Disturbed Spatial WNT Activation-A Potential Driver of the Reticularized Skin Phenotype in Systemic Sclerosis

Abstract

Objective: Little is known on the mechanisms necessary to maintain the physiologic adult human skin integrity. This study aims to quantitatively describe anatomic changes in systemic sclerosis (SSc)-skin compared with controls and investigate the underlying mechanisms.

Methods: Skin morphology was histologically assessed in 23 patients with SSc, 18 controls, and 15 patients with hypertrophic scars. Spatial WNT/β-catenin-activation was analyzed by RNAscope and immunofluorescence staining. Enrichment of reticular marker genes in predefined fibroblast subpopulations was done using Gene Ontology (GO) enrichment and gene set enrichment analysis.

Results: SSc skin showed a decrease in number (P < 0.0001/P = 0.0004), area (P < 0.0001), and height (P < 0.0001) of papillae compared with controls and hypertrophic scars, respectively. The expression of papillary/reticular marker genes shifted toward a reticular expression profile in SSc. On the level of previously defined fibroblast populations, the increase of reticular marker genes was particularly pronounced in the PI16+ and SFRP4+ populations (P < 0.0001, respectively). Mechanistically, the expression of the WNT/β-catenin target AXIN2 and the number of fibroblasts with nuclear β-catenin-staining-pattern increased in the papillary compared with the reticular dermis in healthy skin. This polarization was lost in SSc with a two-fold increase in β-catenin-positive fibroblasts and AXIN2-expressing fibroblasts throughout the dermis (P = 0.0095). Enrichment of genes related to WNT/β-catenin-regulation was found in the PI16+ population that also relocates from the reticular to the papillary dermis in SSc.

Conclusion: We demonstrate an association of the "reticularized" skin phenotype in SSc with a profound loss of physiologic spatial WNT/β-catenin-activation. Rescuing the spatial WNT/β-catenin-activation might help restore the physiologic skin organization in future therapeutic approaches of fibrosing disorders.

Figure 1

Visualization of papillary measurements.

Figure 2

Morphology of the dermal papillae in SSc compared with nondiseased skin and hypertrophic scars. (A) Trichrome staining of skin sections of nondiseased control individuals (n = 18), involved (forearm) and noninvolved (back) skin of patients with SSc (n = 23), patients with hypertrophic scars (n = 15), and corresponding histograms representing collagen fiber orientation in angle frequency of individual collagen fibers. Representative images are shown at 200‐, 400‐, and 1,200‐fold magnification. (B) Quantification of papillae/mm, papillary area/high‐power field, papillary height/high‐power field and collagen fiber alignment coefficient of skin sections of control individuals (n = 18), patients with SSc (n = 23), and patients with hypertrophic scars (n = 15) quantified from trichrome staining images. (C) Papillary and reticular gene scores in Prospective Registry for Early Systemic Sclerosis cohort (GSE130955) (Gene set list from Solé‐Boldo et al ²⁴ ). Data are presented as median ± interquartile range. Statistical significance for each comparison was determined by Mann‐Whitney U‐test. P < 0.0125 was considered significant after Bonferroni correction. hypertr., hypertrophic; p, papillary; r, reticular; SSc, systemic sclerosis. Color figure can be viewed in the online issue, which is available at http://onlinelibrary.wiley.com/doi/10.1002/art.43094/abstract.

Figure 3

Spatial activation of WNT/β‐catenin signaling in healthy and SSc skin. (A) RNAscope in situ hybridization images of COL1A1 (orange) and AXIN2 (green) on skin sections of healthy individuals and patients with SSc. Representative images are shown at 200‐ and 400‐fold magnification. Quantification is shown as bar graph. (B) Immunofluorescence staining and confocal immunofluorescence microscopy for β‐catenin (green) and the pan‐fibroblast marker P4Hβ (magenta) of papillary and reticular fibroblasts from nondiseased control individuals (n = 4) and patients with SSc (n = 6). Representative images are shown at 200‐ and 630‐fold magnification. Data presented as median ± interquartile range. Statistical significance was determined by Mann‐Whitney U‐test. P < 0.05 was considered significant. SSc, systemic sclerosis. Color figure can be viewed in the online issue, which is available at http://onlinelibrary.wiley.com/doi/10.1002/art.43094/abstract.

Figure 4

Analysis of fibroblast populations in relation to the papillary and reticular dermis in SSc compared with controls. (A) Mean expression of reticular marker genes in fibroblast populations defined based on small conditional RNA in SSc compared with control. Adjusted P values using the Bonferroni correction are shown. (B) Immunofluorescence staining for PI16 (magenta) and the pan‐fibroblast marker P4Hβ (green) of nondiseased control individuals (n = 5) and patients with SSc (n = 4). Representative pictures are shown at 200‐fold magnification. Data presented as median ± interquartile range. Statistical significance was determined by Mann‐Whitney U‐test. P < 0.05 was considered significant. (C) NES of gene sets in a PI16‐positive fibroblast cluster. ctrl, control; GSEA, gene set enrichment analysis; NES, normalized enrichment score; SSc, systemic sclerosis; UMAP, uniform manifold approximation and projection. Color figure can be viewed in the online issue, which is available at http://onlinelibrary.wiley.com/doi/10.1002/art.43094/abstract.