MiR-520d-5p modulates chondrogenesis and chondrocyte metabolism through targeting HDAC1

MicroRNAs (miRNAs) play an essential role in the chondrogenesis and the progression of osteoarthritis (OA). This study aimed to determine miRNAs associated with chondrogenesis of human mesenchymal stem cells (hMSCs) and chondrocyte metabolism. MiRNAs were screened in hMSCs during chondrogenesis by RNA-seq and qRT-PCR. MiRNA expression was determined in primary human chondrocytes (PHCs), and degraded cartilage samples. MiRNA mimics and inhibitors were transfected to cells to determine the effect of miRNA. Bioinformatic analysis and luciferase reporter assays were applied to determine the target gene of miRNA. The results demonstrated that miR-520d-5p was increased in hMSCs chondrogenesis. The overexpression and knockdown of miR-520d-5p promoted and inhibited chondrogenesis, and regulated chondrocyte metabolism. Histone deacetylase 1 (HDAC1) was decreased in hMSCs chondrogenesis, and HDAC1 was a targeting gene of miR-520d-5p. CI994, HDAC1 inhibitor, elevated cartilage-specific gene expressions and promoted hMSCs chondrogenesis. In IL-1β-treated PHCs, CI994 promoted AGGRECAN expression and suppressed MMP-13 expression, abolishing the effect of IL-1β on PHCs. Taken together, these results suggest that miR-520d-5p promotes hMSCs chondrogenesis and regulates chondrocyte metabolism through targeting HDAC1. This study provides novel understanding of the molecular mechanism of OA progression.


INTRODUCTION
Osteoarthritis (OA) is the most widespread joint degenerative disease around the world and a leading cause of pain and severe impairment of mobility in adults [1]. It is estimated that OA influences around 18% of females and 10% of males over 60 years old, leading to a substantial socioeconomic burden [2]. OA is characterized by degraded articular cartilage, subchondral bone thickening, synovial inflammation, osteophyte formation, and degeneration of ligaments [3]. Growing studies reveal that the etiology of OA is a complicated multi-factorial mechanism related to heredi-ty, aging, joint injury, as well as obesity [4]. To date, the detailed mechanisms underlying OA progression remain to be thoroughly investigated, and there are no effective interventions to cure degraded cartilage or slow OA progression.
As a primary pathological manifestation, cartilage degeneration takes place in the presence of mechanical damages or low-grade local or systemic inflammatory responses related to obesity, trauma, genetic predisposition, and metabolic syndrome, which are the main risk factors for the initiation and progression of OA [5,6]. In contrast, chondrogenesis is a process of AGING the generation of chondrocytes that occurs both embryogenesis and adulthood, resulting in the formation of cartilage [7]. Chondrocytes originate from the chondrogenic differentiation of mesenchymal stem cells (MSCs) [8], in which a number of cytokines and transcription factors are associated with the chondrocyte differentiation, including Runx protein family (Runx1, Runx2, and Runx3) [9] and Sox family (Sox5, Sox6, and Sox9) [10]. Beyond that, chondrocyte-specific enhancer elements, such as COL2A1, COL9A1, COL10A2, and AGGRECAN, interact with transcription factors to impact cartilage-specific gene expressions, indicating that the chondrogenesis is involved in multiple factors and signaling pathways [11]. Therefore, much attention so far is being drawn toward investigating the cellular and molecular mechanisms underlying the cartilage degeneration and chondrogenesis in OA.
In the past decade, microRNAs (miRNAs), a class of non-coding short RNAs of 18-22 nucleotides [12], are critical regulatory factors for gene expression at the post-transcriptional level, in which miRNAs can inhibit translation through directly binding to 3'UTR of target mRNA and promote mRNA degradation [13]. Numerous studies suggest that miRNAs participate in various cellular processes, such as proliferation, inflammation, stress response, apoptosis, and migration [14]. It has been demonstrated that miRNAs play an essential role in the regulation of cartilage degeneration and chondrogenesis [15,16]. For example, the expression of miR-194 decreases during the chondrogenesis of human adipose-derived stem cells (hASCs), and the downregulation of miR-194 promotes the expression of SOX5, leading to enhanced chondrogenic differentiation [17]. The upregulation of miR-365 in the pre-hypertrophic zone plays a positive role in chondrocyte proliferation by targeting histone deacetylase 4 (HDAC4) [18].
Histone deacetylases (HDACs) is an essential enzyme family associated with the modulation of histone acetylation [19], an important mechanism modulating the gene expression at the transcriptional level [20]. Currently, HDACs are classified into four families (class I, II, III, and IV), according to their function, structure, distribution, as well as expression pattern [21][22][23], of which class I, including HDAC1, 2, 3, and 8, has been reported to be essential for the progression of osteoarthritis and chondrogenesis [24]. HDAC1, in particular, plays a vital role in murine development while it is required for the formation of craniofacial cartilage and pectoral fins in zebrafish [25]. Higher expressions of HDAC1 and HDAC2 are found in the chondrocytes of diseased cartilages of patients with OA, which inhibit the expressions of matrix proteins, such as AGGRECAN and COL2A1 [26]. In addition, HDAC1 promotes TGF-β1-mediated chondrogenesis through downregulating canonical Wnt signaling pathway [27].
In this study, RNA-seq analysis and functional experiments revealed that miR-520d-5p was an essential miRNA associated with the chondrogenic differentiation of human mesenchymal stem cells (hMSCs) and chondrocyte metabolic activities. By bioinformatic prediction, we found that miR-520d-5p might directly target HDAC1. In addition, we demonstrated that miR-520d-5p plays an essential role in chondrogenesis and cartilage degradation by inhibiting HDAC1, thereby modulating the expressions of cartilagespecific genes.

Elevated expression of miR-520d-5p and decreased level of HDAC1 in hMSCs chondrogenesis
By RNA-Seq analysis, a group of differentially expressed miRNAs was found in hMSCs that were induced to differentiate to chondrocytes at 0 days and 21 days ( Figure 1A), of which miR-520d-5p was one of the upregulated miRNAs ( Figure 1A). We then applied qRT-PCR to verify the results obtained from RNA-Seq assay, we found that miR-520-5p showed the highest increasing rate compared with other upregulated miRNAs in hMSCs ( Figure 1B). These results suggested that miR-520d-5p might be an essential regulator in the chondrogenesis of hMSCs. Meanwhile, we also found that mRNA and protein expressions of HDAC1 was decreased in hMSCs at 21 days after chondrogenic induction ( Figure 1C and 1D). In addition, both mRNA and protein expressions of chondrogenic markers AGGRECAN, COMP, COL2A1, and SOX9 and the hypertrophic markers COL10A1 and RUNX2 were elevated in hMSCs at 21 days after chondrogenic induction, compared with those at 0 days of the chondrogenesis ( Figure 1E and 1F).

AGING
In contrast, miR-520d-5p mimics inhibited the expression of HDAC1 while enhanced the expression of SOX9 and COL2A1 in hMSCs ( Figure 2G). Furthermore, as shown in the immunohistochemistry assay for collagen type II, the knockdown of miR-520d-5p played a negative role in the chondrogenic differentiation of hMSCs, which was opposite to the effect of overexpression of miR-520d-5p on the chondrogenesis of hMSCs ( Figure 2H). As such, these results indicate that there is an inverse correlation between miR-520d-5p and HDAC1 and that miR-520d-5p is associated with hMSCs chondrogenesis.

Expressions of miR-520d-5p and HDAC1 in IL-1βtreated PHCs
IL-1β has been demonstrated as a proinflammatory cytokine in OA cartilage degradation and plays an inhibitory role in the chondrogenesis [28,29] and chondrocytes metabolism [30]. Thus, we attempted to determine the effect of IL-1β on the expression of miR-520d-5p. The results showed that IL-1β treatment (5 ng/mL) significantly decreased the expression of miR-520d-5p ( Figure 3A) while increased the mRNA levels of HDAC1 and MMP-13 ( Figure 3B and 3C). In addition, the effect of IL-1β on the expressions of miR-520d-5p, HDAC1, and MMP-13 exhibited a time-and dosedependent manner ( Figure 3D-3I). By western blotting assay, the protein expression of HDAC1 was increased by IL-1β treatments in a dose-dependent manner ( Figure 3J). These results together demonstrate that IL-1β treatment exerts the opposite role in the expressions of miR-520d-5p and HDAC1, compared with those in chondrogenesis.

CI994 promotes chondrogenic differentiation of hMSCs
As an HDAC1 inhibitor [34], we applied CI994 (tacedinaline) in the chondrogenic medium to determine the role of CI994 in the chondrogenic differentiation of hMSCs. Compared with the chondrogenic medium without CI994, the treatment of CI994 (100 nM) increased the mRNA expressions of AGGRECAN, COMP, COL2A1, COL10A1, RUNX2, and SOX9 in hMSCs ( Figure 5A). Meanwhile, the protein expressions of COL2A1 and acetylated histone H3 were also increased by CI994 treatment supplemented in the chondrogenic medium ( Figure 5B). Immunohistochemistry assay revealed that CI994 treatment was associated with more Collagen type II-positive cells compared with those in the chondrogenic medium without CI994 ( Figure 5C).

CI994 reverses the effect of IL-1β on PHCs
Next, we determined the effect of CI994 on PHCs. We found that CI994 treatment increased the mRNA expressions of AGGRECAN, COMP, COL2A1, and SOX9 while decreased the mRNA level of COL10A1 in time-and dose-dependent manner ( Figure 6A and 6C). Also, the protein expressions of AGGRECAN, COL2A1, and acetylated histone H3 were increased by CI994 treatment in time-and dose-dependent manner ( Figure 6B and 6D). Furthermore, based on the observations mentioned above, we uncovered that the effect of IL-1β on PHCs was reversed by CI994 treatment, in which the increased expression of MMP-13 was inhibited by CI994 treatment at 8 and 24 hours post-treatment ( Figure 6E and 6F). The similar effect of CI994 treatment was also observed in the protein expressions of AGGRECAN and MMP-13 ( Figure 6G). Together, the results reveal that CI994 can reverse the effect of IL-1β in PHCs. For each experiment, at least three replicates were available for the analysis. Data were expressed as mean ± standard deviation (SD). *P < 0.05; ** P < 0.01; *** P < 0.001.

AGING
Expressions of miR-520d-5p and HDAC1 in the degraded cartilage To determine the effect of miR-520d-5p in the progression of OA, 21 pairs of non-degraded and degraded cartilage were used to detect the expressions of miR-520d-5p and HDAC1. The results revealed that the expression miR-520d-5p was lower while the level of HDAC1 was higher in degraded cartilage samples compared with those of non-degraded cartilages ( Figure  8A and 8B). Also, the higher expression of miR-520d-5p was found in the fluorescent in situ hybridization assay ( Figure 8C).

DISCUSSION
OA is the most prevalent degenerative joint disease leading to pain and disability [35]. It has been forecast that total lived with disability (YLDs) due to OA rose from 0.84 million in 1990 to 1·97 million in 2017, and around 61.2 million individuals had suffered from OA in 2017 in China [36]. To date, in addition to pain management and surgical intervention, we still lack effective therapeutic treatment for OA. Therefore, there is an urgent need for investigating the underlying mechanism and alternative treatment for OA. In the present study, we reported that miR-520d-5p plays a positive role in hMSCs chondrogenesis through downregulating HDAC1 expression and modulates chondrocytes metabolism through impacting the expressions of cartilage-specific genes.
Many studies so far demonstrated that miR-520d-5p participates in diverse cellular processes. In human dermal fibroblast (NHDF) cells, miR-520d-5p repairs damages induced by ultraviolet B and restore damaged cells to normal senescent state through the survival of CD105-positive cells via c-Abl-ATR-BRCA1/p53 signaling pathways [37]. Given its anti-tumor effect, miR-520d-5p suppresses human glioma cell proliferation through targeting PTTG1 [38] and inhibits tumor metastasis and growth by binding to CTHRC1 in colorectal cancer [39]. In addition, the upregulation of miR-520d-5p is observed in the serum of patients with Parkinson's disease [40]. So far, however, there are no studies reporting the effect of miR-520d-5p on OA. To the best of our knowledge, we first demonstrated the role of miR-520d-5p in OA and chondrogenesis. In this study, we found that miR-520d-5p was upregulated in hMSCs at 21 days after chondrogenic induction. Also, miR-520d-5p regulated the expressions of cartilagespecific genes, such as AGGRECAN, COMP, COL2A1, and SOX9, which was also found in both PHCs and OA chondrocytes. In addition, we observed that IL-1β treatment, a proinflammatory cytokine in OA cartilage degradation [28,29], inhibited the expression of miR-520d-5p in dose-and time-dependent manner. Furthermore, we further demonstrated that the level of For each experiment, at least three replicates were available for the analysis. Data were expressed as mean ± standard deviation (SD). *P < 0.05; ** P < 0.01; *** P < 0.001. AGING miR-520d-5p was higher in the control cartilages, compared with those of degraded cartilages. Collectively, our results suggested that miR-520d-5p plays a positive role in the chondrogenic differentiation and chondrocyte metabolism.
As one of the functional mechanisms, miRNAs exert their effect through binding directly to the specific target genes [41]. To further determine the mechanism underlying the effect of miR-520d-5p on chondrogenesis, we applied bioinformatics analysis and luciferase reporter assay to demonstrated that HDAC1 was a target gene of miR-520d-5p, which can be used to reasonably interpret the opposite expression pattern between miR-520d-5p and HDAC1 in the chondrogenic differentiation and metabolism. As members of the class I HDAC family, HDAC1, HDAC2, and HDAC3 suppress chondrogenesis through downregulating cartilage-specific genes, including COL2A1, SOX9, and AGGRECAN, in human chondrocytes [26]. Also, HDAC1 is found to be involved in leukemia/lymphoma-related factor (LRF)-mediated suppressive effect on the chondrogenic differentiation [42]. Thus, these observations are consistent with the findings of this study, which suggests the negative effect of HDAC1 on chondrogenesis as well as cartilage-specific gene expression. In the present study, we also found that the effect of HDAC1 knockdown on cartilage-specific gene expressions was similar to those of overexpression of miR-520d-5p. This further demonstrated that the functional interaction between miR-520d-5p and HDAC1 in the chondrogenic differentiation. Meanwhile, these findings also suggested that miR-520d-5p might regulate the expressions of the cartilage-specific gene through inhibiting HDAC1 expression. In addition, we collected 21 pairs of non-degraded and degraded cartilage and measured the expressions of miR-520d-5p and HDAC1. The results showed that miR-520d-5p was downregulated, while HDAC1 was upregulated in Data were expressed as mean ± standard deviation (SD). *P < 0.05; ** P < 0.01; *** P < 0.001. degraded cartilages. This agrees with findings in vitro experiments in this study, further suggesting that miR-520d-5p can promote chondrogenesis, but HDAC1 can inhibit chondrogenic differentiation.

Ethics statement
All patients and subjects were informed before their inclusion, and written consent was given. The experimental protocols were approved by the Ethics Committee of Changzheng Hospital (Shanghai).

Cell isolation and culture
Human mesenchymal stem cells (hMSCs) were isolated from the iliac crest bone marrow samples of three healthy volunteer donors (2 males and 1 female; 26-36 years old), as previously described [43]. Briefly, 10 mL bone marrow samples were diluted with 10 mL PBS buffer solution (Sigma-Aldrich, Shanghai, China). Then, hMSCs were fractionated on a Lymphoprep density-gradient through centrifugation at 500 g for 20 mins. The interfacial mononuclear cells were harvested, washed, and resuspended in DMEM (Dulbecco's Modified Eagle Medium) supplemented with 10% fetal bovine serum (FBS) (Gibco, NY, USA). Next, cells were seeded in a 25-cm 2 flask (37 °C, 5% CO2) for 48 hours, and then nonadherent cells were removed by changing the new medium. When cell culture reached 80-90% confluence, cells were trypsinized by 0.25% trypsin (0.53mm EDTA), counted, and plated in a new cell culture dish contained DMEM (10% FBS, 100 μg/mL streptomycin, and 100 IU/mL penicillin). Human cartilage samples were collected from three subjects (1 male and 2 females; 38-48 years old) who underwent knee arthroplasty surgery between May 2017 and May 2018. Patients who were diagnosed as OA, immunological disorders, and tumors were excluded. Primary human chondrocytes (PHCs) were isolated from cartilage samples, as previously described [44,45]. Briefly, cartilage samples were cut into small pieces (1 mm in diameter) and then digested with DMEM/F12 medium supplemented with 10% FBS, 100 μg/mL streptomycin, 100 IU/mL penicillin, and 0.4% Pronase (Gibco, NY, USA) for 90 mins. Then, the remaining cartilage tissue samples were transferred to second-digestion processes in DMEM/12 medium supplemented 100 μg/mL streptomycin, 100 IU/mL penicillin, 5% FBS (Gibco, NY, USA), and 0.025% Collagenase P (Sigma-Aldrich, Shanghai, China) for 7 hours at 37 °C on the plate stirrer. Afterward, chondrocytes were seed in a 25-cm 2 flask with DMEM/F12 medium supplemented with 10% FBS, 100 μg/mL streptomycin, and 100 IU/mL penicillin (Gibco, NY, USA). Human OA chondrocytes were isolated from cartilage samples of three patients (2 males and 1 female; 31-38 years old), as described above. All three patients underwent lower limb amputation surgery between July 2017 and May 2018 and did not have the previous history of rheumatoid arthritis. OA was diagnosed based on the American College of Rheumatology criteria and corresponded to Kellgren-Lawrence OA grades III, and IV. For all cell culture procedures, culture media were replaced every three days, and all cells were cultured at 37 °C with 5% CO2. The cell culture reached 80-90% confluence, cells were detached by 0.05% trypsin/EDTA and then passaged in culture. hMSCs at passage 3 and PHCs at passage 1 were used to subsequent experiments.

Human cartilage samples
Twenty-one pairs of non-degraded and degraded human articular cartilage samples were collected from twentyone subjects who underwent knee arthroplasty surgery between August 2016 and May 2018 (10 males and 11 females; 47-64 years old). Subjects who were diagnosed as rheumatoid, inflammatory arthritis, immunological disorders, and tumors were excluded. Based on the Outerbridge classification scale [46], cartilage samples that were classified as grade 0 were defined as nondegraded cartilage, and cartilage samples that were classified as grade 2 and 3 were defined as degraded cartilage.

qRT-PCR
Total RNAs from cells and cartilage samples were isolated using miRNeasy Mini kit (Qiagen, Hilden, Germany), following the manufacturer's instructions. The purity and concentration of RNAs were determined AGING using Epoch Multi-Volume Spectrophotometer System (BioTek, VT, USA) according to the manufacturer's instructions. Reverse transcription was performed using PrimeScript® miRNA cDNA Synthesis kit (Takara Bio, Otsu, Japan) according to the manufacturer's instructions. Then, qRT-PCR reactions were performed using SYBR® Premix Ex Taq™ II (Takara Bio, Shiga, Japan) on the Bio-Rad IQ5 system platform (Bio-Rad Laboratories, Hercules, USA) according to the manufacturer's instructions. The primers used for target genes were summarized in Table 1. GAPDH and U6 RNA were used as the reference gene. Fold differences of target gene expressions were calculated using 2 -ΔΔCt method [47]. All samples were detected in triplicate.

RNA-Seq
After checking the quantity and purity, total RNAs of hMSCs were used to prepare RNA library using TruSeq™ Small RNA Sample Prep Kits (Illumina, San Diego, USA) according to the manufacturer's instructions. Sequencing was performed on an Illumina Hiseq2500. Common RNA families (rRNA, tRNA, snRNA, snoRNA), adapter dimers, low complexity, and repeats were removed according to previously described [48]. Small RNA sequencing reads were aligned to miRbase build 20 using Bowtie software [49]. Normalized deep-sequencing counts in RPM (NOISeq) was applied to determine differential expression [50]. miRNAs whose expression are more than two-fold change with p< 0.05 were defined as differentially expressed miRNAs

Western blotting
Total proteins were isolated from cells using lysis buffer. The amount of protein was determined using Bradford protein assay according to the manufacturer's instructions (Bio-Rad Laboratories, Hercules, USA). Protein samples (20 ug The blots were incubated with secondary antibodies for 1 hour at room temperature. The blots were visualized by SuperSignal™ West Femto Maximum Sensitivity Substrate (Thermo Scientific™, Waltham, USA) according to the manufacturer's instructions. The intensity of brands was analyzed using Image J [51].

Immunohistochemistry and Alcian blue staining
Immunohistochemistry assay was performed to determine the protein expressions of collagen type II, SOX9, and HDAC1, as previously described [52]. Meanwhile, samples were fixed with 4% paraformaldehyde and then embedded in paraffin. Processed samples were sectioned to 5 μm sections and stained with 1% Alcian blue 8GX for 30 min (Amresco, Solon, USA).

In situ hybridization
Fluorescent in situ hybridization was performed to detect miR-520d-5p expression, as previously described [53]. The fluorescent probe for miR-520d-5p was obtained from Tocris Bioscience (Bristol, United Kingdom). Immunoreactivity was determined using Image-Pro Plus 6.0 analysis software.

Statistical analysis
Data were expressed as mean ± standard deviation (SD). Statistical comparisons between groups were performed using Student's t-test or analysis of variance. Statistical analysis was performed using SPSS 13.0 (SPSS, Inc., Chicago, USA). P < 0.05 was defined as statistical significance. For each experiment, at least three replicates were available for the analysis.

AUTHOR CONTRIBUTIONS
Jiajia Lu, Bin Sun and Zhibin Zhou contributed to the conception or design of the work. Bin Han and Qiang Fu contributed to the acquisition, analysis, or interpretation of data for the work. Wang Yuan drafted the manuscript. Yaguang Han, Zeng Xu and Aimin Chen critically revised the manuscript. All gave final approval and agree to be accountable for all aspects of work ensuring integrity and accuracy.