Reactive Oxygen Species Regulate Endoplasmic Reticulum Stress and ER-Mitochondrial Ca2+ Crosstalk to Promote Programmed Necrosis of Rat Nucleus Pulposus Cells under Compression

Programmed necrosis of nucleus pulposus (NP) cells caused by excessive compression is a crucial factor in the etiopathogenesis of intervertebral disc degeneration (IVDD). The endoplasmic reticulum (ER) and mitochondria are crucial regulators of the cell death signaling pathway, and their involvement in IVDD has been reported. However, the specific role of ER stress (ERS) and ER-mitochondria interaction in compression-induced programmed necrosis of NP cells remains unknown. Our studies revealed that compression enhanced ERS and the association between ER and mitochondria in NP cells. Suppression of ERS via 4-phenylbutyrate (4-PBA) or ER-mitochondrial Ca2+ crosstalk by inhibiting the inositol 1,4,5-trisphosphate receptor, glucose-regulated protein 75, voltage-dependent anion-selective channel 1 complex (IP3R–GRP75–VDAC1 complex) protected NP cells against programmed necrosis related to the poly(ADP-ribose) polymerase (PARP) apoptosis-inducing factor (AIF) pathway. Moreover, excessive reactive oxygen species are critical activators of ERS, leading to mitochondrial Ca2+ accumulation and consequent programmed necrosis. These data indicate that ERS and ER-mitochondrial Ca2+ crosstalk may be potential therapeutic targets for the treatment of IVDD-associated disorders. These findings provide new insights into the molecular mechanisms underlying IVDD and may provide novel therapeutic targets.


Introduction
As the most common musculoskeletal disorder in outpatients, low back pain (LBP) causes huge economic losses in the global health system [1]. In the United States, this acute illness results in a loss of more than $100 billion in annual health care costs [2]. Intervertebral disc degeneration (IVDD) is the most common cause of LBP [3]. Excessive mechanical loads play a significant role in the etiology of IVDD [4]. Unphysiological loading exacerbates disc degeneration by accelerating disc cell death, leading to progressive loss of extracellular matrix and disc bioactivity [5]. However, the mechanisms underlying mechanical load-induced nucleus pulposus (NP) cell death have not been completely elucidated. Therefore, it is paramount to understand the molecular mechanisms of NP cell death under excessive mechanical loading conditions to identify effective therapies for IVDD treatment.
Mounting evidences indicate that programmed necrosis plays a greater role in the development of IVDD than the other two programmed cell death, apoptosis and autophagic cell death [6]. The most intuitive evidence is that necrotic cells in degenerated intervertebral discs account for more than 80% of the total [7]. In our previous study, NP cells showed mainly necrotic morphology changes under harmful stimuli, and inhibition of programmed necrosis by Nec-1 evidently retarded NP cell death [8]. Inhibition of apoptosis did not effectively relieve compression-induced cell death [9]. Therefore, mechanical load-induced NP cell death is mainly attributed to programmed necrosis. However, the underlying molecular mechanisms remain unclear.
The endoplasmic reticulum (ER) is the primary location for synthesis and maturation of proteins in response to cellular stimuli [10]. Additionally, ER is an essential location for intracellular Ca 2+ store that plays a crucial role in signal transduction [11]. Under severe or prolonged ER dysfunction, ER stress (ERS) triggers cell death by the release of Ca 2+ and subsequent triggering of a series of signal transduction pathways. Increasing evidence supports the involvement of ERS-initiated cell death in IVDD [12,13]. Zhao et al. found that disc degeneration was concomitant with increased cell death and upregulation of ERS markers, caspase-12 and the 78 kDa glucose-regulated protein (GRP78) [14]. Wang et al. reported that IVDD at the mild stage showed a strong upregulation of ERS markers, including GRP78, growth arrest-and DNA damage-inducible gene 153, and caspase-12 [15]. However, the specific role of ERS in compressioninduced programmed necrosis of NP cells remains unclear, and it is crucial to understand the underlying mechanisms for developing alternative treatment options for IVDD.
Mitochondrial dysfunction is a common pathophysiological change that occurs under disc overloading and contributes to IVDD [16]. Recent studies have shown that the mitochondria and ER interact physically and functionally to regulate their functions [17]. However, it is unclear how the interaction between ER and mitochondria is involved in compression-induced programmed necrosis of NP cells. Previous studies have confirmed that the ER couples with the mitochondria and an inositol 1,4,5-trisphosphate receptor (IP 3 R), glucose-regulated protein 75 (GRP75), voltagedependent anion-selective channel 1 (VDAC1) complex (IP 3 R-GRP75-VDAC1 complex) is present at the ERmitochondria interface, which is considered critical determinants of cell survival or death by exerting intracellular Ca 2+ efflux into the mitochondria [18]. However, the involvement of the IP 3 R-GRP75-VDAC1 complex in compressioninduced NP cell death has not been clarified.
In the current study, we demonstrated that ERS plays a critical role in compression-induced programmed necrosis of NP cells. Mechanical loading enhanced the association between ER and mitochondria, which in turn stimulated Ca 2+ efflux to mitochondria through the IP 3 R-GRP75-VDAC1 complex, resulting in poly(ADP-ribose) polymerase (PARP)/apoptosis-inducing factor (AIF) pathway-related cell programmed necrosis. This study also clarified the mechanism by which ERS contributes to compression-induced programmed necrosis of NP cells. Our findings may provide a molecular basis for the inhibition of ERS or aberrant Ca 2+ translocation in the treatment of IVDD.

Materials and Methods
2.1. Cell Culture. All animal experiments and protocols were approved by the animal experimentation ethics committee at the Huazhong University of Science and Technology. After euthanasia by intraperitoneal injection of pentobarbital sodium (100 mg/kg), NP tissues (L1-L7) of Sprague-Dawley rats (male, weight 250-300 g) were harvested using forceps, with a horizontal incision on the annulus fibrosis. NP tissues were then digested for 15 min in 0.25% type II collagenase (Gibco, USA) at 37°C. The digested NP tissue was cultured in Dulbecco's modified Eagle medium/Ham's F-12 (Gibco, USA) supplemented with 10% fetal bovine serum (Gibco, USA) and 1% penicillin/streptomycin (Sigma, USA) at 37°C. Cells were passaged till they reached 80-90% confluency. We used the second-generation cells in the following experiments.

Excessive Compression
Loading on Rat NP Cells. A pressure apparatus was used to create excessive mechanical loading. Cells were seeded in 6-well plates (5:0 × 10 4 per well) or 96-well plates (0:5 × 10 4 per well) and cultured for 2 days before starting the experiments. As described previously [6], after pretreatment with 4-phenylbutyrate (4-PBA; Sigma, USA), dantrolene (DAN; Selleck, USA), xestospongin C (XeC; APExBIO, USA), ruthenium red (RR; Sigma, USA), or N-acetylcysteine (NAC; Sigma, USA) for 1 h, NP cells were placed in a pressure apparatus under the compression of 1 MPa. The apparatus was then sealed carefully. Two conduits were opened to transmit oxygen and carbon dioxide, ensuring 20% O 2 and 5% CO 2 . A compressor was applied to create an atmosphere of 1.0 MPa. Cells were cultured under compression for 0, 12, 24, and 36 h. All cells were cultured for the same time span; only the duration of compression and the applied treatments varied.

Transmission Electron Microscopy (TEM)
. TEM was used to examine changes in the ultrastructure of the cells [6]. Briefly, cells were trypsinized, collected, and washed twice with phosphate-buffered saline (PBS). Cell pellets were then created by centrifugation at 1500 × g for 20 min. Next, the cell pellets were treated with 2.5% glutaraldehyde for 2 h and then with 1% osmium tetroxide for 2 h. An ascending ethanol series was applied to dehydrate the pellets, and the cells were embedded in epon 812 (SPI, USA). Finally, after staining, sections were observed using TEM (FEI Company, Holland).

Propidium Iodide
Staining. The cells were stained with 4 mg/mL Hoechst 33258 (Beyotime, China) for 5 min, then washed three times with PBS. The cells were covered with propidium iodide (PI; Beyotime, China), diluted with the corresponding buffer to 5 mg/mL. After incubation for 5 min, the NP cells were observed using a fluorescence microscope (Olympus IX71, Japan).

Flow
Cytometric Analysis of PI Uptake. After treatment, NP cells were collected by trypsinization. Next, the NP cells were resuspended in 5 mg/mL PI for 15 min. The PIpositive cells were then determined by flow cytometry (Becton Dickinson, USA).
2.6. Lactate Dehydrogenase Leakage Assay. NP cells seeded in 96-well plates were exposed to a mechanical load in a pressure apparatus under the compression of 1 MPa. At the end of treatment, the lactate dehydrogenase (LDH) leakage assay 2.14. Detection of 8-Hydroxy-2 ′ -deoxyguanosine (8-OHdG).
The detection of 8-OHdG (Cusabio, China) was performed as previously described [20]. DNA from NP cells was extracted using a DNA Extractor WB kit. The levels of 8-OHdG were detected using an ELISA assay kit according to the manufacturer's protocol.

Reverse-Transcription-Quantitative Polymerase Chain
Reaction (RT-qPCR). A TRIzol reagent (Invitrogen, USA) was used to extract total RNA from cells according to the manufacturer's instructions. RNA was transcribed to generate complementary DNA (cDNA), using HiScript II Q RT SuperMix (Vazyme, China). The primer sequences were as follows: Oxidative Medicine and Cellular Longevity (qPCR) assays were performed using SYBR Green Real-time PCR Master Mix (Toyobo, Japan) in a StepOnePlus Real-Time PCR System (Applied Biosystems, Canada). The data were analyzed using 2 -ΔΔCT calculations.

Statistical Analysis.
Each experiment was performed for at least three biological replicates and presented as the mean ± standard deviation ðSDÞ and analyzed using SPSS 19. Student's t-test was used to analyze the differences between the two groups. Multiple sets of data were analyzed by one-way repeated measures analysis of variance (ANOVA), followed by least significant difference (LSD). P < 0:05 was considered statistically significant.

ERS Is Involved in Compression-Induced Programmed
Necrosis of NP Cells. Increasing evidence supports the involvement of the ERS-initiated cell death pathway in IVDD [12,13]. However, the precise role of ERS in compressionrelated NP death has not been fully elucidated. To confirm the existence of ERS in the pathogenesis of IVDD, we compared the levels of ERS markers in normal and degenerative discs. As seen in Figure 1(a) and Figure S1, the degenerated discs exhibited increased expression of GRP78 and CHOP compared to the nondegenerated discs. Next, we evaluated whether compression induced ERS in vitro. We examined the changes in ERS-related protein (CHOP, GRP78, and p-PERK) expression in NP cells under compression conditions. After treatment with compression for 0, 12, 24, and 36 h, the expression levels of CHOP, GRP78, and p-PERK increased in a time-dependent manner (Figure 1(b)). Consistently, compression significantly increased the mRNA levels of CHOP, GRP78, and PERK (Figure 1(c)). Subsequently, ER swelling was observed by transmission electron microscopy. NP cells without compression showed normal morphology with intact ER, while compression stimulation caused extensive expansion and deformation of ER, as indicated by white arrowheads. The borders between some mitochondria and ER were unclear at 24 h and 36 h, which indicated an interaction between ER and mitochondria; these mitochondria are highlighted by red arrowheads (Figure 1(d)).
To further investigate the effects of ERS on programmed necrosis caused by compression, NP cells were incubated with 4-PBA before compression treatment, and programmed necrosis of cells was examined. The elevation of CHOP, GRP78, and p-PERK levels was notably ameliorated by 4-PBA, suggesting that 4-PBA was sufficient to inhibit ERS in NP cells ( Figure S2 (a)). Meanwhile, we found by flow cytometry (P < 0:01) and PI staining (Figures 1(e) and 1(f); Figure S2 (b)) that 4-PBA significantly attenuated compression-induced programmed necrosis in NP cells. In addition, the detection of HMGB1 expression and LDH release also showed that 4-PBA strongly inhibited programmed necrosis of NP cells induced by the mechanical load (P < 0:01; Figures 1(g) and 1(h)). Our data indicated that ERS was activated under prolonged compression, while its inhibition significantly alleviated compression-induced programmed necrosis in NP cells.

Compression-Induced Ca 2+ Mitochondrial
Translocation from the ER. We have revealed that ERS contributes to compression-induced programmed necrosis of NP cells. However, the underlying mechanisms remain unclear. Several studies have indicated that perturbation of ER Ca 2+ homeostasis is involved in ERS and cell apoptosis [21]. Specifically, Ca 2+ released from the ER translocates to the mitochondria to induce cell death in response to cellular stimuli [22]. Our flow cytometry results indicated that NP cells subjected to mechanical load showed significant upregulation of [Ca 2+ ] m at 24 and 36 h. The P values of each treatment group were 0.142, 0.002, and 0.0005, respectively (Figures 2(a)-2(c)). Interestingly, 4-PBA completely inhibited the upregulation of [Ca 2+ ] m in compression-treated NP cells (P < 0:01; Figures 2(d)-2(f)). However, the specific Ca 2+ channels that determine compression-induced Ca 2+ translocation have not yet been clarified. The ryanodine receptor (RyR) and IP 3 R are the main intracellular Ca 2+ release channels located on the ER of most cell types. To further explain the role of the ER in Ca 2+ redistribution, we employed Ca 2+ -channel inhibitors-xestospongin C (for IP 3 R) and dantrolene (for RyR)-to investigate the change in [Ca 2+ ] m . As shown in Figures 2(g)-2(i), NP cells pretreated with xestospongin C exhibited a decrease in [Ca 2+ ] m levels compared with the compression group (P < 0:01). By contrast, dantrolene was unable to block the compression-mediated increase in [Ca 2+ ] m levels (P > 0:05). Our study indicates that compression induces ER Ca 2+ efflux to the mitochondria via the IP 3 R channel.
3.3. The IP 3 R-GRP75-VDAC1 Complex Is Responsible for the Ca 2+ Translocation between the ER and the Mitochondria. The interaction between ER and mitochondria induced by compression was evident (Figure 1(d), red arrows). This 4 Oxidative Medicine and Cellular Longevity   Oxidative Medicine and Cellular Longevity    Oxidative Medicine and Cellular Longevity interaction may be responsible for compression-induced Ca 2+ translocation. Previous studies have confirmed that the ER couples with the mitochondria and that an IP 3 R-GRP75-VDAC1 complex-essential for intracellular Ca 2+ efflux into the mitochondria-is present at the ERmitochondria interface [23]. As demonstrated above, IP 3 R is located in the ER and is responsible for compressioninduced Ca 2+ release from the ER pool. VDAC1 is a mitochondrial outer membrane protein, and GRP75 is a linker protein that tethers IP 3 R to VDAC1 to generate a molecular bridge for Ca 2+ translocation [19]. To investigate the potential effect of the IP 3 R-GRP75-VDAC1 complex on Ca 2+ accumulation in the mitochondria under excessive compression, we used confocal microscopy to analyze the colocalization of IP 3 R and VDAC1. The yellow areas represent colocalization between IP 3 R and VDAC1; these areas were more obvious under 36 h compression, demonstrating that compression enhanced IP 3 R-VDAC1 colocalization in NP cells (Figure 3(a)). Furthermore, the formation of the IP 3 R-GRP75-VDAC1 complex was confirmed by coimmunoprecipitation using IP 3 R or VDAC1 as bait. As shown in Figures 3(b) and 3(c), compared with the control group, the combination of IP 3 R, VDAC1, and GRP75 was enhanced when cells were subjected to mechanical loading. In summary, our studies demonstrated that the formation of the IP 3 R-GRP75-VDAC1 complex was significantly increased under continuous compression.
To confirm the impact of the IP 3 R-GRP75-VDAC1 complex on Ca 2+ transmission from the ER pool to the mitochondria, we silenced GRP75 and VDAC1 expression by siR-NAs in NP cells, and the levels of [Ca 2+ ] m were observed by Rhod-2 AM staining. As shown in Figures 3(d)-3(f), silencing of GRP75 and VDAC1 reduced the levels of [Ca 2+ ] m in NP cells exposed to mechanical load (P < 0:01). Moreover, the inhibition of IP 3 R by XeC and the silencing of either GRP75 or VDAC1 all reduced the percentage of PI-positive NP cells (P < 0:001; Figures 3(g) and 3(h)). These results indicate that the formation of the IP 3 R-GRP75-VDAC1 complex in NP cells under compression was responsible for the efflux of Ca 2+ from the ER to the mitochondria and compression-induced NP cell death.

Inhibition of Compression-Induced Programmed Necrosis in [Ca 2+ ] m -Protected NP Cells by Improving Mitochondrial
Function and Inhibiting the PARP/AIF Pathway. We confirmed that compression induced ER-mitochondrial Ca 2+ exchange in NP cells. To further explore the role of Ca 2+ redistribution in IVDD, we preincubated NP cells with ruthenium red (RR), a mitochondrial Ca 2+ uptake inhibitor. As seen in Figures 4(a) and 4(b), RR inhibited PI uptake in      Mitochondrial dysfunction plays a critical role in IVDD progression. We first determined whether Ca 2+ overload in mitochondria mediated the compression-induced NP cell necrosis by disturbing mitochondrial homeostasis. Mitochondrial inner membrane potential (ΔΨ m ) is an essential indicator of mitochondrial function. JC-1 indicates mitochondrial polarization by shifting its fluorescence from green to red in a potential-sensitive manner. As shown in Figures 4(e) and 4(f), ΔΨ m was significantly lower in compression-treated cells (P < 0:001); however, the reduction was greatly reduced when RR was present (P < 0:001). In addition, cellular ATP levels were significantly decreased after compression for 36 h (P < 0:001), whereas cotreatment with RR alleviated the compression-mediated ATP decrease (P < 0:01; Figure 4(g)). These results suggest that ERmitochondrial Ca 2+ exchange may mediate compressioninduced programmed necrosis of NP cells by disrupting cellular bioenergetics.
PARP, located at the mitochondrial outer membrane, is an important protein that regulates energy metabolism and mitochondrial membrane potential [24]. Activated PARP leads to continued nicotinamide adenine dinucleotide (NAD + ) depletion and loss of mitochondrial membrane potential, which then triggered AIF translocation from the mitochondria to the nucleus to cause chromatin condensation and DNA fragmentation, which leads to programmed necrosis of cells [25]. In this study, the expression of PARP was significantly decreased, whereas that of cleaved PARP (C-PARP) was elevated in NP cells exposed to compression (Figure 4(h)), indicating the activation of PARP. In addition, compression also induced a decrease in AIF in the mitochondria and increased AIF accumulation in the cell nucleus (Figure 4(i)). RR treatment decreased C-PARP expression and increased PARP expression (Figure 4(h)). Additionally, intranuclear accumulation of AIF was attenuated in the presence of RR (Figures 4(i) and 4(j)). Therefore, these results suggest that Ca 2+ mediated programmed necrosis of NP cells under compression by facilitating the PARP-AIF pathway.

Increase in ROS Levels Contributes to Compression-Induced ERS, [Ca 2+ ] m Overload, and Programmed Necrosis.
It has been reported that ROS are crucial mediators in the pathogenesis of IVDD [26,27]. To assess ROS accumulation and related damage in NP cells subjected to compression, total ROS was detected using the fluorescent dye H 2 DCFH-DA, and a quantification of oxidative DNA damage-8hydroxydeoxyguanosine (8-OHdG) quantitative analysis-was carried out to evaluate ROS damage in NP cells. The results indicated that compression caused ROS accumulation in NP cells and increased levels of 8-OHdG in rat NP cells, suggesting that oxidative damage occurs in NP cells subjected to compression ( Figure S3). We then applied NAC to scavenge ROS and found that compression-induced programmed necrosis was significantly attenuated by NAC when compared with the compression group in the absence of NAC, as detected by PI flow cytometry (P < 0:001), PI staining, and cellular release of HMGB1 and LDH (P < 0:001; Figures 5(a)-5(d)). Interestingly, NAC also efficiently inhibited the upregulation of ERS markers in NP 13 Oxidative Medicine and Cellular Longevity cells exposed to compression (Figures 5(e) and 5(f)). Therefore, compression-induced ROS may serve as an activator of increased ERS, resulting in programmed cellular necrosis. NAC also inhibited compression-induced [Ca 2+ ] m overload (P < 0:01; Figures 5(g)-5(i)). In summary, our results indicate that compression induced ROS accumulation that excited ERS-related [Ca 2+ ] m overload and led to programmed necrosis.

Discussion
In the present study, we first demonstrated that radical ERS contributed to programmed necrosis of NP cells under compression. Under mechanical load, Ca 2+ was released from the ER and translocated to the mitochondria via the IP 3 R-GRP75-VDAC1 complex, which resulted in programmed necrosis through the PARP/AIF pathway. Our findings

16
Oxidative Medicine and Cellular Longevity revealing the molecular basis of programmed necrosis of NP cells induced by compression have tremendous potential for the treatment of IVDD. The ERS response system has dual functions in the execution of cell death. Appropriate ERS protects cells against the death response to a harmful stimulus. However, aggravated ERS promotes cell apoptosis by initiating a proapoptotic mechanism [28]. Previous studies have confirmed that ERS plays a vital role in IVDD pathogenesis. IVDD was alleviated to a certain degree by targeting ERS induced by oxidative stress or advanced glycation end products (AGEs) [12,29]. Nevertheless, whether ERS accelerates NP cell programmed necrosis under compression has not yet been elucidated. Mechanical loading is consistent and inevitable stress on human discs. Compression may be of greater clinical relevance than chemicals [30]. In this study, we observed increased expression of the ERS-related proteins GRP78, CHOP, and p-PERK, along with significant ER deformation in response to mechanical loading. Pretreatment with the chemical ERS inhibitor 4-PBA significantly decreased compression-induced NP cell death. Our study is the first to indicate that ERS contributes to compression-induced NP cell programmed necrosis.
The ER is an important organelle that regulates intracellular Ca 2+ levels [21]. In response to pharmacological and physiological stimuli that disrupt ER homeostasis, Ca 2+ is released by the ER through two ER-specific Ca 2+ channels: RyR and IP 3 R. Because of close physical contact with the ER, the mitochondria acquire a remarkable Ca 2+ fraction [21]. Our results demonstrated that compression induced Ca 2+ elevation in the mitochondria of NP cells; this was attenuated by XeC (IP 3 R inhibitor) but not by dantrolene (RyR inhibitor). Interestingly, RyR and IP 3 R were both found to be responsible for increased Ca 2+ release in NP cells induced by AGEs [31]. AGEs have been reported to crosslink in the annulus fibrosus and lead to increased compressive stiffness, torque range, and failure torque [32], whereas excessive mechanical loading leads to extracellular matrix remodeling and a decreased number of intervertebral disc cells [33]. These pathogenic differences between AGEs and compression may lead to different roles of RyR in regulating Ca 2+ redistribution in NP cells, while further studies may be required to further clarify the roles of RyR.
Increasing data have verified that Ca 2+ translocation from the ER to the mitochondria by the IP 3 R-GRP75-VDAC1 complex results in Ca 2+ overload in the mitochondria, which leads to the opening of mitochondrial permeability transition pores, release of cytochrome c, and activation of caspases to induce cell apoptosis [19,34]. Our results indicate that IP 3 R strikingly interacts with GRP75 and VDAC1 in NP cells under mechanical load conditions, providing the molecular basis for Ca 2+ translocation. To explore the potential impacts of a reinforced IP 3 R-GRP75-VDAC1 complex in [Ca 2+ ] m excess and NP cell death, we inhibited different proteins in the axis for further exploration. The current results clearly demonstrated that inhibition of IP 3 R, GRP75, or VDAC1 prevented compression-induced [Ca 2+ ] m overload and decreased NP cell programmed necrosis. Our results suggest that the IP 3 R-GRP75-VDAC1 complex facilitates NP cell programmed necrosis by stimulating Ca 2+ efflux from the ER to the mitochondria. The mechanism by which compression induces IP 3 R-GRP75-VDAC1 complex assembly on the ER requires further investigation.
Ca 2+ overload in the mitochondria usually serves as intracellular signaling for PARP activation [35,36]. PARP activation leads to PARP-mediated mitochondrial dysfunction, as demonstrated by mitochondrial membrane depolarization, overproduction of superoxide, and loss of cardiolipin content [37]. Besides, in response to damage stimulation, the DNA repair enzyme, PARP, results in the massive synthesis of poly(ADP-ribose) (PAR) from nicotinamide adenine dinucleotide (NAD+), leading to rapid exhaustion of intracellular NAD+ and ATP. Afterwards, AIF escapes from the mitochondria to the nucleus and then combines with DNA and RNA to bring about chromatolysis and cell programmed necrosis [38]. Supportively, we previously reported that the PARP was activated in H 2 O 2 -treated disc cells to exert cell necrosis [39]. Thus, PARP/AIF may be critical in Ca 2+ overload-induced mitochondrial dysfunction and cell death in NP cells under prolonged compression. In our study, we found that Ca 2+ accumulation in the mitochondria induced by compression significantly impaired mitochondrial membrane potential and ATP synthesis. Also, compression increased PARP hyperactivation and the translocation of AIF from the mitochondria to the nucleus, which could be inhibited by RR. Therefore, Ca 2+ efflux to the mitochondria from the ER mediated compression-induced NP cell programmed necrosis via activation of the PARP/AIF pathway.
Excessive accumulation of ROS, including NO, peroxynitrite, and nitrotyrosine, has been identified in degenerative discs [40]. Multiple microenvironmental factors in discs, such as mechanical loading and proinflammatory cytokines, contribute to increased ROS levels in disc cells [40][41][42]. Importantly, excessive ROS accumulation may disrupt protein synthesis and conformation, leading to protein deformation and unfolded protein response [43]. Inhibition of ROS significantly reduced ERS activation and ERS-related Leydig cell apoptosis in rats [44]. These findings were consistent with those of our study in that ERS was downregulated after NAC administration and compression-induced programmed necrosis of NP cells was decreased. Although mitochondrial dysfunction can be caused by either Ca 2+ overload or ROS production [45], it is unclear whether Ca 2+ overload in the mitochondria is regulated by ROS production. Our findings suggest that NAC largely decreases the accumulation of compression-induced Ca 2+ in the mitochondria. Considering NAC downregulated ERS, which was essential in compression-induced [Ca 2+ ] m overload, compressioninduced excessive ROS may increase Ca 2+ accumulation in the mitochondria via radical ERS.

Conclusions
Our results indicate for the first time that compressioninduced ERS plays a crucial role in mediating NP cell programmed necrosis. Under mechanical loading, Ca 2+ is released from the ER and translocated to the mitochondria