Nuclear factor‐kappa B‐dependent X‐box binding protein 1 signalling promotes the proliferation of nucleus pulposus cells under tumour necrosis factor alpha stimulation

Abstract Objectives Tumour necrosis factor alpha (TNF‐α) expressed by nucleus pulposus cells (NPCs) plays a critical role in intervertebral disc (IVD) degeneration. A key unfolded protein response (UPR) component, X‐box binding protein 1 (XBP1) and nuclear factor‐kappa B (NF‐κB) are essential for cell survival and proliferation. The aim of our study was to elucidate the roles of XBP1 and NF‐κB in IVD degeneration (IDD). Materials and methods Rat NPCs were cultured with TNF‐α in the presence or absence of XBP1 and NF‐κB‐p65 small interfering RNA. The associated genes and proteins were evaluated through quantitative real‐time PCR, Western blot analyses and immunofluorescence staining to monitor UPR and NF‐κB signalling and identify the regulatory mechanism of p65 by XBP1. Cell counting kit‐8 assay, cell cycle analysis and related gene and protein expression were performed to examine the proliferation of NPCs. Results The acute exposure of TNF‐α accelerated the proliferation of rat NPCs by activating the UPR/XBP1 pathway. XBP1 signalling favoured the phosphorylation and nuclear translocation of p65 subunit of NF‐κB. The activation of NF‐κB in the later phase also enhanced NPC proliferation. Conclusions Unfolded protein response reinforces the survival and proliferation of NPCs under TNF‐α stimulation by activating the XBP1 pathway, and NF‐κB serves as a vital mediator in these events. The XBP1 signalling of UPR can be a novel therapeutic target in IDD.

phenomenon, we supposed that different cell stress statuses lead to different biological changes of apoptosis or proliferation NPCs under TNF-α stimulation. And whether a healthy IVD can recover after its exposure to TNF-α remains uncertain. 5 These matters all need to be further clarified in our study.
Nuclear factor-kappa B is vital for genes involved in immune response, inflammation, cell survival, adhesion and proliferation. 6 The canonical NF-κB signalling has been considered the central mediator of the inflammatory process. 7 Inflammation and NF-κB have a double-edged role in cells. On the one hand, activation of NF-κB is part of the immune defence, which usually results in the upregulation of anti-apoptotic genes thereby providing cell survival mechanism to withstand the physiological stress that triggered the inflammatory response. 8 Under normal conditions, NF-κB plays a critical role in protecting cells against apoptosis, 9 stimulating cell proliferation 10 and promoting cell migratory. On the other hand, persistent and aberrant activation of NF-κB stimulates IVD cells to synthesize many pro-inflammatory cytokines (such as TNF-α, IL-1, IL-6 and IL-8) to amplify inflammatory responses. Moreover, NF-κB signalling was shown to contribute to the upregulation of matrix metalloproteinases (MMPs) and a disintegrins and metalloproteinases with thrombospondin motifs (ADAMTSs), which exacerbate the loss of ECM, especially aggrecan (ACAN) and collagen Ⅱ (COL2), thereby aggravating the degeneration of IVD. 8 Clearly, either too little or too much NF-κB activity is detrimental; furthermore, NF-κB signalling pathway is also crucial in maintaining a healthy homeostasis.
Nuclear factor-kappa B homodimers and heterodimers are formed by five family members, namely p65 (RelA), p50, p52, RelB and c-Rel, and the most canonical form is the p65/p50 heterodimer. 11 The activity of these dimers is regulated by an inhibitory protein known as the inhibitor of κB (IκB). 12,13 As a pro-inflammatory cytokine, TNF-α can activate NF-κB signalling. In resting cells, the inactive form of NF-κB is sequestered in the cytoplasm combined with IκB. Extracellular stimuli, such as TNF-α, activate the phosphorylation and proteolytic degradation of IκB. The released p65/p50 then rapidly translocates to the nucleus and regulates the transcription of target genes, including cytokines, cell adhesion molecules, hematopoietic growth factors and prosurvival genes. 14 Endoplasmic reticulum (ER) performs important cell functions and plays a crucial role in the folding of newly synthesized proteins. 15 ER stress elicits an unfolded protein response (UPR) to facilitate repair and thus reestablish normal ER functioning through the mediation of three ER-resident transducers, namely inositol-requiring enzyme 1 (IRE1), protein kinase RNA-like ER kinase (PERK) and activating transcription factor 6 (ATF6). 16 Various agents inducing ER stress and UPR activation can trigger NF-κB signalling. 17 The activated NF-κB can translocate to the nucleus and stimulate target gene transcription. As such, a direct ER-nuclear signal transduction pathway is established. 18 Although studies have yet to fully explain the underlying mechanism of NF-κB activation in UPR, studies have revealed that the activation of NF-κB can be induced by PERK-mediated eIF2α phosphorylation, XBP1 splicing, IRE1 phosphorylation, oxidative stress and calcium disturbances. 13,19 Many scholars found that TNF-α has extensive connections with ER stress and UPR. [20][21][22] In addition, our previous studies demonstrated that TNF-α activates ER stress, and the consequent initiation of UPR reinforces the survival and proliferation of NPCs. 23 Considering the importance of NF-κB in immunity, inflammation, anti-apoptosis and cell proliferation, 12 we hypothesized that UPR exerts cytoprotection and promotes NPC proliferation under TNF-α stimulation by activating NF-κB signalling. In the current research, we conducted experiments to verify whether UPR could trigger the NF-κB signalling pathway in NPCs, the function of NF-κB signalling and its association with UPR in TNF-α-induced biological changes in NPCs. Our data suggested that the NF-κB-dependent XBP1 signalling pathway represented an adaptive mechanism that prevented NPC apoptosis and promoted cell proliferation under TNF-α stimulus.

| Cell isolation and culture
Nucleus pulposus cells were collected from the tail discs of Sprague Dawley rats in accordance with previously described methods, 24 resuspended in DMEM/F12 containing 10% FBS and antibiotics and incubated at 37°C with 5% CO 2 in a humidified incubator. The cells after the third passage were identified based on the newly defined NPC phenotype 25 and used in the subsequent studies ( Figure S1).

The experimental protocol was approved by the Institutional Animal
Care and Use Committee of Southeast University (Nanjing, China).

| Cell proliferation assay
Cell proliferation was quantified using a cell counting kit-8 (CCK-8; KeyGen, Nanjing, China). The NPCs were cultured in 96-well plates at a density of 2000 cells/well with or without siRNA transfection.
Then, each well was added with CCK-8 reagent at various time points and incubated at 37°C in 5% CO 2 in air atmosphere for 4 hours.
Absorbance at 450 nm was detected with Multiskan MK3 (Thermo Scientific, Waltham, MA, USA) to calculate cell viability.

| Immunofluorescence staining
Nucleus pulposus cells were seeded into 24-well plates at 8000 cells per well and allowed to adhere to the glass bottom for 24 hours. The cells with or without siRNA interference were treated using TNF-α (10 ng/mL), fixed with 4% paraformaldehyde at room temperature and permeabilized with 0.3% Triton X-100. The treated cells were

| Cell cycle analysis through flow cytometry
The NPCs were harvested through trypsinization, fixed with 70% cold ethanol for more than 2 hours, centrifuged, resuspended in 50 μL of
Purified RNA was reverse transcribed using PrimeScript TM RT Master Mix (TaKaRa, RR036A, Dalian, China). The primer sequences are presented in Table 1. Gene transcripts were quantified using SYBR Premix EX Taq TM (TaKaRa, RR420A) through real-time PCR (Applied Biosystems, Foster City, CA, USA), and relative gene expression was measured using the 2 −ΔΔCt method with β-actin gene as an internal control.

| Western blot analysis
Total protein extracts from NPCs were obtained through a wholecell lysis assay (KeyGen). Protein concentration was determined with a BCA assay (Beyotime, Shanghai, China). Proteins were separated with 10% SDS-PAGE and transferred onto PVDF membranes.
The membranes were blocked with 5% skimmed milk in TBST for 1.5 hours and incubated with primary antibodies at 4°C overnight ( Table 2). The membranes were then incubated with secondary antibodies (Abcam, ab6721; 1:5000) diluted in 5% bovine serum albumin (BSA)/TBST at room temperature for 2 hours. Bands were detected and assessed through densitometric analysis. Protein expression level was normalized to that of vinculin.

| Statistical analysis
Data were presented as mean ± SD of at least three independent experiments. Statistical analyses were carried out using GraphPad Prism 6 (GraphPad Software, Inc, La Jolla, CA, USA). Two-tailed Student's t test was conducted to compare differences among two groups, and one-way ANOVA was conducted to compare differences among multiple groups. P-values <0.05 were considered statistically significant.

| TNF-α promoted the proliferation of NPCs
The rat NPCs were treated with a gradient TNF-α concentration to explore the biological effects of TNF-α on the proliferation of NPCs.
CCK-8 assay was performed to calculate the rate of cell proliferation. After 24 hours, the proliferation rate of NPCs increased upon exposure to the TNF-α stimulus ( Figure 1A). To further validate the role of TNF-α on the proliferation of rat NPCs, we subjected the pro-

| TNF-α activated UPR signalling and XBP1 signalling could be silenced by XBP1 siRNA
Our previous studies showed that the proportion of apoptotic cells increases at the early stage when TNF-α concentrations exceed 20 ng/mL. 23 Therefore, we selected a TNF-α concentration of The value of XBP1s/XBP1u decreased with XBP1 siRNA interference ( Figure 2D,E).

| UPR signalling regulated the TNF-α-induced proliferation of NPCs through XBP1 signalling
To investigate the effect of UPR signalling on NPC proliferation induced by TNF-α, we silenced the XBP1 signalling of UPR by using siRNA. After the treatment with TNF-α (10 ng/mL) or siRNA specific to XBP1 was administered, the proliferation rate of NPCs was meas-

| UPR initiated the secondary activation of NF-κB
The branches of UPR have a crosstalk with the NF-κB pathway. 13 In this study, we aimed to validate whether the activation of UPR triggered the NF-κB signalling in NPCs. We treated the NPCs with tunicamycin (TM), a widely used and recognized ER stressor that inhibits protein N-glycosylation. 26,27 The activation of NF-κB by TNF-α is rapid, often occurring within minutes following stimulation. By contrast, ER stress and UPR are slow, cumulative process that increases as additional proteins accumulate in the ER over time. In our study, NF-κB was rapidly induced by TNF-α. An increase in the protein expression of p-p65 was detected as early as 10 minutes after the treatment.
The maximal activation occurred 20 minutes after stimulation; after 30 minutes, the p-p65 expression was markedly reduced, and the expression increased again at 12-24 hours ( Figure 4A,B). By contrast, TM-induced NF-κB activation initially became detectable 12 hours The results were presented as mean ± SD (n = 3). *P < 0.05; **P < 0.01; ***P < 0.001 F I G U R E 2 Activation of unfolded protein response (UPR) signalling after tumour necrosis factor alpha (TNF-α) treatment in rat nucleus pulposus cells (NPCs) and X-box binding protein 1 (XBP1) siRNA inhibition of the TNF-α-induced activation of the XBP1 pathway. A, After treatment with 10 ng/mL TNF-α, the activation of UPR signalling was assessed in terms of the related protein expression at different time intervals (0, 6, 12 and 24 h). B, UPR markers in NPCs were analysed through Western blot, and the protein expression levels of BiP were normalized to that of vinculin. The protein expression levels of p-eIF2α and XBP1s were normalized to those of total eIF2α (t-eIF2α) and XBP1u. NPCs were treated with control siRNA (siCtrl) or XBP1 siRNA (siXBP1) for 24 h and subsequently exposed to 10 ng/mL TNF-α for another 24 h. C, The mRNA expression of XBP1 in NPCs was quantified through qRT-PCR. The mRNA expression levels were normalized to that of β-actin. D,E, XBP1s was evaluated through Western blot, and the protein expression levels were normalized to those of XBP1u. The results were presented as mean ± SD (n = 3). *P < 0.05; **P < 0.01 showed that additional p-p65 translocated into the nucleus at 12 and 24 hours compared with that at 0 and 6 hours under TNF-α stimulus ( Figure 4E,F). We hypothesized that the slow and steady increase in the kinetics of TNF-α-induced NF-κB activation after 12 hours was activated by the initiation of UPR.

| TNF-α-mediated NF-κB induction in the later phase was inhibited by XBP1 siRNA
To determine the underlying interaction between XBP1 and NF-κB signalling and verify our speculations of NF-κB activity regulation by XBP1, we silenced XBP1 with siRNA in rat NPCs. We performed immunofluorescence staining assay and found in the immunofluorescence images that p-p65 was localized in the cytoplasm under siCtrl treatment. With TNF-α treatment, p-p65 translocated to the nucleus. XBP1 siRNA not only silenced XBP1 signalling, but also inhibited the nuclear translocation of p-p65 ( Figure 5A,B).
The mRNA level of p65 was lower in the XBP1 siRNA-treated cells than in the siCtrl-treated cells under TNF-α stimulation ( Figure 5C).
Furthermore, Western blot analysis showed that the knockdown of XBP1 with siRNA decreased the p65 and p-p65 protein level and also decreased the value of p-p65/p65 considerably, thereby suggesting that XBP1 regulated NF-κB signalling in NPCs ( Figure 5D,E).

| NF-κB signalling regulated the TNF-α-induced proliferation of NPCs
To elucidate the connection of NF-κB signalling and TNF-α-induced NPC proliferation, we knocked down the NF-κB pathway through RNA interference. The silencing by p65 siRNA was verified by the decreased expression levels of p65 and p-p65 proteins compared with that of siCtrl ( Figure 6A,B). After the treatment with TNF-α (10 ng/mL) or p65 siRNA was administered, the proliferation rate of NPCs was measured through CCK-8 assay, and p65 silencing decreased the TNF-α-induced NPC proliferation at different time points ( Figure 6C). We also detected the proliferating antigen Ki67

| D ISCUSS I ON
In our observation, the acute exposure of TNF-α increases the proliferation of rat NPCs and activates UPR signalling and NF-κB signalling. In our previous studies, we demonstrated that ER stress facilitates the survival and proliferation of NPCs under TNF-α stimulus by activating UPR. 23 To the best of our knowledge, our work was the first to link UPR to NF-κB signalling in the proliferation of NPCs. We also confirmed the proliferative effect of XBP1 on NPCs under TNF-α stimulation. Hence, UPR-activated NF-κB signalling and its association with cell proliferation were revealed in our present research.
Inflammatory responses and cellular loss elicited by TNF-α are consid- Various genetic or environmental conditions that perturb the ER lead to misfolded protein accumulation in a condition known as ER stress. In response, cells adapt to such stress via subsequent signalling pathways, including UPR. 32  branch is the main pathway in UPR-associated NF-κB activation.
Once UPR is activated, the phosphorylated IRE1 binds to TRAF2, leading to the phosphorylation and degradation of IκB and the nuclear translocation of NF-κB. 33,39 It is well known that inflammation is closely related to cell viability and cell proliferation. 40 NF-κB plays an important role in cell F I G U R E 4 Kinetics of nuclear factor-kappa B activation by tumour necrosis factor alpha (TNF-α) and tunicamycin (TM). Nucleus pulposus cells (NPCs) were treated with either 10 ng/mL TNF-α or 100 ng/mL TM at various time points. A,B, p65 protein phosphorylation with TNF-α treatment was analysed through Western blot, and proteins were extracted to detect the expression of p-p65. C,D, p65 protein phosphorylation with TM treatment was examined through Western blot. The protein expression levels of p-p65 were normalized to that of p65. E, The nuclear translocation of p-p65 under the TNF-α stimulus was subjected to immunofluorescence studies (original magnification, ×400). F, p-p65-positive cells in the nucleus were quantified. The results were presented as mean ± SD (n = 3). *P < 0.05; **P < 0.01; ***P < 0.001 proliferation as the key regulator of inflammation. Besides that, it has been proved that there are extensive crosstalks between inflammation and UPR. 41  One research reported that NF-κB-p65 activation promoted vascular smooth muscle cell proliferation under inflammation through the microRNA-17/RB pathway. 43 Zhang et al 6 found that the PAK5mediated phosphorylation and nuclear translocation of NF-κB-p65 promote the proliferation of breast cancer cells in vitro and in vivo. In our study, to demonstrate the hypothesis that TNF-α-induced NPC proliferation associated with the NF-κB pathway, we knocked down NF-κB-p65 with p65 siRNA and found that cells blocking the p65 F I G U R E 5 X-box binding protein 1 (XBP1) regulates nuclear factor-kappa B signalling by regulating p65 transcription and p-p65 expression. The cells were exposed to 10 ng/mL tumour necrosis factor alpha with or without pretreatment with siCtrl and siXBP1 for 24 h. A, In immunofluorescence double staining, the cells were stained initially with XBP1 antibody, Alexa Fluor 488 (green), p-p65 antibody and Alexa Fluor 647 (red), and the nucleus was counterstained with DAPI (blue) and examined through fluorescence microscopy (original magnification, ×400). B, XBP1-and p-p65-positive cells were quantified. C, Total RNA was extracted, and the mRNA expression of p65 was measured through qRT-PCR. D,E, Proteins were extracted to detect the expression of p-p65 and p65 through Western blot analysis, and the protein expression levels of p-p65 were normalized to that of p65. The results were presented as mean ± SD (n = 3). *P < 0.05; **P < 0.01 F I G U R E 6 Role of nuclear factor-kappa B (NF-κB) signalling in rat NPC proliferation induced by tumour necrosis factor alpha (TNF-α).
The cells were exposed to 10 ng/mL TNF-α with or without p65 siRNA (siP65) pretreatment for 24 h. A,B, NF-κB-p65 interference was verified through Western blot analysis. Proteins were extracted to detect the expression of p-p65 and p65. The expression levels of p-p65 were normalized to that of p65. C, Cell proliferation was evaluated through CCK-8 analysis at different time points. D,E, Immunofluorescence staining of Ki67 (original magnification, ×200) and quantification of Ki67-positive cells. F,G, The proportions of cells in each phase were examined through flow cytometry. H,I, Cyclin D1 and cyclin B1 expression levels in NPCs subjected to different treatments were measured through Western blot analysis and normalized to that of vinculin. The results were presented as mean ± SD (n = 3). *P < 0.05; **P < 0.01 pathway proliferated more slowly than the control cells did under TNF-α stimulus. All of these studies confirmed that NF-κB-p65 signalling was closely related to cell proliferation.

| CON CLUS I ON S AND PER S PEC TIVE S
Our experiments confirmed that XBP1 signalling was effective in activating NF-κB signalling by upregulating p-p65 expression and nuclear translocation. A regulatory connection between XBP1 and NF-κB survival signalling was determined. The proliferation of NPCs under TNF-α stimulation was regulated via the NF-κB-dependent IRE1/XBP1 pathway. This study suggested that the XBP1/NF-κB signalling could be a novel biological therapeutic target for the treatment of the cellular loss of NPCs under inflammation during IVD degeneration.

ACK N OWLED G EM ENTS
The research was supported by grants from the National Natural