Human Mast Cells Upregulate Cathepsin B, a Novel Marker of Itch in Psoriasis

Mast cells (MCs) contribute to skin inflammation. In psoriasis, the activation of cutaneous neuroimmune networks commonly leads to itch. To dissect the unique contribution of MCs to the cutaneous neuroinflammatory response in psoriasis, we examined their density, distribution, relation to nerve fibres and disease severity, and molecular signature by comparing RNA-seq analysis of MCs isolated from the skin of psoriasis patients and healthy volunteers. In involved psoriasis skin, MCs and Calcitonin Gene-Related Peptide (CGRP)-positive nerve fibres were spatially associated, and the increase of both MC and nerve fibre density correlated with disease severity. Gene set enrichment analysis of differentially expressed genes in involved psoriasis skin showed significant representation of neuron-related pathways (i.e., regulation of neuron projection along with dendrite and dendritic spine morphogenesis), indicating MC engagement in neuronal development and supporting the evidence of close MC–nerve fibre interaction. Furthermore, the analysis of 208 identified itch-associated genes revealed that CTSB, TLR4, and TACR1 were upregulated in MCs in involved skin. In both whole-skin published datasets and isolated MCs, CTSB was found to be a reliable indicator of the psoriasis condition. Furthermore, cathepsin B+ cells were increased in psoriasis skin and cathepsin B+ MC density correlated with disease severity. Therefore, our study provides evidence that cathepsin B could serve as a common indicator of the MC-dependent itch signature in psoriasis.


Introduction
Mast cells (MCs) are uniquely seeded into the skin in early life, occupying stable clonal territories unless or until inflammation occurs [1].They are commonly found in close proximity to nerve fibres [2,3], especially those expressing substance P (SP) and calcitonin gene-related peptide (CGRP) [4][5][6][7].Mast cells make direct membrane-membrane contact with nerve fibres [8,9].Cytokines and granule-associated mediators released by MCs, such as TNF-α, interleukin (IL)-1b, IL-6, and IL-17A, have been shown to act on the nervous system and promote pain.Furthermore, MCs synthesize and release nerve growth Cells 2023, 12, 2177 2 of 24 factor (NGF), known to induce inflammatory hyperalgesia [10] and neurite outgrowth and synaptogenesis in the brain as well as to enhance CGRP expression in dorsal root ganglia sensory neurons [11].
Studies have established the potential for MC activity in psoriasis.Degranulated MCs are consistently found to be increased in involved psoriasis skin [22,23] and during the occurrence of new lesions [24].Serum total IgE and tryptase are raised in psoriasis [25,26], the latter implicated in the generation of the autoantigen Pso p27, found to be increased in lesional skin and possibly involved in disease chronicity [27].In addition, MCs can capture, store, and release exogenous IL-17A [28], and can release extracellular traps and degranulate upon stimulation of IL-23 and IL-1β [29], which are major cytokines involved in psoriasis pathogenesis [30].However, despite the indirect evidence, the individual contribution of MCs in psoriasis remains to be fully elucidated.In psoriatic lesions, a number of inflammatory markers such as kallikreins (KLKs) [31], alarmins such as IL-33 [17], and TNF-α are predictably augmented [32,33].Furthermore, an increase in MC-activating neuropeptides such as SP and vasoactive intestinal peptide (VIP) [34], along with SP-, Transient Receptor Potential Vallinoid 1 (TRPV1)-, and Neurokinin-1 Receptor (NK1R)positive nerve fibres [35][36][37].The neuropeptides SP, CGRP, and VIP induce skin MCs to degranulate and promote the release of pro-inflammatory mediators, including TNF-α and IL-3 [38,39].In human skin, MCs express SP [40] and VIP [41], suggestive of autocrine or paracrine activity.Therefore, during disease the MC-nerve axis may drive feed-forward neurogenic inflammation and regulate allo-and hypersensitivity through as yet undefined neuroplastic mechanisms [42].
Through this study, we sought to better define the involvement of MCs in psoriasis and the potential role of neuroimmune interactions in the pathogenesis of psoriatic itch.

Human Skin Samples
Normal and involved psoriasis skin punch biopsies were obtained from the Dermatopharmacology Department in Salford Royal Hospital through clinical studies according to protocols approved by the local NHS research ethics committee (13/NW/0867 and 10/H1005/77).Normal human skin punch biopsies were obtained through the Manchester skin health biobank (14/NW/0185).All subjects provided written informed consent in accord with the declaration of Helsinki.Study subject characteristics are shown in Supplementary Tables S2 and S3.Three of the healthy volunteer samples had previously been analysed as controls in another study [62].

Acquisition of Human Skin Biopsies
Punch biopsy specimens 6 mm in diameter were obtained under local anaesthesia from sun-protected normal and involved psoriasis buttock skin, and the psoriasis area severity index (PASI) scores for psoriasis patients were calculated by a trained specialist nurse.Patients were asked to score how intensely they felt their skin itching before the biopsy was taken on a scale from 1 to 10, with 1 indicating no itch and 10 indicating the worst itch imaginable.Additionally, a worst itch numerical scale was used whereby each psoriasis patient was asked for the intensity of the most intense itch they had felt in the last 24 h before the biopsy was taken on a scale from 1 to 10, with 1 indicating no itch and 10 indicating the worst itch imaginable.
Images were collected on a Zeiss Axioimager.D2 upright microscope using a 20×/0.5 EC Plan-neofluar objective and captured using a Coolsnap HQ2 camera (Photometrics) through Micromanager software v1.4.23.Images were captured using an Olympus BX53 upright microscope with a 20×/0.75UPlanS Apo objective and an Olympus DP73 camera with Olympus cellSens v1.18 software.Both microscopes utilised specific bandpass filter sets for DAPI, FITC, Cy3, and Cy5 which were used to prevent bleedthrough between channels.Images were then processed and analysed using Image J (v1.53g) software (https://imagej.nih.gov/ij/download.htmlaccessed on 7 January 2021).Images were false coloured with cyan, magenta, and yellow to enable greater colour accessibility.The distance between each mast cell and the closest staining for PGP or SP/CGRP and VIP was measured manually using calibrated images in ImageJ.The density of MCs was calculated as the number of MCs per measured area of dermis in mm 2 taken from n = 10 nonoverlapping calibrated images for each subject, in Image J. Similarly, the density of neuropeptides was calculated from the stained area of nerves as a proportion of the total dermal area in the same number of non-overlapping images.Mean densities per subject were correlated with Psoriasis Area and Severity Index (PASI) using Pearson's correlation statistic.Analysis of cathepsin B positive cells was restricted to the dermis only.Cellular cathepsin B counts were limited to colocalization with MC tryptase in the case of MCs or presence of nuclei in the case of other cellular stainings.Single colour channels of representative composite photomicrographs are shown in the supplementary data (Supplementary Figure S5).

Mast Cell Isolation from Skin
Mast cells were isolated from 6 mm biopsies obtained from 6 mm skin punch biopsies acquired from the buttocks of volunteers with and without psoriasis under local anaesthesia in the Salford Royal Foundation Trust Hospital Dermatopharmacology Unit.Biopsies were stored in ice-cold PBS.Skin biopsies were cut into small pieces in digestion solution containing RPMI 1640 medium (Sigma-Aldrich, Gillingham, UK), 0.5% BSA (ThermoFisher Scientific), 100 U.mL −1 penicillin and 100 µg.mL −1 streptomycin, 0.5 Wunch units.mL−1 of Liberase TM (Roche, Basel, Switzerland), and 100 ng.mL −1 of SCF (Genscript, Piscataway, NJ, USA), then incubated overnight at 37 • C, 5% CO 2 .

Sample Preparation for RNAseq
A SMART-Seq ® Ultra ® Low Input RNA Kit was used to synthesize the full cDNA from 1000 MCs according to the manufacturer's instructions (Takara Bio, London, UK).The analysis was performed on the next-generation sequencing Illumina HiSeq platform (Illumina, Cambridge, UK).

Gene Expression Analysis
Microarray datasets of published studies comparing the expression of genes in involved psoriatic skin with normal skin were acquired from the National Centre for Biotechnology Information GEO Datasets (https://www.ncbi.nlm.nih.gov/gds/).
Two-group comparison analysis comparing involved psoriatic versus normal skin samples was conducted using Qlucore software (v3.6) (Qlucore AB, Lund, Sweden).Lognormalized counts obtained from normal and involved psoriatic skin samples from isolated MCs, as well as the whole skin datasets GDS4602, GSE80047, and GSE78097 were extracted from Qlucore and used for principal component analysis, which was conducted in R using the mixOmics package [69].Three-dimensional PCA plots were generated from the logtransformed normalized matrix from Qlucore using the pca and plotIndiv functions of the mixOmics R package.Heatmaps and volcano plots were generated using the R packages pheatmap [70] and EnhancedVolcano [71], respectively.
Genes with a p-value threshold of 0.05 after false discovery rate adjustment (q) were deemed significant, and log2 fold change values were used to determine upregulation (log2 fold change < 0) or downregulation (log2 fold change > 0) relative to normal skin controls.Venn diagrams were created using the R package ggvenn [72].

Gene Set Enrichment Analysis
Gene set enrichment analysis (GSEA) was carried out on a list of ranked genes based on p-values and the sign of the log2 fold changes for isolated MCs and the whole skin datasets.GSEA was computed using the R packages clusterProfiler [73] and ReactomePA [74].Three ontology databases were used, namely Gene Ontology (GO), the Kyoto Encyclopedia of Genes and Genomes (KEGG), and REACTOME.Significance was set at p-value of < 0.05 adjusted for false discovery rate.Scatterplots were created using the ggplot2 R package.Plots of enrichment scores were created using the R package enrichplot.

Ingenuity Pathway Analysis
Datasets of pre-filtered significant genes (q < 0.05) of isolated MCs and the whole skin datasets GDS4602, GSE80047, and GSE78097 were uploaded to Ingenuity Pathway Analysis (IPA) software v01.20.04 (Qiagen).Core analysis was run individually on each dataset, and analysis of canonical pathways, upstream regulators, and causal networks was conducted.Comparison analysis was then conducted between different datasets to assess the differences in predicted activation z-scores and the significance between isolated MCs and whole skin datasets.Significance was set at p < 0.05 for the Fisher exact test in each analysis.

xCell Analysis
Cell type enrichment analysis of publicly available datasets and isolated MCs from normal skin and psoriasis lesions was performed using the xCell webtool [75] for microarray data and RNA-seq data.xCell scores and significance levels were obtained for each donor/condition.The proportion of samples in which significant (p < 0.05) MC and neuron enrichment was detected was displayed per study.Isolated MCs were added as internal control to assess the efficacy of the algorithm in picking up MC-specific signatures.

Random Forest Analysis
Genes were assessed for importance in predicting psoriatic from control tissue using Boruta (v.7.0.0) [76].Those that were initially determined to be important were assessed again using Boruta to refine the number of genes and avoid overfitting the random forest model.Random forest analysis was conducted on pre-selected itch-associated genes (n = 208) and run using 500 trees with the Rattle v.5.4.0 R package [77].The estimation of the area under the curve (AUC) for sensitivity and specificity analysis of the random forest model and out of bag (OOB) AUC for estimation of the prediction and error rate of the model applied to the validation set were calculated using Rattle.

Peripheral Blood-Derived Human Mast Cell Culture and Cathepsin B ELISA
Human peripheral blood NC24 leukocyte cones obtained from anonymous healthy donors were obtained from NHS blood and transplant (Manchester, UK) and used in accordance with a protocol approved by the University of Manchester Research Ethics Committee (UREC ref 2018-2696-5711).Progenitor cells were isolated and cultured as previously described [78].Mature MCs were activated by SP (Genscript, Oxford, UK) at the indicated concentrations for 1 h.Cell-free supernatants were collected, and human total cathepsin B content was analysed by ELISA (R&D Systems, Abingdon, UK) according to the manufacturer's instructions.

Statistical Analysis
Graphing and analyses were performed using Prism v9.1.2(GraphPad, San Diego, CA, USA).Data were subject to normality tests, and inter-group comparisons were made using appropriate parametric (unpaired t-test) or non-parametric (Mann-Whitney test) methods.Correlation statistics were generated from a simple linear regression model.Concentration response curves were generated from the four-parameter logistic regression least squares fit to the raw data.

In Psoriasis Skin, Mast Cells Are Spatially Associated with Neuropeptide Positive Nerve Fibres and Their Density Correlates with Disease Severity
While MCs are consistently increased [23,79], nerve fibres have been found to be both heightened [36] as well as diminished [80] in psoriasis patients' skin.Colocalisation of MCs with nerve fibres suggests increased cross-talk, indicating exposure of the latter to neurotrophic and itch mediators.However, both small-diameter DRG afferents and cholinergic sympathetic nerve fibres can express neuropeptides [81,82] that activate human MCs [83][84][85].In psoriasis, the role of neuropeptides is unclear, with conflicting findings having been published [34,79,[86][87][88].Therefore, the association between MCs and nerve fibre markers in skin from psoriasis lesions and healthy volunteers was investigated.
While CGRP density and CGRP/PGP 9.5 ratio (Figure 2a,b) were not significantly altered in psoriatic lesions compared to normal skin (p = 0.37 & p = 0.31 respectively), MCs showed significantly increased proximity to CGRP+ nerve fibres 60. 35 1c,d).CGRP density was found to correlate significantly with psoriasis disease severity, (r = 0.65, p = 0.043) (Figure 2e).Representative photomicrographs are shown in Figure 2f,g.This was not the case for the neuropeptides SP or VIP, for which no significant increases in density, MC association, or significant correlation with PASI was observed (Supplementary Figures S1 and S2).
Thus, our findings indicate that in psoriasis, MCs and nerve fibres both increase in number and correlate with disease, while their proximity is increased.MCs were predominantly and uniquely associated with CGRP+ nerve fibres, suggesting a role in disease.Therefore, we proceeded to investigate the potential impact of this association on MCs and itch-related gene expression.

Mast Cells from Psoriasis Skin Exhibit Enriched Expression of Neuronal Development-Associated Genes
To explore the relative contribution of MCs to psoriasis pathogenesis, we analysed the transcriptional signature of MCs isolated from skin by cell sorting (Supplementary Figure S3, Supplementary Table S2).
Differential gene expression analysis of MCs isolated from involved psoriasis and normal skin revealed a total of 249 differentially expressed genes (Supplementary Table S4).Principal Component Analysis showed separation between normal and involved skin MCs, with 25% and 12% of variation between subjects explained by components 1 and 2, respectively (Figure 3a).The volcano plot and heatmap of the top 50 significantly differentially expressed genes are displayed in Figure 3b,c.
To interpret the transcriptomic analysis in the context of curated pathways and predict the underlying mechanisms, we performed canonical pathway and upstream regulator analyses using Ingenuity Pathway Analysis (IPA).Canonical pathway analysis of differentially expressed genes revealed enrichment in sixteen canonical pathways, among which three showed predicted activation, namely, Coordinated Lysosomal Expression and Regulation (CLEAR) Signaling Pathway, Autophagy, and PI3K Signaling in B Lymphocytes (Figure 4a).Common shared genes across the three identified pathways were ATF4 and TLR4 (Figure 4b).Upstream regulators analysis revealed a possible role of IL1B and NFkB in the modulation of differentially expressed genes (Figure 4c), including ATF4, CTSB, CLD12, EIF2AK3, CXCL12, FABP5, LTA, TLR4, and TACR1.Causal network analysis predicted the involvement of the IFN and IL-12 cytokine families, which promote downstream gene activation.These results suggest an influence of the surrounding pro-inflammatory cytokine milieu in the observed MC signature in psoriasis skin, in particular mediated by IL-1β.

Mast Cell Signatures Are Unequally Represented in Psoriasis Whole Skin Studies
To assess the contribution of observed signatures in MCs isolated from psoriasis lesions, we sought to compare the differential gene expression in isolated MCs to the published whole-tissue microarray datasets GDS4602, GSE80047, and GSE78097 from psoriasis patients (Table 1).Mast cells (MCs) were isolated from healthy donors (n = 7) and involved skin of psoriasis patients (n = 6, 2 donors pooled).RNA was extracted and analysed.Microarray data from three separate studies on bulk skin from healthy donors and psoriasis patients were downloaded from the Gene Expression Omnibus database.Two-sample comparison analysis was performed for each study comparing normal and involved samples.Differentially expressed genes (DEGs) were selected based on p-values adjusted for the false discovery rate (q < 0.05).
Among the 249 genes identified in isolated MCs, 127 genes were shared with at least one of the selected microarray datasets (Figure 6a, Supplementary Table S6), with 15 genes shared by all datasets (Figure 6b).In particular, CLDN12, FABP5, IFI44L, and TMEM167B showed similar regulation patterns across datasets.Of note, CLDN12 and FABP5 were among the genes identified in IPA as being modulated by upstream IL1B and NFkB complex activity (Figure 4c).
We then performed canonical pathway and upstream regulator comparison analysis between the isolated MCs and whole-tissue datasets using IPA, and analysed the similarities in the enriched canonical pathways and upstream regulators found in the isolated MCs.We observed similar positive activation z-scores in the Autophagy and CLEAR pathways, as well as a downstream effect of IL1B (Supplementary Tables S7 and S8).
Comparison of causal network analysis showed shared predicted activation mediated by the IL-12 cytokine family only in GSE78097, while no predicted shared effect of the IFN cytokine family were observed (Supplementary Table S8).
In addition, GSEA analysis of whole-skin microarray studies was performed to assess any similarities in enrichment with isolated MCs.Among significantly enriched ontology terms, only four were shared with at least one dataset, namely, cytosolic DNA-sensing pathway, pattern specification process, anterior/posterior pattern specification, and regionalization (Supplementary Table S8).
Hence, both our IPA and GSEA analyses indicate that isolated MCs share marginal overlap of the pathways and gene set enrichment characteristics of the whole skin of psoriasis subjects.
To assess whether the discrepancies observed in IPA and GSEA could be the result of uneven representation of MCs in whole skin samples, we performed xCell enrichment, with a particular focus on the relative distribution of MC-and neuron-associated signatures in normal and involved skin samples.Isolated MCs were included as internal control.
Hence, both our IPA and GSEA analyses indicate that isolated MCs share marginal overlap of the pathways and gene set enrichment characteristics of the whole skin of psoriasis subjects.xCell enrichment revealed that the representation of MCs in different datasets was unequal in normal skin samples of GDS4602 (89.1%),GSE80047 (58.3%), and GSE78097 (50%), and detected only in a small number of samples of involved psoriasis skin (10% and 7.7% of samples in GSE80047 and GSE78097, respectively) (Figure 6c).The same analysis showed that neuronal signatures in involved skin were undetectable in GSE80047 and GSE78097, while they were detected in 50% of isolated MC samples from involved psoriasis skin.
Hence, analysis of isolated MCs is necessary in order to define their specific signature in psoriasis, as it is underestimated in currently published whole skin studies.The reason for the low MC representation in involved psoriasis skin could be due to insufficient MCs in biopsy samples or to a change in the MC-specific signature defined by the xCell algorithm.

CTSB, TACR1, and TLR4 Are Itch-Associated Genes Differentially Expressed by Psoriasis Skin Mast Cells
To explore gene signatures contributing to itch in psoriasis, we compiled a curated list of 208 itch-associated genes after a systematic review of the available literature on genes known to be either positively correlated with perceived itch or scratching, to directly induce itch, or to induce sensitisation to pruritogens (Supplementary Table S9).
To investigate whether these genes were upregulated in psoriasis skin, and as such could contribute to psoriatic itch, their differential expression was investigated in published microarray datasets.Of itch-associated genes, 89 (42.7%), 135 (64.9%), and 42 (20.2%)were significantly differentially expressed in GDS4602, GSE80047, and GSE78097, respectively (Supplementary Table S10).However, in the isolated MCs only three itch-associated genes (1.4%) were statistically different from healthy controls, namely, CTSB, TACR1, and TLR4.These genes were only found to be differentially expressed in the GSE80047 dataset, with CTSB and TLR4 expression significantly upregulated and TACR1 significantly downregulated (Supplementary Table S10).
Taken together, these data imply that involved psoriasis skin MCs are primed to respond to selective immune and neurogenic factors as well as to release proteases that can activate itch responsive neurons but lack the effect mediated by the majority of literaturedefined itch-associated genes and variably represented in whole skin analyses.

Cathepsin B Is a Predictive Indicator of Mast Cell-Dependent Itch in Psoriasis
To identify itch-associated genes that are predictive indicators of psoriasis in comparison with normal skin, random forest analysis was utilised.
CTSB, JAK1, and LTB4R were the only genes found by random forest analysis to be significant indicators of the psoriasis condition in both the whole skin datasets and isolated MCs.CTSB was the only gene to be significantly upregulated in MCs as well, suggesting that Cathepsin B could serve as common indicator of MC-dependent itch signature in psoriasis.Cathepsin B is upregulated in animal models of inflamed skin, where it contributes to aberrant keratinocyte proliferation [90,91].We demonstrated in vitro that cathepsin B is released by human peripheral blood-derived MCs in response to stimulation by neuropeptide SP (Figure 7c).Therefore, we investigated changes in cathepsin B expression in psoriasis skin via immunofluorescence microscopy.Cathepsin B was widely expressed in the epidermis and dermis of normal and psoriasis skin, including in MCs (Figure 7d,e and Supplementary Figure S4a,b 7g), the density of cathepsin B+ MCs did significantly correlate with psoriasis severity (r = 0.74, * p = 0.0225), whereas the total cathepsin B+ cell number did not.Notably these analyses confirmed our earlier findings (Figure 1) in which the total MC number was increased in psoriasis skin and their density correlated with PASI (Supplementary Figure S4c,d).Our analyses did not include the measurement of acellular cathepsin B. However, we detected an increase of extracellular cathepsin B in the form of small spots, indicative of its induced release, which was present in increasing amounts in high-PASI subjects (Supplementary Figure S4e).These findings indicate that an increase in cathepsin B specific to MCs in psoriasis might provide functional cross-talk between peptidergic neurons and MCs.

Discussion
In this study, we sought to better define the involvement of MCs in psoriasis and the potential role of neuroimmune interactions in the pathogenesis of psoriatic itch.To elucidate the contribution of MCs in the generation of itch and explore the potential role of the crosstalk between MCs and neurons, we analysed the transcriptomic landscape of MCs isolated from psoriatic skin and assessed MC-neuron proximity through immunohistochemistry.
Here, we demonstrate increased MC and nerve fibre density which correlates with disease severity in psoriasis and that MCs and CGRP+ nerve fibres are spatially associated.Furthermore, we found that CTSB expression can serve as an indicator of MC involvement in the itch signature of psoriasis, and that MC cathepsin B expression correlates with psoriasis severity.
Mast cells were found to be significantly more frequent in psoriasis, which corroborates previous findings [23].Further, we confirmed that an increased number of MCs and nerve fibres normalised to a specific area is associated with psoriasis severity.The association between the intensity of pruritus and density of MCs remains controversial [92,93], and it is possible that increased numbers alone could increase association between MC and nerves and increase the cutaneous innervation and neuropeptides content in lesions correlated with both disease and pruritus [92,94].Interestingly, involved psoriasis skin from patients with pruritus shows increased MC activation, implying a mechanistic contribution of secreted MC mediators to itch [92].
Close proximity between mucosal MCs and intrinsic CGRP+ sensory neurons has been demonstrated in the mouse colon, where the interaction causes MC activation [95], as well as in psoriasis [6].In human skin, CGRP is present in few sensory nerve fibres.However, CGRP+ nerve fibres are increased in conditions such as prurigo nodularis, and are thought to contribute to neurogenic inflammation and MC recruitment [96].The pruriceptive neuronal subtype NP2 is known to express CGRP in mice [97].Whether CGRP release in psoriasis is fundamental to MC activation and immune cell recruitment, and consequently the initiation of the acute inflammatory process [98], or whether tryptase release by MCs is key in deactivating CGRP [99], thereby promoting resolution of the lesion, is yet to be defined.
The classical MC mediator involved in itch in a variety of inflammatory diseases is histamine [100].However, it seems to play a minor role in psoriatic pruritus [44].According to our findings, MCs derived from psoriasis lesions express higher levels of genes involved in the regulation of neurons, and in particular, receptors that could facilitate communication with sensory nerves, such as the substance P receptor (TACR1).This indicates that the psoriatic microenvironment alters the functionality of skin MCs.Whether these changes lead to increased susceptibility of these cells to neuropeptides remains to be defined.Furthermore, the inflammatory indicators found to be upregulated in psoriasis MCs, such as ATF4, CXCL12, LTA, TACR1, TLR4, and CTSB, interestingly share IL-1β as a common modulator.This cytokine has a key role in psoriasis [101,102], and its abnormal regulation has been suggested to represent one of the early steps in the pathogenesis of the disease [103,104].Thus, our findings suggest that MCs are challenged early on by the cytokine milieu of psoriatic epidermal keratinocytes.
Substance P is known to ligate both the MC Tachykinin receptor 1 (TACR1), sometimes called neurokinin 1 receptor (NK1R), and MRGPRX2 [105][106][107].In our study, TACR1 was upregulated in psoriasis MCs, while MRGPRX2 displayed no significant changes in its expression.While these findings could indicate a prominent role of TACR1 in MCdependent itch in psoriasis, we cannot exclude the contribution of the MRGPRX2-induced signalling pathway, as this receptor is known to be constitutively and preferentially highly expressed in skin MCs [108,109].
Activation of MCs via TLR4 engagement induces the release of pro-inflammatory cytokines [110].Furthermore, TLR4 antagonists administered in a rat model of chronic relapsing itch alleviated both itch and chronic dermatitis [111].Thus, the increased expression of TLR4 could indicate an altered susceptibility of MCs to endogenous ligands such as human beta-defensin 2, which has been shown to not only promote itch through TLR4 signalling in mice [112], but to be overexpressed in psoriasis skin [113].
Interestingly, Th2 cytokines such as IL-4 and IL-13, expressed by MCs, and suggested as signalling bridges between immune sense and nerve fibres [114,115], were not modulated in MCs in psoriasis.Indeed, our analysis of itch-associated genes indicated a specific MC signature in psoriasis and a lack of regulation in the majority of the literature-defined itchassociated genes.Whether additional MC-specific signalling pathways are contributing to itch in psoriasis remains to be investigated.
In our random forest analyses, CTSB was the only gene common to the psoriasis wholeskin datasets and isolated MCs which was a significant indicator of the psoriasis condition.Furthermore, we observed an increase in cathepsin B expression in psoriasis skin, an increase in the density of cathepsin B+ MCs correlating to PASI, and cathepsin B protease release from MCs in response to SP activation.Cathepsin B is a protease with pro-inflammatory functions that is highly expressed in psoriasis plaques [91,116,117].It has been suggested that this may contribute to aberrant keratinocyte growth and differentiation [118].Furthermore, CTSB is considered to be an established itch-associated gene [35].Cathepsin B localises to the secretory granules of primary human skin-derived MCs, and is required for the activation of protryptase into active tryptase [119].MCs have been suggested to contribute to psoriatic pruritus by releasing tryptase [54,120].Because human tryptase is a PAR-2 ligand, the significant upregulation of CTSB in MCs in psoriasis lesions and the correlation of cathepsin B with PASI would suggest a contribution to promoting PAR-2 signalling in sensory nerve fibres.While the downregulation of cathepsin B in keratinocytes has been shown to reduce their proliferation and inflammatory response [90], the silencing of CTSB expression in MCs could become a means of controlling MC-induced itch in psoriasis.
Interestingly, neuronal loss and brain atrophy has been observed in mice lacking CTSB, suggesting a role for cathepsin B in neuronal maintenance [121].However, whether this MC protease plays a role in promoting the increase and outgrowth of nerve fibres in psoriasis is yet to be investigated.
In summary, our findings suggest that bidirectional cutaneous cross-talk between MCs and nerve fibres regulates the growth, morphogenesis, and mediator/receptor expression of the latter.However, the altered microenvironment of neuropeptide, cytokine, and anti-microbial mediators in psoriasis promotes a unique psoriasis-specific itch signature in MCs.In this context, cathepsin B could serve as a biomarker and candidate for interventional therapy.

Figure 1 .
Figure 1.Mast cell and nerve fiber density, spatial association, and correlation with psoriasis severity.Immunofluorescence microscopy was used to identify and measure the density of (a) dermal MC tryptase and (b) PGP 9.5+ nerve fibres in normal (n = 8) and involved psoriasis skin (n = 11).Each subject is represented by a different symbol.(c,d) The distance in µm between MCs and PGP9.5+ nerves and the normalised frequency distribution of MCs within a given distance of a PGP9.5+nerve fibre.(e,f) Representative photomicrographs of normal (e) and involved (f) skin show MC tryptase (cyan) and PGP9.5 (magenta).Discrete areas of PGP9.5 immunofluorescence in the dermis are identified (arrowheads).Scale bar = 20 µm.The tissue density of tryptase-positive MCs (g) and PGP9.5-positive nerve fibres (h) was correlated with the psoriasis severity index (PASI).* p < 0.05, ** p < 0.01, **** p < 0.0001, (unpaired t-test (a,b) Mann-Whitney U test (c), Pearson correlation (g,h)).

Figure 2 .
Figure 2. Calcitonin gene-related peptide density, spatial association with MCs, and correlation with severity of psoriasis.Immunofluorescence microscopy was used to identify and measure the density of MCs (tryptase), nerve fibres (PGP 9.5), and CGRP in normal (n = 6) and involved psoriasis skin (n = 10).(a) CGRP nerve fibre density and (b) proportion of CGRP+ nerve fibres in skin sections.(c) Distance between MC and CGRP+ nerve fibres and (d) normalised frequency distribution of MCs within a given distance of a CGRP+ nerve fibres in skin.(e) Correlation between CGRP density and severity index (PASI).(f,g) Representative photomicrographs of normal (f) and involved (g) skin show MC tryptase (cyan), PGP9.5 (yellow), and CGRP (magenta).The inset panel shows a magnified area with colocalized fluorescence.Scale bar = 20µm.Data are median ± IQR of n = 6 and n = 10 donors.* p < 0.05, **** p < 0.0001, ns: not significant (Mann-Whitney U test (a,c), unpaired t-test (b), Pearson correlation (e)).
12,  x FOR PEER REVIEW 10 of 25 development in psoriasis skin, supporting the evidence of close interaction between MCs and nerve fibres in lesions.

Figure 4 .
Figure 4. Canonical pathway (CP) analysis revealed functional enrichment of the CLEAR and Autophagy pathways in mast cells, with an upstream role for IL1B.Canonical pathway analysis was carried out on 203/249 differentially expressed genes (q < 0.05) using IPA.(a) CP analysis of DEGs, showing a Fisher exact test p-value < 0.05.Activation z-scores are displayed as orange for predicted activation.(b) Venn diagram of shared differentially-expressed genes identified in the CLEAR, Autophagy, and PI3K

Figure 4 .
Figure 4. Canonical pathway (CP) analysis revealed functional enrichment of the CLEAR and Autophagy pathways in mast cells, with an upstream role for IL1B.Canonical pathway analysis was carried out on 203/249 differentially expressed genes (q < 0.05) using IPA.(a) CP analysis of DEGs, showing a Fisher exact test p-value < 0.05.Activation z-scores are displayed as orange for predicted activation.(b) Venn diagram of shared differentially-expressed genes identified in the CLEAR, Autophagy, and PI3K Signaling in B Lymphocytes (PI3K) canonical pathways.(c) Upstream regulator analysis of isolated MCs based on gene expression.Upregulated/downregulated genes (log2 fold change > 0/< 0) are respectively marked in magenta/cyan.Activation z-scores are displayed as orange arrows for predicted activation, blue for predicted inhibition, and grey for no prediction.Inconsistent findings are highlighted in yellow.

Figure 5 .Figure 5 .
Figure 5. Neuron-associated ontology terms in psoriasis mast cells in gene set enrichment analysis.Gene set enrichment analysis of RNA isolated from MCs from healthy (n = 7) and psoriasis lesional Figure 5. Neuron-associated ontology terms in psoriasis mast cells in gene set enrichment analysis.Gene set enrichment analysis of RNA isolated from MCs from healthy (n = 7) and psoriasis lesional skin (n = 6, 2 donors pooled).(a) Scatterplot of significantly enriched ontology terms obtained through the gseGO, gseKEGG, and gsePathway functions of ClusterProfiler/ReactomePA R tools.The cut-off for significance was set to the false discovery rate-corrected p value of < 0.05, and the analyses returned a total of 29 enriched ontology descriptions (23 GO, 2 KEGG, 4 REACTOME), as shown on the left of the plot.The colour of the dots represents the adjusted p-value, while the size of the dots represents the gene counts for each ontology term.The gene ratio is a measure of the total enrichment based on positive hits out of the total number of genes in that pathway.(b) Plot of enrichment scores and rank in the ordered dataset of the neuron-associated GO ontology terms GO:0048814, GO:0061001 and GO:0010975, with the highest-ranked genes showing the most enrichment.

Figure 6 .
Figure 6.Psoriasis mast cells share half of their differentially expressed genes with whole skin, while mast cells and neuronal signatures are unequally represented across studies and conditions.(a) Venn diagram of the number of overlapping differentially expressed genes between healthy donors and psoriasis patients in isolated MCs and microarray datasets.(b) Log2 fold change compared to healthy controls of 15 overlapping genes across datasets.(c) Cell type enrichment analysis of publicly available

Figure 6 .
Figure 6.Psoriasis mast cells share half of their differentially expressed genes with whole skin, while mast cells and neuronal signatures are unequally represented across studies and conditions.(a) Venn diagram of the number of overlapping differentially expressed genes between healthy donors and psoriasis patients in isolated MCs and microarray datasets.(b) Log2 fold change compared to healthy controls of 15 overlapping genes across datasets.(c) Cell type enrichment analysis of publicly available datasets and of MCs isolated from normal skin and psoriasis lesions, performed using xCell.The proportion of samples in which significant (p < 0.05) MC and neuron enrichment was detected is displayed per study.Isolated MCs were added as internal control for MC-specific signatures.

Figure 7 .
Figure 7. Random forest analysis of itch-associated genes and cathepsin B expression in mast cells.Random forest plot of differentially expressed genes in (a) psoriasis whole skin and (b) isolated MCs.Attributes that were significant (green bars), tentative, (yellow bars), or rejected (red bars) as indicative of the psoriasis condition are listed in order of importance.Worst, average and best shadow in each iteration are shown (blue bars).Gene names of significantly indicative attributes are enlarged below each plot.(c) Cathepsin B measured in the cell-free supernatant from human MCs stimulated with SP (concentrations shown) for 1 h.(d,e) Representative photomicrographs of the dermis of healthy (d) and psoriasis skin (e), showing cathepsin B (magenta) and MC tryptase (cyan) with nuclei stained with DAPI.Scale bar = 20 µm.Cathepsin B+ MCs are indicated by arrows.Examples of cathepsin-stained mast cells have been enlarged in each panel.(f,g) Number of total cathepsin B+ cells (f) and cathepsin B+ MCs (g) in normal skin (blue dots) and psoriasis skin (red dots).Data are mean± SEM (f) and median± IQR (g) of n = 9 donors.**** = p < 0.0001, ns: not significant (unpaired t-test).(h,i) total number of cathepsin B+ cells and cathepsin B+ MCs in psoriasis skin correlated with PASI.Significant positive correlation indicated by * p = 0.0225 (Pearson correlation).

Table 1 .
Summary of differential gene expression results from isolated mast cells and whole skin microarray datasets of psoriasis patients.