[Effects of exosomes from hepatocyte growth factor-modified human adipose mesenchymal stem cells on full-thickness skin defect in diabetic mice].

Abstract

Objective: To investigate the effects and mechanism of exosomes from hepatocyte growth factor (HGF)-modified human adipose mesenchymal stem cells (ADSCs) on full-thickness skin defect wounds in diabetic mice. Methods: The experimental study method was adopted. Discarded adipose tissue of 3 healthy females (10-25 years old) who underwent abdominal surgery in the Department of Plastic Surgery of First Affiliated Hospital of Air Force Medical University from February to May 2021 was collected, and primary ADSCs were obtained by collagenase digestion method and cultured for 7 days. Cell morphology was observed by inverted phase contrast microscope. The ADSCs of third passage were transfected with HGF lentivirus and cultured for 5 days, and then the fluorescence of cells was observed by imaging system and the transfection rate was calculated. The exosomes of ADSCs of the third to sixth passages and the HGF transfected ADSCs of the third to sixth passages were extracted by density gradient centrifugation, respectively, and named, ADSC exosomes and HGF-ADSC exosomes. The microscopic morphology of exosomes was observed by transmission electron microscopy, and the positive expressions of CD9, CD63, and CD81 of exosomes were detected by flow cytometry, respectively. Twenty-four 6-week-old male Kunming mice were selected to make the diabetic models, and full-thickness skin defect wounds were made on the backs of mice. According to the random number table method, the mice were divided into phosphate buffer solution (PBS) group, HGF alone group, ADSC exosome alone group, and HGF-ADSC exosome group, with 6 mice in each group, and treated accordingly. On post injury day (PID) 3, 7, 10, and 14, the wounds were observed and the wound healing rate was calculated; the blood flow intensity of wound base was detected by Doppler flowmeter and the ratio of relative blood flow intensity on PID 10 was calculated. On PID 10, the number of Ki67 positive cells in wounds was detected by immunofluorescence method, and the number of new-vascularity of CD31 positive staining and tubular neovascularization in the wounds was detected by immunohistochemistry method; the protein expressions of protein endothelial nitric oxide synthase (eNOS), phosphatidylinositol 3-kinase (PI3K), phosphorylated PI3K (p-PI3K), protein kinase B (Akt) and phosphorylated Akt (p-Akt) in wounds were detected by Western blotting, and the ratios of p-PI3K to PI3K and p-Akt to Akt were calculated. On PID 14, the defect length and collagen regeneration of wound skin tissue were detected by hematoxylin and eosin staining and Masson staining, respectively, and the collagen volume fraction (CVF) was calculated. The number of samples is 3 in all cases. Data were statistically analyzed with repeated measurement analysis of variance, one-way analysis of variance, and Tukey test. Results: After 7 days of culture, the primary ADSCs were spindle shaped and arranged in vortex shape after dense growth. After 5 days of culture, HGF transfected ADSCs of the third passage carried green fluorescence, and the transfection rate was 85%. The ADSC exosomes and HGF-ADSC exosomes were similar in microscopic morphology, showing vesicular structures with an average particle size of 103 nm and 98 nm respectively, and both were CD9, CD63, and CD81 positive. On PID 3, the wounds of mice in the 4 groups were all red and swollen, with a small amount of exudate. On PID 7, the wounds of HGF-ADSC exosome group were gradually reduced, while the wounds of the other three groups were not significantly reduced. On PID 10, the wounds in the 4 groups were all reduced and scabbed. On PID 14, the wounds in HGF-ADSC exosome group were basically healed, while the residual wounds were found in the other three groups. On PID 3, the healing rates of wounds in the four groups were similar (P>0.05); On PID 7 and 10, the wound healing rates in HGF-ADSC exosome group were significantly higher than those in PBS group, HGF alone group, and ADSC exosome alone group, respectively (with q values of 13.11, 13.11, 11.89, 12.85, 11.28, and 7.74, respectively, all P<0.01); on PID 14, the wound healing rate in HGF-ADSC exosome group was significantly higher than that in PBS group, HGF alone group, and ADSC exosome alone group (with q values of 15.50, 11.64, and 6.36, respectively, all P<0.01). On PID 3, there was no obvious blood supply in wound base of mice in the 4 groups. On PID 7, microvessels began to form in the wound base of HGF-ADSC exosome group, while the wound base of the other three groups was only congested at the wound edge. On PID 10, microvessel formation in wound base was observed in the other 3 groups except in PBS group, which had no obvious blood supply. On PID 14, the blood flow intensity of wound base in HGF-ADSC exosome group was stronger than that in the other 3 groups, and the distribution was uniform. On PID 10, the ratio of wound base relative blood flow intensity in HGF-ADSC exosome group was significantly higher than that in PBS group, HGF alone group, and ADSC exosome alone group (with q values of 23.73, 19.32, and 9.48, respectively, all P<0.01); The numbers of Ki67-positive cells and new-vascularity of wounds in HGF-ADSC exosome group were significantly higher than those in PBS group, HGF alone group, and ADSC exosome alone group, respectively (with q values of 19.58, 18.20, 11.04, 20.68, 13.79, and 8.12, respectively, P<0.01). On PID 10, the protein expression level of eNOS of wounds in HGF-ADSC exosome group was higher than that in PBS group, HGF alone group, and ADSC exosome alone group (with q values of 53.23, 42.54, and 26.54, respectively, all P<0.01); the ratio of p-PI3K to PI3K and the ratio of p-Akt to Akt of wounds in HGF-ADSC exosome group were significantly higher than those in PBS group, HGF alone group, and ADSC exosome alone group, respectively (with q values of 16.11, 11.78, 6.08, 65.54, 31.63, and 37.86, respectively, P<0.01). On PID 14, the length of skin tissue defect in the wounds of HGF-ADSC exosome group was shorter than that in PBS group, HGF alone group, and ADSC exosome alone group (with q values of 20.51, 18.50, and 11.99, respectively, all P<0.01); the CVF of wounds in HGF-ADSC exosome group was significantly higher than that in PBS group, HGF alone group and ADSC exosome alone group (with q values of 31.31, 28.52, and 12.35, respectively, all P<0.01). Conclusions: Human HGF-ADSC exosomes can significantly promote wound healing in diabetic mice by increasing neovascularization in wound tissue, and the mechanism may be related to the increased expression of eNOS in wounds by activating PI3K/Akt signaling pathway.