Hypoxia induces adipocyte differentiation of adipose-derived stem cells by triggering reactive oxygen species generation

Materials

N-Acetyl-L-cysteine (NAC), dibenziodolium chloride (DPI), N-nitro-L-arginine methyl ester hydrochloride (NAME), triphenylphosphine (TPP), antimycin and rotenone were obtained from Sigma (St. Louis, MO). S-Nitroso-N-acetyl-D,L-penicillamine (SNAP) and S-nitroso-L-glutathione (GSNO) were purchased from Cayman Chemical (Ann Arber, MI).

Synthesis of mito-carboxy proxyl

Mito-carboxy proxyl (CP) was synthesized by following the procedure of Kalyanaraman (Dhanasekaran et al., 2005) and Boelsterli (Lim et al., 2008).

Cell culture and inhibition study

Sampling of human subcutaneous adipose tissue and isolation of ASCs was done as before (Kim et al., 2007, 2009a). ASCs were characterized by trans-differentiation and analysis of cell surface markers by flow cytometry. ASCs were grown in Minimum Essential Medium-alpha (Hyclone, Thermo Scientific, Logan, UT) with 10% fetal bovine serum (GIBCO), 1% penicillin and streptomycin (GIBCO) at 37°C in a humidified air plus 5% CO₂ (normoxia), or 5% CO₂, 2% O₂ and balanced N₂ (hypoxia). Antimycin (10 nM), rotenone (10 pM), SNAP (10 nM) and GSNO (10 nM) were used in the functional study. For the inhibition of ROS generation, 0.1 or 1 mM NAC, 1 μM DPI, 20 mM Nox4 siRNA transfection, and 2–10 μM mito-CP were used. For the inhibition of RNS generation, 0.1 mM NAME were used under hypoxia.

ROS or RNS generation assay

ROS generation was measured using 2′,7′-dichlorofluorescin diacetate probe (DCF-DA, Molecular Probes, Eugene, OR), and RNS generation was measured using diaminofluorescein-2diacetate (DAF-DA, Sigma). ASCs (4 × 10⁵ cells) were seeded into 60 mm culture dishes. They were pretreated with DCF-DA or DAF-DA (10 μM) for 20 min and incubated under hypoxia for 20 min at 37°C in the dark place before being harvested using trypsin-EDTA. Fluorescence intensity of DAF-DA or DAF-DA was measured and quantified by flow cytometry.

Adipocyte differentiation

ASCs (4 × 10⁴) were seeded in 12-well plates in complete medium. When they had become confluent, the medium was replaced with preadipocyte growth medium-2 (PGM-2, Lonza, Walkersville, MD) for 4–6 days. They were then treated with hypoxia, ROS generators, or RNS generators in preadipocyte differentiation medium-2 (PDM-2, Lonza) for 4–7 days (Figure 1A). Adipocyte differentiation was confirmed by using oil-red-O staining and RT-PCR.

Oil red-O staining

ASCs were washed with PBS and fixed with 10% formalin for 1 h. After fixing, cells were washed with PBS and stained with oil red-O solution for 2 h. This solution was made of 60% oil red-O stock solution (1% oil red-O powder in isopropanol) in distilled water. The stained cells were washed twice with distilled water and examined microscopically. For the quantification of oil red-O, cells were incubated with isopropanol for 30 min, and the color in the supernatant of the decolorized cells was measured by the absorbance at 492 nm with a micro-plate reader (TECAN).

RT-PCR

Total cellular RNA was extracted with TRIzol reagent (Invitrogen, Carlsbad, CA) followed by a reverse transcription with cDNA synthesis kit (Promega, Madison, WI). cDNA was synthesized from 1 µg total RNA using 200 U of reverse transcriptase (M-MLV RT) and 50 ng/μl oligo(dT). Primers were used as follows; PPARγ, forward: 5′-GCGGAGATCTCCAGTGATATC-3′ and reverse: 5′-TCAGCGACTGGGACTTTTCT-3′, C/EBPβ, forward: 5′-GAATCTCCTAGTCCTGGCTC-3′ and reverse: 5′-GATGAGAACAGAACGAGTAC-3′, and GAPDH, forward: 5′-CGAGATCCCTCCAAAATCAA-3′ and reverse: 5′-TGTGGTCATGAGTCCTTCCA-3′. PCRs were performed in a final volume of 25 µl reaction mixture that contained 2 µl of the RT reaction mixture, 15 mM MgCl₂, 1.25 mM dNTP, 20 pM of each primer and 0.5 U of Taq polymerase (Promega). Thermal cycling over 40 cycles consisted of an initial denaturation at 94°C for 5 min, 94°C for 30 s, 56°C for 30 s, 72°C for 30 s, terminated by a final extension at 72°C for 5 min. GAPDH mRNA level was used for sample standardization. After electrophoresis on 1.5% agarose gel, each band was quantified by densitometer (Bio-Rad Laboratories, Hercules, CA).

Phosphokinase chip array

For phosphokinase chip array, ASCs were seeded into serum-free medium in 100 mm dishes. The following day, ASCs were incubated in hypoxia for 10 min. Cells were harvested by lysis buffer (R&D Systems) and proteins were isolated. Phosphorylation of signaling molecules was detected with a phosphokinase array kit (R&D Systems).

Statistical analysis

Data are representative of triplicate independent experiments. The statistical significance of the differences among groups was tested using student's t-test. P < 0.01 was considered to be significant.

Hypoxia at 2% oxygen increased adipocyte differentiation

In a pilot study, ASCs were fully differentiated into adipocytes 2–3 weeks after addition of adipocyte differentiation medium. However, we usually measured the early induction of adipocyte differentiation at 4 and 7. Compared with normoxia and 5% hypoxia, 2% hypoxia significantly increased lipid accumulation of ASCs (Figure 1B). mRNA expression of adipocyte differentiation markers, such as peroxisome proliferator-activated receptor-γ (PPARγ) and CCAAT/enhancer-binding protein-β (C/EBPβ), were upregulated in 2% hypoxia (data not shown).

ROS, but not RNS, generation induced adipocyte differentiation

Flow cytometry was used to measure the free radicals generated by hypoxia. The geometric mean of DCF-DA, a ROS indicator, significantly increased under hypoxia, while that of DAF-DA, a RNS indicator, was unaltered (Figure 2A, P < 0.01). By scavenging every free radical using NAC (a ROS scavenger) and NAME (a RNS scavenger), the former, but not NAME, significantly attenuated hypoxia-induced lipid accumulation (Figure 2B, P < 0.01). Likewise, hypoxia-induced mRNA expression of PPAR-γ and C/EBP-β was not downregulated by NAME, but by NAC treatment (Figure 2C). ROS generators such as antimycin (10 nM) and rotenone (10 pM) induced lipid accumulation (Figure 2D) and upregulated mRNA expression of PPAR-γ and C/EBP-β expression (Figure 2E), but RNS generators such as SNAP (10 nM) and GSNO (10 nM) did not. This suggests that hypoxia increases adipocyte differentiation of ASCs mainly via ROS generation.

Nox4-produced ROS does not induce adipocyte differentiation

We previously demonstrated that Nox4-produced ROS results in an increase in the proliferation and migration of ASCs. Here, we investigated whether or not Nox4 can induce an increase in adipocyte differentiation. However, a pharmacological inhibitor of Nox4 (DPI, 100 nM) or siRNA transfection of Nox4 did not reduce lipid accumulation (Supplemental Figure 1A and 1C, respectively), nor downregulation of PPAR-γ and C/EBP-β expression (Supplemental Figure 1B and 1D, respectively), thus indicating that Nox-induced ROS generation does not mediate adipocyte differentiation.

Hypoxia increases mtROS generation

Since mitochondrial respiration is major ROS source, we investigated whether mtROS generation under hypoxia induces adipocyte differentiation. mtROS generation was assessed by measuring the fluorescence intensity of mito-SOX in a flow cytometry. Hypoxia significantly increased the fluorescence intensity of mito-SOX in ASCs (Figure 3A).

mtROS generation induced adipocyte differentiation

Mito-CP (Figure 3B), a mitochondria specific ROS scavenger, was synthesized to investigate the involvement of mtROS generation in adipocyte differentiation. Compared to TPP treatment (as negative control), scavenging mtROS with mito-CP significantly reduced hypoxia-induced lipid accumulation in a dose-dependent manner (Figure 3C, P < 0.01).

Antimycin and rotenone induce ROS generation by inhibiting mitochondrial respiratory complex III and complex I, respectively (Li et al., 2003; Rieske et al., 1967). We therefore examined if these mtROS generators induce adipocyte differentiation. Different concentrations of antimycin (Figure 3D, P < 0.01) and rotenone (Figure 3E, P < 0.01) significantly induced lipid accumulation in a concentration dependent manner. Antimycin (Figure 3F) and rotenone (Figure 3G) also induced the PPAR-γ and C/EBP-β expression in ASCs.

Akt and mTOR phosphorylation are involved in adipocyte differentiation

Phosphokinase antibody array was carried out to investigate the signaling pathways activated by hypoxia (Figure 4A). Phosphorylation of ERK1/2, Akt, and mTOR was significantly increased by hypoxia (P < 0.01), which had been previously demonstrated by Western blot analysis (Kim et al., 2011). To see whether the PI3K/Akt/mTOR and MAPK pathways are involved in adipocyte differentiation, we found that the inhibitors of PI3K/Akt (LY294002, Figure 4B) and mTOR (rapamycin, Figure 4C) significantly reduced lipid accumulation in ASCs. LY294002 (Figure 4D) and rapamycin (Figure 4E) reduced the PPAR-γ and C/EBP-β expression in ASCs. However, U0126, an inhibitor of the MAPK pathway, did not reduce the adipocyte differentiation (Supplemental Fig. 2). The findings indicate that PI3K/Akt/mTOR pathway plays a pivotal regulatory role in adipocyte differentiation.