SOX2-dependent wound repair signature triggers pro-healing outcome in hyperglycemic wounds

Wound healing is underpinned by the transcriptional regulation of signaling pathways to coordinate proper tissue repair (Eming et al., 2014Eming S.A. Martin P. Tomic-Canic M. Wound repair and regeneration: Mechanisms, signaling, and translation.Science Translational Medicine. 2014; 6265sr6-sr6Crossref PubMed Scopus (2094) Google Scholar, Overmiller et al., 2022Overmiller A.M. Sawaya A.P. Hope E.D. Morasso M.I. Intrinsic Networks Regulating Tissue Repair: Comparative Studies of Oral and Skin Wound Healing.Cold Spring Harb Perspect Biol. 2022; 14PubMed Google Scholar). In a hyperglycemic state, the repair process becomes deregulated and may result in diabetic foot ulcers (DFUs) (Eming et al., 2014Eming S.A. Martin P. Tomic-Canic M. Wound repair and regeneration: Mechanisms, signaling, and translation.Science Translational Medicine. 2014; 6265sr6-sr6Crossref PubMed Scopus (2094) Google Scholar). We previously showed that induction of the oral pro-wound healing transcription factor SOX2 in skin can reprogram cutaneous keratinocytes and prime wounds for accelerated repair through upregulation of epidermal growth factor receptor (EGFR) ligands (Iglesias-Bartolome et al., 2018Iglesias-Bartolome R. Uchiyama A. Molinolo A.A. Abusleme L. Brooks S.R. Callejas-Valera J.L. et al.Transcriptional signature primes human oral mucosa for rapid wound healing.Science Translational Medicine. 2018; 10eaap8798Crossref PubMed Scopus (159) Google Scholar, Uchiyama et al., 2019Uchiyama A. Nayak S. Graf R. Cross M. Hasneen K. Gutkind J.S. et al.SOX2 Epidermal Overexpression Promotes Cutaneous Wound Healing via Activation of EGFR/MEK/ERK Signaling Mediated by EGFR Ligands.Journal of Investigative Dermatology. 2019; 1391809-20.e8Abstract Full Text Full Text PDF Scopus (34) Google Scholar). EGFR signaling plays a crucial role in skin integrity and normal wound healing, and its impairment has been implicated in DFUs (Berlanga-Acosta et al., 2017Berlanga-Acosta J. Fernández-Montequín J. Valdés-Pérez C. Savigne-Gutiérrez W. Mendoza-Marí Y. García-Ojalvo A. et al.Diabetic Foot Ulcers and Epidermal Growth Factor: Revisiting the Local Delivery Route for a Successful Outcome.BioMed Research International. 2017; 20172923759Crossref PubMed Scopus (45) Google Scholar, Bodnar, 2012Bodnar R.J. Epidermal Growth Factor and Epidermal Growth Factor Receptor: The Yin and Yang in the Treatment of Cutaneous Wounds and Cancer.Advances in Wound Care. 2012; 2: 24-29Crossref Google Scholar). Therefore, we decided to investigate the effect of SOX2 upregulation on wound healing with hyperglycemia. Combining tamoxifen (Tam)-inducible K14CreERTM/LSL-SOX2-IRES-GFP (epidermal SOX2-overexpressing) and streptozotocin (STZ)-treated (hyperglycemic) mouse models (Figure 1a), we determined that the mRNA levels of SOX2 were significantly increased in Tam-treated mice compared with vehicle (Veh)-treated mice in both the control and STZ-treated groups, with no significant difference detected between the Tam- and Tam STZ-treated groups or between the Veh- and Veh STZ-treated groups by a Kruskal-Wallis test with Dunn's multiple comparisons test (Figure 1b). All animal studies were carried out according to the protocol approved by the Institutional Animal Care and Use Committee at the National Institute of Arthritis and Musculoskeletal and Skin Diseases (protocol A022-01-01). Protein expression of SOX2 was detected in the epidermis via immunofluorescence (IF) staining in Tam- and Tam STZ-treated mice (Figure 1c). To assess the wound healing process, skin punch biopsies were obtained from age-matched SOX2-overexpressing (Tam and Tam STZ) and control mice (Veh and Veh STZ) mice, and external wound measurements were performed every 2 days for 8 days (Figure 1d). Tam- and Tam STZ-treated mice exhibited significantly greater wound closure at the 2- and 4-day time points compared to the Veh- and Veh STZ-treated groups, respectively, and Tam STZ-treated mice also exhibited significantly greater wound closure than Veh STZ-treated mice at the 8-day time point according to two-way ANOVA with Tukey's multiple comparisons test (Figure 1e). Additionally, a scratch assay performed on primary human epidermal keratinocytes transduced with either GFP (adGFP) or GFP and SOX2 (adGFP/adSOX2) and cultured in media with either no added glucose or with 25 mM added glucose found that adGFP/adSOX2-transduced cells had a significantly higher percentage of migration at 24 hours post-scratch than adGFP-transduced cells in either normal or 25 mM glucose media according to ordinary one-way ANOVA with Bonferroni's multiple comparisons test (Supplementary Figure S1a). IF stainings showed increased protein expression of keratins 6A (K6A) and 17 (K17) in the epidermis of Tam- and Tam STZ-treated mice compared to Veh- and Veh STZ-treated mice, with a particular increase in the Tam STZ-treated group relative to the Veh STZ-treated group (Figure 1f). Additionally, IF stainings revealed the presence of Cd31+ cells and of F4/80+ cells in all experimental groups (Figure 1f). Quantification of Cd31+ cells in the papillary dermis (n = 2) showed a 2-fold increase in their number in Tam- and Tam STZ-treated mice relative to Veh- and Veh STZ-treated controls, respectively (Figure 1g). Quantification of F4/80+ (macrophage marker) cells (n = 2) showed an average of 12.6 ± 17.1 F4/80+ cells in the Veh group, 23.7 ± 9.2 in the Tam group, 3.4 ± 0.3 in the Veh STZ group, and 39.9 ± 4.4 in the Tam STZ group (Figure 1g). Hematoxylin and eosin (H&E) staining of skin sections showed changes in skin morphology between Tam- and Veh-treated mice and between Tam STZ- and Veh STZ-treated mice, including significantly increased epidermal thickness in Tam- and Tam STZ-treated mice relative to Veh- and Veh STZ-treated mice, respectively, with no significant difference between Tam- and Tam STZ-treated mice or between Veh- and Veh STZ-treated mice according to one-way ANOVA with Bonferroni's multiple comparisons test (Supplementary Figure S1b). To investigate transcriptomic differences between SOX2-expressing and non-expressing skin in control and hyperglycemic states, we performed single-cell RNA sequencing analysis on skin biopsies from mice treated with Veh, Tam, Veh STZ, or Tam STZ. A total of 159,365 cells (54,587 from Veh, 29,351 from Tam, 38,406 from Veh STZ, and 37,021 from Tam STZ) that met quality control metrics were grouped into 29 clusters by unsupervised clustering, aggregated into 6 high-level groups according to their expression profiles (Supplementary Figure S2a), and visualized by Uniform Manifold Approximation and Projection (UMAP) (Figure 2a). Proportions of each cell type were tabulated (Figure 2a). Samples were pseudobulked upstream of differential expression analysis using DESeq2; 146 genes were found to be differentially expressed as defined by an absolute fold-change ≥ 1.5 and adjusted p-value < 0.05 between Tam and Veh-treated mice (75 upregulated in Tam, 71 downregulated), and 412 genes were differentially expressed between Tam STZ- and Veh STZ-treated mice (350 upregulated in Tam STZ, 62 downregulated) (Figure 2b, Supplementary Table S1, and Supplementary Table S2). Of these, 317 genes were uniquely upregulated and 56 genes were uniquely downregulated in the Tam STZ-Veh STZ comparison (Figure 2b). Qiagen Ingenuity Pathway Analysis (IPA) performed on these 373 genes predicted an increase in several immune system processes in Tam STZ-treated skin relative to Veh STZ-treated skin, including synthesis of reactive oxygen species, T cell migration, phagocyte migration, granulocyte migration, and cellular infiltration by leukocytes (Figure 2c). Analysis of ligand-receptor interactions using MultiNicheNet of the predominant signaling pathways originating from keratinocytes in each experimental group and comparison of changes between Tam STZ and Veh STZ and between Tam and Veh revealed an increased prevalence of signaling from keratinocytes to immune cells and vascular cells in Tam STZ-treated mice compared to Veh STZ-treated mice than occurred in Tam-treated mice compared to Veh-treated mice, including increased signal sending via S100a8/S100a9 by keratinocytes and increased signal reception via Ccr2, Ccr5, Csf3r, and Itga9 by immune cells (Figure 2d and Supplementary Table S3). The top 25 interactions between keratinocytes and other cell types on a sample-wise basis within each experimental group are shown (Figure 2d). Comparable changes were less pronounced between Tam- and Veh-treated mice using the same method (Supplementary Figure S2b). S100a8 and S100a9 are associated with keratinocyte activation in wounded murine and human skin and chemoattraction of inflammatory myeloid cells (Thorey et al., 2001Thorey I.S. Roth J. Regenbogen J. Halle J.-P. Bittner M. Vogl T. et al.The Ca2+-binding Proteins S100A8 and S100A9 Are Encoded by Novel Injury-regulated Genes.Journal of Biological Chemistry. 2001; 276: 35818-35825Abstract Full Text Full Text PDF PubMed Scopus (206) Google Scholar). Ccr2, Ccr5, and Csf3r signaling are associated with neutrophil signaling and extravasation and Itga9 is associated with neutrophil development (Park et al., 2022Park S.D. Saunders A.S. Reidy M.A. Bender D.E. Clifton S. Morris K.T. A review of granulocyte colony-stimulating factor receptor signaling and regulation with implications for cancer.Front Oncol. 2022; 12932608Crossref Scopus (6) Google Scholar). To investigate these predicted correlations, we performed immunohistochemistry (IHC) staining against Ly6g to detect neutrophil, monocytes and granulocytes. An increased positive staining for Ly6g was determined in Tam-treated mice compared to Veh-treated mice, which was further increased in Tam STZ-treated mice compared to Veh STZ-treated mice (Figure 2e). Quantification of Ly6g+ cells revealed that these increases were statistically significant in Tam STZ-treated mice compared to Veh STZ-treated mice and compared to Tam-treated mice (Supplementary Figure S2c). Herein, we demonstrate that combinatorial upregulation of SOX2 and hyperglycemia in cutaneous keratinocytes triggers a pro-healing signature and increased neutrophil chemoattraction, putatively by Ccr2, Ccr5, and Csf3r signaling. Neutrophil depletion and deregulation through the TREM1/FOXM1 pathway are directly implicated in DFU pathophysiology (Sawaya et al., 2020Sawaya A.P. Stone R.C. Brooks S.R. Pastar I. Jozic I. Hasneen K. et al.Deregulated immune cell recruitment orchestrated by FOXM1 impairs human diabetic wound healing.Nature Communications. 2020; 11: 4678Crossref PubMed Scopus (161) Google Scholar, Sawaya et al., 2022Sawaya A.P. Stone R.C. Mehdizadeh S. Pastar I. Worrell S. Balukoff N.C. et al.FOXM1 network in association with TREM1 suppression regulates NET formation in diabetic foot ulcers.EMBO reports. 2022; 23e54558Crossref PubMed Scopus (19) Google Scholar). Our results highlight SOX2-mediated gene expression as a potential means of addressing the neutrophil-deficient DFU phenotype and keratinocyte non-migratory phenotype, highlighting genes regulated by SOX2 as potential therapeutic targets for DFU treatment. Datasets related to this article can be found in the Gene Expression Omnibus database under access code GSE266638 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE266638). The authors state no conflict of interest. Conceptualization: CGO, APS, MIM; Methodology: CGO, APS, MIM; Software: CGO, SRB, SN, AMO; Investigation: CGO, APS, SM, KH; Validation: CGO, APS, SM, MIM; Data Curation: CGO, APS, SN, AMO, MIM; Formal Analysis: CGO, APS, MIM; Writing - Original Draft Preparation: CGO, MIM; Writing - Review and Editing: CGO, MIM; Visualization: CGO, AMO, MIM; Supervision: MIM; Project Administration: MIM; Funding Acquisition: MIM, Intramural Research Program of the National Institute of Arthritis and Musculoskeletal and Skin Diseases. This work was supported by the Intramural Research Program of the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS), National Institutes of Health (grant AR041124 to MIM). We thank the members of the NIAMS Genomic Technology Section and NIAMS Light Imaging Section. Download .zip (.01 MB) Help with zip files Download .zip (.03 MB) Help with zip files Download .zip (10.03 MB) Help with zip files Mice All animal studies were carried out according to the protocol approved by the Institutional Animal Care and Use Committee at the National Institute of Arthritis and Musculoskeletal and Skin Diseases (protocol A022-01-01). K14CreERTM/LSL-SOX2-IRES-GFP mouse line was generated as previously described (Uchiyama et al., 2019Uchiyama A. Nayak S. Graf R. Cross M. Hasneen K. Gutkind J.S. et al.SOX2 Epidermal Overexpression Promotes Cutaneous Wound Healing via Activation of EGFR/MEK/ERK Signaling Mediated by EGFR Ligands.Journal of Investigative Dermatology. 2019; 1391809-20.e8Abstract Full Text Full Text PDF Scopus (34) Google Scholar). Hyperglycemia (>250 mg/dL fasting blood glucose) was induced with 60 mg/kg streptozotocin (STZ) in citrate buffer (Sigma-Aldrich) injected peritoneally for 5 consecutive days between 3-5 weeks of age; citrate buffer was used as a control. 5 weeks after STZ treatment, Sox2 expression was induced with 1 mg topical tamoxifen applied to shaved dorsal skin for 5 consecutive days between 8-10 weeks of age; 100% ethanol was used as a control. Mice were anesthetized, hair was shaved, and biopsies were created by a sterile disposable 6 mm biopsy punch (Integra Miltex). For analysis, each wound site was photographed with a digital camera at two-day intervals after wounding and wound area was measured using ImageJ software (version 1.48, National Institutes of Health). Sample collections were performed with littermate controls. Cell Suspension Preparation Single-cell suspensions were prepared in five batches prepared from 6 mm punch biopsies of full-thickness mouse skin using the Miltenyi Whole Skin Dissociation Kit (Miltenyi Biotec). Samples were incubated in enzymes P, D, and A for 3 hours at 37°C in gentleMACS C Tubes (Miltenyi Biotec). Samples were diluted with 0.5 mL of 4°C KBM-Gold Keratinocyte Growth Medium (Lonza). Mechanical dissociation was then performed using the h_skin_01 setting on a Miltenyi gentleMACS dissociator (Miltenyi Biotec) for 20 seconds. After brief centrifugation, suspensions were filtered through 30 μm pre-separation filters (Miltenyi Biotec) to remove debris. 4 mL of 4°C KBM-Gold Keratinocyte Growth Medium (Lonza) was applied to the gentleMACS C Tubes and was used to rinse the filters. Samples were centrifuged for 5 minutes at 300 g at 4°C. Supernatant was aspirated and cell pellets were resuspended in 10 mL mL 0.4% BSA (Sigma) prepared in 1x PBS (Gibco). BSA Cell viability was assessed by staining with ViaStain AOPI Staining Solution (Nexcelom Bioscience), and live/dead cells were counted using a Cellometer Auto 2000 (Nexcelom Bioscience). Volume of each suspension containing 50,000 live cells was centrifuged at 300g at 4°C for 5 minutes and resuspended in 50 μL 0.4% BSA prepared as described above. Single-Cell Capture, Droplet-Based Single-Cell RNA Sequencing Analysis, and Data Processing Single cells were captured using the droplet-based microfluidic Chromium Controller system (10x Genomics) according to manufacturer instructions. Cell suspensions were loaded onto separate chip inlets to perform cell encapsulation into droplets. Reverse transcription, cDNA generation, and library preparation were performed using the standard 10x Genomics single-cell protocol (Chromium Single Cell 3' Reagent Kit). Libraries were sequenced on the Illumina NovaSeq 2000, NovaSeq6000, or NovaSeq X platforms (Illumina) following manufacturer instructions to obtain an average of approximately 522,000,000 reads per sample. Single-Cell RNA Sequencing Data Analysis Raw sequencing data were aligned to an GRCm38/mm10 reference genome and counted using CellRanger (version 6.0.1, 6.0.2, and 7.2.0; 10x Genomics). Secondary analysis was performed using R (version 4.3.2, available from https://www.r-project.org); the R packages SoupX (version 1.6.2; Young and Behjati, 2020Young M.D. Behjati S. SoupX removes ambient RNA contamination from droplet-based single-cell RNA sequencing data.GigaScience. 2020; 9Crossref Scopus (391) Google Scholar), Seurat (version 5.0.1; Satija et al., 2015Satija R. Farrell J.A. Gennert D. Schier A.F. Regev A. Spatial reconstruction of single-cell gene expression data.Nature Biotechnology. 2015; 33: 495-502Crossref PubMed Scopus (3221) Google Scholar), sctransform (version 0.4.1; Choudhary and Satija, 2022Choudhary S. Satija R. Comparison and evaluation of statistical error models for scRNA-seq.Genome Biology. 2022; 23Crossref PubMed Scopus (100) Google Scholar), DoubletFinder (version 2.0.4; McGinnis et al., 2019McGinnis C.S. Murrow L.M. Gartner Z.J. DoubletFinder: Doublet Detection in Single-Cell RNA Sequencing Data Using Artificial Nearest Neighbors.Cell Systems. 2019; 8329-37.e4Abstract Full Text Full Text PDF PubMed Scopus (1309) Google Scholar), Harmony (version 1.2.0; Korsunsky et al., 2019Korsunsky I. Millard N. Fan J. Slowikowski K. Zhang F. Wei K. et al.Fast, sensitive and accurate integration of single-cell data with Harmony.Nature Methods. 2019; 16: 1289-1296Crossref PubMed Scopus (2826) Google Scholar), DESeq2 (version 1.42.1; Love et al., 2014Love M.I. Huber W. Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2.Genome Biology. 2014; 15Crossref Scopus (48240) Google Scholar), MultiNicheNet (version 1.0.3; Browaeys et al., 2023Browaeys R. Gilis J. Sang-Aram C. Bleser P.D. Hoste L. Tavernier S. et al.MultiNicheNet: a flexible framework for differential cell-cell communication analysis from multi-sample multi-condition single-cell transcriptomics data.bioRxiv. 2023; Google Scholar); Python (version 3.10; available from https://www.python.org); and the Python package leidenalg (version 0.10.1; Traag et al., 2019Traag V.A. Waltman L. van Eck N.J. From Louvain to Leiden: guaranteeing well-connected communities.Scientific Reports. 2019; 9Crossref PubMed Scopus (1908) Google Scholar). Counts were corrected for background RNA contamination with SoupX according to the recommended protocol (Young and Behjati, 2020Young M.D. Behjati S. SoupX removes ambient RNA contamination from droplet-based single-cell RNA sequencing data.GigaScience. 2020; 9Crossref Scopus (391) Google Scholar). Cells with fewer than 200 genes and/or greater than 5% mitochondrial gene content were discarded. Genes present in fewer than 3 cells were also excluded from analysis. Counts were normalized and scaled using variance-stabilizing transformation through the sctransform package before principal component analysis to compute 50 principal components based on 5,000 highly variable genes found using sctransform. Each sample was clustered using 30 nearest neighbors with the cosine metric and the Leiden algorithm at a resolution of 0.8, then underwent doublet detection using DoubletFinder with 50 principal components (PCs) using the default parameters. Doublets, clusters within each sample containing more than 50% cells with doublet identity, and cells containing greater than 5,000 genes were then discarded, and samples were reprocessed as above before cell cycle scoring; the difference between G2M and S phase scores was regressed out during SCT renormalization. Samples were merged and batch corrected using Harmony, and the 50 PCs corrected with Harmony were used as input to Uniform Manifold Approximation and Projection visualization and reclustering at 0.4 resolution using 30 neighbors with the Leiden algorithm. Cluster markers were determined with the Wilcoxon rank-sum test with Bonferroni correction for adjusted p-values on log-normalized RNA counts; clusters were annotated on the basis of markers significantly upregulated in that cluster relative to all others in each experimental group. Clusters expressing erythrocyte markers Hba-a1, Hba-a2, and Alas2 or clusters without informative marker genes were excluded from further analysis, and PCA and UMAP were recalculated. Differential expression analysis between experimental groups was performed by rounding corrected counts to the nearest integer and testing with DESeq2 within the Seurat FindMarkers() function, using Bonferroni correction for adjusted p-values as before. MultiNicheNet was used to compare changes in the cell-cell communication changes between tamoxifen- and vehicle-treated mice in the control and streptozotocin-treated conditions using empirical p-value and default parameters. Ingenuity Pathway Analysis Genes found to be upregulated (FC ≥ 1.5, adjusted p-value < 0.05) or downregulated (FC ≤ -1.5, adjusted p-value < 0.05) in tamoxifen STZ skin versus vehicle STZ skin but not in tamoxifen skin versus vehicle skin by DESeq2 were used as input to Ingenuity Pathway Analysis (Qiagen) to investigate hyperglycemia- and Sox2-dependent changes in the skin using default analysis parameters. Selected diseases and functions predicted to be activated (bias-corrected z-score ≥ 2) were selected for presentation. Immunofluorescence Stainings Formalin-fixed, paraffin-embedded tissue sections were deparaffinized, rehydrated, and stained with hematoxylin and eosin (H&E) by Histoserv, Inc. (Germantown, MD) or processed for immunostaining as previously described (Nayak et al., 2023Nayak S. Jiang K. Hope E. Cross M. Overmiller A. Naz F. et al.Chromatin Landscape Governing Murine Epidermal Differentiation.Journal of Investigative Dermatology. 2023; 1431220-32.e9Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar). After staining with primary and secondary antibodies of interest, sections were counterstained with DAPI to visualize nuclei and mounted in ProLong Gold antifade reagent (Life Technologies). Immunofluorescence images were taken using a Leica TCS SP8 confocal microscope (Leica) coupled to Leica LAS X software (Leica). Immunofluorescence image analysis was performed with Imaris (Bitplane AG, Zürich, Switzerland). H&E-stained slides were scanned using a Leica Slide Scanner SCN400F and analyzed using NDP.view2 (Hamamatsu Corporation, Bridgewater, NJ). Immunohistochemistry Staining Immunohistochemistry staining against Ly-6g was performed by Histoserv, Inc. Ly-6g-stained slides were scanned using a Leica Slide Scanner SCN400F. Immunohistochemistry image analysis was performed with NDP.view2. RNA Extraction and Quantitative PCR Total RNA was extracted from 6 mm whole-skin full-thickness biopsies using RNeasy kit (Qiagen) according to manufacturer instructions. cDNA synthesis was performed using the SuperScript IV First-Strand Synthesis System (Invitrogen) from 100 ng RNA according to manufacturer instructions. Quantitative PCR was performed using SYBR system (Applied Biosystems, Foster City, CA) using a StepOnePlus real-time PCR system (Applied Biosystems) per manufacturer instructions. Relative expression was normalized to the level of β-actin. All primers were purchased from Integrated DNA Technologies (Coralville, Iowa, USA) and were as follows: Sox2, forward (5'-CGGCACAGATGCAACCGAT-3') and backward (5'-CCGTTCATGTAGGTCTGCG-3'); β-actin, forward (5'-GGCTGTATTCCCCTCCATCG-3') and backward (5'-CCAGTTGGTAACAATGCCATGT-3'). Cell Culture and Transductions Primary human epidermal keratinocytes (ATCC) were grown in KBM-Gold Keratinocyte Growth Medium (Lonza) containing either 5 mM glucose of the medium as a control or 25 mM glucose (Sigma) to model human hyperglycemia. Cells were seeded at 100,000 cells per well in a 24-well plate and incubated overnight. Cells were transduced at approximately 50% confluence with either adeno-GFP as a control or adeno-h-SOX2 (Vector Labs) at a MOI of 5 for 24 hours. 48 hours after transduction, cells were treated with 8 μg/mL of mitomycin C for 1 hour and washed 3 times with KBM-Gold Keratinocyte Growth Medium (Lonza) before a scratch was made using WoundMaker (Essen BioScience, cat. 4563). Images were taken at 0 and 24 hours using an IncuCyteS3 (Essen Bioscience) and IncuCyteS3 instrument software (Essen Bioscience). Images were analyzed using ImageJ (National Institutes of Health). Western Blot Western blotting was performed as previously described (Sawaya et al., 2023Sawaya A.P. Uchiyama A. Hope E. Bajpai D. Worrell S. Cross M. et al.Keratin 75 Is a Component of the LINC Complex and Has an Essential Role in Mediating the SOX2 Rapid Healing Response during Wound Repair.Journal of Investigative Dermatology. 2023; 143: 494-498Abstract Full Text Full Text PDF PubMed Scopus (3) Google Scholar). Antibodies Primary (1°) antibodies (Ab) used in for immunofluorescence staining in this study and the concentrations used were as follows: rabbit anti-human SOX2 pAb (Abcam; ab97959) diluted 1:250; guinea pig anti-mouse keratin 5 pAb (LSBio; LS-C22715) diluted 1:500; rabbit anti-mouse keratin 6A pAb (BioLegend, San Diego, CA; 905701) diluted 1:500; rabbit anti-mouse keratin 17 mAb (Cell Signaling Technology; 4543) diluted 1:500; rabbit anti-mouse CD31 (Cell Signaling Technology; 77699) diluted 1:200; and rabbit anti-mouse F4/80 mAb (Cell Signaling Technology; 70076) diluted 1:100. Secondary (2°) Abs used for immunofluorescence staining are as follows: Alexa Fluor 647-conjugated goat anti-guinea pig pAb (Invitrogen; A-21244) diluted 1:250; and Alexa Fluor 488-conjugated donkey anti-rabbit pAb (Invitrogen; A-11008). 1° Ab used in immunohistochemistry staining was rat anti-mouse Ly-6g mAb (Invitrogen, 16-9668-82) diluted 1:2000. 2° Ab used in immunohistochemistry staining was a proprietary formulation of Histoserv, Inc. 1° Abs used in Western blot are as follows: rabbit anti-human SOX2 mAb (Cell Signaling; 3579) diluted 1:1000; and rabbit anti-human RPS14 pAb (Bethyl Laboratories; 18-8816-31) diluted 1:20,000. 2° Ab used in Western blot is Rabbit TrueBlot Anti-Rabbit IgG HRP (Rockland Immunochemicals) diluted 1:1000. Statistical Analysis Statistical analyses were performed using Kruskal-Wallis test with Dunn's multiple comparisons test (Figure 1b), one-way ANOVA with Bonferroni's multiple comparisons test (Supplementary Figure S1a), one-way ANOVA with Šidák's multiple comparisons test (Supplementary Figure S1b, Supplementary Figure S2c), and two-way ANOVA with Tukey's multiple comparisons test (Figure 1e). All data are presented as mean ± SD. Data analysis was done using GraphPad Prism, version 10 (GraphPad Software).Supplementary Figure S2. Bioinformatic analysis of distinct single cell populations. (a) Dot plot denoting markers used to annotate clusters to high-level cell populations. (b) MultiNicheNet analysis compares signaling from keratinocytes to other skin cells in Veh and Tam groups. Dot plot presents pseudobulk expression product of ligand in keratinocytes and receptor in receiving cell type. Samples are only considered for a specific interaction if both sender and receiver cell types are present in sufficient number (≥ 10 cells) in that sample. (c) Quantification of Ly6g+ cells in 5 randomly selected fields of papillary dermis taken at 40× magnification. n = 3-4 per group. Data expressed as mean ± SD. ***p < 0.001.