External Application of Human Umbilical Cord-Derived Mesenchymal Stem Cells in Hyaluronic Acid Gel Repairs Foot Wounds of Types I and II Diabetic Rats Through Paracrine Action Mode

Abstract

Diabetic foot ulcer (DFU) is a main diabetic complication with unmet treatment needs. This study applied human umbilical cord-derived mesenchymal stem cells-hyaluronic acid (hucMSCs-HA) gel to treat DFU in a noninvasive external way and investigated its paracrine action and mechanism. In this study, after analyzing the physical and biological properties of HA gel, hucMSCs-HA gel was applied in 2 in vivo models (types I and II DFU), and a molecular mechanism was investigated. To evaluate the paracrine action of hucMSCs, hucMSCs-conditional medium (MSC-CM) was collected to treat 1 in vivo model (type I DFU) and 2 in vitro models (high glucose (HG)-injured human umbilical vein endothelial cells (HUVECs) and human skin fibroblasts (HSFs)). The results indicated that HA gel with a porous microstructure underwent over 90% degradation and swelled to the maximum value within 48 h. In vivo, hucMSCs-HA gel accelerated wound healing of DFU rats by improving re-epithelialization, collagen deposition, and angiogenesis, in which a paracrine action of hucMSCs was confirmed and the phosphorylation of p38, ERK1/2, JNK, and Akt was increased. In vitro, MSC-CM improved cell viability, wound healing, migration, tube formation, cell senescence, and abnormal expressions (TNF-α, IL-1β, IL-6, ET-1, p16 genes, and PCNA protein) of HUVECs, also improved cell viability, wound healing, antioxidant stress, and abnormal expressions (COL1, COL3, COL4, SOD1, SOD2 genes, and PCNA protein) of HSFs. Summarily, noninvasive external application of hucMSCs-HA gel shows great perspective against DFU and exerts wound healing effects through the MAPK and Akt pathways-mediated paracrine mechanism.

Graphical Abstract

Significance Statement.

Diabetic foot ulcer (DFU) is a main diabetic complication with unmet treatment needs. Noninvasive external application of mesenchymal stem cells (MSCs) can avoid shortcomings of MSCs injection therapy, indicating a promising therapeutic option for DFU. This study developed an excellent hucMSCs-HA gel dressing and confirmed that its noninvasive external application could promote DFU healing through re-epithelialization, collagen deposition, and angiogenesis, involving the MAPK and Akt pathways-mediated paracrine mechanism.

Introduction

Diabetes mellitus is a globally severe condition accompanied by rapidly increasing incidence (projected 12.2% by 2045) and large health expenditures (projected $1054 billion by 2045).¹ Diabetic foot ulcer (DFU) becomes a major diabetic complication with high mortality and disability rates, which massively decreases the life quality of patients and increases healthcare costs.^2,3 DFU refers to a full-thickness foot wound below the ankle, which increases the risks of diabetic neuropathy and peripheral arterial disease.⁴ Eventually, approximately 85% of DFU patients may suffer limb amputations, along with morbidity of 6.23% (type I diabetes) and 6.72% (type II diabetes).⁵ The conventional treatment for DFU is local wound care with debridement and antibiosis, followed by covered dry dressings (gauze, cotton pads, and bandages), which aims to control infection and facilitate self-healing.^6,7 However, the outcomes remain unsatisfactory regarding tissue regeneration.^8,9 Recently, many adjunctive therapies have been developed including bioengineered skin grafts, hyperbaric oxygen therapy, negative pressure wound therapy, and biomaterial dressings.¹⁰ However, bioengineered skin grafts have low bioavailability or immunological rejection,¹¹ and hyperbaric oxygen and negative pressure rely on special infrastructure, prone to barotrauma complications and increase bacterial infection rate,^12,13 resulting in limitations in clinical application. Biomaterials, such as collagen, chitosan, and hyaluronic acid (HA) gel, are low-priced and provide a moist environment, benefiting wound healing in DFU.¹⁴ However, these materials without living cells lack biological activity and their regenerative outcomes remain unsatisfactory.¹⁵ Collectively, approaches combining biomaterials with living cells would be a promising strategy for DFU wound healing.

Mesenchymal stem cells (MSCs) are a type of pluripotent progenitor cells with high self-renewal, differentiation, tissue regeneration, and immune regulation abilities, obtained from various human tissues, such as umbilical cord, adipose tissue, and bone marrow.^16,17 Currently, MSCs have become an important source of cell-based therapies for various diseases.¹⁸ Clinical trials indicated that bone marrow-derived MSCs effectively improved insulin sensitivity and C-peptide response in type II diabetic patients.¹⁹ Zhou et al found that umbilical cord-derived MSCs ameliorated the symptoms of type II diabetic rats, and improved the islet viability and insulin secretory function by introducing β-cell growth factors (IGF-1, HGF, and PDGF-A).²⁰ With the widespread use of MSCs in diabetes, growing attention has been drawn to the use of MSCs in diabetic complications.^21,22 A clinical trial has applied HA gel loaded MSCs (adipose-derived) to treat DFU, wound closure rate (82%) and Kaplan-Meier median closure time (28.5 days) were all improved compared to the control group with a 53% closure rate of and 63.0 days’ closure time.²³ Paracrine is the main action mode of MSCs in wound repair, which increases epithelialization, granulation tissue formation, and angiogenesis by paracrine pathways.^24,25 For instance, intravenously transplantation of umbilical cord-derived MSCs secreted growth factors (VEGF, bFGF, and HGF) that promoted angiogenesis and collagen deposition, thereby regulating inflammation in wound tissue and accelerating wound healing in diabetic rats.²⁶ To date, the mainstay of MSCs therapy studies for treating DFU focused on the injection routes (intravenous injection, intravenous infusion, or topical injection). Although application of MSCs through injection has achieved certain therapeutic effects, the limitations are also obvious, for example, inconvenient operations, unknown risk of rejection reaction, and secondary damage to skin.²⁷ Currently, few attentions have paid to the noninvasive application of MSCs.

To fill this gap, this study developed an external application of HA-loaded MSCs to treat DFU. Human umbilical cord-derived MSCs (hucMSCs) were applied to prepare hucMSCs-HA gel, in which hucMSCs played a leading role and HA gel acted as a carrier to maintain activity and bio-functions of hucMSCs. The hucMSCs-HA gel was directly applied to cover and improve foot wounds in both types I and II diabetic rat models. To further explore paracrine action and mechanism of hucMSCs, hucMSCs-conditional medium (MSC-CM) was applied in treating type I DFU rat model and HG-injured cellular models (human umbilical vein endothelial cells (HUVECs) and human skin fibroblasts (HSFs)). This is the first study applying noninvasive hucMSCs-HA gel to treat DFU, conquering limitations of MSCs injection as well as showing great perspective in further clinical application.

Materials and Methods

Chemicals and Reagents

Streptozotocin (STZ) was obtained from Sigma (Saint Louis, USA). Sodium hyaluronate powder (USP grade, 1000–1500 kDa) was purchased from Shanghai Yuanye Bio-Technology Co., Ltd. Alpha-minimum essential medium (α-MEM) was purchased from Gibco BRL. Dulbecco’s modified Eagle’s medium (DMEM) was purchased from Zhejiang Senrui Biotechnology Co., Ltd. Trypsin (0.25%) and phosphate-buffered saline (PBS) were purchased from Biosharp. Fetal bovine serum (FBS) was purchased from Gibco. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) was purchased from Sigma-Aldrich. TRIzol reagent and cell culture plates were purchased from Thermo Fisher Scientific Inc. Transwell chambers and Matrigel growth factor reduced basement were purchased from Corning. Masson’s Trichrome Stain Kit (R20381) was purchased from Shanghai Yuanye Technology Co., Ltd. The Malondialdehyde (MDA) Detection Kit was purchased from Nanjing Jiancheng Bioengineering Institute. The miScript SYBR Green PCR kit was obtained from Qiagen. Calcein/PI cell viability/cytotoxicity assay kit (C2015M), senescence β-galactosidase staining kit (C0602), RIPA buffer, BCA protein assay kit (P0010S) and reactive oxygen species assay kit (S0033S) were purchased from Beyotime Institute of Biotechnolog. The antibodies against PCNA (NB500-106) and VEGFA (NB100-664) were purchased from Novus Biologicals, Inc., the antibodies against β-actin (A3854), p-p38 (Thr180/Tyr182, 4511), total p38 (87869), total JNK (9252S), p-JNK (Thr183/Tyr185, 9255S), p-ERK1/2 (p-p44/42, Thr202/Tyr204, 4370), total ERK1/2 (p44/42, 4695), p-Akt (Ser473, 4060T), and total Akt (4691T) were purchased from Cell Signaling Technology, Inc., and horseradish peroxidase–conjugated antibody was purchased from Zhongshan Jinqiao Biotechnology Co., Ltd.

Preparation and Identification of hucMSCs

HucMSCs were obtained from the cell resource bank and integrated cell preparation center of Xiaoshan, incubated in α-MEM containing 10% FBS, L-glutamine, ribonucleosides, deoxyribonucleosides at 37 °C, 5% CO₂, grown to 80%-90% confluence, and passaged 3-4 generations. To ensure the accuracy of subsequent experiments, flow cytometry analysis was applied to identify the biological properties of hucMSCs. HucMSCs (1 × 10⁹ cells/mL) were stained with antibodies of CD34-FITC, CD45-FITC, HLA-DR-FITC, CD73-FITC, CD90-FITC, and CD105-APC, followed by incubation at 4 °C in dark for 30 minutes. Finally, positive rates of these surface markers of stem cells were tested by Accuri C6 flow cytometer (BD Biosciences).

Preparation of MSC-CM

The hucMSCs (3 ×  10⁵ cells/well) were seeded into 10 cm dishes and grown to 80%-90% confluence in α-MEM. After the culture medium was removed, the cells were washed for 3 times with PBS, and incubated with DMEM containing 10% FBS and 25 mM glucose for 48 h. Then the culture medium was collected and centrifuged at 1400 ×g for 10 minutes to remove cell debris. The supernatant was obtained as MSC-CM.

Preparation of HA Gel and MSC-HA Gels

To prepare HA and MSC-HA gels (hucMSCs-HA gel and MSC-CM-HA gel), sodium hyaluronate powder was dissolved in PBS (8 mg/mL) at room temperature. Before solidification, 3 × 10⁶ hucMSCs were mixed with the gel (3 mL) to prepare the MSC-HA gels (1 × 10⁶ cells/mL). As for the MSC-CM-HA gel preparation, 15 mL MSC-CM were concentrated to a volume of 3 mL, and then mixed with 24 mg sodium hyaluronate powder (equal to 8 mg/mL) at room temperature.

Characterization of HA Gel

Internal morphology, degradation ratio, and swelling ratio were assessed to characterize the property of HA gel. Cell viability (by CCK-8 assay) and live/dead cell rate (calcein/PI cell viability/cytotoxicity assay) of hucMSCs in HA gel were evaluated to determine the cytocompatibility. Internal morphology of the HA gel was captured by scanning electron microscopy (SEM, S-3000N E-1010, Hitachi Limited, Tokyo, Japan). The wet weight (Ww) and dry weight (Wd, freeze-drying powder) data were obtained to measure the swelling ratio at 0.5, 1, 2, 4, 8, 16, 24, 36, and 48 h by using the equation: swelling ratio = (Ww − Wd)/Wd. After mixing the HA gel (1 mL) with PBS (3 mL) at 37 °C, the weight of the residual mass (Wr) of HA gel after removing PBS at different times (0, 1, 2, 4, 8, 16, 24, 36, 48, 60, 72, 84, and 96 h) was recorded to calculate the degradation rate: degradation rate (%) = (Ww—Wr)/ Ww × 100%.

For the CCK-8 assay, hucMSCs (3 × 10³ cells/well) mixed with HA gel were seeded into a 96-well plate and incubated with serum-free α-MEM medium for 0, 24, and 48 h at 37 °C. CCK-8 working solution (at the concentration of 10% v/v) was then added and incubated for 2 h at 37 °C. Optical density (OD) values were detected at a wavelength of 450 nm using the microplate photometer (Synergy H1, BioTek,).

For live/dead cell assay, hucMSCs (2 × 10⁵ cells) mixed with HA gel were seeded into 35 mm dishes and incubated with serum-free α-MEM medium for 0, 24, and 48 h at 37 °C. Afterward, the cells were stained with calcein and PI from the calcein/PI cell viability/cytotoxicity assay kit for 30 minutes. Final images were captured under an inverted microscope (Carl Zeiss).

Animals

Male SD, Wistar, and Goto-Kakizaki (GK) rats were purchased from SLAC Laboratory Animal Co. Ltd (Certificate No: SCXK (Shanghai) 2017-0005). Among them, SD rats (8 weeks old, 250-280 g) were used to establish the type I diabetic model, and GK rats (12 weeks old, 300-350 g) were used to establish type II diabetic model. All rats were fed in an SPF animal room with standard environmental conditions (22 ± 2 °C, relative humidity of 55%-60%, and 12-h light/12-h dark cycles). All animal experiments were approved by the Animal Ethics Committee of Zhejiang Chinese Medical University, Hangzhou, China, and met the guidelines for the Care and Use of Laboratory Animals published by the National Institutes of Health Animal (Ethics No: IACUC-20190408-09).

DFU Modeling and Grouping

A total of 32 SD rats were applied for DFU modeling of type I diabetes and evaluation of the hucMSCs-HA gel with the following groups (each n = 8): (1) control group, (2) type I diabetic model (T1DM) group, (3) HA gel treated (HA) group, and (4) hucMSCs-HA gel treated (MSC-HA) group. Another 32 SD rats were applied to testify the paracrine action of the hucMSCs-HA gel with the following groups (n = 8): (1) control group, (2) T1DM group, (3) hucMSCs-HA gel treated (hucMSCs-HA) group, and (4) MSC-CM-HA treated (MSC-CM-HA) group. SD rats in the T1DM group, HA group, MSC-HA group, hucMSCs-HA group, and MSC-CM-HA group were modeled by tail vein injection of STZ in citrate buffer (50 mg/kg, pH 4.2-4.5). After the STZ injection for 7 days, SD rats with blood glucose levels ≥ 16.1 mM were selected as type I diabetic rats and surgically suffered full-thickness dorsal skin wounds (5 × 10 mm) on their feet. SD rats in the control group received an equal volume of blank citric buffer. After the modeling, SD rats in the HA group, MSC-HA group, hucMSCs-HA group, and MSC-CM-HA group were treated with the respective gels.

Totally 24 GK rats were applied for DFU modeling of type II diabetes and 8 Wistar rats were applied as the control rats with the following groups (each n = 8): (1) control group, (2) type II diabetic model (T2DM) group, (3) HA treatment (HA) group, and (4) hucMSCs-HA gel treatment (MSC-HA) group. GK rats with blood glucose levels ≥ 10-20 mM were selected as type II diabetic rats and surgically suffered full-thickness dorsal skin wounds (5 × 10 mm) on their feet. After the modeling, GK rats in the HA group and MSC-HA group were treated with the HA gel and hucMSCs-HA gel, respectively.

According to the degradation characteristics of HA gel, we chose to apply MSC-HA gel every 48 h for a total of 11 times over a period of 21 days. Digital images of all wounds were taken every 3 days until the complete healing, and the foot wound area (mm²) was quantified using Image-J software (Version 1.49, National Institutes of Health). The average wound healing rate was calculated using the formula: wound healing rate (%) = (sq.0—sq.A)/ sq.0 × 100%, where sq.0 represented the original wound area on day 0, and sq.A represented the wound area on day A (3, 6, 9, 12, 15, 18, and 21). In the end, the foot skin samples were bisected from euthanized rats and fixed in 4% paraformaldehyde for further experiments.

Histopathological Staining and Immunohistochemical Analysis

To evaluate the therapeutic effect of hucMSCs on DFU, the pathological morphology, collagen deposition, and VEGFA expression were determined by H&E staining, Masson staining, and immunohistochemical analysis, respectively. All skin samples were dehydrated by the Thermo Fisher Scientific Excelsior AS (Thermo Fisher Scientific Inc.) and embedded in paraffin using Thermo HistoStar (Thermo Fisher Scientific Inc.). Four micrometer sections were cut with Semi Motorized Rotary Microtome (RM2245, Leica). H&E staining was conducted by Auto-stainer (ST5010, Leica), and Masson staining was conducted using Masson’s Trichrome Stain Kit (G1340, Solarbio). For immunohistochemical analysis, sections were deparaffinized, rehydrated, and then received heat-induced epitope retrieval. Endogenous peroxidase was inhibited by 3% hydrogen peroxide. The slides were then incubated with the primary antibody (VEGFA, dilution 1: 50) overnight, rinsed 3 times in PBS for 5 minutes at room temperature, incubated with a biotinylated secondary antibody (dilution 1: 100) for 1 h, and rinsed 3 times in PBS at room temperature. Immunohistochemical detection was performed with 3,3ʹ-diaminobenzidine tetrahydrochloride (DAB). All sections were sealed with neutral gum and photographed using inverted microscope (Nikon Eclipse 80i, Nikon). Quantitative analyses of collagen deposition, VEGFA expression, and number of blood vessels were performed with Image-J software and the Image-Pro Plus software (Version 6.0, Media Cybernetics).

Cell Culture of HUVECs and HSFs

HUVECs and HSFs were purchased from the Chinese Academy of Sciences (Beijing, China). At 37 °C, 5% CO₂, and humidified atmosphere, both HUVECs and HSFs were cultured in DMEM with 25 mM glucose, 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin. After the cell confluence reached approximately 80%, the cells were digested with 0.25% trypsin and applied for the follow-up assays.

Conditioned Medium Preparation of HUVECs and HSFs

HUVECs or HSFs were cultured in 10 cm dishes and grew to 80%-90% confluence in DMEM for conditioned medium collection. First, remove the medium, wash the cells 3 times with PBS, and then incubate cells with DMEM containing 10% FBS and 25 mM glucose for 48 h. The supernatant was collected after centrifugation at 1400 ×g for 10 minutes to remove cell debris as the conditioned medium of HUVECs (HUVEC-CM) and HSFs (HSF-CM).

Cellular Experimentation and MSC-CM Treatment

To further evaluate the effects and mode of hucMSCs, the HUVECs and HSFs were cultured in high glucose (HG) and treated with MSC-CM. Three groups for each cell type were set as follows: control group (cultured in normal DMEM medium with 25 mM glucose in total), model (HG) group (cultured in DMEM containing 50 mM glucose in total), and MSC-CM group (cultured in DMEM containing 50 mM glucose in total). For in vitro modeling, the HUVECs and HSFs were incubated a under high-glucose environment for 48 h. The control group and HG group received HUVEC-CM (or HSF-CM, referring to the testing cells) treatment, while the MSC-CM group received MSC-CM treatment. After the incubation for 48 h, the HUVECs and HSFs were applied for the follow-up assays.

Cell Viability Assay

The HUVECs (3 × 10³ cells/well) and HSFs (3 × 10³ cells/well) were seeded in 96-well plates and were modeled and treated as described above. Subsequently, the HUVECs and HSFs were added with 50 μL MTT solutions, and incubated for 4 h at 37 °C in the dark. After removing the supernatants, 150 μL dimethyl sulfoxide (DMSO) was added into each well, shaking for 10 minutes and dissolving the purple formazan formed by the reduction of MTT. Finally, OD values were detected at a wavelength of 490 nm (HUVECs) and 570 nm (HSFs) using the microplate photometer (Multiskan FC, Thermo Fisher Scientific Inc.).

Wound Healing Assay

The effects of hucMSCs on the horizontal migration capabilities of HUVECs and HSFs were assessed by wound healing assay. The HUVECs (6 × 10⁴ cells/well) and HSFs (8 × 10⁴ cells/well) were seeded in the 6-well plates, treated with different glucose concentrations and MSC-CM as described above. The scratched cell-free zone was manually created across the cell monolayer by a sterile 10 μL pipette tip. The cells were subsequently washed twice with PBS to remove cellular debris and cultured for 48 h in fresh DMEM medium. Then, the migration of HUVECs and HSFs was observed and photographed under an inverted microscope (Carl Zeiss). The wound area was calculated by using the Image-J software. The wound closure percentage was obtained using the following formula: wound closure rate (%) = (A₀ − A_t)/A₀ × 100, where A₀ was the wound area at 0 h and A_t was the remaining area at the designated time.

Transwell Migration Assay

The effect of hucMSCs on the vertical migration capability of HUVECs was assessed by transwell migration assay. The HUVECs (3 × 10⁴ cells/well) were seeded in the 6-well plates and exposed to glucose concentrations and received MSC-CM treatment as described above. After the treatment, the HUVECs were digested by 0.25% trypsin, and then 4 × 10³ cells/well were seeded in the upper chamber of a 24-well transwell plate with a polycarbonate membrane. After culturing in serum-free DMEM medium for 15 h, the migrated HUVECs in the lower chamber were fixed with 4% paraformaldehyde and stained. The migrated cell number was calculated under an inverted microscope (FLEXACAM C1, Leica) by Image-J software.

Reactive Oxygen Species (ROS) Production and Lipid Peroxidation Determination

The effect of hucMSCs on the generation of intracellular ROS in HSFs was assessed with the Reactive Oxygen Species Assay Kit (S0033S). The HSFs (4 × 10³ cells/well) were seeded in the 24-well plates and treated as described above. Following the manufacturerʹs instructions, the HSFs were washed twice with PBS and incubated in serum-free medium with 10 μmoL/L oxidation-sensitive fluorescent probe ­(dichloro-dihydro-­fluorescein diacetate, DCFH-DA) at 37 °C for 20 minutes. Subsequently, the HSFs were washed thrice with serum-free medium and observed under inverted microscope (Carl Zeiss). The ROS fluorescence intensity in random fields was captured by Zen software (Carl Zeiss) and quantified by the Image-J software. The effect of hucMSCs on the lipid peroxidation of HSFs was assessed with the malondialdehyde (MDA) detection kit (A003-1-2). The HSFs (8 × 10⁴ cells/well) were seeded and treated as described above. Finally, the supernatant of HSFs was collected and mixed with TBA, reacting at 90-100°C under acidic conditions according to the manufacturerʹs instructions. The MDA level was measured at 532 nm using the microplate photometer (Multiskan FC, Thermo Fisher Scientific Inc.).

Cell Senescence Staining

The effect of hucMSCs on the senescence of HUVECs was assayed by the senescence-associated galactosidase ­(SA-β-Gal) staining using a senescence β-galactosidase staining kit (C0602). The HUVECs (4 × 10⁴ cells/well) were seeded and treated as described above. According to the manufacturer’s instructions, the treated HUVECs were rinsed twice with PBS, added with fixative for 15 minutes, washed twice with PBS, and subsequently stained with a working solution of β-galactosidase with X-Gal at 37 °C for 2 h. Finally, the SA-β-Gal-positive cells in randomly selected microscope fields (× 400) were observed under an inverted microscope (Carl Zeiss) and quantified by using the Image-J software.

Tube Formation Assay

The effect of hucMSCs on the angiogenesis of HUVECs was assessed by tube formation assay using Matrigel Growth Factor Reduced Basement. The HUVECs (4 × 10⁴ cells/well) were seeded and treated as described above. After the treatment, 50 μL of growth-factor reduced Matrigel was added into each well of the pre-coated 96-well plates and solidified at 37 °C for 30 minutes. Then, the HUVECs were collected after 0.25% trypsin digestion and seeded into the Matrigel pre-coated 96-well plates at 37 °C for 4 h. The tube network formation was observed under an inverted microscope (Carl Zeiss) and quantified by using the Image-J software.

RNA Extraction and Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)

The total RNA of HUVECs and HSFs was extracted with Trizol reagent and quantified by NanoDrop 2000 (Thermo Fisher Scientific, Inc.). qRT-PCR was carried out in Applied Biosystems StepOnePlus Real-Time PCR System (Thermo Fisher Scientific, Inc.) with a running condition as follows: a holding stage (one cycle of 30 s at 95 °C), a cycling stage (40 cycles of 5 s at 95 °C and 40 cycles of 30 s at 60 °C), and a melting curve stage (one cycle of 15 s at 95 °C, one cycle of 30 s at 60 °C and one cycle of 15 sec at 95 °C). The gene expressions of TNF-α, IL-1β, IL6, ET1, and p16 in HUVECs and the expressions of COL1, COL3, COL4, SOD1, and SOD2 in HSFs were analyzed, in which β-ACTIN was used as a reference. All data were calculated by the ^ΔΔCt method and the primer sequences of all genes are listed in Table 1.

Table 1.

Primer sequences used for quantitative real-time PCR analysis.

Gene	Forward primer	Reverse primer
β-ACTIN	TGGCACCCAGCACAATGAA	CTAAGTCATAGTCCGCCTAGAAGCA
TNF-α	CCTCTCTCTAATCAGCCCTCTG	GAGGACCTGGGAGTAGATGAG
IL-1β	ATGATGGCTTATTACAGTGGCAA	GTCGGAGATTCGTAGCTGGA
IL-6	ACTCACCTCTTCAGAACGAATTG	CCATCTTTGGAAGGTTCAGGTTG
ET1	TAGAGTGTGTCTACTTCTGCCA	TTCTTCCTCTCACTAACTGCTG
p16	CATGGTGCGCAGGTTCTTG	CTTCCAAGTCCATACGGAACAA
COL1	GTGCGATGACGTGATCTGTGA	CGGTGGTTTCTTGGTCGGT
COL3	TGCTGGTCCTGCTGGTCCTAAG	CCAGTAGCACCATCATTTCCACGAG
COL4	GGACTACCTGGAACAAAAGGG	GCCAAGTATCTCACCTGGATCA
SOD1	GATGACTTGGGCAAAGGTGGAAATG	CCAATTACACCACAAGCCAAACGAC
SOD2	CGCCCTGGAACCTCACATCAAC	AACGCCTCCTGGTACTTCTCCTC

Western Blot (WB)

The total protein of skin tissues or cells was extracted with RIPA lysis buffer. All lysates were centrifuged at 17 000 ×g for 10 minutes at 4 °C, the supernatant was transferred to the pre-cooled centrifuge tube, and the protein content was determined with a BCA protein assay kit. All proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene difluoride (PVDF) membranes. Blocked with 5% skimmed milk powder solution at room temperature for 1 h, the membranes were washed with Tween in Tris-buffered saline (TTBS) and incubated overnight at 4 °C with primary antibodies (1: 1000 dilution). Followed by incubation with horseradish peroxidase-conjugated secondary antibodies (1: 1000 dilution) at room temperature for 1 h. The blots in membranes were detected by the ChemiDoc Imaging Systems (Bio-Rad Laboratories, Inc.) and analyzed by the Image-J software.

Statistical Analysis

All statistical analyses were performed by SPSS software (Version 26.0, SPSS) and OriginPro Software (Version 2021, OriginLab). All data were given as mean values ± SD and comparisons among 3 or more groups were analyzed by one-way ANOVA followed by least-significant difference (LSD) tests. The results of P-value < .05 were considered statistically significant.

Results

Characterization of hucMSCs

The hucMSCs displayed positive staining of CD73 (99.73%), CD90 (99.96%), and CD105 (99.99%), as well as negative staining of CD34 (0.18%), CD45 (0.19%), and HLA-DR (0.57%; Supplementary Fig. S1).

Characterization of HA Gel

The HA gel displayed a typical loose and porous microstructure (Supplementary Fig. S2A and S2B). The degradation rate of the HA gel was increased with time to reach 92.67% at 48 h and 99.45% at 60 h (Supplementary Fig. S2C). The swelling ratio of the HA gel was rapidly increased within 0.5 h (from 0 to 22.13 ± 0.87) and was slightly increased at 1 h (23.44 ± 1.29) and 24 h (30.75 ± 2.72), and reached a maximum (32.59 ± 1.60) at 36 h (Supplementary Fig. S2D). The cell viability of hucMSCs under the HA gel environment was decreased at 24 and 48 h (P < .01 vs 0 h level; Supplementary Fig. S2E). The ratio of live cells (green fluorescence) was decreased at 24 h (68.34%) and 48 h (51.17%) and the ratio of dead cells (red fluorescence) was increased at 24 h (31.66%) and 48 h (48.83%; P < .05 vs 0 h level; Supplementary Fig. S2F and S2G).

HucMSCs-HA Gel Facilitated Foot Wound Healing of Type I and II Diabetic Rats

Both T1DM and T2DM rats showed a significant delay in foot wound closure, and the hucMSCs-HA gel treatment significantly accelerated wound healing in both T1DM and T2DM rats, while the HA treatment insufficiently repaired the wounds in T1DM rats (Fig. 1A and 1B). Histological analysis indicated that both T1DM and T2DM rats presented incomplete re-epithelialization and severe inflammatory cell infiltration, while the hucMSCs-HA gel completely healed the epidermal area without inflammatory cell infiltration in both T1DM and T2DM rats (Fig. 2A and 3A). Both T1DM and T2DM rats showed a significantly decrease in collagen deposition, while the HA gel and the hucMSCs-HA gel treatments significantly increased the collagen deposition in both T1DM and T2DM rats (P < .01 vs T1DM or T2DM) (Fig. 2B, 2C, 3B, and 3C). Both T1DM and T2DM rats showed significantly decreased expression of VEGFA and significantly reduced a number of blood vessels (P < .01 vs control), which were significantly restored by the hucMSCs-HA gel (P < .01 vs T1DM or T2DM) (Fig. 2D, 2E, 2F, 3D, 3E, and 3F). These results indicated that the hucMSCs-HA gel repaired DFU wounds in T1DM and T2DM rats and exerted better effects than the HA gel.

Figure 1.

Therapeutic effects of the hucMSCs-HA gel on foot wound healing of type I and II diabetic (T1DM and T2DM) rats at different time points. (A) Representative digital images of wound healing surfaces in type I and II diabetic rats, scale bar = 2 mm. (B) Line graph of average wound surface healing rate in type I and II diabetic rats. *P < .05 and **P < .01 vs the control group, ^#P < .05 and ^##P < .01 vs the model (T1DM or T2DM) group, ^$P < .05 and ^$$P < .01 vs the HA group.

Figure 2.

Histopathological features of diabetic foot wound closure in type I diabetic (T1DM) rats on day 21. (A) Representative micrographs of HE staining, scale bar = 500 μm (40×) and 50 μm (400×). (B) Representative micrographs of Masson’s trichrome stained sections, scale bar = 500 μm (40×) and 50 μm (400×). (C) Histogram of collagen deposition from Masson’s trichrome. (D) Representative micrographs of VEGFA antibody immunohistochemistry and blood vessels labeled by yellow allows, scale bar = 500 μm (40×) and 50 μm (400×). (E) Quantification of immunohistochemical staining intensity (VEGFA antibody). (F) Quantification of average number of blood vessels. Data were mean ± SD. *P < .05 and **P < .01 vs the control group, ^#P < .05 and ^##P < .01 vs the model group, ^$P < .05 and ^$$P < .01 vs the HA group.

Figure 3.

Histopathological features of diabetic foot wounds closure in type II diabetic (T2DM) rats on day 21. (A) Representative micrographs of H&E staining, scale bar = 500 μm (40×) and 50 μm (400×). (B) Representative micrographs of Masson’s trichrome stained sections, scale bar = 500 μm (40×) and 50 μm (400×). (C) Histogram of collagen deposition from Masson’s trichrome. (D) Representative micrographs of VEGFA antibody immunohistochemistry and blood vessels labeled by yellow allows, scale bar = 500 μm (40×) and 50 μm (400×). (E) Quantification of immunohistochemical staining intensity (VEGFA antibody). (F) Quantification of average number of blood vessels. Data were mean ± SD. *P < .05 and **P < .01 vs the control group, ^#P < .05 and ^##P < .01 vs the model group, ^$P < .05 and ^$$P < .01 vs the HA group.

Comparison of DFU Healing Effects Between the hucMSCs-HA and MSC-CM-HA Gels

A significant delay in foot wound closure was observed in T1DM rats, and the treatment with hucMSCs-HA gel and MSC-CM-HA gel significantly accelerated wound healing with comparable repair effects (Fig. 4A and 4C). T1DM rats presented a significant increase in the wounded epidermal area and a significant reduction in collagen deposition in the wound bed (Fig. 4B and 4D). By contrast, the hucMSCs-HA and MSC-CM-HA gels repaired the epidermal area and improved collagen deposition in T1DM rats (Fig. 4). The results demonstrated that the hucMSCs-HA and ­MSC-CM-HA gels exerted similar effects in DFU healing.

Figure 4.

Therapeutic effects of the hucMSCs-HA and MSC-CM-HA gels on healing of type I diabetic (T1DM) foot wounds at different time points. (A) Representative digital images of wound healing surfaces in SD rats on days 0, 3, 6, 9, 12, 15, 18, and 21, scale bar = 2 mm. (B) Representative micrographs of H&E staining and Masson’s trichrome stained sections of the wounds on day 21 after wound induction, scale bar = 500 μm (40×) and 50 μm (400×). (C) Line graph of average wound surface healing rate in SD rats. (D) Quantification of collagen deposition from Massonʹs trichrome staining results. Data were mean ± SD. *P < .05 and **P < .01 vs the control group, ^#P < .05 and ^##P < .01 vs the model (T1DM) group.

The Paracrine Effects of hucMSCs on HUVECs and HSFs

The cell viability assay showed that MSC-CM exerted increased the viability of HUVECs and HSFs under a high glucose environment (P < .01 vs model; Fig. 5A and 5B). The transwell assay showed that MSC-CM significantly promoted the vertical migration of HUVECs under a high glucose environment (Fig. 5C and 5D). The wound healing assay showed that MSC-CM significantly increased the mean wound healing rates of HUVECs and HSFs under a high glucose environment from 24 to 48 h (P < .01 vs model; Fig. 5E–5H).

Figure 5.

The effects of MSC-CM on the cell viability, wound healing, and cell migration of HUVECs and HSFs. (A) Cell viability of HUVECs was detected by MTT method. (B) Cell viability of HSFs was detected by MTT method. (C) and (D) showed the number of cell migration and the representative pictures of HUVECs in the transwell experiment (scale bar = 50 μm). (E) and (F) showed wound closure rate and representative pictures in the scratch test of HUVECs at 0, 12, 24, 36, and 48 h (scale bar = 50 μm). (G) and (H) showed wound healing rates and representative pictures in the scratch test of HSFs at 0, 12, 24, 36, and 48 h (scale bar = 50 μm). Data were mean ± SD, **P < .01 vs control level, and ^##P < .01 vs model level.

High glucose significantly increased the MDA and ROS levels of HSFs (P < .05 vs control), which were significantly reduced by MSC-CM (P < .05 vs model) (Supplementary Fig. S3A–S3C). High glucose severely disrupted the tube formation of HUVECs, which was obviously improved after MSC-CM treatment, as demonstrated by significantly increased numbers of nodes and junctions, and total tubule length (P < .05 vs model; Supplementary Fig. S3D–S3G). MSC-CM dramatically decreased the percentage of senescent HUVECs under a high glucose environment (P < .01 vs model; Supplementary Fig. S3H and S3I). These results confirmed that MSC-CM effectively promoted tubule formation of HUVECs and inhibited oxidative stress of HSFs and cell senescence of HUVECs under a high glucose environment. The effective MSC-CM suggested a paracrine action mode of hucMSCs.

Molecular Actions of MSC-CM on HUVECs and HSFs

The qRT-PCR results indicated that high glucose significantly upregulated TNF-α, IL-1β, IL-6, ET-1, and p16 mRNA expressions in HUVECs (P < .05 or P < .01 vs control), and downregulated COL1, COL3, COL4, SOD1, and SOD2 mRNA expressions in HSFs (P < .05 or P < .01 vs control), while these altered gene expressions were significantly reversed by MSC-CM (P < .05 or P < .01 vs model; Fig. 6A and 6B). The Western blot results showed that MSC-CM significantly reversed the abnormal expression of PCNA protein in HUVECs and HSFs under a high glucose environment (Fig. 6C and 6D).

Figure 6.

The role of MSC-CM treatment in mRNA and protein expressions of high glucose-induced HUVECs and HSFs. (A) mRNA expressions of inflammation (TNF-α, IL-1β, and IL-6), tube formation (ET-1), and senescence (p16) related genes in HUVECs. (B) mRNA expressions of collagen synthesis (COL1, COL3, and COL4), oxidative stress (SOD1 and SOD2) related genes in HSFs. (C) Representative protein expression images of PCNA (HUVECs and HSFs). (D) Relative protein levels of PCNA/β-actin (HUVECs and HSFs). Data was presented as mean ± SD, *P < .05 and **P < .01 vs control level, ^#P < .05 and ^##P < .01 vs model level.

HucMSCs-HA Gel Promoted DFU Healing in Type I Diabetic Rats Through Activation of MAPK and Akt Signaling Pathways

To explore the underlying mechanism of the hucMSCs-HA gel, the tissue repair-related MAPK and Akt signaling pathways were investigated. The Western blot results indicated that the hucMSCs-HA gel significantly increased the protein levels of p-p38, t-p38, p-JNK, p-ERK1/2, t-ERK1/2, and p-Akt when compared with the T1DM group (P < .05 or P < .01 vs T1DM), indicating the activation of the MAPK and Akt signaling pathways (Fig. 7).

Figure 7.

Involvement of MAPK and Akt signaling pathways in the pro-healing property of hucMSCs-HA. Proteins were extracted from skin tissues in type I diabetic rats (including the control, T1DM, HA, and MSCHA groups). (A) Total and phosphorylated p38, JNK, ERK1/2, and Akt were assessed by Western blotting assay. Among them, representative images from 3 individual rats (marked as 1^#, 2^#, and 3^#) were shown. (B) Relative protein levels were also analyzed. Data were mean ± SD. *P < .05 and **P < .01 vs the control group, ^#P < .05 and ^##P < .01 vs the T1DM group, ^$P < .05 and ^$$P < .01 vs the HA group.

Discussion

The wound healing is a complex process and pivotal therapeutic target of DFU, for which MSCs therapy has great potential due to its pro-regenerative improvement in tissue cells (eg, endotheliocytes and fibroblasts). Current studies indicate that MSCs could enhance the proliferation and migration of damaged endothelial cells and fibroblasts,^25,28 and promote collagen formation and angiogenesis.²⁹ The commonly used approaches of MSCs for DFU treatment are systematical injection and topical injection.³⁰ The systematical (intravenous) injection facilitates the delivery of MSCs throughout the body, but only a few cells can get through the lungs and a much lesser part of them can arrive at the distal wound sites, resulting in inefficient delivery of cells.^31,32 Moreover, the systematical delivery of MSCs may bring adverse events, such as whole body urticaria, diarrhea, oral ulceration, and elevation of serum creatinine level.³³ The topical injection directly delivers MSCs to wound sites and thereby avoids cell loss among organs and blood vessels.³² However, the topical injection may create secondary damage at wound sites and the efficiency may be influenced by the local ischemia and hypoxia under a high glucose environment.^27,34 To overcome the shortcomings of systematical or topical injection of MSCs, this study tried external administration by directly covering MSCs-gel mixture (hucMSCs-HA gel) on DFU sites and obtained satisfactory results. HA gel was used as the carrier of hucMSCs, since it was cheap and biocompatible, with FDA approved clinical use, and commercially available.^35,36 Previous studies have applied HA gel to carry MSCs in DFU treatment in different ways from this present study. For instance, topical injection of MSCs with HA gel accelerated wound healing rates of refractory diabetic foot ulcers in patients,³⁷ and human adipose-derived stem cells (hADSCs)-containing HA-enhanced diabetic wound healing.³⁸ In this study, the HA gel with a loose and porous microstructure (Supplementary Fig. S2A and S2B) was used to encapsulate hucMSCs, which provided a buffer place for maintaining cell viability within 48 h (Supplementary Fig. S2E and S2F). Without the HA gel, there would be an increased occurrence of cell death within 48 h. Our findings suggested that the HA gel exhibited good biocompatibility with the MSCs and did not contribute to additional cell death, consistent with reports from other studies.^37,38 Within 48 h, hucMSCs and their paracrine products might be continuously released from the HA gel into the foot wounds, thus maintaining the effects of hucMSCs on DFU healing.

In vivo, we established DFU models on both type I and II diabetic rats to evaluate the repairing effects of hucMSCs-HA gel. When compared with the blank HA gel in the HA group, the hucMSCs-HA gel in the MSC-HA group obviously reduced hemorrhage and inflammatory response and promoted granulation tissue formation, collagen deposition, angiogenesis, and re-epithelialization, indicating hucMSCs-enhanced foot wound healing rate of diabetic rats (Fig. 1–3). In vitro, we found that hucMSCs significantly recovered the HG-caused abnormalities of HUVECs and HSFs through MSC-CM, as determined by cell viability, wound healing, migration, tube formation, senescence, ROS, and MDA level assays (Fig. 5 and Supplementary Fig. S3). MSC-CM significantly restored the HG-caused abnormal expressions of inflammatory genes (TNF-α, IL-1β, and IL-6), vasoconstrictive gene (ET-1), senescent gene (p16), and proliferative protein (PCNA) of HUVECs, and also significantly restored the abnormal expressions of pro-fibrotic genes (COL1, COL3, COL4), antioxidant genes (SOD1 and SOD2), and proliferative protein (PCNA) of HSFs (Fig. 6). In sum, this study demonstrated dramatic wound healing efficacy of external application of hucMSCs-HA gel on DFU and discovered its paracrine mechanism of action in regard to the improvement of angiogenesis, collagen deposition, and dermal regeneration. Furthermore, the paracrine effects of hucMSCs also involved anti-senescence of HUVECs and anti-oxidative stress of HSFs.

To confirm the paracrine action of hucMSCs in the HA gel, the MSC-CM-HA gel group was set. Foot wound healing in diabetic rats in both the MSC-CM-HA gel group and hucMSCs-HA gel group had been significantly promoted, evidenced by improved re-epithelialization and collagen deposition, without any significant difference between these 2 groups (Fig. 4). This data implied the paracrine action of hucMSCs in the HA gel. The external use of hucMSCs-HA gel might have the advantages of longer-term and more direct paracrine effects due to the longer time survival of hucMSCs in HA gel as well as the direct delivery of paracrine factors of hucMSCs on the wound sites. Moreover, direct cover of hucMSCs-HA provided a moist environment, avoided injection injury, and prevented exposure infections, which benefited wound repair and regeneration.²⁷ In terms of wound healing time and convenient administration, the external use of hucMSCs-HA gel resulted in a better efficacy than other MSCs injection treatments, since the wound closure time in this study was >3 days shorter than that of other reports using MSCs injection.³⁹ The novelty of this study lies in that: (1) noninvasive application of the hucMSCs-HA gel for treating DFU in type I and II diabetic rats; (2) determination of the paracrine action of MSCs in treating DFU; and (3) clarification of the MAPK and Akt signaling pathways-mediated mechanism of hucMSCs for wound healing in DFU. The shortcomings of conventional application of MSCs can be addressed by this new approach, overcoming potential side effects of MSCs injection as well as the secondary damage in patients receiving autologous MSCs therapy.

The actions of MSCs in wound healing contain 2 main ways: indirect paracrine and direct replacement.⁴⁰ The paracrine of MSCs refers to the release of various secretions (growth factors, chemokines, immune factors, and exosomes, etc.), while direct replacement is the differentiation of MSCs into tissue cells to repair the wound sites.⁴¹ These 2 actions might coexist and function together, but more evidences suggested that the paracrine would be the major way because MSCs hardly survived longer in the recipient tissues.^41,42 In this study, the paracrine action of hucMSCs in the HA gel was confirmed, and the paracrine molecules with targeted pathways are listed (Table 2). The secretome of MSCs (VEGF, TGF-β1, EGF, miR-146a, and LncRNA H19 etc.) could recruit endotheliocytes and fibroblasts to rapidly proliferate and migrate to wound sites of DFU by activating the MAPK and Akt pathway, enhancing angiogenesis and promoting collagen synthesis. In the MAPK pathway, p38 acts as a growth kinase that promotes the closure of the wound through increased migration, proliferation, and collagen production of fibroblasts,⁶⁶ ERK1/2 and JNK favor the proliferation, migration, neovascularization of endotheliocytes.^46,67 Akt is an important signaling regulator for a wide range of cellular functions like cell proliferation, angiogenesis, migration, senescence, etc.⁶⁸ In this study, the paracrine mechanism that the hucMSCs-HA gel promoted DFU healing was mediated by the activation of MAPK and Akt pathways, in which the phosphorylated Akt, p38, ERK1/2, and JNK (Fig. 7) participated in the skin regeneration under a high glucose environment (Figs. 5 and 6, and Supplementary Fig. S3). Further studies are needed for in-depth clarification of the in vivo paracrine mechanism of hucMSCs-HA gel.

Table 2.

Molecular mechanisms of MSCs in treating endotheliocytes and fibroblasts by paracrine action.

Paracrine componentss	Target cell	Signaling pathway	Changes in gene expression.	Outcomes	Ref.
miR-17-5p	Endotheliocytes	PTEN/AKT/HIF-1α/VEGF pathway ↑	PTEN ↓, p-AKT ↑, HIF-1α ↑, VEGF ↑	Proliferation ↑, migration ↑, tube formation ↑, enescence ↓	43
IL1ra, IL10, IL13, TGF-β1, BDNF, GDNF, CNTF, HGF, FGF, EGF, VEGF,	Endotheliocytes	JAK-STAT pathway ↑, PI3K-Akt pathway ↑, MAPK pathway ↑	TNF-α ↓, IL-1β↓, Selectin-E ↓, p65 ↓	Inflammation ↓	44
Exosomes-carried miR-126	Endotheliocytes	PI3K/AKT pathway ↑	PTEN ↓, p-AKT ↑	Proliferation ↑ migration ↑, tube formation ↑	45
Microvesicles	Endotheliocytes	AKT pathway ↑ ERK pathway ↑	Cyclin D2 ↑, cyclin A1 ↑, c-Myc ↑, VEGFA ↑, VEGFR2 ↑, FGF2 ↑, HIF-1A ↑, PDGFA ↑, Cox-2 ↑, ITGB1 ↑, CXCL16 ↑, cyclin D1 ↑, cyclin A2 ↑, PDGFR ↑	Proliferation ↑, migration ↑, tube formation ↑	46
IL-6; VEGF, MCP-1, angiogenin	Endotheliocytes	PI3K/Akt pathway ↑	p-Akt ↑, p-ERK ↑	Proliferation ↑, migration ↑, tube formation ↑, invasion ↑	25
VEGF	Endotheliocytes	VEGF/VEGFR2 pathway ↑	VEGFR2 ↑	Proliferation ↑, apoptosis ↓, inflammation ↓, tube formation ↑	47
SDF-1, MCP-1, TGF-β, PDGF-BB, VEGF, VCAM-1, MCP-1	Endotheliocytes	SDF-1/CXCR4 axis ↑, MCP-1/CCR2 axis ↑	SDF-1↑, CXCR4 ↑, MCP-1 ↑, CCR2 ↑	Proliferation ↑, migration ↑, tube formation ↑	48
exosomes-carried miR-135b-5p and miR-499a-3p	Endotheliocytes	miR-135b-5p/miR-499a-3p-MEF2C axis ↓	MEF2C ↓	Proliferation ↑ migration ↑, tube formation ↑	49
VEGFA, bFGF, Ang-1, aFGF, PDGF	Endotheliocytes	integrin β1/ ERK1/2/HIF-1α/ VEGFA pathway ↑	p-integrin β1 ↑, ERK1/2 ↑, HIF-1α↑, VEGFA ↑	Proliferation ↑ migration ↑, tube formation ↑, senescence ↓	50
Extracellular vesicles	Endotheliocytes	PI3K/AKT/mTOR/HIF-1α pathway ↑	PI3K ↑, AKT ↑, mTOR ↑, p-AKT ↑, p-mTOR ↑, HIF-1α ↑, VEGF ↑	Proliferation ↑ migration ↑, tube formation ↑	51
Extracellular vesicles-carried miR-129	Endotheliocytes	PTEN/PI3K/AKT pathway ↑	TRAF6 ↓, PTEN ↓,	Proliferation ↑ migration ↑, tube formation ↑	45
Exosomes-carried HMGB1	Endotheliocytes	HMGB1/HIF-1α/VEGF pathway ↑, JNK pathway ↑	p-JNK ↑, HIF-1α ↑, VEGF ↑	Proliferation ↑ migration ↑, tube formation ↑	52
FGF2, VEGFA, TGF-β	Endotheliocytes		PLGF ↑, SCF ↑ , VEGFR2 ↑	Proliferation ↑ migration ↑, tube formation ↑	53
LncRNA H19	Endotheliocytes	PTEN/AKT/eNOS pathway ↑	miR-211-3p ↑, PTEN ↓, p-AKT ↑, p-eNOS ↑, PDGF ↑, EGF ↑, bFGF ↑, VEGF ↑, ANG1 ↑	Proliferation ↑ migration ↑, tube formation ↑, invasion ↑	54
lncRNA H19	Endotheliocytes	lncRNA H19/miR-152-3p/PTEN axis ↑ PTEN/PI3K/AKT pathway ↓	PTEN ↑, miR-152-3p ↑, p85 PI3K ↓, AKT ↓	Proliferation ↑, migration ↑, apoptosis ↓, inflammation ↓	55
miR-146a	Endotheliocytes	miR-146a/Src axis ↓	p-Src ↓, p-VE-cadherin ↓, p-caveolin-1 ↓, SASP ↓, p16, ↓ p21 ↓, p53 ↓	Proliferation ↑ migration ↑, tube formation ↑, senescence ↓	56
Exosomes	Endotheliocytes		NOX1↓, NOX4 ↓, cyclin D1 ↓, cyclin D3 ↓, IL-1β ↓, IL-6 ↓, TNF-α ↓	Proliferation ↑, angiogenesis ↑, oxidative stress ↓, inflammation ↓	57
Exosomes	Endotheliocytes	PI3K/AKT/eNOS pathway ↑	PTEN ↓, p-AKT ↑, p-PI3K ↑, p-eNOS ↑, cyclin D1 ↑, cyclin D3 ↑, VEGF↑	Proliferation ↑ migration ↑, tube formation ↑ invasion ↑, nitric oxide ↑	58
Exosomes	Fibroblasts	ERK/MAPK pathway ↑	MMP3 ↑, TIMP1 ↑	Proliferation ↑, migration ↑	59
Exosomes	Fibroblasts	PI3K/Akt pathway ↑	bFGF ↑, TGF-β1 ↑, collagen I ↑, collagen III ↑,	Collagen synthesis ↑	60
Extracellular vesicles-carried miR-27b	Fibroblasts	ITCH/JUNB/IRE1α pathway ↑	ITCH ↓, JUNB ↑, IRE1α ↑	Proliferation ↑ migration ↑	61
IL-8, IL-6, TGF-β , TNFRI, VEGF, EGF	Fibroblasts	TGF-β/SMAD2 pathway ↑	p-SMAD2 ↑	Proliferation ↑, migration ↑	62
Extracellular vesicles-carried miR-27b	Fibroblasts	ITCH/JUNB/IRE1α pathway ↑	ITCH ↓, JUNB ↑, IRE1α ↑	Proliferation ↑ migration ↑	61
Wnt4, G-CSF, PDGF-BB, VEGF, MCP-1, IL-6, IL-8	Fibroblasts	Wnt4/β-catenin pathway ↑ AKT pathway ↑	β-catenin ↑, CK19 ↑, PCNA ↑, collagen I ↑	Proliferation ↑, migration ↑	63
TGF-β1, IL-6, IL-8, MCP-1, RANTES, HGF, SPARC, IGFBP-7	Fibroblasts	HGF/c-met axis ↑	HGF ↑, c-met ↑	Migration ↑	48
bFGF, IGF-1, VEGFA, TGFβ-2, TGFβ-3, IL-1β, IL-6, IL-8	Fibroblasts	TGF-β/SMAD pathway ↑	p-SMAD2/3 ↑, SMAD7 ↑, p21, p16	Oxidative stress ↓ ROS overproduction ↓ senescence ↓ myofibroblast formation ↓	64
EGF, bFGF, TGF-b, PDGF, HGF, EGF, pro-collagen I	Fibroblasts		Collagen I ↑, collagen III ↑, fibronectin ↑, flastin ↑	Proliferation ↑, migration ↑	65
VEGF, FGF2, TGF-β1, IL6	Fibroblasts	TGF-β/SMAD2 pathway ↑, PI3K/AKT pathway ↑	Fibronectin ↑, p-AKT ↑, p-SMAD2 ↑	Proliferation ↑ migration ↑ collagen synthesis ↑	64

Conclusions

This study reported that the external noninvasive use of hucMSCs-HA gel exerted therapeutic effects in treating DFU. The HA gel was a loose, porous, degradable, and swellable microstructure that provided a compatible microenvironment for hucMSCs. The in vivo data demonstrated that the hucMSCs-HA gel promoted re-epithelialization, collagen deposition, and angiogenesis of DFU wounds in both type I and II diabetic rats. Both the hucMSCs-HA and MSC-CM-HA gels exerted similar effects on DFU healing, suggesting that hucMSCs acted in a paracrine mode. The in vitro data further confirmed the paracrine beneficial effects of hucMSCs on HUVECs and HSFs by restoring the abnormalities of cell viability, migration, oxidative stress, angiogenesis, and cell senescence under a high glucose environment. Mechanistically, activation of the MAPK and Akt signaling pathways was involved in the paracrine mechanism of hucMSCs. Further studies are needed to elucidate in-depth mechanisms underlying the paracrine effects of hucMSCs-HA gel. In sum, our findings provided a new and promising strategy of MSCs-based therapy for DFU treatment in clinic.

Funding

This work was supported by National Natural Science Foundation of China (Grant Nos. 81973767 and 82274239) and Hangzhou Science and Technology Development Program (Grant No. 20180533B38).

Conflict of Interest

The authors have declared that no competing interest exists.

Author Contributions

J.A.C.: conception and design, manuscript writing and revision, collection and/or assembly of data, data analysis and interpretation, conducted the main work of this study. Y.L.: conducted to the animal and cellular experiments, collection and/or assembly of data. J.W.Z.: contributed to the cellular and molecular experiments. Y.P.Y.: manuscript revision, data analysis. H.W.L.: conducted to the animal experiment. L.Y.: contributed to the hucMSCs preparation and quality control. T.L.: contributed to the molecular experiments. L.Z.: conception and design, manuscript revision. L.T.S.: conception and design, provision of study material, final approval of manuscript. H.W.: conception and design, financial support, final approval of manuscript.

Data Availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Ethics Approval and Consent to Participate

Great care was taken to minimize their suffering and this study was approved by the Animal Ethics Committee of Zhejiang Chinese Medical University (Animal Ethics No: IACUC-20190408-09).