Single-cell RNA-seq reveals keratinocytes and fibroblasts heterogeneity and their crosstalk via epithelial-mesenchymal transition in psoriasis

doi:10.21203/rs.3.rs-3017107/v1

Download PDF

Article

Single-cell RNA-seq reveals keratinocytes and fibroblasts heterogeneity and their crosstalk via epithelial-mesenchymal transition in psoriasis

https://doi.org/10.21203/rs.3.rs-3017107/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

As a chronic inflammatory autoimmune skin disease with high global prevalence, the pathogenesis of psoriasis remains inconclusive. We performed a high-resolution single-cell RNA sequencing analysis of 94 759 cells from 13 samples including psoriasis and wide-type mouse model. We presented a comprehensive single-cell transcriptional landscape of the skin immune cells in psoriasis, especially the heterogeneity of keratinocytes and fibroblasts. More interestingly, we discovered that special keratinocyte subtypes and fibroblast subtypes could interact with each other through epithelial–mesenchymal transition and validated the results with drug verification. What’s more, we conducted a tentative exploration of the potential involving pathway and disclosed that the IL-17 signaling pathway may be the most relevant one. Collectively, we revealed the full-cycle landscape of key cells associated with psoriasis and provided a more comprehensive understanding of the pathogenesis of psoriasis.

Health sciences/Diseases/Immunological disorders/Autoimmune diseases/Psoriasis

Health sciences/Diseases/Immunological disorders/Inflammatory diseases/Psoriasis

Epithelial-mesenchymal transition

Fibroblasts

Keratinocytes

Psoriasis

Single-cell RNA-seq

Psoriasis is a chronic inflammatory autoimmune skin disease with a high global prevalence of 2% ~ 3%⁽¹⁾ mediated by innate and adaptive immune systems⁽²⁾. Various cells including keratinocytes (KCs), fibroblasts (FBs), plasmacytoid dendritic cells (pDCs), natural killer T cells (NKT), and macrophages are involved in the entire immune process as well as multiple cytokines like TNF-α、IL-12、IL-23、IL-17A^{(3, 4)}. Although the pathogenesis of psoriasis has been investigated extensively, the full-cycle landscape of key cells associated with psoriasis remains inconclusive.

KCs are key cells in the pathogenesis of psoriasis. As the main component of the skin epidermis, KCs play an important role in resisting environmental damage and maintaining skin homeostasis⁽⁵⁾. Stimulated by signaling molecules, the cell cycle of KCs is accelerated leading to cutaneous psoriatic lesions formation⁽⁶⁾. Various factors such as genetic regulation, cytokines, and receptors, transcription factors regulate keratinocytes in both the initiation and maintenance phases of psoriasis. To date, how various factors work together to regulate the functions of keratinocytes remains limited.

In recent years, more and more attention has been attracted to the role of FBs in psoriasis. It’s known that skin FBs are closely bound up with the condition and function of the skin by producing collagen and other intercellular matrix substances⁽⁷⁾. Unlike KCs, the potential effect of FBs in the pathological process of psoriasis has not been explored extensively. Agnieszka G.⁽⁸⁾ recently reported that dermal FBs in psoriasis patients were dysfunctioned possibly by upregulation of proinflammatory factor, antioxidant protein and factors related to signal transduction and proteolytic processes. What’s more, downregulation of proteins in psoriatic FBs mostly accountable for the transcription/translation processes, glycolysis/adenosine triphosphate synthesis and structural molecules. Consequently, epidermal cells such as KCs were promoted to hyperproliferation. Obviously, FBs and KCs interact with each other in the occurrence and development process of psoriasis. However, the detailed intercellular signaling is indefinite.

Epithelial–mesenchymal transition (EMT) is a series of transformation processes in which epithelial cells are transformed from adhesion cell form to free cell form with mesenchymal phenotype and acquire the ability to invade extracellular matrix through specific procedures⁽⁹⁾. EMT and the reverse process mesenchymal-epithelial transition (MET) occur during the entire course of embryonic development and finally promote tissue morphogenesis⁽¹⁰⁾. Under the stimulation of trauma and inflammatory injury, mature epithelial or endothelial cells in the tissue transform into fibroblasts and other related cells, leading to tissue remodeling⁽¹¹⁾. This EMT process terminates when the stimulation is eliminated. EMT is closely associated with tumor formation and metastasis. In the last few decades, EMT has been ascertained in diverse cancers, including breast cancer⁽¹²⁾, head and neck cancer⁽¹³⁾, uveal Melanoma⁽¹⁴⁾, colorectal Cancer⁽¹⁵⁾ and so on. But few research were found to investigate the EMT phenomenon in psoriasis.

Single-cell RNA-sequencing (scRNA-seq) shows a powerful tool to detect cell specificity and inter-cell differences, explore the cooperative mode of inter-cell operation, and study tissue heterogeneity from the perspective of cell mapping. Chenxin Qie et al. revealed the transcriptional landscape and heterogeneity of skin macrophages in Vsir knockout murine psoriasis with scRNA-seq analysis⁽¹⁶⁾. Jared Liu et al. performed scRNA-seq in psoriatic and healthy skin and disclosed the different landscape of CD8 + T cells, including 2 nonexhausted Tc17 cell subpopulations involved in disease severity⁽¹⁷⁾. ScRNA-seq analysis were also applied in psoriatic arthritis by Frank Penkava et al. discovering the clonal expansions of pro-inflammatory synovial CD8 T cells expressing tissue-homing receptors⁽¹⁸⁾. In 2021, Gary Reynolds et al. used scRNA-seq to uncover the cellular heterogeneity and organization of human first trimester prenatal skin, adult healthy skin and skin of patients with AD and psoriasis⁽¹⁹⁾. ScRNA-seq enables precise analysis of single cell sequencing, greatly improves the resolution of cell heterogeneity study, as well as comprehensive, precise studies of cytokines and signaling pathways.

In this study, to research the critical role of KCs and FBs as well as their dynamic crosstalk through EMT in psoriasis, scRNA-seq analysis combined with bulk RNA-seq, gene array and experimental verification were conducted to get comprehensive understanding of the pathogenesis of psoriasis.

Single cell atlas of the skin of psoriasis mice and WT mice

To identify the cell composition of the skin of psoriasis, 13 samples were collected in this study, including 4 IMQ (imiquimod) induced psoriasis samples, 5 WT samples, 2 SHP099 samples, 1 Chrna5 knockout (KO) sample, and 1 Chrna5 KO IMQ induced psoriasis sample (Fig. 1A-C; Supplementary Table 1). With Prior assessment, 94 759 cells were screened, and 13 cell populations were identified, including adipocytes (47), DC (1810), Endothelial (911), FBs (33118), Granulocytes (1 067), KCs (42 073), lymphatic-Endothelial (1 156), Macrophage (6 656), Mast (444), Melanocyte (532), NK (547), Schwann (997), T (5 401) (Fig. 1D-F; Fig.S1A, B). Among these detected cell populations, it can be easily discovered that KCs, FBs, Macrophage and T cells were the main populations differ between IMQ mice and WT mice (Fig. 1E, F).

KCs heterogeneity in psoriasis

To better interpret the heterogeneity of KCs, further restrictive and comprehensive analysis were conducted and 7 KC subtypes were identified including basal 1 KC, basal 2 KC, basal 3 hyperplasia KC, EMT KC, Mitotic KC, Sebaceous KC and Spinous KC (Fig. 2A-C; Fig.S2A, B). The results of the gene set variation analysis (GSVA) suggested that basal 2 KC, basal 3 hyperplasia KC and EMT KC exhibits a high level of EMT (Fig. 2C). Gene expression profiles were then analyzed in each KC subtype. EMT related marker genes such as Vim, Dcn, Mmp, Zeb2 and Twist2 were especially high in EMT KC (Fig. 2D). The EMT score of KC subtype demonstrated that basal 2 KC, basal 3 hyperplasia KC and EMT KC got high EMT score compared with other KC subtypes (Fig. 2E; Fig.S2C). All the above results revealed that basal KC were involved in EMT process in psoriasis.

KC subtypes were analyzed separately among 13 samples (Fig.S2E). The comparison of IMQ group and WT group disclosed that the proportion of EMT KC subtypes increased in IMQ group (Fig. 2G, H). The top differentially expressed genes (DEGs) of EMT KC subtype in IMQ group compared with WT group were then further dissected (Fig. 2F). Most of the DEGs such as Vim, Saa3 and Cebpd were EMT associated genes which furthermore proved that EMT phenomenon occurred in EMT KC in IMQ group more than WT group. Meanwhile, it was found that basal 2 KC and basal 3 hyperplasia KC also expressed higher EMT associated genes in IMQ group. This indicate that EMT KC may transform from basal 2 KC and basal 3 hyperplasia KC. To verify the results, immunohistochemistry (IHC) was performed with mice skin of IMQ and WT group (Fig. 2K). The result showed that the expression of vim was stronger in IMQ group than that of WT group, and the expression of e-cadherin was missing in IMQ group compared with that of WT group, which strongly proved that EMT was happening in IMQ group more than WT group.

To explore the potential signaling pathway related to EMT process, the top DEGs of IMQ group compared with WT group were analyzed with bulk RNA-seq (data source: GSM2299980) and scRNA-seq analysis (Fig. 2I, J; Fig.S2D). S100a8 and S100a9 were highlighted as a result. It was well known that S100a8 and S100a9 were specific marker genes of IL-17 signaling pathway. Here, it can be speculated that IL-17 signaling pathway may be the main pathway regulating the EMT process.

FBs heterogeneity in psoriasis

As one of the cell types that actively engaged in maintaining the condition and function of the skin, FBs was gaining more and more attention in psoriasis. In this study, 7 FB subtypes were recognized, including Angiogenic fibroblast (Angiogeni_Fib), inflammation-associated fibroblasts (iFib1,2), Mitotic fibroblast (Mitotic_Fib), Normal fibroblast (Normal_Fib), Pericytes fibroblast (Pericytes_Fib) and Secreted Phosphoprotein 1 fibroblasts (Spp1_Fib) (Fig. 3A-C; Fig.S3A, B).

As shown in Fig. 3A, iFib1 and iFib2 showed as the biggest two clusters compared to other FB subtypes. Compared to WT group, the proportion of iFib1 and iFib2 increased obviously in IMQ group (Fig. 3D; Fig.S3C). This indicated that iFibs might play an important role in psoriasis. To confirm the result, combined bulk RNA-seq (data source: GSM2299980) and scRNA-seq analysis were implemented between IMQ group and WT group. Inflammation specific marker Saa3 and Lcn2 turned out to be the two genes that differ the most between IMQ group and WT group (Fig. 3F, G; Fig.S3D). Moreover, the difference was notably outstanding in iFib1, iFib2 and Mitotic_Fib, which further demonstrated that iFib was quite associated with the process of psoriasis.

Various FB subtypes were involved in different signaling pathways. As shown in Fig. 3B, except inflammatory response, iFib1 and iFib2 also showed high level of EMT in GSVA result. KEGG pathway map of IMQ vs WT scRNA-seq displayed that IL-17 signaling pathway was the most significantly enriched pathway. Combined with the EMT phenomenon in KCs and its possible relationship with IL-17 signaling pathway, we speculated that iFib can interact with KCs through EMT and IL-17 signaling pathway may be the most relevant pathway.

Dynamic evolution process of EMT in psoriasis

To verify aforementioned speculation, extensive exploration was carried out to probe the dynamic evolution process of EMT in psoriasis. EMT subpopulation of FBs was reconfirmed first (Fig. 4A, B; Fig.S4). IFib1 and iFib2 got the highest EMT score which consist with previous study. EMT score of the 94,759 cells from all 13 samples validated that FBs and KCs clusters were main cell type correlated with EMT process (Fig. 4C). Subsequently, recluster and UMAP visualization of EMT populations were conducted, including iFib1, iFib2, basal 3 hyperplasia KC, EMT KC, Mitotic_KC and Spinous_KC (Fig. 4D). Diffusion map and principal component analysis showed their developmental trajectories, which was sequential process from Spinous_KC to Mitotic_KC, basal 3 hyperplasia KC, EMT KC, iFib2 and iFib1(Fig. 4E). Furthermore, as the KCs marker Krt5 increase, EMT marker Vim and Zeb2 decreased and IL-17 signaling pathway marker S100a9 increase clearly which strongly proved the existence of EMT in psoriasis and its intimately connection of IL-17 signaling pathway (Fig. 4F).

Chrna5 KO reduce EMT KC in psoriasis

It has been reported that Chrna5 is related to cell proliferation, invasion, migration, angiogenesis as well as immunological functions^{(20, 21)}. Our previous study substantiated that Chrna5 was overexpressed on the back skin of mice after IMQ treatment (Fig. 5A)⁽²²⁾. Chrna5 KO inhibits epidermal proliferation and inflammation in mice. What’s more, scRNA-seq analysis revealed that Chrna5 KO significantly reduce the number of KCs. To explore the influence of Chrna5 on each KCs subtypes, further analysis was conducted. As exhibited in Fig. 5B, C, the proportion of EMT KC was decreased obviously. This verified from the other side that EMT KC was the main cells deciding epidermal proliferation in psoriasis. To identify the involved signaling pathway, KEGG pathway enrichment and marker gene analysis were performed again between IMQ group and KO IMQ group. The results showed that IL-17 signaling pathway was downregulated, and MAPK signaling pathway, TGF-β signaling pathway and Th17 differentiation were upregulated (Fig. 5D-F). Combined with above mentioned results, Chrna5 KO may reduce EMT KC expression by suppressing IL-17 signaling pathway.

Drug verification

According to British Association of Dermatologists guidelines for biologic therapy for psoriasis 2020, three commonly used biological agents were selected, including IL-17A inhibitor secukinumab (SEC), IL-12/23 inhibitor ustekinumab (UST) and TNFα inhibitor adalimumab (ADA). To find out the related genes that affected by each biological agent, RNA-seq analysis and expression array were performed (Fig. 6A-D). With these up-regulated and down- regulated genes, associated signaling pathway were enriched (Fig. 6E). As a result, cell cycle and IL-17 signaling pathway turned out to be the most associated pathway that suppressed by three biological agents, which consisted with our scRNA-seq analysis.

To validate the result, we scored the suppressed genes by three biological agents in single cell subpopulation. UMAP plot of IL-17 and its receptor expression evinced that IL-17ra mainly expressed in FBs and IL-17rc mainly expressed in KCs (Fig.S5A). The KC subtype with high SEC score consisted with that highly expressing IL-17rc (Fig.S5B, C). Similarly, the FB subtype with high ADA score, SEC sore and UST score consisted with that highly expressing IL-17ra (Fig.S5D-F). Linked with Fig. 3a, the FB subtype were exactly ifib1 and ifib2, which indicated that ifib subtypes were main subtype affected by these biological agents.

Correlation analysis between drug intervention and EMT

The genes upregulated by three biological agents were scored in EMT associated cells first (Fig. 7A). Obviously, IL-17ra mainly expressed in ifibs which again supported that ifib subtypes were main subtype affected by these biological agents. EMT score and drug intervention score in EMT associated cells were demonstrated respectively in Fig. 7B and Fig. 7C. From the visualized score, it is evident that both EMT score and drug intervention score are high in ifib subpopulations. To further strengthen the correlation, correlation analysis was deployed (Fig. 7D). The relative coefficient of EMT score vs. three drug score was 0.92, 0.74, 0.71 respectively, which indicated that EMT was highly positively correlated with drug intervention.

Psoriasis has been a long-standing excruciating burden for patients around the world because of its repeated attacks, prolonged skin damage, accompanied by itching characteristics. The burden of psoriasis is even greater than that of other chronic skin disease due to the obvious disfiguring damage and stigma associated with it, the uncomfortable treatment process, the fear of prognosis, and the need to depend on family care for loss of work. Although biologic treatments for psoriasis have achieved excellent results, the high cost is not affordable for everyone. So, it’s urgent to further in-depth study the pathogenesis of psoriasis and provide new ideas for more popular treatment. In this study, 13 samples were collected from database published so far and scRNA-seq were performed to construct a single-cell transcriptome map of psoriasis.

As autoimmune disease, various cells including KCs, FBs, pDCs, NKT, and macrophages are involved in the entire immune process of psoriasis^{(3, 4)}. In our scRNA-seq analysis, 13 cell populations were identified, among which KCs, FBs, Macrophage and T cells were found the main populations differ between IMQ mice and WT mice. The proportion of KCs and FBs showed the most significant difference which indicated that KCs and FBs played quite important role in the pathogenesis of psoriasis.

KCs are not only important cytokine producing cells, but also important cytokine target cells and important immune cells⁽²³⁾. The release of a large number of pro-inflammatory cytokines secreted by keratinocytes, recruitment of neutrophils, and angiogenesis form a vicious cycle in psoriatic skin, leading to infiltration of inflammatory cells and abnormal epidermal proliferation and differentiation⁽²⁴⁾. Interestingly, in the process of interpreting the heterogeneity of KCs, we discovered that a subtype of EMT KC increased obviously in IMQ group compared with WT group. What’s more, Chrna5 knock out significantly reduce the number of KCs in psoriasis, EMT KC was the main population that was decreased obviously. This denoted that EMT phenomenon might donate essential part in the development of psoriasis. EMT of KCs has been demonstrated in wound healing in recent years. It was reported that fibroblast growth factor 2 accelerated the EMT process in KCs during wound healing process⁽²⁵⁾. Yan Shi et al. revealed that ROS can promote hypoxia-induced KC EMT by inducing SOX2 expression and activating of Wnt/β-Catenin⁽²⁶⁾. The role of EMT in psoriasis needs to be explored further.

Then in the mapping and identification of FBs, 7 FB subtypes were recognized, among which iFibs exhibited special heterogeneity in psoriasis. Compared with WT group, the proportion of 2 iFib subpopulations increased clearly in IMQ group. Combined bulk RNA-seq (data source: GSM2299980) and scRNA-seq analysis also confirmed the result, indicating the important role of iFibs in psoriasis. Previous studies focused on KCs, but recent studies have shown that FBs are abnormal in psoriasis lesions and may be one of the initiating factors of psoriasis. Current studies have noted that FBs in patients with psoriasis can maintain the hyperproliferative state of the psoriatic epidermis and produce higher levels of IL-6, IL-8 and keratinocyte growth factor^{(27, 28)}. It is suggested that the high level of cytokines released by psoriatic FBs through paracrine may be an important initiating factor for the excessive proliferation of KCs⁽²⁹⁾. Gęgotek et al. reported that FBs were dysfunctioned due to virous factors which may promptly affect intercellular signaling and elevate the hyperproliferation of KCs⁽⁸⁾. Our result further suggested that iFib maybe the right subpopulation that associated with psoriasis initiating.

More interestingly, iFibs also got high EMT score in our scRNA-seq analysis, which encourage us to speculate that KCs and FBs interact with each other via EMT. To explore the dynamic evolution process of EMT in psoriasis, diffusion map and principal component analysis were conducted. Finally, their developmental trajectories were confirmed as sequential process from Spinous_KC to Mitotic_KC, basal 3 hyperplasia KC, EMT KC, iFib2 and iFib1.

EMT is a biological process by which epithelial cells lose polarity and adhesion and acquire the migration and invasion properties of mesenchymal cells⁽³⁰⁾. It plays a key role in the regulation of tumor invasion and metastasis in many human cancers^(31–33). In addition to ensuring cell motility, EMT is also involved in restoring self-renewal properties⁽³⁴⁾. In the process of wound healing, tissue regeneration, and fibrosis, EMT was triggered by inflammation^(35–37). EMT-like event enables KCs transfer to a metastable state which allows KCs to move while maintaining loose contact with the surroundings. Our findings revealed that EMT was truly happening in psoriasis and may be involved in the pathogenesis. The EMT event in psoriasis was a dynamic process between EMT KCs and iFibs. To confirm the EMT process in psoriasis, drug verification and correlation analysis were conducted. The result showed that EMT score was highly positively correlated with drug intervention which indicated that UST, SEC and ADA may improve the condition of psoriasis via intervening the process of EMT. So EMT may play quite essential role in the pathogenesis of psoriasis which needs further investigation.

The process of EMT is regulated by an aggregation of signaling pathways and signaling molecules under different contexts. EMT involved in gastrulation is promoted by the canonical Wnt signaling pathway⁽³⁸⁾. EMT participated in metastasis is induced by variety of signaling molecules like Wnt, JAK/STAT3, FGF, EGF BMP, TGFβ, etc^{(33, 39, 40)}. EMT associated with wound healing is induced by factors such as VEGF and TGFβ⁽³⁷⁾. However, the signaling pathway regulating the EMT process during psoriasis remains unknown. In our whole scRNA-seq analysis process, quite a few evidence prove that IL-17 signaling pathway may involve in the EMT process. While analyzing the heterogeneity of KCs, S100a8 and S100a9 were highlighted as the top DEGs between IMQ group and WT group, which were known as specific marker genes of IL-17 signaling pathway. While analyzing the heterogeneity of FBs, KEGG pathway map of IMQ vs WT scRNA-seq displayed that IL-17 signaling pathway was the most significantly enriched pathway. In the dynamic evolution process of EMT in psoriasis, as the KCs marker Krt5 increase, EMT marker Vim and Zeb2 decreased and IL-17 signaling pathway marker S100a9 increase clearly which strongly proved the existence of EMT in psoriasis and its intimately connection of IL-17 signaling pathway. In drug verification, cell cycle and IL-17 signaling pathway also turned out to be the most associated pathway that suppressed by three biological agents, which consisted with our scRNA-seq analysis. Taken together, IL-17 signaling pathway maybe the most potential pathway that participated in the EMT process of psoriasis.

Collectively, based on an unbiased single-cell RNA-seq analysis, we presented the full-cycle landscape of key cells associated with psoriasis such as KCs and FBs. Most interestingly, we revealed that EMT KC, ifibs and their crosstalk through EMT may play a quite critical role in the pathogenesis of psoriasis, in which IL-17 signaling pathway may be the most potential pathway that involved.

Data collection

The data of four mice samples including WT, KO, IMQ and KO IMQ were obtained from our previous study (GSM5795802, GSM5795803, GSM5795800 and GSM5795801)⁽²²⁾. Other 9 samples were extracted from published datasets. GSM4547483 for C57BL/6 mice treated with IMQ for 5 consecutive days on the ear. GSM5024748 and GSM5024749 for C57BL/6J mice treated with IMQ on the shaved dorsal skin for 4 consecutive days. GSM5024750 and GSM5024751 for IMQ-induced psoriasis mice model treated with SHP099 hydrochlorid for an additional 4 days. GSM4547481 and GSM4547482 for dermis and epidermis of untreated WT mice respectively. GSM5024746 and GSM5024747 for untreated WT mice. Detailed information of the 13 groups summarized in Supplementary Table 1. The bulk RNA-seq data used in Fig. 3F was extracted from GEO with accession number GSM2299980. Other data used in Fig. 6 were extracted from GEO with accession number GSE171012, GSE103489, GSE74697.

RNA-seq analysis

The data of RNA-seq analysis were obtained from GEO with accession number GSM2299980, GSE171012 and GSE74697. FastQC tool (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/) was applied to analyze the raw data. Then Trimmomatic (http://www.usadellab.org/cms/index.php?page=trimmomatic) was deployed to get rid of the low-quality sequences and adaptor-contaminated sequences. The filtered data was mapped to mouse mm10 or the human hg19 genome using HISAT2(https://daehwankimlab.github.io/hisat2/). SAM files were generated and then converted to the BAM format with SAMtools^{(41, 42)}. HTseq⁽⁴³⁾ was employed to calculate the number of reads mapped on each gene. The DEGs were then screened out from the mapping results using the R package DEseq2⁽⁴⁴⁾, and those with P < 0.05 and log2FoldChange value > 1 or log2FoldChange value <-1 were identified as genes with significant differences.

Microarray analysis

GEO2R was used to compare samples between psoriasis group before treated and psoriasis group after treated by UST in order to identify DEGs.

ScRNA-seq analysis

The Seurat package (version 4.2.0) was applied to transform the matrix to Seurat object. Quality filtering were carried out according to respective sequencing characteristics. Samples with cells less than 1000 were eliminated. The data was standardized by CCA to erase the influence of data sources. Then “ScaleData” function was used to further linearly regress against the normalized expression levels for each gene, the total UMI counts and mitochondrial RNA content per cell. Subsequently, principal component analysis (PCA) with “RunPCA” were performed. Eventually, gene expression and clustering results were visualized on a umap plot of the top ten PCs using “RunUMAP”, and “FindNeighbors” and “FindClusters” functions are used to cluster cells.

Annotation of cell types

Cell type is determined based on gene expression of known markers, i.e., Rgs1, Cd3g, Cd3e, Cd69 for T cells; Plp1, Mpz, Gatm, Mbp, Mal for Schwann cells; Gzmb, Klrd1,Ptprcap, Xcl1, Kirk1for NK cells; Dct, Tyrp1, Kit, Ptgds, Pax3 for melanocyte; Hdc, Ccl6, Srgn, Ccl3, Lilr4b for mast cells; Lyz2, Fcer1g, Ctss for macrophages; Ccl21a, Mmrn1, Cldn5, Lvve1 for lymphatic endothelial cells; Krt5, Krt14, Lgals7, Krt17, Sfn for Keratinocyte; Retnlg, Wfdc21, Wfdc17, Cxcl2 for granulocytes; Dcn, Col3a1, Col1a2, Col1a1,Clo6a3 for fibroblasts; Ptptb, Fit1, Vwf, Podxl for endothelial cells; H2-Ab1, H2-Eb1, H2-Aa, H2-DMa for DCs; Cfd, Car3, Cidec, Adig for adipocytes. Annotation of the sub-celltype of KCs and FBs is based on the combination of known marker expression and participated signaling pathway.

EMT signature scores

Based on 194 EMT genes, which was reported by Yutong Sha et al.⁽⁴⁵⁾, “AddModuleScore” (Seurat, version 4.2.0) function was employed to calculate the EMT signature score of all clusters. By estimating EMT meta-signature scores between different groups based on the nonparametric Wilcoxon rank sum test and figuring the average expression level of the 194 genes for each cell, the EMT signature score is acquried.

Gene set variation analysis

To explore the function of different KC subtypes and FB subtypes, we estimated the pathway activity through the gene set variation analysis (GSVA, version 1.30.0), based on hallmark gene sets from the Molecular Signatures Database (MSigDB, version 6.2)

Trajectory inference

Trajectory analysis was performed via the R package monocle (version 2.22.0) to explore the dynamic EMT process between KCs and FBs. The top highly variable genes (HVGs) were differential gene intersection selected from single-cell sequencing, RNA-Seq. Then the “estimateSizeFactors”, “estimateDispersions” functions were used to build statistical models to characterize the data. Additionally, “reduceDimension” function were applied to reduce dimensions, and “orderCells” function were used to place cells onto a pseudotime trajectory.

Correlation analysis

R package PerformanceAnalytics was used for the correlation between EMT and three drugs intervention. Normalize expression II17ra from Seurat as the input data. “AddModuleScore” (Seurat, version 4.2.0) function was used to evaluate the average expression value of the drug induced gene set.

KEGG

KEGG pathway analyses were performed using the DAVID Functional Annotation Bioinformatics Microarray Analysis tool (http://david.abcc.ncifcrf.gov/).

IHC assay

After deparaffinization, slides were hydrated in alcohol. Antigen epitope retrieval was induced by microwave heating. Normal goat serum For Blocking. To examine the expression pattern of candidate antibodies in IMQ and WT tissues, sections were immunostained with primary antibodies overnight at 4°C, the secondary anti- body used for immunostaining was CoraLite488-conjugated anti-rabbit or CoraLite594-conjugated anti-mouse immunoglobulin (Wuhan Sanying, Proteintech), following the protocol of manufacturer. DAPI was used for counterstaining. The antibodies were listed in the Supplementary Table 2.

Competing interests

The authors declare no competing interests.

Author contributions

Ninging Dang conceived and supervised the project. Dianhao Guo designed and performed the bioinformatics analyses and interpreted the analytical data. Xiaokang Li integrated all materials, compiled the figures and wrote the manuscript with edits from Ninging Dang and Shuhong Huang. Shuhong Huang performed the HE experiments. Jing Wang and Xin Liu searched literature and collected related data. All authors discussed and approved the final manuscript.

Acknowledgements

This study was supported by grants from the Shandong Provincial Natural Science Foundation of China (numbers ZR2022MH242), the National Natural Science Foundation of China (number 82273527).

Greb JE, Goldminz AM, Elder JT, Lebwohl MG, Gladman DD, Wu JJ, et al. Psoriasis. Nat Rev Dis Primers. 2016;2:16082.
Griffiths CEM, Armstrong AW, Gudjonsson JE, Barker J. Psoriasis. Lancet. 2021;397(10281):1301–15.
Kanda N. Psoriasis: Pathogenesis, Comorbidities, and Therapy Updated. Int J Mol Sci. 2021;22(6).
Sato Y, Ogawa E, Okuyama R. Role of Innate Immune Cells in Psoriasis. Int J Mol Sci. 2020;21(18).
Zhou X, Chen Y, Cui L, Shi Y, Guo C. Advances in the pathogenesis of psoriasis: from keratinocyte perspective. Cell Death Dis. 2022;13(1):81.
Wang Z, Zhou H, Zheng H, Zhou X, Shen G, Teng X, et al. Autophagy-based unconventional secretion of HMGB1 by keratinocytes plays a pivotal role in psoriatic skin inflammation. Autophagy. 2021;17(2):529–52.
Varani J, Dame MK, Rittie L, Fligiel SE, Kang S, Fisher GJ, et al. Decreased collagen production in chronologically aged skin: roles of age-dependent alteration in fibroblast function and defective mechanical stimulation. Am J Pathol. 2006;168(6):1861–8.
Gęgotek A, Domingues P, Wroński A, Skrzydlewska E. Changes in Proteome of Fibroblasts Isolated from Psoriatic Skin Lesions. Int J Mol Sci. 2020;21(15).
Brabletz S, Schuhwerk H, Brabletz T, Stemmler MP. Dynamic EMT: a multi-tool for tumor progression. Embo j. 2021;40(18):e108647.
Huang Y, Hong W, Wei X. The molecular mechanisms and therapeutic strategies of EMT in tumor progression and metastasis. J Hematol Oncol. 2022;15(1):129.
Amack JD. Cellular dynamics of EMT: lessons from live in vivo imaging of embryonic development. Cell Commun Signal. 2021;19(1):79.
Addison JB, Voronkova MA, Fugett JH, Lin CC, Linville NC, Trinh B, et al. Functional Hierarchy and Cooperation of EMT Master Transcription Factors in Breast Cancer Metastasis. Mol Cancer Res. 2021;19(5):784–98.
Pal A, Barrett TF, Paolini R, Parikh A, Puram SV. Partial EMT in head and neck cancer biology: a spectrum instead of a switch. Oncogene. 2021;40(32):5049–65.
Lv Y, He L, Jin M, Sun W, Tan G, Liu Z. EMT-Related Gene Signature Predicts the Prognosis in Uveal Melanoma Patients. J Oncol. 2022;2022:5436988.
Vu T, Datta PK. Regulation of EMT in Colorectal Cancer: A Culprit in Metastasis. Cancers (Basel). 2017;9(12).
Qie C, Jiang J, Liu W, Hu X, Chen W, Xie X, et al. Single-cell RNA-Seq reveals the transcriptional landscape and heterogeneity of skin macrophages in Vsir(-/-) murine psoriasis. Theranostics. 2020;10(23):10483–97.
Liu J, Chang HW, Huang ZM, Nakamura M, Sekhon S, Ahn R, et al. Single-cell RNA sequencing of psoriatic skin identifies pathogenic Tc17 cell subsets and reveals distinctions between CD8(+) T cells in autoimmunity and cancer. J Allergy Clin Immunol. 2021;147(6):2370–80.
Penkava F, Velasco-Herrera MDC, Young MD, Yager N, Nwosu LN, Pratt AG, et al. Single-cell sequencing reveals clonal expansions of pro-inflammatory synovial CD8 T cells expressing tissue-homing receptors in psoriatic arthritis. Nat Commun. 2020;11(1):4767.
Reynolds G, Vegh P, Fletcher J, Poyner EFM, Stephenson E, Goh I, et al. Developmental cell programs are co-opted in inflammatory skin disease. Science. 2021;371(6527).
Cushing KC, Chiplunker A, Li A, Sung YJ, Geisman T, Chen LS, et al. Smoking Interacts With CHRNA5, a Nicotinic Acetylcholine Receptor Subunit Gene, to Influence the Risk of IBD-Related Surgery. Inflamm Bowel Dis. 2018;24(5):1057–64.
Dang N, Meng X, Qin G, An Y, Zhang Q, Cheng X, et al. α5-nAChR modulates melanoma growth through the Notch1 signaling pathway. J Cell Physiol. 2020;235(11):7816–26.
Wang J, Li X, Zhang P, Yang T, Liu N, Qin L, et al. CHRNA5 Is Overexpressed in Patients with Psoriasis and Promotes Psoriasis-Like Inflammation in Mouse Models. J Invest Dermatol. 2022;142(11):2978-87.e6.
Ortiz-Lopez LI, Choudhary V, Bollag WB. Updated Perspectives on Keratinocytes and Psoriasis: Keratinocytes are More Than Innocent Bystanders. Psoriasis (Auckl). 2022;12:73–87.
Benezeder T, Painsi C, Patra V, Dey S, Holcmann M, Lange-Asschenfeldt B, et al. Dithranol targets keratinocytes, their crosstalk with neutrophils and inhibits the IL-36 inflammatory loop in psoriasis. Elife. 2020;9.
Koike Y, Yozaki M, Utani A, Murota H. Fibroblast growth factor 2 accelerates the epithelial-mesenchymal transition in keratinocytes during wound healing process. Sci Rep. 2020;10(1):18545.
Shi Y, Wang S, Yang R, Wang Z, Zhang W, Liu H, et al. ROS Promote Hypoxia-Induced Keratinocyte Epithelial-Mesenchymal Transition by Inducing SOX2 Expression and Subsequent Activation of Wnt/β-Catenin. Oxid Med Cell Longev. 2022;2022:1084006.
Zalewska A, Głowacka E, Wyczółkowska J, Tchórzewski H, Narbutt J, Sysa-Jedrzejowska A. Interleukin 6 and 8 levels in plasma and fibroblast cultures in psoriasis. Mediators Inflamm. 2006;2006(1):81767.
Capellino S, Cosentino M, Luini A, Bombelli R, Lowin T, Cutolo M, et al. Increased expression of dopamine receptors in synovial fibroblasts from patients with rheumatoid arthritis: inhibitory effects of dopamine on interleukin-8 and interleukin-6. Arthritis Rheumatol. 2014;66(10):2685–93.
Bártolo I, Reis RL, Marques AP, Cerqueira MT. Keratinocyte Growth Factor-Based Strategies for Wound Re-Epithelialization. Tissue Eng Part B Rev. 2022;28(3):665–76.
Pastushenko I, Blanpain C. EMT Transition States during Tumor Progression and Metastasis. Trends Cell Biol. 2019;29(3):212–26.
Wei C, Yang C, Wang S, Shi D, Zhang C, Lin X, et al. Crosstalk between cancer cells and tumor associated macrophages is required for mesenchymal circulating tumor cell-mediated colorectal cancer metastasis. Mol Cancer. 2019;18(1):64.
Nachiyappan A, Gupta N, Taneja R. EHMT1/EHMT2 in EMT, cancer stemness and drug resistance: emerging evidence and mechanisms. Febs j. 2022;289(5):1329–51.
Jin W. Role of JAK/STAT3 Signaling in the Regulation of Metastasis, the Transition of Cancer Stem Cells, and Chemoresistance of Cancer by Epithelial-Mesenchymal Transition. Cells. 2020;9(1).
Savagner P, Arnoux V. [Epithelio-mesenchymal transition and cutaneous wound healing]. Bull Acad Natl Med. 2009;193(9):1981–91; discussion 92.
Debnath P, Huirem RS, Dutta P, Palchaudhuri S. Epithelial-mesenchymal transition and its transcription factors. Biosci Rep. 2022;42(1).
Marconi GD, Fonticoli L, Rajan TS, Pierdomenico SD, Trubiani O, Pizzicannella J, et al. Epithelial-Mesenchymal Transition (EMT): The Type-2 EMT in Wound Healing, Tissue Regeneration and Organ Fibrosis. Cells. 2021;10(7).
Sisto M, Ribatti D, Lisi S. Organ Fibrosis and Autoimmunity: The Role of Inflammation in TGFβ-Dependent EMT. Biomolecules. 2021;11(2).
Howard S, Deroo T, Fujita Y, Itasaki N. A positive role of cadherin in Wnt/β-catenin signalling during epithelial-mesenchymal transition. PLoS One. 2011;6(8):e23899.
Ramesh V, Brabletz T, Ceppi P. Targeting EMT in Cancer with Repurposed Metabolic Inhibitors. Trends Cancer. 2020;6(11):942–50.
Sabouni E, Nejad MM, Mojtabavi S, Khoshduz S, Mojtabavi M, Nadafzadeh N, et al. Unraveling the function of epithelial-mesenchymal transition (EMT) in colorectal cancer: Metastasis, therapy response, and revisiting molecular pathways. Biomed Pharmacother. 2023;160:114395.
Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics. 2009;25(16):2078–9.
Li H. A statistical framework for SNP calling, mutation discovery, association mapping and population genetical parameter estimation from sequencing data. Bioinformatics. 2011;27(21):2987–93.
Anders S, Pyl PT, Huber W. HTSeq–a Python framework to work with high-throughput sequencing data. Bioinformatics. 2015;31(2):166–9.
Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550.
Sha Y, Wang S, Zhou P, Nie Q. Inference and multiscale model of epithelial-to-mesenchymal transition via single-cell transcriptomic data. Nucleic Acids Res. 2020;48(17):9505–20.

(Not answered)

SupplementaryInformation.pdf

Download PDF

Editorial decision: revise
20 Jul, 2023
Review #1 received at journal
18 Jul, 2023
Review #2 received at journal
12 Jul, 2023
Reviewer #2 agreed at journal
06 Jul, 2023
Reviewer #1 agreed at journal
06 Jul, 2023
Reviewers invited by journal
05 Jul, 2023
Submission checks completed at journal
12 Jun, 2023
First submitted to journal
09 Jun, 2023
Unknown event
07 Jun, 2023
Editor assigned by journal
03 Jun, 2023

You are reading this latest preprint version

Single-cell RNA-seq reveals keratinocytes and fibroblasts heterogeneity and their crosstalk via epithelial-mesenchymal transition in psoriasis

Status:

Version 1

Abstract

Figures

Introduction

Result

Discussion

Materials and Method

KEGG

Declarations

Competing interests

Author contributions

Acknowledgements

References

Additional Declarations

Supplementary Files

Status:

Version 1