The animal ethics committee of Guangzhou Red Cross Hospital approved the research. Following previously published methods (28), a 3 cm × 3 cm wound was made on the back skin of the rats. The wound was disinfected every day, and the rats were provided with enough food and water to ensure their normal activities and survival. After 1 week, 0.8% pentobarbital was injected into the abdominal cavity of these animals for anesthesia. After anesthetization, approximately 5 mL of abdominal aortic blood was collected using a fine-needle approach, and the blood sample was diluted to 1:1 by PBS. Mononuclear cells (MNCs) were separated and collected with Ficoll separation solution (GBCBIO Technologies, Guangzhou, China) and centrifuged at 2000 rpm for 35 min. The middle layer was pipetted with a thin tube and washed twice with phosphate-buffered saline (PBS). MNCs (2×10⁶/mL) were seeded onto the T-25 flask with 10 mL of complete Dulbecco’s Modified Eagle Medium (Gibco, MA). The complete medium contained 1% penicillin/streptomycin (Gibco), 20 ng/mL bFGF (R&D Systems, MN), and 20% fetal bovine serum (Gibco). With 21 days of culture, the cell convergence was 80%, and the third-generation cells digested by 0.25% trypsin were used for subsequent experiments. Representative bright-field images were captured by an inverted phase-contrast microscope (Nikon ECLIPSE Ts2, Nikon).

The cell immune phenotypes of third-generation PBMSCs (P3 PBMSCs) were identified by flow cytometry (FCM). CD73, CD105, and CD90 (R&D Systems, US) were selected as positive markers of rPBMSCs, whereas CD14, CD19, CD34, CD45, and HLA-DR (BD biosciences, US) were chosen as negative markers of rPBMSCs. rPBMSCs (2×10⁵ cells/mL) were resuspended in PBS and mixed in CD14, CD19, CD34, CD45, CD73, CD90, CD105, and HLA-DR antibody solutions for 30 min, and the cell samples were then loaded on the machine for analysis.

rPBMSCs treated under normoxic and hypoxic (5% O₂) conditions for 24 h were collected, and rPBMSCs were then fixed with 70% (V/V) ethanol overnight. Moreover, 50 μg/mL propidium iodide (PI) (Beyotime Biotechnology, Shanghai, China) was diluted by PBS solution containing 1% Triton X-100. Cells were fully infiltrated in the freshly prepared PI solution for 30 min and were analyzed by a BD FACScan flow cytometer (BD Company, CA).

P3 rPBMSCs were seeded into a 24-well plate at a density of 2 × 10⁴/well and cultured at 37°C in an incubator with 5% CO₂. When the cells grow to 70% confluence, chondrogenesis induction, osteoinduction, and adipogenesis tests were performed. For chondrogenesis, cells were induced for 21 days in a chondrogenesis induction medium kit (RAXMX-90041, Cyagen Biosciences, US). The differentiation was evaluated by alcian blue staining. For osteogenesis, the cultures were induced with an osteogenesis induction medium kit (RAXMX-90021, Cyagen Biosciences, CA). After culture for 21 days, alizarin red staining was performed to evaluate the osteogenic products. For adipogenesis, cells were induced for 21 days in an adipogenesis induction medium kit (RAXMX-90031, Cyagen Biosciences). The formation of lipid vacuoles was assessed by Oil Red O staining. All images were captured under an inverted phase-contrast microscope (Nikon ECLIPSE Ts2, Nikon).

P3 rPBMSCs were seeded into a 24-well plate at a density of 2 × 10⁴/well and cultured at 37°C in a 21% O₂ and 5% CO₂ incubator or a 5% O₂ and 5% CO₂ incubator. The induction medium and experimental procedures performed in the subsequent experiments were the same as the methods described in “Multilineage Differentiation Potential Assay.”

P3, P5, and P6 rPBMSCs (2× 10³/well) were inoculated in microplates (24-well) in 5% CO₂ incubators with a gradient concentration of oxygen at 37°C. The experiments were set up as the control group (21% O₂ and 5% CO₂), 19% O₂ and 5% CO₂ hypoxia group, 5% O₂ and 5% CO₂ hypoxia group, and 1% O₂ and 5% CO₂ hypoxia group, with three replicate wells in each group. Starting from the next day, each group of cells was digested and counted accurately with a cell counter at each time point (Days 1–8). The growth curves of each cell group were made according to the number of cells. For the MTS assay, the above-mentioned groups of cells were planted on the well plate after Day 8, and the absorbance was measured at 450 nm by a multifunctional microplate reader (BioTek, US).

After the cells reached 80-90% confluency, cells were passaged and counted. Calculate the cumulative population doubling (CPD) value using the following formula (29):

CPD was plotted against time in culture and performed in triplicate for each counting procedure.

Moreover, 500 rPBMSCs were cultured in 6-well plates in an incubator capable of adjusting oxygen concentration for 14 days. After fixation with paraformaldehyde for 15 min, 1 mL of crystal violet staining solution was added to the culture plate for staining clones for 30 min. Under an inverted phase-contrast microscope (Nikon ECLIPSE Ts2, Nikon), the number of clones containing more than 50 cells was counted.

rPBMSCs were collected after normoxic and hypoxic (5% O₂) treatments for 5 days, and whole-cell lysates were prepared for Western blotting in radioimmunoprecipitation assay buffer. Then, 30 μg of protein was loaded into the sample well, dispersed in the gel according to the molecular weight, and directly transferred to the poly(vinylidene fluoride) membrane (Bio-Rad, CA) in a band-to-band manner through the semi-dry transfer method. The membranes were immersed in a square dish filled with primary antibody diluent. These antibodies (HIF-1α, 36169; β-catenin, 8400; SOX2, 3579; CyclinE, 4132; Nanog, 8822; GAPDH [glyceraldehyde 3-phosphate dehydrogenase], 5174) were purchased from Cell Signaling Technology (MA), and when used, the dilution ratio was 1:1000. On the next day, membranes were incubated with secondary anti-rabbit/mouse IgG, HRP-linked antibody (#7074/7076, 1:3000, Cell Signaling Technology). The electrochemiluminescence detection mixture was used to detect the protein on the membranes. ChemiDoc XRS imaging system with Image Lab software (Bio-Rad) was used to analyze the graphs.

Furthermore, 10⁴/well rPBMSCs were seeded in glass slides placed in plates treated under normoxia and hypoxia (5% O₂) for 5 days. After sequential fixation, blocking, incubation of primary (β-catenin,1:200, 8242, Cell Signaling Technology; HIF-1α, 1:200, #36169, Cell Signaling Technology) and secondary (1:200, ZF0311, OriGene Technologies, MD) antibodies, a fluorescence microscope (Ti2-U, Nikon) was used to observe and capture pictures of interest.

The culture method of rPBMSCs was the same as with Western blot. Total RNA obtained by the TRIzol method was reversed into cDNA in the PrimeScript RT Master mix reaction system (Takara Bio, Japan). With reference to the instructions, SYBR-Green reagent (Takara Bio) was used to perform real-time PCR in triplicate in a fluorescence quantitative PCR instrument (Jena, Germany). GAPDH was used as a control to analyze relative gene expression in the 2^-ΔΔCt formula (30). Primer sequences are presented in Table 1 .

Data in three replicates are presented as mean ± standard deviation. Student’s t-test or one-way analysis of variance was used to analyze differences between the two groups and among multiple groups; P < 0.05 was used to mark significant differences.

On the day after inoculation, round or polygonal adherent cells were observed in the primary culture. After 7 days, colonies gradually formed. After approximately 16 days, the cell coverage area was 70%–80% of the bottom of the culture flask. At approximately 21 days later, the cell growth reached 100% ( Figure 1A ). Flow cytometry experiments revealed that rPBMSCs had high expression of CD73, CD90, and CD105, extremely low expressions of CD14, CD19, CD34, CD45, and HLA-DR ( Figure 1B ). Oil Red O staining indicated that rPBMSCs can differentiate into adipocytes embellished by red-stained lipid droplets. Alician blue staining demonstrated that after 21 days of induction, rPBMSCs could differentiate into chondrocytes embellished by blue-stained proteoglycans. Alizarin red staining presented that rPBMSCs could differentiate into osteoblasts embellished by red-stained bone nodules under osteogenic conditions ( Figure 1C ).

The results of the experiments are displayed in Figures 2A, B . In the first 2 days of culture, different concentrations of O₂ had no noticeable effects on the proliferation of third-generation PBMSCs (P3 rPBMSCs), fifth-generation PBMSCs (P5 rPBMSCs), and sixth-generation PBMSCs (P6 rPBMSCs). After 3 days, hypoxia (5% O₂) significantly increased the number of cells and proliferation rate of P3, P5, and P6 rPBMSCs. After 8 days, the cells approached the plateau stage. At this time, the number of P3 rPBMSCs in the control group, 19% O₂ hypoxia, 5% O₂ hypoxia, and 1% O₂ hypoxia groups were 55 × 10³, 70 × 10³, 96 × 10³, and 71 × 10³, respectively. Statistical analysis showed that compared with the number of P3 rPBMSCs in the control group, those in the 19% O₂ hypoxia, 5% O₂ hypoxia, and 1% O₂ hypoxia groups were increased significantly (P < 0.05). Compared with 19% O₂ hypoxia and 5% O₂ hypoxia, 1% O₂ hypoxia further increased the number of P3 rPBMSCs (P < 0.05). Similar to the growth curve, 5% O₂ hypoxia significantly promoted the absorbance of P3 rPBMSCs seeded at Day 8. The CPD curve of P6 PBMSCs proved that the CPD value of PBMSCs in the 5% hypoxia group was significantly higher than that in the normoxia group. Compared with the normoxia group, the CPD values for P6 PBMSCs in the 19% O₂ hypoxia and 1% O₂ hypoxia groups did not change significantly on Day 7. The shape of the growth curve and viability of P5 and P6 rPBMSCs were similar with those of P3 rPBMSCs, but the amounts of rPBMSCs and optical density values of P5 and P6 on Day 8 were lower than those of P3 rPBMSCs. Based on the cell growth curve and MTT assay results, 5% O₂ hypoxia was selected for subsequent experiments. As presented in Figures 2C , D , 5% O₂ hypoxia significantly promoted the formation of rPBMSC colonies. The number of rPBMSC colonies in the 5% O₂ hypoxia group was increased by 53% compared with that in the control group (P < 0.05).

FCM was used to investigate the cell cycle transition of rPBMSCs treated under normoxia and hypoxia (5% O₂). Hypoxia exerted a significant increase and decrease in the number of S phase and G1 phase cells, respectively ( Figures 3A, B ). Moreover, 5% O₂ hypoxia increased the percentage of rPBMSCs in the S phase from 27.26% to 46.32% (P < 0.05) and reduced the percentage of rPBMSCs in the G1 phase from 58.24% to 47.19% (P < 0.05). These data indicated that 5% O₂ hypoxia increased the DNA synthesis and cell cycle of rPBMSC progression at the S phase. After 21 days of culture with a differentiation agent under hypoxia or normoxia, the effect of hypoxia on the pluripotency of rPBMSCs was investigated. Figure 3C illustrates that hypoxia increased the ability of induced cells to differentiate into three lines, including osteoblasts, chondrocytes, and adipocytes.

As displayed in Figure 4A , 5% O₂ hypoxia significantly increased the HIF-1α (red) nuclei expression in rPBMSCs, compared with the control rPBMSCs. Simultaneously, the nuclei expression for β-catenin (green) in rPBMSCs was also upregulated significantly with 5% O₂ hypoxia. Immunohistochemistry results ( Figure 4B ) that 5% O₂ hypoxia stimulated the upregulation of HIF-1α and β-catenin expressions are consistent with the promotion of HIF-1α and β-catenin stabilization and nuclear translocation in immunofluorescence experiments ( Figures 4C, D ).

For the objective investigation that hypoxia regulated the self-renewal and stemness of rPBMSCs, pluripotency factors and Cyclin E/CDK2 were primarily selected as indicators. Compared with normoxia, hypoxia significantly triggered upregulation of mRNA and protein expression for β-catenin, CDK2, and Cyclin E in rPBMSCs ( Figures 5A –C ). Moreover, the mRNA and protein expressions of HIF-1α, Nanog, and SOX2 were significantly increased by hypoxia ( Figures 5D–F ).