Exosomal miRNA-155-5p from M1-polarized macrophages suppresses angiogenesis by targeting GDF6 to interrupt diabetic wound healing

Unprogrammed macrophage polarization, especially prolonged activation of proinflammatory macrophages, is associated with delayed wound healing in diabetic objectives. Macrophage-derived exosomes cargo a variety of microRNAs (miRNAs), participating in different stages in wound healing. Here, exosomes were isolated from naive bone marrow–derived macrophages (BMDMs) (M0-Exos), interferon-γ plus lipopolysaccharide-polarized BMDMs (M1-Exos), and interleukin-4-polarized BMDMs (M2-Exos). M1-Exos impaired migration and tube formation in human umbilical vein endothelial cells (HUVECs) compared to M0-Exos, whereas M2-Exos exhibited the opposite effects. High-throughput sequencing was performed to decipher the miRNA expression profiles in M0-Exos, M1-Exos, and M2-Exos. A total of 63 miRNAs were identified to be differentially expressed in exosomes derived from polarized BMDMs. Among them, miRNA-155-5p is highly expressed in M1-Exos, which interrupted angiogenesis in HUVECs. Furthermore, miRNA-155-5p directly binds to the 3′ UTR of growth differentiation factor 6 (GDF6) mRNA to suppress its protein expression. Lastly, local administration of a temperature-sensitive hydrogel Pluronic F-127 loading miRNA-155-5p antagomiR promoted angiogenesis and accelerated wound healing in diabetic db/db mice via enhancing GDF6. In summary, this study deciphered the miRNA expression profiles in exosomes from polarized macrophages. M2-like macrophage-derived exosomes and miRNA-155-5p inhibitors could be promising therapeutics against diabetic foot ulcers.


INTRODUCTION
Diabetes is a chronic metabolic disorder with a series of severe complications.Among them, impaired wound healing, particularly diabetic foot ulcers (DFUs), leads to significant morbidity, which ultimately causes nontraumatic limb amputation. 1 The risk of developing DFUs for diabetic patients across their lifetime has been estimated to be 30%. 2Unfortunately, no effective medical tools for DFUs are available in clinic.Wound healing is a complex process involving a series of coordinated and overlapping stages, including coagulation, inflammation, angiogenesis, and remodeling of the cell matrix.Delay or stagnation in any of these steps causes impaired wound healing. 3During wound healing, macrophages clear pathogens and regulate the proliferative phase of the repair process directly or indirectly through paracrine modes. 4In diabetic patients and mice, about 80% cells at the edge of chronic wounds are proinflammatory M1 macrophages, and the transition to an anti-inflammatory M2like phenotype does not occur in time. 5The M2-like macrophages regulate tissue remodeling and inflammation resolution, which underpin their critical role in maintaining microenvironmental homeostasis in diabetic wounds.Therefore, the impaired macrophage polarization is one of the main causes of healing stagnation in the inflammatory phase and poor wound healing in diabetic patients.
Exosomes are extracellular nanovesicles with a diameter of 30 to 100 nm, which mediate universal cell-to-cell communication by transferring specific components to recipient cells, including proteins, lipids, and nucleic acids. 6Emerging evidence suggests that exosomes play a key role in regulating matrix microenvironment homeostasis. 7n diabetic mice, M1 macrophages at the wound site directly transform into M2 macrophages after subcutaneous administration of M2 macrophage-derived exosomes, which accelerates wound healing by enhancing angiogenesis and epithelial regeneration. 8croRNAs (miRNAs) are a class of 17-to 25-nt small noncoding RNAs (ncRNAs) that regulate posttranscriptional gene silencing by binding to the 3 0 UTR or open reading frame region of target mRNAs. 9miRNAs stably exist in body fluids and can be packaged into exosomes or microvesicles to protect them from degradation.Given the transportability of vesicles, the information transmission through vesicles is a promising way of cell-to-cell communication. 10herefore, exosomal miRNAs from polarized macrophages may participate in the regulation of diabetic wound healing.
Here, we deciphered the miRNA expression profiles in exosomes derived from polarized bone marrow-derived macrophages (BMDMs), either proinflammatory M1-like or anti-inflammatory M2-like.Among the differentially expressed miRNAs, miRNA-155-5p, highly enriched in exosomes from M1-like macrophages, was identified as interrupting endothelial cell function and angiogenesis, and the miRNA-155-5p antagomiR was found to accelerate wound healing in diabetic db/db mice.

BMDMs-derived exosomes modulate angiogenesis in HUVECs
Macrophages orchestrated vessel sprouting and regeneration in wound angiogenesis.BMDMs can differentiate into mature macrophages (M0) in the presence of growth factors and other signaling molecules, which can be further induced to proinflammatory M1 macrophages by a variety of stimuli such as lipopolysaccharides (LPS), interferon-g (IFN-g), and pathogen-associated molecular patterns, or M2 anti-inflammatory macrophages under stimuli such as interleukin-4 (IL-4). 11To disclose their roles in mediating cellular interaction and manipulating angiogenesis, BMDM-derived exosomes were first isolated.To obtain BMDM-derived exosomes, bone marrow-derived stem cells (BMSCs) were isolated from the femurs and tibias of C57BL/6J mice, differentiated into naive BMDMs (M0-BMDMs), and then induced to M1-BMDMs and M2-BMDMs by LPS plus IFN-g and IL-4, respectively (Figure 1A).Under the microscope, the M1-BMDMs showed characteristic fried egg shapes, whereas the M2-BMDMs showed mixed populations of fried eggs and spindles (Figure 1B), which were consistent with a previous report. 12ext, exosomes were isolated from M0-BMDMs (M0-Exos), M1-BMDMs (M1-Exos) and M2-BMDMs (M2-Exos).Transmission electron microscopy (TEM) images showed that the exosomes derived from three types of BMDMs showed a standard bowl-shaped vesicle morphology, and particle size distribution results suggested these exosomes as having a diameter ranging from 80 to 150 nm (Figure 1C).The protein expression of the exosomal markers Alix and TSG101 further confirmed the successful isolation of exosomes (Figure 1D).Collectively, the above evidence indicated that exosomes were successfully isolated from BMDMs with characteristic morphology and similar size.
Next, human umbilical vein endothelial cells (HUVECs) were incubated in serum-free medium supplied with 50 mg/mL exosomes from different types of BMDMs (Figure 1E).The confocal images showed that BMDM-derived exosomes could be indiscriminately ingested by HUVECs within 7 h (Figure 1F).Interestingly, M0-Exos and M2-Exos significantly increased the proliferation of HUVECs, but not M1-Exos, assessed by Cell Counting Kit 8 (CCK-8) assay (Figure 1G).As anticipated, the migration potentials were significantly enhanced in HUVECs treated with M0-Exos or M2-Exos, but not those treated with M1-Exos, assessed by cell scratch assay (Figure 1H).Next, the tube formation assay was conducted.Quantitative analysis of key parameters in tube formation including the total length, total segments length, total branches length, number of branches, number of segments, and number of nodes showed that the treatment of M0-Exos or M2-Exos promoted angiogenesis, but not M1-Exos (Figure 1I).Taken together, BMDMs-derived exosomes can be taken up by HUVECs, and M2-BMDM-or M0-BMDM-derived exosomes promote the proliferation, migration, and tube formation of HUVECs.

miRNA expression profiles in BMDM-derived exosomes
Exosomes cargo a variety of components, including miRNAs, DNAs, proteins, and lipids.Among them, miRNAs play important roles in cellular communication through posttranscriptional patterns.To decipher the miRNA expression profiles in exosomes derived from polarized BMDMs, miRNA sequencing was performed.Small RNA (sRNA) distribution results showed that the known miRNAs in M0-Exos, M1-Exos, and M2-Exos accounted for 1.52%, 3.72%, and 3.47 of their total sRNAs, respectively (Figure S1A).The correlation difference between samples in the same group was negligible (Figure S1B).To identify the conserved miRNAs, all of the ncRNA reads from different types of BMDM-derived exosomes libraries were compared with the known mouse miRNAs in miRBase version 20.To focus on the highly represented miRNAs, sequences less than 10 reads were removed.Finally, 385, 338, and 384 miRNAs were identified in M0-Exos, M1-Exos, and M2-Exos, respectively.Venn analysis indicated that, among them, 330 miRNAs commonly exist in M0-Exos, M1-Exos, and M2-Exos; the numbers of miRNAs specifically existing in M0-Exos, M1-Exos, and M2-Exos were 81, 60, and 92, respectively (Figure 2A).The structural analysis of known miRNAs revealed that there are base changes in some sites of partial sequences, which may make the target genes diverse (Figure S1C).Principal-component analysis (PCA) showed that miRNAs were unambiguously segregated into three tight clusters (Figure 2B).
Next, we compared the miRNA expression levels in different types of BMDM-derived exosomes.Using a 2-fold change and p < 0.05 as the threshold cutoff, 63 miRNAs were identified as being differentially expressed in different types of BMDM-derived exosomes (Figure 2C).Among them, 18 miRNAs were upregulated and 20 miRNAs were downregulated in M1-Exos, compared with M0-Exos; 16 miRNAs were upregulated and 12 miRNAs were downregulated in M2-Exos, compared with M0-Exos.The stacking diagram of expression levels showed the distribution of the top 10 miRNAs in each group (Figure 2D).To better understand the potential biological processes associated with the differences in miRNA profiles in different types of BMDM-derived exosomes, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed using the online DAVID 6.8 bioinformatics resource.The GO database classifies and annotates genes and gene products in terms of their biological process, molecular function, and cellular component. 13mong biological process terms, the differently expressed genes were mainly enriched in the regulation of cellular component biogenesis, biological regulation, cell aggregation, and extracellular part (Figure 2E).KEGG contains seven major pathways: metabolism, genetic information processing, environmental information processing, cellular processes, organismal systems, human diseases, and drug development.KEGG annotation and enrichment analysis showed that the main functions of miRNA target genes were related to cell cycle, cell growth, and cell death (Figure 2F).The analysis of miRNA expression between M1-Exos and M2-Exos showed many differently expressed miRNAs (Figure 2G; Table S1).Among them, miRNA-155-5p showed the most significant difference between M1-Exos and M2-Exos (Figure 2G; Table S1).The qRT-PCR results further supported that miRNA-155-5p is highly expressed in M1-Exos compared to M0-Exos and M2-Exos (Figure 2H).Taken together, the present study deciphered the miRNA expression profiles in different types of BMDM-derived exosomes and identified that miRNA-155-5p is highly expressed in M1-Exos.
miRNA-155-5p exerts antiangiogenic effects via GDF6-Akt axis miRNAs exert their biological effects via blocking translation or inducing the degradation of target mRNAs by partial or complete pairing of seed sequence bases with mRNA recognition sites. 14To identify the target genes of miRNA-155-5p mediating the antiangiogenic effect in HUVECs, a virtual screen was performed using Targetscan and miRanda.As shown in Figure 4A, 11 hits were identified as potential miRNA-155-5p target genes, such as GDF6, Lat2, Golph3l, and Slc16a2.KEGG annotation (Figure 4B) and GO enrichment (Figure 4C) analysis of miRNA-155-5p target genes indicated that they were involved in tube morphogenesis and regulation of epithelial cell proliferation.Strikingly, the sequence of miRNA-155-5p in mice and humans are highly similar, and GDF6 may be one of the target genes involved in the antiangiogenic effect of miRNA-155-5p (Figure 4D).Previous studies reported that angiogenesis is closely related to the GDF6 gene; GDF6 protein can be recognized by the vascular endothelial growth factor receptor. 15,16Specifically, human miRNA-155-5p was found to target GDF6 mRNA 3 0 UTR at one binding site of base pair 1994 th to 2001 th .Subsequently, miRNA-155-5p overexpressed HUVECs were generated by the transfection of miRNA-155-5p agomiR.The abovementioned binding site of GDF6 mRNA was cloned into a luciferase reporter plasmid and transfected into HEK293T cells.miRNA-155-5p agomiR significantly reduced the luciferase activity of GDF6 reporter, compared to the NC agomiR (Figure 4E), which directly indicated that miRNA-155-5p could partially bind to the 3 0 UTR region of GDF6 mRNA.Furthermore, miRNA-155-5p overexpression led to the suppression of GDF6 protein expression and subsequently impaired Akt phosphorylation (Figure 4F).Thus, miRNA-155-5p suppresses angiogenesis in HUVECs via the GDF6-Akt axis.

DISCUSSION
The unprogrammed macrophage polarization is one of the main physiological features of poor and delayed wound healing. 17Diabetic wounds show dysregulated and persistent proinflammatory M1-like macrophages, whereas normal wounds display a transition to prohealing M2-like macrophages at around 3 days postwounding. 4everal studies have shown that blood glucose affects the wound microenvironment and macrophage polarization. 18The transition of M1 to M2 macrophages, a dynamic process depending on the surrounding microenvironment, is associated with impaired wound closure, poor angiogenesis, and decreased collagen deposition. 19It still remains elusive how macrophages orchestrate wound healing.After injury, a large number of monocytes and precursor macrophages from bone marrow are recruited around the wound, and these recruited cells exceed the resident macrophage population manifold. 20Therefore, macrophages derived from BMSCs possess incomparable advantages.Bone marrow monocytes released into the microenvironment express different chemokines and adhesion molecule receptors, and they are preferentially recruited to inflammatory lesions, where they can differentiate into macrophages. 21In the characterization of activated macrophages, the combination of IFN-g and tumor necrosis factor signals results in an activated or "inflammatory phenotype" of the macrophage population. 22It is widely believed that IL-4 rapidly converts resident macrophages into a population of cells programmed to promote wound healing.Meanwhile, IL-4 stimulates arginase activity in macrophages, and promotes collagen and extracellular matrix production in wounding. 23Thus, BMDMs from nondiabetic mice were recruited to produce exosomes, which were polarized to anti-inflammatory M2-like macrophages and proinflammatory M1-like macrophages via IL-4 and LPS + IFN-g, respectively.Many studies, either animal or human, have indicated that sustained high glucose exposure reduces macrophage phagocytic activity and renders BMDMs a proinflammatory phenotype. 24Plasma exosomal miRNAs involved in the adiponectin pathway were significantly dysregulated in diabetic patients. 25Thus, BMDMs from diabetic mice are more than like M1 macrophages, not M0 macrophages, which could not satisfy our current research purpose.Isolation of BMDMs from diabetic mice and evaluation of exosomal miRNA expression profiles from them could be interesting.It could help to identify miRNAs differently expressed in BMDMs from diabetic mice and develop them as therapeutic agents against diabetes.
Emerging evidence indicates that exosomes are involved in wound healing through multiple pathways, including angiogenesis and collagen synthesis.Exosomes derived from adipose stem cells accelerate diabetic wound healing by promoting angiogenesis and collagen synthesis. 26Fibroblast-derived exosomes induce keratinocyte migration and proliferation in vitro and accelerate wound healing in diabetic mice. 27In addition, macrophage-derived exosomes accelerate wound healing in diabetic rats by reducing the secretion of proinflammatory cytokines. 28Moreover, exosomes derived from other cells modulate diabetic wound healing by inducing or suppressing macrophage polarization in an indirect manner. 27,29In a recent study, M2 macrophage-derived exosomes promote M1 to M2 macrophage transform in the wound site, which accelerates wound healing by enhancing angiogenesis and epithelial regeneration. 8Our study showed that the exosomes from polarized BMDMs mediate cellular communication between macrophages and endothelial cells.Thus, exosomes regulate the functions of macrophages and endothelial cells in various stages of wound healing, which could be a promising therapeutic strategy against DFUs.
As critical posttranscriptional regulators, miRNAs participate in many physiological and pathological processes.One miRNA can simultaneously target multiple genes, forming a network amplification effect. 30miRNA is relatively stable and feasible for clinical application.Several studies have reported the critical roles of miRNAs in wound healing.The inhibition of miRNA-26a in diabetic wounds promotes the formation of granulation tissue and blood vessels. 31iRNA-27b promotes proliferation, migration, and tube formation of bone marrow-derived angiogenic cells to accelerate wound healing in diabetic mice. 32The inhibition of miRNA-29a increases collagen content and corrects the impaired biomechanical properties of diabetic skin. 33Our study deciphered the miRNA profiles in exosomes derived from polarized BMDMs.A series of differently expressed miRNAs were identified.Among them, miRNA-155-5p was found to be highly expressed in M1-Exos.The overexpression of miRNA-155-5p impairs the migration and angiogenesis of HUVECs, and the inhibition of miRNA-155-5p promotes wound healing in diabetic mice.Dysregulation of miRNA-155 expression was found to affect the development of diabetic nephropathy and retinopathy. 34Thus, miRNAs are considered to be attractive candidates for the treatment of DFUs.
In animals, miRNAs bind to specific sequences in the 3 0 UTR, causing degradation or translation arrest of the target mRNAs.GDF6 was predicted to be one of the targets of miRNA-155-5p.GDF6, a member of the superfamily transforming growth factor-b, induces vascular smooth muscle cell growth and remodeling to promote endothelial vascular integrity. 15,16GDF6 treatment forms new tendons and promotes tendon regeneration in bone marrow mesenchymal stem cells. 35High-temperature requirement factor A1 regulates angiogenesis by GDF6 in the retinas of mice. 36In addition, another predicted target of miRNA-155-5p was fibroblast growth factor 7 (FGF7).We found that the protein level of FGF7 was decreased in miRNA-155-5p overexpressed HUVECs (Figure S4).FGF7 participated in various processes such as angiogenesis and wound healing. 37A previous study suggested that miRNA-155 inhibition restores FGF7 expression in diabetic skin and decreases wound inflammation. 38Thus, miRNA-155-5p could target multiple mRNAs, including GDF6 and FGF7 to regulate angiogenesis and tissue remodeling.Systemic delivery, typically oral or intravenous, is the most common method of drug administration.Unfortunately, systemic drug delivery relies on the adequate perfusion of target tissue, and many chronic wounds lack this critical blood supply. 39Owing to the risk Molecular Therapy: Nucleic Acids of multiorgan toxicity and the unpredictability of drug delivery to target tissue, scientists are increasingly turning their research to the local delivery of drugs to promote wound healing.Local drug delivery can deliver the bulk drug substance to the target area, thereby maximizing the therapeutic potential and reducing drug toxicity in other organs. 40With the development of therapeutic nucleic acids, there are many ways to deliver nucleic acid drugs for wound healing.Exosomes carrying special nucleic acids have been collected for wound healing. 41Unfortunately, exosome-mediated nucleic acids delivery still need to overcome some challenges for clinical translations, such as insufficient active ingredients and low exosome activity. 42he use of protective substrates such as hydrogel represents one way to overcome these challenges. 39,43Accumulating evidence has indicated that Pluronic F-127, a thermosensitive hydrogel, could serve as a promising vehicle for miRNA encapsulation.The release of miRNA-148b mimics from Pluronic F-127 hydrogel enhances endothelial cell function, thereby increasing angiogenesis. 44The treatment of miRNA-129-2-3p mimics in Pluronic F-127 hydrogel accelerates wound healing. 45In the present study, miRNA-155-5p agomiRand antagomiR-loaded Pluronic F-127 hydrogels were prepared, which was a safe and effective method for the local delivery of miRNAs.
Compared with miRNA mimics and inhibitors, agomiR for miRNA overexpression and antagomiR for miRNA suppression are more suitable for in vivo interference experiments.They exhibit higher stability and inhibitory effect in animal studies.In addition, miRNA ago-miR and antagomiR are double-stranded, which can be directly loaded into RNA-induced silencing complexes and function immediately without Dicer enzyme processing.Instead of delivering the miRNA-155-5p-related products by conventional PBS injection, we creatively used temperature-sensitive hydrogel as carrier and dressing.miRNA was incorporated into thermosensitive hydrogels to form nucleic acid hybrid hydrogels, which showed the advantages of sustained release and practicability. 46The hydrogel loaded with miRNA-155-5p antagomiR was locally administrated, which promoted collagen accumulation and reepithelialization at the wound site, and significantly accelerated wound healing in diabetic mice.
In conclusion, our study deciphered the miRNA expression profiles in exosomes derived from polarized BMDMs, and identified that miRNA-155-5p, highly expressed in M1-Exos, suppresses endothelial cell functions and angiogenesis.miRNA-155-5p antagomiR could be a promising nucleic acid therapy against DFUs.

Isolation and characterization of exosomes
BMDMs were incubated for 2 days in serum-free DMEM.Subsequently, the cell culture supernatant was collected and centrifuged at 1,000 Â g for 30 min at 4 C to remove cell debris.Total exosome isolation reagent (Thermo Fisher Scientific, Carlsbad, CA) was added to the cell culture medium (1:2).After incubation overnight at 4 C, the mixture was centrifuged at 10,000 Â g for 1 h.After removing the supernatant, the precipitate was resuspended in PBS.Exosome size distribution was determined with dynamic light scattering (Zetasizer Nano, Malvern, UK).For TEM analysis, 10 mL exosome solution was dropped onto the copper net.The copper mesh was stained with 3% uranyl acetate for 2 min.After drying for 10 min, the copper mesh was observed using a JEM-1200EX TEM (JEOL, Tokyo, Japan).

Labeling and cellular uptake of exosomes
Exosomes were labeled with a membrane-labeling dye, DiI (Thermo Fisher Scientific), according to the manufacturer's protocol.Exosomes were incubated in a DiI staining solution (5 mg/mL) for 10 min at room temperature.After being washed with PBS, the exosomes were centrifuged at 10,000 Â g for 5 min.HUVECs were treated with the labeled exosomes for 3 h at 37 C. Cell nuclei were stained with DAPI (Sigma-Aldrich), and microfilaments were stained with Actin-Tracker Green-488 (Beyotime, Shanghai, China).The fluorescent images were captured by a confocal fluorescence microscope (Leica TCS SP8, Leica, Weztlar, Germany).

Cell culture and transfection with miRNA
HUVECs were obtained from the American Type Culture Collection (Rockville, MD).Cells were cultured in complete endothelial cell medium (ECM) supplemented with 5% FBS, 1% endothelial cell growth supplement, and 1% penicillin-streptomycin in a humidified atmosphere containing 5% CO 2 at 37 C.For transfection, HUVECs were seeded in 6-well plates and cultured overnight.A total of 100 nM miRNA agomiR or NC (IGE, Guangzhou, China) were transfected into HUVECs using Lipofectamine 3000 reagent (Invitrogen, Waltham, MA).Cells at passages between two and five were used in all of the experiments.

CCK-8 assay
CCK-8 was used to determine cell viability.HUVECs were seeded into 96-well plates at a density of 2 Â 10 3 cells per well.Cells were treated with exosomes for 6 h.Subsequently, 10 mL CCK-8 solution (Beyotime) was added to each well, followed by incubation for 1 h at 37 C. Absorbance at 450 nm was measured using a microplate reader (FlexStation 3 Multi Mode, San Jose, CA).

EdU assay
HUVECs were seeded in 6-well plates.EdU staining was conducted using BeyoClick EdU Cell Proliferation Kit (with Alexa Fluor 555, Beyotime) according to the manufacturer's instruction.Cell nuclei were stained with Hoechst (Beyotime).The images were captured by the Leica DMi8 microscope (Leica).

Cell scratch assay
HUVECs were seeded into 6-well plates at a density of 6 Â 10 5 cells per well.A single scratch was made using a sterile 200-mL pipette tip.The nonadherent cells were removed by rinsing with PBS three times.Then, the cells were incubated with serum-free ECM for 24 h.Images of the scratches were captured with IncuCyte real-time dynamic cell imaging analysis system (Essen Bioscience, Ann Arbor, MI) at 10Â magnification.The areas of the scratches were analyzed using ImageJ (NIH, Bethesda, MD).

Transwell migration assay
Transwell migration assay was performed using 24-well transwell plates and inserts with a filter of 8 mm pore (Corning, Glendale, AZ).HUVECs were seeded into the upper chamber of the insert at a concentration of 1.5 Â 10 5 cells in serum-free ECM, and complete ECM was added to the lower chamber.After incubation for 6 h, the cells on the top surface of the membrane were wiped off, and the cells on the lower surface were fixed with 4% paraformaldehyde.Then, the fixed cells were stained with DAPI.Fluorescent images were captured with the Leica DMi8 microscope (Leica) at 10Â magnification.The cell numbers were counted with ImageJ.

Tube formation assay
Matrigel (BD Biosciences, Glendale, AZ) was added to 96-well plates at 50 mL per well.HUVECs was seeded to each well at a concentration of 4 Â 10 4 in ECM (50 mL).HUVECs were treated with exosomes or miRNAs and then incubated for 5 h at 37 C.The tube formations were observed using IncuCyte real-time dynamic cell imaging analysis system at 4Â magnification.The results were analyzed by ImageJ.

qRT-PCR
Total RNAs were extracted using Trizol reagent (Invitrogen) according to a previous publication. 47The cDNA was synthesized from 1 mg RNA using the TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems, Waltham, MA) or the SuperScript III First-Strand Synthesis System (Invitrogen) following the manufacturer's instructions.The qPCR experiments were conducted on the Step-One plus real-time PCR System using SYBR Green PCR Master Mix (Invitrogen) with gene-specific primers (Table S2).The qRT-PCR assay was biologically repeated at least three times.b-Actin was applied as the internal standard.For miRNA expression, Ct values were normalized to U6 as the reference gene.The expression levels of genes were finally quantitated using the standard 2 ÀDDCt method.

miRNA library construction and sequencing
Total RNAs from exosomes were used for miRNA library preparation and sequencing by the Majorbio group (Shanghai, China).Briefly, total RNA samples were fractionated on a 15% Tris-borate-EDTA polyacrylamide gel (Invitrogen), and sRNAs ranging between 18 and 30 nt were used for library preparation.sRNAs were reverse transcribed and amplified by PCR.The PCR products were sequenced using the Illumina HiSeq 2500 platform.The RNA sequencing raw data were deposited in GEO: GSE234953.

Bioinformatics analysis
The Quantifier script in the miRDeep2 software was used to analyze the miRNA expression.Venn correlation and PCA analysis were performed according to the expression of miRNA among different samples.Three software programs, DESeq2, DEGseq, and edgeR, were used to analyze the differences in miRNA expression.The false discovery rate correction with Benjamini-Hochberg method was used to correct the statistical test results of miRNA expression differences.The predicted miRNA target genes were compared with six major databases (GO, KEGG, Clusters of Orthologous Genes [COG], NR, SwisS-PROT, and Pfam) to obtain comprehensive functional annotation information about the target genes.The target genes were sequentially aligned with the NR, SwisS-ProT, and COG databases using DIAMOND software.BLAST2GO was used for sequence alignment with the GO database.KEGG Orthology results were obtained using KOBAS2.1.

Dual-luciferase reporter assay
HEK293T cells were cotransfected with the GDF6 promoter luciferase reporter plasmid and the hsa-miRNA-155-5p agomiR (Genepharma, Shanghai, China) in the presence of an empty vector.The luciferase activities were assessed using the Dual-Luciferase Reporter Assay System (Promega, Madison, WI) according to the manufacturer's protocol.The results were standardized with the activity of Renilla luciferase.Molecular Therapy: Nucleic Acids Vol.34 December 2023 Wound healing in db/db mice All of the animal experiments were approved by the Animal Ethical and Welfare Committee of the University of Macau (no.UMARE-033-2017).All of the procedures involved in the animal experiments were carried out in accordance with the approved guidelines and regulations.Male Lepr em2Cd479 /Gpt diabetic mice (db/db) were purchased from Gempharmatech (Nanjing, China) and housed in the animal facility of the University of Macau, maintained at 23 C ± 1 C (50% ± 5% relative humidity) with 12-h light/dark cycles with free access to water and a regular chow diet.miRNA-155-5p agomiR, agomiR negative control (agomer NC), miRNA-155-5p antagomiR, and antagomiR NC were purchased from Genepharma.Pluronic F-127 (Sigma-Aldrich) hydrogel powder was mixed with diethyl pyrocarbonate water to obtain a 25% (w/v) solution.All of the procedures were carried out under aseptic and nonenzymatic conditions.db/db mice (10-12 weeks old, fasting blood glucose >11.1 mmol/L) were randomly divided into 5 groups of 6 mice each.The 5 groups of db/db mice were locally administrated with 20 mL 25% Pluronic F-127 solution, the same volume of 1.25 mM miRNA-155-5p agomiR in 25% Pluronic F-127 solution, 1.25 mM agomiR NC in 25% Pluronic F-127 solution, 1.25 mM miRNA-155-5p antagomiR in 25% Pluronic F-127 solution, and 1.25 mM antagomiR NC in 25% Pluronic F-127 solution, respectively, per wound.The administration was performed on days 0 and 3 of the wound closure experiment.

Assessment of wound area closure
Mice were anesthetized via inhalation of 3% isoflurane (RWD, Shenzhen, China).Before excision for wounds, the dorsal hair was shaved with an electric clipper followed by a depilatory cream.The skin was rinsed with alcohol, and two full-thickness wounds extending through the panniculus carnosus were created on the dorsum on each side of midline, using a 6-mm biopsy punch.Digital photographs were recorded on the day of surgery and every other day postinjury at fixed focal length and distance.Changes of wounds were quantified by assessing the wound area on days 0, 3, 5, 7, 9, and 11 using ImageJ software.The wound closure rates were calculated as the following formulation: (wound area on day 0 À wound area on day X)/wound area on day 0 Â 100%.On day 11, the mice were euthanized, and wound tissues were dissected.Part of the wound tissues was fixed in 4% paraformaldehyde at 4 C overnight and then embedded in paraffin.The remaining tissues were frozen in liquid nitrogen and stored in À80 C.

Histological analysis and microvessel density assay
For histological evaluation, sections (5 mm) were deparaffinized and rehydrated, followed by H&E and Masson's trichrome staining, according to standard protocols as described previously. 49Epidermal thickness was measured on images of H&E staining slides taken at 20Â magnification.Collagen quantification was determined on Masson trichrome-stained skin areas using ImageJ software.For immunohistochemical staining of CD31, the wound tissue sections were deparaffinized and stained with CD31 antibody (Abcam, ab182981, 1:800).Five randomly high-power field areas with the highest microvessel density were selected for each section.The average was calculated as the microvessel density.

Immunohistochemistry
Paraffin-embedded sections of skin tissues were stained using the avidin-biotin complex.Sections were incubated with polyclonal rabbit anti-GDF6 antibody at 4 C overnight, followed by treatment with horseradish peroxidase-labeled anti-rabbit immunoglobulin G.The streptavidin-biotin complex was used to visualize the staining.For each section, five high-power fields were observed.Scoring was conducted according to the immunoreactive score (IRS) standard. 46

Statistical analysis
Data were analyzed using GraphPad Prism version 8.3.0 (GraphPad Software, San Diego, CA).All of the experimental data were expressed as mean ± SD, and the sample size for each experiment corresponds to three biological replicates.Two groups were compared via the Student's t test, and more than two groups were compared using one-way ANOVAs, considering p < 0.05 as significant differences.Where statistical significance is evaluated, variance between groups is confirmed to be similar between comparison groups (control versus experimental) and the statistical analysis is deemed appropriate.

Figure 1 .
Figure 1.BMDMs-derived exosomes manipulated angiogenesis in HUVECs (A) The experimental procedure for polarized BMDMs.(B) Morphology of polarized macrophages.Scale bar, 100 mm (top) or 20 mm (bottom).(C) The ultrastructure of exosomes observed by TEM and the average particle size distribution of exosomes.Scale bar, 200 nm.(D) The protein expression of exosome markers, including Alix and TSG101.(E) Treatment of exosomes derived from polarized macrophages on HUVECs.(F) The ingestion of exosomes into HUVECs.Exosomes were stained with red fluorescent dye DiI, and actin in HUVECs was stained with actin-Tracker Green-488.Scale bar, 10 mm.(G) The cell viability of HUVECs treated with exosomes from polarized BMDMs, assessed by CCK-8 assay.(H) The migration capacity of HUVECs treated with exosomes from polarized BMDMs, assessed by cell scratch assay.Scale bar, 100 mm.(I) The tube-formation capacity of HUVECs treated with exosomes from polarized BMDMs.Scale bar, 150 mm.Data are means ± SDs. n = 6.*p < 0.05;**p < 0.01; ***p < 0.001; Exos versus NC.

Figure 2 .
Figure 2. Sequencing and bioinformatics analysis of exosomal miRNAs (A) miRNA variety in different groups was analyzed by Venn.(B) PCA of miRNAs from different groups of BMDMs.(C) Representative heatmap of differentially expressed miRNAs in macrophage-derived exosomes.(D) Top 10 highly expressed miRNAs in each group.(E) GO annotations and enrichment analysis of the target genes of miRNAs.(F) KEGG annotations and enrichment analysis of the target genes of miRNAs.(G) Representative volcano map of the differentially expressed miRNAs in M1-Exos and M2-Exos.(H) The expression level of miRNA-155-5p in exosomes from polarized BMDMs.Data are means ± SDs. n = 6.*p < 0.05; **p < 0.01; ***p < 0.001; M1-Exos or M2-Exos versus M0-Exos.