ADSCs were isolated from subcutaneous adipose tissues of the abdomen of patients undergoing caesarean section in Good Moonhwa Hospital, Busan, Korea. Informed consent was obtained from patients and all procedures were approved from hospital ethic committee and Institutional Review Board (2013-02). Tissues were washed three or four times with phosphate buffered saline (PBS) and digested with same volume of PBS supplemented 0.2% collagenase type I (Gibco BRL, Gaithersburg, MD, USA) for 60 minutes at 37℃ in shaking incubator. And then the tissues were filtered through the 100 µm nylon cell strainer (BD Biosciences Pharmingen, San Diego, CA, USA) and then centrifuged at 1000 rpm for 8 minutes to obtain stromal vascular fraction (SVF). The SVF pellets were resuspended and cultured at 37℃, 5% CO₂ in growth media (GM) (Dulbecco's modified Eagle's medium [DMEM] low glucose supplemented with 10% [v/v] fetal bovine serum [FBS; Hyclone, Logan, UT, USA] and 1% [v/v] 100 units/mL of penicillin, and 100 ng/mL streptomycin [Gibco BRL]). This initial culture was referred as passage 0. Fresh media were replaced every third day. When monolayer of adherent cells reached 75% to 90% confluence, the cells were trypsinized (0.25% trypsin-ethylenediaminetetraacetic acid [EDTA]; Gibco BRL) and sub-cultured to passage 4 to 5.

An analysis of cell surface molecule was executed on passage 4 cultures of human ADSCs using flow cytometry. Briefly, after the media were removed from dish, the cell layer was washed with dPBS and treated with 0.25% trypsin-EDTA to detach cell. ADSCs were harvested by centrifugation and washed with BD PharmigenTM stain buffer (FBS; BD Biosciences). Then, the cells were incubated with fluorescein isothiocyanate (FITC) mouse-anti human CD90, FITC mouse-anti human CD105, FITC mouse anti-human CD73, FITC MSC negative cocktail (FITC mouse-anti human CD31, FITC mouse-anti human CD34, FITC mouse-anti human CD35) and FITC mouse IgG1, Isotype Control (BD Biosciences) for 50 minutes, after which the cells washed with BD PharmagenTM stain buffer (FBS) three times. And then the cells were analyzed on FC500 flow cytometer (Beckman Coulter GmbH, Krefeld, Germany).

For the differentiation into osteoblasts and adipocytes, ADSCs were cultured with an osteogenic induction media (OIM; GM supplemented with 100 nM dexamethasone, 50 µM ascorbate acid, and 10 mM β-glycerophosphate sodium [Sigma-Aldrich, St. Louis, MO, USA]) and an adipogenic induction media (AIM; GM supplemented with 1 µM dexamethasone, 1 mM 3-isobutyl-1-methylxanthine, 10 ng/mL insulin and 60 µM indomethacin [Sigma]), respectively. In all experiments, media were changed every three days.

GLP-1 (PeproTech, Rocky Hill, NJ, USA) was treated to the ADSCs at a concentration of 10 nM and 100 nM. For the study concerning the effects of protein kinase inhibitors, cells were pretreated with 20 µM PD98059 (Gibco BRL) for one hour, followed by treatment with GLP-1 in the presence of the inhibitor.

After ADSCs were grown to confluence in GM, the cells were induced in OIM for 21 days. And then, the cells were estimated using Alizarin red S staining, which indicates the extracellular matrix calcification. Briefly, the media were replaced and then ADSCs were fixed with 70% iced ethanol for one hour at room temperature. After washing in distilled water twice, the cells were stained with 40 mM Alizarin Red S (Sigma, pH 4.1-4.6) for 5 minutes at room temperature. Alizarin Red S solution was removed and the cells were rinced twice with distilled water. Images of cells stained with Alizarin Red S were obtained with a G12 digital camera (Cannon, Tokyo, Japan) or were captured by an optical microscope with a G12 digital camera. Matrix mineralization was quantified by eluting Alizarin Red S staining with 20% methanol, 10% acetic acid and 70% distilled water at room temperature for 10 minutes. The absorbance of the supernatants was measured at 405 nm using a Thermo Multiskan FC Microplate Reader (Scientific).

After ADSCs were cultured in AIM for two weeks, the cells were fixed with 10% formalin for one hour at room temperature. Fixed cells were washed in 60% isopropanol and stained for 10 minutes with Oil Red O (Sigma) at room temperature. After five consecutive washes in distilled water, the presence of red stained lipid droplets were documented using bright-field microscopy. Images of cells stained with Oil Red O were acquired with a G12 digital camera. In order to elute stained lipid droplet, 100% isopropanol was added to each well followed by 10 minutes at room temperature and quantified by measuring the optical absorbance at 500 nm using a Thermo Multiskan FC Microplate Reader (Scientific).

Total RNA was extracted using a Trisol reagent® (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. Complementary DNA (cDNA) was synthesized from 5 µg of total RNA with Moloney murine-leukaemia virus (MMLV) Reverse Transcriptase (Promega, Madison, WI, USA) using a random hexamer (Bioneer, Daejeon, Korea) at 42℃ for one hour. Template cDNA was subjected to PCR amplification using gene-specific sense and antisense primers (Table 1). RT-PCR conditions were denatured at 95℃ for 5 minutes, followed by 28 to 35 cycles of 95℃ for 30 seconds, annealed at 57℃ for 30 seconds in a thermal cycle. The PCR products were visualized by electrophoresis on 1.5% agarose gel.

Real-time PCR was performed with the SYBR Green I Light cycle system (Roche, Mannheim, Germany). The reaction mixtures were prepared using Light Cycle Fast DNA master mixture for SYBR Green I, 0.5 µM of each primer, 4 mM MgCl₂ and 2 µL of cDNA in a final volume of 20 µL. The reaction condition consisted of denaturation at 95℃ for 10 minutes, followed by 40 cycles of 95℃ for 10 seconds and 60℃ for 10 seconds, followed by melting curve analysis. For each sample, PCR were performed in duplicate. The quantitative amount of each gene was normalized against the house-keeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH).

For cellular detection of GLP-1R, the cells were fixed with 4% paraformaldehyde in PBS. Rabbit anti-GLP-1R antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA) was used, with diluted 1 : 100 in blocking solution. Fluorescence detection was performed by incubation with FITC conjugated goat anti-rabbit antibody (Santa Cruz Biotechnology) with diluted 1 : 400 in blocking solution for GLP-1R detection. Nuclei staining and mounting were performed with ProLong^@Gold antifade reagent with DAPI (Invitrogen, Oregon, USA).

At day 10 after passage, ALP histochemistry was performed using diagnostic kit 85 (Sigma). ALP activity was determined using a colorimetric end point assay measuring the enzyme conversion of p-nitrophenyl phosphate (pNPP) to the yellowish product, p-nitrophenol (pNP), in the presence of ALP.1 In brief, cells were rinsed with cold PBS, scraped into 1.5 ml Eppendorf tubes in 1ml of lysis buffer (10 mM Tris-HCl, 1 mM magnesium chloride, and 0.05% Triton X-100, pH7.5), and then homogenized on ice for 10 minutes. Afterwards, the resulting mixture was centrifuged at 12,000 rpm for 10 minutes at 4℃. The cell lysates (0.1 mL of the above supernatants) were mixed with 0.5 mL of pNP phosphate solution (Sigma) and 0.5 mL of alkaline buffer solution (Sigma). After incubation at 37℃ for 15 minutes, the above mixture was added to 1 mL of 0.5 N NaOH to stop the reaction and the absorbance was then measured spectrophotometrically at 405 nm. A standard curve of known concentration of pNP was generated concurrently and used to determine sample concentrations.

Statistical analysis was performed with the student t-test using SPSS (version 12.0; SPSS Inc., Chicago, IL, USA). A P < 0.05 was considered to be statistically significant.

ADSCs obtained from human adipose tissues exhibited a fibroblast-like cells morphology and adhered to plastic. From one day of ADSCs culture, attachment of spindle-shaped cells was observed. After five days, spindle-shaped cells reached about 80% confluency. Morphology of the cells changed gradually according to passage numbers.

In order to assess whether ADSCs have MSC-specific characteristics, flow cytometry was performed at passage 4. The cells were positively stained for MSC-specific surface markers (CD73, CD90 and CD105), but they were negative for markers of the endothelial cells (CD31) and hematopoietic cells (CD34 and CD45) (Fig. 1A).

To test the multi-lineage potential of ADSCs, the cells were induced in adipogenic or osteogenic differentiation media. The adipogenic and osteogenic differentiation potential was determined by each differentiation specific gene expression and staining. Seven days after adipogenic or osteogenic induction, specific osteoblast markers, ALP and runt-related transcription factor 2 (RUNX2) were positively expressed (Fig. 1B). The expression of adipogenic-specific genes, PPAR gamma (PPAR-γ), adipocyte protein 2 (AP2) and lipoprotein lipase (LPL) were also observed (Fig. 1D). Fourteen days after adipogenic or osteogenic induction, ADSCs revealed Alizarin Red S positive staining for calcium deposition (Fig. 1C) and Oil Red O positive staining for the lipid droplet (Fig. 1E). These results mean that the isolated ADSCs have the potential of MSCs.

Prior to the examination of the effect of GLP-1 on adipogenic and osteogenic differentiation, whether GLP-1R is expressed in ADSCs was tested by immunofluorescence staining using anti-GLP-1R antibody. Fig. 2 showed GLP-1R positive stained cells.

To examine whether GLP-1 has an effect on osteoblast differentiation, two doses of GLP-1 (10 and 100 nM) were treated to ADSCs cultured in OIM, and then Alizarin Red S staining and ALP staining were assessed after 10 days and 21 days of culture, respectively as the first indicator for osteogenic differentiation. The areas of ALP positive staining (Fig. 3A) and calcified nodules by Alizarin Red S staining (Fig. 3B) were increased in a dose-dependent manner in response to GLP-1, and the significant maximal effect was reached at a concentration of 100 nM GLP-1. Quantitative analysis of calcium deposition showed two-fold and five-fold increase in calcium deposition in 10 nM and 100 nM GLP-1, respectively (Fig. 3C).

Next, to additionally demonstrate the effect of GLP-1 on osteoblast differentiation, 100 nM GLP-1 was treated to ADSCs cultured in OIM, and the expression of ALP mRNA transcript and ALP activity were examined. ALP mRNA expression was increased with culture period and it was significantly increased in GLP-1 treatment compared to the OIM on day 7 of culture (Fig. 3D). After 10 days of induction, the ALP activity also significantly increased in the 100 nM GLP-1-treated cells compared to the OIM group (Fig. 3E). These data indicate that GLP-1 stimulates osteogenic differentiation of ADSCs.

To examine the effect of GLP-1 on adipogenic differentiation in ADSCs, ADSCs cultured in AIM were treated with two doses of GLP-1 (10 nM and 100 nM). And then Oil-Red O staining and adipogenic specific gene expression were determined as an indicator for adipogenic differentiation. After 14 days of culture, the lipid accumulation by Oil Red O staining and quantitative analysis of lipid-droplet formation was decreased in a dose-dependent manner in response to GLP-1 treatment. Especially, a concentration of 100 nM GLP-1 significantly decreased the lipid accumulation compared to AIM without GLP-1 (Fig. 4A, 4B). Next, when the ADSCs were treated with 100 nM GLP-1 on day 0, 3 and 7 of culture, the expressions of AP2, PPAR-γ and LPL were significantly down-regulated at day 7 compared to AIM group (P < 0.05) (Fig. 4C). These data indicate that adipogenic differentiation of ADSCs may be inhibited by GLP-1.

To understand the signaling pathway involved in the inhibitory effect of GLP-1 on the adipocyte differentiation of ADSCs, ADSCs cultured in AIM in the presence or absence of GLP-1 were treated with ERK inhibitor, PD98059. GLP-1 treatment decreased expression of PPAR-γ, LPL and AP2 mRNA levels and this decreased effect caused by GLP-1 was significantly reversed by the treatment of ERK inhibitor (P < 0.05) (Fig. 5).