Adipose tissues were obtained from 3 healthy donors undergoing tumescence liposuction (age range, 32-52 years; 2 males and 1 female) who underwent surgery at the Shanghai General Hospital of Nanjing Medical University between January, 2018 and April, 2018. Clinical and biochemical examinations confirmed that these subjects did not suffer from acute inflammation, cancers, endocrine diseases or infectious diseases. The hADSCs were isolated as previously described (16). hADSCs were cultured in basal MSC culture medium (bM) [Dulbecco's modified Eagle's medium (DMEM), 10% FCS penicillin/streptomycin, L-glutamine] (Lonza Group, Ltd.). The cells from the 3 donors were pooled together for all experiments and the experiments were performed 3 times. The present study was approved by the Ethics Committee of Shanghai General Hospital of Nanjing Medical University (IRB approval no. 2018-0032). Informed consent was obtained from all patients prior to study inclusion.

hADSCs were grown in basal MSC culture medium at 37°C for 2 days, followed by differentiation into osteoblasts, adipocytes and chondrocytes. For osteogenic differentiation, the culture medium was exchanged for osteogenic differentiation medium (odM) [DMEM, 100 nM dexamethasone, L-glutamine, 50 mM ascorbate, 1% penicillin/streptomycin, mesangial cell growth supplement (MCGS) and 10 mM β-lycerophosphate] (Lonza Group, Ltd.). The medium was changed every 3 days during the course of incubation (21 days) and the cells were harvested and analyzed at 7, 14 and 21 days. Cells cultured in bM medium served as a control. Alizarin Red S staining was performed to assess the occurrence of calcium phosphate deposition. Briefly, hADSCs were washed twice with PBS and fixed with 4% paraformaldehyde for 10 min. The fixed cells were stained with 500 µl of Alizarin Red solution (Sigma-Aldrich; Merck KGaA) for 5 min at room temperature. For adipogenic differentiation, the hADSCs were induced using the StemPro adipogenesis differentiation kit media (Gibco; Thermo Fisher Scientific, Inc.). The differentiation medium was changed every 3 days. After 21 days, Oil Red O staining was performed for the assessment of lipid/fatty acid droplets in adipocytes. Briefly, the fixed cells as above were stained with filtered 0.6 mg/ml Oil Red O solution for 30 min at room temperature. For chondrogenic differentiation, the hADSCs were induced using an MSC Chondrogenic Differentiation kit (Cyagen Bioscience, Inc.) according to the manufacturer's protocol. The differentiation medium was changed every 3 days. After 21 days, Alcian Blue staining was performed to assess the chondrogenic differentiation potential of hADSCs. Briefly, cells were rinsed in PBS and then fixed in methanol for 30 min, following Alcian blue staining for 30 min at room temperature. All these images were viewed by a microscopy at ×400 magnification.

The osteoblast phenotype was evaluated by determining ALP activity. Briefly, hADSCs were collected and lysed with RIPA buffer, and the total protein concentration was determined with Pierce BCA Protein Assay kit (Thermo Fisher Scientific). Relative ALP activity was normalized by the protein concentration. Then, the ALP activity of the hADSCs was detected with the Colorimetric Alkaline Phosphatase Assay kit (BioVision), according to the manufacturer's instructions.

miRNA was extracted from the hADSCs using an miRNeasy extraction kit (Qiagen, Inc.) according to the manufacturer's instructions. cDNA was synthesized using TaqMan™ MicroRNA Reverse Transcription kit (Applied Biosystems; Thermo Fisher Scientific, Inc.). To quantify the mRNA levels of runt-related transcription factor 2 (Runx2), osteopontin (OPN), osteocalcin (OCN) and bone sialoprotein (BSP), total RNA was isolated using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. cDNA was synthesized using the PrimeScript RT reagent kit (Takara). Quantitative PCR was performed using a standard protocol from the SYBR-Green PCR kit (Toyobo) on an ABI 7500 thermocycler (Applied Biosystems; Thermo Fisher Scientific, Inc.). All reactions were performed in triplicate or duplicate. The relative levels of miRNAs and mRNA in cells were normalized to U6 and GAPDH, respectively. 95°C for 10 min followed by 45 cycles consisting of 95°C for 15 sec, 60°C for 30 sec and 68°C for 60 sec were amplification conditions. The sequences of the primers were as follows: miR-29b forward, 5′-GGGGTAGCACCATTTGAA-3′ and reverse, 5′-TGC GTG TCG TGG AGTC-3′; U6 forward, 5′-GCT TCG GCA GCA CAT ATA CTA AAAT-3′ and reverse, 5′-CGC TTC ACG AAT TTG CGT GTCAT-3′; Runx2 forward, 5′-GTC TCA CTG CCT CTC ACTTG-3′ and reverse, 5′-CAC ACA TCT CCT CCC TTCTG-3′; OCN forward, 5′-ACA GAC AAG TCC CAC ACA GCAGC-3′ and reverse, 5′-TGA AGG CTT TGT CAG ACT CAG GGC-3′; BSP forward, 5′-GCC AGA GGA GCA ATC ACCAA-3′ and reverse, 5′-CAG GCT GGA GGT TCA CTGGT-3′; GAPDH forward, 5′-GTC TCA CTG CCT CTC ACTTG-3′ and reverse, 5′-GTG GTG AAG ACG CCA GTGGA-3′; OPN forward, 5′-TTG GCT TTG CAG TCT CCT GCGG-3′ and reverse, 5′-AGG CAA GGC CGA ACA GGC AAA-3′ and GAPDH forward, 5′-CGA GCC ACA TCG CTC AGACA-3′. The relative expression of each gene was calculated using the 2^−ΔΔCq method (17).

Total RNA was isolated using a miRNAeasy mini kit (Qiagen, Inc.), followed by labeling and hybridization with the miRCURYTM LNA Array (v.16.0, Exiqon), which contained capture probes targeting all miRNAs for human, mouse and rats registered in the Sanger miRBase 21.0 database (www.mirbase.org/). The procedure and imaging processes were as described previously (18).

Agomir-29b, antagomir-29b, phosphatase and tensin homolog (PTEN) overexpression plasmid and negative controls were all provided by RiboBio Co., Ltd. When the hADSCs in the 12-well plate grew to approximately 80% confluency, agomir-29b (0.1 nmol), antagomir-29b (0.1 nmol), or 2 µg pcDNA-PTEN were transfected into the cells at 37°C for 24 h, using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.).

TargetScan Release 7.0 (http://targetscan.org/), Miranda (http://miranda.org.uk) and PicTar (https://pictar.mdc-berlin.de) were used to search for the putative targets of miR-34a.pGL3-PTEN wild-type (Wt) or pGL3-PTEN mutant type (mut) were co-transfected with agomir-29b (0.1 nmol), antagomir-29b (0.1 nmol) into 293T cells (ATCC) in 24-well plates (2×10⁵/well) using Lipofectamine 2000 (Invitrogen). At 24 h post-transfection, the double luciferase activities were analyzed using the Dual-Luciferase Reporter Assay system (Promega Corp.). The pRL-TK plasmid (Promega Corp.) was used as a normalizing control.

Western blot analysis was performed as previously described (19). Briefly, the cells were lysed with RIPA buffer, and the total protein concentration was determined with Pierce BCA Protein Assay kit (Thermo Fisher Scientific). A total of 40 µg protein samples were separated by 12% SDS-PAGE, and then transferred onto polyvinylidene difluoride (Millipore) membranes at 200 mA for 1 h on ice. After blocking with 5% skim milk for 2 h at 4°C overnight, the membranes were incubated with the following primary antibodies: Anti-PTEN (cat. no. 9188, Cell Signaling Technology, Inc. 1:1,000), anti-Runx2 (cat. no. 12556, Cell Signaling Technology, Inc. 1:1,000), anti-OPN (cat. no. ab75285, Abcam, 1:1,000), anti-OCN (cat. no. ab13420, Abcam, 1:1,000), anti-BSP (cat. no. ab52128, Abcam, 1:1,000), anti-AKT (cat. no. 4691, Cell Signaling Technology, Inc. 1:1,000), phosphorylated-AKT (cat. no. 9611, Cell Signaling Technology, Inc. 1:1,000), glycogen synthase kinase (GSK)-3β (cat. no. 12456, Cell Signaling Technology, Inc. 1:1,000), phosphorylated-GSK-3β (Ser9) (cat. no. 8466, Cell Signaling Technology, Inc. 1:1,000), β-catenin (cat. no. 8480, Cell Signaling Technology, Inc. 1:1,000), phosphorylated-β-catenin (cat. no. 4176, Cell Signaling Technology, Inc. 1:1,000) at 4°C overnight. After washing with PBST, the membranes were incubated with horseradish peroxidase-conjugated secondary antibody (1:2,000; cat. no. 7074; Cell Signaling Technology, Inc.) for 1 h at room temperature. Finally, antibody labeling was detected by enhanced chemiluminescence (Thermo Fisher Scientific, Inc.). Band intensity was evaluated using ImageJ v1.48 U software (National Institutes of Health).

All data were presented as the means ± standard deviation (SD). The comparisons among data were calculated by one-way ANOVA followed by Tukey's post hoc test by using SPSS software, version 13.0 (SPSS, Inc.). P<0.05 was considered to indicate a statistically significant difference.

To assess the potential of hADSCs to differentiate into multiple lineages, several conventional analyses were conducted following the induction of osteogenesis, adipogenesis and chondrogenesis of hADSCs. As shown in Fig. 1A, hADSCs were able to differentiate into osteogenic cells (Alizarin Red S staining), adipocytes (Oil Red O staining) and chondrocytes (Alcian blue staining), which suggested that the hADSCs successfully differented into 3 different lineages. Alizarin Red S staining was performed to visualize calcium deposition in osteoblasts at 7, 14 and 21 days following the induction of osteogenesis. It was found that calcium deposition in osteoblasts was significantly increased, and these promoting effects were time-dependent (Fig. 1B). Moreover, the osteoblast phenotype was evaluated by determining ALP activity. The results revealed that ALP activity was also increased in a time-dependent manner (Fig. 1C). Several osteogenic markers, (Runx2 and BSP as early marker genes), (OPN and OCN as late marker genes) were also examined at the genomic level. As expected, the expression levels of these osteogenic markers were markedly increased in a time-dependent manner (Fig. 1D-G), suggesting that the osteogenic differentiation of hADSCs was induced successfully.

Previous studies have suggested that miRNAs play a role in osteogenic differentiation (20,21). In the present study, using microarray assay, it was found that the levels of 24 miRNAs were increased and those of 14 miRNAs were decreased in the differentiated cells compared with the undifferentiated control cells (Fig. 1H). Of the upregulated miRNAs, miR-29b expression displayed the most alteration in this array. Of relevance, miR-29b has been identified as a key regulator of the osteogenic differentiation of MSCs by targeting anti-osteogenic factors or modulating bone extracellular matrix proteins in a mineralogenic cell system (ABSa15) or in MC3T3 cells (22,23). However, whether miR-29b affects the osteogenic differentiation of hADSCs remains unclear. To further verify the expression of miR-29b, the expression of miR-29b was quantified by RT-qPCR at different time points during the osteogenesis process. As shown in Fig. 1I, the results revealed that miR-29b expression was increased in a time-dependent manner compared with the undifferentiated control cells. These results suggest that miR-29b may play an important role in the osteogenic differentiation of hADSCs.

To determine whether miR-29b affects the osteogenic differentiation of hADSCs, the hADSCs was transfected with agomir-29b, agomir-negative control (agomir NC), antagomir-29b and antagomir NC during osteogenic differentiation. As shown in Fig. 2A, miR-29b expression was significantly increased (or decreased) in the hADSCs following agomir-29b (or antagomir-29b) transfection, compared with the agomir (or antagomir) NC group. At 16 h after transfection, the medium was exchanged for osteogenic medium and the effects of miR-29b on the hADSCs induced to differentiate into osteoblasts were evaluated. It was observed that the overexpression of miR-29b significantly increased the mineralization of the extracellular matrix and ALP activity, while miR-29b inhibition produced the opposite results (Fig. 2B and C). Additionally, supporting the above-mentioned observations, agomir-29b increased the expression of bone-specific markers, such as Runx2, BSP, OPN and OCN at the mRNA and protein level, whereas the expression of bone-specific markers was significantly suppressed by antagomir-29b infection (Fig. 2D-H). More importantly, the effects of miR-29b on the tri-lineage differentiation potential of hADSCs were examined. As shown in Fig. 2I, miR-29b upregulation led to a significant increase in the adipogenic and chondrogenic differentiation of hADSCs, while miR-29b downregulation exerted an opposite effect, which is consistent with the findings of a previous study (24). These data suggest that miR-29b plays a regulatory role in the adipogenic and chondrogenic differentiation of hADSCs. Notably, previous studies have paid increasing attention to the osteogenic differentiation of hADSCs as an effective treatment for musculoskeletal disorders, such as osteoarthritis (OA) (25) and rheumatoid arthritis (RA) (26). Therefore, the present study focused on the progression of the osteogenic differentiation of hADSCs regulated by miR-26. In the future, the authors aim to further investigate the mechanisms underlying the regulation of the adipogenic and chondrogenic differentiation of hADSCs by miR-29b.

Through bioinformatics online programs, such as TargetScan, miRanda and PicTar, PTEN, a negative regulator of the AKT/mTOR signaling pathway, was found to have a putative target site of miR-29b in its 3′-UTR (Fig. 3A). To validate the possibility that PTEN was directly targeted by miR-29b, a luciferase reporter assay was performed. Luciferase reporter assay revealed that miR-29b mimics markedly inhibited the luciferase activity in the constructs containing the wild-type binding site of PTEN-3′UTR, and that the miR-29b inhibitor caused an increased luciferase activity. However, the luciferase activity of the constructs containing the mutant binding site exhibited no difference upon miR-29b expression compared with the negative control (Fig. 3B). Subsequently, to further validate the suppressive effect of miR-29b on PTEN, the expression of PTEN was measured in the hADSCs by western blot analysis. As shown in Fig. 3C, the PTEN levels were significantly downregulated following transfection with agomir-29b, whereas they were increased following transfection with antagomir-29b (Fig. 3C). In addition, the expression of PTEN decreased in a time-dependent manner during the osteogenic differentiation of hADSCs, as determined by western blot analysis (Fig. 3D). These results indicated that miR-29b can bind to PTEN through specific sites and can downregulate PTEN expression during osteogenic differentiation.

PTEN is an important upstream regulator of the AKT signaling pathway that has been implicated as an important regulator of osteoblast differentiation in MSCs (27,28). However, the association between miR-29b and the PTEN/Akt pathway in the osteogenic differentiation of hADSCs remains unclear. As shown in Fig. 4, the introduction of agomir-29b into the hADSCs significantly suppressed the expression of PTEN, but promoted p-AKT protein expression. As is already known, the AKT/GSK-3β/β-catenin signaling pathway is important for MSC differentiation, including osteogenic differentiation (6,29). Therefore, the present study further investigated whether miR-29b affects the AKT/GSK-3β/β-catenin signaling pathway during the osteogenic differentiation of hADSCs. As shown in Fig. 4, the overexpression of miR-29b in the hADSCs induced the expression of phospho-GSK3β and phospho-β-catenin compared with the cells infected with agomir-NC. These results suggest that miR-29b promotes the osteogenic differentiation of hADSCs via the PTEN/AKT/GSK-3β/β-catenin pathway.

To further investigate whether PTEN is involved in the miR-29b-mediated osteogenic differentiation of hADSCs, the hADSCs were co-transfected with agomir-29b and pcDNA-PTEN during osteogenic differentiation. It was found that the protein expression of PTEN was significantly increased in the hADSCs when they were transfected with the pcDNA-PTEN plasmid (Fig. 5A). As was expected, the protein expression of PTEN was also significantly increased in the miR-29b-overexpressing hADSCs when they were transfected with the pcDNA-PTEN plasmid (Fig. 5B). Subsequently, the ALP activity, the changes in matrix mineralization and the levels of osteogenic marker genes were assessed by ALP staining, Alizarin Red S staining and RT-qPCR assays, respectively. The results revealed that the upregulation of miR-29b by agomir-29b significantly enhanced the ALP activity and the mineralization of extracellular matrix, and led to an increase in Runx2, OPN, OCN and BSP mRNA expression, whereas the promoting effects of agomir-29b were reversed by the overexpression of PTEN (Fig. 5C-H). Collectively, these data suggest that miR-29b promotes the osteogenic differentiation of hADSCs by targeting PTEN.