Silencing NDC80, RAD21 and BUB1B ameliorates intervertebral disc degeneration by promoting proliferation and inhibiting apoptosis of nucleus pulposus cells

Intervertebral disc degeneration (IVDD) is a commonly occurring musculoskeletal disorder, which is closely associated with low back pain. Accumulating evidence has demonstrated that dysregulated genes expression proles play important roles in pathogenesis of IVDD. Hence, the current study was aimed to identify key genes to understand underlying mechanisms and therapeutic targets of IVDD. Microarray datasets of GSE34095, GSE63492 and GSE45856 were downloaded to identify the hub genes that participate in the IVDD pathogenesis. After establishment of rat IVDD models, the expressions of NDC80, BUB1B and RAD21 in rat IVDD samples were evaluated by reverse transcription quantitative PCR (RT-qPCR) and immunochemistry. Subsequently, we assessed the proliferation, cycle and apoptosis of nucleus pulposus (NP) cells that transfected with siRNA-NDC80, siRNA-BUB1B and siRNA-RAD21. Our results showed indicated that NDC80, BUB1B and RAD21 were the key pathogenic genes with higher expression in IVDD rats, and silencing of NDC80, BUB1B and RAD21 gene could promote the aggrecan and collagen II synthesis, cell cycle and proliferation of NP cells, and inhibit NP cells apoptosis. RAD21 and BUB1B genes ameliorates intervertebral disc degeneration by promoting proliferation and inhibiting apoptosis of nucleus pulposus cells. These ndings suggest that NDC80, BUB1B and RAD21 may serve as potential therapeutic targets for IVDD.


Conclusion
Our study suggests that silencing NDC80, RAD21 and BUB1B genes ameliorates intervertebral disc degeneration by promoting proliferation and inhibiting apoptosis of nucleus pulposus cells.

Background
Intervertebral disc degeneration (IVDD) is a musculoskeletal disorder characterized by imbalanced extracellular matrix synthesis and breakdown, consequently contributing to herniated disks, spine instability and spinal stenosis [1,2]. Chronic lower back pain is considered to result from intervertebral disc degeneration, which brings a heavy economic burden to the society and families [3,4]. At present, the medical treatment of IVDD is still limited to pain relief and delaying surgery, due to lacking of understanding of underlying pathogenesis [5,6]. Therefore, in order to explore novel therapeutic approaches, there is an urgent need to identify the underlying molecular mechanisms of IVDD.
Except using classical experimental approaches to de ne IVDD, genetic factors have been considered to play emerging roles in the multifactorial etiology of IVDD [7,8]. Recently, the dysregulated genes expression pro les of IVDD have attracted increasing attention of researchers. Among them, MAP2K6 and RHOBTB2 are identi ed as two potential therapeutic targets of IVDD [9]. Periostin has also been proven to take part in the progression of human IVDD [10]. Hence, using high-throughput technology to screen out IVDD biomarkers in human samples has been shown to be promising to identify more precise criteria of disease diagnostic, classi cation and prognosis [11][12][13].
In this study, we identi ed NDC80, BUB1B and RAD21 genes as hub genes in the pathogenesis of IVDD utilizing the bioinformatics method. Besides, increased expression of NDC80, BUB1B and RAD21 in IVDD rats veri ed these genes associations with IVDD. Furthermore, we found that NDC80, BUB1B and RAD21 genes silencing could elevate the expression of aggrecan and collagen II, promote cell cycle and proliferation, and inhibit apoptosis of NP cells, which provides insight into the underlying mechanisms and therapeutic targets of IVDD.

Ethics statement
This study was performed with the approval of the Ethics Committee of Tianjin Medical University General Hospital. All animal experiments were performed under strict adherence with the Guide for the Care and Use of Laboratory Animal by International Committees.

Microarray Data Analysis
Gene expression data sets GSE34095, GSE63492 and GSE45856 were downloaded from Gene Expression Omnibus. For human NP cells in GSE34095, three degenerative samples were collected from IVDD patients and 3 non-degenerative samples were considered to be control group. MicroRNA expression pro ling of GSE63492 was obtained from 5 IVDD patients, compared with 5 cadaveric discs.
In relation to the miRNAs in GSE45856, three IVDD specimens and 3 traumatic intervertebral disc specimens were individually analyzed using microarray techniques. After data conversation and normalization [14], we performed differential analysis between degeneration samples and control [15]. We set |Log (fold change)| > 1 and adjust P value < 0.05 as the thresholds to screen out DEGs and differentially expressed miRNAs. Then, the target genes of the differentially expressed miRNAs were predicted by combining two online miRNA databases, miTarBase [16] and targetscan (http://www.targetscan.org).

Functional enrichment analysis of DEGs and target genes
In order to identify the underlying biological functions in IVDD, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis was performed on the DEGs using Database for Annotation, Visualization, and Integration Discovery (DAVID) database (https://david.ncifcrf.gov/) [17].
Protein-protein interaction (PPI) network construction The Search Tool for Retrieval of Interacting Genes (STRING) online tool was utilized to construct the PPI of DEGs. A con dence score > 0.4 was representative of statistical signi cance. Then, the visualization of PPI network was performed by Cytoscape software.

Animal grouping and modeling
A total of 25 male Sprague-Dawley (SD) rats (3 months old, 180 ± 20 g) were from the medical science experimentation center, SUN YAT-SEN UNIVERSITY (Guangzhou, China). The rats were raised individually and freely to eat and drink under 25°C conditions with humidity of 50~70% for one month. Next, 15 rats were randomly included in the IVDD group, while the remaining 10 rats were included in the normal group.
The rats were anaesthetized by intraperitoneal injection with 3% pentobarbital sodium (40 mg/kg, P3761, Sigma-Aldrich Chemical Company, St Louis MO, USA). A longitudinal incision was made along the mid line of the rats back. After incision of subcutaneous tissues, the fascia surrounding the intervertebral disc was isolated and the sacrospinalis was then stripped. The spinous process (L1-6) was excised with the interlaminar ligaments cut in order to establish the IVDD rat model. Rats of the normal group were only subject to an incision of the subcutaneous tissue and suture after their operation followed by a return to normal feeding. The rats were injected with 80, 000 U/d penicillin over a period of three days. After 3 months, the rats were euthanized through an intraperitoneal injection of excessive anesthesia for followup experiments.

Assessment of IVDD models
Three months after the operation, microcomputed tomography (micro-CT) and magnetic resonance imaging (MRI) were used to assess the alterations to disc height and vertebral endplate with degeneration. Three rats chosen randomly from both IVDD group and normal group were euthanized through an intraperitoneal injection of excessive anesthesia. Subsequently, micro CT and sagittal T2weighted images were obtained using SkyScan 1172 (SkyScan, Belgium) and 3.0-T MRI scanner (GE, Chicago, USA), respectively. Then, their respective intervertebral discs (L1-6) and adjacent vertebrae were collected. Histological staining including hematoxylin-eosin (HE) and safranin-O/Fast-greening staining was performed for assessment of cell morphology, extent of endplate ossi cation, and cartilage matrix of the endplate.

Real-Time quantitative Polymerase Chain Reaction
Three rats were randomly selected from both the IVDD and normal groups. Then, total RNA was extracted using the trizol method. RNA was reversely transcribed into cDNA using the Takara reverse transcription kit (RR047, Takara Biotechnology Ltd., Dalian, China). RT-qPCR was performed with instructions of SYBR®Premix Ex TaqTM II kit (Takara Biotechnology Ltd., Dalian, China). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the internal reference.

Isolation and cultivation of NP cells
The NP cells were collected from the remaining IVDD rats intervertebral discs (L1-6) that was digested with 0.25% trypsin (25200-056, Gibco Company, Grand Island, NY, USA) for 15 min at 37°C. The collected cells were then seeded into a 25 mm 2 culture ask with a density of 2 × 10 4 cells/mL and cultured in a 5% CO 2 incubator added with dulbecco's minimum essential medium (DMEM) medium (C11330500BT, Invitrogen, Car, Cal, USA) containing 3% fetal bovine serum (FBS) at 37°C. The third generation of NP cells were used for the following experiments. The NP cells were then implanted into a 6-well plate at a density of 2 × 10 5 cells/well with each well added with 1.3 mL serum-free Opti-MEM (31985-070, Gibco Company, Grand Island, NY, USA). Next, 5 µL lipofectamine 2000 (Invitrogen Inc., Carlsbad, CA, USA) with 3 μg of siRNA expressing plasmid were dissolved with 100 µL serum free Opti-MEM culture medium respectively, mixed and permitted to stand at 37°C for 5 min. After that, products were added into the corresponding wells with each well containing 200μL corresponding products. Then, the products were placed in incubator for 5 h, and exchanged with a fresh complete medium. Finally, the cells were collected after 48-h of transfection.
3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay Cells at the logarithmic growth phase and in a well-grown state were seeded into a 96-well culture plate at a volume of 100 μL per well with 1 × 10 4 cells per well. After cell adherence for 24 h, they were placed into a 5% CO 2 incubator at 37°C for further 24 h, 48 h and 72 h incubation. Later, 20 μL MTT solution (5 mg/mL) was added into each well for incubation purposes at 37°C for 4 h. After removal of the supernatant, each well was added with 100 μL dimethylsulfoxide (DMSO) and oscillated for 10 min. The OD value of each well at 490 nm was used to drawn the growth curve.

Flow cytometry
After 48h of transfection, the culture medium was discarded and the cells concentration was adjusted to 1 × 10 6 cells/mL. The cells stained with propidium iodide (PI) containing RNase (GR1-25, SBS Genetech, Beijing, China) were detected at 488 nm using a ow cytometer (FACS Calibur, Becton, Dickinson and Company, New Jersey, USA) to cell cycle. The transfected cells at the logarithmic growth phase were used to detect cell apoptosis. According to the instructions of Annexin-V-FITC/PI apoptosis kit (KA3805, Abnova, Walnut, CA, USA), ow cytometry at 488 nm using the 515 nm and 560 nm band pass lter to examine the FITC uorescence and PI uorescence, respectively.

Statistical analysis
One-way analysis of variance (ANOVA) was used for comparisons among multiple groups with SPSS 21.0 software. Data were expressed as mean ± standard deviation, and p < 0.05 was regarded as statistically signi cant.

RAD21, NDC80 and BUB1B were identi ed as DEGs
Initially, bioinformatic analysis was applied to identify the differentially expressed genes (DEGs) of IVDD. In regard to GSE34095, a total of 153 DEGs were screened out, including 111 up-regulated genes and 42 down-regulated genes. Among them, the RAD21, NDC80 and BUB1B genes were upregulated (Fig. 1a). In addition, seven up-regulated miRNAs and 8 down-regulated miRNAs were obtained from GSE63492, among which miR-5100 was one of the down-regulated miRNAs (Fig. 1b). Besides, six up-regulated miRNAs and 9 down-regulated miRNAs were obtained from GSE45856. Among them, miR-1246 and miR-3908 were detected as down-regulated miRNAs (Fig. 1c). Based on the prediction analysis, the targeted genes of miR-5100, miR-1246 and miR-3908 were con rmed to be RAD21, NDC80 and BUB1B respectively. Therefore, RAD21, NDC80 and BUB1B were signi cantly upregulated in degenerative NP cell samples.
RAD21, BUB1B and NDC80 were the hub genes related with cell cycle To further elucidate the molecular mechanisms of IVDD, we constructed a PPI network of DEGs from NP cells. Furthermore, it was noted that the NDC80, BUB1B, and RAD21 were the hub genes of NP cells within the PPI network (Fig. 1d).
GO enrichment analysis revealed that the RAD21 and BUB1B genes participate in the cell cycle, mitotic cell cycle, cell apoptosis, programmed cell death and cell death processes. The NDC80 gene was determined to be related to cell cycle as well as in the mitotic cell cycle (Table 1). KEGG pathway enrichment analysis revealed that the RAD21 and BUB1B genes were involved in cell cycle-related signaling ( Table 2).

Veri cation Of Ivdd Rats
To assess whether a successful IVDD rat model was established, the radiologic imaging and histological staining were analyzed. Compared to the normal, the IVDD group of micro-CT scans showed decrease in disc height, increase in osteophyte formation at the vertebral edge. Besides, T2-weighted images showed signi cant reduction in signal intensity (Fig. 2b). As illustrated in Fig. 2c, IVDD group exhibited a signi cant intervertebral disc degeneration including a shrunken NP, decreased number and uneven distribution of NP cells, unclear boundary of NP and annulus brosis, swelling and fractured inner annulus brosis, irregular proliferation and calci cation of the cartilage endplate. These ndings proved the successful IVDD rat models.
Increased expressions of NDC80, BUB1B and RAD21 in IVDD rats RT-qPCR and immunohistochemistry analysis were performed to examine the expression of NDC80, BUB1B and RAD21 in the intervertebral disc tissues of rats (Fig. 3). The results demonstrated that IVDD rats exhibited an elevated mRNA and protein expression of NDC80, BUB1B and RAD21, which was in agreement with the bioinformatics results.
Silencing of NDC80, RAD21 and BUB1B increased the expression of collagen II and aggrecan After transfection for 48 h, compared to no matter the blank or NC groups, the siRNA-NDC80-1, siRNA-BUB1B-2 and siRNA-RAD21-2 groups exhibited the most distinctively down-regulated mRNA expression of NDC80, BUB1B and RAD21, which were used for observing collagen II content (Fig. 4a-c). More importantly, after transfection, the NP cells exhibited a signi cantly increased content of collagen II aggrecan in the siRNA-NDC80, siRNA-BUB1B and siRNA-RAD21 groups (Fig. 4d-f). These results con rmed the silence e ciency, and showed that silencing of NDC80, BUB1B and RAD21 genes could increase the synthesis of collagen II and aggrecan.
Silencing NDC80, BUB1B and RAD21 promoted proliferation of NP cells As shown in Fig. 4G, compared with the blank group and the NC group, the proliferation rate of the siRNA-NDC80 group, the siRNA-BUB1B group and the siRNA-RAD21 group was markedly increased at the 48 h and 72 h time points (all p < 0.05). The above results demonstrated that silencing of NDC80, BUB1B and RAD21 genes could promote the proliferation of NP cells.
Silencing NDC80, BUB1B and RAD21 accelerated cell cycle and inhibited cell apoptosis In terms of the cell cycle distribution and apoptosis rate between the blank and NC groups, no signi cant difference was found (p > 0.05). However, the G1/G0 phase in the siRNA-NDC80, siRNA-BUB1B and siRNA-RAD21 groups were decreased, while the S phase in these gene silencing groups were increased (p < 0.05). There were no obvious changes in the G2/M phase (p > 0.05). Furthermore, after the NDC80, BUB1B and RAD21 gene silencing, the cell apoptosis rate decreased signi cantly. Collectively, these results suggested that silencing od NDC80, BUB1B and RAD21 genes accelerated cell cycle, which may result in the inhibition of cell apoptosis (Fig. 5).

Discussion
Although much effort and resources have been invested, the underlying pathogenesis of IVDD still remains poorly understood. Recently, the dysregulated genes expression pro les of IVDD have attracted increasing attention of researchers. To investigate key pathogenic mechanism of IVDD, we identi ed NDC80, BUB1B and RAD21 as hub genes in the degenerative NP cells utilizing the bioinformatics analysis. Subsequently, increased expressions of NDC80, BUB1B and RAD21 were con rmed in IVDD rats, and silencing NDC80, BUB1B and RAD21 played a protective role in intervertebral disc degeneration by promoting proliferation and inhibiting apoptosis of nucleus pulposus cells.
According to GO enrichment analysis and KEGG analysis, NDC80, BUB1B and RAD21 were predicted to participate in several biological processes and pathways including cell death, apoptosis and cell cycle. The protein encoded by NDC80 con nes itself to kinetochore and subsequently acts to mediate the formation of the kinetochore-microtubule structure, which is crucial for the stable kinetochore-microtubule interaction [18][19][20]. Recent studies have provided evidence highlighting a correlation between the overexpression of NDC80 and the over-activation of the mitotic checkpoint [21]. Mitotic checkpoints could prevent separation errors by adjusting the anaphase time until all of chromosomes have been precisely been attached to the spindle microtubules [22]. In mammal cells, BUB1B gene encodes BubR1 protein, which is also a vital component of mitotic checkpoints that ensures the synchrony of chromosome segregation [23]. Yamamoto et al. found that the up-regulation of BUB1B may be a physiologically compensatory mechanism to the absence of normal checkpoints function [24]. Besides, exogenous BUB1B prevents chromosomal instability in a manner of triggering apoptosis [25]. Moreover, RAD21 is one of the four subunits of cohesion, participating in repairing of DNA double-strand breaks, and growth of mitosis [26]. C-terminal cleavage of RAD21 could form degradation products, translocation of which to the cytoplasm initiates apoptosis and causes ampli cation of the cell death signal [27][28][29]. Hence, considering the importance of three interrelated genes in chromosomal stability and increased expressions of NDC80, BUB1B and RAD21 in IVDD rats, the dysregulation of NDC80, BUB1B, and RAD21 may contribute to the disruption of mitosis and subsequent apoptosis of NP cells as IVDD progresses.
Furthermore, in vitro transfection results veri ed that NDC80, BUB1B and RAD21 gene silencing accelerated cell cycle progression and suppressed NP cells apoptosis. Consistent with our ndings, previous studies have suggested that inhibiting of BUB1, NDC80 and RAD21 expression could regulate the cell cycle progression and apoptosis [30][31][32][33]. Additionally, the loss of extracellular matrix (ECM) such as collagen II and aggrecan has been highlighted as a marker of IVDD [34,35]. We found that silencing NDC80, BUB1B and RAD21 could enhance synthesis of ECM component, ameliorating disc degeneration in a rat model of IVDD.
There are also some limitations in our study, which should be taken into consideration when interpreting the results. First, the underlying mechanisms on how NDC80, BUB1B and RAD21 genes silencing promoted proliferation and inhibited apoptosis of nucleus pulposus cells were not thoroughly identi ed. Secondly, whether there are interactions between NDC80, BUB1B and RAD21 genes or how they interact with each other are not clear. In the end, the role of NDC80, BUB1B and RAD21 genes in human intervertebral disc degeneration also needs more research.

Conclusion
In conclusion, our study identi es three hub genes of NDC80, BUB1B and RAD21 via bioinformatics analysis, and indicates that silencing of NDC80, RAD21 and BUB1B genes ameliorates intervertebral disc degeneration by promoting proliferation and inhibiting apoptosis of nucleus pulposus cells. These ndings suggest that NDC80, BUB1B and RAD21 may serve as potential therapeutic targets for IVDD. Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests

Funding
The study was supported by grant projects: International Cooperation Program of National Natural Science Foundation of China (81620108018) and National Natural Science Foundation of China (81930070, 81802197, 81772342) provided funds in the experimental design, animal purchase and feeding, and reagent purchasing.
Hebei Provincial Natural Science Foundation (H2019110028) and Tianjin key research and development plan, key projects for science and technology support (19YFZCSY00660, 19JCZDJC36300) provided funds in the data analysis and interpretation, and manuscript polishing.
Authors' contributions BZ, YYG, WG and WXL: experimental work, and manuscript writing. PP, CX and JYH: data analysis, experimental work. CS and RWD: manuscript editing. XHK and SQF: research design. All authors read and approved the nal manuscript.  The effect of silencing NDC80, BUB1B and RAD21 on the expression of collagen II and aggrecan and proliferation of NP cells. a-c, Silencing e ciency of three pairs of siRNAs in the siRNA-RAD21, siRNA-NDC80, and siRNA-BUB1B group. d-e, Silencing NDC80, BUB1B and RAD21 increased the expression of collagen II of NP cells, determined by immunohistochemistry (n = 6). f, Silencing NDC80, BUB1B and RAD21 increased aggrecan content of NP cells, determined by RT-qPCR (n = 6). g, Silencing NDC80, BUB1B and RAD21enhanced proliferation of NP cells, determined by MTT assay. *p < 0.05 compared with the blank and NC groups. NC, negative control; NP, nucleus pulposus Figure 5 Silencing NDC80, BUB1B and RAD21 accelerated cell cycle and inhibited cell apoptosis of NP cells. a, cell cycle distribution in the blank, NC, siRNA-NDC80, siRNA-BUB1B and siRNA-RAD21 groups, detected by ow cytometry. b, cell apoptosis in the blank, NC, siRNA-NDC80, siRNA-BUB1B and siRNA-RAD21 groups, detected by ow cytometry. *p < 0.05 compared with the blank and NC groups. NC, negative control; NP, nucleus pulposus

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. TheARRIVEGuidelinesChecklist.pdf