MicroRNA-122-5p is upregulated in diabetic foot ulcers and decelerates the transition from the inflammatory to the proliferative stage

Abstract

BACKGROUND

Shifting from the inflammatory to the proliferative phase represents a pivotal step during managing diabetic foot ulcers (DFUs); however, existing medical interventions remain insufficient. MicroRNAs (miRs) highlight notable capacity for accelerating the repair process of DFUs. Previous research has demonstrated which miR-122-5p regulates matrix metalloproteinases under diabetic conditions, thereby influencing extracellular matrix dynamics.

AIM

To investigate the impact of miR-122-5p on the transition from the inflammatory to the proliferative stage in DFU.

METHODS

Analysis for miR-122-5p expression in skin tissues from diabetic ulcer patients and mice was analyzed using quantitative real-time polymerase chain reaction (qRT-PCR). A diabetic wound healing model induced by streptozotocin was used, with mice receiving intradermal injections of adeno-associated virus -DJ encoding empty vector or miR-122. Skin tissues were retrieved at 3, 7, and 14 days after injury for gene expression analysis, histology, immunohistochemistry, and network studies. The study explored miR-122-5p’s role in macrophage-fibroblast interactions and its effect on transitioning from inflammation to proliferation in DFU healing.

RESULTS

High-throughput sequencing revealed miR-122-5p as crucial for DFU healing. qRT-PCR showed significant upregulation of miR-122-5p within diabetic skin among DFU individuals and mice. Western blot, along with immunohistochemical and enzyme-linked immunosorbent assay, demonstrating the upregulation of inflammatory mediators (hypoxia inducible factor-1α, matrix metalloproteinase 9, tumor necrosis factor-α) and reduced fibrosis markers (fibronectin 1, α-smooth muscle actin) by targeting vascular endothelial growth factor. Fluorescence in situ hybridization indicated its expression localized to epidermal keratinocytes and fibroblasts in diabetic mice. Immunofluorescence revealed enhanced increased presence of M1 macrophages and reduced M2 polarization, highlighting its role in inflammation. MiR-122-5p elevated inflammatory cytokine levels while suppressing fibrotic activity from fibroblasts exposed to macrophage-derived media, highlighting its pivotal role in regulating DFU healing.

CONCLUSION

MiR-122-5p impedes cutaneous healing of diabetic mice via enhancing inflammation and inhibiting fibrosis, offering insights into miR roles in human skin wound repair.

Key Words: MicroRNA-122-5p; Diabetic foot ulcer; Wound healing; Inflammation; Fibrosis

Core Tip: The shift from the inflammatory to the proliferative stage is essential for effective treatment of diabetic foot ulcers (DFUs). Through bioinformatics analysis and experimental studies, we evaluated the role of microRNA (miR)-122-5p in DFU. MiR-122-5p was found to delay wound repair in DFU by disrupting re-epithelialization and intensifying inflammation throughout the healing process.

INTRODUCTION

Diabetic foot ulcers (DFUs) are long-lasting wounds impacting around 15%[1] individuals managing diabetes, serving as a major cause of amputations and significantly elevating mortality rates[2,3]. Healing these ulcers is exceptionally challenging, with a high risk of recurrence, and impose substantial financial burdens on patients, significantly diminishing their overall well-being[4]. Managing diabetic foot lesions continues to pose significant challenges in clinical practice. Despite advances, existing treatments for refractory DFUs remain inadequate, underscoring the urgent need for reliable tools to diagnose and evaluate the prognosis of DFUs.

The process of wound repair progresses through four interconnected and sequential phases: Hemostasis, inflammation, proliferation, and tissue remodeling[5-7]. The intricate mechanism is tightly regulated by various cell types, encompassing cell migration, proliferation, extracellular matrix deposition, and tissue remodeling[8]. Fibroblasts, immune cells, endothelial cells, along with growth factors, inflammatory cytokines, and matrix metalloproteinases (MMPs) work in concert to ensure proper wound healing[9]. Macrophages, activated by signals from neutrophils, produce inflammatory mediators, including hypoxia inducible factor-1α (HIF-1α), that are crucial for the healing process[10]. In DFUs, an imbalance occurs, characterized by a dominance of M1 inflammatory macrophages and a deficiency of M2 anti-inflammatory macrophages, leading to prolonged inflammatory response. Such an imbalance disrupts fibroblast migration as well as proliferation, hinders the formation of new connective tissue, ultimately resulting in significant delays in wound recovery[11].

The proliferative stage of wound repair encompasses three critical mechanisms: Formation of new blood vessels, the shrinking of wound edges, and the regeneration of the epithelial layer[12]. At that stage, fibroblasts along with keratinocytes move to an injury bed and form an extracellular matrix that serves functions as a framework for cellular migration[13]. The formation of fresh vascular networks within the newly formed connective tissue, known as angiogenesis, is a vital component of this phase[14]. However, the angiogenic protein vascular endothelial growth factor (VEGF) shows markedly lower levels of expression in diabetes-related wounds, highlighting their reduced healing potential[15]. Connective tissue possessing immune capabilities, keratinocytes and fibroblasts start to migrate and proliferate immediately following an injury to close the wound gap[16]. Despite these insights, the molecular pathways affecting the progression between the inflammatory and proliferative phases in diabetic wounds continue to be poorly understood.

Recent breakthroughs in genetic science are greatly expanding the research of diabetes and the role of MicroRNAs (miRs)[17-19]. These small RNA molecules are crucial modulators of immune cell activities, with alterations of them influencing numerous physiological processes and contributing to various mammalian diseases[20-22]. Notably, miRs are accessible from serum or pathological specimens, making them accessible as part of research and clinical applications. For instance, miR-122-5p is predominantly expressed in hepatic tissue and has been consistently detected in patients with liver cancer[23]. MiR-122-5p, positioned at 18q21.31 in the human genome, is closely linked to metabolic processes, including insulin release, glucose metabolism, and diabetes susceptibility[24]. While substantial studies have explored miRs, their role in DFUs remains underexplored. Investigating miRs involved in DFUs is essential for uncovering potential therapeutic targets to enhance tissue repair. It has been found that miR-122-5p regulates MMPs in diabetes, impacting the extracellular matrix remodeling necessary for wound healing[25]. Moreover, circulating afamin, which is associated with complications of type 2 diabetes mellitus, shows a concentration positively correlated with the levels of miR-122, especially in long-term diabetic patients[26]. Given the critical roles of miRs in immune modulation and tissue repair, investigating miR-122-5p in the context of DFUs may provide novel insights into the molecular mechanisms of wound repair of diabetes and contribute to advancing targeted therapeutic strategies.

The research explored miR-122-5p, which was identified as upregulated in DFUs. Using a diabetic ulcer mouse model, we screened for miRs linked to wound healing and further analyzed the functions and underlying pathways.

MATERIALS AND METHODS

Human

The participants were recruited from Shuguang Hospital, Affiliated with The Shanghai University of Traditional Chinese Medicine (Shanghai, China). The group of healthy donors consisted of three individuals, each of whom underwent a 3 mm punch biopsy to create an excisional wound. The same number of patients were included in the DFU group. Lidocaine injections were used to administer local anesthesia. Signed informed consent was obtained from all participants during sample collection. This study was approved by the Ethics Committee of Shanghai University of Traditional Chinese Medicine (Approval No. 2024-1443-026-01) on March 11, 2024, and adhered to the principles of the Declaration of Helsinki. Table 1 showed DFU patients and healthy donors information.

Table 1 Information of diabetic foot ulcer patients and healthy donors.

Groups	Numbers	Gender	Age (year)	Duration of DFU (day)	Wagner grade
DFU patients	1	Male	45	15	2
	2	Male	52	20	3
	3	Female	55	60	3
Healthy donors	4	Male	43	NA	NA
	5	Male	46	NA	NA
	6	Female	50	NA	NA

DFU: Diabetic foot ulcer; NA: No appliable.

Animal model

A total of 63 C57BL/6 mice, all adult males weighing 21 g-25 g and aged 8-10 weeks, were procured from the Shanghai Model Organisms Center, Inc. The mice were kept under specific pathogen-free environments at the Experimental Animal Center of Shanghai University of Traditional Chinese Medicine. They were maintained on a 12-hour light/dark schedule, with food and water available at all times. Animals were randomly divided into groups of 20 mice each. One group received phosphate-buffered saline (PBS) treatment as a control, while another was a diabetic ulcer (DU) group, which was further subdivided based on AAVDJ-miR-122-5p upregulation or absence thereof. Furthermore, 3 mice received injections of AAVDJ-encoding empty vector to validate the successful modeling of miR-122-5p upregulation. At each time point (3, 7, and 14 days), five mice from each group will be sacrificed, and five mice will be kept as reserves in each group to address any unexpected issues. After a one-week acclimatization period, the DU groups were transitioned to a high-fat diet comprising 60% of calories (FB-D12451), provided by Wuxi Fan Bo Biotechnology Co., Ltd., which they consumed for one month. This was followed by daily intraperitoneal injections of streptozotocin a dose of 40 mg/kg (Cat. No. 2196GR001, BioFRoxx) for seven consecutive days[27]. Type 2 diabetes in mice was confirmed when their fasting blood sugar levels surpassed 11.1 mmol/L[28]. During the course of the experiment, blood glucose levels in DU mice were maintained between 11.1 and 30 mmol/L, with insulin administered as needed for levels exceeding this range.

Preparation AAVDJ vector

AAVDJ-miR-122-5p up and AAVDJ-encoding empty vector were constructed by GeneChem Co., Ltd. (Shanghai, China). The virus solution was delivered via intradermal injection along the wound margins employing a 1 mL syringe fitted with a needle with a 30-gauge size. Seven injections were administered to each wound, with 20 μL per injection, totaling around approximately 10¹¹ vector particles[29,30]. Following 3 weeks, 3 mice from each of the two virus-injected groups were sacrificed to validate the development of the AAVDJ-miR-122-5p up model using quantitative real-time polymerase chain reaction (qRT-PCR). Full-thickness skin excision surgery was performed to create wounds measuring 1 cm × 1 cm on the dorsal surface[31].

Intervention

A 1 cm² gauze saturated with PBS was used to cover the ulcers, with the dressings changed daily. On days 3, 7, and 14 after treatment, animals from each group were anesthetized using isoflurane (Cat. No. R510-22-10, RWD Life Science Co., LTD). We assessed whether the mice exhibited intolerable pain or significant declines in mobility, which were defined as humane endpoints. After ensuring complete loss of consciousness, they were euthanized by cervical dislocation. The skin samples were segmented across three parts, each 0.2 cm in width. This experiment was approved by the Animal Ethics Committee of Shanghai University of Traditional Chinese Medicine (Approval No. PZSHUTCM2212140011). This design and reporting adhere to the requirements of the ARRIVE guidelines.

Assessment of wound healing rate

The ulcer healing status on the backs was evaluated through capturing images on days 0, 3, 7, and 14 with a Nikon digital camera (Tokyo). The healing rate was analyzed with ImageJ software (Bethesda, MD, United States) and determined using the formula: t₀ - t/t₀ × 100%, where t represent wound healing was assessed, and t₀ was the initial wound area.

Bioinformatics

Fourteen days post-treatment, the ulcer samples were collected for analysis at OE Biotech (Shanghai, China). MiR expression was assessed using the Quantifier script from miRDeep2 software (version 0.1.2), and comparisons across samples were made using Venn diagrams and principal component analysis. The Benjamini-Hochberg method was applied for multiple testing corrections. The Kyoto encyclopedia of genes and genomes (KEGG) database was used to interpret the predicted target genes of miRs to gain functional knowledge[32]. Additionally, gene set enrichment analysis software (version 4.3.2) was utilized to identify cellular functions potentially modulated by miRs.

Data filtering and processing

Metabolomic data were analyzed using Progenesis QI (version 3.0), which involved baseline correction, peak detection, integration, retention time adjustment, alignment, and normalization. The “limma” R package (version 3.52.4) was used to preprocess and standardize gene expression data obtained from the gene expression omnibus and KEGG repositories.

Cell system

The American Type Culture Collection provided the RAW264.7 and NIH3T3 cell lines. Both cell lines were cultured in Dulbecco’s modified eagle medium containing 10% fetal bovine serum (FBS) (preheated to inactivate the serum), 100 U/mL penicillin, and 100 μg/mL streptomycin, and maintained at 37 °C with 5% carbon dioxide in a standard incubator. M1 macrophage polarization was triggered with 100 ng/mL lipopolysaccharide (LPS) (Cat. No. L2630, Sigma Aldrich, United States). Supernatants derived from RAW264.7 cell cultures were subsequently used to culture NIH3T3 cells.

Cellular transfection

In basal medium, the cell density was adjusted to 3 × 10⁴ cells/mL, and the cells were then seeded into six-well plates. With a final dosage of 50 nM, NIH3T3 cells were introduced into a miR-122-5p mimic or inhibitor, accompanied by the corresponding mock samples (Ribobio Co., Guangzhou, Guangdong Province, China). Employing lipofectamine 3000 (Invitrogen, United States), genetic modifications were conducted following the manufacturer’s protocols. 72 hours post-transfection, cells medium protein and supernatant samples were extracted.

Scratch wound healing assay

With a cell count of 5 × 10⁵ cells/mL, cells were plated into six-well plates and placed in medium supplemented with 10% FBS for overnight incubation. Employing a 10 μL sterile tip, a gap was induced, and the treated cells were cultured thereafter under normoxic conditions for 0, 24, and 36 hours. At every time interval, images were recorded with the OLYMPUS BX53 microscope. Cell motility formula: (Gap width at 0 hour gap width at the specific time)/gap width at 0 hour × 100%.

Histopathological assessment

Fixation of lesion samples was preserved using 4% formalin (Cat. No. P0099, Beyotime) over a 48-hour period. Following standard paraffin tissue processing as well as sectioning, ulcer tissues were subjected to hematoxylin and eosin (HE), (Cat. No. C01105M, Beyotime, China) and Masson’s trichrome stain (Cat. No. G1340, Solarbio Life Sciences, China). The histological analysis was performed by a pathologist at magnification levels ranging from × 20 to × 100. A whole-slide scanning system (Precipoint M8) was used to enable detailed examination of the stained sections.

Immunohistochemistry

Immunohistochemistry (IHC) analysis was conducted on tissue slices embedded in paraffin, which were exposed to specific antibodies at 4 °C for 12 hours: Tumor necrosis factor (TNF)-α (1:300, Cat. No. YT4689, Immunoway), MMP9 (1:500, Cat. No. 10375-2-AP, Proteintech), HIF-1α (1:500, Cat. No. 20960-1-AP, Proteintech), fibronectin 1 (FN1) (1:300, Cat. No. 15613-1-AP, Proteintech), α-smooth muscle actin (α-SMA) (1:200, Cat. No. 55135-1-AP, Proteintech), and VEGF (1:500, Cat. No. 19003-1-AP, Proteintech). For IHC staining, the tissue sections were treated with a secondary immunoglobulin (1:200, Cat. No. GB23303, Servicebio). Staining visualization was achieved with 3,3’-Diaminobenzidine substrate, and hematoxylin was applied for counterstaining. The method of image scanning was the same as pathological analysis.

Immunofluorescence

For IHC, 6 μm tissue sections were prepared and treated with specific antibodies targeting F4/80 (1:800, CST, 30325), ARG1 (1:1000, CST, 93668S), and inducible nitric oxide synthase (1:800, CST, 13120S). Secondary antibodies labeled with Alexa Fluor^® 488 (1:2000, anti-rabbit IgG, CST, 4412) and Alexa Fluor^® 594 (1:2000, goat anti-rabbit IgG, CST, 8889) were utilized in visualization. 4’,6-diamidino-2-phenylindole (DAPI) was applied for nuclear labeling. Images were captured with a Leica TCS SP8 STED 3X confocal microscope with ultra-high resolution (Ernst-Leitz, Wetzlar, Germany).

Western blot

Samples were lysed with radio immunoprecipitation assay buffer (Cat. No. P0013C, Beyotime, China), supplemented with phosphatase and protease inhibitor cocktails (Cat. Nos. P1005 and P1045, Beyotime, China). Proteins (20 μg per sample) were separated on 7.5% sodium dodecyl sulfate polyacrylamide gel electrophoresis gels (Cat. Nos. PG111 and PG112, EpiZyme, China) alongside protein size markers (Cat. Nos. 26619 and 26625, Thermo Fisher, United States). The proteins were conveyed to polyvinylidene difluoride membranes (Cat. No. IPVH00010, Millipore) and incubated with 5% bovine serum albumin (BSA) for 2 hours to block. The blots were incubated with the target antibodies at 4 °C for 12 hours, including rabbit anti-TNF-α (1:2000, Cat. No. YT4689, Immunoway), anti-MMP9 (1:1500, Cat. No. 10375-2-AP, Proteintech), anti-HIF-1α (1:1000, Cat. No. 20960-1-AP, Proteintech), anti-FN1 (1:2000, Cat. No. 15613-1-AP, Proteintech), anti-α-SMA (1:2000, Cat. No. 55135-1-AP, Proteintech), anti-VEGF (1:500, Cat. No. 19003-1-AP, Proteintech), and anti-glyceraldehyde-3-phosphate dehydrogenase (1:5000, Cat. AF1186, Beyotime).

qRT-PCR

Total RNA was extracted from skin tissue by the RNeasy mini kit (Qiagen GmbH). Bulge-Loop™ miR qPCR primer sets for miR-122-5p (MQPSO000123-1-100) and U6 small nuclear RNA (snRNA) (MQPSO000002-1-100) were obtained from Guangzhou RiboBio Co., Ltd. Reverse transcription was performed using the PrimeScript™ RT reagent kit (Catalog No. RR047A; Takara Bio). qRT-PCR was conducted with the iQ™ SYBR^® green super mix (Catalog No. 1708882AP; Bio-Rad Laboratories). The thermal cycling program consisted of an initial step at 95 °C for 3 minutes, followed by 40 cycles of 95 °C for 20 seconds, 60 °C for 30 seconds, and 70 °C for 30 seconds. U6 snRNA served as the internal control for miR-122-5p quantification.

Total RNA was isolated using the cell/tissue total RNA isolation kit V2. The RNA levels were measured with a NanoDrop-1000 spectrophotometer. III RT SuperMix was utilized to reverse transcription and further elimination of residual gDNA. With the SYBR Green method, qRT-PCR was conducted on an Applied Biosystems 7500 system (Foster City, CA, United States). RNA expression levels were determined with the 2^-∆∆Ct method. The detailed primer sequences were listed in the Supplementary Table 1.

Enzyme-linked immunosorbent assay

Wound tissues from mice were homogenized into a slurry. PBS was incorporated at a ratio of 1:9 (weight to volume), and the mixture was processed at 2500-3000 rpm while kept in an ice bath. The resulting mixture was then spun at high speed for 10 minutes, with the resulting supernatant employed to quantify TNF-α, MMP9, HIF-1α, FN1, α-SMA, and VEGF levels. Similarly, enzyme-linked immunosorbent assay (ELISA) kits (Multi Sciences, Hangzhou, Zhejiang Province, China) were employed to measure these markers in the supernatant of NIH3T3 cell cultures. The assay was evaluated with an ELX800 ELISA reader (BioTek Instruments, United States).

Fluorescence in situ hybridization

Tissue paraffin slices were initially incubated using a pre-hybridization solution and then hybridized overnight with the miR-122-5p probe (sequence: 5’-UGGAGUGUGACAAUGGUGUUUG-3’, 500 nM) in G3045 solution. A warmed branch probe solution was introduced, and the samples were subsequently rinsed with heated saline-sodium citrate buffer. Next, a signal probe was introduced, and blocking was performed on the sections using 3% BSA. For immunostaining, the slices were incubated using a principal antibody against clip3 (rabbit, 1:100), rinsed in PBS, and treated with a fluorescein-labeled secondary antibody (488-labeled goat anti-rabbit). Lastly, DAPI was used for nuclear counterstaining, and the samples were imaged.

Statistical analysis

All data were collected from three independent experiments and processed by GraphPad Prism (GraphPad, United States). One-way and two-way analysis of variance were applied for multi-group differences, while pairwise differences among groups were performed using the student’s t-test. A P value of < 0.05 was deemed statistically significant.

RESULTS

The miRs in DFU subjects and DU models

In DU models, miRs serve a critical role in mediating cellular interactions via post-transcriptional regulation. Regarding DU, we performed miR sequencing and bioinformatics analysis. Comparing miR levels in normal vs DU mouse revealed numerous differentially regulated miRs, with miR-122-5p exhibiting the highest variation (Figure 1A). A total of 22 differentially expressed miRs were identified with a twofold change and P < 0.05 as the threshold. Among these, the five most upregulated miRs were miR-31-5p, miR-154-5p, miR-122-5p, miR-206-3p, and miR-375-3p, while the five most downregulated included miR-142a-3p, miR-129-5p, miR-203b-3p, miR-335-3p, and miR-708-3p (Figure 1B). A grouped bar plot displayed the expression profiles of the top 10 miRs across groups (Figure 1C). To explore the biological processes associated with miR differences, we performed gene ontology (GO) and KEGG pathway evaluations by the Database for Annotation, Visualization, and Integrated Discovery 6.8 bioinformatics platform. The GO database categorizes and annotates genes and their associated proteins based on biological activities, cellular structures, and molecular roles[33,34]. The genes with differential expression were mainly engaged in the formation of cellular components, biological control, cell clustering, and the extracellular environment (Figure 1D). The KEGG database organized biological processes into seven categories, including cellular functions, human diseases, and drug development. KEGG pathway analysis and enrichment assessment highlighted the involvement of target genes with pathways associated with cell division, cellular proliferation, and apoptosis, offering insights into gene interactions (Figure 1E). We analyzed the levels of miR-122-5p in DFU patients and DU mice, and found that miR-122-5p was markedly upregulated in both (Figure 1F).

Figure 1 Sequencing and bioinformatics analysis of MicroRNAs. A: In diabetic ulcer mice wounds, wound tissues were harvested 14 days postinjury, and total RNA was extracted. Subsequently, MicroRNAs (miR)-seq was performed, followed by the generation of a volcano plot; B: Representative heatmap of differentially expressed miRs in the skin tissues of diabetic ulcer mice; C: The top 10 miRs with the highest expression levels in each group; D: Gene ontology annotations and enrichment analysis of the target genes of miRs; E: Kyoto encyclopedia of genes and genomes annotations and enrichment analysis of the target genes of miRs; F: Relative expression levels of miR-122-5p in wound tissue from streptozotocin-induced diabetic mice [2^-∆∆Ct = 2.30, 95% confidence interval (CI): 1.81-2.72] showed a significant increase compared to normal mice (2^-∆∆Ct = 1.12, 95%CI: 0.64-1.31, P < 0.001) by quantitative real-time polymerase chain reaction; Further experiments demonstrated that the miR-122-5p overexpression group (2^-∆∆Ct = 3.82, 95%CI: 3.45-4.37, P < 0.001) exhibited significantly higher expression compared to the diabetic ulcer group; miR-122-5p levels were significantly increased in patients with diabetic foot ulcer (2^-∆∆Ct = 1.86, 95%CI: 2.14-3.51) compared with healthy individuals (2^-∆∆Ct = 1, 95%CI: 0.96-1.23, P < 0.001). ^cP < 0.001. miR: MicroRNA; FC: Fold change; DU: Diabetic ulcer; GO: Gene ontology; KEGG: Kyoto encyclopedia of genes and genomes; DFU: Diabetic foot ulcer; AAVDJ: Adeno-associated virus-DJ.

AAVDJ-miR-122-5p up impaired ulcer repair in DU mice

To evaluate the impact of miR-122-5p, we monitored wound healing progression and collected skin samples on days 3, 7, and 14 post-treatment (Figure 2A). All groups showed a gradual reduction in wound size; however, by day 7, wound contraction was evident in normal and DU mice, while the AAVDJ-miR-122-5p mice exhibited minimal alterations, highlighting its inhibitory effect on wound recovery. During the proliferative stage, wounds with miR-122-5p overexpression developed a hard film on the surface, but complete healing was not achieved (Figure 2B and C). The AAVDJ-miR-122-5p group exhibited the smallest repaired areas, measuring 2.87 cm² ± 0.02 cm² on day 7 and 1.20 cm² ± 0.12 cm² on days 14, which were significantly lower than those in the other two groups. In comparison, the DU mice achieved 2.01 cm² ± 0.17 cm² and 1.02 cm² ± 0.14 cm² on the same days (Figure 2D and E). HE results showed that the ulcers in the AAVDJ-miR-122-5p mice had a thinner epidermis and dermis and exhibited more inflammatory cell infiltration (e.g., polymorphonuclear leukocytes and plasma cells) compared to the other groups post-treatment (Figure 2F). Additionally, Masson staining showed poorer granulation tissue formation, reduced collagen accumulation, and less organized arrangement (Figure 2G), suggesting which miR-122-5p impaired DU by hindering reepithelialization.

Figure 2 Adeno-associated virus-DJ-microRNA-122-5p up-decelerated wound healing in diabetic mice. A: Schematic diagram of the timeline of mice tests on the therapeutic effect of wound; B-D: A full-thickness skin wound was created on the dorsal area of the mice (1 cm × 1 cm). Animals were randomized into three groups and treated with phosphate-buffered saline. Optical pictures and related quantification of the wound closure rate in the control, diabetic ulcer (DU), and adeno-associated virus (AAV)-DJ-microRNA (miR)-122-5p groups at days 0, 3, 7, and 14 after the skin operation (n = 5); E: Gross view of wounds and wound area among the three groups of mice; F: Hematoxylin-eosin staining images of wound tissues in the control, DU, and AAVDJ-miR-122-5p groups at day 14 (n = 5; scale bar = 500 μm for 10 × and 100 μm for 40 ×); G: Masson trichrome staining at day 14 post operation (n = 5, scale bar = 500 μm for 10 × and 100 μm for 40 ×). ^aP < 0.05. ^bP < 0.01. DU: Diabetic ulcer; miR: MicroRNA; AAVDJ: Adeno-associated virus-DJ; STZ: Streptozotocin; HE: Hematoxylin-eosin.

MiR-122-5p enhanced the expression of inflammatory factors

We evaluated the levels of inflammatory factors in the early inflammatory stage, commonly occurring within the first 3 days. IHC revealed elevated levels of MMP9, TNF-α, and HIF-1α in the AAVDJ-miR-122-5p mice, while the other groups demonstrated notable suppression of these markers (Figure 3A and B). Proinflammatory markers were further assessed by qRT-PCR, confirming the expected results (Figure 3C). Western blot (WB) analysis confirmed the high expression of inflammatory proteins in DU, with further activation observed in the presence of miR-122-5p overexpression (Figure 3D and E). All findings were in line with the cell experiment results (Figure 3F and G).

Figure 3 MicroRNA-122-5p upregulated matrix metalloproteinase 9 to target inflammatory factors during the transition from the inflammatory phase to the proliferative phase. A and B: Levels of matrix metalloproteinase (MMP) 9, tumor necrosis factor (TNF)-α, and hypoxia inducible factor (HIF)-1α in wound tissues of mice after 14 days were detected using immunohistochemistry (20 ×); C: The expression of MMP9, TNF-α, and HIF-1α in wound tissues of mice was assessed using quantitative real-time polymerase chain reaction; D and E: Expressions of MMP9, TNF-α, and HIF-1αproteins were tested in wound tissues (n = 3); F and G: Expressions of MMP9, TNF-α, and HIF-1α proteins were tested in NIH3T3 cells in different groups (n = 3). ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. DU: Diabetic ulcer; miR: MicroRNA; AAVDJ: Adeno-associated virus-DJ; TNF-α: Tumor necrosis factor-α; HIF-1α: Hypoxia inducible factor-1α; MMP9: Matrix metalloproteinase 9; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; NC: Normal control.

MiR-122-5p inhibited vascular endothelial growth and slowed fibrosis progression

We also evaluated the levels of VEGF and fibrosis in the proliferative stage, commonly occurring within 14 days. IHC indicated that VEGF, FN1, and a-SMA were negative in the AAVDJ-miR-122-5p mice, while the normal mice exhibited clear activation of these markers (Figure 4A and B). qRT-PCR analysis of VEGF and fibrosis indicators (FN1 and α-SMA) confirmed results consistent with expectations (Figure 4C). WB showed reduced levels of VEGF, and fibrosis indicators in DU, with additional inhibition of these proteins observed under miR-122-5p overexpression (Figure 4D and E), corroborating the cell experiment findings (Figure 4F and G).

Figure 4 MicroRNA-122-5p reduced vascular endothelial growth and delayed fibrosis during the transition from the inflammatory phase to the proliferative phase. A and B: Levels of vascular endothelial growth factor (VEGF), fibronectin (FN) 1, and α-smooth muscle actin (α-SMA) in wound tissues of mice after 14 days were detected using immunohistochemistry (20 ×); C: The expression of VEGF, FN1, and α-SMA in wound tissues of mice was assessed using quantitative real-time polymerase chain reaction; D and E: Expressions of VEGF, FN1, and α-SMA proteins were tested in wound tissues (n = 3); F and G: Expressions of VEGF, FN1, and α-SMA proteins were tested in NIH3T3 cells in different groups (n = 3). ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. DU: Diabetic ulcer; miR: MicroRNA; AAVDJ: Adeno-associated virus-DJ; SMA: Smooth muscle actin; VEGF: Vascular endothelial growth factor; FN: Fibronectin; NC: Normal control.

MiR-122-5p up hindered the progression from the inflammatory stage to the proliferative phase

Fluorescence in situ hybridization analysis revealed elevated miR-122-5p expression in diabetic mice compared to normal controls, with signals predominantly localized to epidermal keratinocytes and fibroblasts (Figure 5A). Moreover, miR-122-5p promoted M1 macrophage accumulation and inhibited M2 macrophage differentiation in diabetic wounds (Figure 5B). A wound healing assay revealed that miR-122-5p reduced the motility of NIH3T3 cells (Figure 5C). At 36 hours, the wound closure rate was about 44.0% in the miR-122-5p inhibitor group, compared to 10.33% in the miR-122-5p mimic group (Figure 5D). ELISA results demonstrated that miR-122-5p led to elevated levels of inflammatory cytokines in DFU and NIH3T3 cells, while VEGF and fibrosis-related factor expression were reduced (Figure 5E and F).

Figure 5 Adeno-associated virus-DJ-microRNA-122-5p up-decelerated the transition from the inflammatory to the proliferative stage. A: In situ hybridization was performed using a microRNA (miR)-122-5p-specific probe. Green indicates miR-122-5p expression, and blue indicates 4’,6-diamidino-2-phenylindole (20 ×); B: Tissue immunofuorescence staining of F4/80, ARG1, and inducible nitric oxide synthase (20 ×); C and D: The migration capacity and quantitative analysis of NIH3T3 cultured with RAW264.7 media in different groups were assessed using a cell scratch assay (n = 3); E and F: Matrix metalloproteinase 9, vascular endothelial growth factor, inflammation factors, and fibrosis factors were detected in vivo by enzyme-linked immunosorbent assay assays (n = 3). ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. miR: MicroRNA; DU: Diabetic ulcer; AAVDJ: Adeno-associated virus-DJ; DAPI: 4’,6-diamidino-2-phenylindole; iNOS: Inducible nitric oxide synthase; TNF-α: Tumor necrosis factor-α; HIF-1α: Hypoxia inducible factor-1α; MMP9: Matrix metalloproteinase 9; SMA: Smooth muscle actin; VEGF: Vascular endothelial growth factor; FN: Fibronectin; NC: Normal control.

DISCUSSION

In our investigation, an imbalance between increased M1 macrophages and reduced M2 anti-inflammatory macrophages significantly contributes to hinder ulcer repair. The shift of macrophages, which depend on the local microenvironment, was closely linked to aggravated inflammation, impaired angiogenesis, and reduced collagen deposition[35,36]. This further limited fibroblast proliferation and motility, contributing significantly to the delayed generation of granulation tissue[37]. The exact molecular mechanisms responsible for the disrupted transition between these phases remain unclear. MiRs are crucial in modulating the shift from chronic inflammation to epithelialization in wound healing. Recent findings on miR-185, miR-195, and miR-205 blockers or promoters underscore their regulatory effects on DFU[38-40]. Despite previous evidence highlighting the therapeutic potential of miRs in DFU, only a limited have progressed to clinical applications[41,42]. Additional studies are required to explore and enhance the application of miRs for better healing outcomes. Our investigation identified a strong association between miR-122-5p and DFU, revealing significantly elevated miR-122-5p levels in DU mice.

MiR-122-5p was upregulated in DFU, aligning with research findings that elevated miR-122 levels significantly contribute to the development of metabolic syndrome. The hallmark features of metabolic syndrome are glucose metabolism disorders[43]. MiR-122 has been studied as a biomarker for hepatocyte injury, with elevated levels detected even prior to the onset of liver function decline[44]. Regmi et al[45] reported abnormal miR-122 expression in diabetic nephropathy, which was significantly associated with renal function, blood glucose, and other clinical parameters. Additionally, Wang and Yu[46] demonstrated that miR-122-5p is elevated in coronary-related disorders, suggesting its utility as a marker for assessing coronary atherosclerosis[46]. These findings collectively highlight miR-122-5p’s involvement in metabolism-related diseases, emphasizing its broader relevance as a biomarker and regulatory factor.

MiRs act as key posttranscriptional regulators, playing essential roles in various disease. Due to their stability, miRs are considered promising candidates for therapeutic implementation. As an illustration, inhibiting miR-26a in DU promotes granulation tissue and blood vessel formation[47], while inhibiting miR-29a increases collagen and improves the impaired characteristics of DU[48]. The involvement of miR-122-5p in DU was also explored in our study.

Target genes of miR-122-5p are enriched within pathways and biological processes identified through GO and KEGG analyses, which are closely related to diabetic wound healing. MiR-122 regulated the cell cycle by targeting key regulatory factors such as cyclin G1, and its anti-cancer properties are closely related to this function[49]. This offered a compelling explanation regarding cell proliferation in DFU. Its enrichment in the cell cycle regulatory pathway may accelerate wound repair by modulating the proliferation of fibroblasts. As well, miR-122-5p helps alleviate the chronic inflammatory environment in diabetic wounds by targeting and regulating genes in the nuclear factor kappa-B (NF-κB) signaling pathway, thereby improving the conditions for healing[50]. Furthermore, its regulation of angiogenesis-related mechanisms, like the VEGF signaling network, could enhance neovascularization and increase vascular perfusion to wound tissues[51]. GO analysis indicated that miR-122-5p target genes notably participate in biological processes related to extracellular matrix remodeling, which may support collagen production and improve tissue organization at the wound site[52]. These mechanisms collectively suggested that miR-122-5p assumes a multifaceted and pivotal function in the regulation of DFU.

To investigate the involvement of miR-122-5p in DFU, we focused on macrophages and fibroblasts, key cells during the inflammatory and proliferative phases. MiR-122-5p expression, along with inflammatory and fibrosis markers, was analyzed in NIH3T3 cultured with LPS-stimulated RAW264.7 cell supernatants. The findings revealed elevated levels of TNF-α, MMP9, and HIF-1α and reduced VEGF, FN1 and α-SMA, suggesting a link between miR-122-5p and processes connected to inflammation and epithelialization. Drawing from in vitro predictions of miR-122-5p mimic, we substantiated the involvement of miR-122-5p in slowing diabetic tissue repair through delayed recovery following AAVDJ-miR-122-5p in DU.

Studies have indicated that the miR-122-5p-mediated p38 mitogen-activated protein kinase signaling pathway regulates the levels of pro-inflammatory cytokines in HIBECs, which aligns with the inflammatory expression observed during the inflammatory phase of DFU in this study[53]. Additionally, delivering miR-122-5p to alveolar macrophages promotes M1 polarization through the suppressor of cytokine signaling-1/NF-κB pathway, contributing to the development of pulmonary damage[54]. Reducing miR-122-5p levels in neutrophil-derived exosomes can decrease the immunomodulatory function of macrophages in Behçet’ s disease by targeting IRF5 expression[55]. This evidence support that miR-122-5p modulates immune function and inflammatory cytokine levels through a complex signaling network; however, additional studies are required to identify the upstream signals and specific targets of miR-122-5p. LINC01296 functions by sequestering miR-122, regulating MMP-9 expression, suggesting that miR-122 could be instrumental in the development of gastric carcinoma[56]. Furthermore, miR-122-5p inhibition has been shown to reduce TNF-α levels. Studies have found elevated TNF-α levels in patients with type 2 diabetes, are linked to the upregulation of miR-122-5p[57]. Previous research has highlighted the interaction between miR-122 and HIF-1α, demonstrating the miR-122 acting as a regulator in oxygen-deprivation mechanisms[58]. Both HIF-1α and miR-122 have been identified as potential contributors to cancer[59]. However, the connection between HIF-1α-mediated regulation of miR-122 and its role in DFU remains unexplored.

MiR-122 has demonstrated potential as a biomarker for liver fibrosis associated with chronic liver disease caused by hepatitis C virus[60]. Miravirsen, the first anti-miR therapy to reach clinical evaluation, is designed to target miR-122 for the treatment of hepatitis C virus. In its phase II clinical study (NCT01200420), the therapy showed significant effectiveness and a favorable safety profile in addressing hepatitis C virus genotype 1[61]. It is important to emphasize that this differs from the fibrosis observed in DFU in vitro. The research revealed that elevated miR-122-5p expression suppresses fibrosis markers like FN1 and α-SMA, both of which are associated with delayed wound repair. While various cytokines influence angiogenesis, VEGF serves as a key regulator[62]. MiR-122-5p functions as a strong pro-angiogenic factor by stimulating VEGF signaling, thereby enhancing angiogenesis. Knockdown of circulating miR-122-5p significantly diminished exercise-induced pro-angiogenic effects in skeletal muscles while enhancing muscle performance[63].

Overall, we conducted bioinformatics and experimental analyses to investigate the role of miR-122-5p in DU. MiR sequencing comes with certain limitations. The data utilized were derived from public databases that are subject to ongoing updates and may impose potential distortions. Specifically, the current experiments did not provide direct evidence that miR-122-5p influences fibrosis during wound healing, and the precise molecular mechanisms underlying its regulation of inflammatory responses remain to be fully elucidated. Thus, further studies are needed to validate whether miR-122-5p’s influences the shift from the inflammatory to the proliferative phase in DFU.

CONCLUSION

MiR-122-5p is upregulated in DFUs, leading to elevated levels of HIF-1α, MMP9, and TNF-α during the inflammatory phase. MiR-122-5p targets VEGF, reducing fibrosis markers FN1 and α-SMA in DFUs, and inhibits fibroblast migration and proliferation, thereby hindering the transition from the inflammatory to the proliferative phase and slowing ulcer repair.