Umbilical cord–derived mesenchymal stromal cell–conditioned medium exerts in vitro antiaging effects in human fibroblasts

ARTICLE IN PRESS Cytotherapy, 2016; ■■: ■■–■■

Umbilical cord–derived mesenchymal stromal cell–conditioned medium exerts in vitro antiaging effects in human fibroblasts

MEIRONG LI1,2,*, YALI ZHAO1,2,*, HAOJIE HAO1, LIANG DONG1, JIEJIE LIU1, WEIDONG HAN1 & XIAOBING FU1 1

Wound Healing and Cell Biology Laboratory, Institute of Basic Medical Science, Chinese PLA General Hospital, Beijing, China, and 2Trauma Treatment Center, Central Laboratory, Hainan Branch, Chinese PLA General Hospital, Sanya, China Abstract Background aims. Chronic wounds are a common complication of diabetes. Fibroblast-myofibroblast differentiation is important for wound repair, which is commonly impaired in non-healing wounds, and the underlying mechanisms need to be further elucidated. Methods. We used high glucose (HG) to simulated the diabetes microenvironment and explored its effects on the biological features of fibroblasts in vitro. Results. The results showed that prolonged HG induced senescence in fibroblasts through activation of p21 and p16 in a reactive oxygen species (ROS)-dependent manner, further delayed the viability and migration in fibroblasts and also depressed fibroblast differentiation through the TGF-β/Smad signaling pathway. However, mesenchymal stromal cell–conditioned medium (MSC-CM) counteracts the effects of HG. Treatment of fibroblasts with MSC-CM decreased HG-induced ROS overproduction, ameliorated HG-induced senescence in fibroblasts and reversed the defects in myofibroblast formation. Our results may provide clues for the pathogenesis of chronic wounds and a theoretical basis to develop MSC-CM as an alternative therapeutic method to treatment of chronic wounds. Key Words: chronic wounds, high glucose, myofibroblast, mesenchymal stromal cell-conditioned medium, senescence, oxidative stress

Introduction With the expansion of the diabetes mellitus (DM) epidemic, the incidence of chronic wounds is dramatically increasing. Chronic wounds are important complications of diabetes that can lead to amputation and even death [1]. Fibroblasts, cells that are involved in wound healing, participate in granulation tissue formation and provides a scaffold for repopulating cells through the secreting of extracellular matrix (ECM) [2]. More importantly, fibroblasts differentiate into myofibroblasts, enhancing wound contraction, and then promote wound closure [2]. Transforming growth factor beta (TGF-β1), a pleiotropic factor, participates in the whole process of skin repair, and it is also considered to be the major growth factor inducing fibroblast differentiation [3].The TGF-β signaling pathway activates and inhibits target genes, mainly through canonical Smaddependent signaling. Deregulation of the canonical Smad-dependent signaling pathway of TGF-β impairs myofibroblast formation. Reduced myofibroblast de-

velopment has been discovered in chronic wounds in diabetic patients [4], and the mechanisms underlying this developmental deficiency are not fully understood. Impairment in regulating glucose metabolism and elevated glucose levels are the main etiology of DM [5]. Substantial evidence suggests that high-glucoseinduced oxidative stress formation is closely related with senescence in a variety of cells, such as keratinocytes, endothelial cells and fibroblasts [6–8]. For example, hyperglycaemia-induced generation of reactive oxygen species (ROS) accelerates the shortening of telomere length of endothelial cells [9–11]. The replicative life span of skin fibroblasts derived from diabetic subjects is reduced compared with controls [12].Therefore, we speculated that oxidative stress and senescence may be pathogenically linked with the development of chronic wounds. Effective strategies for treating non-healing wounds are still lacking. Using stem cells, especially mesenchymal stromal cells (MSCs), may be a promising means for chronic wound therapy. A large body of

*These authors contributed equally to this work. Correspondence: Weidong Han, MD, PhD, and Xiaobing Fu, MD, PhD, Institute of Basic Medical Science, Chinese PLA General Hospital, 28 Fuxing Road, HaiDian District, Beijing 100853, P. R. China. E-mail: [email protected]; [email protected] (Received 19 July 2016; accepted 5 December 2016) ISSN 1465-3249 Copyright © 2017 International Society for Cellular Therapy. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcyt.2016.12.001

ARTICLE IN PRESS 2

M. Li et al.

evidence indicates that MSCs can promote wound closure of chronic wounds in animal models and in preclinical studies [13,14]. Umbilical cord–derived MSCs (UC-MSCs), which are required with noninvasive harvesting procedures, meet the criteria for MSCs and exhibit a more prominent cytokine secretion profile than MSCs from other sources [15]. Additionally, increasing evidence suggests that the mechanism underlying UC-MSCs efficacy depends mostly on their paracrine activity [16]. All the bioactive factors and cytokines in MSCs secretions constitute can be collected in the conditioned medium. Accumulating evidence suggests that MSC–conditioned medium has similar therapeutic effects to MSCs. MSCCM can enhance the repair of myocardial infarction [17], enhance wound healing [16] and reduce fibrotic kidney injury [18,19]. In addition, research reporting the ability of MSC-CM to repair dysfunctional cells in a diabetic environment is just emerging [20,21]. Whether MSC-CM can ameliorate the biological function of fibroblasts exposed to a diabetic microenvironment is not fully understood, and the potential mechanisms need to be further elucidated. We hypothesized that the pathological stress of high-glucose conditions increased senescence markers in human fibroblasts that were correlated with functional impairment. Thus, in this study, we provide important evidence for the mechanism underlying the pathogenesis of non-healing wounds under the pathological state of diabetes. Additionally, we further demonstrate that MSC-CM exposure recovers the function of fibroblasts through regulation of the canonical Smad-dependent signaling pathway. These results provide a theoretical basis for the clinical application of MSC-CM in diabetic wound healing. Methods Fibroblast culture and treatment Fibroblasts were isolated from human foreskin through routine circumcision. All patients provided written informed consent. Collected foreskin specimens were washed with a phosphate-buffered saline (PBS) supplement with 1% penicillin-streptomycin solution (PS). All samples were then cut into small pieces and incubated in collagenase type I (Sigma) for 1 h at 37°C. The digestion was terminated, and the supernatant was collected for centrifugation. The cell pellets were gently resuspended in Dulbecco’s Modified Eagle’s Medium (DMEM) (Invitrogen) containing 10% fetal bovine serum (FBS), and the culture medium was released every 2 days. Thomas et al. [22] have shown that the glucose concentration in subcutaneous tissue is similar to that

in plasma (~6 mmol/L).Therefore, fibroblasts were cultured in normal glucose (6 mmol/L) as the control. Many previous studies have adopted 26 mmol/L glucose to simulate a high-glucose condition. In the study, fibroblasts are cultured in a glucose concentration of 26 mmol/L (HG1), and 30 mmol/L (HG2) were used to mimic hyperglycemic conditions as in other studies [6,8]. For investigating the effect of high glucose on foreskin fibroblasts (FFs) differentiation, FFs were incubated with glucose at the final concentration of 6 mmol/L (control), 26 mmol/L (HG1) and 30 mmol/L (HG2) for 1, 3 and 5 days, and then the cells further incubated with recombinant human TGFβ1 (5 ng/mL, 48 h) for inducing differentiation. To explore the regulatory role of MSC-CM in high glucose treated fibroblasts, fibroblasts were pretreated with UCMSC-CM (at two concentrations, 1% [CM1] or 5% [CM2], respectively) and then incubated with high glucose (HG1 and HG2) for corresponding times. All experiments were performed three times, and results are reported as means ± SE.

Cell proliferation assay Cell viability was assessed by the CCK8 assay. Fibroblasts were plated in 96-well plates (Costar), and 12 h later, the concentration of glucose in the culture medium was adjusted to 26 mmol/L and 30 mmol/L in the two high-glucose experimental groups. After incubation in high-glucose medium for 1, 3 or 5 days, CCK-8 was added to the media, and the cells were incubated for 4 h. The plates were read on an enzyme immunoassay analyzer at 450 nm. The CCK8 assays were performed in triplicate, and the mean values were used (n = 5).

Scratch assay The motility properties of fibroblasts after exposure to high-glucose treatment for three different durations was evaluated by scratch wound assay. Fibroblasts were inoculate in six-well plates and cultured in DMEM containing 10% FBS for 12 h. The culture medium was then changed to DMEM containing 0.5% FBS, and the concentration of glucose in the culture system was modulated. When they reached 80% confluence, the cells were scraped with a p200 pipette tip. After washing the cells with PBS three times, serumfree DMEM was immediately added. To monitor wound closure, photographs were taken under a phasecontrast microscope (0 and 24 h). The distance of fibroblast migration was evaluated by a migration index. The scratch assays were performed in triplicate, and the mean values were used.

ARTICLE IN PRESS UC MSC medium exerts in vitro antiaging effects Preparation of UC-MSC-CM UC-MSCs were cultured in DMEM/F12 containing 10% FBS at passage 3.When the cells reached 70% confluence, they were washed with PBS and then cultured in DMEM/F12 for an additional 48 h. The conditioned medium was collected and centrifuged at 2000 rpm for 5 min. The supernatant was collected as UC-MSC-CM, which was further concentrated 10fold using an ultrafiltration membrane with a molecular weight cutoff of 5 kDa (Millipore). The concentrated product was filtered through a 0.2-μm filter and stored at −80°C until use. UC-MSC-CM from MSCs derived from three umbilical cords were detected in this study. Measurement of intracellular ROS Fibroblasts were cultured for 1, 3 and 5 days with the specified medium, washed with PBS and incubated with 10 μmol/L 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA, Sigma) in a light-protected humidified chamber for 20 min. The cells were washed again and left until observation. Moreover, fibroblasts were pretreated with N-acetylcysteine (NAC) (10 μmol/L) for 4 h, or CM1 or CM2 for 12 h and then incubated with high-glucose (HG1 and HG2) for corresponding times.The cells that were incubated with 100 mmol/L H2O2 were as positive controls (data are not shown), and the probe was omitted as negative control.The accumulation of ROS in cells was viewed on a confocal microscope and imaged (Leica DMI6000). The data are the means from independent experiments performed at least three times. Senescence-associated β-galactosidase staining A SA-β-gal-kit was adopted to process the SA-β-gal staining of fibroblasts. Briefly, the cells were fixed for 10 min and then incubated with a SA-β-gal stain solution for 24 h at 37°C (without CO2). The stain solution was then removed, and the cells were rinsed with PBS. The cells were observed under a phasecontrast microscope and then imaged. The ratio of positive cells was determined by counting the blue cells versus total cells. Western blot analysis Treated fibroblasts were lysed using RIPA buffer (Invitrogen) containing a protease inhibitor cocktail and were lysed on ice for 30 min.The lysates were purified, and the protein concentrations were determined. Equal amounts of protein from different experimental groups were separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred to polyvinylidene fluoride membranes. The membranes were then blocked with 5% non-fat milk

3

and incubated overnight with antibodies, including α-SMA, collagen I, Smad2/3, phosphorylated Smad2/3 and Smad7. A horseradish peroxidase–conjugated antibody was used as the secondary antibody. The proteins were visualized using a chemiluminescence detection system according to the manufacturer’s instructions (Pierce, Rockford, IL). Immunofluorescence staining The treated fibroblasts were washed in PBS and fixed with 4% paraformaldehyde. The fixed cells were blocked with 10% goat serum and then incubated overnight at 4°C with anti-α-SMA antibody (Sigma) followed by incubation with a DyLight 488-conjugated secondary antibody (Santa Cruz) and Hoechst 33342 (Sigma) to reveal the nuclei.The cells were viewed and photographed with a fluorescence microscope in five randomly selected fields. Negative control experiments were performed by omitting the primary antibody. Statistical analysis All data are expressed as the means ± SEM. At least three independent experiments were performed for each experiment. Statistical analysis was performed using one-way analysis of variance (ANOVA) for comparison of group means. A P value <0.05 was used to indicate statistical significance. Results Characterization of UC-MSCs The isolated UC-MSCs exhibited spindle shape and could differentiate into adipocytes and osteoblasts (supplementary Figure S1A and B). Moreover, flow cytometry data showed that UC-MSCs were high expression of CD105, CD90 and CD73, and no expression of CD45, CD34, CD11α and HLA-DR (supplementary Figure 1SC), showing the specific characteristics of MSCs. Few SA-β-gal-positive cells were found in cultured UC-MSCs (supplementary Figure S1D). Representative UC-MSC sample data are shown in supplementary Figure S1. Fibroblasts showed different proliferation and migration responses to acute hyperglycemia and chronic hyperglycemia Several studies have demonstrated that the proliferation and migration of fibroblasts are critical to wound healing [23,24], and non-healing wounds are often associated with fibroblast dysfunction [8].To investigate whether high glucose is involved in regulating the proliferation and migration of fibroblasts, the CCK8 and scratch wound healing assays were used to deter-

ARTICLE IN PRESS 4

M. Li et al. glucose, the viability and motility of fibroblasts were significantly suppressed. Diabetes is a chronic disease that develops slowly. Thus, with the progress of diabetes, there is a high potential risk of impairment of fibroblast function. ROS significantly increased in the prolonged high-glucose stimulation

Figure 1. The effect of high glucose on the proliferation and motility of fibroblast. (A) After incubation fibroblasts with high glucose for 1, 3 and 5 days, cell viability was evaluated using the CCK8 assay. (B) Fibroblast migration was determined using the in vitro wound closure assay. Migration indices indicated the distance moved by fibroblasts to close the scratch wound was used to evaluate the migration ability of fibroblasts. The results were given as the mean ± SD (n = 5) and are representative of three independent experiments. *P < 0.05; **P < 0.01 versus control (CON) group.

mine cell proliferation and migration in fibroblasts with different treatments. First, we evaluated whether exposure to high-glucose affects the viability of fibroblasts. As showed in Figure 1, after treatment with high glucose at 26 and 30 mmol/L, the viability of fibroblast was not significantly affected at day 1, although the proliferation of these cells was slightly higher compared with the control group (Figure 1A). However, as time went on, the cell viability of the high-glucose experimental groups was reduced at day 3 (Figure 1A). In addition, comparison of the control group with the experimental groups indicated further significant differences in cell proliferation at day 5 (Figure 1A). A scratch wound healing assay was used to assess the effect of high-glucose on fibroblast mobility. Migration indices represented the distance traveled by experimental or control fibroblasts to close the scratch wound. These indices were used to evaluate the migration ability of fibroblasts after treatment with control, HG1 and HG2. As the results in Figure 1B and supplementary Figure S2 show, the experimental groups HG1 and HG2 had increased fibroblast migration at day 1 compared with the control group. Nevertheless, 3 days post–high-glucose treatment, the migratory capacity of fibroblasts decreased by 11% and 28% in the HG1 and HG2 groups, respectively. As time went on, this high-glucose-induced suppression of fibroblast migration was even more pronounced at day 5 compared with the control group (23% and 39% in HG1 and HG2, respectively). The results illustrated that high-glucose stimulation promoted the viability and motility of fibroblasts for a short period time. However, after prolonged stimulation with high

Increased ROS generation is a hallmark of diabetes [25]. Fibroblasts were treated with high glucose with two concentrations (HG1 and HG2) for three durations, and the intracellular ROS generation in fibroblasts was determined by DCFH-DA fluorescence. The results showed that intracellular ROS generation was gradually increased after exposure to high glucose for 1, 3 and 5 days (Figure 2). In addition, the intracellular ROS levels also increased in a glucoseconcentration-dependent manner, reaching maximum levels with 30 mmol/L high glucose as opposed to lower concentrations of glucose when cells exposed for the same durations were compared. High-glucose treatment groups had more cells with high fluorescence intensities than the control group (Figure 2). Moreover, it was also found that the antioxidant NAC significantly decreased the intracellular level of ROS after high-glucose treatment (supplementary Figure S3B). These data suggest that high glucose induced the production of intracellular ROS in a timeand dose-dependent manner. Prolonged high-glucose stimulation accelerates fibroblast senescence via upregulation of p16 and p21 Cellular senescence is associated with upregulation of cyclin-dependent kinase inhibitors, such as p16 and p21, and with the appearance of SA-β-gal activity [26]. The results showed that the expression levels of p53, p21 and p16 in fibroblasts were increased by treating the cells with both high glucose concentrations (HG1 and HG2) in a dose- and time-dependent manner. The expression of p16, p21 and p53 was slightly upregulated in high-glucose-treated fibroblasts at day 1, was upregulated in high-glucosetreated fibroblasts at day 3 and was further increased at day 5 (Figure 3B). In agreement with this finding, compared with the control group, the SA-β-galpositive cells were noticeably increased after prolonged treatment with HG1 and HG2 in a dose- and timedependent manner (Figure 3A). Taken together, our data suggest that fibroblasts exhibit senescent behavior under chronic hyperglycemic conditions and that this senescence might be regulated through the activation of p53/p21 and p16. In addition, pre-incubation with antioxidant NAC not only inhibited the intracellular ROS production

ARTICLE IN PRESS UC MSC medium exerts in vitro antiaging effects

5

Figure 2. High-glucose-induced ROS generation in fibroblasts. Fibroblasts were treated with high glucose with two concentrations (HG1, HG2) for 1, 3 and 5 days, and then the cells were labeled with DCFH-DA, followed by fluorescence microscopy analyses. Data are shown as the mean ± SD of three independent experiments. *,# P < 0.05; **,## P < 0.01 versus control (CON) group. Scale bar = 50 μm.

but also reduced high-glucose-induced SA-β-galstaining in fibroblasts (supplementary Figure S3).

Figure 3. High-glucose-stimulation accelerates fibroblast senescence via upregulation of p16 and p21. (A) SA-β-gal-kit was adopted to evaluate the SA-β-gal activity in fibroblasts. The SA-β-Gal-positive cells were increased after treatment with HG1 and HG2 in a dose- and time-dependent manner. (B) Western blot analysis demonstrated that the expression levels of p53, p21and p16 were increased by treating with HG1 and HG2 in a dose- and time- dependent manner. The results were given as the mean ± SD (n = 5) and are representative of three independent experiments. *,#P < 0.05; **,##P < 0.01 versus control (CON) group. Scale bar = 100 μm.

ARTICLE IN PRESS 6

M. Li et al.

Chronic hyperglycemia inhibited fibroblast differentiation through TGF-Smad-dependent signaling pathway Fibroblast differentiation into myofibroblast is essential for wound healing, which is usually defective in chronic wounds [4]. We sought to determine whether a high-glucose condition can modulate fibroblast differentiation. In standard culture medium, 60.75 ± 8.92% of fibroblasts spontaneously differentiate into myofibroblasts in the presence of TGF-β1. Under an acute hyperglycemia condition, the number of a-SMA-positive cells was up to 66.5 ± 7.18% (HG1) and 68.25 ± 6.24% (HG2), respectively (supplementary Figure S4), with exposure to TGF-β1. Compared with fibroblast differentiation under normal culture conditions, exposure to a hyperglycemia condition reduced the fraction of differentiated myofibroblasts to 31.24 ± 8.27% and 25 ± 9.32% (HG1 and HG2,

respectively) at day 3 post–high-glucose treatment and to 18.75 ± 5.24% and 15.25 ± 5.17% (HG1 and HG2, respectively) in fibroblast at day 5 Post-high-glucose treatment (Figure 4A). Smad signaling has been shown to be the principal signaling pathway for TGF-β1-induced myofibroblast formation [27]. We wanted to explore whether high glucose stimulation affects the phosphorylation of Smad2/3. Fibroblasts were grown in monolayer under HG1, HG2 or control condition for 1, 3 or 5 days.The cells were grown to confluence and serum starved for 12 h before TGF-β1 was added.We evaluated the expression and phosphorylation of Smad2 and Smad3 on fibroblasts by immunoblotting. As shown in Figure 4B, the phosphorylation level of Smad 2/3 was enhanced in the experimental group compared with the level observed in the normal condition at day 1. In contrast, HG stimulated Smad 2/3

Figure 4. High-glucose regulated fibroblast differentiation through TGF-Smad-dependent signaling pathway. (A) Pre-incubated with highglucose for one day, the a-SMA-positive cells (green) were increased after TGF-β1 treatment. However, the a-SMA-positive cells were obviously decreased when pre-treated with high-glucose for 3 or 5 days. (B) Western blot assay was performed to determine the expression levels of myofibroblast markers (a-SMA and collagen I) and TGF-β signaling pathway.The change in protein (a-SMA, collagen I, Smad2/3, p-Smad2/3 and Smad7) expression, as assessed by densitometry, was quantified compared with control (CON). Quantitation of changes in protein expression are expressed as the mean ± SD. *,#P < 0.05; **,##P < 0.01 versus control group.

ARTICLE IN PRESS UC MSC medium exerts in vitro antiaging effects phosphorylation was significantly suppressed at days 3 and 5 in the presence of both concentrations of HG. Moreover, Smad7, the natural inhibitor of the TGF-β/Smad signaling pathway, showed the opposite expression pattern compared with phosphorylated Smad2/3 but was not changed significantly (Figure 4B).These observations indicated that chronic hyperglycemia may inhibit the TGF-β-mediated myofibroblast formation by regulating the TGF-β1Smad pathway. MSC-CM reduces HG-induced oxidative stress, senescence in fibroblasts MSC-CM has been proven play a positive role in wound healing [13,28].There are a variety of cytokines and regulatory factors in the MSC-CM, so its role in restoring the function of injured cells and tissues is complicated. First, we investigated the potential role of MSC-CM in regulating intracellular ROS. Compared with NAC, pretreatment with MSC-CM inhibited the high-glucose-induced intracellular ROS generation (Figure 5). Furthermore, after pretreatment of fibroblasts with MSC-CM, the number of SAβ-gal positive cells and the expression of p53, p21 and p16 were significantly reduced (Figure 6), whereas FFCM pretreatment of fibroblasts showed no significant effects on these detection index (supplementary Figure S5). The data showed that the mechanism through which MSC-CM inhibits cell senescence might be regulated through a decrease in the level of ROS in fibroblasts.

7

MSC-CM resists chronic hyperglycemia-induced fibroblast differentiation dysfunction To evaluate whether MSC-CM can reverse the inhibitory effects of high-glucose in fibroblast differentiation, we pretreated fibroblasts with MSC-CM for 12 h. We found that the percentage of a-SMA-positive fibroblasts was much higher in the prolonged high-glucose group that was pretreated with MSC-CM than in the group that was treated with high-glucose alone, which had no significant difference in the percentage of a-SMApositive fibroblasts compared with the control group. Therefore, MSC-CM had a therapeutic effect on myofibroblast formation deficiency, and these effects occurred in a dose-dependent manner. We further explored the possible mechanisms associated with this effect.We determined that chronic hyperglycemia could inhibit myofibroblast formation through the Smad signaling pathway. In this article, we further explore whether MSC-CM restores fibroblast function through regulation of this pathway. As shown in Figure 7A and B, MSC-CM could prevent the downregulation of phosphorylation of smad2/3 and the upregulation of Smad7 in chronic hyperglycemic conditions.These data suggest that MSC-CM can prevent chronic hyperglycemiasuppressed myofibroblast formation via activation of the TGF-β/Smad pathway. Discussion Hyperglycemia has been implicated as the main pathological and physiological symptom of diabetes mellitus

Figure 5. MSC-CM reduced high-glucose-induced oxidative stress in fibroblasts. Fibroblasts were pre-treated with MSC-CM, and then treated with high-glucose with HG1 and HG2 for 3 and 5 days. The cells were labeled with DCFH-DA, followed by fluorescence microscopy analyses. Data are shown as the mean ± SD of three independent experiments. *,#P < 0.05; **,##P < 0.01 versus control (CON) group. Scale bar = 50 μm.

ARTICLE IN PRESS 8

M. Li et al.

Figure 6. MSC-CM inhibited high-glucose-induced senescence in fibroblasts. (A) Fibroblasts were pre-treated with MSC-CM, and then treated with high-glucose with HG1 and HG2 for 3 and 5 days. SA-β-gal-kit was used to evaluate the SA-β-gal activity. SA-β-gal-staining cells were counted in at five randomized fields. The results shown here are from on triplicate experiments. *P < 0.05, **P < 0.01 versus control (CON) group. Scale bar = 100 μm. (B) Western blot assay was performed to determine the expression levels of p53, p21 and p16. Pre-incubated with MSC-CM, the protein expression of p53, p21 and p16 was significantly inhibited in the high-glucose-treatment groups. Quantitation of changes in protein expression are represented as the mean ± SD. *P < 0.05; **P < 0.01 versus control group. GADPH, glyceraldehydes-3-phosphate.

[29]. Diabetic wounds are often affected by fibroblast differentiation disorder, which negatively affects wound healing [4]. This study aimed to evaluate the effect of high glucose on the physiological function of fibroblasts in a diabetic microenvironment. It has been

reported that fibroblasts are more resistant to environmental stress in high FBS concentrations than in low FBS concentrations [8]. Fibroblasts are cultured in media with an FBS concentration of 10%. Therefore, in this experiment, 2% of FBS was used,

ARTICLE IN PRESS UC MSC medium exerts in vitro antiaging effects

9

Figure 7. MSC-CM resists prolonged high-glucose-induced fibroblast differentiation dysfunction. (A) MSC-CM inhibited the prolonged high-glucose caused fibroblast differentiation depression in term of α-SMA staining and myofibroblast markers (α-SMA and collagen I) expression. Western blot assay was performed to determine the protein levels of TGF-β/Smad signaling pathway, showing the expression levels of p-Smad2/3 was not been down-regulated after prolonged high-glucose treatment. (B) Pre-incubated with MSC-CM, the inhibitory effect of HG2 on fibroblast differentiation was also eliminated. Immunofluorescence staining for α-SMA and Western blot assay for myofibroblast markers (α-SMA and collagen I) and TGF-β/Smad signalling pathway were used for evaluated the MSC-CM’s role in fibroblast differentiation. Quantitation of changes in protein expression are represented as the mean ± SD. *,#P < 0.05; **, ##P < 0.01 versus control group. GADPH, glyceraldehydes-3-phosphate.

ARTICLE IN PRESS 10 M. Li et al. which is sufficient to meet the nutritional needs of fibroblasts and avoids any potential impact of a high FBS concentration on the results. In addition, the fibroblasts were stimulated with two different concentrations of high-glucose for several durations to simulate acute and chronic hyperglycemia. Fibroblasts play a positive role in wound healing through rapid proliferation and migration to fill the wound, and they differentiate into myofibroblasts, which contract the wound, thereby promoting wound repair. Defects in any biological characteristics of fibroblasts may affect the healing process. In our study, we found that fibroblast viability and motility was enhanced in an acute hyperglycemic condition. Nevertheless, the proliferation and migration of fibroblasts was substantially inhibited in a chronic hyperglycemic condition. Prolonged stimulation with high glucose not only impaired the proliferation and migration of fibroblasts but also depressed the ability of fibroblast differentiation in the presence ofTGF-β1.TGF-β1 has emerged as the major modulator of myofibroblast formation [30,31]. As noted in many studies, we chose a-SMA as a marker of the myofibroblast phenotype. The experimental results were as hypothesized; chronic hyperglycemia decreased the percentage of cells stained with a-SMA.We further explored the molecular mechanism of these effects. Smad signaling has been shown to be the principal signaling pathway for TGF-β1induced myofibroblast formation [32]. As expected, prolonged exposure of high glucose blocked Smad2/3 phosphorylation (Figure 3B).This inhibition of Smad signaling suggests that chronic hyperglycemia suppressed fibroblast differentiation by deregulating the TGF-β1/Smad2/3 signaling pathway. Smad7 is known to be an inhibitor that attenuates TGF-β signal transduction [33]. The current results showed that the expression level of Smad7 was also affected. Therefore, Smad7 may affect the phosphorylation of Smad 2/3 under chronic hyperglycemic conditions in diabetic patients. In addition to signaling via Smad proteins, TGF-β1 also activates other signaling pathways, including the MAP kinase (ERK, p38 and JNK), PI3K/ Akt and Rho GTPase pathways [33]. Whether these signaling pathways are involved in hyperglycemiaregulated fibroblast differentiation is unknown and requires further study. In addition, we note that acute hyperglycemia not only increases the proliferation and migration of fibroblasts but also slightly increases the rate of fibroblast differentiation, as indicated by more α-SMA positive cells and increased phosphorylation of Smad 2/3. These results further corroborate that fibroblasts are activated under acute hyperglycemic condition. Accumulating evidence suggests that hyperglycemia induces ROS overproduction in a variety of cells, including human tubular epithelial cells [34], endo-

thelial cells [35] and fibroblasts. Sustained high levels of ROS induce premature senescence in these cells [36–38]. Senescence is thought to be a main contributor to disease-related tissue malfunction [6,39,40]. Substantial evidence suggests that endothelial cell senescence may be involved in chronic wound formation. Senescence can be mediated by the p53-p21-pRB and p16-pRB pathways [41]. The current findings demonstrate that exposure to chronic hyperglycemia significantly increases senescence in fibroblasts in a dose- and time-dependent manner. The senescenceassociated proteins, including p53, p21 and p16, were upregulated in fibroblasts after chronic treatment with HG1 and HG2. Additionally, the percentage of SAβ-gal-staining cells was relatively higher in fibroblasts cultured under chronic hyperglycemic conditions than in fibroblasts cultured under control conditions.These results demonstrated that chronic hyperglycemiainduced fibroblast senescence might be regulated by both the p53/p21 pathway and the p16 pathway. We further explored the relationship between oxidative stress and fibroblast senescence. In our study, pretreating fibroblasts with NAC, an antioxidant, significantly inhibited the SA-β-gal activity observed in cells cultured in high-glucose media, further illustrating that the accumulation of intracellular oxidative stress might be related to fibroblast senescence.These results are consistent with other data that the replicative life span of skin fibroblasts derived from diabetic subjects is reduced compared with those derived from controls. A number of recent studies have indicated that ROS can be used as a messenger to mediate inhibition or activation of the downstream signaling molecules [42,43]. In this study, acute high-glucose-induced low levels of ROS in fibroblasts and chronic high-glucoseinduced high levels of ROS in fibroblasts. Meanwhile, the biological function of fibroblasts appeared to increase first with acute high-glucose and then decrease with chronic high-glucose exposure. After fibroblasts were exposed to the antioxidase NAC, these bidirectional regulation effects of high-glucose culture were almost abolished. In addition, our previous study showed that short or prolonged stimulation of keratinocytes with high-glucose media presented the same changing trend in terms of the proliferation and migration. All of these data indicated that the level of intracellular ROS has a regulatory effect on the biological function of fibroblasts and that the related regulatory mechanisms need to be further studied. It is possible that further understanding of this mechanism will provide a therapeutic target for improving the function of cells in a chronic high-glucose environment. Additionally, in this study, fibroblast senescence is correlated with a decrease of fibroblast function, which is in accord with the results of

ARTICLE IN PRESS UC MSC medium exerts in vitro antiaging effects previous research that indicated that the activity and function of aged cells is lower than that of young cells [44,45]. Therefore, high-glucose-induced oxidative stress contributes to the induction of senescence and to the dysfunction of fibroblasts. Whether senescence is the direct cause of fibroblast inactivation remains unclear and warrants additional study. Furthermore, previous work suggests that senescent cells not only display reduced function but also have an altered gene expression profile. They secrete inflammatory cytokines and degradative enzymes that lead to chronic inflammation and extracellular matrix degradation, thereby impairing tissue repair [46]. Additional studies are needed to determine the secretory characteristics of aged fibroblasts in a diabetic microenvironment. Several studies have shown that MSC-CM improves the quality of healed skin in acute and chronic wounds in vivo [13,28,47]. Furthermore, MSC-CM can promote the viability and migration of a variety of cells, such as keratinocytes, endothelial cells and fibroblasts in vitro [48,49]. Few studies focus on the therapeutic effects of MSC-CM on dysfunctional cells. In the current study, the results showed that MSCCM could significantly suppress the intracellular level of ROS induced under chronic hyperglycemia conditions. Additionally, genes related to senescence were down-regulated, and SA-β-gal activity decreased in fibroblasts under MSC-CM pretreated conditions. Therefore, treatment with MSC-CM has the effect of antioxidant and antiaging activity in fibroblasts. More important, MSC-CM antagonizes chronic hyperglycemia-induced dysfunctions in myofibroblast formation through specific activation of the TGF-βSmad signaling pathway. Therefore, we preliminarily conclude that MSC-CM can enhance the ability of fibroblasts to resist stress and to maintain normal biological function. The role of MSC-CM in regulating cellular biological functions is complex because it contains a variety of cytokines and active factors. In this study, we mainly considered the role of the TGF-β/Smad signaling pathway in regulating myofibroblast formation, but other signaling pathways may also be involved, such as Wnt/βcatenin and PI3K/Akt signaling pathways [50,51], which have been reported to be involved in MSCCM-regulated tissue repair.Therefore, MSC-CM may be a practical and feasible treatment for acute and chronic wound healing because it promotes fibroblast differentiation. The use of MSC-CM has more advantages compared with the use of stem cells: (i) it is much easier to produce and save in large quantities, (ii) there is no need to match human leukocyte antigen between donor and recipient, (iii) it is easy to handle for clinicians when applying to wounds compared with living

11

cells, and (iv) it is much safer than the use of stem cells because of the malignant transformation risk during in vitro expansion is avoided. However, there are issues that should be addressed before MSCCM is made available for clinical use. First is the methods of applying MSC-CM to the wound bed. Direct intradermal injection into the adjacent skin is a commonly used method for evaluating MSC-CMinduced wound repair [13,47,52]. In addition, cream or Matrigel-containing MSC-CM has also been used topically on skin wounds [53,54]. Incorporating proteins into tissue engineering scaffolds is an emerging and promising approach for wound repair [55]. This method can sustain and control protein release, thereby fully and effectively use of MSC-CM. Second, quality monitoring of the collected MSC-CM. Several studies show that the paracrine activity of MSCs changes during in vitro MSC expansion or induced senescence [56]. Senescent MSC-CM plays completely different roles—for example, limiting the regenerative potential of stem cells, impairing the anti-tumor activity and sensitizing young cells to senescence [57,58]. Moreover, secretome variability of MSCs in different donors, passages and culture condition was also noted [54,59]. Therefore, standard procedures for collection, production, identification and application methods for MSC-CM need to be defined. In summary, the current findings demonstrate that acute and chronic high-glucose environments had different effects on fibroblasts.The levels of intracellular ROS were correlated with markers of senescence and reduced function in fibroblasts. In addition, this study suggests that MSC-CM plays multifaceted roles in the wound-healing process. Incubating fibroblasts in MSCCM in a high-glucose environment promoted the function of fibroblasts, including suppressing intracellular ROS, decreasing fibroblast senescence and activating fibroblasts through the TGF-β/Smad signaling pathway. MSC-CM treatment may be a novel strategy for antiaging and an alternative therapeutic method for promoting wound healing in diabetic patients in the future. Acknowledgments This research was supported in part by the National Basic Science and Development Program (2012CB518103, 2012CB518105), the 863 Projects of Ministry of Science and Technology of China (2013AA020105 and 2012AA020502), the National Natural Science Foundation of China (81201479, 81121004 and 81230041), the Military Medical Foundation (AWS11J008), the Hainan Province Key Sciences and Technology Project (ZDZX2013003), and the China Postdoctoral Science Foundation Funded Project (2016M592929).

ARTICLE IN PRESS 12 M. Li et al. Disclosure of interests: The authors have no commercial, proprietary or financial interest in the products or companies described in this article. References [1] Wong SL, Demers M, Martinod K, Gallant M, Wang Y, Goldfine AB, et al. Diabetes primes neutrophils to undergo NETosis, which impairs wound healing. Nat Med 2015;21: 815–19. [2] Darby IA, Hewitson TD. Fibroblast differentiation in wound healing and fibrosis. Int Rev Cytol 2007;257:143–79. [3] Carthy JM, Sundqvist A, Heldin A, van Dam H, Kletsas D, Heldin CH, et al. Tamoxifen inhibits TGF-beta-mediated activation of myofibroblasts by blocking non-smad signaling through ERK1/2. J Cell Physiol 2015;230:3084–92. [4] Elliott CG, Forbes TL, Leask A, Hamilton DW. Inflammatory microenvironment and tumor necrosis factor alpha as modulators of periostin and CCN2 expression in human non-healing skin wounds and dermal fibroblasts. Matrix Biol 2015;43:71–84. [5] Wertheimer E, Spravchikov N, Trebicz M, Gartsbein M, Accili D, Avinoah I, et al. The regulation of skin proliferation and differentiation in the IR null mouse: implications for skin complications of diabetes. Endocrinology 2001;142: 1234–41. [6] Maeda M, Hayashi T, Mizuno N, Hattori Y, Kuzuya M. Intermittent high glucose implements stress-induced senescence in human vascular endothelial cells: role of superoxide production by NADPH oxidase. PLoS ONE 2015;10:e0123169. [7] Terashi H, Izumi K, Deveci M, Rhodes LM, Marcelo CL. High glucose inhibits human epidermal keratinocyte proliferation for cellular studies on diabetes mellitus. Int Wound J 2005;2:298–304. [8] Xuan YH, Huang BB, Tian HS, Chi LS, Duan YM, Wang X, et al. High-glucose inhibits human fibroblast cell migration in wound healing via repression of bFGF-regulating JNK phosphorylation. PLoS ONE 2014;9:e108182. [9] Arunachalam G, Samuel SM, Marei I, Ding H, Triggle CR. Metformin modulates hyperglycaemia-induced endothelial senescence and apoptosis through SIRT1. Br J Pharmacol 2014;171:523–35. [10] Chen F, Chen B, Xiao FQ, Wu YT, Wang RH, Sun ZW, et al. Autophagy protects against senescence and apoptosis via the RAS-mitochondria in high-glucose-induced endothelial cells. Cell Physiol Biochem 2014;33:1058–74. [11] Yamagishi S, Fukami K, Matsui T. Crosstalk between advanced glycation end products (AGEs)-receptor RAGE axis and dipeptidyl peptidase-4-incretin system in diabetic vascular complications. Cardiovasc Diabetol 2015;14:2. [12] Blazer S, Khankin E, Segev Y, Ofir R, Yalon-Hacohen M, Kra-Oz Z, et al. High glucose-induced replicative senescence: point of no return and effect of telomerase. Biochem Biophys Res Commun 2002;296:93–101. [13] Chen L, Xu Y, Zhao J, Zhang Z, Yang R, Xie J, et al. Conditioned medium from hypoxic bone marrow-derived mesenchymal stem cells enhances wound healing in mice. PLoS ONE 2014;9:e96161. [14] Hou C, Shen L, Huang Q, Mi J, Wu Y, Yang M, et al. The effect of heme oxygenase-1 complexed with collagen on MSC performance in the treatment of diabetic ischemic ulcer. Biomaterials 2013;34:112–20. [15] Bai L, Li D, Li J, Luo Z, Yu S, Cao S, et al. Bioactive molecules derived from umbilical cord mesenchymal stem cells. Acta Histochem 2016;118:761–9.

[16] Linero I, Chaparro O. Paracrine effect of mesenchymal stem cells derived from human adipose tissue in bone regeneration. PLoS ONE 2014;9:e107001. [17] Timmers L, Lim SK, Hoefer IE, Arslan F, Lai RC, van Oorschot AA, et al. Human mesenchymal stem cellconditioned medium improves cardiac function following myocardial infarction. Stem Cell Res 2011;6:206–14. [18] da Silva AF, Silva K, Reis LA, Teixeira VP, Schor N. Bone marrow-derived mesenchymal stem cells and their conditioned medium attenuate fibrosis in an irreversible model of unilateral ureteral obstruction. Cell Transplant 2015;24:2657–66. [19] Du T, Zou X, Cheng J, Wu S, Zhong L, Ju G, et al. Human Wharton’s jelly–derived mesenchymal stromal cells reduce renal fibrosis through induction of native and foreign hepatocyte growth factor synthesis in injured tubular epithelial cells. Stem Cell Res Ther 2013;4:59. [20] Faulknor RA, Olekson MA, Nativ NI, Ghodbane M, Gray AJ, Berthiaume F. Mesenchymal stromal cells reverse hypoxiamediated suppression of alpha-smooth muscle actin expression in human dermal fibroblasts. Biochem Biophys Res Commun 2015;458:8–13. [21] Li M, Zhao Y, Hao H, Dai H, Han Q, Tong C, et al. Mesenchymal stem cell-conditioned medium improves the proliferation and migration of keratinocytes in a diabetes-like microenvironment. Int J Low Extrem Wounds 2015;14:73–86. [22] Thomas K, Kiwit M, Kerner W. Glucose concentration in human subcutaneous adipose tissue: comparison between forearm and abdomen. Exp Clin Endocrinol Diabetes 1998;106:465–9. [23] Jeon YK, Jang YH, Yoo DR, Kim SN, Lee SK, Nam MJ. Mesenchymal stem cells’ interaction with skin: wound-healing effect on fibroblast cells and skin tissue. Wound Repair Regen 2010;18:655–61. [24] Schreier T, Degen E, Baschong W. Fibroblast migration and proliferation during in vitro wound healing. A quantitative comparison between various growth factors and a low molecular weight blood dialysate used in the clinic to normalize impaired wound healing. Res Exp Med (Berl) 1993;193:195– 205. [25] Lamers ML, Almeida ME, Vicente-Manzanares M, Horwitz AF, Santos MF. High glucose-mediated oxidative stress impairs cell migration. PLoS ONE 2011;6:e22865. [26] Hayflick L. Living forever and dying in the attempt. Exp Gerontol 2003;38:1231–41. [27] Schiller M, Javelaud D, Mauviel A. TGF-beta-induced SMAD signaling and gene regulation: consequences for extracellular matrix remodeling and wound healing. J Dermatol Sci 2004;35:83–92. [28] Mehanna RA, Nabil I, Attia N, Bary AA, Razek KA, Ahmed TA, et al. The effect of bone marrow-derived mesenchymal stem cells and their conditioned media topically delivered in fibrin glue on chronic wound healing in rats. Biomed Res Int 2015;2015:846062. [29] Tiwari P. Recent trends in therapeutic approaches for diabetes management: a comprehensive update. J Diabetes Res 2015;2015:340838. [30] Barrientos S, Stojadinovic O, Golinko MS, Brem H, Tomic-Canic M. Growth factors and cytokines in wound healing. Wound Repair Regen 2008;16:585–601. [31] Werner S, Grose R. Regulation of wound healing by growth factors and cytokines. Physiol Rev 2003;83:835–70. [32] Biernacka A, Dobaczewski M, Frangogiannis NG. TGF-beta signaling in fibrosis. Growth Factors 2011;29:196–202. [33] Finnson KW, Arany PR, Philip A. Transforming growth factor beta signaling in cutaneous wound healing: lessons learned from animal studies. Adv Wound Care (New Rochelle) 2013;2:225–37.

ARTICLE IN PRESS UC MSC medium exerts in vitro antiaging effects [34] He T, Guan X, Wang S, Xiao T, Yang K, Xu X, et al. Resveratrol prevents high glucose-induced epithelialmesenchymal transition in renal tubular epithelial cells by inhibiting NADPH oxidase/ROS/ERK pathway. Mol Cell Endocrinol 2015;402:13–20. [35] Kanikarla-Marie P, Jain SK. L-Cysteine supplementation reduces high-glucose and ketone-induced adhesion of monocytes to endothelial cells by inhibiting ROS. Mol Cell Biochem 2014;391:251–6. [36] Chang J, Li Y, Huang Y, Lam KS, Hoo RL, Wong WT, et al. Adiponectin prevents diabetic premature senescence of endothelial progenitor cells and promotes endothelial repair by suppressing the p38 MAP kinase/p16INK4A signaling pathway. Diabetes 2010;59:2949–59. [37] Sasaki M, Kajiya H, Ozeki S, Okabe K, Ikebe T. Reactive oxygen species promotes cellular senescence in normal human epidermal keratinocytes through epigenetic regulation of p16(INK4a.). Biochem Biophys Res Commun 2014;452: 622–8. [38] Zhuge CC, Xu JY, Zhang J, Li W, Li P, Li Z, et al. Fullerenol protects retinal pigment epithelial cells from oxidative stressinduced premature senescence via activating SIRT1. Invest Ophthalmol Vis Sci 2014;55:4628–38. [39] Burton DG, Krizhanovsky V. Physiological and pathological consequences of cellular senescence. Cell Mol Life Sci 2014;71:4373–86. [40] Olivieri F, Bonafe M, Spazzafumo L, Gobbi M, Prattichizzo F, Recchioni R, et al. Age- and glycemia-related miR-126-3p levels in plasma and endothelial cells. Aging 2014;6:771–87. [41] Kong JG, Park JB, Lee D, Park EY. Effect of high glucose on stress-induced senescence of nucleus pulposus cells of adult rats. Asian Spine J 2015;9:155–61. [42] He T, Quan T, Shao Y, Voorhees JJ, Fisher GJ. Oxidative exposure impairs TGF-beta pathway via reduction of type II receptor and SMAD3 in human skin fibroblasts. Age (Dordr) 2014;36:9623. [43] Landry WD, Cotter TG. ROS signalling, NADPH oxidases and cancer. Biochem Soc Trans 2014;42:934–8. [44] James EL, Michalek RD, Pitiyage GN, de Castro AM, Vignola KS, Jones J, et al. Senescent human fibroblasts show increased glycolysis and redox homeostasis with extracellular metabolomes that overlap with those of irreparable DNA damage, aging, and disease. J Proteome Res 2015;14:1854–71. [45] Lim H, Park H, Kim HP. Effects of flavonoids on senescenceassociated secretory phenotype formation from bleomycininduced senescence in BJ fibroblasts. Biochem Pharmacol 2015;96:337–48. [46] Zhu Y, Armstrong JL, Tchkonia T, Kirkland JL. Cellular senescence and the senescent secretory phenotype in agerelated chronic diseases. Curr Opin Clin Nutr Metab Care 2014;17:324–8. [47] Fong CY, Tam K, Cheyyatraivendran S, Gan SU, Gauthaman K, Armugam A, et al. Human Wharton’s jelly stem cells and its conditioned medium enhance healing of excisional and diabetic wounds. J Cell Biochem 2014;115:290–302.

13

[48] Shen C, Lie P, Miao T, Yu M, Lu Q, Feng T, et al. Conditioned medium from umbilical cord mesenchymal stem cells induces migration and angiogenesis. Mol Med Rep 2015;12:20–30. [49] Walter MN, Wright KT, Fuller HR, MacNeil S, Johnson WE. Mesenchymal stem cell-conditioned medium accelerates skin wound healing: an in vitro study of fibroblast and keratinocyte scratch assays. Exp Cell Res 2010;316:1271–81. [50] Jin F, Qu X, Fan Q, Wang L, Tang T, Hao Y, et al. Regulation of prostate cancer cell migration toward bone marrow stromal cell-conditioned medium by Wnt5a signaling. Mol Med Rep 2013;8:1486–92. [51] Jun EK, Zhang Q, Yoon BS, Moon JH, Lee G, Park G, et al. Hypoxic conditioned medium from human amniotic fluidderived mesenchymal stem cells accelerates skin wound healing through TGF-beta/SMAD2 and PI3K/Akt pathways. Int J Mol Sci 2014;15:605–28. [52] Li M, Luan F, Zhao Y, Hao H, Liu J, Dong L, et al. Mesenchymal stem cell-conditioned medium accelerates wound healing with fewer scars. Int Wound J 2016; doi:10.1111/iwj.12551. In press. [53] Arno AI, Amini-Nik S, Blit PH, Al-Shehab M, Belo C, Herer E, et al. Human Wharton’s jelly mesenchymal stem cells promote skin wound healing through paracrine signaling. Stem Cell Res Ther 2014;5:28. [54] Kusindarta DL, Wihadmadyatami H, Fibrianto YH, Nugroho WS, Susetya H, Musana DK, et al. Human umbilical mesenchymal stem cells conditioned medium promote primary wound healing regeneration. Vet World 2016;9:605–10. [55] Hou Y, Hu J, Park H, Lee M. Chitosan-based nanoparticles as a sustained protein release carrier for tissue engineering applications. J Biomed Mater Res A 2012;100:939–47. [56] Ozcan S, Alessio N, Acar MB, Mert E, Omerli F, Peluso G, et al. Unbiased analysis of senescence associated secretory phenotype (SASP) to identify common components following different genotoxic stresses. Aging 2016;8:1316–29. [57] Ozcan S, Alessio N, Acar MB, Toprak G, Gonen ZB, Peluso G, et al. Myeloma cells can corrupt senescent mesenchymal stromal cells and impair their anti-tumor activity. Oncotarget 2015;6:39482–92. [58] Severino V, Alessio N, Farina A, Sandomenico A, Cipollaro M, Peluso G, et al. Insulin-like growth factor binding proteins 4 and 7 released by senescent cells promote premature senescence in mesenchymal stem cells. Cell Death Dis 2013;4:e911. [59] Zhukareva V, Obrocka M, Houle JD, Fischer I, Neuhuber B. Secretion profile of human bone marrow stromal cells: donor variability and response to inflammatory stimuli. Cytokine 2010;50:317–21.

Appendix: Supplementary material Supplementary data to this article can be found online at doi:10.1016/j.jcyt.2016.12.001.