Prurigo Nodularis Is Characterized by Systemic and Cutaneous T Helper 22 Immune Polarization

Abstract

Prurigo nodularis (PN) is an understudied, chronic inflammatory skin disease that disproportionately affects African Americans and presents with intensely pruritic nodules of unknown etiology. To better characterize the immune dysregulation in PN, PBMCs and skin biopsies were obtained from patients with PN and healthy subjects (majority African American) matched by age, race, and sex. Flow cytometric analysis of functional T-cell response comparing patients with PN with healthy subjects identified increased γδT cells (CD3⁺CD4^-CD8^-γδTCR⁺) and Vδ2⁺ γδT enrichment. Activated T cells demonstrated uniquely increased IL-22 cytokine expression in patients with PN compared with healthy controls. CD4+ and CD8+ T cells were identified as the source of increased circulating IL-22. Consistent with these findings, RNA sequencing of lesional PN skin compared with nonlesional PN skin and biopsy site‒matched control skin demonstrated robust upregulation of T helper (Th) 22‒related genes and signaling networks implicated in impaired epidermal differentiation. Th22‒related cytokine upregulation remained significant, with stratifications by race and biopsy site. Importantly, the expression of the IL-22 receptors IL22RA1 and IL22RA2 was significantly elevated in lesional PN skin. These results indicate that both systemic and cutaneous immune responses in patients with PN are skewed toward a Th22/IL-22 profile. PN may benefit from immunomodulatory therapies directed at Th22‒mediated inflammation.

Figure 1.. PN lesional skin is characterized histologically by marked inflammatory cell infiltrate, robust architectural changes, and greater immunofluorescence signal intensity of IL1A and SERPINB4.

a, H&E staining of lesional PN, non-lesional PN, and matched healthy control skin tissue biopsies from African American females. b, Average epidermal thickness as measured at the thickest suprapapillary epidermal plate. c, Blinded dermatopathologist assessed degree of inflammation (0 = no, 3 = severe). d, Blinded dermatopathologist assessed presence (1) or absence (0) of specific inflammatory cell infiltrate across different skin samples with the average value shown. e, Immunofluorescence staining (IF) of DAPI, IL1A, and SERPINB4 in lesional PN and matched healthy skin from African American females with overlapping areas highlighted in gold. *p-value < .05, ** p-value < .01, and *** p-value < .001. PML, polymorphonuclear leukocyte; L, lymphocytes; N, neutrophils; E, eosinophils; M/H, macrophages / histiocytes; MC, mast cells; PC, plasma cells. 5x scale bar = 200um; 10x scale bar = 100um; 20x scale bar = 50um; 40x scale bar = 20um; IF scale bar = 200um.

Figure 2.. Vδ2⁺γδ T and iNKT-cells are enriched in PN patients.

PBMCs from PN (n=4) and healthy subjects (n=5) (67% African American) were stimulated with PMA and Ionomycin in combination with protein transport inhibitor for 4hrs. The differentiation of T-cell subtypes was assessed by flow cytometry. a, Representative flow cytometry plots for γδ T-cells (CD3⁺CD4⁻ CD8⁻γδTCR⁺ cells). b, Percentage of CD3⁺ γδ T-cells (median ± ICR). c, Representative flow cytometry plots for Vδ1⁺ and Vδ2⁺ subsets of γδ – T cells. d, Percentage of sub-population of γδ-T cells (CD3⁺γδ-TCR⁺Vδ1⁺ or Vδ2⁺) (median ± ICR). e, Representative flow cytometry plots for iNKT-cells (CD3⁺CD4⁻CD8⁻CD56⁺). f, Percentage of CD3⁺ iNKT – cells (median ± ICR). g, Heat map representing geometric mean expression of T-cell populations from healthy and PN subjects as represented by CD4⁺ T helper (CD3⁺CD4⁺CD8⁻) cells, CD8⁺ T cytotoxic cells (CD3⁺CD8⁺ CD4⁻), γδ T-cells (CD3⁺CD4⁻CD8⁻ TCR⁺ cells), iNKT-cells (CD3⁺CD4⁻CD8⁻ CD56⁺). h, Representative flow cytometry plots for CD8 Naïve (CD3⁺CD8⁺CD45RA⁺) cells and CD8 Memory (CD3⁺CD8⁺CD45RO⁺) cells. i, Percentage of CD8 Naïve and CD8 Memory cells (median ± ICR). *P<0.05, healthy controls versus PN subjects, as calculated by a non-parametric Mann Whitney U-test. ns = non-significant.

Figure 3.. Activated T-cells exhibit an IL-22 phenotype in PN.

PBMCs from PN (n=4) and healthy subjects (n=5) were stimulated with PMA and ionomycin combined with protein transport inhibitor for 4hrs. The cells were analyzed for intracellular cytokine expression by flow cytometry. a, Representative flow cytometry histograms of intracellular IL-22 cellular expression versus unstained control (Ctrl). b, Mean fluorescence intensity (MFI) of IL-22 expressing CD3⁺ cells (median ± ICR). c, Percentage of IL-22 expressing T-cell subsets including CD4, CD8, γδ T-cells, and iNKT – cells ± SEM. d, Representative flow cytometry histograms of intracellular TNF cellular expression versus unstained control (Ctrl). e, Mean fluorescence intensity (MFI) of TNF expressing CD3⁺ cells (median ± ICR). f, Percentage of TNF expressing T-cell subsets including CD4, CD8, γδ T-cells, and iNKT-cells ± SEM. g, Representative flow cytometry histograms of intracellular IL-4 cellular expression versus unstained control (Ctrl). h, Mean fluorescence intensity (MFI) of IL-4 expressing CD3⁺ cells (median ± ICR). i, Percentage of IL-4 expressing T-cell subsets including CD4, CD8, γδ T-cells, and iNKT-cells ± SEM. *P<0.05, healthy controls versus PN subjects, as calculated by a non-parametric Mann Whitney U- test. ns = non-significant.

Figure 4.. Analyses of differentially expressed genes (DEGs) in lesional PN (n=13), non-lesional PN (n=13), and matched healthy skin (n=13).

DEGs defined as coding genes with a log base 2 fold change value less than −1 or greater than 1 and FDR adjusted p value less than 0.05 (−1 > logFc > 1, p<0.05). a, Heatmap of all DEGs by RNA-seq in lesional PN, non-lesional PN, and matched healthy control skin. Red, greater expression; blue, lower expression. b-d, Volcano plot compared gene expression in lesional to non-lesional PN skin (b) lesional to healthy skin (c) and non-lesional to healthy skin (d). e, Venn diagram comparison of DEGs. f-h, Plot comparison of DEG fold changes. L/H / L/NL (f). NL/H / L/NL (g). NL/H / L/H (h). L, lesional PN samples; NL, non-lesional PN samples; H, matched healthy control samples.

Figure 5.. Cutaneous mRNA analyses indicate Th22 immune polarization.

a, A heat map of cutaneous mRNA expression of select Th22/IL-22 related genes. Red, greater expression; blue, lower expression. Fold change shown is log base 2 fold-change. p-value shown is a false discovery rate adjusted p value. b, GeneMANIA functional association gene network for Th22-associated genes. The co-expression and physical interaction between genes are expressed as purple and green lines, respectively. Stronger associations are shown with thicker lines. Gene names shown in red are upregulated, while names in black show no significant difference in expression in PN lesional skin compared to healthy controls. c, Gene Set Variation Analyses (GSVA) comparison of immune mediators in lesional PN (n=13) to non-lesional (n=13) and healthy skin (n=13) found lesional PN skin is characterized by significantly increased expression of Th17/Th22-markers, mildly elevated Th1, and no significant difference in expression of Th2 related genes. *p-value < .05, ** p-value < .01, and *** p-value < .001. L, lesional PN samples; NL, non-lesional PN samples; H, matched healthy control samples.