If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
School of Biomedical Engineering (Suzhou), Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, Anhui, ChinaSuzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, Jiangsu, China
School of Biomedical Engineering (Suzhou), Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, Anhui, ChinaSuzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, Jiangsu, China
School of Biomedical Engineering (Suzhou), Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, Anhui, ChinaSuzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, Jiangsu, China
School of Biomedical Engineering (Suzhou), Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, Anhui, ChinaSuzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, Jiangsu, China
School of Biomedical Engineering (Suzhou), Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, Anhui, ChinaSuzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, Jiangsu, ChinaXuzhou Medical University, Xuzhou, Jiangsu, ChinaZhengzhou Zhongke Academy of Biomedical Engineering and Technology, Zhengzhou, Henan, China
School of Biomedical Engineering (Suzhou), Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, Anhui, ChinaSuzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, Jiangsu, ChinaXuzhou Medical University, Xuzhou, Jiangsu, China
MSCs in FG scaffold (MSC/FG) promoted VEGF-mediated angiogenesis in diabetic wounds in an Akt/mTOR-dependent way.
•
Metformin exerted dose-dependent effects on MSCs-promoted Akt/mTOR activation.
•
Metformin impaired the survival but not wound healing effects of MSCs/FG in diabetic mice.
•
Metformin coordinated with MSCs/FG to promote VEGF-mediated angiogenesis in diabetic wound.
Abstract
Introduction
Cell therapy with mesenchymal stem cells (MSCs) and biomaterials holds great potential for the treatment of diabetic ulceration; however, the underlying mechanism as well as its compatibility with the first-line anti-diabetic drug, metformin (MTF), has not been well elucidated.
Methods
MSCs derived from the umbilical cord were labeled with fluorescent proteins, followed by transplantation in a fibrin scaffold (MSCs/FG) onto the STZ-induced diabetic wound in a C57BL6/J mouse model. MTF was administered by oral gavage at a dose of 250 mg/kg/day. The wound healing rate, epithelization, angiogenesis, and underlying mechanism were evaluated in MSCs/FG- and MTF-treated diabetic wounds. Moreover, the dose-dependent effects of MTF and involvement of the Akt/mTOR pathway were analyzed in keratinocyte and fibroblast cultures.
Results
MSCs/FG significantly promoted angiogenesis in diabetic wound healing without signs of differentiation or integration. The recruitment of fibroblasts and keratinocytes by MSCs/FG promotes migration and vascular endothelial growth factor (VEGF) expression in an Akt/mTOR-dependent manner. MTF, which is generally considered a mTOR inhibitor, displayed dose-dependent effects on MSC-unregulated Akt/mTOR and VEGF expression. Oral administration of MTF at an anti-diabetic dosage synergistically acted with MSCs/FG to promote Akt/mTOR activation, VEGF expression, and subsequent angiogenesis in diabetic wounds; however, it reduced the survival of MSCs.
Conclusions
Our study identifies that MTF coordinates with mesenchymal cells to promote Akt/mTOR activation and VEGF-mediated angiogenesis during diabetic wound healing. These findings offer new insights into MSCs engraftment in FG scaffolds for diabetic wound healing and provide support for the promotion of MSCs therapy in patients prescribed with MTF.
As one of the most common complications of diabetes mellitus (DM), diabetic foot ulcer (DFU) lacks any definitive healing protocol and remains a leading cause of non-traumatic limb amputations [
]; MSCs can be collected from the bone marrow, umbilical cord, placenta, and adipose tissue, and have been proven to promote healing in all phases of wound repair. Many clinical trials have used MSCs for the treatment of various diseases because of their ease of access, high proliferation, and potent paracrine capacities [
]. Owing to extensive cell death in the unfavorable milieu of ischemic diabetic wounds, MSCs are often applied with biomaterial scaffolds to favor survival [
Topical administration of allogeneic mesenchymal stromal cells seeded in a collagen scaffold augments wound healing and increases angiogenesis in the diabetic rabbit ulcer.
]. Fibrin (FG) is a biocompatible, non-cytotoxic, and naturally biodegradable carrier that has been used as a delivery system for tissue regeneration [
]. However, the cell fate and underlying mechanism of MSCs loaded in the FG scaffold (MSCs/FG) for diabetic wound healing needs to be further elucidated.
MTF is the first-line and most commonly prescribed anti-diabetic drug [
]. Its anti-diabetic effect is primarily due to the inhibition of hepatic gluconeogenesis through the activation of adenosine monophosphate-activated protein kinase (AMPK) and the subsequent inhibition of mammalian target of rapamycin (mTOR) [
]. Recently, numerous studies have shown that MTF exhibits anti-tumor and anti-aging effects, which are also dependent on the inhibition of mTOR activation through multiple signaling pathways [
Metformin induces cell cycle arrest, reduced proliferation, wound healing impairment in vivo and is associated to clinical outcomes in diabetic foot ulcer patients.
]. Considering that patients with DFU are often prescribed MTF, which is generally considered a mTOR inhibitor, it is important to clarify whether a clinically relevant dosage of MTF interferes with MSCs-promoted diabetic wound-healing effects.
To further explore the underlying mechanism of MSC therapy as well as its compatibility with MTF in terms of diabetic wound healing, the healing rate, epithelization, and angiogenesis, as well as the involved signaling pathways, were evaluated in MSCs/FG- or MTF-treated diabetic wounds. By labeling MSCs derived from the umbilical cord with lentivirus carrying green fluorescent protein (GFP), we found that MSCs/FG significantly promoted wound closure in a streptozotocin (STZ)-induced diabetic mouse model, without signs of differentiation and migration. The recruitment of fibroblasts and keratinocytes to promote migration and angiogenesis is dependent on activation of the protein kinase B (Akt)/ mTOR pathway. MTF showed dose-dependent inhibitory effects on Akt/mTOR activation. More importantly, the clinically relevant anti-diabetic dosage of MTF synergistically acted with MSCs/FG therapy to further enhance vascular endothelial growth factor (VEGF) expression and angiogenesis in diabetic wound healing. These data will be helpful in understanding the underlying mechanism of FG-based MSCs therapy for diabetic wound healing and provide insights for the clinical application of MSCs therapy in patients with DFU prescribed MTF.
2. Materials and methods
2.1 Preparation of the primary MSCs, keratinocytes, and fibroblasts culture
All studies involving human samples were performed in accordance with the ethical guiding principles of human embryonic stem cell research and the Declaration of Helsinki. Umbilical cord and skin samples were donated with signed informed consent and ethical approval from the First Affiliated Hospital of Soochow University (2019-136). MSCs were isolated and cultured as previously described [
]. Briefly, small pieces of Wharton's jelly were inoculated into Dulbecco's modified Eagle medium (DMEM)/F12 medium (D/F; Gibco; C11330500BT) containing 10 % fetal bovine serum (FBS; Gibco; 10099-141) and 1 % penicillin and streptomycin (P/S; Solarbio; P1400).
For human keratinocytes and fibroblasts preparation, the skin samples were incubated with 2 mg/mL dispase (Roche; 04942078001) overnight at 4 °C to separate the epidermis and dermis. Minced epidermis and dermis were further processed and cultured as described for keratinocyte [
To prepare the conditioned medium (CM), MSCs (P3-P5) were washed and re-fed with serum-free D/F medium, and 48 h later, the medium was collected and centrifuged at 2000 ×g for 5 min. The supernatant was then harvested and stored at −80 °C. D/F medium incubated without MSCs for 48 h was considered as the control (Ctrl).
Keratinocytes or fibroblasts were washed and incubated with D/F medium containing 50 % CM or Ctrl for 48 h. Rapamycin (rapa; 1 μM; Abmole, M1768) or MTF (0.03 mM, 0.3 mM, 0.6 mM, 1.0 mM, 2.0 mM, or 3.0 mM; Abmole; M2049) was added before treatment with CM or Ctrl.
2.3 Scratch assay
The migration of fibroblasts and keratinocytes was evaluated using the scratch assay. The cells were scratched with a 200 μL pipette tip in 6-well plates, and then changed to D/F medium containing 50 % CM or Ctrl. After different incubation times, the cells were fixed, stained with 4′,6-diamidino-2-phenylindole (DAPI), and scratches were recorded under a fluorescent microscope (100× magnification; VOVEL, NIB900). The uncovered area was analyzed using Image J (NIH, USA), and the migration rates were calculated as follows: migration rate = % (initial area − uncovered area) / initial area.
2.4 Lentivirus production and preparation of GFP-labeled MSCs
The enhanced GFP (EGFP) lentivirus was generated in human embryonic kidney (HEK) 293 T cells co-transfected with the pLVX-CMV-EGFP lentiviral vector (Miaoling, China) and three package plasmids [
]. The viral particles were concentrated by ultracentrifugation in a Thermo LYNX-6000 centrifuge with an A27-8 50 rotor at 50,000 ×g for 120 min at 4 °C. The pellet was gently resuspended in PBS, aliquoted, and stored at −80 °C.
MSCs at ∼70 % confluence were infected with EGFP lentivirus at MOI = 1 in D/F medium supplemented with 8 μg/mL of polybrene (Yeasen; 40804ES) for 5–6 h, and replaced with D/F medium containing 10 % FBS for another 24 h. Subsequently, 48 h later, direct GFP fluorescence and GFP immunofluorescence staining were used to determine the efficiency of GFP labeling in MSCs. Moreover, cell counting kit-8 (CCK8; CK04, Dojindo) assay and Ki67 immunostaining were used to evaluate the proliferative activity of the labeled MSCs. GFP-labeled MSCs were used for subsequent transplantation experiments when the average GFP-labeling efficiency was over 70 %.
2.5 Preparation of the diabetic mice wound model
All animal experiments were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the Biological Research Ethics Committee of the Chinese Academy of Sciences (ref. 2018-A21). The mice were maintained in a specific pathogen-free environment under a 12/12 h light-dark cycle. Adult male C57BL/6 J mice (7–8 weeks) (SPF Biotechnology, China) were used as diabetic models as described previously [
]. Briefly, mice were fed a 60 kcal% fat diet (FBSH, China) for 1-week, followed by an intraperitoneal injection of streptozotocin (STZ; 50 mg/kg/day; Cayman, USA; 13,104) for 5 days (Fig. 1A). Blood glucose levels were monitored every fifth day after STZ injection from the tail vein using a glucometer (Accu-Chek Performa, Roche Diagnostics, USA). Mice with fed blood glucose ≥16.7 mM (300 mg/L) for 2 weeks were considered diabetic and selected for skin wounds [
]. The mice were 12–13 weeks old at the start of wounding.
Fig. 1MSCs in FG scaffold (MSCs/FG) promote wound closure and re-epithelization in STZ-induced diabetic mice. (A) Schematic diagram of the experimental design studying the effects of MSCs/FG on diabetic wounds. Mice (7–8 weeks) were injected with STZ (50 mg/kg/day, dissolved in citrate buffer, pH 4.5) for 5 days after being fed with high-fat diet for 1 week to prepare diabetic models. Mice with fed blood glucose ≥16.7 mM (300 mg/L) for 2 weeks were considered to be diabetic and selected for creating skin wounds. Mice were 12–13 weeks old at the start of wounding. Two full-thickness skin wounds were made on the back of each mouse. PBS, FG, MSCs, or MSCs/FG were randomly assigned to the wounds on the day of surgery (D0) and 7 days (D7) after the surgery. The wound size was monitored, and the mice were sacrificed for the subsequent analysis on D3, D7, and D14 after the surgery. (B) The blood glucose levels were monitored every fifth day after STZ or vehicle (citrate buffer, pH 4.5) injection. Diabetes was defined as fed blood glucose ≥16.7 mM (indicated by blue dotted line). STZ-treated mice became diabetic 5 days after treatment. ***p < 0.001 at all time points between STZ-treated (n = 36) and vehicle-treated (n = 10) mice, two-tailed unpaired t-test. (C) Representative images showing the dynamic changes in the wounds when treated with PBS, FG, MSCs, or MSCs/FG in diabetic wound models. (D) MSCs/FG treatment significantly reduced the wound area in diabetic mice model at different time points, as compared to the PBS, FG, or MSCs treatment. The wound healing effects were evaluated by the dynamic change in wound area percentage (%), that is, the wound area at different time points compared to the corresponding starting area. D3 and D7: n = 10, 11, 8 and 9 for PBS, FG, MSCs and MSCs/FG, respectively; D14: n = 7, 8, 5 and 6 for PBS, FG, MSCs and MSCs/FG, respectively. One-way ANOVA followed by Tukey's post-hoc test. (E) Representative images showing immunostaining against Keratin 14 (K14), a marker of the basal keratinocytes of epidermis, of the regenerated tissues 14 days after PBS, FG, MSCs, or MSCs/FG treatment. DAPI was used for counterstaining. The area in the yellow dashed box was enlarged and shown in the lower panel. Epithelial gap (EG) indicated that the wound area has not been covered by K14 epidermis. (F) Statistical analysis revealed that the EG was significantly smaller in the MSCs/FG-treated diabetic wounds than that of the PBS, FG, or MSCs-treated wounds. n = 4 for PBS, FG and MSCs, n = 5 for MSCs/FG. One-way ANOVA followed by Tukey's post-hoc test. Scale bar: 2 mm in the upper panel and 200 μm in the lower ones of E. All data are presented as mean ± SEM. ⁎p < 0.05, ⁎⁎p < 0.01, ⁎⁎⁎p < 0.001. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
To avoid the effect of mouse skin contraction, we developed a two-layer patch method (Fig. S1) for wound preparation and subsequent transplantation experiments. Mice were depilated, anesthetized with 3 % inhaled isoflurane (RWD, China), and coated with a 3 × 2 cm antibacterial film (Drape Antimicrob, USA; REF6640) on the back skin. Two full-thickness wounds (Φ = 8 mm) were created along the midline of the dorsum by using a skin biopsy instrument (Honglong, China) within the film area of each mouse. Two wounds were cut at a distance of at least 1 cm to minimize interference between wounds. After the engrafting experiment, the wounds were covered with another 3 × 2 cm film to maintain sterility and prevent skin contraction. Each mouse was raised in a separate cage after the operation. Each wound was considered an experimental unit to avoid individual and locational differences.
2.6 Animal treatment and wound area evaluation
All diabetic wound mice (n = 36) were divided randomly (RAND; Microsoft Excel) into two groups. One group was treated with MTF (Beijing Jingfeng, China) at 250 mg/kg/day by oral gavage from the day of wound preparation (D0) until sacrifice (D14) (n = 11), and the other group received gavage with drinking water (n = 25). The MTF dose was calculated according to the human dose based on body surface area conversion [
]. MTF was dissolved in drinking water and administered to the mice using gavage feeding needles (Beijing Heli Kechuang, China; HL-GWZ-8).
The wounds of the mice receiving either MTF or drinking water were further randomly (RAND; Microsoft Excel) divided into four groups, treated with either PBS (20 μL), FG (10 μL of 10 mg/mL fibrinogen (Solarbio, China; F8051) mixed with 10 μL of 25 U/mL thrombin (Solarbio, China; T8021)), MSCs (106 in 20 μL of PBS), or MSCs/FG (106 MSCs in 10 μL of fibrinogen, mixed with 10 μL of thrombin). Each wound was treated with same engrafting materials on the day of wound preparation (D0) and D7 to ensure the therapeutic effects. All wounds (n = 72) were divided into eight groups: PBS (n = 13), FG (n = 14), MSCs (n = 11), MSCs/FG (n = 12), PBS + MTF (n = 5), FG + MTF (n = 6), MSCs + MTF (n = 5), and MSCs/FG + MTF (n = 6). Reagent administration was performed carefully within one wound area without splashing or flowing to the other. After cell transplantation, cyclophosphamide (50 mg/kg/day; Sigma, USA; C0768) was administered intraperitoneally for three days to minimize the immune response to xenografting.
The wound size was recorded with a digital camera at the vertical angle of view on D0, D3, D7 (before engraftment), and D14 after the treatment and calculated using Image J software version 1.8.0 with a unified calibration.
2.7 Immunohistochemistry
Mice were sacrificed by CO2 inhalation; the wound specimens (Φ = 8 mm) containing the wound bed and regenerated surrounding skin tissue were excised and cut in half along the midline. The right half was used for the subsequent total RNA or protein preparation. The left half was fixed with paraformaldehyde (4 %) and paraffin-sectioned over the wound tissue (6 μm thickness). Sections were processed for hematoxylin and eosin (H&E) staining or immunostaining using the primary antibodies listed in Table S1. Immunoreactivity was visualized using the appropriate Alexa Fluor-conjugated secondary antibodies and observed using a confocal microscope (Nikon, Japan, A1R HD25). Results were derived from the evaluation of cells or structures in randomly selected fields within a defined center or transitional area of the wound (400× magnification; 2 microscopic fields per wound). The survival of GFP-labeled MSCs was evaluated by analyzing the GFP fluorescence intensity (100× magnification) or GFP+ cells (400× magnification) in randomly chosen fields.
2.8 Quantitative PCR (q-PCR)
Total RNA from wound skin samples (right half) or cells receiving various treatments was harvested with Trizol (Thermo, 15596018) and quantified using NanoDrop2000 (Thermo). qPCR was performed on a Bio-Rad CFX96 PCR System using TB Green Premix Taq (TAKARA; RR042B) and appropriate primers (Table S2). The housekeeping gene GAPDH was used as an internal control.
2.9 Western blot
Wound tissue (right half) or cells were lysed in cell lysis buffer (Beyotime, Shanghai, China; P0013). After determining the protein concentration (Lowry method), samples were separated on 10 %–15 % polyacrylamide gels and transferred to polyvinylidene fluoride (PVDF) membranes. Membranes were blocked and incubated overnight with the primary antibodies listed in Table S1. Specific protein bands were visualized using enhanced chemiluminescence (GE Life Sciences). The bands were quantified using Image Pro Plus software (Media Cybernetics, USA) after subtraction of the local background.
2.10 Statistical analysis
Numerical data are presented as mean ± SEM. The data were subjected to unpaired t-tests or analysis of variance and an appropriate post-hoc analysis using GraphPad Prism version 8.0 (GraphPad Software, USA). The level of significance was set at p < 0.05. All the sample sizes and statistical information are listed in Table S3.
3. Results
3.1 MSCs/FG promoted diabetic wound healing with insignificant migration and integration
The effect of MSCs/FG on full-thickness wounds in STZ-treated mice was evaluated according to the experimental paradigm shown in Fig. 1A. STZ-treated mice exhibited a persistently high blood glucose level (≥16.7 mM), indicating successful establishment of the diabetic model (Fig. 1B). Compared to the PBS-, FG-, and MSC-treated counterparts, MSCs/FG-treated diabetic mice exhibited significantly smaller wound areas on the 3rd, 7th, and 14th days after treatment (Fig. 1C and D). Furthermore, immunostaining against cytokeratin 14 (K14), a marker of epithelial cells, revealed that MSCs/FG application led to the smallest epithelial gap (EG) among the four treatment groups (Fig. 1E and F). H&E staining also clearly showed that the epidermis migrated from the edge to the center of the wound area in all four groups, and the EG in the MSCs/FG group was the smallest (Fig. S2).
The biological activity of MSCs in the presence of fibrin has also been evaluated. Flow cytometry showed that MSCs and MSCs/FG highly expressed markers for mesenchymal cells, such as CD73, CD90, and CD105, whereas few of them expressed hematopoietic markers CD34, CD45, and HLA-DR (Fig. 2A–B ). Both MSCs and MSCs/FG differentiated into osteogenic cells (Fig. S3A) or adipogenic (Fig. S3B) lineages under specific induction conditions. The CCK8 assay revealed that the cell numbers did not differ significantly when MSCs were cultured with or without FG (Fig. 2C). Altogether, FG showed good biocompatibility, exhibiting no interference with the surface antigen expression, differentiation, and proliferation capacity of MSCs.
Fig. 2MSCs in FG scaffold significantly retains at the wound site, showing no signs of migration or differentiation. (A–B) Flow cytometry analysis of MSCs (A) and MSCs incubated with FG leaching solution (MSCs in FG) (B) for 48 h. 105 cells were fixed, stained with antibodies conjugated with different fluorescent dyes, and analyzed with BD LSR Fortessa. The blue peak represents cells stained positive for the corresponding antibodies, while the red peak represents the isotype control. Note that most of the cells were immunoreactive for CD73, CD90, and CD105 (markers of the mesenchymal cells), but not for HLA-DR (marker of the leucocytes) and CD34 and CD45 (markers of the endothelial and hematopoietic cells) in MSCs and MSCs/FG populations. The average percentage of the positive population is shown in each histogram. (C) No significant difference was observed in the cell number between MSCs and MSCs incubated with FG leaching solution (MSCs in FG) at different points. n = 9 for MSCs and MSCs in FG. Two-tailed unpaired t-test. (D) Statistical analysis of the GFP fluorescence intensity revealed that MSCs transplanted in FG scaffold was significantly retained at the wound site at each time point, as compared to the MSCs transplantation. n = 3 for MSCs and MSCs/FG. Two-tailed unpaired t-test. (E–F) 106 GFP-labeled MSCs (E) or MSCs in FG scaffold (MSCs/FG; F) were engrafted onto the diabetic wound, and the survival was evaluated by using double staining against GFP and K14 on D3, D7, and D14 after transplantation. In the diabetic mice transplanted with MSCs alone, few GFP+ MSCs were observed on D3, while they were untraceable on D7 and D14 (E). In comparison, more GFP+ MSCs were observed at the wound site in MSCs/FG-treated group at each time points, which was gradually covered by the migrating K14+ epithelial tongue (F). There were no signs of migration into the deep tissue (F). Images in the yellow dotted box are amplified and shown in the magnified right panel. (G–I) Representative images of double staining against GFP and αSMA (G), CD34 (H), or vimentin (I), showing that the diabetic wound was treated with GFP-labeled MSCs/FG for 14 days, and subjected to immunostaining for GFP and αSMA, CD34, or vimentin. Images in the yellow dotted box are amplified and shown as insets. The yellow, white, and red arrowheads indicate the GFP-labeled MSCs, the αSMA+, or CD34+ vessel-like structures, respectively. GFP+ cells did not co-label with αSMA and CD34, but with vimentin. Scale bar: 500 μm in E-F and 100 μm in the magnified right panel; 100 μm in G–H. All data are presented as mean ± SEM. ⁎p < 0.05. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
The migration and integration of the transplanted MSCs in the diabetic wound healing process were further studied by genetically labeling MSCs with lentivirus encoding GFP (Fig. S4). The efficiency of GFP labeling in MSCs was ∼70 %, and labeling showed insignificant effects on the proliferative activity of MSCs (Fig. S4B–C). Double staining for K14 and GFP in diabetic wounds 3, 7, and 14 days after MSCs or MSCs/FG treatment revealed that the transplanted cells diminished rapidly with time in both groups, whereas more labeled MSCs were retained in the wound area in the MSCs/FG group at each time point (Fig. 2E–F). Only sparse GFP+ cells could be observed 3 days after MSCs engraftment; in comparison, there were ∼5 fold of transplanted MSCs residing throughout the MSCs/FG-treated diabetic wound with a segment underneath the migrating epithelial tongue (Fig. 2D). On the 14th day, the surviving cells were also rare in the MSCs/FG group; they were clustered under the K14+ epidermis without obvious signs of migration or diffusion.
The phenotypes of MSCs/FG were characterized after being applied to diabetic wounds for 14 days. GFP-labeled MSCs were negative for either K14 (Fig. 2F) or CD34 (Fig. 2G), a marker of endothelial cells, or α-smooth muscle actin (α-SMA) (Fig. 2H), a marker of pericytes. Most GFP+ MSCs stained for vimentin (Fig. 2I), a marker of mesenchymal-derived cells. Taken together, these data indicate that MSCs in the FG scaffold contribute to diabetic wound healing through a paracrine mechanism rather than direct incorporation into the wound tissue.
3.2 MSCs/FG promoted angiogenesis and activation of Akt/mTOR-VEGF pathway in diabetic wound
The vascular insufficiency is closely related with the onset and prognosis of DFU [
], MSCs/FG treatment was shown to significantly increase the number of CD34+ and α-SMA+ vessels in the transitional area of the diabetic wound, as revealed by double staining against CD34 and α-SMA. Note that the CD34+ and α-SMA+ cells aggregated to develop a ring-shaped structure, confirming the formation of vasculature in the MSC/FG-treated diabetic wound (Fig. 3A and C ). There were CD34+ endothelial cells but not α-SMA+ pericytes in the central area (Fig. 3A and B), reflecting spatial-temporal angiogenesis in the regenerated tissues. In comparison, few CD34+ or α-SMA+ cells were found in the PBS or FG groups, and only a slight increase in the number of positive cells was observed in the MSCs group (Fig. 3B and C). In line with these observations, western blot (WB) analysis showed that the expression levels of CD31 and CD34, markers of vascular endothelial cells, were significantly upregulated in the MSC/FG-treated wound tissue compared to all other groups (Fig. 3D and E).
Fig. 3MSCs/FG treatment significantly promotes angiogenesis as well as Akt/mTOR activation in the diabetic wound. (A) Representative immunofluorescence staining against CD34 and α-SMA in the center area (W) and the transitional area (E) of the wounds treated with PBS, FG, MSCs, or MSCs/FG for 14 days. The arrows indicate the CD34+ vessel-like structure in the center area of the wound; the arrowheads indicate the vessels positive for both CD34 and α-SMA in the transitional area. There were no α-SMA+ cells in the center area of the diabetic wound in all four groups. (B–C) Statistical analysis revealed that MSCs/FG results in a significant increase in vessel-like structures positive for CD34 (B) or CD34 and α-SMA (C) in the central area or the transitional area of wound, respectively. n = 8 (2 fields/wound, 4 wounds/group). One-way ANOVA followed by Tukey's post-hoc test. (D) Representative WB images showing the CD31 and CD34 expression levels in wounds treated with PBS, FG, MSCs, or MSCs/FG. β-actin was used as an internal control. (E) Semi-quantitative results of the CD31 and CD34 immunoblotting analysis. n = 3 biological replicates. One-way ANOVA followed by Tukey's post-hoc test. (F) Quantitative PCR results showing the mRNA levels of VEGF-A in wounds treated with PBS, FG, MSCs, or MSCs/FG. n = 9 (3 replicates/wound, 3 wounds/group). One-way ANOVA followed by Tukey's post-hoc test. (G-H) Representative WB images showing the VEGF-A expression level (G) and p-Akt (S473), p-mTOR (S2448), Akt, and mTOR expression levels (H) in diabetic wounds treated with PBS, FG, MSCs, or MSCs/FG. GAPDH or β-actin was used as an internal control. (I) Semi-quantitative immunoblotting analysis of VEGF-A, p-Akt, and p-mTOR. Note that MSCs/FG treatment significantly increased the expression levels of VEGF-A, p-Akt, and p-mTOR when compared to PBS, FG, and MSCs treatment. All the experiments were performed 14 days after the PBS, FG, MSCs, or MSCs/FG engraftment. n = 3 biological replicates. One-way ANOVA followed by Tukey's post-hoc test. Scale bar: 100 μm in A. Data are presented as the mean ± SEM. ⁎⁎p < 0.01, ⁎⁎⁎p < 0.001.
Next, we analyzed the underlying mechanism of the angiogenesis-promoting effects of MSC/FG treatment. Vascular endothelial growth factor-A (VEGF-A), the main isoform of VEGF in wounds, promotes the outgrowth of blood vessels by directly binding to its receptors on endothelial cells [
]. In agreement with the observation that MSCs/FG treatment promoted angiogenesis, we found a significant increase in both VEGF-A mRNA (Fig. 3F) and VEGF-A protein (Fig. 3G and I) levels in MSCs/FG-treated wounds, as compared to PBS-and FG- MSC-treated groups.
Akt and mTOR pathways play a critical role in the regulation of VEGF expression and angiogenesis [
]. It is noteworthy that the markers of vascular endothelial cells, CD31 and CD34, the expression level of VEGF-A, as well as the Akt and mTOR pathways, were all significantly impaired in the diabetic wound compared to the normal wound (Fig. S5A–F). WB analysis showed that the Akt and mTOR pathways were impaired in diabetic wounds compared to those in normal wound on the 14th day (Fig. S5C and D). MSCs/FG treatment significantly increased the phosphorylation levels of Akt (p-Akt) and mTOR (p-mTOR) in diabetic wounds compared to PBS-, FG-, or MSC-treated wounds (Fig. 3H and I), indicating the significant activation of the Akt/mTOR signaling pathway in MSC-promoted diabetic wound healing.
3.3 MSCs promoted VEGFA expression and migration through an Akt/mTOR-dependent way
We next investigated whether the two main cell types in the wound, keratinocytes and fibroblasts, contributed to the re-epithelization and angiogenesis of MSC-induced diabetic wound healing. WB results revealed that CM of MSCs activated the Akt/mTOR pathway by upregulating p-Akt and p-mTOR levels in both fibroblasts and keratinocytes (Fig. 4A and B ). The total levels of Akt and mTOR remained unchanged in both cell types. CM significantly promoted VEGF-A mRNA transcription in fibroblasts and keratinocytes (Fig. 4C), and this upregulation was inhibited by pretreatment with the Akt/mTOR inhibitor rapa. Similarly, data from the cell scratch assay showed that the migration rates in CM-treated fibroblast and keratinocyte cultures were significantly higher than those in the control at different time points (Fig. 4D–F), and that pretreatment with rapa abolished CM-induced migration in these two cell types. Taken together, CM was shown to promote migration and VEGF-A expression in keratinocytes and fibroblasts in an Akt/mTOR-dependent manner, revealing the critical role of the Akt/mTOR pathway in MSC-regulated re-epithelization and angiogenesis in diabetic wound healing.
Fig. 4The conditioned medium of MSCs (CM) promotes the VEGF expression and migration of fibroblasts and keratinocytes in an Akt/mTOR-dependent manner. (A) Representative Western blots showing the effects of CM on the levels of p-Akt (S473), Akt, p-mTOR (S2448), and mTOR in fibroblasts and keratinocytes for 48 h. (B) Semi-quantitative analysis demonstrated a significant increase in p-Akt (S473) and p-mTOR (S2448) levels in CM-treated fibroblasts and keratinocytes, as compared to the control (Ctrl). The total levels of Akt and mTOR remained unchanged. n = 3 biological replicates. Two-tailed unpaired t-test. (C) The mRNA levels of VEGF-A in fibroblasts and keratinocytes were significantly high when incubated with CM for 48 h, as compared to the Ctrl. Moreover, the CM-induced increase could be abolished by 1 h pre-incubation with 1 μM rapamycin (rapa), a mTOR inhibitor. GAPDH was used as the internal reference gene. n = 9 (3 replicates/biological sample, 3 biological replicates /group). Two-way ANOVA (rapa ∗ CM) followed by Tukey's post-hoc test. (D–E) Time-lapse microscopy images showing the effects of CM and rapa on the migration of fibroblasts (D) and keratinocytes (E). The area between solid lines defines the original scratched area, and the area within the dotted lines is not covered by the migrating fibroblasts (D) or keratinocytes (E) at different time points. Cell nuclei were visualized with 4′,6-diamidino-2-phenylindole (DAPI). (F) Quantification of the migration rate revealed that CM significantly increased the migration of fibroblast or keratinocyte, and the pretreatment with rapa reversed the migration-promoting effects of CM. n = 6 (2 fields/biological sample, 3 biological replicates /group). Two-way ANOVA (rapa ∗ CM) followed by Tukey's post-hoc test. All data are presented as mean ± SEM. ⁎p < 0.05, ⁎⁎p < 0.01, ⁎⁎⁎p < 0.001.
] dosage of MTF did not induce an increase in p-Akt and p-mTOR in keratinocytes or fibroblasts, and the CM-promoted p-Akt and p-mTOR levels were not attenuated or boosted by MTF (Fig. 5A–C ). Accordingly, MTF at these two anti-diabetic dosages did not attenuate or boost the CM-induced VEGF-A upregulation (Fig. S6). In contrast, CM-induced Akt and mTOR activation was completely abolished by a supra-pharmacological dosage (3.0 mM) of MTF [
]. Consistent with these observations, the levels of p-AMPK in fibroblasts and keratinocytes were not affected by CM and the anti-diabetic dosage of MTF, but were significantly upregulated by 3.0 mM of MTF (Fig. 5A–C, E). The dose-dependent inhibition of MTF on CM-promoted Akt and mTOR phosphorylation could be observed in fibroblasts and keratinocytes with IC50 values ranging from 1.1 to 1.4 mM (Figs. 5D, F and S7).
Fig. 5The dose-dependent effects of metformin (MTF) on the Akt/mTOR activation and migration in fibroblasts and keratinocytes. The cells were pre-treated with different dosage of MTF (ranged from 0.03 mM to 3 mM) for 4 h, followed by incubation with Ctrl or CM for 48 h. (A–B) Representative Western blot images showing the effects of anti-diabetic (0.03 mM and 0.3 mM) or supra-pharmacological (3.0 mM) dosage of MTF on CM-regulated expression of p-Akt (S473), Akt, p-mTOR (S2448), mTOR, p-AMPK (Thr172), and AMPK in fibroblasts (A) and keratinocytes (B). (C, E) Semi-quantitative analysis showed that the CM-induced increase of p-Akt (S473) and p-mTOR (S2448) in fibroblasts or keratinocytes was significantly inhibited by supra-pharmacological (3.0 mM) dosage of MTF, but not by the anti-diabetic (0.03 mM and 0.3 mM) dosage. The levels of p-AMPK in fibroblasts (C) and keratinocytes (E) were not affected by CM as well as the anti-diabetic dosage of MTF treatment, but significantly upregulated by 3.0 mM of MTF. n = 3–4 biological replicates. Two-way ANOVA (MTF ∗ CM) followed by Tukey's post-hoc test. (D, F) Dose-effect curve showing the dose-dependent inhibition of MTF on CM-promoted Akt and mTOR phosphorylation in fibroblasts (D) and keratinocytes (F). The cells were pre-treated with 0.3 mM, 0.6 mM, 1.0 mM, 2.0 mM, and 3.0 mM of MTF for 4 h, followed by incubation with Ctrl or CM for 48 h. (G-H) The time-lapse microscopy images showing the effects of MTF at anti-diabetic or supra-pharmacological dosage on CM-promoted migration in fibroblasts (G) and keratinocytes (H) by using the scratch assay. The area between solid lines defines the original scratched one, and the area within the dotted lines is not covered by the migrating fibroblasts (G) or keratinocytes (H). (I) Statistical analysis revealed that the migrating rate in fibroblasts or keratinocytes was significantly increased by CM in the presence of anti-diabetic (0.03 mM and 0.3 mM) instead of supra-pharmacological (3.0 mM) dosage of MTF. n = 6 (2 fields/biological sample, 3 biological replicates /group). Two-way ANOVA (MTF ∗ CM) followed by Tukey's post-hoc test. All data are presented as mean ± SEM. ⁎p < 0.05, ⁎⁎p < 0.01, ⁎⁎⁎p < 0.001.
Similarly, MTF (0.03 mM and 0.3 mM) did not show any attenuation on the migrating rate effects of CM in both keratinocyte and fibroblast cultures, while 3.0 mM of MTF abolished such effects in these two types of cells (Fig. 5G–I). Taken together, MTF exerted dose-dependent effects on MSC-promoted Akt/mTOR activation and migration of fibroblasts and keratinocytes.
3.5 MTF impaired the survival but not wound healing effects of MSCs/FG in diabetic mice
We next evaluated the effects of MTF at an anti-diabetic dose on MSCs/FG-promoted wound healing (Fig. 6A ). The validity of MTF by oral gavage was confirmed by the significantly lower blood glucose levels in MTF-treated diabetic mice as compared to those in untreated mice (Fig. 6D). In the presence of MTF, diabetic mice treated with MSCs/FG still exhibited significantly smaller wound areas on the 3rd, 7th, and 14th days after engraftment as compared to the PBS-, FG-, or MSCs-treated counterparts (Fig. 6B and C). HE staining (Fig. S8) and immunostaining against K14 (Fig. 6E and F) further revealed that MSCs/FG treatment promoted epithelization and led to the smallest EG in all MTF-gavaged mice (Fig. 6F). Moreover, no significant difference was observed in EGs between MTF-untreated and MTF-treated counterparts (PBS vs. MTF + PBS; FG vs. MTF + PBS, MSCs vs. MTF + PBS; MSCs/FG vs. MTF + MSCs/FG) (Fig. 6F).
Fig. 6The anti-diabetic dosage of MTF impairs the survival but not wound healing effects of MSCs/FG in diabetic mice. (A) Schematic diagram of the experimental design studying the effects of MSCs/FG and MTF on diabetic wounds. The preparation of diabetic wound model, administration of FG, MSCs, or MSCs/FG were identical as described previously. Additionally, the diabetic mice were treated with MTF (250 mg/kg/day) by oral gavage from the surgery day (D0). The wounds were photographed on the D3, D7, and D14, and mice were sacrificed on D14 for the subsequent analysis. (B) Representative images showing the dynamic change of the wounds when treated with PBS, FG, MSCs. or MSCs/FG in diabetic wound models administrated with MTF. (C) MSCs/FG treatment significantly reduced the wound area in the diabetic mice model prescribed with MTF, as compared to the PBS, FG, or MSCs treatment. n = 5 for PBS and MSCs, n = 6 for FG and MSCs/FG. One-way ANOVA followed by Tukey's post-hoc test. (D) The MTF-treated diabetic mice showed significant reduction in the blood sugar level, while vehicle-treated ones exhibited persistent high blood sugar level (≥16.7 mM). D7: n = 14 for vehicle, n = 16 for MTF; D14: n = 10 for vehicle, n = 14 for MTF. Two-tailed unpaired t-test. (E) Representative images showing the immunostaining against K14 in the regenerated tissues of diabetic mice treated with MTF as well as PBS, FG, MSCs, or MSCs/FG. DAPI was used for counterstaining. Epithelial gap (EG) indicates the wound area not covered by K14+ epidermis on D14. (F) Statistical analysis of EG. n = 4 for all the groups. Two-way ANOVA (MTF ∗ wound treatment) followed by Tukey's post-hoc test. (G) Representative images showing double staining against K14 and GFP in the diabetic wounds 14 days after MSCs/FG treatment combined with MTF gavage. The area in the yellow dashed box is magnified and shown in the right panel. Note that the GFP-labeled MSCs distributed in cluster, and showed no overlapping with K14+ epidermis (Epi). (H) The analysis of GFP-labeled MSCs in the regenerated tissue revealed that the survived MSCs in mice receiving MTF gavage was significantly less than that of those not received MTF. n = 6 (2 fields/wound, 3 wounds/group). Two-tailed unpaired t-test. (I–K) Representative images of double staining against GFP and αSMA (I), CD34 (J), or vimentin (K) in the regenerated tissue. Mice were treated with GFP-labeled MSCs in FG scaffold as well as MTF by oral gavage for 14 days. Images in the yellow dotted box are amplified and shown as insets. The yellow arrowheads indicate the GFP-labeled MSCs, while the white and red arrowheads indicate the αSMA+ and CD34+ vessel-like structures, respectively. GFP+ cells did not co-label with αSMA and CD34, but with vimentin. Scale bar: 1 mm in E and the left panel of G, 100 μm in the right panel of G, I–K. Data are presented as the mean ± SEM. ⁎p < 0.05, ⁎⁎p < 0.01, ⁎⁎⁎p < 0.001. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
The survival and cell fate of MSCs/FG in the presence of MTF were evaluated. As shown in Fig. 6G–H, GFP-labeled MSCs were significantly less in MTF-treated mice than in the untreated ones. The residual MSCs were distributed in clusters, showing phenotypes positive for vimentin and negative for K14, CD34, and αSMA (Fig. 6G–K), similar to the distribution and phenotypes of MSCs/FG in MTF-untreated mice. These data indicate that MTF did not interfere with the wound healing effects of MSCs/FG; however, it reduced the survival of implanted MSCs in FG.
3.6 The synergistic effects of MTF and MSCs on angiogenesis and Akt activation in diabetic wound healing
In all MTF-gavaged mice, MSC/FG treatment led to a substantial increase in CD34+ (Fig. S9A and B), and αSMA+ vessel-like structures (Fig. S9A and C) compared to their PBS-, FG-, or MSCs counterparts. Furthermore, a ∼4-fold increase was observed in CD34+ vessel-like structures in the center area, and ∼1.5-fold increase in CD34− and αSMA+ vessels in the transitional area of the wounds treated with MTF and MSCs/FG in comparison with those treated with MSCs/FG alone (Fig. 7A–C ), suggesting that MTF acted in coordination with MSCs/FG for angiogenesis in the diabetic wound. WB analysis of CD31 and CD34 expression levels further confirmed the enhancement of vascular formation in the regenerated wound tissue treated with MSCs/FG and MTF compared to those treated with MSCs/FG (Fig. 7D and E).
Fig. 7MSCs/FG and anti-diabetic dosage of MTF synergistically promote Akt/mTOR activation, VEGF-A expression, and angiogenesis in the diabetic wound healing. (A) Representative images showing immunostaining against CD34 and α-SMA in the center and transitional area of the diabetic wounds on D14. The wounds were treated with MSCs/FG, together with or without MTF (250 mg/kg/day) gavage. The arrows indicate the CD34+ vessel-like structure in the center, while the arrowheads indicate the vessel-like structure positive for both CD34 and α-SMA in the transitional area of the wound. The area in the dashed box is amplified and shown as insets. (B–C) Statistical analysis revealed that mice treated with MSCs/FG and MTF showed significantly increased CD34+ vessel-like structure in the center (B), as well as significantly increased vessel-like structure positive for CD34 and α-SMA in the transitional area of wounds (C), as compared to the mice treated with MSCs/FG alone. n = 8 (2 fields/wound, 4 wounds/group). Two-tailed unpaired t-test. (D) Representative WB images showing the CD31 and CD34 expression levels in the diabetic wounds treated with PBS, FG, MSCs, and MSCs/FG in the presence or absence of MTF on D14. β-actin was used as an internal control. (E) Semi-quantitative results revealed that mice treated with MSCs/FG and MTF exhibited significantly increased CD31 and CD34 expression levels in the diabetic wounds, as compared to the mice treated with either PBS and MTF, FG and MTF, MSC and MTF, as well as MSCs/FG alone. n = 3 biological replicates. Two-way ANOVA (MTF ∗ wound treatment) followed by Tukey's post-hoc test. (F–G) Quantitative PCR analysis showing the effects of MTF on the VEGF-A mRNA levels in PBS- (F) or MSCs/FG-treated diabetic wounds (G). MTF specifically increased the VEGF-A mRNA levels in the PBS-treated diabetic wounds (F); further, it exhibited synergistic effects with MSCs/FG treatment on the VEGF-A expression (G). The levels in PBS-treated wounds were set as control in G. n = 9 (3 replicates/wound, 3 wounds/group). Two-tailed unpaired t-test. (H) Representative WB images showing the VEGF-A expression levels in the diabetic wounds treated with PBS, FG, MSCs, and MSCs/FG in the presence or absence of MTF on D14. (I) Semi-quantitative results revealed that mice treated with MSCs/FG and MTF exhibited significantly increased VEGF-A expression levels when compared to the mice treated with either PBS and MTF, FG and MTF, MSCs and MTF, and MSCs/FG alone. n = 3 biological replicates. Two-way ANOVA (MTF ∗ wound treatment) followed by Tukey's post-hoc test. (J) Representative WB images showing the influence of MSCs/FG and MTF on the regulation of p-Akt (S473), p-mTOR (S2448), total Akt, and mTOR protein levels in the regenerated diabetic wound on D14. (K) Semi-quantitative analysis revealed that mice treated with MTF by oral gavage showed significant increased levels of p-Akt (S473) and p-mTOR (S2448) in the local diabetic wound. n = 3 biological replicates. Two-tailed unpaired t-test. Scale bar: 100 μm in A. All data are presented as mean ± SEM. ⁎p < 0.05, ⁎⁎p < 0.01, ⁎⁎⁎p < 0.001.
Indeed, MTF was found to significantly promote VEGF-A expression in untreated (Fig. 7F) and MSCs/FG-treated diabetic wounds (Fig. 7G). Accordingly, in the presence of MTF, diabetic mice treated with MSCs/FG showed significantly upregulated VEGF-A protein (Fig. 7H–I) levels in the regenerated wound tissue compared to the MTF-untreated counterparts.
Consistent with these observations, MTF at the anti-diabetic dosage induced a moderate but significant increase in Akt/mTOR phosphorylation. More importantly, MTF cooperated with MSCs/FG to significantly upregulate p-AKT and p-mTOR levels in the local diabetic wound, as compared to the MSCs/FG-treated wound (Fig. 7J–K). The total Akt and mTOR levels remained unchanged. These results indicate that MTF and MSCs/FG coordinately promote angiogenesis in diabetic wound healing; the synergistic effects on VEGF-A expression and Akt/mTOR activation are involved in the underlying mechanism.
4. Discussion
Stem cell therapy is a promising strategy for the treatment of DFU [
]. Given that the state of autologous MSCs is often impaired in patients with DM, allogeneic MSCs from the umbilical cord are a good candidate for cell therapy. In the present study, we showed that MSCs/FG derived from the umbilical cord significantly promotes diabetic wound healing. Different from the observation that autologous MSCs in fibrin scaffold differentiated into the blood vessels [
], here we found that xenogeneic MSCs remained mesenchymal character in the scaffold, showing no obvious integration or migration. Similarly, another in vitro migration assay revealed that MSCs in FG were localized in the scaffold and did not migrate to the injured site [
]. This difference might be due to the pore size of FG and its cell type specificity. Considering that the presence of immunogenicity after differentiation reduced the therapeutic effects of allogeneic MSCs, the FG scaffold offered a localized niche for MSCs survival and exertion of therapeutic effects. In addition, no signs of integration or migration indicated the safety of topical application of MSCs/FG therapy.
Angiogenesis plays an important role in the proliferation phase of wound healing. Newly formed endothelial tubes are initially unstable and subsequently stabilized through their association with pericytes [
]. The appearance of both endothelial cells and pericytes in the wound area, together with the increase in pro-angiogenic factors VEGF-A, indicated that MSCs/FG treatment established a functional microvasculature. In contrast to the positive regulation of MSCs/FG, MTF showed paradoxical effects on angiogenesis. In general, it was considered to downregulate angiogenesis in the context of tumors but was associated with enhanced VEGF expression [
] in disease models such as DM. In agreement with these observations, we observed that MTF at an anti-diabetic dosage significantly boosted MSC-induced endothelial reconstitution as well as VEGF-A expression in vivo, indicating that MTF-promoted angiogenesis is more focused on the recruitment of endothelial cells. Although MTF reduced the survival of MSCs in the FG scaffold, the systematic improvement of glucose homeostasis, as well as the synergistic angiogenesis-promoting effects, probably overwhelmed the growth inhibition effects, resulting in comparable wound closure rates with improved angiogenesis. The beneficial effects of the combined administration of MTF and MSCs on diabetic wound healing will be instructive for the promotion of MSCs therapy in patients with DFU prescribed MTF.
mTOR, the major regulator of cell growth and metabolism, is cross-talked with other energy sensor pathways, such as AMPK and Akt [
]. Instead of an inhibitory effect of AMPK on mTOR, Akt was reported to activate mTOR pathways, mediating downstream responses, including cell survival, growth, migration, and angiogenesis, which are critical for tissue regeneration and wound repair. We showed that the recruitment of fibroblasts and keratinocytes for migration and VEGF expression is mTOR-dependent, reinforcing the critical role of mTOR in MSC/FG-promoted diabetic wound healing. Indeed, Castilho [
] reported that diverse molecular signaling circuits involved in the process of wound healing ultimately converge on the activation of mTOR. Akt/mTOR signaling has been shown to be activated during normal wound healing and weakened in refractory diabetic wounds [
]; to our knowledge, the exact role of MTF on the Akt/mTOR pathway and its interaction with MSCs therapy have not been well elucidated in terms of diabetic wound healing. It is worth noting that MTF exerts its effects through multiple mechanism in a dose-dependent manner [
]. The widely reported mechanisms of MTF's anti-aging and anti-tumor action, that is, AMPK activation and subsequent mTOR inhibition, have mainly been observed in the context of supra-pharmacological (>1 mM) concentrations [
]. Here, we also observed that AMPK activation occurred only at the mM levels of MTF in keratinocytes and fibroblasts, consistent with previous reports [
]. Although inhibition of hepatic glucose production through AMPK is a critical contributor to the anti-diabetic dosage of MTF lowering hyperglycemia, recent evidence indicates that improved insulin resistance and glucose uptake in peripheral tissues is identically important [
] in the context of metabolic syndrome. In the present study, we reported that MTF exerted dose-dependent effects on skin cells; MSCs-induced Akt/mTOR activation was inhibited by MTF at supra-pharmacological concentrations, while it remained unaffected at the anti-diabetic dosage. MTF at an anti-diabetic dosage improved DM-impaired Akt/mTOR impairment, and more importantly, cooperated with MSCs/FG for prominent Akt activation in the local diabetic wound, which was in line with the observations of increased VEGF expression and angiogenesis by combined application of MSCs/FG and MTF.
It is noteworthy that the recommended hypoglycemic dose of MTF (0.25–2.5 g/day) resulted in plasma concentrations ranging from 5 to 20 μM in patients with DM [
]. This was mainly due to the marked diversity between mammalian species, such as basal metabolism, circulating plasma proteins, and renal function, and the rodent equivalent dose was more appropriate for calculation according to the FDA-recommended body surface area normalization method [
]. The dose range of 200–350 mg/kg in rodents is equivalent to the hypoglycemic dose in the clinic and has consistently been used for the glucose-lowering effects of MTF in rodent experiments [
]. However, the difference in plasma MTF concentrations between humans and rodents exhibiting glucose-lowering effects reminds us to be cautious in translating MTF action from rodent models into clinical strategies.
In summary, MTF and MSCs synergistically promoted the repair of diabetic wounds and angiogenesis. Clarification of the effects of MTF and MSCs therapy on the signaling pathways of wound healing will provide more insights for the application of new strategies in DFU treatment.
Funding sources
This work was funded by the National Key Research and Development Program of China (No. 2021YFA1101103), Strategic Priority Research Program of the Chinese Academy of Sciences (No. XDA16020807), Major Innovative Research Team of Suzhou, China (No. ZXT2019007), Young Medical Talents Supported by the Jiangsu Provincial Improving Medical and Health Care by Science and Education Program (QNRC2016861), 14th “Six Talents Peaks” Foundation of Jiangsu Province (WSW-054), and Fifth Gusu Health Stratification Training Youth Talents Program (GSWS2019038).
CRediT authorship contribution statement
Fangzhou Du: Conceptualization, Methodology, Validation, Data curation, Formal analysis, Writing - Original draft preparation. Mengmeng Liu and Jingwen Wang: Investigation, Validation. Lvzhong Hu: Resources. Shaocong Zhou, Dongao Zeng: Validation, Formal analysis. Lixing Zhang: Conceptualization. Meijia Wang, Xi Xu: Validation. Chenglong Li, Jingzhong Zhang, Shuang Yu: Writing - Review & Editing, Supervision, Funding acquisition.
Data access
The data supporting the findings of this study are available from the corresponding author upon reasonable request.
Topical administration of allogeneic mesenchymal stromal cells seeded in a collagen scaffold augments wound healing and increases angiogenesis in the diabetic rabbit ulcer.
Metformin induces cell cycle arrest, reduced proliferation, wound healing impairment in vivo and is associated to clinical outcomes in diabetic foot ulcer patients.
00001-X&id=ga1.jpg){kind=link}
00001-X&id=gr1.jpg){kind=link}
00001-X&id=gr2.jpg){kind=link}
00001-X&id=gr3.jpg){kind=link}
00001-X&id=gr4.jpg){kind=link}
00001-X&id=gr5.jpg){kind=link}
00001-X&id=gr6.jpg){kind=link}
00001-X&id=gr7.jpg){kind=link}