, we conducted functional enrichment analysis to investigate the enriched GO items and significant pathways based on these FRDEGs. These FRDEGs are mainly related to response to chemical, response to stress, and cellular response to chemical stimulus, etc. It is worth noting that the FRDEGs are closely associated with several important pathways in IDD, including TNF signaling pathway, HIF-1 signaling pathway, and IL-17 signaling pathway. TNF-α, a member of the tumor necrosis factor (TNF) superfamily, was reported to be dysregulated in degenerated IVDs (27). Moreover, TNF signaling is deeply involved in various pathological process during IDD, including extracellular matrix (ECM) degradation, apoptosis, autophagy inflammatory responses (27). Hypoxia-inducible factors (HIFs) are transcription factors that that plays essential roles in the cellular response to low oxygen (28). The hypoxia inducible factor 1 subunit alpha (HIF1A), one of the key FRDEGs identified in this study, was reported to be decreased with the disc degeneration and participated in the IDD process through interacting with autophagy (29). However, another research work by Wang et al. (30) demonstrated that HIF1A expression was dysregulated in cartilaginous endplate and annulus fibrosus tissues of IDD patients and mouse models. They also found that aberrant activation of HIF1A in EP and AF tissues was a pathological factor for DDD, and inhibition of its aberrant activation prevented the IDD development in animal models. Interleukin-17 (IL-17), namely IL-17A, is a key cytokine of IL-17 family and is primarily secreted by T helper 17 (Th17) cells. IL-17 can trigger various signal pathways to exert regulating effects on mammalian cells (31, 32). Importantly, the expression of IL-17 was positively correlated the degree of IDD, and it could promote the IDD progression by regulating ECM metabolism, inflammatory responses, neo-angiogenesis, and NP cell autophagy and proliferation (33). It is suggested that the FRDEGs might play roles in regulating these critical signaling pathways during IDD initiation and progression, which needs further investigations in the future.

Moreover, we conducted the PPI network analysis to further identify the key and hub genes among these FRDEGs. We determined ten most important hub FRDEGs, as follows: HMOX1, KEAP1, MAPK1, HSPA5, TXNRD1, IL6, PPARA, JUN, HIF1A, and DUSP1. Then we performed functional and pathway enrichment analysis to further explore the potential molecular functions of the ten hub genes. Importantly, these hub FRDEGs are primarily involved in response to oxidative stress and cellular response to oxidative stress. It is suggested that these hub FRDEGs are closely related to oxidative stress. Oxidative stress has played key roles in the pathogenesis and development of IDD, as reported in previous studies from ours and others (34–37). Interestingly, oxidative stress is closely associated with ferroptosis. In 2009, Reardon et al. (38) reported that iron injections could upregulate skeletal muscle iron content and promote oxidative stress in mice. Recently, researchers have found that iron overload was an independent risk factor for IDD and it accelerated IDD development via oxidative stress and ferroptosis in endplate chondrocytes (22). Besides, oxidative stress could induce ferroptosis in AF cells and NP cells in an autophagy‐dependent way during IDD process (39). Therefore, the interaction between oxidative stress and ferroptosis has significantly contributed to the disc degeneration. In the pathway enrichment analysis, the most meaningful and significantly enriched pathways include Th17 cell differentiation and HIF-1 signaling pathway, and related contents have been discussed above.

We further performed ROC analysis to evaluate the predictive value of the hub FRDEGs, and also established IDD cell model to validate these genes expression in the current study. Results of ROC analysis showed that six out of the ten hub FRDEGs might be potential signature genes for IDD, including KEAP1, MAPK1, HSPA5, TXNRD1, JUN, and HIF1A. KEAP1, the Kelch like ECH associated protein 1, has been found to exert promoting functions on ferroptosis (40). KEAP1 is a redox sensor for ROS and electrophiles and negatively modulates the activation of nuclear factor E2-related factor 2 (NRF2) signaling (41). In fact, the KEAP1-NRF2 complex is a redox-sensitive transcriptional regulatory system to protect cells against oxidative stress injury (42). In our previous review article, we have comprehensively discussed the roles of KEAP1-NRF2 system in IDD progression (4). This crucial antioxidant defense system could regulate the NP cell apoptosis, senescence, extracellular matrix (ECM) metabolism, inflammatory responses, and EP calcification in IDD. MAPK1, mitogen-activated protein kinase 1, is a core signal transductor of the MAPK/ERK signaling pathway, and it could facilitate ferroptosis through regulating ROS production (43). Evidence has shown that the MAPK1 expression was increased in the degenerated disc tissues and it could accelerate the development of disc degeneration (44). HSPA5, heat shock protein family A (Hsp70) member 5, was found to resist ferroptosis in cancer cells (45). Mechanistically, HSPA5 can bind to GPX4 protein and suppress the degradation of GPX4, which is a vital suppressor of intracellular lipid peroxidation and ferroptosis as described above. However, whether and how HSPA5 participates in IDD process is unclear so far. Though the effects of TXNRD1 (thioredoxin reductase 1) on ferroptosis is not fully understood, accumulating evidence has shown that TXNRD1 might also take part in the lipid peroxidation and ferroptosis (46, 47). The roles of TXNRD1 in IDD is not so clear either. JUN, namely Jun proto-oncogene, was demonstrated to inhibit erastin-induced ferroptosis in Schwann cells (48). A previous work has revealed that the c-Jun could alleviate IDD by modulating TGF-β (49). Previous studies have reported that HIF1A could exert promoting or inhibiting effects in regulating ferroptosis (50, 51). Moreover, as mentioned above, HIF1A has exerted critical regulatory functions in the disc degeneration process, and it could be also an important target to prevent the IDD development (29, 30, 52). Therefore, these key FRDEGs might be closely involved in the initiation and progression of IDD.

Accumulating studies have attempted to explore the underlying mechanisms of IDD by bioinformatic analysis, including inflammatory response, immune infiltration, mitochondrial dysfunction, etc (53, 54). The present study focused on the association between ferroptosis with IDD, and provided novel insights into the pathogenesis of IDD. However, certain limitations should be noted when interpreting our findings. First, we did not perform microarrays or RNA sequencing in this study. The gene expression data was obtained from the GEO database, and especially, the validation dataset was from blood samples of IDD patients and healthy controls. Second, the IDD cell model used in our study could not reflect the intact IDD microenvironment in vivo, which was far more complexed than in vitro models. And this may be part of the cause for some hub FRDEGs with different expression patterns in the microarray analysis and in the IDD cell model results. Lastly, the cellular and molecular biology experiments did not unveil specific mechanisms of ferroptosis in IDD, which needed more in-depth research to explore in the future.

In summary, this study illuminated that the expression pattern of ferroptosis related genes in IDD was distinct from controls. We have determined 80 FRDEGs dysregulated in IDD. Furthermore, ten most important OSRDEGs were identified and validated, including HMOX1, KEAP1, MAPK1, HSPA5, TXNRD1, IL6, PPARA, JUN, HIF1A, DUSP1. These hub FRDEGs had close relationships with oxidative stress and some other crucial signaling pathway, which were deeply involved in IDD initiation and development. These identified hub FRDEGs might be potential signature genes for IDD. This work reveals that ferroptosis might provide a novel and promising strategy for the diagnosis and treatment of the disc degeneration.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

WL designed and supervised the study. QX and YZ performed the experiments, analyzed the data, and wrote the manuscript. All authors contributed to the article and approved the submitted version.