Mesenchymal stem cell culture
Mesenchymal stem cells (MSCs) were provided by the Guangdong umbilical blood hematopoietic stem cell bank (Guangzhou, China). The phenotype of the MSCs was verified by flow cytometry, cell surface antigen markers including CD34, CD45, CD73, CD90, and CD105. MSCs were cultured in serum-free Dulbecco’s Modified Essential Medium (DMEM)-F12 (HyClone, Logan, UT, USA) supplemented with either 2 mg/ml glucose (Control Group), 2 mg/ml glucose plus 20 ng/ml VCAM-1 (ADP5, R&D, USA) (VCAM-1 Group), 2 mg/ml glucose plus 20 ng/ml TNF-α (T6674, Sigma, Germany) (TNF-α Group), or 2 mg/ml glucose plus 20 ng/ml IL-6 (200-06-20, PeproTech, USA) (IL-6 Group). The culture supernatant was collected after 48 hours of incubation.
Isolation and purification of exosomes from MSCs
Exosomes were isolated from the collected culture supernatant using ultracentrifugation by referring literatures. Briefly, to remove cellular debris and proteins, the supernatant was sequentially centrifuged at 300g for 5 min, 2000g for 10 min, and then 10000g for 20 min. The supernatant was then filtered using 0.22 μm filter (UFC810096-1, Merck Millipore, Germany) to remove particles with over 0.22 μm in size. The filtrate was then ultra-centrifuged at 100000g for 120 min at 4°C (Optima XE-90 ultracentrifuge with a swing rotor, Type 70Ti; Beckman Coulter, CA, USA). The enriched MSCs-exo was resuspended in PBS for further studies.
Morphological visualization using transmission electron microscopy (TEM)
To analyze the morphological features of MSCs-exo, transmission electron microscopy (JEOL 2100F, Japan) was used here. For sample preparation, the MSCs-exo suspension was dropped onto the copper grid and deposited for 1 min, following by treatment of 10 μl uranyl acetate (phosphotungstic acid) for 1 min. Then, MSCs-exo sample stood by for 20 min in a dry environment, and then washed with 100 μl PBS three times for 2 min before TEM observation.
Nanoparticle tracking analysis (NTA)
The MSCs-exo size and concentration were assessed using NanoSight LM10 system (Malvern). All the measurements were performed in triplicate at room temperature, with the detection threshold fixed to 20 - 300 particles per field of view and the camera level set at standard mode for 1 min. Data analyses were performed using NanoSight NTA software.
Sequencing analysis of exosomal miRNA
To reveal the miRNA components of exosomes, the MSCs-exo (8×1010) was analyzed by BGISEQ-500 technology. Then the acquired data were compared with miRBase and other non-coding databases. We used Transcripts Per Kilobase Million (TPM) to normalize miRNA expression and predict target genes using RNAhybrid, miRanda, and TargetScan. Functional heat-maps were used to exhibit hierarchical clustering analysis of miRNA expressions. The hypergeometric test was then used to analyze significantly enriched gene ontology terms based on database (refers to http://www.geneontology.org/). Pathway-based analyses were used to discover the functional target genes using KEGG database.
Knockdown(KD) and overexpression(OE) of miRNA-21-5p of MSCs
According to miRNA sequencing results, we found significant high expression of miRNA-21-5p of exosomes secreted from all MSC groups, therefore, Knockdown(KD) and overexpression(OE) of miRNA-21-5p of MSCs were conducted here. Briefly, the virus packaging system was a three-plasmid system: pspax2, pMD2G, and pCDH. The epithelioid cell line 293T was used as lentivirus packaging cells, and Escherichia coli strain DH5α was used as vectors of lentivirus and plasmid. The expression of miRNA-21-5p was verified by qRT-PCR and the Takara kit.
To observe interactions between MSCs-exo and HUVECs, laser scanning confocal microscopy was used here. For fluorescence staining of exosomes, MSCs-exo (8×109), miRNA-21-5p-KD-exo (8×109), and miRNA-21-5p-OE-exo (8×109) were respectively incubated with green fluorescent dye PKH67 (MINI67-1KT, SIGMA) for 5 min at room temperature, followed by twice PBS wash using ultracentrifugation, as described previously10. Then, human umbilical vein endothelial cells (HUVECs) were co-cultured with either PBS (control), MSCs-exo, miRNA-21-5p-KD-exo or miRNA-21-5p-OE-exo for 4 hours at 37°C with 5% CO2. Subsequently, the cytoskeleton (tubulin-α) of Human umbilical vein endothelial cells (HUVECs) was stained with fluorescent mouse anti-human monoclonal antibody (1:100, CY-SC006, Cytoskeleton) according to the standard protocol provided by reagent producer Cytoskeleton Inc. Cell nucleus was stained with Hoechst (0.1 ul added with 200 ul). The cells were washed twice with PBS between each step. Finally, cell samples were observed using laser scanning confocal microscopy (TCS SP8, Leica, Germany).
In vitro cellular proliferation assay
To evaluate how MSCs-exo affects cell proliferation of HUVECs, 1 x 106 HUVECs were treated with PBS (control), MSCs-exo (3 x 1010), miRNA-21-5p-KD-exo (3×1010), or miRNA-21-5p-OE-exo (3×1010) in serum-free medium for 24 hours. Then, cell proliferation of HUVECs was analyzed using the Ki67 assay kit and flow cytometry (FACsverse, BD corporation, America) by following standard protocol.
In vitro wound-healing migration assay
1 x 106 HUVECs were in 6-well plate with serum-free medium. A scratching wound was manually generated by a sterile 200-μl pipette tip. Pictures of the wound area were taken after 0, 8, 16, and 24 hours of MSCs-exo treatments. The migration was analyzed using the ImageJ software.
In vitro endothelial cell tube formation assay
96-well culture plates were pre-coated with 50 μL of growth-factor-reduced Matrigel (BD) (356234, Matrigel; Corning) at 37˚C for 30 minutes. Then, 4 x 104 HUVECs cells per well were co-cultured with either PBS (control), or MSCs-exo (5 x 109), miRNA-21-5p-KD-exo (5 x 109), or miRNA-21-5p-OE-exo (5 x 109) with the pre-treated plates at 37°C in 5% CO2 for 4 hours. Afterward, cells were visualized and tube formation was quantified. Tube number and branch points were statistically analyzed using ImageJ software.
Western Blot analysis
To test expressions of functional proteins, Western Blot was used here by following the standard protocols of our lab. HUVECs (5 × 106) were pre-treated with MSCs-exo (5 × 109) after 24 hours. VEGF, AKT, and MAPK signaling pathway molecules were then tested, including: VEGFR1 (1:1000 dilution, 2893S, Cell Signaling Technology, Danvers, MA, United), AKT (1:1000 dilution, 4691S, Cell Signaling Technology, Danvers, MA, United States), Phospho-AKT (1:1000 dilution, 13038S, Cell Signaling Technology, Danvers, MA, United States), ERK (1:2000 dilution, 4370S, Cell Signaling Technology, Danvers, MA, United States), and Phospho-ERK (1:1000 dilution, 4695S, Cell Signaling Technology, Danvers, MA, United States).
Chorioallantoic membrane assay
The pro-angiogenesis ability of MSCs-exo was examined using the chorioallantoic membrane (CAM) methodology. Fertilized chicken eggs (Gallus gallus) were rinsed and soaked in l:1000 benzalkonium bromide solution for three minutes. They were randomly divided into four groups and subsequently incubated at 37.8°C. After seven days, a small opening was made in the shell. The chorioallantoic membrane was treated with either PBS (control), MSCs-exo (5 x 109), miRNA-21-5p-KD-exo (5 x 109), or miRNA-21-5p-OE-exo (5×109), and the picture were taken. The openings in the eggs were covered with adhesive tape and the eggs were returned to the incubator. After nine days, pictures were taken again to evaluate the pro-angiogenesis ability of the MSCs-exo.
Hindlimb ischemia rat diabetic foot model
9-week male Sprague–Dawley rats (300 - 350 g; grade: SPF; license #: 1100111911009085) were purchased from Charles River Laboratory Animal Technology Co., Ltd. (Beijing, China; License: SCXK 2016-0011) to establish rat diabetic model. All animal experiments were conducted in accordance with the Institutional Animal Care and the Use Committee of Jinan University (License: SYXK 2017-0174). The Animal Experiment Protocol (Approval No.: IACUC-20190403-03) has been reviewed and approved by the Laboratory Animal Ethics Committee of Jinan University. Diabetes of rats was induced by intraperitoneally injected with 40 mg/kg (in 0.01 M sodium citrate, pH 4.3) streptozotocin. The blood glucose level of rats was controlled between 11.1 and 31.5 mmol/L. Subsequently, the femoral artery and all arterial branches of rat diabetic model were ligated to establish diabetic foot model. In addition, 5 mm full-thickness skin of the dorsal hind foot of the diabetic rats was wounded to simulate diabetic injury model.
In vivo treatment of diabetic foot rats using MSCs-exo
According to RNA sequencing, we found miRNA-21-5p altered the most significantly. To address whether miRNA-21-5p played crucial role on the function of pro-angiogenesis of exosomes in the inflammatory microenvironment, we performed knockdown (KD) and overexpression (OE) of miRNA-21-5p in MSCs. The surface markers of the exosomes in all the groups were identified using flow cytometry.
Then, diabetic foot rats were randomly divided into four groups: control group (PBS), MSCs-exo group, miRNA-21-5p-KD-exo group, and miRNA-21-5p-OE-exo group. Rats from each group received intramuscular injections of 1 ml PBS (control), MSCs-exo (5 × 1010, mixed with 1 ml PBS), miRNA-21-5p-KD-exo (5×1010, mixed with 1ml PBS), or miRNA-21-5p-OE-exo (5×1010, mixed with 1ml PBS) in the local area of ischemic hindlimb and foot ulceration. Afterward, we assessed the effect of MSCs-exo on proliferation signal molecules by measuring the expression of AKT and MAPK signaling pathway proteins by Western blot. The angiogenic VEGF signaling pathway proteins were measured by Western blot as well.
Additionally, the rat ischemic hindlimb severity degree of all groups including ischemic hindlimb complexion, skintemperature, and limb motor function were analyzed at D1, D7, and D14 after MSCs-exo treatments. The rat foot ulcer size of all rats was measured at D1, D3, and D7 after treatments, and then the ulcer area was analyzed using Image J.
Ultrasonic Doppler perfusion imaging and contrast ultrasonography
Ultrasonic Doppler imaging was performed to assess angiogenesis at D1 and D7 after treatment with MSCs-exo through the animal ultrasound micro imaging system (MYLAB 30CVVET, Esaote, Italy). Rats were anesthetized with 2% isoflurane, and ultrasound images of the right ischemic hindlimb were then immediately acquired using B mode Doppler perfusion imaging system. The blood perfusion ratio was analyzed by comparing the mean blood flow velocity and vessel number before and after treatments.
Digital subtraction angiography
Ischemic hindlimb angiogenesis was evaluated by Digital Subtraction Angiography (DSA) at D14 after treatment with MSCs-exo. The brief procedures include: isolation of femoral artery after incision of the rat’s thigh skin, following by ioversol (H20067896, Hengrui, Jiangshu, China) injection by puncture with a tiny needle; subsequently, hindlimb vessels of the rat were observed by the digital subtraction angiography machine (INNOVA-3100-IQ, General Electric Company, USA). The vessel numbers of the pictures taken by DSA were analyzed by Image J.
Histology and immunohistochemistry
At day 14, muscle and ulceration tissue from the ischemic limb was harvested, Then, the tissues were fixed in 10% formalin, paraffin-embedded and sliced. The sliced tissues (3 μm in slice thickness) were stained with hematoxylin and eosin (H&E). To conduct immunohistochemical examinations of granulation tissue angiogenesis, rabbit polyclonal anti-CD31 antibody (1:300 dilution, GB12063, Servicebio, China), anti-VEGF antibody (1:100 dilution, GB11034, Servicebio, China), and anti-TGFβ antibody (1:1000 dilution, GB11271-1, Servicebio, China) were used for sample staining by following standard protocols.
All acquired research data were expressed as mean ± standard deviation (SD). Significance between experimental groups and control group were calcuated using one-way ANOVA, and p < 0.05 was considered statistically significant.