The effects of mechanical stretch on the biological characteristics of human adipose‐derived stem cells

Abstract Adipose‐derived stem cells (ADSCs) are a subset of mesenchymal stem cells (MSCs), which have promised a vast therapeutic potential in tissue regeneration. Recent studies have demonstrated that combining stem cells with mechanical stretch may strengthen the efficacy of regenerative therapies. However, the exact influences of mechanical stretch on MSCs still remain inconclusive. In this study, human ADSCs (hADSCs) were applied cyclic stretch stimulation under an in vitro stretching model for designated duration. We found that mechanical stretch significantly promoted the proliferation, adhesion and migration of hADSCs, suppressing cellular apoptosis and increasing the production of pro‐healing cytokines. For differentiation of hADSCs, mechanical stretch inhibited adipogenesis, but enhanced osteogenesis. Long‐term stretch could promote ageing of hADSCs, but did not alter the cell size and typical immunophenotypic characteristics. Furthermore, we revealed that PI3K/AKT and MAPK pathways might participate in the effects of mechanical stretch on the biological characteristics of hADSCs. Taken together, mechanical stretch is an effective strategy for enhancing stem cell behaviour and regulating stem cell fate. The synergy between hADSCs and mechanical stretch would most likely facilitate tissue regeneration and promote the development of stem cell therapy.

could be considered as a kind of ideal seed cell for translational medicine research and regenerative medicine therapies.
Mechanical stretch is one of physical cues from an organismal level to a subcellular level, which is associated with many physiological and pathological processes such as beating heart or bone remodelling processes. 7,8 It has been indicated that in vitro mechanical stretch can modulate the balance of self-renewal and differentiation in various cell types. For example, cyclic stretch induces increased proliferation of human skin keratinocytes and dermal fibroblasts. 9,10 It also induces the cardiogenesis and vasculogenesis of embryonic stem cells (ESCs). 11,12 It is noteworthy that mechanical stretch has been widely used as an important newfangled pretreatment method in tissue engineering. For instance, cyclic stretch preconditioning improves engineered muscle contraction. 13 Proper mechanical stimulation may promote the cell-based ligament regeneration and success rate of transplantation. 14 Therefore, mechanical stretch may promote tissue regeneration by influencing cell biological characteristics.
In plastic and reconstructive surgery, ADSCs have been used for a variety of different applications such as autologous fat grafting or treatment of chronic wounds. 15,16 On the other hand, tissue regeneration through continuous mechanical stretch has received widely recognized during the last decades for repair of soft tissue defects. 17 Recent studies have found that combining stem cells with mechanical stretch may strengthen the effect of tissue regeneration. It is reported that the transplantation of ADSCs could possibly promote mechanical stretch induced skin regeneration and mitigate traditional limitations of skin expansion. 18 ADSCs also play a key role in external volume expansion induced by negative pressure. 19 When primed with stiff substrates, ADSCs stimulated by mechanical stimulus exhibited increased therapeutic efficacy in pathological wound healing. 20 In conclusion, mechanical stretch could enhance the therapeutic efficacy of ADSCs, and ADSCs has synergistic effect with mechanical stretch in promoting tissue regeneration. However, the exact influences of mechanical stretch on ADSCs still remain inconclusive.
Given this, our study aimed to investigate the effects of mechanical stretch on the biological characteristics of human ADSCs (hADSCs). In this study, hADSCs were applied cyclic stretch stimulation under an in vitro stretching model and the cellular proliferation, apoptosis, adhesion, migration, cytokine production, as well as differentiation potentials of hADSCs were examined. We further investigated the phenotypic characteristics of hADSCs cultured under cyclic stretch stimulation during long-term cultivation and mechanical stretch related cell signalling events.

| Isolation and culture of hADSCs
The adipose tissue was minced and digested with 0.2% collagenase type I (Sigma-Aldrich, USA) at 37°C for 1 hour. After filtration, centrifugation and resuspension, the pelleted hADSCs were treated in Dulbecco's Modified Eagle Medium (DMEM) (Gibco, Life Technique, NY) supplemented with 10% foetal bovine serum (FBS) (Hyclone, Thermo Fisher, MA) and 1% penicillin/streptomycin (Gibco, Life Technique, NY). After 48 hours, the non-adherent cells were removed and the medium was then changed every 3 days. The adherent spindle-shaped cells were further propagated for three passages.

| Immunophenotypic analysis of hADSCs
For flow cytometric analysis, P3 hADSCs were incubated with monoclonal PE-conjugated antibodies for CD34, CD45 and HLA-DR or with FITC-conjugated antibodies for CD73, CD90 and CD105 at room temperature for 30 min (BD Pharmingen, San Diego, CA).
Isotype control IgG was used to stain the cells as control. The cells were subsequently washed with phosphate-buffered saline (PBS), fixed with 4% formaldehyde and analysed on a FACScan flow cytometer (Becton-Dickinson, San Jose, CA).

| Multipotential induction
Multilineage differentiation capacity of hADSCs was detected as requested by the International Society for Cellular Therapy. Briefly, cells were cultured in adipogenic induction medium (Cyagen Biosciences, USA) for designated time and were stained by Oil Red-O staining. For osteogenic differentiation, cells were cultured in osteogenic inducing medium (Cyagen Biosciences, USA) for designated time and were stained by alizarin red S. Chondrogenic differentiation was performed using the micromass culture technique maintained in chondrogenic medium (Cyagen Biosciences, USA) for up to 5 weeks, and were stained by Alcian blue.

| Application of mechanical stretch
Mechanical stretch was applied to hADSCs by FX-5000T™ Flexcell Tension Plus (Flexcell International Corporation, Hillsborough, NC) ( Figure 1C). Briefly, hADSCs were seeded on six-well collagen I coated BioFlex™ culture plates. Cyclic stretch was applied in a sinusoidal pattern with 10% amplitude at 0.5 Hz for designated duration (stretch group). Cells cultured in the same plates but left non-stretched were used as controls (static group). The experimental protocol is summarized ( Figure 1D).

| Flow cytometry for cell cycle, apoptosis and cell size analyses
Cells were sedimented and fixed in 70% ethanol for overnight at 4°C. RNA was digested with RNase A Solution (5 Prime GmbH, Hilden, Germany), and DNA was stained with propidium iodide (Sigma, Deisenhofen, Germany) for 40 min at 37°C. Assessment of cellular apoptosis was done with Annexin V Apoptosis Detection Kit (BD Biosciences, New York).
Cell cytrometry analysis was performed using a BD FACSCalibur (BD Biosciences, Heidelberg, Germany) and analysed by FlowJo version 9.3.2 (Tree Star, Inc, Ashland). FSC-A parameters of the software were used for cell size measurements, as recommended by BD.

| Immunofluorescence for Ki67
Cells were fixed in 4% paraformaldehyde in PBS for 20 min at room

| TUNEL assay
The TUNEL method was performed with the one step TUNEL kit according to the manufacturer's instructions (Beyotime Institute of Biotechnology, China). Briefly, cells were fixed in 4% paraformaldehyde in PBS for 20 minutes at room temperature. Then the cells were permeabilized with 0.3% Triton X-100 in PBS followed by TUNEL for 1 hour at 37°C. Nuclei were stained using DAPI. Images were visualized under a confocal microscope (Zeiss, Germany) by using 488-nm excitation and 530-nm emission.
Quantitative analyses were performed by counting the positively stained cells under microscopy within at least five randomly chosen fields.

| Assays for cell adhesion and migration
For adhesion assay, cells were plated at a density of 1 × 10 5 cells/well in a six-well plate and were incubated for 30 minutes at 37°C. The wells were washed three times, and the number of attached cells was counted under microscope within at least five randomly chosen fields. Adhesion is expressed as fold change over the controls.

| Western blot analysis
Total proteins were extracted from cells utilizing Radio immunoprecipitation assay (RIPA) lysis buffer. After determining the concentrations of protein by bicinchoninic acid (BCA) assay (Thermo

| Enzyme-linked immunosorbent assay
After cell stretching for 24 h, the culture medium was collected and centrifuged at 2,500 rpm for 10 min. The supernatants were used and the concentrations of cytokines TGF-β1, IGF-1, VEGF, HGF, KGF, bFGF, SDF-1α and IL-6 were measured by enzyme-linked immunosorbent assay (ELISA) kits from the R&D system (Minneapolis, MN, USA) in accordance with the instruction manual. The number of ADSCs were counted and the results have been shown in the unit of "pg/mL/10 5 cells".

| RNA isolation and real-time polymerase chain reaction
TRIzol reagent (Invitrogen, Mulgrave, Australia) was utilized to ex-  Table 1. The housekeeping gene GAPDH was used for normalization.

| Antibody array
The Phospho Explorer Antibody Array (CSP100) was used to screen for differential expression of more than 130 protein phosphorylation sites (H-Wayen Biotechnologies, Shanghai, China). Analysis was carried out according to the protocol provided. Briefly, proteins were extracted and labelled by biotinylation. Then biotin-labelled proteins were applied to the array for conjugation. The conjugation-labelled protein was detected by Cy3-Streptavidin. The analysed results were expressed by the ratio of phosphorylation/unphosphorylation. 23,24

| Statistical analyses
Statistical differences among groups were assessed using two-tailed Student's t test. P < 0.05 was considered statistically significant.
Results are presented as the mean ± SD.

| Flow cytometry and multipotency analysis of isolated hADSCs
hADSCs were isolated from human subcutaneous adipose tissues and expanded easily when cultured in regular medium in vitro, which exhibited morphologically heterogeneous and fibroblastic shape. They were TA B L E 1 QRT-PCR primer sequence confirmed positive for CD73, CD90 and CD105, negative for CD34, CD45 and HLA-DR according to flow cytometry analysis of stem cellrelated surface markers ( Figure 1A). This is consistent with the typical mesenchymal stem cell surface marker profile. To verify the ability of the generated hADSCs cultures to differentiate into multiple cell types, the cells were analysed for osteogenic, adipogenic and chondrogenic potentials in vitro. This was confirmed by using oil red O staining for lipid droplets formation, alizarin red S staining for calcium deposits formation and Alcian blue staining for sulphated proteoglycans formation ( Figure 1B).

| Mechanical stretch enhanced the proliferation of hADSCs
To investigate the effect of mechanical stretch on the cellular proliferation, hADSCs were cultured under cyclic stretch stimulation for 24 hours to 72 hours. Quantitative analysis showed that the cell numbers were significantly more in the stretch group than controls at all examined time-points (P < 0.05, Figure 2A), although the cell counts were similar between both groups at baseline. Additionally, metabolic activity was analysed by Alamar Blue assay. We found that hADSCs cultured under cyclic stretch stimulation exhibited an increased activity compared with controls (P < 0.05, Figure 2B). Furthermore, the cellular proliferative capacity was determined by Ki67 immunofluorescent staining ( Figure 2C). Compared to the static group, the stretch group showed a significant increase in Ki67 positive cells (P < 0.01, Figure 2D), which indicated that hADSCs cultured under cyclic stretch stimulation clearly exhibited a higher proliferation rate compared with controls. Moreover, flow cytometric analysis of DNA content demonstrated the same trend that mechanical stretch led to a decrease in cell population in the G 0 /G 1 phase and an increase in the S phase and G 2 /M phase (P < 0.05, Figure 2E,F). Together, our results indicated that mechanical stretch enhanced the proliferation of hADSCs.

| Mechanical stretch inhibited the glucose deprivation induced apoptosis of hADSCs
To explore the effect of mechanical stretch on the glucose deprivation induced cellular apoptosis, hADSCs preconditioned with cyclic stretch for 12 hours, then subjected to glucose deprivation under static conditions for 8 hours. Apoptotic cell population was analysed by flow cytometry (Annexin V staining). Compared to the static group, the stretch group showed a significant decrease in apoptotic cells (P < 0.01, Figure 3A,B). In addition, the TUNEL assay was utilized to further validate the apoptotic cell proportion and the result was consistent with the results obtained from flow cytometry assay (P < 0.05, Figure 3C,D). Further Western blot analysis showed that the expression of cleaved caspase-3 was significantly lower in the stretch group than controls (P < 0.01, Figure 3E,F). These data suggested that mechanical stretch inhibited the glucose deprivation induced apoptosis of hADSCs.

| Mechanical stretch improved the adhesion and migration of hADSCs
We next evaluated the effect of mechanical stretch on the adhesion and migration of hADSCs. hADSCs preconditioned with cyclic stretch for 12 hours were replanted onto coverslips and cultured under static conditions. After 30 minutes of incubation, a portion of the cells were attached to the plate of the well, which adopted a uniform round-shaped population. More cells were attached in the stretch group compared with controls (P < 0.01, Figure 4A,B). The cellular migration was detected using a Transwell Boyden Chamber.
After 24 hours of incubation, more cells migrated to the underside of the filters in the stretch group compared with controls (P < 0.01, Figure 4C,D). Western blot analysis showed that the expression of integrin-β1 and p-FAK, two critical adhesion molecules for cell motility, increased significantly after mechanical stretch (P < 0.01, Figure 4E,F). In addition, the expression of the ECM proteins was also enhanced by mechanical stretch (P < 0.05, Figure 4G). These findings demonstrated the significant function of mechanical stretch in improving the adhesion and migration of hADSCs.

| Mechanical stretch inhibited adipogenesis, but enhanced osteogenesis of hADSCs
After adipogenic induction, the number of Oil Red O positive lipid droplets was decreased under cyclic stretch stimulation ( Figure 6A).  Figure 6D). From the above mentioned findings, hADSCs cultured under cyclic stretch stimulation had inhibited adipogenesis, but enhanced osteogenesis.

| Long-term mechanical stretch promoted ageing of hADSCs, but did not alter the cell size and typical immunophenotypic characteristics
To study whether long-term cyclic stretch stimulation affects the biological characteristics of hADSCs, P3 hADSCs were passaged under cyclic stretch stimulation once per week for 8 weeks. We found that more SA-β-gal positive cells were detected in hADSCs cultured under mechanical stretch stimulation than controls (P < 0.001, Figure 7A,B). In addition, hADSCs were still positive for CD73 and CD90, but negative for CD45 and HLA-DR as before ( Figure 7C).
And there were no significant differences in cell size between both groups ( Figure 7D). These results implied that long-term mechanical stretch promoted ageing of hADSCs, but did not alter the cell size and typical immunophenotypic characteristics.

| PI3K/AKT and MAPK signalling pathways may participate in the effects of mechanical stretch on the biological characteristics of hADSCs
To research the signalling events, we used a Phospho Explorer antibody microarray covering most typical known signalling pathways.
hADSCs were cultured under cyclic stretch stimulation for 30min and the phosphorylation states of proteins were then detected using antibody-based arrays ( Figure S1). We found that activated factors or proteins have been concentrated in PI3K/AKT and MAPK cellular signalling pathways (Tables S1 and S2). To determine whether cyclic stretch could activate the PI3K/AKT and MAPK pathways, which might contribute to the effects of mechanical stretch on the biological characteristics of hADSCs. hADSCs were exposed to cyclic stretching for 0, 5, 10, 15, 30, 45 minutes or 1, 3, 6, 24 hours, and the expression of p-ERK, p-AKT, PI3K and p-JNK was detected by western blot. We observed that the expression of all the proteins above increased rapidly after stretching and peaked at 30 minutes and then gradually fell back to the control level after 6 hours ( Figure 7E).
These results demonstrated that cyclic stretch stimulation could activate the PI3K/AKT and MAPK signalling pathways in hADSCs.

| D ISCUSS I ON
Nowadays, regenerative medicine as well as tissue engineering is searching for novel methods, which can promote the regenerative process of different tissue injuries and organ damages. Increasing evidence have demonstrated that ADSCs promote tissue repair through both direct regeneration and indirect mechanisms. 25 In addition, exogenous mechanical stretch has been proved to play an important role in regulating structure and function of various cells. 26,27 Mechanical stretch pretreatment has been reported to have positive effect on increasing therapeutic efficacy of cell transplantation in tissue engineering, such as dermal fibroblast. 28 In plastic and reconstructive surgery, synergistically combining ADSCs and mechanical stretch, has been used for skin and adipose regeneration. 29,30 However, the underlying mechanisms by which synergy between ADSCs and mechanical stretch induced tissue regeneration were not investigated. In this study, we explored the effects of mechanical stretch on the biological characteristics of hADSCs, including proliferation, apoptosis, adhesion and migration, differentiation, cytokine production and phenotypic characteristics during long-term cultivation.
First of all, we analysed the survival and viability that are most important factors directly correspond with regenerative process. 31 We found that hADSCs cultured under cyclic stretch stimulation Investigators have explored ways to increase MSCs migration and adhesion into injuries portion to promote tissue repair. 37 In the current study, we found that mechanical stretch improved the hesion molecules which are known to be critical for cell motility. 38,39 In our study, mechanical stretch could activate the FAK signals and up-regulate the expression of integrin-β1 in hADSCs, which are consistent with previous studies in cardiomyocytes and vascular smooth muscle cells. 40,41 What is more, the expression of the ECM proteins was enhanced in the stretch group, which is also important to cell survival and motility. 42 Our study confirmed the function of mechanical stretch on the adhesion and migration of hADSCs, which may be a novel method to increase MSCs delivery and efficacy.
A growing body of evidence have illustrated that the effect of promoting wound healing by ADSCs partly attributed to the production of diverse growth factors. 43 Previous studies showed that conditioned medium of ADSCs (ADSCs-CM) has the ability to promote tissue regeneration. 44 Our data showed that many kinds of tissue regeneration It has been reported that the transduction pattern of external mechanical signals into the intracellular biological signals determines the cell fate. 49,50 However, the complete pathways relating specific mechanical stimuli to stem cell fate remain to be elucidated. In the present study, we found that cyclic stretch stimulation could activate the PI3K/AKT and MAPK signalling pathways in hADSCs. PI3K and MAPKs have been reported to play vital roles in regulating mechanotransduction mechanisms. 51,52 The PI3K/ AKT signalling has been identified to regulate stem cell growth, proliferation, differentiation, motility and intracellular trafficking. 53 MAPKs are specific serine/threonine protein kinases that are involved in regulation of cell growth and differentiation, inflammation and apoptosis. 54 It is noteworthy that the crosstalk between PI3K/AKT and MAPK signalling pathways is important to regulate stem cell fate. 55 Taken together, we clearly demonstrated that mechanical stretch significantly promoted the proliferation, adhesion and migration of hADSCs, suppressing cell apoptosis and increasing the production of pro-healing cytokines. For differentiation of hADSCs, mechanical stretch inhibited adipogenesis, but enhanced osteogenesis. Long-term stretch could promote ageing of hADSCs, but did not alter the cell size and typical immunophenotypic characteristics. Furthermore, we revealed that PI3K/AKT and MAPK pathways might participate in the effects of mechanical stretch on the biological characteristics of hADSCs. This is the first time (to our knowledge) that the in vitro cyclic stretch model was successfully used to investigate how mechanical stretch affects the multitudinous biological characteristics of hADSCs. These findings provide the first direct evidence that mechanical stretch is an effective strategy for regulating stem cell fate and could serve as an important newfangled pretreatment in stem cell therapy and regenerative medicine.

ACK N OWLED G EM ENTS
This work was supported by National Natural Science Foundation of China (No. 81701917) and Natural Science Foundation of Shanghai (No. 17ZR1416500).

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.