Exosomes from adipose-derived stem cells overexpressing Nrf2 accelerate cutaneous wound healing by promoting vascularization in a diabetic foot ulcer rat model

Abstract

Diabetic foot ulcers (DFU) increase the risks of infection and amputation in patients with diabetes mellitus (DM). The impaired function and senescence of endothelial progenitor cells (EPCs) and high glucose-induced ROS likely exacerbate DFUs. We assessed EPCs in 60 patients with DM in a hospital or primary care setting. We also evaluated the therapeutic effects of exosomes secreted from adipose-derived stem cells (ADSCs) on stress-mediated senescence of EPCs induced by high glucose. Additionally, the effects of exosomes and Nrf2 overexpression in ADSCs were investigated in vitro and in vivo in a diabetic rat model. We found that ADSCs that secreted exosomes promoted proliferation and angiopoiesis in EPCs in a high glucose environment and that overexpression of Nrf2 increased this protective effect. Wounds in the feet of diabetic rats had a significantly reduced ulcerated area when treated with exosomes from ADSCs overexpressing Nrf2. Increased granulation tissue formation, angiogenesis, and levels of growth factor expression as well as reduced levels of inflammation and oxidative stress-related proteins were detected in wound beds. Our data suggest that exosomes from ADSCs can potentially promote wound healing, particularly when overexpressing Nrf2 and therefore that the transplantation of exosomes may be suitable for clinical application in the treatment of DFUs.

Fig. 1. Characterization of endothelial progenitor cells (EPCs) by confocal microscopy and flow cytometric analysis.

a EPCs have a typical cobblestone-like morphology. b EPCs were stained with FITC-labeled or DiI-labeled UEA-1 and Ac-LDL and then detected using a confocal microscope. Scale bar, 20 μm. c Flow cytometric analysis of EPC surface markers (CD34, KDR, CD133, and CD14). The results presented are typical of those obtained from three separate experiments

Fig. 2. Expression of SMP30 and oxidative stress-related proteins on circulating CD34-positive cells of patients with diabetes mellitus (DM) and healthy controls.

a Expression of senescence marker protein 30 (SMP30) was measured by western blot. GAPDH served as an internal control. The relative protein levels were analyzed, and data are presented as the mean ± SD (n = 3). ***P < 0.001 vs. normal group. b Phosphorylation of VEGFR2 was measured by western blot. GAPDH served as an internal control. The relative protein levels were analyzed, and data are presented as the mean ± SD (n = 3). ***P < 0.001 vs. normal group. c Levels of the ROS-related proteins NOX1 and NOX4 were measured by western blot. GAPDH served as an internal control. The relative protein levels were analyzed, and data are presented as the mean ± SD (n = 3). ***P < 0.001 vs. normal group. d The tube formation capability of endothelial progenitor cells (EPCs) from DM patients or subjects without DM was measured. The results show that the tube formation capability of EPCs decreased in DM patients

Fig. 3. High glucose (HG) accelerates stress-induced premature senescence in endothelial progenitor cells (EPCs).

a EPCs from healthy subjects incubated with increasing concentrations of glucose (15 and 30 mM) for 24–72 h. Cell viability was measured by the MTT assay. The percentage of viable cells was analyzed, and data are presented as the mean ± SD (n = 3). *P < 0.05, **P < 0.01 vs. normal glucose (NG, 5.5 mM). b The intracellular ROS level of EPCs was determined by immunofluorescence using DCH-DA stain after exposure to HG (30 mM) for different lengths of time. Representative immunofluorescence images of ROS formation (red fluorescence) and DAPI (blue fluorescence) in EPCs. (Magnification ×200). Scale bar, 40 μm. c The average fluorescence intensity of DHE in each group. n = 5, **P < 0.01, ***P < 0.001 vs. NG. d–i Western blot analysis shows the protein expression levels of SMP30, VEGF, VEGFR2, and NOX1/4. GAPDH served as an internal control. The relative protein levels were analyzed, and data are presented as the mean ± SD (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001 vs. NG group. j Apoptosis was determined by Annexin-V/PI staining after induction with 30 mM HG for 24–72 h. Percentage of apoptotic cells was measured, and data are presented as the mean ± SD (n = 3). ***P < 0.001 vs. NG glucose. k–m Inflammatory cytokine levels of IL-1β, IL-6, and TNF-α were measured by ELISA. The relative protein levels were analyzed, and data are presented as the mean ± SD (n = 3). ***P < 0.001 vs. NG group

Fig. 4. Exosomes derived from adipose-derived stem cells (ADSCs) decreased endothelial progenitor cell (EPC) senescence under high glucose (HG) conditions.

a The apoptosis rate of EPCs was determined by Annexin-V/PI staining after 24 h induction with HG (30 mM) with or without GW4869 (2.5 μM) pretreatment for 8 h. b The percentage of apoptotic cells was measured, and data are presented as the mean ± SD (n = 3). ***P < 0.001 vs. HG; ^###P < 0.001 vs. HG + ADSCs. c The tube formation capability increased in EPCs stimulated with ADSCs, but the promoted effects decreased after pretreatment with GW4869 (2.5 μM) for 8 h. d The intracellular ROS levels were determined by immunofluorescence using DCH-DA stain. (×200). Scale bar, 20 μm. e The average fluorescence intensity of DHE in each group. n = 5, ***P < 0.001 vs. HG; ^###P < 0.001 vs. HG + ADSCs. f–k Western blot analysis shows the protein expression levels of SMP30, VEGF, VEGFR2, NOX1, and NOX4. GAPDH served as an internal control. The relative protein level was analyzed, and data are presented as the mean ± SD (n = 3). ***P < 0.001 vs. HG; ^###P < 0.001 vs. HG + ADSCs. l–n Inflammatory cytokine levels of IL-1β, IL-6, and TNF-α were measured by ELISA. The relative protein levels were analyzed, and data are presented as the mean ± SD (n = 3). ***P < 0.001 vs. HG; ^###P < 0.001 vs. HG + ADSCs

Fig. 5. Isolation and identification of exosomes from adipose-derived stem cell (ADSC) culture medium.

a The ultrastructure of ADSC-derived exosomes by transmission electron microscopy. Scale bar, 100 nm. b Size distribution of Exo-ADSCs determined by dynamic light scattering. c Expression of the exosome markers CD4, CD63, and TSG101 and β-actin confirmed by immunoblotting. Exosome lysate was loaded into the left lane and cell lysate into the right. d Representative micrographs of EPCs or EPCs incubated with Dil-labeled Exo (+ExoDil) or with Dil-labeled EVs denatured by boiling (+dExo-Dil). Scale bar, 10 μm

Fig. 6. Exosomes (Exos) derived from adipose-derived stem cells (ADSCs) overexpressing Nrf2 prevent high glucose (HG)-induced senescence by inhibiting ROS and inflammatory cytokine expression.

a,b The level of Nrf2 in exosomes secreted from ADSCs with or without Nrf2 overexpression. c Proliferation of EPC cells was analyzed by the MTT assay with different concentrations of Exos from ADSCs (ADSC-Exo) or Nrf2 overexpression ADSCs (Nrf2-ADSC-Exo) after exposure to high glucose (HG, 30 mM) for 48 h. d Tube formation capability detected in endothelial progenitor cells (EPCs) stimulated with 50 μg/ml ADSC-Exo or Nrf2-ADSC-Exo for 12 h under HG (30 mM) condition. e–j Western blot analysis and protein expression levels of SMP30, VEGF, VEGFR, and NOX1/4 after stimulation with ADSC-Exo or Nrf2-ADSC-Exo for 12 h under HG (30 mM) condition. GAPDH served as an internal control. The relative protein levels were analyzed, and data are presented as the mean ± SD (n = 3). ***P < 0.001 vs. normal group. k–m Inflammatory cytokine levels of IL-1β, IL-6, and TNF-α were measured by ELISA. The relative protein levels were analyzed, and data are presented as the mean ± SD (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001 vs. NG group

Fig. 7. Comparison between treatments by EPCs combined with or without exosomes from adipose-derived stem cells (ADSCs) and Nrf2-ADSCs.

a Representative images of full-thickness skin defects in a diabetic rat model treated with EPCs alone or combined with exosomes derived from ADSCs (ADSCs-Exos) or ADSCs overexpressing Nrf2 (Nrf2-ADSCs-Exo) for 0, 7, and 14 days postoperatively. Scale bar: 10 mm. Percentage wound closure of different groups. *P < 0.05 compared with control. b Granulation tissues in the different treatment groups were measured with H&E stain. c Collagen deposition and tissue fibrosis were assessed with Masson Trichrome staining. d Intracellular ROS levels were determined by immunofluorescence using DCH-DA stain. Scale bar, 40 μm. e Immunofluorescence with CD31 staining shows microvascular formation. Scale bar, 40 μm. The relative protein levels were analyzed, and data are presented as the mean ± SD (n = 6). *P < 0.05, **P < 0.01, ***P < 0.001 vs. PBS; ^#P < 0.05, ^###P < 0.001 vs. EPCs