MicroRNA‐34a‐5p promotes the progression of osteoarthritis secondary to developmental dysplasia of the hip by restraining SESN2‐induced autophagy

Osteoarthritis (OA), a late‐stage complication of developmental dysplasia of the hip (DDH), is a key factor leading to further degeneration of joint function. Studies have shown that Sestrin2 (SESN2) is a positive regulator in protecting articular cartilage from degradation. However, the regulatory effects of SESN2 on DDH‐OA and its upstream regulators remain obscure. Here, we first identified that the expression of SESN2 significantly decreased in the cartilage of DDH‐OA samples, with an expression trend negatively correlated with OA severity. Using RNA sequencing, we identified that the upregulation of miR‐34a‐5p may be an important factor for the decrease in SESN2 expression. Further exploring the regulation mechanism of miR‐34a‐5p/SESN2 is of great significance for understanding the mechanism of DDH occurrence and development. Mechanistically, we showed that miR‐34a‐5p could significantly inhibit the expression of SESN2, thereby promoting the activity of the mTOR signaling pathway. We also found that miR‐34a‐5p significantly inhibited SESN2‐induced autophagy, thereby suppressing the proliferation and migration of chondrocytes. We further validated that knocking down miR‐34a‐5p in vivo resulted in a significant increase in SESN2 expression and autophagy activity in DDH‐OA cartilage. Our study suggests that miR‐34a‐5p is a negative regulator of DDH‐OA, and may provide a new target for the prevention of DDH‐OA.


| INTRODUCTION
Developmental dysplasia of the hip (DDH) is one of the most common bone and cartilage disorders, affecting 1%-3% of all newborns. 1,2Osteoarthritis (OA) is a late-stage complication of DDH, that placed a heavy economic burden on families and society. 3,4Even if infant patients with DDH receive early and effective treatment, a considerable number of patients eventually develop OA. 2,5 Articular cartilage degeneration occurs at an early stage of DDH, [6][7][8] but can be reversed at a certain stage. 9During OA progression, chondrocyte apoptosis has been suggested to be related to the severity of cartilage degeneration. 10,11Recent studies have identified that enhanced autophagy can reduce chondrocyte apoptosis, thereby inhibiting joint degeneration and delaying the progression of OA. 12,13 Thus, further study of the regulatory mechanism of chondrocyte autophagy is conducive to a deeper understanding of the mechanism of DDH-induced OA (DDH-OA), and provides new strategies for the clinical prevention and treatment of DDH-OA.
Recently, some studies have found that the expression of SESN2 in OA cartilage is significantly reduced, resulting in a significant decrease in chondrocyte autophagy. 14Sestrin2 (SESN2) is a multifunctional protein with antiaging properties. 157][18] Overexpression of SESN2 can inhibit the mTORC1 signaling pathway by upregulating the activity of AMPK. 17,19The mTORC1 pathway is a key pathway to regulate the biosynthesis of proteins and ribosomes, the absorption of nutrients, autophagy, and other processes, and is, therefore, an important factor in regulating cell growth, metabolism, proliferation, and apoptosis. 20,21The mTORC1 inhibitor can prevent cell and matrix damage by enhancing autophagy, which is a protective mechanism in normal cartilage to avoid cell death caused by aging and trauma. 22These findings indicated that SESN2 plays a pivotal role in regulating OA progression by enhancing chondrocyte autophagy.However, the regulatory mechanism underlying decreased expression of SESN2 in DDH-OA remains unclear.
Recently, microRNAs (miRNAs) are complementary to their target messenger RNAs (mRNAs), allowing them to negatively regulate gene expression at the posttranscriptional level. 23,24cent studies have found that miR-34a-5p is closely related to autophagy. 25,26In this study, we also showed that the expression of miR-34a-5p was significantly increased in the cartilage tissue of DDH OA, and was significantly correlated with the expression level of SESN2.In terms of regulatory mechanisms, we confirmed that miR-34a-5p can inhibit autophagy of chondrocytes by regulating the SESN2/AMPK/mTORC1 pathway, and then promote apoptosis of articular chondrocytes.Our study suggests that miR-34a-5p is a negative regulator of DDH-OA and may be regarded as a new target for the clinical treatment of DDH-OA.

| Sample collecting
Patients who underwent total hip arthroplasty due to DDH were included in the experimental group (Supporting Information: Table 1, Level of Evidence III).Based on the degree of hip arthritis, we further divided these patients into OA-III (n = 10, age: 63.40 ± 8.45, 8 females and 2 males) and OA-IV (n = 10, age: 63.90 ± 8.54, 8 females and 2 males) groups, according to the Kellgren-Lawrence grade. 27

| miRNA sequencing
MiRNA sequencing was performed by Xiuyue Bio.Briefly, the total RNA of three DDH-OA cartilage samples and three control samples was extracted using the Trizol method.The small RNAs were then reverse-transcribed to cDNA.Raw data were generated by high-throughput sequencing using Hiseq Single-End mode based on the Illumina platform.The sequencing data was then analyzed and annotated based on the miRBase database.
The differentially expressed miRNAs were screened based on the thresholds with fold change ≥2 and p < 0.05, which were then visualized by heatmaps and volcano plots.

| Quantitative polymerase chain reaction (qPCR)
Total RNA in samples or cells was extracted by the Trizol method according to the manufacturer's instructions, and then reversetranscribed into cDNA.The cDNA was amplified by fluorescence qPCR using a StepOne Plus quantitative PCR system (Applied Biosystems).The primers used in this study were designed by Sangon Biotech Co. Ltd.U6 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the internal reference genes of miRNA and mRNA, respectively (Supporting Information: Table 2).After overnight incubation at 4°C, the PVDF membranes were incubated with corresponding secondary antibodies (A32731; Invitrogen) for 1 h at 25°C.Finally, a drop ECL exposure solution (Biyotime) was added to the PVDF membranes to visualize the protein bands.

| Immunofluorescence
Chondrocytes were seeded into 24 well plates with slides and fixed in ice-cold 4% paraformaldehyde after 48 h transfection.
Cells were permeated by phosphate-buffered saline (PBS) with 0.25% Triton X-100 for 10 min and blocked with 10% goat serum for 1 h.Next, the fixed cells were then incubated with a SESN2 antibody (1:200; Abcam) at 4°C for a night.Then cells were incubated with fluorescein isothiocyanate (cy3)-conjugated secondary antibody (1:200; Abcam) for 1 h in the dark.The 4,6diamidino-phenyindole (DAPI) (Abcam) was used to stain the nuclei.Finally, the slides were examined under a Leica DMi8 microscope (Carl Zeiss MicroImaging GHBH).

| Chondrocyte culture
Chondrocyte cell lines were purchased from ATCC and cultured in Dulbecco's minimal Eagle's medium (DMEM; Gibco) containing 10% fetal bovine serum (FBS; Gibco) and 1% penicillin and streptomycin.All the cells were cultured in an incubator with 5% CO 2 and 95% humidity.

| Luciferase reporter gene assay
Luciferase reporter gene plasmids containing sequencing of wild-type or mutant type of SESN2-3'UTR were constructed by Han Bio.293T cells were cultured in a 24-well plate.Luciferase reporter gene plasmids were transfected into 293T cells with or without miR-34a-5p mimics.Renilla luciferase was used as the internal reference.The luciferase activity in each group was detected using a fluorescence microplate reader.The luciferase activity of miR-34a-5p mimics was regarded as the reference group and then the related luciferase activity of each group was calculated.
The small interfering RNA (si-SESN2, s38097; Invitrogen) and overexpression plasmid of SESN2 (ov-SESN2) were designed and synthesized by Genepharma Biotech Co. Ltd.Lipofectamine™ 3000 transfection reagent was used to transfect the plasmids and RNAs into the cells.

| Cell proliferation and migration
The transfected chondrocytes seeded into a 96-well plate at 5 × 10 3 /well density.After 0, 24, 48 h incubation, add 10 μL CCK-8 reagent into each well which was then incubated at 37°C with 5% CO 2 for 2 h.The OD value of each well at 450 nm was measured by Bio Rad enzyme-linked immunosorbent assay.The relative OD value was used to indicate the proliferation of chondrocytes.

| DDH-OA model
A total of 15 neonatal Wistar rats were used in this study.In brief, the left hind leg of each rat was used to construct DDH model which induced by the fixation of in hip adduction and extension with medical tape for 10 days.The right hind leg was used as control.The fixation was loosened once daily for 1 h.Minor movements of hip and knee joints were permitted. 28DDH model rats were examined at the age of 1 month.The depth, length, and width of the acetabulum and the length and width of the femoral head were then measured.
Additionally, to observe the in vivo effect of miR-34a-5p in preventing the degeneration of shares, a group of DDH model rats was injected with atagomiR-34a-5p by tail vein injection (10 mg/kg, once a week for 4 weeks).

| Statistical analysis
The expression data were described as mean ± standard deviation (x ¯± s), which was analyzed and visualized by GraphPad Prism 7.5.The difference between the two groups was compared using a Student's t test and p < 0.05 was considered to indicate a statistically significant difference.Comparisons between three or more groups were analyzed using one-way analysis of variance.Pearson's correlation coefficient was used to analyze the correlation between miR-34a-5p and SESN2.

|
The expression of SESN2 is significantly reduced in the cartilage of DDH-OA SESN2, which is downregulated in OA cartilage, has been identified as a positive regulator in OA progression. 29,30We detected the SESN2 expression at different stages of DDH-OA.The results showed that the mRNA and protein level of SESN2 was significantly reduced in the cartilage of DDH-OA patients compared with that in healthy controls (p < 0.001, Figure 1A,B).This reduction was more significant in stage IV patients than in stage III (p < 0.001, Figure 1A,B).We further detected the protein expression levels of the downstream target genes of SESN2, including p-mTORC1 and p-AMPK1.The results showed that the expression of p-AMPK1 was significantly reduced in the cartilage of DDH-OA patients compared with that in healthy controls, with a more significant deduction trend in stage IV than in stage III (Figure 1B,C).
However, a significant increase was observed in the expression of p-mTORC1 in the cartilage of DDH-OA patients compared with that in healthy controls, which was more significant in stage IV than in stage III (Figure 1B,C).Pearson's correlation analysis showed that the level of SESN2 was positively correlated with p-AMPK (R = 0.972, p < 0.001) and negatively correlated with p-mTORC1 (R = −0.912,p < 0.001).
These results indicate that SESN2 expression is significantly correlated with the progression of DDH-OA.

| miR-34a-5p is negatively correlated to SESN2 in DDH-OA samples
Studies have shown that miRNAs are critical for the regulation of gene expression at the posttranscriptional level.A total of 244 miRNAs were predicted to be the regulatory miRNAs of SESN2 using the miRWalk, mircoT-CDS, and TargetScan databases (Figure 2A,B).In addition, RNA sequencing was performed to identify the dysregulated miRNAs in DDH-OA patients (Figure 2C).A total of 21 miRNAs were found to be dysregulated in DDH-OA patients (Figure 2D).To explore the relationship between these dysregulated miRNAs and SESN2, a Venn diagram was generated to identify the potential regulatory miRNA of SESN2 (Figure 2E).Finally, miRNA-34a-5p, an upregulated miRNA in the cartilage of DDH-OA patients, was found to be a potential regulatory miRNA of SESN2.Next, we detected the expression of miRNA-34a-5p in the cartilage of DDH-OA patients using qPCR.The results showed that the expression of miRNA-34a-5p significantly increased in the cartilage of DDH-OA patients, an effect that was more significant in stage IV than in stage III.The expression trend of miRNA-34a-5p was also negatively correlated with SESN2 expression (p < 0.001, R = -0.726, Figure 3A,B).
F I G U R E 4 miR-34a-5p could promote the mTOR/AMPK signaling pathway by suppressing SESN2 expression.(A) Overexpression of miR-34a-5p suppressed the mRNA level of SESN2, which could be partly reversed by co-transfecting the SESN2 overexpression plasmid (***p < 0.001, n = 3).(B) Immunofluorescence staining showed that the protein level of SESN2 significantly decreased after miR-34a-5p overexpression, which could be reversed by co-transfecting the SESN2 overexpression plasmid (n = 3).(C, D) Western blot analysis showed that overexpression of miR-34a-5p suppressed the protein level of SESN2 and p-AMPK but promote the protein level of p-mTOR, an effect which can be reversed by co-transfecting SESN2 overexpression plasmid (***p < 0.001, n = 3).
To further verify the interaction between miRNA-34a-5p and SESN2, we verified the binding sites of miR-34a-5p on SESN2.
Luciferase activity was significantly reduced in 293T cells cotransfected with miR-34a-5p mimics and luciferase reporter gene plasmids containing the wild-type sequence of SESN2, whereas no significant change was observed in the mutant type group (Figure 3C,D).Additionally, a significant reduction in SESN2 expression was observed when miR-34a-5p was overexpressed using the mimics (Figure 3E).In contrast, the knockdown of miR-34a-5p resulted in the overexpression of SESN2 (Figure 3F).Thus, we predicted and verified the interaction between miR-34-5p and SESN2.

| MiR-34a-5p regulates the mTORC1 signaling pathway
To explore the effect of miR-34a-5p on inhibiting SESN2-induced autophagy in chondrocytes, we conducted gain/loss-of-function assays combined with SESN2 knockdown or overexpression as a remedial strategy.Overexpression of miR-34a-5p significantly inhibited the mRNA expression of SESN2, but this effect was reversed using the SESN2 overexpression plasmid (Figure 4A).The protein levels of SESN2 and its target gene p-AMPK significantly reduced after miR-34a-5p overexpression, which could be reversed by co-transfecting the SESN2 overexpression plasmid (Figure 4B-D).However, the protein level of p-mTORC1 significantly increased after miR-34a-5p overexpression but could be remarkably reversed by overexpressing SESN2 (Figure 4C,D).
F I G U R E 5 Knocking down miR-34a-5p could inhibit the mTOR/AMPK signaling pathway by promoting SESN2 expression.(A) Knocking down miR-34a-5p promoted the mRNA level of SESN2, which could be partly reversed by co-transfecting si-SESN2 (***p < 0.001, n = 3).(B) Immunofluorescence staining showed that the protein level of SESN2 significantly increased after knocking down miR-34a-5p, which could be reversed by co-transfecting si-SESN2 (n = 3).(C, D) Western blot analysis showed that knocking down miR-34a-5p increased the protein level of SESN2 and p-AMPK, but decreased the protein level of p-mTORC1, an effect which can be reversed by co-transfecting si-SESN2 (***p < 0.001, n = 3).
On the contrary, the miR-34a-5p knockdown significantly promoted SESN2 expression thereby promoting the expression of p-AMPK and inhibiting the expression of p-mTORC1, which could also be reversed by using si-SESN2 (Figure 5A-D).

We also measured autophagy indicators LC3-II/LC3-I and Beclin-1by
Western blot analysis.The data showed that the overexpression of miR-34a-5p decreased the ratio of LC3-II/LC3-I and Beclin-1, suggesting a reduction of autophagy (Figure 6A-C).Nevertheless, this effect was significantly reversed by overexpressing SESN2 (Figure 6A-C).
Moreover, overexpression of miR-34a-5p decreased the chondrocyte proliferation and migration ability, an effect that could also be reversed by overexpressing SESN2 (Figure 6D-F).In contrast, miR-34a-5p suppression remarkably promoted the activity of chondrocyte autophagy, proliferation, and migration (Figure 6G-L).However, this effect could also be significantly reversed by SESN2 knockdown (Figure 6G-L).

| DISCUSSION
SESN2 has already been identified to be downregulated in OA cartilage. 30The downregulation of SESN2 may be a key factor in OA progression. 14,30Lou et al. found that SESN2 regulates the activity of the mTOR signaling pathway, thereby upregulating chondrocyte autophagy and then suppressing OA progression. 14Shen et al. also found that the expression of SESN2 declined in OA cartilage. 30These findings suggest that SESN2 protects chondrocytes by promoting autophagy activation.
The end-stage of DDH is characterized by hip OA. 2,4 We first identified that the expression of SESN2 and its downstream p-AMPK significantly decreased in cartilage samples of patients with DDH-OA.Meanwhile, the protein levels of p-mTORC1, downstream of AMPK, 31,32 significantly increased in the DDH-OA cartilage.These findings are consistent with those of other OA studies.However, the upstream regulatory mechanism leading to the dysregulation of SESN2 expression in DDH-OA cartilage remains unclear.
Studies have shown that miRNAs are involved in posttranscriptional regulation. 24,33MiRNAs, assisted by ago protein, can recognize the 3'UTR end of the mRNA and induce its degradation. 34,35udies have shown that many miRNAs are differentially expressed in OA articular cartilage and play key regulatory roles. 36,37However, the expression profiles of miRNAs in DDH-OA have not yet been reported.
In this study, we detected the differential expression profiles of miRNAs in DDH using high-throughput sequencing and screened a total of 21 differentially expressed miRNAs.We further examined the correlation between these miRNAs and SESN2 and found that the expression trend of miR-34a-5p was significantly correlated with SESN2.Endisha et al. found that miR-34a-5p level elevated in the synovial fluid of patients with late-stage OA and promoted joint destruction. 38Endisha et al. found that miR-34a-5p can inhibit the proliferation but promote the apoptosis of chondrocytes. 25Knockdown of miR-34a-5p can protect articular cartilage.However, the relationship between miR-34a and SESN2 expression remains unclear.
To further clarify the regulatory relationship between miR-34a-5p and SESN2, we first verified that miR-34a-5p can bind to SESN2 by constructing luciferase reporter plasmid loaded with SESN2 3'UTR sequence.We confirmed that miR-34a-5p knockdown promoted the expression of SESN2 and p-AMPK, and then inhibited the expression of p-mTORC1, thereby promoting chondrocyte autophagy.In contrast, we confirmed that overexpression of miR-34a-5p inhibited the expression of SESN2 and p-AMPK, but promote the expression of p-mTORC1, and finally suppressed chondrocyte autophagy.Our data suggested that SESN2 can inhibit mTORC1 by activating AMPK, and then participate in the regulation of cell autophagy (Figure 8).Since previous studies have shown that enhanced autophagy can inhibit apoptosis thereby promoting the viability of chondrocytes, 12,20 we further confirmed that miR-34a-5p can suppress chondrocyte proliferation and migration, which indicated that miR-34a-5p is a negative regulator of DDH-OA by inhibiting SESN2-induced autophagy.
To further confirm our hypothesis, we constructed miR-34a-5p inhibitor (antagomiR-34a-5p) to verify the protective in vivo effect of miR-34a-5p knockdown on articular cartilage.The results showed that miR-34a-5p knockdown significantly promoted the SESN2 expression in the articular cartilage of DDH rats.These results suggest that the injection of miR-34a-5p inhibitor in vivo can inhibit the progression of OA by promoting SESN2-induced chondrocyte autophagy.
In conclusion, our study reveals the mechanism by which the miR-34a-5p/SESN2 axis regulates autophagy-mediated DDH-OA degeneration.MiR-34a knockdown can significantly promote the expression of SESN2, thereby enhancing chondrocyte autophagy and maintaining cartilage morphology.Therefore, miR-34a-5p knockdown may be a new strategy for preventing DDH-OA.
In addition, patients who underwent total hip arthroplasty due to femoral neck fractures were included in the control group (n = 10, age: 64.30 ± 8.96, 8 females and 2 males).Patients with other metabolic bone diseases were excluded from this study.Cartilage tissue samples were collected from the femoral head of patients with femoral neck fractures or DDH-OA.Briefly, sterile blades were used to separate the articular cartilage (100-200 mg) on the surface of the femoral head from DDH-OA patients who are undergoing joint replacement surgery.The samples were then immediately stored in liquid nitrogen after removing the soft and bone tissue on the surface.All experiments in the present study were under the ethical standards formulated in the Helsinki Declaration and were approved by the ethics committee of the First Affiliated Hospital of Weifang Medical University.

F
I G U R E 1 The expression level of SESN2 significantly decreased in the DDH-OA cartilage samples.(A) The mRNA and protein levels of SESN2 were reduced in DDH-OA cartilage samples, which was more significant in the DDH-OA IV group.(B-C) The protein levels of SESN2, p-mTOR, and p-AMPK in DDH-OA cartilage samples (***p < 0.001, n = 10).DDH, dysplasia of the hip; OA, osteoarthritis.F I G U R E 2 SESN2 was predicted as a target of miR-34a-5p.(A, B) The potential interacting miRNA of SESN2.(C) The miRNAs expression profile in DDH-OA cartilage samples.(D) The Heatmap showed the dysregulated miRNAs in DDH-OA cartilage samples.The right columns stand for the fold change of each miRNA in each sample.(E) MiR-34a-5p was predicted as the interacting miRNA of SESN2.DDH, dysplasia of the hip; OA, osteoarthritis.