This article is mentioned in:

Introduction

Hypertrophic scars (HS) are fibroproliferative disorders resulting from abnormal wound healing processes, characterized by excessive extracellular matrix (ECM) accumulation and ongoing inflammation. These scars develop in 30–90% of individuals, with higher incidences reported among adolescents and pregnant women and a markedly high prevalence following deep dermal injuries such as burns (1). Particularly in patients with full thickness burns, the prevalence of HS can be as high as 70%, highlighting a substantial clinical challenge due to the significant functional and aesthetic disruptions and the psychological distress these scars often cause (2). The presence of HS substantially affects the quality of life of patients, causing physical symptoms such as severe itching, tenderness and pain, as well as psychological effects such as sleep disturbances, anxiety and depression (3). These symptoms are known to markedly contribute to morbidity, particularly following burn injuries, leading to disruptions in daily activities and psychosocial impairments (4).

Despite the variety of treatments available for HS, such as corticosteroid injections, laser therapy, pressure therapy, silicone applications, botulinum toxin, surgical interventions and radiation therapy, there remains a significant demand for more specialized and efficacious therapeutic choices (5). TGF-β1 inhibitors, as one of the emerging treatments, target a crucial pathological element of HS by blocking the activity of TGF-β, a cytokine known to contribute to fibrosis by stimulating fibroblast proliferation and collagen production (6). This advance toward targeted therapies emphasizes the necessity of continuing research dedicated to improving the precision and effectiveness of treatments for HS and to lessen the extensive effect these scars have on individuals. Within this research frontier, non-coding RNAs (ncRNAs) emerge as crucial regulators in the post-transcriptional network, influencing a wide array of biological pathways. Although ncRNAs hold substantial promise for HS therapy due to their regulatory capabilities, the development of ncRNA-based treatments is challenging. The complexity of the ncRNA-mediated regulatory networks, coupled with an incomplete understanding of their mechanistic roles in scar formation and remodeling, poses significant scientific challenges. Practical issues related to the stability, delivery and specificity of ncRNA-based interventions further complicate their clinical translation (7).

Adipose-derived stem cells (ADSCs) have gathered attention in regenerative medicine for their versatility and therapeutic potential. Derived from human adipose tissue, these mesenchymal stem cells offer multipotency, abundant sources and paracrine effects conducive to regenerative applications ranging from muscle repair to cardiac rejuvenation (8). ADSCs have also shown potential in differentiating into chondrocyte and osteocyte analogues, hinting at new possibilities for bone and cartilage restoration, particularly when paired with scaffold technologies (9). Refining the efficacy of ADSCs for clinical application is an area full of potential for innovation (10). Exosomes derived from ADSCs (ADSCs-Exos) have been identified as powerful agents in enhancing wound healing. They promote regenerative processes, such as cell proliferation and collagen synthesis, thus accelerating wound closure and improving healing quality, which are crucial to both clinical and cosmetic skin repair (11,12). However, as promising as ADSCs-Exos are, challenges remain, particularly in understanding their precise mechanistic roles in healing and in optimizing their use with biomaterials for enhanced therapy (13).

The present study used a rabbit ear model to conduct an in-depth examination of the dynamics of ncRNAs during hypertrophic scar formation. By analyzing the differential expression of ncRNAs, their interactions with mRNAs and the enrichment of specific pathways, the aim of the present study was to uncover potential therapeutic targets for HS, with particular attention to micro (mi)RNA 194 as a notable candidate.

Materials and methods

Establishment of rabbit ear hypertrophic scar model

A total of six female New Zealand white rabbits, aged 3–4 months and weighing 2.5–3.0 kg, were sourced from Shanghai Xinova Medical Research Co., Ltd. The rabbits were acclimated for 1 week under controlled conditions: Temperature was maintained at 22±2°C, relative humidity was kept at 50–60%, and the rabbits were housed under a 12-h light/dark cycle. No abnormalities in the morphology of either ear and no breaks in the soft tissues of the skin were observed. Anesthesia was administered intravenously using sodium pentobarbital (30 mg/kg). Following aseptic preparation, six full-thickness wounds were created on the ventral ear surface, avoiding the perichondrium and major blood vessels. Following surgery, the rabbits were administered antibiotics and analgesics. Each rabbit was housed individually post-operatively. Rabbits were randomly assigned to either the control group or the ADSCs-Exos treatment group, with three rabbits in each group. The health and behavior of the rabbits were monitored every other day throughout the duration of the experiment.

Procaine penicillin (cat. no. 6130-64-9; Merck KGaA) was administered subcutaneously at a dose of 60,000 IU/kg every 48 h for 7 days (14). Concurrently, buprenorphine (cat. no. 52485-79-7; Merck KGaA) was administered subcutaneously at a dose of 0.05 mg/kg every 8 h for 2 days (15,16). The wound care involved the application of either saline or ADSCs-Exos (100 µg/2 ml) over the first 3 days from injury. The micrON hsa-miR-194 agomir (micrON miR-194) or micrON NC agomir (micrON NC) was administrated subcutaneously at a dose of 0.2 µM. Scar progression was evaluated at specified intervals, with tissue collection for histology on days 3, 7, 14 and 21.

Following the administration of anesthesia using sodium pentobarbital at a dose of 30 mg/kg, potassium chloride was rapidly administered intravenously at a concentration of 150 mg/kg (17). The rabbits were subsequently assessed for death based on the following criteria: i) Absence of respiration, pulse and heartbeat for >5 min, verified both by auscultation with a stethoscope and palpation of the cardiac region of the thorax; ii) loss of corneal reflexes; iii) pupil dilation; and iv) absence of neural reflexes. The presence of these conditions confirmed the animal's death. Ethical approval was obtained from the Ethics Committee of The First Affiliated Hospital of Harbin Medical University (approval no., XNM-YX-20220728-01).

ADSCs-Exos were obtained from Shanghai Xinova Medical Research Co., Ltd. The negative control, micrON agomir NC (cat. no. miR4N0000001-4-5) and micrON hsa-miR-194 agomir (cat. no. miR40000460-4-5) were purchased from Guangzhou RiboBio Co., Ltd.

Hematoxylin and eosin (H&E) and Masson's trichrome staining

Histological analyses were conducted on scar tissue samples from both groups. H&E staining was performed to observe cellular structures, while Masson's trichrome staining provided insights into collagen distribution. The experiments were performed according to the established protocols (18).

Briefly, slides were placed in staining can and deparaffinized by immersion in three sequential baths of absolute xylene for 4 min each. This was followed by a series of ethanol washes at concentrations of 100, 100, 95, 90 and 70%, with each wash lasting 4 min. Subsequently, the slides were rinsed under running tap water for 2 min. They were then stained with hematoxylin (cat. no. C0105, Beyotime) for 2 min, followed by another 2-min rinse under tap water. To decolorize, the slides were briefly dipped into 1% acid alcohol three times and rinsed again for 2 min. After decolorization, the slides were immersed in 2% potassium acetate for 3 min and rinsed once more for 2 min. The staining process was completed by submerging the slides in eosin for 2 min, followed by a final 2-min wash under running tap water. All of the aforementioned procedures were carried out at room temperature. The slides were then air-dried for 24 h at 38°C. Prior to observation, the slides were dipped in absolute xylene for 1 min and permanently mounted with a cover slip using DPX mounting medium (cat. no. MM1410; Shanghai Maokang Biotechnology Co., Ltd.).

Slides were deparaffinized in staining cans using three 4-min immersions in absolute xylene, followed by a descending ethanol series from 100–70% in 5% decrements, each for 4 min. They were then submerged in 60°C Bouin's solution (cat. no. HT101128; Merck KGaA) for 45 min and rinsed until clear under running tap water. Nuclei were differentiated with 8 min in modified Weigert's hematoxylin, then rinsed for 2 min. Cytoplasm and erythrocytes were stained with acid fuchsin (cat. no. HT15; Merck KGaA) for 5 min, rinsed for 2 min, treated with phosphomolybdic acid (cat. no. 51429-74-4; Merck KGaA) for 10 min, then stained with methyl blue (cat. no. 28983-56-4; Merck KGaA) for 5 min and rinsed. The slides were treated with 1% acetic acid for 1 min, dehydrated in an alcohol series from 100–70%, each for 1 min, dipped in xylene for 1 min and mounted with a cover slip using DPX (Shanghai Maokang Biotechnology Co., Ltd.). All of the aforementioned procedures were carried out at room temperature.

Immunohistochemical (IHC) staining

Initially, these 4-µm paraffin-embedded tissue sections were incubated at 60°C for 20 min, followed by a deparaffinization process. The antigen retrieval process entailed immersing the slides in a Tris/EDTA buffer (pH 9.0), followed by heating at 100°C for 40 min using a microwave oven. The slides were allowed to cool at room temperature for 30 min before being washed twice with PBS with Tween 20 (PBST) buffer (0.1% Tween 20 in PBS; pH 7.4) for 5 min per wash. The sections were treated with a hydrogen peroxide blocking reagent (cat. no. ab64218; Abcam) for 10 min, followed by two 5-min washes using PBST buffer. The sections were further incubated at room temperature for 1 h in a blocking buffer containing 1% DMSO and 1% bovine serum albumin (cat. no. ab64009; Abcam) in PBS buffer. The TGF-β1 (cat. no. ab215715; 1:500; Abcam) or VEGFA (cat. no. ab52917; Abcam) antibody in blocking buffer were used for primary staining. The sections were incubated with the primary antibodies overnight at 4°C in a humidified chamber. The sections underwent six PBST washes (5 min each) and were then incubated with a secondary goat anti-rabbit HRP antibody (cat. no. ab6112; 1:750; Abcam) in blocking buffer for 1 h at room temperature, followed by three washes in PBST (5 min each). Visualization of the staining was achieved by incubating the sections with 3,3′-diaminobenzidine (cat. no. ab209101; Abcam) for 10 min, according to the manufacturer's instructions. Excess DAB was removed by washing under running deionized water for 5 min. Counterstaining was performed with hematoxylin for 5 min at room temperature, followed by a rinse in distilled water and pat-drying prior to mounting. A Nikon H600L (Nikon Corporation) microscope in normal brightfield mode was used to obtain images. The images of H&E staining, Masson's trichrome staining and IHC staining were captured at a ×100 magnification using a Nikon H600L light microscope (Nikon Corporation).

RNA library preparation and sequencing

The RNA sequencing library from rabbit ear scar tissue on post-operative day 21 was constructed and sequenced by Wuhan HealthCare Biotechnology Co., Ltd. Briefly, total RNA was extracted from samples with an achieved yield of 1 µg per sample. Integrity and purity were confirmed using agarose gel electrophoresis (28S:18S ≥1.5) and Nanodrop spectrophotometry (OD 260/280 ratio between 1.8 and 2.2), respectively. RNA concentration was quantified using a Qubit 4 Fluorometer (Thermo Fisher Scientific, Inc.), with minimum readings of 500 ng/µl.

For small RNA libraries (18–30 nt), adaptors were added to both ends of small RNAs and cDNA was synthesized through reverse transcription, followed by purification using QIAseq miRNA NGS Beads (cat. no. 333923; Qiagen GmbH). For long ncRNAs (lncRNAs) and circular RNAs (circRNAs), RNA was fragmented into ~200 bp segments after rRNA removal. Strand-specific cDNA was synthesized, end-repaired, A-tailed and purified. Single-stranded cDNA libraries were prepared by digesting double-stranded cDNA with USER enzyme (cat. no. 5508; New England BioLabs, Inc.) and enriched by PCR for 15 cycles.

Sequencing was performed on the NovaSeq6000 (Illumina, Inc.) using 2×150 bp paired-end reads (sequencing by synthesis). Post-sequencing, libraries underwent quality control and alignment for differential expression analysis.

Analysis of genetic variability

In the analysis of genetic variability, several key steps were meticulously followed. Initially, read normalization was conducted to ensure uniformity in the number of reads. Subsequently, a statistical model was employed to calculate both the P-value and the fold change for hypothesis testing. To enhance the robustness of the results, a multiple testing correction method was used to obtain the adjusted P-value (adjP). Differentially expressed RNAs were then identified based on specific criteria: A log2 (Fold Change) >1 and a adjP<0.05. This rigorous approach allowed for the identification of miRNAs with statistically significant differences in expression levels among the samples, categorizing them as either upregulated or downregulated. Corresponding P-values and fold changes were obtained and a volcano plot was created to visualize these findings.

Functional enrichment analysis

The functional enrichment analysis of selected differential RNA target genes was conducted using Gene Ontology (GO) (https://geneontology.org/) and Kyoto Encyclopedia of Genes and Genomes (KEGG) (release 109.0) (https://www.genome.jp/kegg/). The analysis was conducted using the clusterProfiler R package (version 4.4.4) in R (version 4.2.1) (https://www.r-project.org/), which facilitated the identification of the primary functions and biological processes (BP) associated with these genes (19). This led to the identification of enriched GO terms and pathways, indicating the significant roles and impacts of differential miRNA target genes on various cellular mechanisms. The top 30 influenced pathways were visualized for detailed insights. GO terms and KEGG pathways were considered statistically significant with P-values <0.05. The gene count thresholds were set to ≥20 for GO terms and ≥4 for KEGG pathways according to the established protocol.

Protein-protein interaction (PPI) analysis

The construction of the PPI network was facilitated using the STRING database (version 12.0, http://string-db.org/), while visualization of the network was performed using Cytoscape (version 3.10.1; National Resource for Network Biology) (20). Interactions among proteins encoded by differentially expressed genes were identified based on predefined confidence criteria, with the network's significant modules being elucidated using analytical tools. This methodology underscored the complex interactions and potential functional pathways influenced by treatment.

Prediction of miRNA-mRNA interaction

TargetScanHuman (Release 7.2, http://www.targetscan.org/vert_72/) (21). miRDB (Version 6.0, http://mirdb.org/) (22) and STARBASE (https://rnasysu.com/encori/) (23) were used to predict interactions between miRNA and mRNA.

Cell culture and transfection

Human keloid fibroblasts (cat. no. CP-H235) were purchased from Procell Life Science & Technology Co., Ltd. CP-H235 cells were cultured in DMEM/F12 culture medium (cat. no. D8437; Merck KGaA), which was supplemented with 10% FBS (cat. no. 12006C; Merck KGaA), 100 U/ml penicillin (cat. no. P3032; Merck KGaA) and 100 µg/ml streptomycin (cat. no. S9137; Merck KGaA). Cells were maintained at a temperature of 37°C in an environment containing 5% CO2. The absence of mycoplasma contamination was confirmed by the supplier.

Cell seeding was conducted in 6-well plates 24 h prior to the commencement of the experiment, ensuring that the cultivation proceeded until the cell density approached ~75% of each well's capacity. In preparation for transfection, 2.75 µl of the pre-diluted liposome reagent (Lipofectamine®; cat. no. L3000001; Thermo Fisher Scientific, Inc.) was mixed with miRNAs at a final concentration of 20 nM, including negative control (NC), miR-194 mimic, inhibitor NC or inhibitor, alongside untreated cells for comparison. This mixture was then allowed a 20-min incubation period to ensure adequate preparation. Following incubation, the mixture was gently introduced to the growth medium and evenly distributed across the wells. The plates were subsequently incubated at 37°C in a CO2-enriched environment for 4–6 h.

The oligonucleotide sequences for the miR-194 mimic, its NC, miR-194 inhibitor and the miR-194 inhibitor NC were produced by Guangzhou RiboBio Co., Ltd. The sequences are as follows: miR-194 mimic, sense strand sequence: 5′-CCAGUGGAGCUGCUGUUACUUC-3′; antisense strand sequence: 5′-GAAGUAACAGCAGCUCCACUGG-3′. miR-194 mimic NC, sense strand sequence: 5′-UUCUCCGAACGUGUCACGU-3′, antisense strand sequence: 5′-ACGUGACACGUUCGGAGAA-3′. miR-194 inhibitor, sequence: 5′-CAGAUAACAGCAGCCCCACUGG-3′, miR-194 inhibitor NC, sequence: 5′-CAGUACUUUUGUGUAGUACAA-3′. Subsequent experiments were conducted 48 h post-transfection.

RNA isolation and reverse transcription-quantitative PCR (RT-qPCR)

RNA was extracted from rabbit scar tissue on post-operative day 21 using 1 ml TRIzol® reagent (cat. no. 15596026; Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocols and homogenized with a rotor-stator. For RT, 2 µg total RNA, components of the reaction mix and water were combined to a total volume of 20 µl. The mixture was incubated through specific temperature cycles to synthesize cDNA, according to the manufacturer's instructions (QuantiTect Rev. Transcription Kit, cat. no. 205311; Qiagen GmbH). The reaction mix of qPCR included 50 ng template DNA/cDNA, 0.5 µl of each primer, 2 µl PCR buffer, 1 U of Taq DNA polymerase and 0.3 µl PowerTrack SYBR Green Master Mix (cat. no. A46109; Thermo Fisher Scientific, Inc.), adjusted to 20 µl with deionized water. Thermal cycling conditions comprised an initial denaturation at 95°C for 5 min, followed by 35–40 cycles of denaturation at 95°C for 30 sec, primer-specific annealing and extension at 72°C for 60 sec, concluding with a final extension at 72°C for 5 min. qPCR was performed with Sybr Green on an Exicycler 96 instrument (Bioneer Corporation). RT-qPCR was conducted in triplicate to ensure reproducibility. The primers for the genes of interest are listed in Table I. Gene expression was normalized to human or rabbit β-actin to analyze the relative expression using the previously described 2−ΔΔCq method (24).

Table I. Primers for quantitative PCR assays and corresponding targeted genes.

Table I.

Primers for quantitative PCR assays and corresponding targeted genes.

Gene	Accession number	Forward primers (5′-3′)	Reverse primers (5′-3′)
Human TGF-β1	XM_054321897.1	TACCTGAACCCGTGTTGCTCTC	GTTGCTGAGGTATCGCCAGGAA
Rabbit TGF-β1	XM_051836380.1	GCCTGCAGAGGCTCAAGTTA	CAACCACTCTGCTGTGTTGC
Human β-actin	NM_001017992.4	CATGATAGGGCGTCCTCGAC	TGAGGAGGATGGGATGCTCA
Rabbit β-actin	NM_001101683.1	TCTCGACGAAACCTAACGGC	GTCACCTTCACCGTTCCAGT
Hsa-miR-194-3p	MIMAT0004671	AAGAATCCAGTGGGGCTGC	GTCGTATCCAGTGCAGGGT
Human U6	NR_004394.1	CTCGCTTCGGCAGCACA	AACGCTTCACGAATTTGCGT
Rabbit U6	XM_008267384.3	CGGACGACCAGTTGTGGTAA	CAGGGTCTTCACATTCGCCT

The stem-loop RT-PCR method was used as outlined by the sRNAPrimerDB protocol from Huazhong Agricultural University (http://www.srnaprimerdb.com/protocolC) (25). Initially, the stem-loop RT primer 5′-GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACCAGATA-3′ was specifically designed to hybridize with hsa-miR-194-3p, facilitating its reverse transcription. The cDNA product was then amplified in real-time assays using the forward primer 5′-AAGAATCCAGTGGGGCTGC-3′ and the reverse primer 5′-GTCGTATCCAGTGCAGGGT-3′. The reverse-transcription, cDNA synthesis and qPCR assays were performed on an Exicycler 96 instrument (Bioneer Corporation) following the manufacturers' instructions.

Western blotting

Whole-cell protein extracts were obtained using 100 µl NP-40 lysis buffer (comprising 150 mM sodium chloride, 1.0% NP-40 and 50 mM Tris, pH 8.0) supplemented with ProteoGuard™ protease inhibitor cocktail (cat. no. 635672; Takara Bio, Inc.) in accordance with the manufacturer's instructions. Pierce™ BCA Protein Assay Kit (cat. no. 23225; Thermo Fisher Scientific, Inc.) was used for protein quantification according to the manufacturer's instructions. A total of 25 µg protein from each sample was then separated by SDS-PAGE on 10% gels and was subsequently transferred to nitrocellulose membranes (cat. no. 88018; Thermo Fisher Scientific, Inc.) by electroblotting. The membranes were blocked for 1 h with a 1% bovine serum albumin solution (cat. no. J61089.AP; Thermo Fisher Scientific, Inc.) at room temperature before being incubated overnight at 4°C with the following primary antibodies: Anti-TGF-β1 (cat. no. ab215715; 1:1,000; Abcam), anti-VEGFA (cat. no. ab52917; 1:1,000; Abcam), anti-collagen I (cat. no. ab138492; 1:500; Abcam) and anti-GAPDH (cat. no. ab8245; 1:2,000; Abcam). The membranes were then washed with PBST three times (10 min each time), before being incubated for 1 h at room temperature with the corresponding HRP-conjugated secondary antibodies, goat anti-rabbit IgG H&L (cat. no. ab6721; 1:2,500; Abcam) and goat anti-mouse IgG H&L (cat. no. ab205719; 1:2,500; Abcam). Visualization of the proteins was achieved using the Gel Doc EZ Gel Documentation System (Bio-Rad Laboratories, Inc.). The densitometric analysis of the protein bands was performed using ImageJ software (version 1.4; National Institutes of Health). The expression levels of the target proteins were normalized to GAPDH expression to ensure accurate quantitative analysis.

Luciferase reporter assay

To generate the wild-type and mutant variants of the target nucleotides, a segment spanning nucleotides 40262 to 40504 in Homo sapiens TGF-β1 was selected, including the predicted sponging region of hsa-miR-194-3p and TGF-β1 3′UTR. Initially, PCR was performed using the initial forward primer pairs. Mutagenic primers for the 3′UTR incorporating ‘GGTGAG’ were designed using PrimerX (https://www.bioinformatics.org/primerx/index.htm). These primers were then used to generate mutants by employing the Q5 Site-Directed Mutagenesis Kit (cat. no. E0554; New England Biolabs, Inc.) and using the Exicycler 96 instrument (Bioneer Corporation) for the PCR amplification. The mutants were created following the manufacturer's detailed instructions. To connect the multiple cloning site to both the wild-type and mutant target sequences, the forward primer with an EcoRI site and the reverse primer with an XhoI site were used. Next, the amplified products were inserted into the pCMV-Green Renilla Luc Vector (cat. no. 16153; Thermo Fisher Scientific, Inc.) for cloning. The primers and corresponding products are summarized in Table II. In the transfection process, a mixture of 0.5 µg Renilla Luc-TGF-β1 3′ UTR [or TGF-β1 mutant (MUT)], 20 nM miR-194 (or NC) and 2.5 µl of the liposome reagent (cat. no. L3000001; Thermo Fisher Scientific, Inc.) was diluted in 2 ml DMEM/F12 culture medium. The transfection was then performed in accordance with the manufacturer's instructions. The detection of luciferase activity was conducted 48 h after transfection.

Table II. Primers and the corresponding products in establishing the wild-type and mutant TGF-β1 3′UTR.

Table II.

Primers and the corresponding products in establishing the wild-type and mutant TGF-β1 3′UTR.

Primer name	Sequence (5′-3′)	Products
Initial forward primer	GTGGTTGCCAGCATAATCC	Used in the first round PCR
Initial reverse primer	AGAGTCCTGAGGATATTCTAGA	Used in the first round PCR
Wild-type forward primer	GAATTCGTGGTTGCCAGCATAATCC	Contains EcoRI site for the second round PCR
Wild-type reverse primer	CTCGAGAGAGTCCTGAGGATATTCTAGA	Contains XhoI site for the second round PCR
Mutant forward primer	GAATTCGTGGTTGCCAGCATAATCC	Contains EcoRI site for the second round PCR
Mutant reverse primer	CTCGAGGTTGTTGGTGAGGTGGAGAGTCCTGAGGATATTCTAGA	Contains XhoI site and mutation ‘GGTGAG’

[i] Wild-type TGF-β1 3′UTR, pCMV-Green Renilla Luc-wild-type TGF-β1 3′UTR; Mutant TGF-β1 3′UTR, pCMV-Green Renilla Luc-mutant TGF-β1 3′UTR (TGF-β1 MUT).

For the purpose of initiating cell lysis, 5X passive lysis buffer was diluted to a 1X concentration using distilled water. This diluted solution was then dispensed into 96-well plates, at 100 µl per well. The disrupted cell solution was subsequently transferred into 1.5 ml centrifuge tubes and centrifuged at 13,200 × g for 10 min at a temperature of 4°C. The supernatant was decanted into a fresh tube, Next, Luciferase Assay Reagent II (cat. no. E1910; Promega Corporation) and the cell lysate were thoroughly equilibrated to room temperature before mixing. A volume of 100 µl Luciferase Assay Reagent II was combined with 20 µl prepared cell lysate in a new 96-well plate. Subsequently, 100 µl Dual-Glo Stop & Glo Buffer (cat. no. E314B-C; Promega Corporation) was added to the mixture, which was gently mixed 2–3 times, with a 2-sec wait before analysis. This procedure enabled the measurement of Renilla luciferase activity, providing a quantifiable luminescence readout of reporter gene expression. The reaction products of the Dual Luciferase Assay were quantified by measuring the absorbances at 560 nm (A560) and 480 nm (A480) on a 25-344S BioTek Fluorescence Microplate Reader (Lonza Group Ltd.).

Enzyme-linked immunosorbent assay (ELISA)

TGF-β1 Human ELISA Kit (cat. no. BMS249-4; Invitrogen; Thermo Fisher Scientific, Inc.), IL-1β Human ELISA Kit (cat. no. BMS224-2; Invitrogen; Thermo Fisher Scientific, Inc.) and TNF-α Human ELISA Kit (cat. no. BMS223-2HS; Invitrogen; Thermo Fisher Scientific, Inc.) were used to measure the concentrations of cytokines in both groups. Antigen dilution was performed according to the manufacturer's instructions, followed by sealing the wells with closure solution and incubating at 37°C for 40 min. Wells were washed three times with the provided washing solution, each for 3 min. Diluted samples were dispensed into enzyme-coated wells, with a minimum of two wells per sample, each receiving 100 µl of the sample. Incubation was performed at 37°C for 60 min, followed by a washing step identical to the first. Development involved adding 100 µl of MB-hydrogen peroxide urea solution to each well, incubating for 5 min at 37°C in a light-protected environment and stopping the reaction with 50 µl of termination solution. Concentrations of IL-1β, TGF-β1 and TNF-α were determined by comparing optical density values to a standard curve provided with each kit. The results of ELISA were detected by measuring the absorbances at 450 nm (A450) and 620 nm (A620) on a 25-344S BioTek Fluorescence Microplate Reader (Lonza Group Ltd.).

Statistical analysis

Data are presented as the mean ± standard deviation. Statistical evaluations were conducted using GraphPad Prism software version 8.0 (Dotmatics). Experiments were conducted a minimum of three times to ensure robustness and reproducibility of the findings. For the analysis of differences between two distinct groups, a two-tailed Student's t-test, either paired or unpaired depending on the data structure, was employed. When comparing three or more groups, a two-way ANOVA followed by Tukey's multiple comparisons test was used to identify significant differences among groups. P<0.05 was considered to indicate a statistically significant difference.

Results

Establishment of rabbit ear hypertrophic scar model

To assess the hypertrophic scar formation in rabbit ears, gross examination and histological measurement were performed. In the comparative analysis of hypertrophic scar development over 21 days, the ADSCs-Exos-treated group demonstrated a markedly moderated inflammatory response relative to the control group. This was evidenced by a consistently lower degree of inflammation, redness and swelling. The wound margins in the ADSCs-Exos-treated group healed more quickly compared with the control group, resulting in a faster decrease in scar size. Furthermore, the coloration of scar tissue in the treatment group transitioned rapidly from red to pink, ultimately achieving a similar color to the surrounding skin, signifying an accelerated reduction in blood supply and a more rapid progression through the scar maturation phases (Fig. 1A).

Figure 1. Evaluation of hypertrophic scar formation. (A) Gross morphological changes in rabbit ear hypertrophic scars, comparing control with ADSCs-Exos treated ears. Sequential depiction of hypertrophic scar formation in rabbit ears, showcasing the transition from early post-injury inflammation to late-stage scar maturation at intervals of 3, 7, 14 and 21 days. (B) H&E staining demonstrates the cellular and structural progression of scar tissue. The control group shows denser cellularity and disorganized collagen arrangement compared to ADSCs-Exos treated group over time. (C) Masson's trichrome staining highlighted the collagen deposition within the scar matrix, with marked differences in collagen organization and density between control and ADSCs-Exos treated groups. (D) IHC analysis indicates TGF-β1 expression in scar tissue. The ADSCs-Exos treated group exhibits a gradual decrease in TGF-β1 expression compared to the control. (E) IHC staining for VEGF expression in the hypertrophic scar tissue. The ADSCs-Exos treatment is associated with a reduced VEGF expression. Scale bars represent 100 µm in all images. ADSCs-Exos, Adipose-derived stem cell exosomes; H&E, hematoxylin and eosin; IHC, immunohistochemical.

H&E staining was conducted to elucidate the cellular and structural changes within the scar tissue. Throughout the experiment, the treatment group demonstrated a more regulated arrangement of collagen fibers and less vascular proliferation, indicating a neater and less aggressive healing process compared with the control. The treatment appeared to modulate the wound healing curve, leading to a more structured and less fibrotic scar formation, as evidenced by the more orderly ECM and reduced vascular proliferation seen in the H&E-stained sections (Fig. 1B). Throughout the 21-day study period, Masson's trichrome staining revealed a temporal progression of tissue healing characterized by changes in the inflammatory response and collagen deposition. Across all stages, the treatment group consistently showed fewer and more regularly arranged collagen fibers, suggesting a more controlled healing process and a potential modification of the typical fibrotic response, resulting in the formation of a less dense scar (Fig. 1C). IHC staining showed that TGF-β expression levels in the control group remained high over the 21-day period. By contrast, the ADSCs-Exos-treated group exhibited a marked reduction in TGF-β1 expression (Fig. 1D). In addition, scars in the control group had consistently high levels of VEGF expression. By contrast, the ADSCs-Exos treatment was associated with a significant decrease in VEGF staining, suggesting a potential regulatory effect on angiogenic signaling within the scar tissue (Fig. 1E).

In brief, the evidence in Fig. 1 indicated that adipose-derived stem cell exosomes significantly inhibited the formation of hypertrophic scarring in rabbit ears.

lncRNA expression profiles and regulatory networks in ADSCs-Exos-treated hypertrophic scar formation

High-throughput sequencing of lncRNA from tissues uncovered a broad spectrum of lncRNAs in both the ADSCs-Exos-treated and control groups. The lncRNAs expression profiles are summarized in Table SI. These were effectively visualized in a volcano plot (Fig. 2A), which segregated the upregulated and downregulated lncRNAs, presenting a clear disagreement in the response to treatment. This visualization was complemented by a heatmap (Fig. 2B), which offered a detailed perspective of the expression levels, accentuating the differences in lncRNA profiles between the treatment and control groups. The advanced analysis (Fig. 2C and D) built on the previous expression patterns, revealing that the complex regulatory networks were controlled by these differentially expressed lncRNAs.

Figure 2. Integrative analysis of lncRNA expression profiles, regulatory networks and functional pathways in ADSCs-Exos treated hypertrophic scar formation. (A) Volcano plot representing the differential expression of lncRNAs between Control Group (G1) and ADSCs-Exos treated Group (G2). Significantly upregulated lncRNAs are indicated in cyan, downregulated lncRNAs in purple and non-significant changes in grey. The x-axis represents the log2 fold change and the y-axis shows the-log10 (P-value), highlighting the magnitude and statistical significance of expression differences. (B) Heatmap illustrating the expression patterns of lncRNAs in samples from G1 and G2. A z-score normalized color gradient indicates expression levels, with red for upregulated and blue for downregulated lncRNAs. The hierarchical clustering on the top dendrogram groups samples based on the similarity of their lncRNA expression profiles. (C) Bar graph showing the top 10 significantly enriched GO terms within the BP, CC and MF categories that are associated with differentially expressed lncRNAs. Only GO terms with a P-value ≤0.05 are included. The horizontal axis indicates the statistical significance of the enrichment, represented as-log10 P-value, while the vertical axis lists the names of the significantly enriched GO terms. (D) Network diagram of the top 20 enriched GO terms among genes associated with differentially expressed lncRNAs, organized by the statistical significance-log10 (P-value). The size of each node corresponds to the count of associated candidate genes. This visualization demonstrates the interconnectedness of significant BPs, and MFs, with node size directly relating to the gene count for each GO category. (E) Histogram of KEGG pathways significantly enriched in connection with genes linked to differentially expressed lncRNAs. The horizontal axis lists the names of the significantly enriched pathways, while the vertical axis represents the-log10 (P-value), indicating the level of enrichment significance. Pathways with a P-value ≤0.05 are marked with red bars, signaling considerable enrichment, while blue bars indicate pathways with lesser significance. lncRNAs, long non-coding RNAs; ADSCs-Exos, Adipose-derived stem cell exosomes; GO, Gene Ontology; BP, biological process; CC, cellular component; MF, molecular function; KEGG, Kyoto Encyclopedia of Genes and Genomes.

GO functional analysis systematically categorized the roles of genes influenced by the lncRNAs. A reference set of genes was analyzed to determine statistically significant enrichments, using Fisher's exact test for initial significance testing. The resulting P-values were subsequently adjusted for multiple comparisons, applying a controlled false discovery rate to enhance the robustness of the findings. This analysis revealed significant enrichments across BP, cellular components (CC) and molecular functions (MF), highlighting the extensive networks and pathways at play. To explore the broader implications of these lncRNA-mediated changes, KEGG pathway analysis was conducted. This annotated the modulated genes within known metabolic pathways, linking them to established KEGG pathways and merged this with differential expression data for a comprehensive understanding of the metabolic pathways involved (Fig. 2E).

As well as the lncRNA expression profiles, the differentially expressed circRNAs were examined (Table SII). The results were visualized using both heatmap and volcano plot representations (Fig. S1). Although distinct expression patterns were evident in the heatmap (Fig. S1A), the corresponding volcano plot demonstrated that the observed differences in circRNA expression between the treatment and control groups did not reach statistical significance (Fig. S1B). These findings indicated that, under the given experimental conditions, the alterations in circRNA expression were not statistically significant.

Collectively, these insights suggested that the application of ADSCs-Exos exerted a modulatory effect on lncRNA expression, which in turn may play a pivotal role during HS formation. The identified GO terms and KEGG pathways provided a deeper grasp of the BP and MF associated with the treatment, laying the groundwork for uncovering potential therapeutic targets.

miRNA regulation in ADSCs-Exos-treated scar formation

An integrative analysis was performed on the miRNA expression profiles in HS treated with ADSCs-Exos (Table SIII). The volcano plot (Fig. 3A) outlined the significantly modulated miRNAs, separating those upregulated from those downregulated. Furthermore, the heat map (Fig. 3B) detailed the expression levels across samples, exhibiting distinctive patterns. Functional enrichment analysis was conducted using GO, highlighting the involvement of the miRNAs in BP, CC and MF (Fig. 3C). Regulatory networks were also visualized to display the complex interactions between miRNAs and their target genes (Fig. 3D).

Figure 3. Integrative analysis of miRNA expression profiles, functional enrichment and regulatory networks in ADSCs-Exos treated hypertrophic scar formation. (A) Volcano plot visualizing the differential expression of miRNAs between the control group (G1) and the ADSCs-Exos treated group (G2). Purple points indicate miRNAs significantly downregulated in the ADSCs-Exos treated group (G2) compared to the control (G1) and cyan points represent miRNAs significantly upregulated in G2 relative to G1. Grey points correspond to miRNAs with no significant difference in expression between the two groups. (B) Heatmap detailing the expression levels of miRNAs across individual samples in both groups. Red indicates higher expression and blue denotes lower expression levels. The hierarchical clustering on the y-axis groups miRNAs with similar expression patterns, while the x-axis separates the samples into control (G1) and treated (G2) groups. (C) GO enrichment in BP, CC and MF associated with the differentially expressed miRNAs. The length of each bar corresponds to the-log (P-value), signifying the enrichment significance of the GO terms. (D) Network diagram of the most significant GO terms connected with the differentially expressed miRNAs. The size of the nodes reflects the number of miRNA-related genes within each GO term and the layout of the network illustrates the interconnectedness and relevance of these terms to the ADSCs-Exos treatment effects. miRNA. microRNA; ADSCs-Exos, Adipose-derived stem cell exosomes; GO, Gene Ontology; BP, biological process; CC, cellular component; MF, molecular function; KEGG, Kyoto Encyclopedia of Genes and Genomes.

Briefly, the results demonstrated significant changes in miRNA expression profiles and related biological pathways due to ADSCs-Exos treatment of hypertrophic scar formation.

mRNA expression and pathway analysis in ADSCs-Exos-treated HS

An extensive analysis was undertaken to understand the modulation of mRNA expression in HS post-ADSCs-Exos intervention. The mRNA expression profiles are summarized in Table SIV. Statistical analysis was performed, as evidenced by the volcano plot (Fig. 4A), which delineated differentially expressed mRNAs with statistical rigor. mRNAs with a two-fold increase and P<0.05 were identified, where cyan and purple points represented statistically significantly upregulated and downregulated genes, respectively. A heatmap (Fig. 4B) further illustrated these differences, with gene expression variations between two groups, the control and the treated group, evident through distinctive red (upregulated) and blue (downregulated) color gradients. The GO categorization and KEGG pathway analysis (Fig. 4C and D) revealed patterns and themes within the data. Notably, ‘cardiac muscle tissue morphogenesis’ and ‘sarcomere structures’ emerged as significantly enriched terms within the BP and CC categories. The scatter plot within the enrichment analysis visually underscored these themes, with the size of the markers correlating to gene counts and hues representing categorical significance. Contrasts between the observed patterns and the expected outcomes highlighted the effect of ADSCs-Exos treatment on the association between mRNA expression and hypertrophic scar pathophysiology.

Figure 4. Integrative analysis of mRNA expression profiles, enrichment pathways and regulatory networks in ADSCs-Exos treated hypertrophic scar formation. (A) Volcano plot illustrating the differential expression of mRNAs in rabbit ear scars, comparing the control group (G1) and the ADSCs-Exos treated group (G2). Purple points denote mRNAs significantly downregulated in G2, while cyan points represent mRNAs significantly upregulated in G2, relative to G1. Grey points indicate mRNAs with no significant expression changes. (B) Heatmap showing the mRNA expression levels across samples, with red indicating upregulation and blue signifying downregulation. The expression data are hierarchically clustered to group similar expression patterns in G1 and G2. (C) Stacked bar chart of pathway enrichment based on differentially expressed mRNAs. GO categories include BP, CC, MF and KEGG pathways, with the significance of enrichment represented by the-log10 (P.adj) value. (D) Network diagram of the top enriched pathways associated with the differentially expressed mRNAs. The size of each node reflects the number of mRNAs involved in that particular pathway and the diagram highlights the most significantly altered pathways in the ADSCs-Exos treated group. ADSCs-Exos, Adipose-derived stem cell exosomes; GO, Gene Ontology; BP, biological process; CC, cellular component; MF, molecular function; KEGG, Kyoto Encyclopedia of Genes and Genomes.

miR-194 orchestrates an extensive regulatory network affecting gene expression

The analysis of the present study began by mapping the potential targets of significantly changed miRNAs following ADSCs-Exos treatment (Table III). The Venn diagram (Fig. 5A) provided a visual summary of the extensive regulatory capacity of these miRNAs. The genes found in the intersecting region all bind to the three miRNAs, suggesting that they collectively play a role in regulating scar formation. This subset was determined through the intersection of predictions and experimentally validated miRNA interactions, providing a strong foundation for subsequent analyses. In addition, GO enrichment analysis (Fig. 5B) was performed on the intersected gene set, revealing significant BP, CC and KEGG categories that the mRNAs candidates may influence. In the BP category, ‘protein localization to cell’ was markedly enriched, whereas ‘postsynaptic specialization’ was the primary CC affected. The KEGG analysis indicated significant enrichment in ‘TGF-β signaling pathway’. Furthermore, the PPI network diagram (Fig. 5C) showed the putative PPIs among the targets shared by the miRNAs. The network described numerous connections between the target genes, with nodes representing genes and edges denoting the predicted regulatory associations. Particularly, TGFB1, EGFR and BMPR2 emerged as focal nodes within the network, each exhibiting a high degree of connectivity, suggesting their significance as influential hubs within the PPI landscape.

Figure 5. miR-194-mediated regulatory networks on gene expression. (A) Venn diagram depicting the shared target genes of the indicated miRNAs, illustrating the extensive regulatory reach of these miRNAs within the gene expression landscape. The diagram highlights the number of unique and overlapping target genes predicted by multiple databases. (B) Bar graph of GO and KEGG pathway enrichment analysis among the intersecting targets. The categories of BP, CC and significant KEGG pathways are represented, with the significance of enrichment indicated by-log10(P.adj) values. (C) Network of protein-protein interactions among the intersecting targets, revealing potential regulatory pathways that miRNAs may modulate. (D) Schematic representation of the predicted miR-194 seed regions within the 3′ UTR of the TGF-β1 gene. This alignment suggests a direct regulatory mechanism by which miR-194 may post-transcriptionally silence TGF-β1 expression. (E) The expression of TGF-β1 and miR-194 in CP-H235 cells were measured by quantitative PCR. (F) The relative dual luciferase activity of TGF-β1 3′ UTR (TGF-β1) and TGF-β1 3′ UTR mutant (TGF-β1 MUT) was detected by luciferase reporter assay. The relative luciferase activity of TGF-β1 is reduced in the presence of miR-194, while that of TGF-β1 MUT alters little. *P<0.05; NS vs. NC group. (G) Results of western blotting show the TGF-β1 is reduced in the miR-194 group vs. NC group. miRNA/miR. microRNA; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; BP, biological process; CC, cellular component; MUT, mutant; NS, not significant; NC, negative control.

Table III. Differential expression of miRNAs after ADSCs-Exos treatments.

Table III.

Differential expression of miRNAs after ADSCs-Exos treatments.

ID	FoldChange_Log2	P-value
ocu-miR-144-3p	−1.309525013	6.407×10−4
ocu-miR-144-5p	−1.347413066	1.976×10−4
ocu-miR-181c-3p	−1.201633861	1.003×10−4
ocu-miR-206-3p	−5.247927513	6.750×10−4
ocu-miR-215-3p	1.378511623	2.821×10−4
ocu-miR-215-5p	3.790431619	5.560×10−4
ocu-miR-376b-5p	1	2.713×10−4
ocu-miR-451-5p	−1.470746932	4.324×10−4
ocu-miR-486-5p	−1.168227713	3.874×10−4
ocu-miR-194-3p	7.48762561	8.040×10−4

[i] miR, microRNA; Ocu, Oryctolagus cuniculus.

According to the aforementioned results and the fact that TGF-β-mediated signaling pathways dominate the proliferation of wound healing, TGF-β1 was selected for further investigations. Schematic evidence of the action of miR-194 on the TGF-β1 gene (Fig. 5D) was provided. The diagram displays the predicted binding of miR-194 to the seed regions within the 3′ UTR of TGF-β1. This suggested a direct post-transcriptional regulatory mechanism, where miR-194 can silence TGF-β1 expression, a critical gene in fibrogenesis.

The Renilla Luc-TGF-β1 3′UTR or Renilla Luc-TGF-β1 3′UTR MUT, along with NC or miR-194 mimic were introduced into cells before the luciferase reporter assay. The expression of miR-194 was validated by PCR assays (Fig. 5E). Finally, the results of the luciferase reporter assay (Fig. 5F) confirmed the functional effect of miR-194 on TGF-β1 translation. When miR-194 was present, the luciferase activity linked to the TGF-β1 3′ UTR was significantly reduced, which was not observed with the MUT promoter. This reduction was consistent with the findings of western blotting, which showed a decrease in TGF-β1 protein levels upon the introduction of miR-194, validating the repression effect of miR-194 on TGF-β1 expression (Fig. 5G).

In conclusion, miR-194 strongly influenced gene regulation networks, with a specific suppression of TGF-β1, suggesting its potential as a therapeutic target for diseases characterized by aberrant fibrogenesis.

miR-194 modulates hypertrophic scar formation by targeting TGF-β1

The efficacy of miR-194 intervention on hypertrophic scar formation was evaluated by in vitro and in vivo experiments. The transfection efficacy was confirmed by qPCR assays, demonstrating that a significant increase in miR-194 expression was observed in the mimic group, whereas a decrease was noted in the inhibitor group compared to their respective NCs. In the untreated, NC and inhibitor NC groups, the miR-194 expression remained relatively stable (Fig. 6A). Consequently, the miR-194 mimic and inhibitor were employed in subsequent experiments. A significant reduction in TGF-β1 RNA expression was demonstrated by the miR-194 mimics, similar to that observed with ADSC-Exos treatment (Fig. 6B). Concurrently, the expression of pro-inflammatory cytokines IL-1β, TNF-α and IL-10 was diminished, as measured by ELISA (Fig. 6C). A reduction in TGF-β1 protein levels was also observed (Fig. 6D). Conversely, an increase in TGF-β1 expression and cytokine levels was observed with miR-194 inhibitor treatment (Fig. 6E-G).

Figure 6. miR-194 inhibits hypertrophic scar formation via suppression of TGF-β1. (A) Expression of TGF-β1 and miR-194 at the RNA level measured by quantitative PCR. *P<0.05 vs. NC group. (B) The relative RNA expression of TGF-β1 and miR-194 are determined by quantitative PCR. *P<0.05; NS vs. ADSC-Exos+NC group. (C) Concentrations of specified cytokines measured by ELISA analysis. *P<0.05; NS vs. NCADSC-Exos group. (D) The protein levels of TGF-β1 detected by western blotting. (E) RNA levels of TGF-β1 and miR-194 are determined by quantitative PCR. *P<0.05 vs. miR-194 inhibitor NC group. (F) Cytokine concentrations detected by ELISA analysis. *P<0.05 vs. miR-194 inhibitor NC group. (G) The protein levels of TGF-β1 detected by western blotting. (H) Rabbit ear scars are observed grossly over a period of 21 days, control miRON NC vs. miR-194 overexpressed tissues. A reduction in scar erythema, flattening of scar borders and diminution of scar size are noted with miR-194 overexpression, indicating a mitigated hypertrophic response. (I) Top: Significant differences between groups are shown by H&E staining. Disorganized collagen fibers and abundant inflammatory cells are displayed by the control miRON NC group. More organized collagen, fewer inflammatory cells and improved cellular architecture result from miRON miR-194 treatment. Bottom: Masson's trichrome staining suggests that dense collagen, abundant fibroblasts and active angiogenesis are exhibited by the control miRON NC group. Reduced collagen density, fewer fibroblasts and decreased angiogenesis are displayed by the miR-194 group, indicating attenuated scar formation. (J) Expression of miRON miR-194 and TGF-β1 in scar tissues is validated by quantitative PCR. *P<0.05 vs. miRON NC group. A graph illustrating TGF-β1 and miR-194 RNA levels between groups confirms the inhibition of miR-194 overexpression on TGF-β1 expression. (K) Cytokine concentrations in scar tissues are measured by ELISA. (L) The expressions of TGF-β1, VEGF and COL-1 determined by western blotting, with GAPDH serving as a loading control. The attenuation of key inflammatory factors involved in scar formation by miR-194 overexpression is shown. *P<0.05 vs. miRON NC group. Untreated, cells that have not received treatment; NC, the negative control for the miR-194 mimic; miR-194 inhibitor NC, the negative control for the miR-194 inhibitor; H&E.

The effect of miR-194 on hypertrophic scar formation was further investigated in vivo over a 21-day period. Wounds treated with miR-194 mimics exhibited faster closure compared with the control group, with notable diminutions in scar erythema, scar borders and scar size being noted, indicating a mitigated hypertrophic response (Fig. 6H). By H&E staining, a lowering in the scar elevation index in the miR-194-treated group was revealed, suggesting lessened scar tissue formation (Fig. 6I, top). Complementary Masson's trichrome staining supported these findings, showing a significant cutback in collagen deposition in the miR-194-treated samples (Fig. 6I, bottom). Quantitative assessments of post-intervention gene expression confirmed a marked downturn in TGF-β1 mRNA in the miR-194-treated group (Fig. 6J). This was accompanied by a decrease in the levels of TGF-β1, IL-1β and TNF-α, as indicated by ELISA (Fig. 6K). By western blotting, a decrease in TGF-β1 protein levels and a decline in the expression of fibrotic markers VEGF and type I collagen (COL-1) in the miR-194 group were validated, confirming the inhibitory effects on key inflammatory factors involved in scar formation (Fig. 6L).

Collectively, a significant inhibitory on hypertrophic scar development was exerted by miR-194, primarily through the suppression of TGF-β1.

Discussion

In the present study, rabbit ear models were used to investigate the effects of adipose-derived stem cell exosomes (ADSCs-Exos) on HS, a common and particularly challenging type of proliferative scarring. HS is characterized by raised, often discolored scar tissue that extends beyond the boundaries of the original wound, significantly affecting the patient's self-image, causing physical discomfort and diminishing overall quality of life. Current HS management strategies, including surgical excision, laser therapy and various pharmacological interventions such as glucocorticoids, are associated with a number of drawbacks, ranging from the risk of further scar formation following surgery to inconsistent outcomes and the potential for adverse effects, such as skin atrophy and hyperpigmentation. Such treatments, while partly effective, do not provide a definitive solution and are frequently associated with a high rate of scar recurrence (4). This highlights the pressing need for novel and more effective treatments for HS, which was the motivation for the present exploration of the therapeutic promise of ADSCs-Exos in this challenging clinical area.

In exploring the intricate regulatory landscape of ncRNAs within the context of proliferative scarring, the present study delved into the roles of lncRNAs, miRNAs and circRNAs. These ncRNAs, which do not encode proteins, are nonetheless pivotal in regulating gene expression and consequently have a profound effect on disease development, including scarring processes (26). LncRNAs such as H19 are implicated in exacerbating scarring by promoting fibroblast proliferation and collagen synthesis, highlighting their effect on scar development through numerous signaling pathways (27). Concurrently, adipose-derived stem cell exosomes (ADSCs-Exos) have shown promise in reducing scarring by modulating these BP. They transport proteins, lipids and RNAs, effectively attenuating inflammation, enhancing fibroblast activity and accelerating wound healing and scar remodeling (28,29) miRNAs (miRs), short single-stranded ncRNAs, are critical in post-transcriptional gene regulation by targeting mRNAs for degradation or translational repression. For example, miR-21 is overexpressed in proliferative scarring, activating the PI3K/AKT pathway by suppressing phosphatase and tensin homolog expression, thus promoting fibroblast proliferation and collagen synthesis (30). The sequencing data from scar model tissues and post-treatment samples revealed an array of differentially expressed miRNAs, implicating their regulatory capacity across various BP and signaling pathways. Of note, the expression of TGF-β1, a protein significantly implicated in scar formation, was markedly influenced by ADSCs-Exos via ncRNA mediation. Furthermore, miRNAs within ADSCs-Exos, such as miR-29, target fibroblast activities crucial to scarring, affecting collagen production and fibroblast differentiation (31). For instance, miR-124, reduces inflammation by modulating macrophage activity and cytokine release, which are crucial for mitigating the inflammatory response integral to scar formation (32). CircRNAs also contribute by functioning as miRNA ‘sponges’, exemplified by circRNA CDR1as, which absorbs miR-7 to promote collagen synthesis and fibroblast proliferation (33). However, the present findings from rabbit ear scar model tissues showed that, while circRNAs were abundantly expressed, they lacked statistical significance. Overall, the present study underscored the complex interactions of ncRNAs in ADSCs-Exos and their promising role in the innovative treatment of proliferative scarring.

The present study revealed that ADSCs-Exos significantly reduced proliferative scarring in rabbit ear models. This was accompanied by a marked change in the expression of miRNAs and mRNAs, particularly the post-treatment upregulation of miR-194, which is implicated in regulating proteins including TGF-β1, a protein downregulated following exosome treatment. The overexpression of TGF-β1, crucial in proliferative scar formation, enhances fibroblast proliferation and ECM deposition, thereby exacerbating scarring (34). The regulatory effect of miR-194 on TGF-β1 suggested the therapeutic potential of adipose-derived stem cell exosomes in scarring, as evidenced by predictive software and validated through luciferase reporter assays showing the ability of miR-194 to downregulate TGF-β1 activity. Previous studies have shown that several miRs, including miR-124-3p, miR-26a, miR-29b, miR-145-5p and miR-485-5p, contribute to the suppression of HS formation by downregulating TGF-β1 (35–38). By contrast, miR-15a-5p enhances hypertrophic scar development by boosting the TGF-β1 signaling pathways (39). The verification of miR-15a-5p and TGF-β1 sponge would be necessary to investigate the hyperactivity of TGF-β1. The differences between the findings of previous studies suggested that the mechanisms underlying miRNA-regulated TGF-β1 on HS initiation and development need further exploration. Subsequent experiments further confirmed the role of miR-194 in the modulation of TGF-β1 expression, indicating a promising therapeutic avenue in proliferative scarring management through miR-194 and TGF-β1 regulation.

While TGF-β1 primarily promotes fibroblast proliferation and collagen synthesis, VEGF-mediated angiogenesis is beneficial for normal wound healing. The balance of VEGF-mediated signaling pathways is critical for effective wound healing (40). The present study provided evidence that VEGF expression was decreased in the miR-194-treated group. Similarly, Zhou et al (41) noted that miR-188-5p suppresses VEGF expression, thereby restraining keloid formation. By contrast, Chen et al (42) demonstrated that ADSC treatments improved diabetic wound closure by activating the VEGF-ERK pathway, accompanied by the suppression of inflammatory factors. Qi et al (37) found that the inhibition of miR-145-5p enhances TGF-β2 expression and VEGF secretion, reducing HS development. Liu et al (31) showed that ADSCs-promote miR-21 and miR-29b releases VEGF by activating the PI3K/AKT pathway, facilitating wound healing. The discrepancies between studies may be caused by various interactions, including the interplay of lncRNA-miRNA-mRNA, or by different conditions, such as hypertrophic scarring vs. keloid fibrosis. These controversial results suggest that the global networks of miRNA-mediated VEGF expression require further investigation.

In addition, the crosstalk between TGF-β1 and VEGF cannot be ignored. Although specific research directly connecting TGF-β1 and VEGF in hypertrophic scarring has not been addressed, several studies have emphasized the potential for crosstalk between these factors. Consistent with the current findings, Zhang et al (43) demonstrate that VEGF and collagen fibrosis are significantly increased in HS compared with adjacent skin. Kinashi et al (44) illustrated that TGF-β1 regulates VEGF expression in peritoneal fibrosis, promoting lymphangiogenesis. Komi et al (45) highlight that mast cell-derived VEGF, TGF-β and other mediators trigger scar formation, providing a promising therapeutic target to improve wound healing. However, the specific relationship between TGF-β1 and VEGF in HS has not been fully elucidated. Further investigations into the interplay of TGF-β1 and VEGF in hypertrophic scarring would help clarify the mechanisms involved.

The development of proliferative scars is closely associated with the activation and infiltration of inflammatory factors, which are critical to the post-traumatic inflammatory response, fibrosis and scar tissue remodeling. Key inflammatory cytokines such as IL-1 and TNF-α play instrumental roles in these processes. TNF-α, as an early mediator of inflammation, enhances the recruitment and activation of immune cells, increases vascular permeability and initiates the inflammatory response. Simultaneously, IL-1 stimulates fibroblast proliferation and collagen synthesis, directly linking inflammation to the fibrotic changes observed in proliferative scars (46). The present study highlighted the significant regulatory effect of miR-194 on these inflammatory pathways. It was found that miR-194 inhibitors significantly increased the levels of inflammatory cytokines, whereas miR-194 mimics decreased them. This suggested that miR-194 not only has the ability to suppress TGF-β1 expression, which is known to drive fibroblast proliferation and collagen synthesis, but also plays a crucial role in modulating the inflammatory response itself.

In addition, the balance between collagen synthesis and degradation, which is critical during normal wound healing, is disrupted in proliferative scars. Here, there is a marked increase in type I collagen synthesis coupled with reduced degradation, leading to collagen accumulation within the scar tissue (47). This imbalance is largely driven by the abnormal activation of fibroblasts into myofibroblasts, potentiated by growth factors such as TGF-β and platelet-derived growth factor. These myofibroblasts exhibit enhanced synthetic capabilities, significantly contributing to the excessive deposition of type I collagen seen in HS (45). TGF-β, in particular, plays a pivotal role by promoting the increased production of type I collagen, further exacerbating the fibrotic component of scarring (48). The findings that miR-194 can modulate both TGF-β1 levels and the inflammatory milieu offer promising therapeutic insights into managing the complex pathophysiology of proliferative scarring.

ADSC-Exos have garnered considerable interest for their ready availability, straightforward isolation techniques and cost-effectiveness. In the present study, Exo-miR-194 showed significant potential in modulating inflammatory responses, maintaining angiogenic balance and reducing the overactivation of TGF-β-mediated signaling pathways. These properties highlight its promising role in the management of HS formation. A comparative analysis of Exo-miR-194 with standard HS therapies could provide deeper insights into its efficacy and utility, potentially leading to more effective treatment strategies for HS. Further research is warranted to delineate the specific mechanisms through which Exo-miR-194 influences these pathways and to compare its effects with those of established treatments.

While the present study demonstrated the potential of exosomal miR-194 from ADSCs in attenuating hypertrophic scar formation, it is important to acknowledge the limitations of the phenotype verification. The current focus on histological and molecular markers, although indicative, may not fully capture the complexity of hypertrophic scar phenotypes. Despite observing significant reductions in collagen density, fibroblast abundance and angiogenesis in the miR-194 treated group, a more comprehensive phenotypic analysis is required. Future directions include implementing functional assays to evaluate mechanical properties of scar tissue, such as tensile strength and elasticity, over 6–12 months. Extending the observation period will help assess long-term effects on scar maturation and remodeling. Detailed molecular pathway analyses using RNA sequencing and proteomics will explore how miR-194 modulates scar formation. Employing additional animal models with different wound healing profiles will enhance the generalizability of the findings of the present study. Investigating various delivery methods for ADSCs-Exos, such as topical application, local injection, or systemic administration and optimizing dosage and frequency will determine the most effective strategies. These steps will validate the therapeutic potential of miR-194 and ADSCs-Exos in treating hypertrophic scars and facilitate clinical application.

In conclusion, the present findings suggested that ADSCs-Exos, can significantly inhibit scar formation and reduce the expression of pro-inflammatory and fibrotic markers such as TGF-β1, particularly through the modulation of miR-194. This positions ADSC-Exos as a promising therapeutic approach for managing proliferative scarring, offering a novel avenue for intervention in conditions characterized by excessive fibrosis and inflammation.

Supplementary Material

Supporting Data

Supporting Data

Supporting Data

Supporting Data

Supporting Data

Acknowledgements

Not applicable.

Funding

Funding: No funding was received.

Availability of data and materials

The datasets used and/or analyzed during the current study are available in the GEO database under accession numbers GSE271671 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE271671) and GSE271672 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE271672).

Authors' contributions

Conceptualization was performed by ZX and LH. Formal analysis was conducted by ZX and YT. Validation was performed by YT. ZX wrote the original draft and LH was responsible for writing, reviewing and editing the manuscript. ZX and LH confirm the authenticity of all the raw data. All authors read and approved the final version of the manuscript.

Ethics approval and consent to participate

The animals were treated according to protocols approved by the Ethics Committee of The First Affiliated Hospital of Harbin Medical University (approval no., XNM-YX-20220728-01).

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Glossary

Abbreviations

Abbreviations:

References