Physioxia: a more effective approach for culturing human adipose-derived stem cells for cell transplantation.

BACKGROUND: Although typically cultured at an atmospheric oxygen concentration (20-21%), adipose-derived stem cells (ASCs) reside under considerable low oxygen tension (physioxia) in vivo. In the present study, we explored whether and how physioxia could be a more effective strategy for culturing ASCs for transplantation.
METHODS: After isolation, human ASCs were cultured under physioxia (2% O2) and hyperoxia (20% O2) until assayed. WST-8, Transwell, tube formation, β-galactosidase staining, and annexin V-FITC/PI assays were used to evaluate cell proliferation, migration, angiogenesis, senescence, and apoptosis, respectively. Survivability was determined by an ischemia model in vitro and nude mouse model in vivo, and the underlying metabolic alterations were investigated by fluorescence staining, flow cytometry, and real-time polymerase chain reaction.
RESULTS: Compared with those in the hyperoxia group, cells in the physioxia group exhibited increased proliferation, migration, and angiogenesis, and decreased senescence and apoptosis. The increased survival rate of ASCs cultured in physioxia was found both in ischemia model in vitro and in vivo. The underlying metabolic reprogramming was also monitored and showed decreased mitochondrial mass, alkalized intracellular pH, and increased glucose uptake and glycogen synthesis.
CONCLUSIONS: These results suggest that physioxia is a more effective environment in which to culture ASCs for transplantation owing to the maintenance of native bioactivities without injury by hyperoxia.

Background

Since first isolated in 1964 [1], human adipose-derived stem cells (ASCs) have garnered increasing attention [2]. Especially in the recent two decades, after the discovery of their stemness in 2001 [3], a growing body of research has indicated that ASCs possess properties of repair and regeneration, which include angiogenesis [4], multilineage differentiation [5], immunosuppression [6], and homing to ischemic tissues [7]. Consequently, there is great interest in and demand for utilizing ASCs in several clinical applications, such as osteoarthritis, heart failure treatment and wound healing, according to the clinicaltrials.gov database.

However, there are still several problems to resolve, such as the donor choice [8], therapeutic safety [9], and standard protocol for expanding ASCs [10]; among these problems, the most suitable strategy for culturing and expanding ASCs in vitro has been continuously studied. Several factors should be considered, such as the culture medium, serum replacements, and seeding density [11]. However, there is an extremely appropriate standard to which can be referred, the stem cell niche, which is the surrounding microenvironment and intrinsic factors that control the self-renewal and differentiation of stem cells [12, 13].

A distinct difference between “standard culture conditions” and the ASC niche is the oxygen level [14]. Cell culture is typically performed at an atmospheric O2 concentration (20–21%), i.e., the normoxia recognized by most researchers. However, there is a “mental shortcut” [15] neglecting the fact that the normoxia of 20–21% O2 reflects the pathology of humans or animals, while the practical oxygen concentration of the ASC niche is lower, at 2% [16], which is called physioxia [17]. In other words, atmospheric normoxia represents a hyperoxic state for ASCs.

Many biological alterations occur when culturing such cells under hyperoxia (atmospheric normoxia), particularly with respect to metabolism [18], generating changes in cell proliferation [19] and differentiation [20], among others [21]. Underlying these discrepancies is the impact of hypoxia-inducible factor 1 (HIF-1), which is degraded at O2 levels over 5% [15].

By comparison, previous studies have commonly used physioxia at 2% O2 to culture ASCs [22-24] as a transitory approach to increase the expansion and angiogenesis of ASCs rather than as a culture standard through the entire in vitro period, except for some studies [25-29]; yet, these studies did not examine the angiogenesis or survival of ASCs under an ischemic environment. Thus, the aim of the present study was to explore the superiority of physioxia (2% O2) compared with hyperoxia (20% O2) throughout the in vitro culture of ASCs by examining discrepancies in proliferation, migration, senescence, apoptosis, angiogenesis, and survivability, as well as the underlying mechanism.

Methods

Cell isolation and culture

Subcutaneous adipose tissue was collected from the abdomen of four healthy females (age, 25 ± 5 years, body mass index [BMI]: 19–22) after obtaining their consent. After washing with phosphate-buffered saline (PBS), the tissue was minced and digested with 0.2% collagenase (Sigma-Aldrich, St. Louis, MO, USA)/PBS for 40 min at 37 °C. The mixture was washed with PBS and centrifuged at 1000 rpm for 5 min, and the remaining pellet was cultured in α-modified Eagle’s medium (α-MEM; HyClone, GE Healthcare, Marlborough, MA, USA), 10% fetal bovine serum (FBS; Gibco, San Jose, CA, USA), 100 IU penicillin, and 100 mg/mL streptomycin (Solarbio, Beijing, China). Cells in the physioxia group were cultured with 2% O2 (using a modular chamber, Sanyo, Osaka, Japan) and 5% CO2 at 37 °C (physioxia ASCs, P-ASCs) until further analysis in the following tests at passage 3, with 20% O2 and 5% CO2 at 37 °C as a control (hyperoxia ASCs, H-ASCs). Cells from different donors were mixed at passage 2 to explore the general effect on ASCs.

Cell characterization

Flow cytometric analysis

Flow cytometry was used to analyze the surface markers of the ASCs. After detaching, 1 × 105 cells were incubated with PE- or FITC-conjugated antibodies against CD31, CD34, CD73, CD90, CD105, and HLA-DR for 30 min at 4 °C. All antibodies were obtained from Abcam Biotechnology (Abcam, Cambridge, MA, USA). The cells were then analyzed using a BD Accuri™ C6 flow cytometer (BD Biosciences, San Jose, CA, USA).

Adipogenesis

The cells were seeded onto six-well plates. After reaching 80% confluence, the culture medium was changed to α-MEM supplemented with 10% FBS, 1 mmol/L dexamethasone (Sigma-Aldrich, St. Louis, MO, USA), 10 mmol/L insulin (Sigma-Aldrich, St. Louis, MO, USA), 200 mmol/L indomethacin (Sigma-Aldrich, St. Louis, MO, USA) and 0.5 mmol/L 3-isobutyl-1-methylxanthine (IBMX; Sigma-Aldrich, St. Louis, MO, USA) for 7 days. Lipid clusters were stained with oil red O.

Western blotting

Western blotting was performed as previously described [30], with slight modifications. After being dissolved in radioimmunoprecipitation assay (RIPA) buffer (KeyGEN, Nanjing, Jiangsu, China), 30 μg of protein, as detected by bicinchoninic acid (BCA) assay, was separated on a 10% polyacrylamide gel and blotted onto a polyvinylidene fluoride (PVDF) membrane. The membrane was blocked with 5% skim milk and then treated with primary antibodies against HIF-1 (1:1000, 14,179, Cell Signaling Technology, Beverly, MA, USA) and β-actin (1:1000, ab3280, Abcam, Cambridge, MA, USA) overnight at 4 °C, followed by 1 h of incubation with horseradish peroxidase (HRP)-conjugated secondary antibodies at room temperature. The signals were detected with Amersham ECL Select Western Blotting Detection Reagent (GE, Waukesha, WI, USA) according to the manufacturer’s protocol. The signals were visualized using an ImageQuant LAS 4000 mini (GE, Waukesha, WI, USA).

WST-8

Cell Counting Kit 8 (WST-8; Dojindo, Kumamoto, Japan) was used to determine the proliferation of P-ASCs and H-ASCs. The cells (1 × 103) were seeded onto 96-well plates, and after 1, 2, 3, 4, 5, 6, and 7 days, the culture medium was replaced with 100 μL of WST-8 dye solution (90 μL of α-MEM with 10 μL WST-8) for 2 h at 37 °C. Subsequently, the medium was discarded, and the absorbance at 450 nm was detected using a spectrophotometer (Multiskan GO, Thermo Fisher Scientific, Waltham, MA, USA).

Cell doubling curve

ASCs were seeded onto six-well plates at a concentration of 3× 104 per well. The cells were collected at the indicated time points (1, 2, 3, 4, 5, 6, and 7 days), and the cell numbers were measured using an Automated Cell Counter (Bio-Rad, Hercules, CA, USA).

Determination of reactive oxygen species (ROS), mitochondrial mass, and glucose uptake

The ROS level, mitochondrial mass and glucose uptake were determined by staining with 1 μM dihydrodichlorofluorescein diacetate (H2DCFDA, Sigma-Aldrich, St. Louis, MO, USA), 10 nM nonyl acridine orange (NAO, Sigma-Aldrich, St. Louis, MO, USA) and 150 μM 2-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl) amino]-2-deoxy-D-glucose (2-NBDG, Life Technologies, Gaithersburg, MD, USA), respectively, for 30 min at 37 °C. The results were acquired by fluorescence microscopy (Olympus, Hamburg, Germany) and BD Accuri™ C6 (BD Biosciences, San Jose, CA, USA) with a minimum of 5000 events per sample. ROS inhibition was produced using 100 μM butylated hydroxyanisole (BHA, Sigma-Aldrich, St. Louis, MO, USA).

Transwell assay

After incubation in serum-free medium for 24 h, 1 × 105 cells were transferred to the upper chamber of a Transwell (Corning, Corning, NY, USA). Medium containing 10% FBS was added to the lower chamber as a chemoattractant. After 24 h, nonmigratory cells in the upper chamber were removed. The migrated cells were fixed with 4% paraformaldehyde for 30 min, followed by 0.1% crystal violet (Sigma-Aldrich, St. Louis, MO, USA) staining for 15 min. After images were captured, the crystal violet in the cells was extracted by 10% acetic acid for 15 min, and the absorbance at 600 nm was measured using a spectrophotometer (Multiskan GO, Thermo Fisher Scientific, Waltham, MA, USA).

Cell senescence

We assayed the ASCs for senescence-associated β-galactosidase (SA-β-Gal) activity using a Senescence β-Galactosidase Staining Kit (Beyotime, Shanghai, China) according to the manufacturer’s instructions. The SA-β-Gal+ area was calculated by ImageJ (NIH, Bethesda, MD, USA) using the ratio of Periodic acid-Schiff (PAS)+ area to the total area of the image.

Cell apoptosis

ASC apoptosis was measured using an Annexin V-FITC/PI Apoptosis Detection Kit (KeyGEN, Nanjing, Jiangsu, China) according to the manufacturer’s instructions. Flow cytometry was conducted using the BD Accuri™ C6 flow cytometer (BD Biosciences, San Jose, CA, USA).

Tube formation assay

ASCs (2 × 105) were seeded onto 96-well plates coated with Matrigel (Corning, Corning, NY, USA). After incubation at 37 °C for 6 h, the cells were imaged under a microscope (Olympus, Hamburg, Germany). The images were quantified using ImageJ (NIH, Bethesda, MD, USA).

Real-time polymerase chain reaction (RT-PCR)

RNA was extracted using RNAiso Plus (TaKaRa Biotechnology, Dalian, Liaoning, China) according to the manufacturer’s instructions, followed by cDNA synthesis using a First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Waltham, MA, USA). Quantitative RT-PCR was performed using the Eco Real-Time PCR System (Illumina, San Diego, CA, USA) and SYBR Premix Ex Taq (TaKaRa Biotechnology, Dalian, Liaoning, China) with the following conditions: 2 min at 95 °C, followed by 40 cycles of 5 s at 95 °C and 30 s at 60 °C. The relative expression levels were calculated by the 2-ΔΔct method and normalized to the housekeeping gene HPRT. The primer sequences are displayed in Table 1.

Table 1

Primer sequences

Gene	Forward (5′ to 3′)	Reverse (5′ to 3′)
HPRT	CCTGACCAAGGAAAGCAAAG	GACCAGTCAACAGGGGACAT
VEGF	AGGGAAGAGGAGGAGATGAG	GCTGGGTTTGTCGGTGTT
VEGFR2	CTGGCTACTTCTTGTCATCATCCTACG	TGGCATCATAAGGCAGTCGTTCAC
vWF	ACCTTGGTCACATCTTCACATTCACTC	AAGTCATTGGCTCCGTTCTCATCAC
BNIP3	AGGGCTCCTGGGTAGAACTG	ACTCCGTCCAGACTCATGCT
COX4I1	GCCATGTTCTTCATCGGTTT	CATCCTCTTGGTCTGCTTGG
COX4I2	CCCTACACCAACTGCTATGC	CTTCCCTTCTCCTTCTCCTTC
PDK1	AATCACACAGACGCCTAGCA	CATCCTCTTGGTCTGCTTGG
LDHA	ATCTTGACCTACGTGGCTTGGA	CCATACAGGCACACTGGAATCTC
MCT4	ATCTGCTTTGCCATCTTTGC	GTCCAGAAAGGACAGCCATC
NHE2	TTCATGCCACGGATAAATGA	TTCTCTTCAGGCCAGCAAAT
NHE3	AGGTCCATGTCAACGAGGTC	ACTATGCCCTTCACGCAGTC
CAR9	GTCTCGCTTGGAAGAAATCG	ACAGGGCGGTGTAGTCAGAG
GLUT1	CATAGCCACCTCCTGGGATA	AATCACACAGACGCCTAGCA
GLUT3	GCACATAGCTATCAAGTGTGC	AGTGAGAAATGGGACCCTGC
PGM	TGGAAATACGGAATGCTGAA	GCTGCCTTTGATGGAGATG
GYS1	ACCCACCTTGTTAGCCACCT	AACCGCACTTTGTCCATGTC
PYGL	CCAAAGCAGCCACATCATC	GCCCTAACTATCGGGACCAT

Cell survival

The survival assay was conducted as previously described [31]. Briefly, four harsh conditions (ischemic [1% O2, pH 6.4 and 0.56 μM glucose], hypoxic [1% O2, pH 7.4 and 5.6 μM glucose], acidic [20% O2, pH 6.4 and 5.6 μM glucose], and nutrient-depleted [20% O2, pH 7.4 and 0.56 μM glucose] environments) were generated using a modular chamber (Sanyo, Osaka, Japan) and N-2-hydroxyethylpiperazine-N′-2-ethanesulphonic acid-buffered Tyrode’s solution. After incubating the ASCs for 24 h on 96-well plates (1 × 104 per well), live/dead staining and WST-8 were applied to determine the survival of the P-ASCs and H-ASCs.

Intracellular pH detection

ASCs (1 × 104) on 96-well plates were stained with 5 μM 2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein, acetoxymethyl ester (BCECF-AM, Millipore, Billerica, MA, USA) for 30 min at 37 °C. A Multimode Reader (Thermo Fisher Scientific, Waltham, MA, USA) was employed to measure the intracellular pH at excitation and emission wavelengths of 500 nm and 530 nm, respectively. A calibration curve was produced by dyeing ASCs with 5 μM BCECF-AM for 30 min, with subsequent application of an Intracellular pH Calibration Buffer Kit (Thermo Fisher Scientific, Waltham, MA, USA) under different pH values (4.5, 5.5, 6.5, and 7.5) in the presence of 10 μM K+/H+ ionophore nigericin (Thermo Fisher Scientific, Waltham, MA, USA).

PAS staining

We used PAS staining to explore the different expression levels of glycogen in P-ASCs and H-ASCs. Cells on six-well plates were fixed with 4% paraformaldehyde and incubation for 5 min with 0.5% periodic acid (Solarbio, Beijing, China), followed by Schiff’s reagent for 15 min. After the cells were imaged, the PAS+ area was quantified by ImageJ (NIH, Bethesda, MD, USA).

Extracellular lactate assay

The lactate content of the culture medium was measured using a Lactate Assay Kit (KeyGEN, Nanjing, Jiangsu, China) according to the manufacturer’s protocol.

In vivo experiment

Fibrin gel and TdT-mediated dUTP-biotin nick end labeling (TUNEL) assays were conducted as previously described [31]. The fibrin gel was composed of 25 mg/mL fibrinogen, 20 mM CaCl2, and 2.5 U/mL thrombin. The cells (1 × 106) were mixed with 80 μL of fibrin gel and subcutaneously transplanted into the dorsum of nude mice under deep anesthesia. After 24, 48 and 72 h, the constructs were removed and immediately fixed with 4% paraformaldehyde for paraffin embedding. Subsequently, 5-μm-thick sections were cut and subjected to TUNEL assay using an In Situ Cell Death Detection Kit (KeyGEN, Nanjing, Jiangsu, China) to measure the ASC death. The number of TUNEL+ cells was analyzed using Image-Pro Plus. Animal studies were conducted according to the protocol approved by the Ethics Committee of the State Key Laboratory of Oral Diseases, West China School of Stomatology, Sichuan University, China.

Statistics

Data were analyzed with GraphPad Prism 5.02 (GraphPad Software, San Diego, CA, USA) and are expressed as the mean ± standard deviation. Unpaired Student’s t tests were performed, and statistical significance was considered at P < 0.05. At least three replicates were analyzed in each experiment.

Results

Identification of P-ASCs and H-ASCs

Flow cytometric analysis indicated that the P-ASCs (physioxia ASCs) and H-ASCs (hyperoxia ASCs) were positive for CD73, CD90, and CD105 and negative for CD31, CD34, and HLA-DR (Fig. 1a). Both the P-ASCs and H-ASCs exhibited a typical spindle-shaped morphology (Fig. 1b) and adipogenic ability (Fig. 1c and d). Compared with the H-ASCs, the P-ASCs exhibited upregulated HIF-1 protein expression, as determined by Western blotting (Fig. 1e).

Fig. 1

Characterization of P-ASCs and H-ASCs. Human ASCs were cultured under physioxia (2% O2, P-ASCs) or hyperoxia (20% O2, H-ASCs) for the entire in vitro period until assayed at passage 3. a Flow cytometry was applied to demonstrate the immunophenotype of P-ASCs and H-ASCs. b Morphology of ASCs cultured in α-modified Eagle’s medium (α-MEM) with 10% fetal bovine serum (FBS). c After inducing adipogenesis for 7 days, lipid clusters were stained by oil red O. d Oil red O staining was quantified by the ratio of oil red O+ area to total image area for three fields. Data are presented as the mean ± SD, *P > 0.05. e Western blot of hypoxia-inducible factor 1 (HIF-1) and β-actin (reference gene) expression. Scale bar = 50 μm. ASCs adipose-derived stem cells, H-ASCs hyperoxia ASCs, P-ASCs physioxia ASCs

Physioxia enhanced ASC proliferation and migration through ROS upregulation

Using WST-8 and cell doubling curves, P-ASCs exhibited increased proliferation (Fig. 2a) accompanied by an increased ROS level (Fig. 2b and d). After ROS inhibition in P-ASCs by BHA (Fig. 2b, d), the enhanced P-ASC proliferation was decreased (Fig. 2c). Similarly, the Transwell assay (Fig. 2e, f) revealed reduced migration in H-ASCs and P-ASCs (BHA).

Fig. 2

Physioxia enhanced ASC proliferation and migration through ROS upregulation. a The proliferation of P-ASCs and H-ASCs measured by WST-8 and cell doubling curves. b and d P-ASCs were treated with 100 μM BHA to inhibit ROS, as detected by flow cytometry. The relative MFI was quantified by the ratio of the MFI for P-ASCs and P-ASCs (BHA) to that of H-ASCs. c The proliferation of P-ASCs, H-ASCs and P-ASCs (BHA) measured by WST-8 and cell doubling curves. e Transwell assays were used for determining cell migration, and the migrated cells were stained by 0.1% crystal violet. f The crystal violet in migrated cells was extracted by 10% acetic acid, and the optical density values were determined. The cell doubling curve was produced by dividing the cell number by 104 and then transforming the values to log2. Data are presented as the mean ± SD, *P < 0.05, **P < 0.01, Student’s t tests, scale bar = 100 μm. ASCs adipose-derived stem cells, BHA butylated hydroxyanisole, H-ASCs hyperoxia ASCs, MFI mean fluorescence intensity, P-ASCs physioxia ASCs, ROS reactive oxygen species

Physioxia inhibited ASC senescence and apoptosis

SA-β-Gal staining revealed that physioxia inhibited ASC senescence (Fig. 3a), with a significant difference in the SA-β-Gal+ area (1.53 ± 0.22% vs. 6.50 ± 0.40%, P < 0.01, Fig. 3b). Cell viability was significantly increased under physioxia compared with hyperoxia (95.27 ± 0.50% vs. 91.33 ± 0.85%, P < 0.05, Fig. 3c, d).

Fig. 3

Physioxia inhibited ASC senescence and apoptosis. a Microscopy images of senescent cells shown by SA-β-Gal staining. b SA-β-Gal staining results were quantified by the ratio of SA-β-Gal+ area to the total image area for three fields. c Cell apoptosis measured by flow cytometry using annexin V-FITC/PI double staining. Q1-UL, mechanical error; Q1-UR, late apoptotic or necrotic cells; Q1-LL, viable cells; Q1-LR, early apoptotic cells. d The ratio of viable cells acquired from Q1-LL. Data are presented as the mean ± SD, * P < 0.05, ** P < 0.01, Student’s t tests, scale bar = 20 μm. ASCs adipose-derived stem cells, H-ASCs hyperoxia ASCs, P-ASCs physioxia ASCs, SA-β-Gal senescence-associated β-galactosidase

Angiogenic activities of ASCs were promoted under physioxia

Tube formation induced by Matrigel was employed to examine the angiogenic activities of the cells. The P-ASCs generated more meshes than the H-ASCs (Fig. 4a), and statistical analysis revealed significantly increased total mesh (Fig. 4b), branching length (Fig. 4c) and junction (Fig. 4d) values for P-ASCs than for H-ASCs (2.20-, 1.29-, and 1.41-fold greater, respectively). RT-PCR showed increased expression of the angiogenic genes vascular endothelial growth factor (VEGF), vascular endothelial growth factor receptor 2 (VEGF-R2) and von Willebrand factor (vWF) (Fig. 4e) in P-ASCs.

Fig. 4

Physioxia promoted angiogenic ability of ASCs. ASCs (2 × 104) were seeded onto 96-well plates coated with 50 μL of Matrigel and cultured for 6 h. a Mesh-like structures resulting from tube formation assay. b, c and d Total mesh, branching length, and junction values per field of view were quantified by ImageJ. Five fields were quantified. e Expression levels of mRNA encoding VEGF, VEGFR2, and vWF as measured by qRT-PCR. Data are presented as the mean ± SD, *P < 0.05 (P-ASCs/H-ASCs), **P < 0.01 (P-ASCs/H-ASCs), Student’s t tests, n = 3, scale bar = 100 μm. ASCs adipose-derived stem cells, H-ASCs hyperoxia ASCs, P-ASCs physioxia ASCs, qRT-PCR quantitative real-time polymerase chain reaction, VEGF vascular endothelial growth factor, VEGFR2 vascular endothelial growth factor receptor 2, vWF von Willebrand factor

Survival of P-ASCs was strengthened under ischemic condition

After incubation in an ischemic environment (Fig. 5a) for 24 h, P-ASCs showed increased survival (Fig. 5B) and decreased death rates (Fig. 5A). A minor but significant difference was also detected under the hypoxic (Fig. 5b), acidic (Fig. 5c), and nutrient-depleted conditions (Fig. 5d).

Fig. 5

Physioxia improved ASC survivability under ischemic conditions. ASCs (1 × 104) were seeded onto 96-well plates and incubated in four hostile environments for 24 h: (a) ischemic model, 1% O2, pH 6.4 and 0.56 μM glucose; (b) hypoxic model, 1% O2, pH 7.4 and 5.6 μM glucose; (c) acidic model, 20% O2, pH 6.4 and 5.6 μM glucose; (d) nutrient-depleted model, 20% O2, pH 7.4 and 0.56 μM glucose. (A) Fluorescent images showing the cell death rate by live/dead cell staining. The cell death rate was obtained by the ratio of dead cells to total cells. Three fields were quantified. (B) The cell survival rate was detected by WST-8 presented as the ratio of OD24 to OD0. Data are presented as the mean ± SD, *P < 0.05, **P < 0.01, Student’s t tests, scale bar = 200 μm. ASCs adipose-derived stem cells, H-ASCs hyperoxia ASCs, OD0, optical density value at 0 h, OD24, optical density value at 24 h, P-ASCs, physioxia ASCs

Variations in mitochondrial and pH metabolism of ASCs under physioxia

By NAO staining, we measured a 43% decrease in the mitochondrial mass of P-ASCs (Fig. 6a, b), and the extracellular lactate concentration was much higher compared with that of H-ASCs (7.07 ± 0.54 vs. 4.60 ± 0.16, P < 0.05, Fig. 6d). Underlying these changes was the apparently upregulated mRNA expression of BCL2/adenovirus E1B 19 kDa protein-interacting protein 3 (BNIP3), cytochrome c oxidase subunit 4 isoform 2 (COX4I2), pyruvate dehydrogenase kinase 1 (PDK1) and lactate dehydrogenase A (LDHA), as detected by RT-PCR (Fig. 6c).

Fig. 6

Variations in mitochondrial and pH metabolism of P-ASCs. a and b Fluorescent images and flow cytometry results of mitochondrial mass determined by staining ASCs with NAO; the relative MFI was determined as the MFI of P-ASCs versus that of H-ASCs. c Expression of HIF-1 target genes evaluated by qRT-PCR. d Extracellular lactate concentration of P-ASCs and H-ASCs. e Cells cultured in the acidic model for 24 h were stained with BCECF-AM to determine the intracellular pH. Data are presented as the mean ± SD, *P < 0.05 (P-ASCs/H-ASCs), **P < 0.01 (P-ASCs/H-ASCs), Student’s t tests, n = 3, scale bar = 100 μm. ASCs adipose-derived stem cells, BCECF-AM 2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein, acetoxymethyl ester, BNIP3 BCL2/adenovirus E1B 19 kDa protein-interacting protein 3, mitophagy regulator, COX4I1 cytochrome c oxidase subunit 4 isoform 2, metabolic enzyme, COX4I2 cytochrome c oxidase subunit 4 isoform 2, metabolic enzyme, H-ASCs hyperoxia ASCs, LDHA lactate dehydrogenase A, glycolysis, MCT4 monocarboxylate transporter 4, lactate discharge, NAO nonyl acridine orange, NHE2 sodium-hydrogen exchanger 2, H+ discharge, NHE3 sodium-hydrogen exchanger 3, H+ discharge, P-ASCs physioxia ASCs, PDK1 pyruvate dehydrogenase kinase 1, inactivating pyruvate dehydrogenase

Cells were treated under acidic conditions (pH 6.4) for 24 h, and distinct alkalization in H-ASCs was determined by intracellular pH detection (7.48 ± 0.15 vs. 6.61 ± 0.17, P < 0.05, Fig. 6e). Additionally, the transcript levels of sodium-hydrogen exchangers (NHE2 and NHE3), carbonic anhydrase 9 (CAR9) and monocarboxylate transporter 4 (MCT4) were increased.

Ascending glucose uptake and reserve in P-ASCs

P-ASCs showed significantly increased glucose uptake, as measured by 2NBDG staining (1.20-fold greater, Fig. 7a, b), along with augmented mRNA levels of glucose transporters (GLUT1 and GLUT3), as demonstrated by RT-PCR (Fig. 7d). Increased glycogen reserves were found in P-ASCs, as detected by PAS staining (Fig. 7c), and the expression of glycogen synthesis (phosphoglucomutase [PGM] and glycogen synthase 1 [GYS1]) and breakdown genes (liver isoform of glycogen phosphorylase [PYGL]) were also upregulated (Fig. 7d).

Fig. 7

Increased glucose uptake and reserve of P-ASCs. a and b Fluorescent images and flow cytometry results of glucose uptake in ASCs determined by staining with 2-NBDG; the relative MFI was determined by MFI of P-ASCs versus that of H-ASCs. Scale bar = 100 μm. c Intracellular glycogen detected by PAS staining; the PAS+ area was calculated as the PAS+ area versus the total image area. Three fields were quantified. Scale bar = 50 μm. d Expression of HIF-1 target genes evaluated by qRT-PCR. Data are presented as the mean ± SD, *P < 0.05 (P-ASCs/H-ASCs), **P < 0.01 (P-ASCs/H-ASCs), Student’s t tests, n = 3. 2-NBDG 2-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl) amino]-2-deoxy-D-glucose, ASCs adipose-derived stem cells, GLUT1 glucose transporter 1, glucose uptake, GLUT3 glucose transporter 3, glucose uptake, GYS1 glycogen synthase 1, glycogen synthesis, H-ASCs hyperoxia ASCs. PAS periodic acid-Schiff, P-ASCs physioxia ASCs, PGM phosphoglucomutase, glycogen synthesis, PYGL liver isoform of glycogen phosphorylase, glycogen breakdown

Increased survivability of P-ASCs in vivo

The number of dead cells 24, 48, and 72 h after implantation with fibrin gel was detected by TUNEL assay (Fig. 8a); compared to H-ASCs (47.46 ± 8.58%, 57.35 ± 7.41% and 63.70 ± 3.32%), P-ASCs (18.04 ± 3.13%, 27.56 ± 2.20% and 27.62 ± 5.13%) showed a significantly lower death rate (Fig. 8b).

Fig. 8

Physioxia increased ASC survivability in vivo. After mixing with 80 μL of fibrin gel, 1 × 106 P-ASCs or H-ASCs were subcutaneously transplanted into the dorsum of nude mice. The implants were extracted after 24, 48, and 72 h. a TUNEL assay was used to stain the nucleus of dead cells. The black arrows indicate dead cells. b The TUNEL+ cell rate was determined by the ratio of TUNEL+ cells versus total cells. Three fields were quantified. Data are presented as the mean ± SD, *P < 0.05, **P < 0.01, Student’s t tests, scale bar = 100 μm. ASCs adipose-derived stem cells, H-ASCs hyperoxia ASCs, P-ASCs physioxia ASCs, TUNEL TdT-mediated dUTP-biotin nick end labeling

Discussion

Offering safe and effective cell therapy products for clinical applications is consistent with good manufacturing practice (GMP) guidelines, which should be followed during the entire process of isolating, expanding and transplanting ASCs [32]. The present study compared ASCs cultured under hyperoxia (20% O2) and physioxia (2% O2, oxygen concentration in situ) and provides compelling evidence that the latter could be a more effective approach owing to the advantages of retaining cell proliferation, migration, survival in ischemia and angiogenesis, and suppressing senescence and apoptosis.

There are no differences between P-ASCs and H-ASCs in terms of immunophenotype, morphology or adipogenesis, and a previous study [25] revealed that culturing ASCs under physioxia does not increase the risk of tumourigenesis associated with ASCs, indicating that P-ASCs are safe for clinical therapy. Physioxia promoted cell proliferation and migration, and many studies have attributed this effect to the stabilization of HIF-1 in the lack of O2 [33, 34]. However, the ROS level was also increased in P-ASCs, suggesting that transient physioxia can restore proliferation and migration through the augmentation of ROS [35]. Furthermore, we showed that without the injury caused by hyperoxia, physioxia is an appropriate condition for maintaining ASC proliferation and migration.

The relationship between physioxia and ROS is complicated [36]. In principle, HIF-1 decreases the ROS level [37, 38], which should be lower in P-ASCs, but the results show the opposite effect. The underlying mechanism remains unknown, especially in stem cells.

Many studies have shown the ability of HIF-1 to enhance angiogenesis under transient physioxia [39, 40], but consistent with most studies on physioxia and ASCs, the cells were isolated from a physioxic niche and then cultured under atmospheric hyperoxia, which could injure the bioactivity of the cells. Thus, the discrepancy of such bioactivity between P-ASCs and H-ASCs is not due to the acceleration of physioxia but reflects the damage caused by hyperoxia. Although transient physioxia preconditioning would be applied prior to transplantation for recovery, in the present method, culturing cells under physioxia through the entire in vitro period may be a better approach; however, further research is required.

To acquire an excellent stem cell product, cell viability should also be considered. Physioxia evidently suppressed senescence and apoptosis under nonstressful condition. The required survival of cells implanted in an ischemic environment composed of low oxygen, glucose, and pH levels is a main barrier for cell therapy [41]. Thus, we established an ischemic model and observed increased adaptability in P-ASCs; the same effect was observed in hypoxic, acidic, and nutrient-depleted environments, explaining the superiority of P-ASCs under these conditions and resulting in preferable adaptability in an ischemic environment.

The underlying mechanisms induced by HIF-1 and described in a previous study using an HIF-1 activator [31] were also observed in P-ASCs, but with an inverse trend in the ROS level. Briefly, more efficient aerobic oxidation (switch of cytochrome c oxidase subunit COX4I1 to COX4I2) and a switch to glycolysis (declined mitochondrial mass (Fig. 6a and b) caused by BNIP3 and increased glycolysis by PDK1 and LDHA) indicate adaptability to hypoxia (Fig. 5b). Additionally, enhanced glucose uptake (GLUT1 and GLUT3 (Fig. 7a and b)), glycogen synthesis (PGM and GYS1 (Fig. 7c)), and glycogen breakdown (PYGL) demonstrated cell adaptation to nutrient depletion (Fig. 5d), while an alkalescent intracellular pH (Fig. 6e) (CAR9, NHE2 and NHE3 [export H+] and MCT4 [export lactate] (Fig. 6d)) indicated adaptability to acidic conditions (Fig. 5c).

This study shows for the first time that culturing ASCs under physioxia for the entire in vitro term could induce metabolic alterations and improve ASC survival in ischemic environment. Our observations illustrate the molecular, cellular, and in vivo biological effects induced by physioxia in ASCs, presenting a significant mechanistic basis for culturing ASCs under physioxia for cell therapy. However, longer culture periods should be examined to guarantee the security of cell properties under this condition. Moreover, specific cell therapy models should be constructed to verify the ultimate efficacy of the cells, including when applied in adipose regeneration, heart failure treatment, and wound healing.

Conclusions

In summary, the present results suggest that culturing ASCs under physioxia (2% O2) for the entire in vitro period, not under conventional hyperoxia (20% O2), could be a more effective approach for cell therapy applications owing to the improvements in proliferation, migration, survival and angiogenesis, and suppression of senescence and apoptosis.