Overexpression of BNIP3 in Rat Intervertebral Disc Cells Can Trigger Autophagy-Related Apoptosis

Background: Intervertebral disc degeneration (IDD) is the important cause of low back pain. Excessive apoptosis of intervertebral disc cells is the primary cause of IDD. BNIP3 is a member of Bcl-2 family and is widely involved in cell autophagy and apoptosis. However, the roles and mechanisms of BNIP3 in intervertebral disc cell autophagy and apoptosis are unclear. Results: In this study, the primary rat intervertebral disc cells were prepared and applied to study the effect of BNIP3 overexpression on their autophagy and apoptosis. RT-PCR and western blotting assay showed that BNIP3 overexpression signicantly up-regulated the expression of autophagy and pro-apoptotic related proteins, including HIF-1α, Apaf-1, cleaved caspase 3, LC-3 and Beclin-1, while down-regulated the expression of anti-apoptotic protein Bcl-2. Subsequent cell staining detection of autophagy and apoptosis showed that BNIP3 overexpression signicantly increased the autophagy and apoptosis of rat intervertebral disc cells. Furthermore, BNIP3 interference assay revealed that the effects of BNIP3 overexpression can be reversed by BNIP3 interference. Conclusions: The above ndings suggested that BNIP3 enhanced autophagy of intervertebral disc cells and further triggered the apoptosis of intervertebral disc cells, eventually led to the development of intervertebral disc degeneration. cleaved caspase 3, Bcl-2, LC-3 and Beclin-1 in rat intervertebral disc cells after transfected with BNIP3 overexpression or interference vectors. The RT-PCR results showed that the expression of HIF-1α, Apaf-1, cleaved caspase 3, LC-3 and Beclin-1 mRNA was upregulated signicantly when the rat intervertebral disc cells transfected with BNIP3 overexpression vector compared to control or blank group. However, the expression of anti-apoptosis protein Bcl-2 was decreased in BNIP3 overexpression group compared to the control or blank group. Moreover, the expression of autophagy and apoptosis-related proteins in rat intervertebral disc cells after BNIP3 RNAi was also detected by RT-PCR, the results showed that the expression of HIF-1α, Apaf-1, cleaved caspase 3, LC-3 and Beclin-1 was decreased signicantly when the rat intervertebral disc cells transfected with BNIP3 interference vector compared to control or blank group. Similarly, the


Background
As a common spinal disease, low back pain has become a very serious public health problem in the world, which seriously affecting the physical and mental health of patients and bringing a heavy economic burden to society (1). Although the detailed pathological mechanism of low back pain is elusive up to now, the currently accepted fact is that intervertebral disc degeneration (IDD) is an important cause of low back pain (2,3). The occurrence and development of IDD are related to many factors, such as genetic factors, lifestyle habits (such as smoking or alcohol consumption) and aging. It is worth mentioning that apoptosis of intervertebral disc cells, degradation of extracellular matrix and cell phenotype alterations are still the direct causes of IDD (4). Autophagy is a lysosome-dependent cell degradation pathway that exists exclusively in eukaryotic cells (5). As a type II programmed cell death pathway, autophagy promotes or inhibits apoptosis in different ways. It has reported that autophagy play a crucial role in the development of IDD (6,7).
Bcl-2 and adenovirus E1B 19-kDa interacting protein 3 (BNIP3) is a member of the BH3-only subfamily of the Bcl-2 family, which is closely related to apoptosis triggered by ischemia and hypoxia (8)(9)(10). Present studies suggest that hypoxia-inducible factor-1α (HIF-1α) is the primary regulatory protein of BNIP3 (11), so the expression of BNIP3 is largely upregulated under hypoxic conditions. BNIP3 not only regulate cell apoptosis directly, but also affect the cell survival and apoptosis indirectly by regulating the level of autophagy or mitophagy (12), therefore BNIP3 play vital roles in ischemia or hypoxia related diseases (13)(14)(15). Recent studies also show that the BNIP3 expression is positively correlated with the severity of IDD, that is, patients with severe IDD usually have higher expression of BNIP3 (16,17), suggest that BNIP3 play an important role in the development of IDD. However, the detailed molecular mechanism of BNIP3 in the development of IDD is unclear.
In this study, the methods of overexpression and RNA interference were used to study the effect of BNIP3 on autophagy and apoptosis in the isolated rat intervertebral disc cells. Our ndings showed that overexpression of BNIP3 signi cantly enhanced autophagy of rat intervertebral disc cells and further led to increased apoptosis, resulted in a decrease in the number of intervertebral disc cells and absence in cell function, which ultimately causes low back pain. Similarly, the opposite results were observed in BNIP3 interference assay. These ndings suggested that BNIP3 enhanced autophagy-related apoptosis of intervertebral disc cells and promoted the development of intervertebral disc degeneration.

Identi cation of cultured rat intervertebral disc cells
To study the role of BNIP3 in the autophagy and apoptosis of rat intervertebral disc cells, the primary rat intervertebral disc cells was prepared. The cultured rat intervertebral disc cells were stained with toluidine blue, we found the cytoplasm of the cell was stained to blue and the nucleus was stained to bluish violet ( Fig. 1). Furthermore, the cells under the light microscope have the morphological characteristics of intervertebral disc cells, suggested that the isolated and cultured cells were rat intervertebral disc cells.
BNIP3 overexpression decreased the viability of rat intervertebral disc cells The transfection method was used to study the effect of BNIP3 on the viability of rat intervertebral disc cells. First, the transfection e ciency in the cells was detected. The expression level of BNIP3 mRNA in BNIP3 overexpression group (oeBNIP3) was increased signi cantly compared to the overexpression control group (oeControl) or blank group. Similarly, the expression level of BNIP3 mRNA in BNIP3 interference group (siBNIP3) was decreased signi cantly compared to the interference control group (siControl) or blank group ( Fig. 2A). Subsequently, the MTT assay was employed to detect the viability of rat intervertebral disc cells. Our ndings showed that although the cell viability in all groups was increased with time, the cell viability of BNIP3 overexpression group was lower than overexpression control or blank group, while BNIP3 interference group had higher cell viability than the interference control or blank group (Fig. 2B). These ndings suggested that BNIP3 negatively regulated the viability of rat intervertebral disc cells.
BNIP3 overexpression enhanced the autophagy of rat intervertebral disc cells Previously study has reported that BNIP3 could induce autphagy in various types of cells, however, it was unknown that whether BNIP3 could trigger the autophagy of rat intervertebral disc cells. Here, the autophagy of rat intervertebral disc cells was detected by acridine orange staining. We found the rat intervertebral disc cells with BNIP3 overexpression were concentrated and the nuclei were fragmented into dots, and they were stained into uneven, dense and deep stained green particles under light microscope compared to control and interference group (Fig. 3), which was known as the occurrence of autophagy. Similarly, the results of autophagy detected by MDC method also showed that signi cant autophagy occurred in BNIP3 overexpressed rat intervertebral disc cells, while no obvious autophagy was observed in BNIP3 interference group or control (Fig. 4). The above ndings suggested that BNIP3 overexpression triggerred the autophagy of rat intervertebral disc cells.
BNIP3 overexpression or interference disturbed the cell membrane potential The cell membrane potential was closely related to cell states such as autophagy and apoptosis. Here, the cell membrane potential of BNIP3 transfected cells was detected by ow cytometry. Our ndings showed that the membrane potential of BNIP3 overexpression group was decreased signi cantly compared to the overexpression control or blank group, while the membrane potential of BNIP3 interference group was increased signi cantly compared to the interference control or blank group (Fig. 5), suggested that BNIP3 affected the cell membrane potential by triggering autophagy of rat intervertebral disc cells.

BNIP3 induced increased expression of autophagy and apoptosis-related molecules
In above study, we have observed BNIP3 overexpression could activate the autophagy and apoptosis of rat intervertebral disc cells, and vice versa. Therefore, can BNIP3 affect the expression of autophagy and apoptosis-related proteins? First, the RT-PCR assay was employed to detect the mRNA expression differences of autophagy and apoptosis-related proteins including HIF-1α, Apaf-1, cleaved caspase 3, Bcl-2, LC-3 and Beclin-1 in rat intervertebral disc cells after transfected with BNIP3 overexpression or interference vectors. The RT-PCR results showed that the expression of HIF-1α, Apaf-1, cleaved caspase 3, LC-3 and Beclin-1 mRNA was upregulated signi cantly when the rat intervertebral disc cells transfected with BNIP3 overexpression vector compared to control or blank group. However, the expression of antiapoptosis protein Bcl-2 was decreased in BNIP3 overexpression group compared to the control or blank group. Moreover, the expression of autophagy and apoptosis-related proteins in rat intervertebral disc cells after BNIP3 RNAi was also detected by RT-PCR, the results showed that the expression of HIF-1α, Apaf-1, cleaved caspase 3, LC-3 and Beclin-1 was decreased signi cantly when the rat intervertebral disc cells transfected with BNIP3 interference vector compared to control or blank group. Similarly, the expression of anti-apoptosis protein Bcl-2 was increased in BNIP3 interference group compared to the control or blank group (Fig. 6).
In addition, western blotting analysis of autophagy and apoptosis-related proteins revealed that the expression of HIF-1α, Apaf-1, cleaved caspase 3, LC-3 and Beclin-1 was increased when the rat intervertebral disc cells transfected with BNIP3 overexpression vector compared to the overexpression control and blank group. Interestingly, the expression of anti-apoptosis protein Bcl-2 was decreased in BNIP3 overexpression group compared to the control or blank group. In addition, the BNIP3 RNAi assay revealed that the expression of HIF-1α, Apaf-1, cleaved caspase 3, LC-3 and Beclin-1 was decreased signi cantly when the rat intervertebral disc cells transfected with BNIP3 interference vector compared to control or blank group. Similarly, the expression of anti-apoptosis protein Bcl-2 was increased in BNIP3 interference group compared to the control or blank group (Fig. 7A). To better display the expression differences of HIF-1α, Apaf-1, cleaved caspase 3, Bcl-2, LC-3 and Beclin-1among different groups, the gray scale analysis of western blotting bands was performed and the similar results were obtained (Fig. 7B).
These ndings indicated that BNIP3 activate the expression of autophagy and apoptosis-related proteins and nally lead to the occurrence of autophagy and apoptosis of rat intervertebral disc cells.
Taken together, our ndings showed that BNIP3 triggered autophagy and apoptosis of rat intervertebral disc cells via stimulating the expression of autophagy and apoptosis-related proteins. The augment of autophagy and apoptosis of rat intervertebral disc cells led to development of intervertebral disc degeneration and nally exacerbated the process of low back pain.

Discussion
Apoptosis of lumbar intervertebral disc cells is one of the important causes of IDD (4,20). Numerous studies have shown that the apoptosis rate of nucleus pulposus cells in IDD patients or IDD model animals is signi cantly higher than that in nucleus pulposus cells of normal humans or animals (4). In addition, many factors that can induce IDD such as ischemia and hypoxia, nutritional de ciencies, stress load, high glucose and hypertonicity, not only activated autophagy but also induced apoptosis (21)(22)(23).
The present perspective believed that autophagy can inhibit apoptosis by clearing damaged cells, but excessive autophagy can also induce autophagic apoptosis (24,25). Autophagy is a lysosomedependent catabolism pathway that exclusively exists in eukaryotic cells and composed of a series of autophagy-related genes, which can degrade and recycle damaged, senescent and denatured organelles or macromolecular substances in cells (26). Finally, autophagy provides su cient energy for cell survival.
Autophagy mainly includes macrophage, microautophagy and molecular chaperone-mediated autophagy. The process of autophagy mainly divides into ve stages: autophagy induction, autophagy vesicle nucleation, extension, autophagosome formation and autophagy lysosome formation (27). It has reported that autophagy played many different important roles in the development of IDD. Ye et al. (28) found that the number of autophagosomes and the ratio of autophagy-related proteins LC-3II/LC-3I in rat intervertebral disc nucleus pulposus cells increased with the age of rats. Xu et al. (29) found that the autophagy level of endplate cells in IDD patients was lower than that in normal humans, while Gruber et al. (30) showed that the autophagy level of the annulus brosus in patients with IDD was higher than that in normal humans. The above studies indicated it's discriminative on the exact roles of autophagy in IDD.
One of the possible reasons was that the patients adopted in these studies have different severity of intervertebral disc degeneration. The other possible reason was the tissues used were from different sites. However, the detailed role and molecular mechanisms of autophagy in the development of IDD were unclear. Therefore, present study aimed to explore the role and mechanisms of autophagy in the apoptosis of rat intervertebral disc cells.
The BNIP3 gene is a protein discovered by Boyd and other scholars in 1994 through yeast hybridization technology that can interact with the adenovirus E1B19kD protein and can encode 194 amino acids. The encoded protein has two dimers. Subsequent studies con rmed that BNIP3 is a pro-apoptotic protein of the Bcl-2 family (31). Its C-terminus can bind to the mitochondrial membrane, which increases the permeability of the mitochondrial membrane and reduces the potential of the mitochondrial membrane, which in turn promotes apoptosis (32). The biological role of BNIP3 under hypoxic conditions mainly depends on the regulation of HIF-1α, which can bind to the HRE response element of the BNIP3 promoter, thereby regulating the transcription of BNIP3. Therefore, scholars believe that the hypoxic environment can signi cantly induce the expression of BNIP3, and the moderate expression of BNIP3 can activate autophagy and maintain the survival of cells, while excessive autophagy causes apoptosis (13)(14)(15). Current research showed that the positive expression rate of BNIP3 gradually increases with the severity of IDD, and the expression of BNIP3 was closely related to the apoptosis of intervertebral disc cells during the development of IDD (16). Previous studies showed that BNIP3 play important roles in the development of IDD, but the detailed mechanism was elusive.
In this study, the BNIP3 overexpression and RNA interference methods were employed to explore the role and mechanisms of BNIP3 in IDD process. First, the success of transfection was veri ed by RT-PCR.
Second, the CCK-8 method was used to detect the cell viability of rat intervertebral disc cells after transfected with different vectors and revealed that BNIP3 overexpression remarkable decreased the cell viability and BNIP3 interference remarkable increased the cell viability (Fig. 2B). These ndings were consistent with the evidence that BNIP3 closely related with the apoptosis of intervertebral disc cells in IDD development reported previously (33), suggested that BNIP3 expression signi cant affected the viability of rat intervertebral disc cells. Subsequently, the autophagy of rat intervertebral disc cells detected by acridine orange staining and MDC staining methods, the ndings showed that BNIP3 overexpression increased the autophagy and BNIP3 interference decreased the autophagy of rat intervertebral disc cells (Figs. 3 and 4). Flow cytometry detection of cell membrane potential revealed that BNIP3 overexpression signi cant decreased the membrane potential while BNIP3 interference increased the membrane potential of rat intervertebral disc cells (Fig. 5), suggested that BNIP3 activated autophagy related apoptosis and further induced a decreased number of intervertebral disc cells, eventually led to the occurrence of IDD.
Cell apoptosis and autophagy are closely associated with apoptosis-or autophagy-related proteins and pathways. Bcl-2 is one of the most important anti-apoptosis proteins that can form a dimer with the pro-apoptotic protein Bax, then inhibit apoptosis signaling and promote cell survival (34). Beclin-1 is an autophagy-related gene that is homologous with yeast autophagy gene 6. Beclin-1 can recruit other autophagy-related proteins to localize on the membrane of autophagosome, which plays a positive regulatory role on the formation of autophagosomes (35). Furthermore, Beclin-1 also interacts with Bcl-2 through its BH3 domain to maintain the dynamic equilibrium of cell autophagy and apoptosis (36). LC3 is the homolog of mammalian autophagy gene 8 that contains two forms (LC-3I and LC-3II), especially the content of LC-3II to a certain extent re ects the cell's autophagy activity. BNIP3 can competitively bind to Bcl-2 with Beclin-1 because both of them contains BH3 domain. The overexpression of BNIP3 causes the release of Beclin-1 from the Beclin-1/Bcl-2 complex. The free Beclin-1 then forms a complex with PI3K and further regulates the expression of various autophagy-related genes downstream of PI3K/AKT, eventually promote the occurrence of mitochondrial autophagy (37,38). In addition, the BNIP3 overexpression increases the mitochondrial membrane permeability and membrane potential, then leads to the release of cytochrome C. The binding of cytochrome C with Apaf-1 can activate caspase pathway, then induces a series of cascades and triggers apoptosis. In this study, RT-PCR and western blotting methods were employed to study the expression differences of autophagy and apoptosis-related proteins of rat intervertebral disc cells after transfected with BNIP3 overexpression or interference vectors. Our ndings showed that the expression of pro-apoptotic or autophagic proteins such as HIF-1α, Apaf-1, cleaved caspase 3, LC-3 and Beclin-1 were increased signi cantly after the cells overexpressed with BNIP3, while the expression of anti-apoptotic protein such as Bcl-2 was decreased signi cantly (Figs. 6  and 7). Similarly, the expression of pro-apoptotic or autophagic proteins such as HIF-1α, Apaf-1, cleaved caspase 3, LC-3 and Beclin-1 were decreased signi cantly after the cells transfected with BNIP3 terference vector, while the expression of anti-apoptotic protein such as Bcl-2 was increased signi cantly (Figs. 6 and 7). These ndings suggested that BNIP3 overexpression increased the autophagy-related apoptosis of intervertebral disc cells by regulating the expression of autophagy-and apoptosis-related proteins, thereby triggered autophagic apoptosis.
In conclusion, BNIP3 overexpression triggered autophagic apoptosis by activating the expression of autophagy-and apoptosis-related proteins in intervertebral disc cells, then decreased the viability of intervertebral disc cells and led to the occurrence of intervertebral disc degeneration, caused low back pain.

Conclusions
In this study, the primary rat intervertebral disc cells were prepared and applied to study the effect of BNIP3 overexpression on their autophagy and apoptosis. Our ndings showed that BNIP3 overexpression in rat intervertebral disc cells signi cantly up-regulated the expression of autophagy and pro-apoptotic related proteins, and then enhanced the autophagy and apoptosis of rat intervertebral disc cells. Moreover, the effects of BNIP3 overexpression could be reversed signi cantly by BNIP3 interference. These ndings indicated that BNIP3 enhanced autophagy of intervertebral disc cells and further triggered the apoptosis of intervertebral disc cells, eventually led to the development of intervertebral disc degeneration.

Materials And Methods
Animals 20 healthy adult Sprague-Dawley (SD) rats weighting 200 to 250 g were used in this study. The animals were purchased from Beijing Charles River Laboratory Animal Co., Ltd. The rats were housed in individual cages under a constant temperature (23 ± 2 °C) and had access to food and water ad libitum throughout the study. All procedures, care, and handling of the rats were approved by the Ethics Committee of Nanchang Hongdu hospital of traditional Chinese medicine.

Antibodies and Reagents
The SuperSignal™ West Pico PLUS chemiluminescent substrate (34580) was purchased from Thermo Isolation and culture of rat intervertebral disc cells Twenty SD rats were used to isolate the intervertebral disc cells. In brief, the rats were anesthetized and killed by intraperitoneal injection of 10% chloral hydrate, then isolate the spine under aseptic conditions, following removal of ligaments and nucleus pulposus around the spine. The isolated spine was cut into small block with the size of 1 mm3, after digestion with 0.25% trypsin for 20 min and 0.2% type II collagenase digestion for 2 h, the digested cells were collected by cell sieve and cultured in the DMEM/F12 medium containing 10% fetal bovine serum, 100 U/ml penicillin and 100 µg/ml streptomycin under an atmosphere of 5% CO2 at 37 °C, the cultured medium was changed every 2-3 days. When the degree of cell fusion reached more than 80%, digested the cells at a ratio of 1:2 for passage. At the same time, the second generation cells were taken for the detection of toluidine blue staining.

Toluidine blue staining
The above cultured rat intervertebral disc cells grown to the second generation, then inoculated them on 6-well plate, removed the medium and washed with phosphate buffer solution (PBS) for 5 min, then the cells xed with 4% paraformaldehyde at room temperature for 10 min. Subsequently, the xation solution was removed and the cells were stained with toluidine blue dye for 2 min, the staining solution was removed and the cells were washed with ultrapure water for 3 min. The stained and washed cells then air dried and imaged under optical microscope.

Overexpression and RNA interference assay
To study the role of BNIP3 in the autophagy and apoptosis of rat intervertebral disc cells, the overexpression and RNA interference assay were performed as described previously (18). Brie y, the rat intervertebral disc cells were placed in 6-well tissue culture plates and growth to 60% con uence prior to transfection. Then the cells were transfected with 4 µg of plasmids (built recombinant BNIP3-pcDNA 3.1 vector and pcDNA 3.1 empty vector) for 48 h. The Lipofectamine 3000 transfection reagent was used. For RNA interference (RNAi) experiments, the cells were growth to 40-50% con uence prior to transfection. The recombinant BNIP3 silenced plasmid sh-BNIP3-pLVshRNA (siBNIP3) and control vector (siCtrl) were transfected into cells by using the Lipofectamine 3000 transfection reagent for 48 h. After BNIP3 overexpression and RNAi, the expressions of BNIP3 mRNA in rat intervertebral disc cells were detected by real-time quantitative PCR.

Cell viability determination by CCK8 method
To determine the viability of rat intervertebral disc cells after tranfected with BNIP3 overexpression vector or interference vector, the CCK8 method was employed. Brie y, the transfected cells were placed in 96-well plates and added 150 µL freshly prepared toxicity test solution containing 10 µL CCK8 to each well and cultured for 4 h, the absorbance at 450 nm wavelength in 0 12 24 and 72 h were measured by microplate reader. The cell viability assay was repeated in three times and the data was presented as Means ± SD. The cell growth inhibition rate (%) = (OD450Control -OD450Experiment)/ OD450Control × 100%. The cell growth inhibition rate curve was drawed, the groups as abscissa and cell growth inhibition rate (%) as ordinate.
Cell autophagy determination by acridine orange staining The above transfected cells were collected and prepared to cell suspension at the concentration of 1 × 106 cells/ml. Every 100 µL cell suspension mixed with acridine orange staining solution with a nal concentration of 15 µg/mL, stained at room temperature in the dark for 20 min, then drop it on a glass slide and observed under uorescent microscope. The normal cells were presented to uniform yellow-green uorescence, while autophagic cells' chromatins were concentrated and the nuclei were fragmented into dots, and they were stained into uneven, dense and deep stained green particles.

Cell autophagy determination by MDC method
The transfected cells collected by trypsin digestion and washed with PBS, then suspended in 1 × wash buffer with the concentration of 1 × 106 cells/ml. 90 µL cell suspension was taken out and transferred to new eppendorf tube. Whereafter, 10 µL MDC dye was added into the cell suspension, mixed and stained at room temperature in the dark for 45 min, the stained cells were collected by centrifuged. The cells were washed two times with 1 × wash buffer and resuspended in 100 µL collection buffer. The cell suspension was loaded on the slide and observed under uorescent microscope. The normal cells were presented to uniform yellow-green uorescence, while autophagic cells' chromatins were concentrated and the nuclei were fragmented into dots, and they were stained into uneven, dense and deep stained green particles.
Cell membrane potential was detected by ow cytometry The membrane potential detection of transfected cells by using ow cytometry method. Brie y, the transfected cells collected by trypsin digestion and washed two times with PBS, then suspended in JC-1 working buffer with the concentration of 1 × 106 cells/ml. The cell suspensions incubated at 37 °C for 20 min, centrifuged, collected the cells and resuspended in 1 × Incubation buffer. Finally, the cells were analyzed by using a ow cytometer.

Real-time uorescent quantitative PCR (RT-PCR)
Real-time uorescent quantitative PCR was used to detect the expression of HIF-1α, Bcl-2, Apaf-1, cleaved caspase 3, LC-3 and Beclin-1 mRNA of transfected rat intervertebral disc cells. In brief, the total RNA of transfected cells was extracted by commercial RNA extraction kit and the RNA purity were determined by the ratio of A260/A280 and agarose gel electrophoresis. Total RNA was synthesized the rst strand cDNA by rst strand cDNA synthesis kit. Using the synthesized rst-strand cDNA as template, the speci c primers for HIF-1α, Bcl-2, Apaf-1, cleaved caspase 3, LC-3 and Beclin-1 were designed as the previous method (19) and listed on Table 1 and applied to RT-PCR reaction. The RT-PCR reaction condition as follows: 95 °C for 10min,95 °C for 5 and 60 °C for 1min 40 cycles. The RT-PCR reaction system was 20 µL, which contained 2 µL cDNA template, 0.8µL the primer F and primer R, 10 µL SYBR Green solution and 7.2 µL ddH2O. The data of RT-PCR was calculated and analyzed by the method of 2 -ΔΔCT. Western blotting Western blotting was used to detect the expression level of HIF-1α, Bcl-2, Apaf-1, cleaved caspase 3, LC-3 and Beclin-1 proteins of transfected rat intervertebral disc cells. Brie y, the cells were collected and washed, then lysed by adding RIPA lysate contained protease inhibitor cocktail in ice-bath for 20 min. The above cell lysate was centrifuged at 10000 rpm for 15 min, the supernatant after centrifugation is the total proteins. The concentrations of proteins were determined by BCA method. The total proteins were loaded and separated by 12% SDS − PAGE, then electrotransferred onto PVDF membranes. The PVDF membranes were subsequently blocked with 3% BSA and subsequently incubated with rabbit anti-HIF-1α (1:1000 dilution), anti-Bcl-2 (1:1000 dilution), anti-Apaf-1 (1:1000 dilution), anti-cleaved caspase 3 (1:1000 dilution), anti-LC-3 (1:500 dilution), anti-Beclin-1 (1:1000 dilution) polyclonal antibody and mouse anti-GAPDH monoclonal antibody (1:10000 dilution), respectively. The incubation of PVDF and primary antibodies was overnight at 4 °C. After incubated with primary antibodies, the PVDF membrane washed with TBS buffer contained 1‰ Tween-20 three times, then incubated with HRP-conjugated goat antirabbit or goat anti-mouse secondary antibody (1:5000 dilution). Finally, the protein bands were visualized with the SuperSignal™ West Pico PLUS chemiluminescent substrate.

Statistical analysis
All experimental values were presented as the Mean ± SD. Each experiment was repeated at least 3 times independently. Statistical analysis of data was performed using SPSS 17.0 software. One-way ANOVA was used to analyse the differences between more than two groups, and non-paired student's t test was used to analyse the difference between two groups. P < 0.05 was considered statistically signi cant.

Declarations
Ethical Approval and Consent to participate      The expression differences of HIF-1α, Bcl-2, Apaf-1, cleaved caspase 3, Beclin-1 and LC-3 mRNA of rat intervertebral disc cells in different treatment groups were detected by RT-PCR. The expression differences of HIF-1α (A), Bcl-2 (B), Apaf-1 (C), cleaved caspase 3 (D), Beclin-1 (E) and LC-3 (F) mRNA were showed and analyzed by 2-△△CT method. The above mRNA expressions in the samples was normalized against the expression of GAPDH, and the mRNA expression in untreated cells (blank group) was normalized to 1.