Both traumatic and non-traumatic spinal cord injury (SCI) associated with unfavorable consequences. According to data from American National Spinal Cord Injury Database, the proportion of SCI patients in the United States is 0.0054%. In addition, the age of patients with SCI is gradually getting younger, which deserves our attention (1). A study by a Chinese team showed that the number of SCI patients in China is increasing at a rate of approximately 10% per year, and will gradually comprise a relatively large population in the future (2). Morever, the latest hospital-based retrospective study from China, it shows that the number of people with spinal cord injury between 35 and 45 approximately equal to the number of people over 65 years old, and most people with spinal cord injury are over 45 years old (3). So, there was an aging trend in spinal cord injury patients, and the older population was the most common population for sarcopenia (4). Again, It is vital to focusing on these SCI patients with sarcopenia. SCI can have serious effects on physiology, psychology, and family. There is a strong correlation between the cost of medical treatment after SCI and the level of tetraplegia; the cost of a high level of tetraplegia is the highest regarding both the initial hospitalization and subsequent treatment plan compared to other levels of tetraplegia (1).

When SCI occurs, the motor system, central nervous system, including the sensory system with its proprioceptive component, as well as the sympathetic nervous system, and other related systems, will all exhibit varying degrees of change. Preventing the deterioration of these conditions is one of the core treatments of SCI. The restoration of spinal cord function and remodeling of neural pathways is one of the research hotspots at present; however, there are still many challenges from application and transformation to implementation (5). Therefore, recovery after SCI also remains a concern. This is particularly true in patients with SCI at the neck level and thoracic level, who will experience a longer rehabilitation process after SCI and face a more severe rehabilitation challenge. Among these sequelae, skeletal muscle is a major concern. After SCI, the paralyzed muscles have reduced metabolism and motor capability (6). This feature has been reported in many studies, in addition to skeletal muscle capability changes from different perspectives, including from the aspects of a decline in quality, performance, strength (7–9).

Sarcopenia is characterized by progressive and widespread skeletal muscle dysfunction with adverse consequences (10). Sarcopenia can be diagnosed with several tests, which include the grip strength test, assessment of the appendicular skeletal muscle mass by dual-energy X-ray absorptiometry, the timed-and-go test, and other methods (10). As mentioned above, patients with SCI can present with a decrease in muscle mass, strength, and performance; however, we still lack the awareness to recognize, prevent, and change it. One study on patients with SCI with an average age of 38 years found that sarcopenia was present in many of them (11). Another study also found that sarcopenia and sarcopenic obesity were common in patients with SCI (12). However, at present, unique criteria and biomarkers for the early sensitive identification and specific diagnosis of SCI are lacking, which not only affects the life and quality of life of patients, but also results in a poor prognosis and complications (11, 13).

Research on the molecular mechanisms of the association between sarcopenia and SCI remains sparse. Studying the molecular relationship between the two diseases could result in early diagnosis, later treatment, and rehabilitation of these patients. Furthermore, it also meets the requirements for precision medicine. The aim of this study was to explore the co-pathogenesis between sarcopenia and SCI. We identified hub genes associated with the two diseases and uses various bioinformatics methods to explore the biological mechanisms. The hub genes identified for sarcopenia and SCI in this study are expected to provide new insights and ideas for the diagnosis, treatment, and rehabilitation of both diseases.

The framework of this work shown in Supplementary Figure S1, and all steps will clearly illustrate in subsequent sections.

The correlation between SCI and sarcopenia has gained increasing attention in recent years. Following SCI, skeletal muscle atrophy is approximately 30–60%, and changes, such as decreased muscle endurance and increased fatigability, often occur (35). Patients with SCI are more likely to develop sarcopenia than the general population due to limb dysfunction, especially those with high spinal segmental injury, long-term bed state, or reduced nutritional intake (11, 13, 36, 37). In the 2019 European consensus, the criteria for sarcopenia included low muscle strength, low muscle quantity or quality, and low physical performance (10). In recent years, people have also become increasingly aware of the emergence of sarcopenia after SCI and have implemented many rehabilitation measures to improve muscle changes, such as electrical stimulation, exercise training, and other rehabilitation therapies (7, 38).

After the acute phase of treatment ends for many patients with SCI, most need to enter the rehabilitation phase. This link is extremely important for patients, not only to improve the remaining function, but also to replace the lost function learning. When a SCI occurs, muscle mass decreases significantly after 6 weeks; however, muscle mass can be increased by body weight supported treadmill training (BWSTT) or functional electrical stimulation (FES) (39). Low muscle quality and strength also decrease after SCI, and patients with a high injury plane showed a greater decrease (40, 41). Infiltration of intramuscular fat is also associated with muscle loss; however, FES and resistance training can improve these two symptoms to a certain extent (41). In addition to exercise training, the use of some drugs, such as testosterone and B2-adrenergic agonists, have been shown to improve muscle loss (42). In addition, in patients with SCI, the vastus lateralis show higher fibrosis compared with healthy controls, and muscle training, such as resistance training, could reduce muscle fibrosis (43, 44). However, there is still a lack of research on the effects of training at the genetic level.

Normal muscle tissue has a degree of regeneration that resists muscle loss, but beyond this level (especially after disease or aging), this function does not compensate for loss (45). The regeneration of muscle tissue is a coordinated result of multiple cells, among which satellite cells, FAP cells, have shown important regeneration potential in recent years (46, 47). FAPs plays an important role in muscle homeostasis and regeneration (33, 34). In a muscle where no injury has occurred, FAPs are in a quiescent condition. Furthermore, they are the predominant mononuclear cell population and an essential mesenchymal progenitor for supporting function (48). After muscle injury, FAPs will wake related cell populations to participant in regeneration. In an animal model experiment, researchers used Pdgfrα knock-out mice to simulate depletion of skeletal muscle, which showed a muscle regeneration delay, significant regenerative deficit, and a decrease in the muscle index (49). Furthermore, in sarcopenia-related research, when the depletion of FAPs occurs, it causes a reduction in muscle strength and weight (50). This is why FAPs was selected as a sub-cell cluster to conduct further analysis in this study.

In our study, we focus on the mechanism of SCI and sarcopenia in the muscle tissue. After validation, we obtained three hub genes: DCN, FSTL1, and COL12A1. According to the clinic information supplied by the datasets, we performed a correlation analysis between the hub genes and clinical information. The result showed age and BMI had a positive relationship with the hub genes.

DCN is an avid collagen-binding protein and, and in many skeleton and muscle diseases, DCN plays different roles due to its role in the pericellular matrix and extracellular matrix (51–53). DCN has been widely shown to regulate cartilage degeneration, bone matrix in osteoporosis, tendon architecture, and functional activity (54–56). In contrast, research on the association of DCN in SCI and sarcopenia is lacking. In SCI, few researchers focus on the muscular changes; most assess the transformation of the central nervous system. Because DCN can suppress scarring effects and inhibit fibrogenesis, researchers want to use it to reverse the scarring response after nervous system injury (57, 58). As previously mentioned, skeletal muscle loss in SCI patients can result in fibrosis; however, consistent rehabilitation training can alleviate this effect. It could be assumed that DCN may work in this training process and suppress fibrosis. In out results, DCN showed significant changes after training. Furthermore, in the KEGG enrichment analysis of DCN, it was found to play a crucial role in the TGF-beta signaling pathway, which may promote fibrosis, and if the pathway is suppressed, it will prevent fibrosis and improve regeneration (59–61). When the expression of DCN is elevated, the effects on the TGF-beta signaling pathway decline. Furthermore, DCN also participates in process of muscle regeneration. According to the single-cell RNA-seq analysis, the expression of DCN showed a significantly difference in the FAPs cell population, with an increasing trend in pseudotime trajectory. In the GO enrichment, analysis the result showed that DCN is related to mitochondrial function. There are few studies to show a correlation between DCN and mitochondrial function in muscle tissue; however, mitochondrial function, including maximal mitochondrial ATP production, dysregulation of mitochondrial dynamics, and mitochondrial proteolysis and mitophagy, may affect the muscle mass, function, and quality (62). In the skeletal muscle of patients with SCI, a mitochondrial oxygenation capability deficit has been observed, which is related to components such as succinate dehydrogenase decrease; rehabilitation training could increase mitochondrial function (63).

FSTL1 is a secreted factor from skeletal muscle; it participates in revascularization during ischemia and is related to the function of angiogenesis in skeletal muscle (64). For the capability of normal tissue, blood supply is critical, which is determined by vascular formation (65). In muscle regeneration, when mesenchymal or other stem cell’s function, the decisive factor is whether there is existing vasculature (66). In addition, after exercise, FSTL1 expression is elevate in the serum (67). However, evidence is lacking regarding a correlation between FSTL1 and indicies of muscle, including mass, quality, and performance. In the GO enrichment analysis, FSTL1 was shown to be related to the BMP signaling pathway, endothelial cell differentiation, and development. Research shows that BMP signaling controls muscle mass by the Smad family, and perturbed BMP signaling would lead to muscle loss (68, 69). Furthermore, BMP signaling relies on satellite cells to promote muscle regeneration. The links between FSLT1, BMP signaling, and satellite cells are elusive (70). There were no outcomes in the KEGG analysis of FSTL1; however, some studies have shown that it is related to Samd2/3 signaling and integrin β3/Wnt signaling (71, 72).

COL12A1 belongs to the family of fibril-associated collagens with interrupted triple helical domains and regulates cell communication, tissue repair, and regeneration (73, 74). COL12A1 is widely expressed in mesenchymal tissues in the embryo, which is consistent with our results in the GO enrichment analysis (75). In COL12A1-related myopathic Ehlers–Danlos syndrome, patients usually present with global muscle weakness and atrophy (76). In the KEGG analysis, COL12A1 was related to protein digestion and absorption. The skeletal muscle stores approximately 75% of the protein in the whole body, so when muscle loss or other sarcopenic situations begin, the balance between protein production and consumption is lost (77, 78). In patients with SCI, they may deplete more protein; therefore, it is vital for them to pay more attention to protein intake, and muscle loss is one of the recommendation standards for increasing protein intake (78). Thus, it is a worth exploring markers of skeletal muscle-related dysfunction.

In the immune infiltration landscape in SCI muscle tissue, there is no difference when comparing the time point of day 2 and day 5. We deem that the distance between the two time points is too short to make the difference significant. In sarcopenia datasets, compared with healthy control, the B cell naïve and B cell memory change significantly. Research shows that the capacity of B cells relates to skeletal regeneration and muscle strength (79). DCN and FSTL1 significantly correlated with M1 and M2 macrophages. Myogenesis and myotube fusion correlate with M2-biase macrophages and M1 macrophages convert to M2 macrophages in skeletal muscle regeneration (80). Furthermore, the infiltration of neutrophils will lead to fibrosis in muscle fiber regeneration, which significantly correlates with DCN and FSTL1 (79).

There are some limitations in our study should be acknowledged. Firstly, in microarray datasets or bulk-RNA seq datasets, patients with SCI are not formally diagnosed with sarcopenia. In that situation, we used the related sarcopenia datasets. Secondly, clinical information is lacking in many datasets, which made it hard for us to find the exact evidence and correlations between the hub genes and clinical features. Even although we still used limited clinical information to perform the correlation, a larger sample is needed to validate its value. In our future research, we will validate these hub genes in vitro and clearly explore the mechanism between SCI and sarcopenia in muscle tissue.

In our study, we identified three hub genes (DCN, FSTL1, and COL12A1) by bioinformatics analysis of GEO datasets. In patients with SCI, these hub genes show the underlying mechanism of the development of sarcopenia and could be a diagnostic and prognostic marker. We also provide the predictive values in a rehabilitation view to understand the clinical values of the hub genes, which will help in clinical decision-making. Furthermore, we present the immune change in SCI with sarcopenia. In general, the new sights provided by our study may promote the exploration of the comorbidity of SCI and sarcopenia. DCN, FSTL1, and COL12A1 are new candidate biomarkers for the comorbidity of SCI and sarcopenia.