SETD2 epidermal deficiency promotes cutaneous wound healing via activation of AKT/mTOR Signalling

Abstract Objectives Cutaneous wound healing is one of the major medical problems worldwide. Epigenetic modifiers have been identified as important players in skin development, homeostasis and wound repair. SET domain–containing 2 (SETD2) is the only known histone H3K36 tri‐methylase; however, its role in skin wound healing remains unclear. Materials and Methods To elucidate the biological role of SETD2 in wound healing, conditional gene targeting was used to generate epidermis‐specific Setd2‐deficient mice. Wound‐healing experiments were performed on the backs of mice, and injured skin tissues were collected and analysed by haematoxylin and eosin (H&E) and immunohistochemical staining. In vitro, CCK8 and scratch wound‐healing assays were performed on Setd2‐knockdown and Setd2‐overexpression human immortalized keratinocyte cell line (HaCaT). In addition, RNA‐seq and H3K36me3 ChIP‐seq analyses were performed to identify the dysregulated genes modulated by SETD2. Finally, the results were validated in functional rescue experiments using AKT and mTOR inhibitors (MK2206 and rapamycin). Results Epidermis‐specific Setd2‐deficient mice were successfully established, and SETD2 deficiency resulted in accelerated re‐epithelialization during cutaneous wound healing by promoting keratinocyte proliferation and migration. Furthermore, the loss of SETD2 enhanced the scratch closure and proliferation of keratinocytes in vitro. Mechanistically, the deletion of Setd2 resulted in the activation of AKT/mTOR signalling pathway, while the pharmacological inhibition of AKT and mTOR with MK2206 and rapamycin, respectively, delayed wound closure. Conclusions Our results showed that SETD2 loss promoted cutaneous wound healing via the activation of AKT/mTOR signalling.


| INTRODUC TI ON
Mammalian skin is composed of the inner and outer epidermis, separated by a basement membrane. 1 The skin is a natural physical and immune protective barrier that prevents not only the penetration of harmful microorganisms, but also dehydration. 2,3 Once the skin barrier is damaged, wound-healing process begins immediately.
Cutaneous wound healing is a complex and dynamic process that involves three phases: inflammation, re-epithelialization and tissue remodelling. 4,5 Even though these three phases occur sequentially, they overlap, and the extracellular matrix (ECM), soluble growth factors and multiple cell types, such as immune cells, fibroblasts and keratinocytes (KCs), participate in these phases. 6 Re-epithelialization is the formation of new epithelium and covering of the wound surface, which requires the migration and proliferation of KCs. 7 Epigenetic regulation controls the transcriptional activation or silencing of genes without changing the DNA sequence, governing the phenotypic plasticity of individual cells, organs or a whole organism. 8 In the skin, epigenetic regulation mechanisms play an important role in its development, homeostasis and wound repair. 9 Histone modification is a type of epigenetic regulation that affects the transcriptional activity of genes and is involved in normal physiological processes and diseases. 10,11 Several studies have reported that histone modification and related enzymes play an essential role in skin wound healing. For example, H3K27me3 demethylase JMJD3 interacts with NF-κB, resulting in increased expression of inflammatory, matrix metalloproteinase and growth factor genes, while the inactivation of JMJD3 leads to delayed wound healing. 1,12 The hair follicle cells of mice lacking EZH1 and EZH2 histone H3K27 tri-methylases have defective cell proliferation and wound healing, even though epidermis continues to hyperproliferate. 13 Furthermore, the hypomethylation of histone H3 K4/9/27me3 is beneficial for the differentiation and growth of hair follicles (HFs) and promotes wound healing. 14 SETD2 (SET domain-containing protein 2) was first identified as the protein associated with Huntington's disease (HD). 15 Currently, SETD2 is the only known H3K36 tri-methylase; it interacts with RNA polymerase II to mediate transcriptional extension, resulting in changes in gene transcription levels. 16,17 As a tumour suppressor, SETD2 plays an important role in gene transcription regulation, DNA damage repair and alternative splicing. [18][19][20][21][22] Recently, SETD2 has been extensively studied in various biological processes and diseases. Loss of SETD2 function has been investigated in several human tumours, including GI stromal tumours, renal cell carcinoma, pancreatic ductal adenocarcinoma (PDAC), prostate cancer, breast cancer, leukaemia and high-grade gliomas. [23][24][25][26][27][28][29] Moreover, the role of SETD2 in hematopoietic stem cell self-renewal, sperm development, bone marrow mesenchymal stem cell differentiation, V(D)J recombination, maternal epigenome and embryonic development has been examined. [30][31][32][33][34] In addition, four non-histone substrates of SETD2: α-tubulin, STAT1, EZH2 and actin were discovered. 23,[35][36][37] However, the role of SETD2, an important histone-modifying enzyme, in skin wound healing is still not understood.
In this study, to investigate the role of SETD2 in skin wound healing, we generated epidermis-specific Setd2-knockout mice and showed that SETD2 deficiency promoted cutaneous wound healing through the activation of AKT/mTOR pathway.

| Mice
Setd2 fl/fl mice were generated by Shanghai Biomodel Organism Co. using conventional homologous recombination in embryonic stem (ES) cells as previously described. 19,29 Tg-CK5CreERT2; R26R-CAG-lsl-Tomato mice were purchased from the Jackson Laboratory. Setd2 -KO mice were generated by crossing Setd2-floxed mice with K5CreERT2 mice, and tamoxifen was intraperitoneally injected at 100 mg kg −1 body weight. All mice were maintained in a specific-pathogen-free (SPF) facility, and all experimental proce-

| In vivo wound-healing experiments
Mice (8 to 10-week-old littermates) were anaesthetized by intraperitoneal injection of tribromoethanol, and their backs were shaved.
Four 4-mm full-thickness cutaneous biopsy punch wounds were made on the back of each mouse. The entry wounds were photographed on days 0, 1, 3, 5, 7 and 9. The wound areas were measured using ImageJ software. Mice were euthanized 1, 3 and 5 days after wounding by carbon dioxide inhalation, the wounds were excised, fixed overnight at 4°C with 4% paraformaldehyde and then embedded in paraffin.

| Isolation of primary mouse keratinocytes
Primary mouse keratinocytes were isolated from the skin as previously described. 38,39 Briefly, the skin of 10-week-old mice was treated with dispase II (Gibco,17105-041) overnight at 4°C. The epidermis was separated and digested for 10 min at 37°C with 0.25% trypsin/EDTA (Gibco, 25200-072) and strained through a 70μm filter. The supernatants were centrifuged, and cells were collected.

| Histology, H&E staining and immunohistochemistry (IHC)
Tissues were fixed in 4% paraformaldehyde overnight at 4°C, dehydrated and embedded in paraffin. Sections (5 μm) were cut and stained with haematoxylin and eosin (H&E). For IHC staining, sections were deparaffinized, rehydrated, subjected to antigen retrieval in citrate buffer, and endogenous peroxidases were quenched with 3% H 2 O 2 . Blocking was performed with 5% BSA for 1 hour at room temperature. Next, the samples were incubated with primary antibodies for 12-16 hours at 4°C. The primary antibodies used were anti-Ki67 (Abcam, ab15580) and anti-K5 (Abcam, ab52635). After three washes in PBS, the sections were incubated with an HRPconjugated secondary antibody for 1 hour at room temperature and then counterstained with haematoxylin. Images were acquired using a Leica microscope, and staining intensities were calculated using ImageJ software. The antibodies used for staining are listed in Table S2.

| Immunofluorescence
Skin samples were fixed in 4% paraformaldehyde for 30 minutes at 4°C, transferred to 30% sucrose overnight and then embedded in OCT. Sections (7 μm) were cut, permeabilized with Triton X-100 and blocked with 5% BSA. Next, the sections were incubated with primary antibodies (anti-SETD2 (LS-C332416), anti-p-AKT (CST, #4060) and anti-p-mTOR (CST, #5536)) at 4°C for 12-16 hours, followed by the incubation with the secondary antibodies at 37°C for 1 hour. Nuclei were counterstained with DAPI. All antibodies used for immunofluorescence are listed in Table S2.

| Scratch wound-healing assay
HaCaT cells were plated at 1 × 10 5 cells/well in triplicate in 6-well plates. Scratch assays were performed using completely confluent HaCaT cells. Scratches were generated using a 200 μL plastic pipette tip. Suspended cells were washed off, and the remaining cells were cultured in the medium without FBS to inhibit cell proliferation.
Images were acquired immediately after scratches were generated, as well as 12 hours and 24 hours after scratching. The plates were washed with PBS before imaging to reduce the number of dead cells in the field of view.

| RNA isolation and quantitative RT-PCR
Total RNA was extracted from cultured cells or tissues using an RNA extraction kit (BioTeke) following the manufacturer's protocol.
RNA was reverse transcribed using an RT reagent kit (Takara). The cDNA was subsequently subjected to TB Green-based real-time PCR analysis. GAPDH was used to normalize the results, and the data were presented as the mean ± SD. The P-value was calculated using Student's t test. The primers used for the qPCR analysis are listed in Table S1.

| Western blot analysis and antibodies
Cell and tissue samples were lysed in RIPA buffer (Beyotime, P0013B) supplemented with protease and phosphatase inhibitors (MCE). Protein concentrations were measured using the BCA Protein Assay (Thermo Fisher Scientific). Proteins were separated using 6% and 10% SDS-PAGE gels and then transferred to polyvinylidene fluoride (PVDF) membranes or nitrocellulose membranes (Millipore). The membranes were blocked with 5% skim milk in TBST for 1.5 hours at room temperature and subsequently incubated with primary antibodies overnight at 4°C, followed by  Table S2.
Rapamycin or MK2206 treatments were initiated 7 days before the wounding experiment.

| RNA-seq and analyses
Skin tissue mRNA was obtained from 12-week-old Setd2 -KO and Setd2 fl/fl mice. Differential gene expression was analysed using DESeq2 package. The list of significance was determined by setting a false discovery rate (FDR) threshold at a <0.05, and |log2FC| > 0.585.
All differentially expressed genes were subsequently analysed using GO and pathway analyses.

| ChIP-Seq and analyses
Mouse primary keratinocytes (>1 × 10 7 ) were crosslinked with 1% formaldehyde for 5 minutes at room temperature and quenched with 0.125 mol/L glycine. The fragmented chromatin fragments were pre-cleared and then immunoprecipitated with protein A + G magnetic beads coupled to the anti-H3K36me3 (ab9050) antibody.
After reverse crosslinking, ChIP and input DNA fragments were end-repaired and A-tailed using the NEBNext End Repair/dA-Tailing Module (E7442, NEB) followed by adaptor ligation with the NEBNext Ultra Ligation Module (E7445, NEB). The DNA libraries were amplified for 15 cycles and sequenced using Illumina NextSeq 500 with single-end 1 × 75 as the sequencing mode. Raw reads were filtered to obtain high-quality clean reads by removing sequencing adapters, short reads (length <50 bp) and low-quality reads using Cutadapt (v1.9.1) and Trimmomatic23 (v0.35). Next, FastQC was used to ensure high reads quality. The clean reads were mapped to the mouse genome (assembly GRCm38) using the Bowtie2 (v2.2.6) software.
Peak detection was performed using the MACS (v2.1.1) peak finding algorithm with .01 set as P-value cut-off. Annotation of peak sites to gene features was performed using the ChIPseeker R package.

| Statistical analysis
All experiments were repeated at least three times. Unless otherwise indicated, data were presented as the mean ± SD and analysed for statistical significance by two-way ANOVA or Student's t test using GraphPad Prism 8.0.2 software. Statistical significance was set at P < .05. *P < .05, **P < .01, ***P < .001 and ****P < .0001.

| Generation of epidermis-specific Setd2deficient mice
To determine whether SETD2 plays a role in re-epithelialization after injury, we analysed microarray data obtained from the NCBI GEO Datasets GSE30355. 44 The results showed a decrease in Setd2 in injured KCs compared to normal KCs ( Figure 1A). RT-qPCR analysis also showed decreased expression of Setd2 in the full-thickness wound tissues of wild-type (WT) mice after injury ( Figure 1B). These results suggested that SETD2 played a role in skin wound repair. To further elucidate the role of SETD2 in skin wound healing, we crossed Setd2flox (Setd2 fl/fl ) mice with K5CreERT2 mice to obtain the epidermisspecific Setd2 knockout (Setd2 -KO ) mice ( Figure 1C). 40,44 In this system, Cre expression is driven by the K5 promoter, which directs gene expression in the basal cells of the skin, cells believed to be the stem cells of the skin. The immunofluorescence results showed that keratin 5 was mainly expressed in the epidermis and hair follicles; SETD2-positive cells were also observed in the epidermis and hair follicles ( Figure 1D). Furthermore, there was colocalization between SETD2 and keratin 5. Therefore, in our model system, Setd2 was de-

| Setd2 deficiency promotes cutaneous wound healing and thickens the wound epithelium in mice
Next, we investigated the effect of SETD2 deletion on cutaneous wound healing. Skin punch biopsies were obtained from Setd2 -KO and age-matched Setd2 fl/fl (control) mice to assess the wound-healing process every 2 days for up to 9 days. Three days after wounding, the wound areas in Setd2 -KO mice were significantly reduced compared to Setd2 fl/fl mice as measured by image analysis, suggesting accelerated wound healing in Setd2 -KO mice ( Figure 2A). Furthermore, the differences in the wound area between Setd2 -KO and Setd2 fl/ fl mice were statistically significant on days 3 and 5 after wounding ( Figure 2B). Increased thickness of the wound epithelium (WE) contributed to the re-epithelialization process. The H&E staining of injured tissues showed that the thickness of WE was significantly increased in Setd2 -KO mice compared to control mice ( Figure 2C,D).
These results indicate that the absence of Setd2 accelerates skin wound healing in mice.

| Accelerated wound closure in Setd2 -KO mice is associated with enhanced keratinocytes proliferation and migration
Since KC proliferation and migration are involved in the reepithelialization process, the process essential for wound healing, [45][46][47] and SETD2 is primarily expressed in skin KCs, we further hypothesized that accelerated wound healing is caused by the increased proliferation and migration of these cells. To test this hypothesis, we evaluated the differences in the migration rate of KCs from Setd2 -KO and Setd2 fl/fl mice by K5 immunostaining. KCs migrated from the edge to the centre of the wound in both Setd2 fl/ fl and Setd2 -KO mice. However, re-epithelialization occurred significantly faster in Setd2 -KO mice compared to control mice ( Figure 3A). Figure 3B, the distance between the migration edge in Setd2 -KO KCs was significantly shorter than in controls. Next, to determine whether accelerated wound closure was due to increased cell proliferation, we used Ki67 (a marker of proliferation) immunohistochemical staining and counted the number of Ki67-positive KCs in wound areas on different days. The results demonstrated that cell proliferation was significantly enhanced on days 3 and 5 after injury ( Figure 3C,D). These results confirmed that Setd2 deletion accelerated the proliferation and migration of KCs in vivo.

| Loss of Setd2 promotes scratch closure and proliferation of KCs in vitro
To further analyse the effects of SETD2 loss on KC proliferation and migration in vitro, we performed CCK8 and scratch wound-

| Setd2-deficient skin displays hyperactive AKT/ mTOR Signalling
To understand how SETD2 loss enhances wound healing, we carried  Figure 5D). The mRNA expression levels of these genes were validated using RT-qPCR ( Figure 5E). Furthermore, IF staining showed that the protein expression levels of p-AKT and p-mTOR were enhanced in WE areas on days 1 and 3 after injury in Setd2 -KO mice ( Figure 5F). These data suggested that SETD2 deficiency resulted in the activation of AKT/mTOR signalling.
Since SETD2 mainly regulates the expression of downstream genes through H3K36me3, we performed chromatin immunoprecipitation experiments followed by next-generation sequencing (ChIPseq) assays using a H3K36me3-specific antibody in primary KCs H, Quantification of migration area; n = 6/group. Data are presented as the mean ± SD; statistical significance was determined using a twotailed Student's t test and two-way ANOVA; n.s, no significance, ***P < .001 and ****P < .0001. Thbs3 by ChIP-qPCR assay ( Figure 5J), and found that the intensity of H3K36me3 binding within these gene loci significantly increased in the H3K36me3 group compared to the IgG group. These results suggested that SETD2 deletion might activate the AKT/mTOR signalling pathway via the regulation of these H3K36me3-directly occupied genes.

| Activation of AKT/mTOR in the epithelial compartment accelerates wound healing
To further elucidate the causal link between SETD2 and the AKT/ mTOR pathway in wound healing, we transfected HaCaT cells with sh-Setd2 plasmids and then treated these cells with an AKT inhibitor

| D ISCUSS I ON
This study investigated the mechanistic role of SETD2 in epidermal KCs during cutaneous wound healing. We identified two novel find- Results are presented as percent wound area at day 3 relative to the original wound area in Setd2 -KO mice treated with the vehicle control or inhibitors (rapamycin or MK2206); n = 4 wounds/group. The analysis was performed using ImageJ software. G, H&E staining of wound tissues of Setd2 -KO mice treated either with the vehicle control or inhibitors (rapamycin and MK2206) on day 3 after injury. The arrows indicate the boundary of the wound. Scale bars: 500 μm. Data are presented as the mean ± SD; statistical significance was determined using a two-tailed Student's t test and two-way ANOVA; **P < .01 and ****P < .0001 showed, for the first time, that SETD2 plays an essential role in the re-epithelialization process during cutaneous wound healing. RNA-seq and ChIP-seq data indicated that seven misregulated genes associated with the PI3K/AKT/mTOR pathway showed direct H3K36me3 occupancies. We verified the mRNA expression levels of these genes including Fn1, Gys1, Itga7, Pik3r3, Gng4, Lama4 and Thbs3 by RT-qPCR. ChIP-qPCR validated the existence of H3K36me3 binding at these genes. These results suggested that SETD2 might regulate these downstream genes through H3K36me3, leading to the activation of the AKT/mTOR pathway, and the specific regulatory mechanism needs to be further elucidated.
As a tumour suppressor, SETD2 has been extensively studied in cancers of different tissues. Here, we utilized an animal model of a conditional epidermis-specific knockout of Setd2. However, Setd2 -KO mice showed no tumour-associated phenotype in their skin for the first 10 months. Therefore, the biological function of SETD2 in skin cancer could be further studied either by crossing Setd2 -KO mice with other mutant mice, or by stimulating Setd2 -KO mice with different environmental factors. In summary, our study contributes to a better understanding of the wound-healing process and suggests that SETD2 may be considered as a therapeutic target to improve skin wound healing.

ACK N OWLED G EM ENTS
This study was supported by funds from National Natural Science

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.