Tent5a modulates muscle fiber formation in adolescent idiopathic scoliosis via maintenance of myogenin expression

Abstract Objective Paravertebral muscle asymmetry may be involved in the pathogenesis of adolescent idiopathic scoliosis (AIS), and the Tent5a protein was recently identified as a novel active noncanonical poly(A) polymerase. We, therefore, explored the function of the AIS susceptibility gene Tent5a in myoblasts. Materials and methods RNA‐seq of AIS paravertebral muscle was performed, and the molecular differences in paravertebral muscle were investigated. Twenty‐four AIS susceptibility genes were screened, and differential expression of Tent5a in paravertebral muscles was confirmed with qPCR and Western blot. After the knockdown of Tent5a, the functional effects of Tent5a on C2C12 cell proliferation, migration, and apoptosis were detected by Cell Counting Kit‐8 assay, wound‐healing assay, and TUNEL assay, respectively. Myogenic differentiation markers were tested with immunofluorescence and qPCR in vitro, and muscle fiber formation was compared in vivo. Results The AIS susceptibility gene Tent5a was differentially expressed in AIS paravertebral muscles. Tent5a knockdown inhibited the proliferation and migration of C2C12 cells and inhibited the maturation of type I muscle fibers in vitro and in vivo. Mechanistically, the expression of myogenin was decreased along with the suppression of Tent5a. Conclusions Tent5a plays an important role in the proliferation and migration of myoblasts, and it regulates muscle fiber maturation by maintaining the stability of myogenin. Tent5a may be involved in the pathogenesis of AIS by regulating the formation of muscle fiber type I.

The etiology of AIS is multifactorial, including abnormal bone metabolism, endocrine hormone dyscrasia, or neuromuscular system imbalance, 2 and the genetic basis cannot be ignored in the pathogenesis of AIS. 4 A functional variant of AIS, Lbx-As1, was identified in a genome-wide association study based on 4,317 Chinese AIS patients. 5 Many susceptibility genes related to skeletal muscle development, such as Lbx1, Pax3, and Tbx1, have been reported in AIS. [6][7][8] Recently, a large number of studies have reported paravertebral muscle asymmetry based on imaging changes, histological characteristics, and electrophysiological differences, [9][10][11][12] which indicates that paravertebral muscle asymmetry may be involved in the pathogenesis of AIS.
Recently, 20 loci significantly associated with AIS have been reported from 79,211 Japanese individuals, and Tent5a was one of the susceptibility genes for AIS. 13 Tent5a was recently identified as a novel active noncanonical poly(A) polymerase. 14 Poly(A) length is highly regulated in the nucleus and cytoplasm, and poly(A) tails help regulate gene expression and probably play a critical role in cell differentiation. 15,16 Tent5a has previously been found to be associated with a variety of human diseases, including retinitis pigmentosa, skeletal disorders, Alzheimer's disease, and obesity. [17][18][19] However, the function of the Tent5a gene in muscle development is still unclear.
Based on the asymmetry of paravertebral muscle in AIS, we explored the function of the AIS susceptibility gene Tent5a in myoblasts, as well as the formation of muscle fibers. We revealed that

| Multifidus specimens of AIS patients
This study was approved by the institutional research ethics board (HX-2019-607). Twenty female AIS patients with a right thoracic curve who underwent spinal corrective surgery between the ages of 12 and 18 were included in this study ( Figure 1A). In addition, we collected muscle samples from five teenagers who underwent spinal surgery due to spinal cord trauma as the normal control. Detailed information for curve type, age, gender, and Cobb angle of the AIS patients is presented in Table S1. Muscle tissues (~1.0 × 1.0 × 0.5 cm) were obtained from the multifidus at the apical region of concavity and convexity before tissue cautery. Specimens were divided into two parts. One part was frozen in precooled isopentane for immunofluorescence staining, and the other part was flash-frozen in liquid nitrogen for RNA and protein testing.

| RNA-seq and differential expression analysis
After total RNA was extracted and quantified, library preparation of each sample was constructed according to the manufacturer's protocol with 1 μg total RNA. After isolation of the poly(A) mRNA, mRNA fragmentation and priming were performed using NEBNext Quality control, mapping, and expression analysis were performed, and the Padj of genes was set to <0.05 to detect differentially expressed genes. GOSeq (v1.34.1) was used to identify Gene Ontology terms that annotate a list of enriched genes with a significant Padj less than 0.05. In addition, topGO was used to plot DAG. We used scripts in-house to enrich significantly differentially expressed genes in KEGG pathways.

| Expression analysis of AIS susceptibility genes
In this study, 24 susceptibility genes reported in East Asian AIS patients according to two large sample size genome-wide association studies were selected. 5 The knockdown and control lentiviruses were purchased from Wuhan BrainVTA Co., Ltd. The density of C2C12 cells was approximately 80% after inoculation for 24 h, the growth medium was replaced, and concentrated lentivirus was added (MOI = 100).
Auxiliary infection reagents were also added, and the growth medium was replaced the next day. Puromycin was used to select Tent5a-inhibited C2C12 cells. EGFP fluorescence was detected with fluorescence microscopy 48 h after transfection, and qPCR and Western blotting were used to verify the knockdown effectiveness.

| Primary cultures and cell transplantation
Satellite cells were isolated by a method described by Watanabe et al. 20 Briefly, the quadriceps femoris muscles of 4-week-old C57BL/6 mice (Dossy Experimental Animals Co., Ltd.) were cut off, and nonmuscular tissues, including fat, fascia, nerves, blood vessels, and tendons, were carefully removed using an asana microscope.
The muscle was repeatedly pruned to approximately 1 mm 3 in size with ophthalmic scissors and digested with 0.1% type I collagenase (C8140, Solarbio) for 1 h. After enzymatic digestion, filtration, and resuspension, satellite cells were incubated in high serum growth medium. Satellite cells were further purified by the differential adhesion method and verified by Pax7 staining (Pax7, DSHB). P3-5 myoblasts were selected for lentivirus transfection, and the knockdown efficiency of Tent5a expression was confirmed with qPCR and Western blot.
One day before cell transplantation, 4-week-old C57BL/6 mice were selected to create a muscle injury model. 21 Briefly, 50 µl of cardiotoxin (10 µM) was multipoint injected into the tibialis anterior F I G U R E 1 RNA sequencing of AIS paravertebral muscles and differential expression analysis. (A) Representative radiograph of AIS patients who underwent spinal corrective surgery. Muscle tissues were obtained from the multifidus at the apical region of concavity and convexity during surgery. (B) Clustergram of differential expression genes. (C) Volcano Plot of differential expression genes. (D) KEGG pathway of differential expression genes. (E) GO term of differential expression genes muscle using a 31G insulin syringe. Then, myoblasts transfected with knockdown or control lentivirus were prepared, and 50 µl of singlecell suspension within 5 × 10 5 transfected myoblasts was multipoint injected into the tibial anterior muscle. 22 Four weeks after myoblast transplantation, the tibialis anterior muscles were collected for subsequent experiments.

| Immunofluorescence staining
The thickness of the frozen section for AIS samples was 8 µm.
After fixation with 4% paraformaldehyde and then blocking with

| qPCR and Western blot
After the muscle samples were pulverized using a mortar and pestle in liquid nitrogen or the cells were washed with PBS, RNAiso (108-95-2, Takara) was added, and total RNA was isolated with Qiagen RNAeasy Mini Kits (1062832, Qiagen) per the manufacturer's guidelines. The PrimeScript™ RT reagent kit with a gDNA eraser was used to produce cDNA. PowerUp™ SYBR™ green master mix (A25742, Thermo Fisher Scientific) was added to detect the expression of related genes, and the gene-specific primers are listed in Table S3 and   Table S4. Quantitative real-time PCRs were performed with an ABI QuantStudio 3 machine (Thermo Fisher Scientific).
RIPA lysis solution with 1% PMSF was used to extract total protein. Protein quantification was performed using the BCA method.

| Statistical analysis
Data were analyzed and plotted using GraphPad Prism 8.4.3 (GraphPad Software Inc.), and all quantitative data are presented as the mean ± SD. Statistical analyses were carried out using t-tests or rank-sum tests when comparing continuous variables and the chi-square test or Fisher's exact test for dichotomous variables.
Statistical tests were chosen based on sample size and normality of distribution. A p value of < 0.05 was considered statistically significant.

| RNA sequencing of AIS paravertebral muscles
To investigate the molecular differences of the paravertebral muscle in AIS, RNA sequencing of convex and concave paravertebral muscles was performed ( Figure 1B). The results showed that the expression of 10 genes was upregulated on the concavity side, and 48 genes were downregulated ( Figure 1C). KEGG analysis showed that the differentially expressed genes were enriched mainly in the insulin signaling pathway, PPAR signaling pathway, and cAMP signaling pathway ( Figure 1D). GO analysis showed that the differentially expressed genes were enriched mainly in extracellular space, glucose metabolic process, and glucose homeostasis ( Figure 1E).

| Differential expression of Tent5a in AIS paravertebral muscles
We screened for the expression of previously reported AIS susceptibility genes from RNA sequencing data, including Mtmr11, Arf1, Epha4, Ajap1, and Bcl2. 5,13 Only the Tent5a gene showed differential expression in the two-sided paravertebral muscles of AIS patients ( Figure 2). The differential reads for Tent5a were also confirmed with the use of the IGV tool ( Figure S1). We selected another 10 pairs of paravertebral muscles from AIS patients and verified the differential expression of Tent5a with qPCR ( Figure 3A) and Western blot ( Figure 3B,C).

| Asymmetry of fiber type in AIS
Compared with the convex side, a significantly lower proportion of type I muscle fibers was found on the concave side ( Figure 3F,G), but the proportion of type IIA muscle fibers was higher ( Figure 3H,I). In addition, the expression of MHC and MHC I proteins in the paravertebral muscles of AIS patients was detected by Western blot.
There was no significant difference in the relative expression of MHC between the concavity and convex sides ( Figure 3B,D), and the MHC I protein on the concave side of the paravertebral muscle was significantly lower than the MHC I protein on the convex side ( Figure 3B,E). These results support the asymmetry of fiber type in AIS. In addition, the relative mRNA expressions of myofiber-related genes were compared among the convex side, concave side, and control muscles. The mRNA expressions in the concave side of AIS, but not the convex side, were closer to the normal control, including Myh1, Myh7, Tnni1, Tnni2, Tnnt3, and Tent5a ( Figure 3J).

| Tent5a knockdown inhibited the proliferation and migration of C2C12 cells
To confirm the effect of Tent5a on myoblasts, we produced Tent5a knockdown C2C12 cells by transfection with lentivirus. Cell counting showed that the cell number of the Tent5a knockdown group was significantly lower than the cell number of the Tent5a knockdown group of the negative control group on Day 3 ( Figure 4C). In addition, the Cell Counting Kit-8 assay showed lower cell viability in the Tent5a knockdown group ( Figure 4D).
The effect of Tent5a knockdown on the migration ability of C2C12 cells was detected with a wound-healing assay. After 48 h of culture, the migration area in the Tent5a knockdown group was significantly lower than the migration area in the Tent5a knockdown group in the control group ( Figure 4E,F). A TUNEL assay was used to detect the effect of Tent5a knockdown on the apoptosis of C2C12 cells. The percentage of apoptosis was comparable between the two F I G U R E 2 Expression of AIS susceptibility genes in paravertebral muscles. The expression of 24 AIS susceptibility genes was compared between convex and concave sides in paravertebral muscles, and differential expression of Tent5a was found. Each bar represents mean ± SD. Paired sample t-test. The relative mRNA expression of myofiber-related genes among the convex side, concave side, and control muscles. One-way ANOVA test. *p < 0.05 groups. (Figure 4G,H). These results suggest that Tent5a knockdown inhibited the proliferation and migration of C2C12 cells.

| Tent5a knockdown inhibited myogenic differentiation of C2C12 cells
After 5 days of myogenic differentiation, the expression of the myogenic protein was detected with immunofluorescence staining. Compared with the negative control group, the intensity and density of immunofluorescence for MHC were significantly lower in the Tent5a knockdown group ( Figure 5A). The percentage of MHC (+) nuclei was also decreased in the Tent5a knockdown group ( Figure 5B). Myogenin, a myogenic transcription factor essential for myogenic differentiation, was decreased in the Tent5a knockdown group (Figure 5D-F). These results suggested that Tent5a knockdown inhibited the myogenic differentiation of C2C12 cells.
Intriguingly, we found that the deregulation of Tent5a led to a significant decrease in the mRNA expression of Myh7, Tnni1, and Tnnc1 ( Figure 5G). Furthermore, the percentage of Myh7 (+) nuclei was also decreased in the Tent5a knockdown group ( Figure 5C). These results suggested that Tent5a knockdown inhibited mainly myogenic differentiation and especially impaired the maturation of genes related to type I muscle fibers.

| Tent5a knockdown inhibited the maturation of type I muscle fibers in vivo
To investigate the effect of Tent5a on myogenic differentiation in vivo, myoblasts were labeled with EGFP using lentiviral transfection and transplanted into the tibial anterior muscle of mice ( Figure 6A). EGFP-positive muscle fibers represented the myogenic differentiation of transplanted cells. Four weeks later, we found that the percentage of MHC I/EGFP muscle fibers in the Tent5a knockdown group was significantly lower than the percentage of MHC I/EGFP muscle fibers in the control group ( Figure 6B,C). In the Tent5a knockdown group, the relative mRNA expression of Myh7 and Tnnc1 was also significantly decreased, while the relative expression of the Tnnc2 gene was increased ( Figure 6D). These results confirmed that

F I G U R E 4 Tent5a knockdown inhibited the proliferation and migration of C2C12 cells. (A and B) qPCR and Western blots were
conducted to verify Tent5a knockdown in C2C12 cells. Each bar represents mean ± SD. One-way ANOVA with multiple comparison test. ***p < 0.001. (C and D) Cell counting and Cell Counting Kit-8 assay showing lower proliferation viability in the Tent5a knockdown group. Each bar represents mean ± SD. One-way ANOVA with multiple comparison test. *p < 0.05, **p < 0.01. (E and F) Representative woundhealing assay showing lower migration viability in the Tent5a knockdown group. Each bar represents mean ± SD. One-way ANOVA with multiple comparison test. **p < 0.01. (G and H) Representative TUNEL assay showing a comparable percentage of apoptosis among groups. Each bar represents mean ± SD. One-way ANOVA with multiple comparison test. The scale bar represents 200 µm Tent5a knockdown mainly inhibited the maturation of type I muscle fibers in vivo.

| Regulatory effects of Tent5a on myogenic regulatory factors
We detected the mRNA expression of Tent5a, myogenic regula-

| DISCUSS ION
Because of the genetic complexity of AIS, its pathogenesis has not been clarified. In contrast to congenital scoliosis, no significant vertebral dysplasia is observed in the early stages of AIS. 23,24 Recently, an increasing number of studies have speculated that paravertebral muscle asymmetry may be involved in the pathogenesis of AIS. 25,26 Here, we investigated the role of the susceptibility gene Tent5a in the pathogenesis of AIS, and we found that Although paravertebral muscle asymmetry has been reported in AIS, the molecular differences in the paravertebral muscle of AIS are unclear. We conducted RNA-seq to compare the convex and concave paravertebral muscles. Fifty-eight genes were differentially expressed in AIS paraspinal muscle, and these genes were mainly enriched in the insulin signaling pathway, PPAR signaling pathway, and cAMP signaling pathway. The insulin signaling pathway is crucial for the regulation of glucose levels and energy metabolism, and its downstream molecules play important roles in glycogen synthesis, protein synthesis, glucose transport, and cell apoptosis. 27,28 Glucose metabolism and insulin sensitivity are essential for muscle contraction, and previous evidence has shown that muscle disorders could cause abnormal lipid deposition. 29 Recently, FR030200-Tent5a signaling has been reported to participate in the insulin resistance of the vascular injury. 30 In addition, Tent5a has been validated as a trans-regulator for leptin, and Tent5a knockdown increased insulinstimulated leptin secretion from the adipocytes. 19 PPAR belongs to One of its subtypes, PPARα, is involved in lipid metabolism and regulates the expression of many myogenic genes. [31][32][33] Similarly, scholars recently revealed the differential expression of H19 and ADIPOQ in the paravertebral muscles of AIS patients, and they speculated that transcriptome differences in AIS patients might be the reason for the structural and functional imbalance of paravertebral muscles. 34 Investigating the function of AIS susceptibility genes could help us to further understand the pathogenesis of AIS. Recently, several AIS susceptibility genes, such as Lbx1 and Pax3, have been identified to be associated with skeletal muscle function. 5,8,35,36 We screened 24 AIS susceptibility genes and verified the differential expression of showing dynamic expression of Tent5a and myogenin during myogenic differentiation. Each bar represents mean ± SD. Unpaired t-test. *p < 0.05, **p < 0.01, ***p < 0.001 F I G U R E 8 Schematic drawing of Tent5a modulates muscle fiber formation in adolescent idiopathic scoliosis. The poly(A) polymerase Tent5a promotes myogenic differentiation by maintaining the mRNA stability of myogenin, and it is involved in the pathogenesis of AIS by modulating muscle fiber formation combination is essential for complex body movement and postural behavior. 43,44 In AIS patients, we found different proportions of type I and type IIa muscle fibers between the concave and convex sides of the paravertebral muscles, which supported the asymmetry of paravertebral muscle in AIS. In addition, we found that Tent5a knockdown inhibited myogenic differentiation and especially impaired the maturation of genes related to type I muscle fibers. The composition of muscle fibers is regulated by many factors. Exercise could promote an increased proportion of type I muscle fibers and enable the transition from fast to slow muscles. 45 Tbx1 was reported to regulate the muscle fiber type and oxidative metabolism in the myotube and stimulate the differentiation of myoblasts. 46 Taken together, these results indicate that Tent5a may be involved in the pathogenesis of AIS by disturbing muscle fiber formation.
Tent5a has recently been identified as an active atypical poly(A) polymerase. 14 Poly(A) polymerase plays a highly modulatory role in the nucleus and cytoplasm, and different lengths of poly(A) tails could influence the stability, transport, and translation of mRNAs. 15,16 Recently, Tent5a was identified as a trans-regulator of leptin through proteomic analysis. 19 Patients identified with mutation of Tent5a were associated with severe osteogenesis imperfecta, and skeletal dysplasia was found in the mouse with Tent5a mutation. 47,48 In addition, Tent5a was expressed in hematopoietic system cells and played an important role in heme-induced hemoglobinization. 49 Deletion of the Tent5a gene led to abnormal formation of eye and body color in Xenopus. 50 In this study, we first reported the function of Tent5a in myoblasts and myogenic differentiation in vitro and in vivo.
As the core member of myogenic regulator factors, myogenin regulates the fate of myoblasts and culminates in the formation of mature muscle fibers. 51 In this study, we found that Tent5a knockdown mainly restrains the expression of myogenin. Previous studies have shown that myogenin processed myoblast fusion and determined the number and size of muscle fibers. 52 MyoD and myogenin are an overlapping group of myogenic regulatory factors that maintain the expression of each other in a feed-forward cycle.
Rather than MyoD, late maturation depends mainly on myogenin, and MyoD cannot compensate for the loss of myogenin in vivo. 53 In addition to controlling the maturation of skeletal muscle, myogenin also affected the location and distribution of resident satellite cells, which demonstrated a multilevel contribution of myogenin to muscle homeostasis throughout life. 54 Therefore, the poly(A) polymerase Tent5a might play an important role in maintaining the mRNA stability of myogenin. The risks and health economics related to spinal correction in AIS are thorny issues, and Tent5a might be a new regulatory target for muscle development. Our findings provide new therapeutic opportunities to correct muscle asymmetry with nonoperative management. Some limitations could not be ignored in this study. Firstly, the mechanism for Tent5a in maintaining the mRNA stability of myogenin deserves further study. Secondly, it is not an easy task to collect back muscle samples from normal teenagers, only five control muscle samples have been collected in this study, and a larger sample size control study is needed to verify the muscle difference between AIS and normal teenagers. Finally, changes in mRNA and protein expression for Tent5a and myofiber-related genes after spinal fusion for AIS are still unclear.

| CON CLUS IONS
The poly(A) polymerase Tent5a plays an important role in the proliferation and migration of myoblasts, and it might regulate muscle fiber maturation by maintaining the stability of myogenin. As the susceptibility gene of AIS, Tent5a may be involved in the pathogenesis of AIS by regulating the formation of muscle fiber type I. These findings provide new therapeutic opportunities to correct muscle asymmetry for AIS with nonoperative management. University.

CO N FLI C T O F I NTE R E S T
The authors have declared that no competing interest exists.

AUTH O R CO NTR I B UTI O N S
YMS and SSH designed the project. ML, HLY, DWW, and XHY collected and analyzed the data. ML and HLY wrote the manuscript. All authors reviewed the manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
I confirm that my article contains a Data Availability Statement even if no data is available (list of sample statements) unless my article type does not require one (e.g., Editorials, Corrections, Book Reviews, etc.).