NPCs were isolated from eight-week-old male Sprague-Dawley rats, which were brought from Animal Center of the Xinqiao Hospital. All experiments were approved by the Ethics Committee of the Army Medical University (Code: AMUWEC20211846). Rats were sacrificed after appropriate anesthesia. Nucleus pulposus tissue was separated from caudal vertebra and cut into small pieces. After digestion with 0.1% collagenase for 6 h, the suspension was centrifuged at 300 g for 5 min and then resuspended in DMEM-F12 (BI, Kibbutz Beit-Haemek, Israel) media with 10% fetal bovine serum (FBS, LONSA, Canelones, UY) and 100 U/mL penicillin-streptomycin. Afterwards, the partially digested tissue was cultured in an incubator containing 5% CO${}_2$, 20% O${}_2$ with humidified atmosphere at 37 ${}^{\circ }$C. After 1-week incubation, NPCs migrated from the partially digested tissue, then incubated in culture flasks in humidified atmosphere containing 5% CO${}_2$, 20% O${}_2$ at 37 ${}^{\circ }$C. Nonadherent cells were removed by replacing the media after 48 h. At 90% confluence, NPCs were washed with phosphate-buffered saline (PBS) twice, and then passaged in a 1:2 ratio. Culture media was changed every 3 days. Cells used in this research were Passage 3–Passage 5.

NPCs in all groups were cultured in DMEM-F12 media with 10% fetal bovine serum and 100 U/mL penicillin-streptomycin. CoCl${}_2$ (Sigma, Saint Louis, MO, USA) was dissolved in DMEM-F12 for a stock concentration of 10 mmol/L and then diluted in culture media for required concentrations. For physical hypoxia, NPCs were cultured in an incubator with humidified atmosphere containing 1% O${}_2$, 5% CO${}_2$ at 37 ${}^{\circ }$C. For mimetic-hypoxia, NPCs were incubated with different concentrations of CoCl${}_2$ solution (50 $\mu$M, 100 $\mu$M, 200 $\mu$M, 300 $\mu$M, 400 $\mu$M) and cultured in an incubator with humidified atmosphere containing 20% O${}_2$, 5% CO${}_2$ at 37 ${}^{\circ }$C.

To measure the cell viability of NPCs, a CCK-8 assay (CCK-8, Beyotime, Shanghai, China) was performed according to the manufacturer’s instructions. CCK-8 is a colorimetric reaction-based assay that yields an orange formazan dye to an extent proportional with the cell number. The cell viability of NPCs was calculated by evaluating the absorbance at 450 nm on a spectrophotometer [20]. NPCs were seeded into 96-well plates (2000 cells/100 $\mu$L) and cultured in groups of normoxia, physical hypoxia (1% O${}_2$) and CoCl${}_2$ mimetic-hypoxia (50 $\mu$M, 100 $\mu$M, 200 $\mu$M, 300 $\mu$M, 400 $\mu$M). After 24 h, 48 h and 72 h, cell culture media was replaced with 110 $\mu$L CCK-8 solution (10% concentration) per well and incubated for 2.5 h, respectively. The OD value of each well was measured to obtain Ac (absorbance of control well), As (absorbance of the experimental well), Ab (absorbance of blank well) by a microplate reader at 450 nm (Spectra Max M2, Molecular Devices, Sunnyvale, CA, USA). Cell viability (%) = [(As-Ab) / (Ac-Ab)] $\times$ 100.

Apoptosis was detected using the Annexin V-PE/7-AAD apoptosis detection kit (BD Pharmingen, Franklin Lakes, NJ, USA) according to the manufacturer’s instructions. After treatments of normoxia, physical hypoxia (1% O${}_2$) and CoCl${}_2$ mimetic-hypoxia for 24 h, 48 h and 72 h, NPCs were collected and suspended in 1$\times$ binding buffer at a concentration of 1 $\times$ 10${}^6$ cells/mL. 100 $\mu$L of this cell suspension (1 $\times$ 10${}^5$ cells) was then transferred to a 5 mL culture tube. After addition of 5 $\mu$L of Annexin V-PE and 5 $\mu$L of 7-AAD, the solution was gently mixed and incubated for 15 min at room temperature in the dark. Subsequently, 400 $\mu$L of 1$\times$ binding buffer was added to each tube. Flow cytometry of NPCs was performed within 1 h, with Annexin V positive cells representing the occurrence of apoptosis.

Scratch tests were conducted to determine the effect of normoxia, physical hypoxia (1% O${}_2$), and CoCl${}_2$ mimetic-hypoxia on NPCs migration. A total of 3$\times$10${}^5$ NPCs cells were seeded into 6-well plates and cultured overnight in media containing 10% FBS. A scratch wound was generated in the center of each well by a sterile 200 $\mu$L pipette tip. After washing twice with PBS, fresh serum-free culture media was added to the plates. Cells were then cultured in physical hypoxia (1% O${}_2$) and CoCl${}_2$ mimetic-hypoxia, respectively. Images were taken at 0 h, 12 h, 24 h using an inverted phase contrast microscope and cell migration area was quantified using ImageJ software (National Institutes of Health, New York, NY, USA).

Cells were cultured in 6-well plates and incubated for 24 h, 48 h, 72 h under normoxia (20% O${}_2$), physical hypoxia (1% O${}_2$) and CoCl${}_2$ mimetic-hypoxia (50 $\mu$M, 100 $\mu$M, 200 $\mu$M, 300 $\mu$M, 400 $\mu$M). The conditioned cell media was collected and analysed with a pH meter (Sartorius, Goettingen, Germany). Data were collected and analysed to detect the extracellular pH change.

Intracellular accumulation of ROS was measured by DCFH-DA (Solarbio, Beijing, China) following the manufacturer’s instructions. After physical hypoxia and CoCl${}_2$ treatments for 6 h, cell samples were washed three times with serum-free culture media and stained with 10 $\mu$M of DCFH-DA in serum-free culture media for 20 min at 37 ${}^{\circ }$C in the dark. The cells were washed three times with serum-free culture media to remove unbound probe outside of the cells. Cells were then collected in 5 mL polystyrene tubes and the mean fluorescence intensity of DCF was analyzed with an excitation wavelength of 488 nm and an emission wavelength of 525 nm.

Following treatment of CoCl${}_2$ (0 $\mu$M, 50 $\mu$M, 300 $\mu$M) and hypoxia (1% O${}_2$), NPCs were washed twice with PBS and then treated with the RNAiso Plus (Takara, Tokyo, Japan) to extract total RNA from samples based on manufacturer’s protocols. PrimeScript RT reagent kit (Takara, Tokyo, Japan) was used for RNA reverse transcription to synthesize cDNA. SYBR Premix Ex Taq II Kit (Takara, Tokyo, Japan) was used for amplification and detecting the relative mRNA expression of target genes with the Real-Time PCR System (Cobas z 480, Basel, Switzerland). The primers for QRT-PCR are presented in Table 1. $\beta$-actin was used to normalize target gene mRNA and we utilized formula 2${}^{-\mathrm{\Delta }\mathrm{\Delta }Ct}$ to measure the relative mRNA expression.

NPCs were incubated under normoxia, with selected concentrations of CoCl${}_2$ (50 $\mu$M, 300 $\mu$M) and physical hypoxia for 24 h, respectively. Total protein was extracted by RIPA buffer (Beyotime Biotechnology, Shanghai, China) containing 1% PMSF (Beyotime Biotechnology, Shanghai, China) and then centrifuged at 12,000 g at 4 ${}^{\circ }$C for 5 min to discard the cell debris. Protein concentrations were determined by standard bicinchoninic acid (BCA; Beyotime Biotechnology, Shanghai, China) method. A total of 40 $\mu$g of protein was loaded into each well and separated by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by transfer to polyvinylidene fluoride membranes (PVDF; Merck Millipore, Darmstadt, Germany). Membranes were blocked by 5% nonfat milk in TBST, incubated with specific antibodies overnight at 4 ${}^{\circ }$C, and then washed with TBST three times. The membranes were then incubated with horseradish peroxidase-labeled secondary antibody (dilution 1:5000) for 1.5 h and visualized using an enhanced chemiluminescence substrate (Bio-Rad, Hercules, CA, USA) and Bio-Rad Chemidoc (Hercules, CA, USA). The relative level of target protein to $\beta$-ACTIN was calculated by using the ImageJ software. The antibodies used were as follows: $\beta$-ACTIN (Abcam, Cambridge, UK, 1:1000 dilution), COL2/ACAN/SOX9/MMP1/TIMP1/BAX/BCL2 (Proteintech, Wuhan, China, 1:500 dilution), P53 (Proteintech, Wuhan, China, 1:200 dilution), HIF1$\alpha$/GLUT1 (Abcam, Cambridge, UK, 1:1000 dilution).

Data are presented as mean $\pm$ standard deviation (SD) for at least three independent experiments. SPSS 23.0 software (SPSS Inc. IL, New York, NY, USA) was used to conduct statistical analysis. Multiple comparison of data among the groups were determined by a one-way ANOVA followed by the least significant difference test (Fisher test) and statistical significance was evaluated by an unpaired Student’s test for comparisons between two means. Differences were considered statistically significant when p 0.05.

A dose response test of CoCl${}_2$ was performed to explore the effects of various concentrations on cell viability. When CoCl${}_2$ concentration was no more than 100 $\mu$M, no significant difference in cell viability was found in NPCs with normoxia (20% O${}_2$) or physical hypoxia (1% O${}_2$) at 24 h, 48 h and 72 h, respectively. However, when the concentration of CoCl${}_2$ increased (over 200 $\mu$M), the cell viability of NPCs under CoCl${}_2$ were significantly inhibited compared with NPCs in normoxia or physical hypoxia at 24 h, 48 h and 72 h, respectively (p 0.05) (Fig. 1).

Fig. 1.

The cell viability of NPCs measured by CCK-8 assay in normoxia (20% O${}_{\mathrm{2}}$), physical hypoxia (1% O${}_{\mathrm{2}}$), and CoCl${}_{\mathrm{2}}$-mimetic hypoxia. Data are expressed as mean $\pm$ SD from three independent experiments (*p 0.05, **p 0.01, ***p 0.001, ****p 0.0001 vs. normoxia; No statistical significance/ns).

Flow cytometry results demonstrated that there were no significant differences between the apoptosis rates of NPCs under CoCl${}_2$ (50 $\mu$M and 100 $\mu$M), 1% O${}_2$ and normoxia (p $>$ 0.05). When the concentration of CoCl${}_2$ was greater than 200 $\mu$M, cell apoptosis rates increased (Fig. 2A,B).

Fig. 2.

The apoptosis rates of NPCs measured by flow cytometry assay in normoxia (20% O${}_{\mathrm{2}}$), physical hypoxia (1% O${}_{\mathrm{2}}$), and CoCl${}_{\mathrm{2}}$-mimetic hypoxia. (A) Cell apoptosis detected by flow cytometry analysis after Annexin V-PE/7-AAD double-staining. Apoptotic rate was represented as a percentage of total cell populations. The proportion of dead cells (annexin V-/7AAD+), live cells (annexin V-/7AAD-), early apoptotic cells (annexin V+/7AAD-) and late apoptotic cells (annexin V+/PI+) was measured for comparison; (B) Histograms showing the apoptosis rate (the sum of early and late apoptotic cells) of NPCs treated for 24 h, 48 h and 72 h in normoxia, physical hypoxia, and CoCl${}_2$-mimetic hypoxia. The values are expressed as mean $\pm$ SD from three independent experiments (***p 0.001, ****p 0.0001 vs. normoxia; No statistical significance/ns).

The scratch test was conducted to explore the effect of physical hypoxia, and CoCl${}_2$-mimetic hypoxia on NPCs migration. 1% O${}_2$ stimulated hypoxia elicited stronger NPC migration than CoCl${}_2$-mimetic hypoxia and normoxia, as shown by the NPCs healing area/wounded area at 0 h, 12 h and 24 h after scratch injury (p 0.05). The migration speed of NPCs under CoCl${}_2$ was negatively correlated with the concentration of CoCl${}_2$ (p 0.05) (Fig. 3A,B).

Fig. 3.

Effect of physical hypoxia, and CoCl${}_{\mathrm{2}}$-mimetic hypoxia on NPC migration and extracellular pH. (A) Images illustrating NPCs migration detected by scratch test at 0 h, 12 h, and 24 h; (B) Histogram showing the migration ratio of NPCs treated with normoxia, physical hypoxia, and CoCl${}_2$-mimetic hypoxia for 12 h and 24 h; (C) Extracellular pH of NPCs culture media collected at 24 h, 48 h, and 72 h under normoxia, physical hypoxia, and CoCl${}_2$-mimetic hypoxia; (D) Extracellular pH detected by pH meter at 24 h, 36 h and 72 h, respectively. pH of 1% O${}_2$ group was of the lowest at 24 h, 48 h and 72 h. Data are expressed as mean $\pm$ SD from three independent experiments (*p 0.05, **p 0.01, ***p 0.001, ****p 0.0001 vs. normoxia; No statistical significance/ns).

The pH of culture media was determined at 24 h, 48 h and 72 h after normoxia, physical hypoxia, and CoCl${}_2$-mimetic hypoxia. In each group, extracellular pH was time-dependently downregulated. The pH of culture media under 1% O${}_2$ were lower than those under normoxia and CoCl${}_2$ treatment at 24 h, 48 h and 72 h (p 0.05) (Fig. 3C,D).

The intracellular ROS level was detected by flow cytometry. Compared with normoxia, NPCs under physical hypoxia and CoCl${}_2$ mimetic-hypoxia exhibited upregulated ROS generation. ROS fluorescence intensity was dose-dependent under CoCl${}_2$ culture (Fig. 4A,B).

Fig. 4.

The intracellular ROS of NPCs detected by flow cytometry after DCFH-DA staining. (A) ROS detected by flow cytometry after treated with 1% O${}_2$ and CoCl${}_2$ (50 $\mu$M, 100 $\mu$M, 200 $\mu$M, 300 $\mu$M, 400 $\mu$M) for 6 h and DCFH-DA staining for 20 min (Red histogram represents group of normoxia); (B) Histogram showing the ROS inside NPCs treated with 1% O${}_2$ and CoCl${}_2$ for 6 h compared with normoxia. Data are expressed as mean $\pm$ SD from three independent experiments (***p 0.001, ****p 0.0001).

As 50 $\mu$M CoCl${}_2$ appeared to have similar cell viability compared with 1% O${}_2$, while 300 $\mu$M induced cell apoptosis of NPCs, we analyzed the mRNA expression of several biomarkers of NPCs under normoxia, physical hypoxia and CoCl${}_2$ mimetic-hypoxia at low concentration (50 $\mu$M) and high concentration (300 $\mu$M). Compared with normoxia, both 1% O${}_2$ and CoCl${}_2$ groups upregulated the transcription of Hif1$\alpha$. The elevation of Hif1$\alpha$of CoCl${}_2$ groups was concentration-dependent. Acan, Col2a1, Sox9, Mmp1 and Timp1 had the same tendency with Hif1$\alpha$ while Glut1 was upregulated most in the physical hypoxia group. The ratio of Mmp1/Timp1 was upregulated in CoCl${}_2$ (300 $\mu$M) while it was downregulated in physical hypoxia (Fig. 5A).

Fig. 5.

The relative mRNA and protein expression of Hif1$\mathrm{\alpha }$, Glut1, Acan, Col2a1, Sox9, Mmp1 and Timp1 of NPCs under normoxia, physical hypoxia, and CoCl${}_{\mathrm{2}}$-mimetic hypoxia. (A) The relative mRNA expressions of target genes with QRT-PCR. $\beta$-actin was used to normalize target gene mRNA and formula 2${}^{-\mathrm{\Delta }\mathrm{\Delta }Ct}$ were utilized to measure the relative mRNA expression compared to normoxia. Data are expressed as mean $\pm$ SD from three independent experiments; (B) The Western blot bands of target protein HIF1$\alpha$, GLUT1, ACAN, COL2A1, SOX9, MMP1 and TIMP1; (C) Histogram exhibiting the relative protein level compared to normoxia group. Data are expressed as mean $\pm$ SD from three independent experiments (*p 0.05, **p 0.01, ***p 0.001, ****p 0.0001; No statistical significance/ns).

The protein levels of several biomarkers of NPCs were analysed under normoxia, physical hypoxia and CoCl${}_2$-mimetic-hypoxia at low concentration (50 $\mu$M) and high concentration (300 $\mu$M). Compared with normoxia (20% O${}_2$), proteins such as HIF1$\alpha$, GLUT1, SOX9, ACAN, COL2, MMP1 and TIMP1 had similar increased levels under physical hypoxia and CoCl${}_2$-mimetic-hypoxia (Fig. 5B,C). Under 1% O${}_2$ hypoxia, the level of GLUT1 was upregulated more than that under CoCl${}_2$-mimetic-hypoxia. The BAX/BCL2 ratio was upregulated in CoCl${}_2$ compared with normoxia or physical hypoxia (p 0.05). P53 was detected in CoCl${}_2$ treated groups and the trend was coincident with HIF1$\alpha$ expression, while it was expressed at low level under normoxia or physical hypoxia (Fig. 6A,B).

Fig. 6.

Relative expression of protein P53, BAX, BCL2 of NPCs under normoxia, physical hypoxia, and CoCl${}_{\mathrm{2}}$-mimetic hypoxia. (A) The Western blot bands of target protein P53, BAX, BCL2; (B) Histogram exhibiting for statistical analysis of the relative protein expression compared to normoxia group. Data are expressed as mean $\pm$ SD from three independent experiments (*p 0.05, **p 0.01, ***p 0.001, ****p 0.0001; No statistical significance/ns).

SOX9 is considered as one of the biomarkers for NPC [22, 23, 24]. In our study, physical hypoxia and CoCl${}_2$-mimetic hypoxia all upregulated the level of SOX9. This result is consistent with previous studies [6] and that the upregulation of SOX9 under hypoxia may be via HIF1$\alpha$ pathway [25].

Nucleus pulposus tissue contain abundant ECM including collagen II (COL2) and aggrecan (ACAN) that are responsible for maintaining the mechanical load of IVD [26, 27]. Over the years, many studies have shown that hypoxia could induce various cell types to enhance COL2 and ACAN expression via HIF1$\alpha$, including NPCs [10, 28, 29, 30, 31, 32]. Matrix metalloproteinases (MMPs) are endopeptidases of the ECM that have the ability to degrade almost all known components of the ECM in IVDs. Among these, MMP1 is a collagenase that is precisely regulated by its endogenous protein inhibitors, the tissue inhibitors of metalloproteinases 1 (TIMP1) [33, 34]. In our study, both physical hypoxia and CoCl${}_2$ -mimetic- hypoxia upregulated the levels of COL2 and ACAN compared with normoxia. In addition, both groups upregulated MMP1 and upregulated the level of TIMP1, which is consistent with previous studies [35, 36]. The level of MMP1 was coincident with the increased expression of HIF1$\alpha$. In the group treated with CoCl${}_2$ at 300 $\mu$M concentration, the MMP1/TIMP1 ratio was higher than normoxia, 50 $\mu$M CoCl${}_2$, and physical hypoxia, which may be attributed to the high level of HIF1$\alpha$ as has been previously reported [21].

It has been confirmed that hypoxia could slightly promote cell viability for NPCs via HIF1$\alpha$ [7, 37]. But few studies have focused on the impact of CoCl${}_2$ on NPCs cell viability. He et al. [38] performed CCK-8 tests and found that 10 $\mu$M to 100 $\mu$M CoCl${}_2$ was safe in mimetic-hypoxia for the in vitro study of NPCs. Jiang et al. [39] treated NPCs with 200 $\mu$M CoCl${}_2$ for mimetic hypoxia. In our study, both physical hypoxia and low concentration CoCl${}_2$ (50 $\mu$M, 100 $\mu$M) slightly enhanced the cell viability detected by CCK-8 assay but these differences were not statistically significant. Nevertheless, high concentrations of CoCl${}_2$ ($>$200 $\mu$M) appeared to inhibit cell viability. Thus, we conclude that as a chemical compound, CoCl${}_2$ has some toxic effects on NPCs viability, but the effects may be minimal at low concentrations. Once the critical concentration of 200 $\mu$M is exceeded, CoCl${}_2$ will be detrimental to the NPCs.

Cell migration is a fundamental biological process involved in tissue homeostasis and is an important part of cell transplantation. It has a complex mechanism which is still not clear and may be related to the MAPK signaling pathway, phosphatidylinositol signaling pathway and cytokine–cytokine receptor pathway [40, 41]. Many studies have shown that cells migrate faster under hypoxia than normoxia, including mesenchymal stem cells and carcinoma cells [42, 43, 44]. Magdaleno et al. [43] compared the migration of renal carcinoma cells under hypoxia and CoCl${}_2$ and found that physical hypoxia promoted migration but CoCl${}_2$ failed to achieve similar effect. They concluded that HIF1$\alpha$ independent mechanisms modulate the divergent outcomes in assembly of fibronectin, which is a core matrix protein that assembles to promote cell migration. Heirani-Tabasi et al. [41] compared the effects of three hypoxia-mimicking agents on migration-related signaling pathways in mesenchymal stem cells and found that CoCl${}_2$ failed to promote cell migration and the mechanism of which may be partly related to the MAPK signaling pathway through IL8/CXCR2 axis or similar mechanisms. In addition, some studies found that CoCl${}_2$ inhibited cell migration compared with normoxia [2, 45]. In our study, it was obvious that NPCs migrated faster under physical hypoxia compared with normoxia, while NPCs migrated slower in all groups under CoCl${}_2$ mimetic-hypoxia. Our data indicated that stabilization of HIF1$\alpha$ under mimetic-hypoxia is not sufficient to enhance cell migration and it requires the synergistic contribution of some other signaling pathways driven by physical hypoxia to affect this phenotype, the detailed mechanism of which still needs further research.

Under hypoxia, energy metabolism is switched from oxidative phosphorylation to glycolysis by upregulating the expression of glycolytic enzymes and glucose transporters [46]. Studies have shown that NPCs keep HIF1$\alpha$ under normoxia and generate energy through anaerobic glycolysis [19, 47]. Under hypoxia, NPCs could express more GLUTs to facilitate glucose transport [48]. In our study, both physical hypoxia and CoCl${}_2$ mimetic-hypoxia upregulated the expression of GLUT1, but interestingly the physical hypoxia group expressed the highest GLUT1 and downregulated the extracellular pH the most. This may indicate that physical hypoxia promoted glycolysis in NPCs compared with normoxia but CoCl${}_2$ did not. This result concurs with a recent research report by Zhigalova et al. [49]. They performed RNA-seq experiments to explore transcriptomes of human Caki-1 cells under real hypoxia and CoCl${}_2$ treatment and found that glycolysis was not controlled by HIF1$\alpha$, indicating that CoCl${}_2$ failed to affect some of the essential downstream consequences of hypoxia, particularly the glycolysis/gluconeogenesis pathway. There may be some underlying mechanisms which trigger the downstream events of NPCs glycolysis in hypoxia apart from HIF1$\alpha$.

As our results have shown, both physical hypoxia and CoCl${}_2$ mimetic-hypoxia upregulated the ROS level of NPCs. For the CoCl${}_2$ groups, ROS generation had a positive correlation with concentration ($\le$400 $\mu$M), which is consistent with previous research that demonstrated that CoCl${}_2$ could stimulate cells to generate more ROS and thus cause a negative impact on cell survival [50]. Though the result may be similar, the mechanism could be different. In research of Hep3B cells and wild-type Hep3B cells, using either hypoxia (1.5% O${}_2$) or CoCl${}_2$ incubation, it was found that physical hypoxia activates ROS generation through a mitochondria-dependent signaling pathway, while CoCl${}_2$ stimulating ROS generation via a mitochondria-independent mechanism [51]. The precise details of this mechanism of ROS generation under physical hypoxia and CoCl${}_2$ for NPCs still needs more in-depth study.

As is shown in our study, the apoptosis rates of NPCs under both physical hypoxia and CoCl${}_2$ mimetic-hypoxia (200 $\mu$M) had no obvious difference compared with normoxia. But when the concentration of CoCl${}_2$ was larger than 200 $\mu$M (including 300 $\mu$M, 400 $\mu$M), the apoptosis rates rose significantly. This is in line with the findings of Bae et al. [52] that enhanced hypoxia by further increasing CoCl${}_2$ concentrations can promote cell apoptosis. We found that both the ratio of BAX/BCL2 and P53 expression have a positive correlation with the concentration of CoCl${}_2$, but they were all downregulated under physical hypoxia. Previous studies reported that CoCl${}_2$ induces cell apoptosis via different pathways. For example, CoCl${}_2$-induced HIF1$\alpha$ expression correlated with apoptosis and may be related to the PI3K/Akt pathway [53]. Other researchers revealed that apoptosis associated with oxidative stress and DNA damage [54, 55, 56, 57], may involve P53. Rana et al. [58] found that P53 in breast cancer cells is HIF1$\alpha$-dependent and overexpression of HIF1$\alpha$-dependent BAX ultimately leads to apoptosis. They speculated that hypoxia affects the P53-dependent pathway in a HIF1$\alpha$-dependent manner, thereby targeting the genes involved in P53 pathway which alters the expression of pro-apoptotic genes. Additionally, Lee et al. [59] reported that CoCl${}_2$ induced apoptosis, through both mitochondria and death receptor-mediated pathways, is regulated by the BCL2 family in mES cells. In our study, NPCs under hypoxia down-regulated P53 but CoCl${}_2$ up-regulated it. According to Zhang et al. [60], hypoxia appears to regulate P53 and is related to the severity of hypoxia, resulting in the increase or decrease of P53 levels and activities in cells. We thus speculated that the way that high concentrations of CoCl${}_2$ induces apoptosis in NPCs is through excessive HIF1$\alpha$, P53, superfluous ROS or cell toxicity of CO${}^{2+}$. The mechanisms of these processes need to be further explored in future studies.

There are some shortcomings in our experiments. Firstly, we just compared the similarity and difference of some primary phenotypes between physical hypoxia and CoCl${}_2$ mimetic-hypoxia but didn’t investigate further underlying mechanisms. Secondly, the hypoxia incubator we used could only be set at 1% O${}_2$ as hypoxia, and the effects of different O${}_2$ concentrations on NPCs were not examined.

Generally, this research has presented an experimental study of the CoCl${}_2$ for mimetic-hypoxia environment for culturing NPCs in vitro. This may bring convenience and enlightenment for other researchers studying NPCs and IVVD.