Long non-coding HCG18 promotes intervertebral disc degeneration by sponging miR-146a-5p and regulating TRAF6 expression

Intervertebral disc degeneration (IDD) is associated with the deterioration of nucleus pulposus (NP) cells due to hypertrophic differentiation and calcification. Emerging studies have shown that long noncoding RNAs (lncRNAs) play critical roles in the development of IDD. Using bioinformatics prediction, we hereby sought to identify the lncRNAs that regulate the expression of microRNA-146a-5p (miR-146a-5p), an IDD-related inflammatory factor. Our study demonstrated that lncRNA HCG18 acted as an endogenous sponge to down-regulate miR-146a-5p expression in the NP cells by directly binding to miR-146a-5p. In addition, HCG18 expression was up-regulated in the patients with IDD, bulging or herniated discs, and its level was positively correlated with the disc degeneration grade. In vitro, miR-146a-5p up-regulation HCG18 retarded the growth of NP cells by decreasing S phase of cell cycle, inducing cell apoptosis, recruitment of macrophages and hypercalcification. Conversely, down-regulation of miR-146a-5p exerted opposite effects. Furthermore, we elucidated that TRAF6, a target gene by miR-146a-5p, was modulated by HCG18 expression. Restore of TRAF6 expression by virus infection reserved the effect of HCG18 on the NP cells. Altogether, our data indicated that HCG18 suppressed the growth of NP cells and promoted the IDD development via the miR-146a-5p/TRAF6/NFκB axis.

Our previous study has demonstrated that microRNA-146a-5p (miR-146a-5p) inhibits macrophages recruitment and protects the NP cells from TNF-α-induced apoptosis by targeting TRAF6. We hereby aimed to explore the potential lncRNA that sponging miR-146a-5p in IDD. Firstly, bioinformatics prediction was utilized to identify the lncRNA. The results showed that HCG18 modulated the miR-146a-5p expression in the NP cells, and the elevated HCG18 was found in patients with IDD or herniated disc. Next, we assessed the role of HCG18 on the proliferation and apoptosis of the NP cells, osteogenic differentiation, and macrophages recruitment. Our findings suggested that HCG18 serves as a stimulus in the development of IDD by targeting miR-146a-5p.
To determine the direct binding between HCG18 and miR-146a-5p, we analyzed the potential binding sequence for miR-146a-5p in HCG18 and cloned the wild type/mutant fragments including the paired bases into a pmiR-GLO vector. The luciferase reporter assay indicated that miR-146a-5p mimics significantly reduced the luciferase activity of wild type pmirGLO-HCG18, but not of mutant pmirGLO-HCG18 in NP cells (Fig. 1C).
qPCR was performed to confirm the expression of HCG18 in the patients with spinal cord injury, bulging, herniated intervertebral disc and IDD. The expression of HCG18 was significantly up-regulated in the NP tissues from the patients with bulging or herniated discs compared with the controls (Fig. 1D). Similarly, degenerative NP tissues exhibited significantly higher expression of HCG18 when compared to the controls (Fig. 1E). In addition, the expression of HCG18 was positively correlated with disc degeneration grade (Fig. 1F).

HCG18 regulates the proliferation and apoptosis of NP cells and recruitment of macrophages.
Given that the aberrant expression of HCG18 in degenerative NP tissues, we investigated the role of HCG18 on the proliferation and apoptosis of NP cells and recruitment of macrophages by transfection with HCG18 recombinant plasmid or shRNA ( Fig. 2A). The MTT assay and Ki67 immunofluorescence staining indicated that overexpression of HCG18 in NP cells significantly inhibited the cell growth, whereas down-regulation of HCG18 increase the cell growth (Fig. 2B,C). Moreover, we assessed the cell cycle and apoptosis after transfection by flow cytometry. Overexpression of HCG18 in NP cells significantly decreased the percentage of S phase and induced apoptosis compared with the control group, while down-regulation of HCG18 exerted the opposite effect (Fig. 2D,E). The correlation between HCG18 level and miR-146a-5p expression was measured in NP tissues from the patients with bulging and herniated discs (n = 60). Spearman's correlation coefficient r = 0.4796, P = 0.0006. (C) Luciferase reporter activity in NP cell was detected after co-transfection miR-con/miR-146a-5p mimics (25 nM) and luciferase vector (pmiR-GLo) containing the wild type/mutant HCG18 (100 ng/L). (D) HCG18 expression was measured by qPCR from control, bulging and herniated NP tissues (n = 90). (E) The relative expression of HCG18 in NP tissue from IDD patients and controls (n = 90). (F) The correlation between the expression of HCG18 and patients' pfirrmann score (n = 60). Spearman's correlation coefficient r = 0.7245, P = 0.0001. *P < 0.05.
Scientific RepoRts | 7: 13234 | DOI:10.1038/s41598-017-13364- 6 We also evaluated the role of HCG18 on TNF-α induced macrophage migration. The transwell migration assay indicated that up-regulation of HCG18 obviously elevated the number of migrated macrophages. In contrast, the number of migrated macrophages was significantly reduced compared with the controls after HCG18 expression was knockdown (Fig. 2F).

Effect of HCG18 on osteogenic differentiation.
Recent studies indicate that the process of calcification is associated with the process of IDD. We also investigated the effect of HCG18 on the osteogenic differentiation in NP cells by ALP staining and alizarin red staining. The results demonstrated that overexpression of HCG18 inhibited the osteogenic differentiation of NP cells. Conversely, as expected, knockdown of HCG18 promoted the osteogenic differentiation of NP cells (Fig. 3A,B).
HCG18 exerts its biological effect by controlling miR-146a-5p/TRAF6/NFκB axis. A previous study indicated that miR-146a-5p directly inhibits TRAF6 gene expression via targeting its 3′UTR 14 . As expected, the expression of TRAF6 levels in NP tissues with high HCG18 level was significantly higher than those in NP tissue with low HCG18 level (Fig. 4A). The HCG18 level showed a positive correlation with TRAF6 expression in the NP tissues (Fig. 4B). In vitro, the expression of TRAF6, p-NFκB (Ser536) and NFκB was significantly enhanced after HCG18 was overexpressed in NP cells. However, the inhibition of HCG18 suppressed the expression of TRAF6 and NFκB (Fig. 4C,D). To investigate whether HCG18 exerts its biological effect by modulating the miR-146a-5p/TRAF6/NFκB axis, we repressed or restored the TRAF6 expression by TRAF6 overexpressing or interfering virus infection in NP cells. As expected, TRAF6 expression was decreased after TRAF6 interfering virus infection, and increased after TRAF6 overexpressing virus infection (Fig. 5A). The MTT assay and Ki67 immunofluorescence staining showed that down-regulation of TRAF6 abolished HCG18-mediated inhibition of the growth of NP cells (Fig. 5B,C). Similarly, infection with TRAF6 interfering virus increased the percentage of S phase, and reduced cell apoptosis in NP cells (Fig. 5D,E). Furthermore, HCG18-induced macrophage invasion ( Fig. 5F) and osteogenic differentiation ( Fig. 5G,H) were suppressed due to TRAF6 interfering virus infection. As expected, reintroduction of TRAF6 in low HCG18 expressing NP cells revealed a reverse result.

Discussion
IDD is the most common chronic, prevalent and age-related degenerative musculoskeletal disorder 5 . LncRNAs have been shown to be differentially expressed in human degenerative NP tissue, and involved in the pathological processes of IDD, including inflammatory responses, apoptosis, proteoglycans degradation and ECM degeneration 3,8,15 . Therefore, the investigation of precise regulatory mechanism of lncRNA at the initial stage of IDD may be conductive to develop new diagnostic and therapeutic strategies for IDD and to improve the quality of life in IDD patients.
Emerging studies indicate that lncRNAs play an essential role in the regulation of gene expression by acting as miRNA sponges 13,[16][17][18] . Previous our study suggested that miR-146a-5p is frequently down-regulated in human degenerative NP tissues (such as bulging and herniated discs) 14 . The dysregulation of miR-146a-5p can be attributed to inflammatory response in IDD. Bioinformatics prediction indicated that HCG18 may act as a miR-146a-5p sponge. The luciferase reporter assay confirmed that miR-146a-5p directly bound to wild type pmirGLO-HCG18, but not mutant pmirGLO-HCG18 in NP cells. In addition, we detected the expression of HCG18 in NP tissues from patients with bulging or herniated discs or IDD and spinal cord injury tissues. HCG18, a 2430-bp lncRNA that maps to chromosome 6p22.1, is expressed from NR_024052 locus. However, it is rarely reported especially the biological role of HCG18. The precise biological function and regulatory molecular mechanism of HCG18 are still unknown and need further investigation. Our result first demonstrated that the HCG18 level was up-regulated in NP tissues from patients with bulging or herniated discs. HCG18 expression was positively correlated with the disc degeneration grade and the hernia size, suggesting that HCG18 might act as a promoter in the development of IDD. Recent studies have revealed some aberrantly expressed lncRNAs in human IDD 8 . In fact, lncRNAs have been regarded to function in IDD development 3 . The expression of the profiling of lncRNA reflects the true scenarios in human lumbar disc diseases and low back pain. In addition, current study demonstrated that the HCG18 inhibited the proliferation of NP cells and induced cell apoptosis, macrophage recruitment ability and osteogenic differentiation. These results are in accordance with our previous data, indicating that HCG18 is a novel booster in the progression of IDD by functioning as a competing endogenous miR-146a-5p. During the moderate and late stages of IDD, fibrocartilage-like tissue, bone formation, as well as nerve and blood vessels are found in the intervertebral disc 7,19 . A recent study shows that the presence of progenitor/stem cells can be attributed to this phenomenon [20][21][22] . NP cells have been demonstrated to be able to differentiate into cartilage, fibrocartilage cells, osteoblasts, neurons, and endothelial cells in response to different stimuli 23 . Our data first indicated that HCG18 promotes the osteogenic differentiation which may contribute to chronic low-back pain and low quality of life in IDD patients.
To elucidate the molecular mechanism of HCG18 in IDD progression, we validated the TRAF6/NFκB signaling pathway which plays an important role in inflammatory response and pro-inflammatory cytokines release 24,25 . We found that HCG18 level was positively associated with TRAF6 expression in NP tissue. Overexpression of HCG18 activated the TRAF6/NFκB signaling pathway, leading to the recruitment of macrophages and apoptosis of NP cells in intervertebral discs. A recent study suggested that overexpression of miR-146a significantly decreased the levels of pro-inflammatory cytokines, including IL-1b, TNF-α, and IL-6 by targeting the TARF6/NFκB pathway. It is indicated that miR-146a ameliorates inflammation via the TRAF6/NF-kB pathway in intervertebral disc cells 26,27 . Our results lend credence to the previous study suggesting that HCG18, acting as a miR-146a-5p sponge, accelerate IDD progression via the miR-146a-5p/TARF6/NFκB axis. Moreover, down-regulation of TRAF6 abolished HCG18-mediated effect on the proliferation and apoptosis of NP cells, macrophage recruitment and osteogenic differentiation. These results imply that HCG18/miR-146a-5p/TARF6/ NFκB axis exert a critical function in the pathogenesis of IDD.
In conclusion, our results suggested that HCG18 acts a sponge of miR-146a-5p in NP cells, and the HCG18 level was up-regulated in IDD. Furthermore, HCG18 plays a crucial role in the proliferation and apoptosis of NP cells, macrophage recruitment, and osteogenic differentiation via the miR-146a-5p/TARF6/NFκB axis. Taken together, HCG18 represents a novel early diagnostic marker of IDD and an efficient anabolic strategy for IDD patients. In order to clarify the potential of clinical translation of HCG18 in IDD, an in vivo murine IDD model was used to explore the clinical application of HCG18 in the future work.

Materials and Methods
Patients and tissue samples. The human lumber NP specimens were obtained from 120 patients with IDD (n = 30), bulging discs (n = 30), herniated discs (n = 30), and spinal cord injury (control, n = 30) between August 2010 and June 2016 from the Department of orthopedics, Changzheng hospital, Second Military Medical University (Shanghai, China). All included patients had typically clinical symptoms, and the degree of IDD was evaluated on magnetic resonance imaging (MRI) scan according to a modified pfirrmann grading classification. The specimens were first isolated within 30 min, and then divided into two parts (frozen in liquid nitrogen for store or isolated for NP cell culture). This study (No. SMMU2010023) was approved by the Ethics Review Board of Changzheng hospital, Second Military Medical University. Written informed consent was gathered from all participants. In this study, all methods were performed in accordance with the relevant guidelines and regulations. RNA Extraction and lncRNAs expression assay. Total RNA from NP tissue and cell were extracted using Trizol reagent (Invitrogen, USA) according to the manufacturer's instructions. The RNA concentration was measured using a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Inc., Wilmington, DE, USA). Then, the agarose gel electrophoresis was used to determine RNA integrity. cDNA was generated using the Reverse Transcription Kit (Invitrogen, Carlsbad, CA, USA). Real-time quantitative polymerase chain reactions (RT-qPCR) were performed using the SYBR Green PCR kit protocol in the 7000 Sequence Detection System (Applied Biosystems, Carlsbad, CA, USA). U6 snRNA was used for normalization, and the relative lncRNA expression was calculated using the 2 −ΔΔCt method.
Culture of NP cells. The NP tissue was gently separated from the disc under aseptic condition, washed by phosphate-buffered saline (PBS) for three times, and then cut into small pieces with ophthalmic scissors. Subsequently, tissue was digested with PBS containing 0.025% type II collagenase (Invitrogen, Carlsbad, CA, USA) for 4 h followed by filtration and centrifugation at 500 g for 10 min. The supernatant was removed, the NP cells were seeded into culture dishes in DMEM/F12 medium (GE Healthcare Life Sciences, Logan, Utah, USA) containing 15% fetal bovine serum (FBS, Gibco, Shanghai, China) and 100 U/ml streptomycin/penicillin under 5% CO 2 , saturated humidity at 37 °C for 3 days. The culture medium was changed three times a week, and NP cells were subcultured at a ratio of 1:3 after reaching 80% confluence. Transwell invasion assays. The RAW264.7 macrophage cell line, purchased from the American Type Culture Collection (ATCC, Rockville, MD, USA), was cultured in DMEM supplemented with 10% FBS and 100 U/ml penicillin/streptomycin in a cell culture incubator at 37 °C and 5% CO 2 . The macrophages transwell invasion assay was performed as previously described. Briefly, a total of 1 × 10 6 macrophages were seeded in the upper wells of chambers (8-µm pore size; Corning Inc., Corning, NY, USA) with DMEM containing 0.1% FBS. 1 × 10 6 NP cells in the lower well were grown in 1 ml of DMEM containing 10 ng/ml TNF-α (Sigma, St. Louis, MO, USA). Twenty hours later, the invasive cells were stained with 2% crystal violet (Sigma), and counted in 5 high-power fields under the microscopic fields.
Western Blot. The cells were lysed using the RIPA buffer (Sigma, St. Louis, MO, USA). Total protein was separated by 12% sodium dodecyl sulphate-polyacrylamide (SDS-PAGE) electrophoresis and transferred to nitrocellulose membranes (Millipore, Billerica, MA, USA), followed by incubation with TRAF6, p-NFκB (Ser536), NFκB and GAPDH (Cell Signaling Technology, Boston, USA) overnight at 4 °C. After being rinsed thrice, the membranes were further incubated with a HRP-conjugated anti-IgG for 1 h at 37 °C. The protein was detected using an ECL system (Amersham Pharmacia, Piscataway, NJ, USA) were analyzed using the Quantity One software (BIO-RAD, USA).

TRAF6 viral infection.
Human TRAF6 virus and TRAF6 shRNA were constructed as our previous study described 14 . All assays were conducted 48 h after viral infection or shRNA transfection. Statistical analysis. Statistical analysis was performed using the StatView 5.0 software (SAS Institute, Cary, NC). All the data were expressed as mean ± standard deviation (SD). Student's t-tests were performed to compare the differences between two groups, and differences among three or more groups were evaluated by one-way ANOVA. The correlation between the expression of HCG18 and miR-146a-5p level, duration of symptoms and TRAF6 expression was determined by Spearman's correlation analysis. All experiments were performed independently in triplicate. P < 0.05 was considered statistically significant.