The ANA-1 murine macrophage cell line (SUNNCELL, Wuhan), originally derived from C57BL/6 mice, and the normal human dermal fibroblast (NHDF) cell line (Pricella, Wuhan, CP-H103) were cultured in RPMI 1640 medium (Gibco, United States) supplemented with 10% fetal bovine serum (FBS, Gibco, United States), 100 μg/mL streptomycin, and 100 U/mL penicillin. Cells were maintained in an incubator (Thermo Fisher Scientific Inc., Waltham, MA, United States) at 37 °C with 5% CO₂. When the cell density reached approximately 75%, transient transfection was performed using Lipofectamine 2000 (Invitrogen) with the following plasmids obtained from China Biotechnology Co., Ltd. (Beijing, China): overexpression (OE)-KDM5A, short hairpin RNA (sh)-KDM5A (sh1-KDM5A [5′-GGAACUGGGUCUCUUUUGA-3′], sh2-KDM5A [5′-GCAAAUGAGACAACGGAAA-3′]), sh-KDM5A+ shSocs1, and their corresponding controls (shNC, shKDM5A+ shNC, OE-NC). Following transfection, cells were cultured for an additional 48 h before collection.

Damaged ANA-1 cells were generated via lipopolysaccharide (LPS) treatment. ANA-1 cells were first seeded into 6-well plates and cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum for 24 h until complete confluence. The medium was replaced with fresh medium containing 100 ng/mL LPS (Sigma-Aldrich). Cells were harvested at 0, 4, and 8 days post-treatment, and total RNA was extracted. KDM5A mRNA expression levels were detected via qRT-PCR to observe the dynamic changes in KDM5A under the inflammatory microenvironment.

Adult BALB/c mice were procured from Beijing HFK Bioscience Co., Ltd (China). Following established protocols (He et al., 2019), general anesthesia was induced through intraperitoneal administration of 1% pentobarbital sodium (200 mg/kg), after which the dorsal fur was aseptically removed. Full-thickness excisional wounds were surgically created by removing circular skin segments (diameter: 1.2 cm) from the dorsal region of each animal (n = 6 per experimental group, determined using G*Power 3.1 software). Concurrently, the mice were randomized to be in the shNC or shKDM5A groups, and each group received an injection of 2 × 10⁶ ANA-1 cells/mL, the injection was performed subcutaneously around the wound area from day 0 using a multiple-point injection technique to ensure even distribution of the cells around the wound margin. Every day, the wound area was observed, and at days 0, 4, and 8, the rate of wound healing was computed. The wound healing score was calculated based on three histological criteria: (1) degree of re-epithelialization, (2) granulation tissue formation, and (3) collagen deposition. Each criterion was scored on a scale from 0 to 3, with higher scores indicating better healing. The mice were euthanized at predetermined intervals, and skin samples were taken for additional examination. Sections of the wound samples were frozen and fixed in paraffin for use in histology and molecular investigations later on. In the final verification experiment, another set of mice with surgically induced skin defects (n = 6) were established and randomly assigned to the following groups: shNC group, shKDM5A group, shKDM5A + shNC group, and shKDM5A + shSocs1 group. Similar to the previous experiment, all groups were injected with 2 × 10⁶ cells/mL. The injection was performed around the wound margin via a multiple-point subcutaneous injection method on wound day 0.

In this experiment, all general images were captured uniformly to ensure consistency and accuracy of the observations. Images were captured at three specific time points, namely, 0 days post-injury (dpi), 4 dpi, and 8 dpi, to closely monitor the progress of wound healing. When capturing images, a well-lit location was selected, and photos were taken from an angle perpendicular to the defective skin on the back of the mouse. In instances where new hair growth occurred on the intact skin of the mouse during the repair process, covering the damaged area, hair clippers were used to carefully trim the hair before capturing photos. This step ensures that the wound site is fully visible and not obscured by hair, enabling precise evaluation of wound healing. Subsequently, ImageJ was employed to measure the mouse skin wounds. Specifically, the area of dark-colored (red, red-black) parts in the images was uniformly counted as the unrepaired area.

Tissue sections were baked at 70 °C for 2 h to ensure proper fixation and adhesion to the slides. Perform dewaxing and hydration using a fully automatic staining machine, which typically takes approximately 45 min. Rinse the tissue sections with tap water for 15 min at a moderate flow rate to remove any residual wax or other impurities. Use HE staining kit (Beyotime, C0105S) to perform staining according to the instructions. Let the slides dry naturally and then mount them. Hematoxylin was used to stain the cell nuclei bright blue, the cartilage matrix and calcium salt particles dark blue, and the mucus gray-blue. Eosin was used to stain the cytoplasm in various tones of pink to peach. The eosinophilic granules within the cytoplasm exhibit vivid red coloration and potent reflecting qualities. Red blood cells were orange, and collagen fibers were a pale pink color. The color changes with the life cycle and pathological changes of the tissue or cell, as well as with the kind of tissue or cell.

Utilize the Masson trichrome staining kit (Beyotime, C0189S) according to the manufacturer’s instructions to stain the tissue sections. After staining, quickly dehydrate the tissue sections in a series of ethanol solutions (70%, 80%, 90%, and absolute ethanol) for 10 s each to remove excess water and enhance staining intensity. Clear the tissue sections in xylene three times for 2 min each. Once the slides are dry, seal them with neutral resin, ensuring to avoid generating air bubbles that may interfere with microscopy. Collagen fibers, mucus, and cartilage should appear blue (or green if stained with light green solution), while the cytoplasm, muscle, cellulose, and glial cells should appear red. The nucleus should appear black or blue. Finally, observe and photograph the stained tissue sections under an optical microscope (Nikon, Tokyo, Japan) to visualize the staining patterns and structures. Utilize ImageJ software for further analysis quantification of collagen deposition.

Each group’s wound tissue slices were prepared for immunohistochemistry examination. First, alcohol gradients were used to dewaxe and dehydrate the paraffin sections. To stop endogenous peroxidase activity, the slices were then submerged in 3% methanol H2O2 for 20 min. The portions were submerged in a water bath containing a repair solution to retrieve the antigen. After antigen retrieval, the sections were dried for 20 min at room temperature before being blocked with regular goat serum blocking solution (C-0005, Haoran Bio, Shanghai, China). After that, the sections were incubated with primary antibody anti-CD31 (1:1,000; ab182981; Abcam, Cambridge, United Kingdom) for a whole night at 4 °C. Following the incubation of the primary antibody, the sections were treated for 1 hour at room temperature with a horseradish peroxidase-conjugated polymer anti-rabbit secondary antibody (1:1,000; Dako, Glostrup, Denmark). The sections were then stained with hematoxylin (PT001, Bogoo, Shanghai) for 1 minute after being treated with diaminoamphetamine (DAB; ST033, Whiga Biotechnology, Guangzhou, China) for visualization. After that, the sections underwent regular infiltration and mounting, were dehydrated with graded alcohol, and were blued with 1% ammonia. Ultimately, an optical microscope and a Nikon Eclipse 90i digital camera were used to see the slices. Using ImageJ software (NIH), immunohistochemical staining was examined, and the number of anti-CD31-positive vessels was used to calculate capillary density. To make sure the results were reliable, three parts from each sample were examined.

Analogous to immunofluorescence staining, ANA-1 cells were grown on coverslips as well as wound tissue slices. The primary antibody was incubated for a whole night at 4 °C, and the secondary antibody was incubated for an hour at room temperature, as per the standard procedures for immunofluorescence staining. DAPI staining solution (P0131; Beyotime, Shanghai, China) was used to see the nuclei of the cells. Anti-iNOS (abcam, ab178945) and Anti-CD163 (abcam, ab316218) were the primary antibodies employed in this investigation. The secondary antibody used for detection was an anti-rabbit antibody (1:1,000; 8889S; CST) coupled with Alexa Fluor^® 594. A confocal fluorescence microscope (Leica, Wetzlar, Germany) and the Leica LAS X system were used to image and capture sections and coverslips. For analysis, three sections from each sample were randomly selected and evaluated. Additionally, fluorescence intensity was quantitatively assessed using ImageJ software.

ChIP detection was performed according to the method of Du using Magna ChIP A/G kit (Merck) (Du et al., 2020). ANA-1 cells were transfected with OE-NC, OE-KDM5A, si-NC, and si-KDM5A-1 for 48 h. The cells were fixed with 1% formaldehyde (Merck) at room temperature for 10 min, followed by the addition of glycine solution and cooling on ice for 5 min to stop the cross-linking reaction. After washing with PBS, the cell pellet was obtained and chromatin DNA was disrupted on ice. After centrifuged, the supernatant was mixed with ChIP buffer that contained protease inhibitors and a blocking solution was added. In order to decrease non-specific binding, the mixture was placed in a 4 °C incubation for 30 min. ChIP was performed by incubating the chromatin with 2 μg of KDM5A antibody (Abcam, China), 2 μg of H3K4me3 antibody (Abcam, ab8580), or 2 μg of H3K27ac antibody (Abcam, ab4729), along with 20 μL of Magna ChIP A/G magnetic beads, at 4 °C overnight. Following immunoprecipitation, the chromatin was eluted from the beads using ChIP elution buffer and RNase A mixture at 30 °C for 37 min, followed by proteinase K treatment at 62 °C for 2 h to reverse the cross-links. After purification of the DNA, the samples were subjected to qRT-PCR analysis to determine the enrichment of specific DNA sequences bound by the target proteins.

Total RNA was extracted from wound skin tissues and ANA-1 cells of mice in different groups using TRIzol reagent (Invitrogen, Carlsbad, CA, United States). Oligo dT primers and PrimeScript^® RTase (Takara, Dalian, China) were used to reverse transcribe the isolated total RNA. Takara Holdings Inc. (Kyoto, Japan) was responsible for designing and synthesizing the primers of KDM5A (Table 1). Quantitative RT-PCR (qRT-PCR) was performed using SYBR®Premix Ex TaqTMII (Takara, Dalian, China) on a C1000TM Thermal Cycler. GAPDH served as the internal reference, and relative expression levels were calculated using the 2^(-△△Ct) method.

Total protein was extracted from mouse wound skin tissues and ANA-1 cells across various experimental groups utilizing radioimmunoprecipitation lysis buffer. The concentration of the extracted proteins was quantified using a BCA protein concentration assay kit (Beyotime, P0012). Following extraction, the proteins were separated via electrophoresis and subsequently transferred onto polyvinylidene fluoride membranes. The membranes were then incubated overnight with primary antibodies against KDM5A (1:10,000, Abcam, ab194286) and Socs1 (1:1,000, Abcam, ab280886). The subsequent day, the membranes were reprobed with a horseradish peroxidase-conjugated goat anti-rabbit IgG secondary antibody (1:1,000, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, United States). Visualization of the membrane was performed, and the protein bands were quantified using the Bio-Rad ChemiDoc™ System in conjunction with ImageJ software. The relative protein expression was assessed by calculating the ratio of the gray value of the target band to that of Actin (1:10,000, rabbit, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, United States).

To prepare the 24-well plate for the tube formation assay, add 250 µL of basement membrane extract (BME, BD Biosciences, 354230) to each well, ensuring that the gel completely covers the bottom of the well. Incubate the plate for 30 min at 37 °C in an atmosphere containing 5% CO₂ to facilitate the solidification of the BME. Next, 300 µL of culture medium was mixed with 10 µL of resuspended fibroblasts (approximately 3 × 10⁴ cells). Pre-coated Matrigel plates were carefully removed from the incubator, and 100 μL of cell suspension (approximately 10,000 cells) was gently and slowly added onto the surface of the solidified Matrigel in a 96-well plate. The plate was then carefully returned to a 37 °C, 5% CO₂ incubator for culture, movement or shaking of the plate was avoided to prevent disruption of the forming tubular structures. HDFs were cultured in DMEM with 10% FBS and 1% penicillin–streptomycin. HUVECs (ATCC PCS-100-013) served as angiogenic cells. Growth factor-reduced Matrigel (Corning #354230) was added to 96-well plates (50 µL/well) and polymerized at 37 °C for 30 min. HUVECs (2 × 10⁴/well) were seeded on Matrigel either alone or co-cultured with HDFs (1 × 10⁴/well) by direct mixing (2:1) or by treatment with fibroblast-conditioned medium. Following the incubation period, employ a fluorescence microscope (Carl Zeiss, Oberkochen, Germany) to capture images of the cells in order to observe the formation of tubes. Subsequently, utilize ImageJ software to quantify the tube formation by assessing the average tube length along with other pertinent parameters.

Cells were cultured in 96-well plates at a density of 3 × 10^3 cells per well. After allowing sufficient time for the cells to adhere completely, the culture medium was meticulously removed, and 100 μL of a 10-fold diluted CCK-8 reagent (Cell Counting Kit-8, C0038, Beyotime) was introduced to each well. The plates were subsequently incubated at 37 °C for a duration of 2 h. Upon completion of the incubation, the optical density (OD) of each well was assessed at a wavelength of 450 nm utilizing a microplate reader (SpectraMAX® 190, Molecular Devices, Sunnyvale, CA, United States). The relative OD value was determined by subtracting the OD value of the blank control from that of the experimental group. The blank group consisted of cell culture medium and CCK-8 solution without the addition of cells, serving as a baseline control for background absorbance.

Following the cultivation of fibroblasts to the logarithmic growth phase, the cells were subjected to digestion using 0.25% trypsin for a duration of 3 min and subsequently resuspended in serum-free Opti-MEMI medium (Invitrogen, United States). The cell density was adjusted to a range of 1–10 × 10^5 cells/mL. The resulting cell suspension was then introduced into the upper chambers of 8 μm Transwell inserts (Corning, NY, United States) at a volume of 100 μL per well, utilizing a total of three wells. Concurrently, 600 μL of culture medium supplemented with 10% RPMI-1640 was added to the lower chamber of a 24-well plate to facilitate incubation. The plate was subsequently placed in an incubator for a period of 12 h to promote cell migration. After incubation, the culture medium was aspirated from the upper chambers, and the cells that had migrated to the underside of the membrane were fixed with 600 μL of 4% paraformaldehyde for 30 min. The fixed cells were then stained with 800 μL of a 0.2% crystal violet dye solution for 15 min. After the staining process, any excess dye solution was discarded, and the membranes were gently washed with phosphate-buffered saline (PBS) to eliminate non-specific staining. Finally, the number of stained cells was quantified by counting in five randomly selected fields using an inverted microscope (Leica DMi8-M, Co. Ltd, Solms, Germany), and the average value was calculated to assess the migratory capacity of the fibroblasts.

The means ± standard error of the mean (SEM) is used for statistical analysis of between-group mean comparisons, while the means ± standard deviation (SD), indicating the variability of the raw data, the relevant values have been annotated in the results and figure captions. All continuous data were tested for normality using the Shapiro-Wilk test. Given the normal distribution, we used unpaired two-tailed Student’s t-tests for comparisons between two groups and one-way ANOVA for multiple groups. Bonferroni correction was applied post-ANOVA to adjust for Type I errors when comparing >2 groups. All statistical analyses were performed utilizing GraphPad Prism 5.0 (GraphPad Software, La Jolla, CA, United States). A significance level of P < 0.05 was established for all tests.

Following induction of dorsal cutaneous wounds in mice, quantitative analysis revealed a significant upregulation of both KDM5A gene and protein expression levels at day 8 post-injury (P < 0.01; Figures 1A,B), suggesting its potential functional involvement in the wound repair process. In striking contrast, ANA-1 cells exhibited a progressive downregulation of KDM5A mRNA expression at the 4- and 8-day post-injury time points (P < 0.01; Figure 1C), demonstrating active transcriptional suppression of KDM5A during wound healing progression.

In this investigation, we employed a well-established murine dorsal skin wound model to systematically evaluate the functional consequences of KDM5A suppression on wound repair processes. Successful KDM5A knockdown was quantitatively verified through qRT-PCR and WB (Figures 1D,E). Comparative assessment of wound closure kinetics between shNC and shKDM5A groups revealed distinct temporal patterns. At the early repair phase (4 dpi), the shKDM5A cohort exhibited a modest but statistically significant reduction in wound closure relative to controls (20.64% ± 0.95% vs. 23.97% ± 1.24%, P < 0.05). However, by 8 dpi, KDM5A-deficient mice demonstrated markedly enhanced wound resolution, with residual wound areas measuring 4.57% ± 0.81% compared to 8.51% ± 0.55% in controls (P < 0.01; Figure 1F). Histomorphometric analysis of HE-stained sections confirmed substantial neodermal formation and adipocyte recruitment in shKDM5A specimens, accompanied by significantly improved wound healing scores (P < 0.01; Figure 1G).

The critical role of collagen matrix deposition in tissue regeneration was evidenced by Masson’s trichrome staining, which revealed significantly increased collagen content in shKDM5A wounds (15.13% ± 0.48% vs. control, P < 0.01; Figure 1H). Furthermore, immunohistochemical quantification demonstrated enhanced CD31⁺ microvascular density in KDM5A-deficient tissue (P < 0.01; Figure 1I), suggesting improved angiogenic potential. Immunofluorescence profiling identified a distinct macrophage polarization shift in shKDM5A wounds, characterized by reduced iNOS signal intensity concomitant with elevated CD163 expression (P < 0.01; Figure 1J), indicative of preferential M2 macrophage activation. Cytokine analysis revealed significant downregulation of pro-inflammatory mediators (TNF-α, IL-12) with concomitant upregulation of reparative factors (TGF-β1, VEGF) in the shKDM5A group (Figure 1K). These collective findings establish KDM5A as a critical epigenetic regulator of cutaneous wound repair through modulation of macrophage polarization and extracellular matrix remodeling.

Immunofluorescence analysis of ANA-1 cells demonstrated markedly attenuated CD163 signal intensity in both LPS-treated and LPS + shNC control groups relative to PBS controls. In contrast, LPS-stimulated cells with KDM5A knockdown exhibited significantly enhanced CD163 fluorescence (P < 0.01; Figure 2A), indicating that KDM5A suppression promotes M2 macrophage polarization-a finding consistent with the elevated CD163 expression observed in murine dermal wound tissues (Figure 1J). Quantitative cytokine profiling revealed that while LPS challenge robustly induced pro-inflammatory responses, shKDM5A transfection substantially mitigated these effects. Specifically, LPS exposure significantly increased TNF-α and IL-12 secretion while reducing TGF-β1 and VEGF production compared to PBS controls (P < 0.01). Notably, KDM5A-depleted cells showed significant reversal of these cytokine alterations when compared to LPS + shNC counterparts (P < 0.01; Figure 2B). These collective findings demonstrate that KDM5A silencing attenuates inflammatory cascades while fostering an anti-inflammatory microenvironment conducive to cellular repair processes.

Fibroblasts exhibit functional parallels with vascular endothelial cells in their capacity to support neovascularization. We successfully knocked down KDM5A in NHDFs using distinct shRNAs targeting KDM5A (Figure 2C). In vitro angiogenesis assays demonstrated that LPS treatment significantly impaired tubular network formation, whereas KDM5A knockdown in NHDF cells markedly augmented their angiogenic potential, as reflected by increased vascular length (P < 0.01; Figure 2D). Subsequent CCK-8 proliferation analysis revealed that while LPS exposure substantially attenuated NHDF cell growth, shKDM5A transfection significantly enhanced cellular proliferative capacity (P < 0.01; Figure 2E). Transwell invasion assays further established that KDM5A suppression promoted NHDF cell invasiveness (P < 0.01; Figure 2F). Consistent with these observations, temporal migration analysis demonstrated progressive enhancement of NHDF cell motility at both 12 h and 24 h post-KDM5A silencing (P < 0.01; Figure 2G). These data collectively indicate that KDM5A ablation potentiates fibroblast-mediated vascularization while concomitantly augmenting their proliferative, migratory and invasive properties.

Initial validation of KDM5A modulation efficacy was performed through comprehensive assessment of mRNA and protein expression profiles in siKDM5A-1, siKDM5A-2, and OE-KDM5A groups using WB and quantitative PCR analyses. Notably, KDM5A knockdown consistently upregulated Socs1 protein expression, establishing its regulatory role in Socs1 expression control (Figures 3A–C). Chromatin immunoprecipitation assays subsequently revealed KDM5A’s direct binding to the Socs1 promoter region, with KDM5A overexpression significantly reducing H3K4me3 and H3K27ac histone mark deposition, while KDM5A depletion conversely enhanced these activating modifications (Figure 3D), demonstrating KDM5A’s function as a negative regulator of Socs1-associated histone modifications.

Pharmacological inhibition experiments further corroborated these findings. Compared to OE-NC controls, KDM5A-overexpressing ANA-1 cells exhibited diminished H3K4me3 and H3K27ac enrichment at the Socs1 promoter concomitant with increased KDM5A occupancy (P < 0.01; Figure 3E). Importantly, KDM5A inhibitor treatment effectively rescued the histone modification profile, restoring H3K4me3 and H3K27ac levels (P < 0.01). These data collectively establish that KDM5A-mediated recruitment to the Socs1 promoter orchestrates epigenetic silencing through erasure of activating histone marks, thereby suppressing Socs1 transcriptional activation in ANA-1 cells.

We found that the regulatory effect of KDM5A is mainly achieved by affecting the expression of Socs1. Therefore, based on stably expressing shKDM5A, the Socs1 gene was silenced to verify the regulatory effect of KDM5A on Socs1 in macrophages and fibroblasts. Compared to the LPS + shKDM5A + shNC group, the LPS + shKDM5A + shSocs1 group led to a significant decrease in CD163 fluorescence in ANA-1 cells (P < 0.01), promoting polarization of macrophages towards the M1 phenotype (Figure 4A). This led to elevated concentrations of the inflammatory cytokines TNF-α and IL-12, while concurrently resulting in reduced levels of the growth factors TGF-β1 and VEGF (Figure 4B). This further aggravates the inflammatory response. Concurrently, the inhibition of KDM5A and Socs1 led to a marked decrease in the capabilities of NHDF cells regarding blood vessel formation, as well as cell proliferation, migration, and invasion (P < 0.01) (Figures 4C–F). These findings indicate that the KDM5A-Socs1 pathway is crucial in modulating immune responses and angiogenesis.

It was verified in the back skin injury model of mice stably expressing shKDM5A and shKDM5A + shSocs1. Statistics on the wound repair area of mice at 4 dpi and 8 dpi showed that both the shKDM5A group and the shKDM5A + shNC group were lower than the shNC group (P < 0.01). Compared with the shKDM5A + shNC group, the repair rate of the shKDM5A + shSocs1 group increased (P < 0.01; Figure 5A). It was also found that the wound healing score, collagen deposition area, and the proportion of CD31 in skin tissue were significantly lower in the shKDM5A + shSocs1 group than in the shKDM5A + shNC group (P < 0.01; Figures 5C–E). This suggests that KDM5A may have a specific function in the regulation of Socs1. When KDM5A is silenced, wound repair is accelerated, while silencing Socs1 reverses this accelerated effect and affects the wound repair process.

Immunofluorescence analysis of wound tissues revealed that combined KDM5A/Socs1 silencing significantly reduced CD163 signal intensity (P < 0.01; Figure 5F), indicative of M1-polarized macrophage predominance. This phenotypic shift correlated with elevated pro-inflammatory mediators (TNF-α, IL-12) and iNOS expression (P < 0.01; Figures 5B,F), while concurrently suppressing reparative factors (TGF-β1, VEGF; P < 0.01; Figure 5B). Although M1 macrophages facilitate early wound debridement, their sustained activation promotes tissue damage and chronic inflammation. These findings collectively demonstrate that KDM5A-mediated wound healing acceleration is mechanistically dependent on Socs1 expression, with Socs1 ablation reinstating inflammatory dominance and impairing tissue regeneration.