Neutrophil Extracellular Traps Induce Human Th17 Cells: Effect of Psoriasis-Associated TRAF3IP2 Genotype

Abstract

Psoriasis lesions are rich in IL-17-producing T cells as well as neutrophils, which release webs of DNA-protein complexes known as neutrophil extracellular traps (NETs). Because we and others have observed increased NETosis in psoriatic lesions, we hypothesized that NETs contribute to increased T helper type 17 (Th17) cells in psoriasis. After stimulating peripheral blood mononuclear cells with anti-CD3/CD28 beads for 7 days, we found significantly higher percentages of CD3⁺CD4⁺IL-17⁺ (Th17) cells in the presence versus absence of NETs, as assessed by flow cytometry, IL-17 ELISA, and IL17A/F and RORC mRNAs. Memory, but not naïve, T cells were competent and monocytes were required for CD3/CD28-mediated Th17 induction, with or without NETs. Th17 induction was enhanced by the T allele of rs33980500 (T/C), a psoriasis risk-associated variant in the TRAF3IP2 gene encoding the D10N variant of Act1, a key mediator of IL-17 signal transduction. Global transcriptome analysis of CD3/CD28-stimulated peripheral blood mononuclear cells by RNA sequencing confirmed the stimulatory effects of NETs, demonstrated NET-induced enhancement of cytokine gene expression, and verified that the effect of Act1 D10N was greater in the presence of NETs. Collectively, these results implicate NETs and the Act1 D10N variant in human Th17 induction from peripheral blood mononuclear cells, with ramifications for immunogenetic studies of psoriasis and other autoimmune diseases.

Figure 1.. NETs promote CD3/CD28 induction of Th17 cells from PBMCs.

Unstimulated PBMCs, and PBMCs activated with anti-CD3/CD28 beads in presence of NETs or LPS-induced NETs, were incubated for 7 days followed by (a) flow cytometry to determine the percentage of IL-17+T-cells; (b) ELISA to determine secreted IL-17A protein; (c) qPCR to determine IL-17A, IL-17F and RORC mRNA levels; and (d) flow cytometry to determine the percentage of CD161+T-cells. Means are indicated by horizontal lines. Statistics: one-way repeated measures ANOVA was performed on transformed data (A and D, logit; B log (ln); C log₂), with all p-values corrected to maintain experimentwise type I error rate for the tested factor. * indicates p < 0.05, ** indicates p < 0.01, *** indicates p < 0.001. Graphical results for the transformed data and a summary of statistically-significant outcomes are presented in Figure S1.

Figure 2.. Induction of Th17 by NETs cannot be fully explained by expansion of T-cells.

(a) Percentage of Th17 cells in the starting population (day 0) compared to 7 days of culture, either in the presence of NETs alone, with CD3/CD28 stimulation alone, or both. NETs were prepared with or without 5 ng/ml LPS, as indicated below the figure. Statistics: Asterisks (*) indicate p = 0.05 by one-way repeated measures ANOVA of untransformed data, with Dunnett’s test for multiple comparisons. (b) Th17 proliferation index. PBMC were labeled at the time of seeding with carboxyfluorescein succinimidyl ester (CSFE) and the number of Th17 cell divisions were measured as a proliferation index after 7 days of culture as described in Methods. Statistics: analysis of untransformed data by one-way ANOVA with Tukey’s correction.

Figure 3.. Monocytes, memory CD4+ T cell and cell-cell contact are required for NET-induced Th17 induction.

(a-c), monocyte depletion experiment. (a) Percentage of IL-17+T-cells in whole PBMC (filled squares) or monocyte-depleted PBMC (open squares), as determined by flow cytometry. (b) Secreted IL-17 protein, as determined by ELISA. (c) IL-17A, IL-17F and RORC mRNA levels, as determined by qPCR. Statistics: two-way repeated measures ANOVA of transformed data (a, logit, b, log (ln), c, log₂) with all p-values corrected to maintain experimentwise type I error rate for the tested factor. * indicates p < 0.05, ** indicates p < 0.01, *** indicates p < 0.001. (d) T-cell depletion experiment. Memory or naïve T-cells were removed from PBMC prior to 7 days of CD3/CD28 stimulation. Closed circles represent whole PBMC, open squares indicate depletion of naïve T-cells, and gray squares indicate depletion of memory T-cells. Statistics: analysis of non-transformed data by two-way ANOVA with Tukey’s test for multiple comparisons, * indicates p < 0.05, ** indicates p < 0.01. (e) Cell-cell contact experiment. Percentage of IL-17+T-cells in whole PBMC, CD4 T cell, CD4 T cell with monocytes or CD4 T cell or monocytes seeded in transwell determined by flow cytometry. Brackets ([ ]) indicate the cell type confined to the transwell. Statistics: analysis of non-transformed data by two-way ANOVA with Tukey’s test for multiple comparisons, * indicates p < 0.05, ** indicates p < 0.01, *** indicates p<0.001.

Figure 4.. The Act1 D10N variant promotes Th17 induction.

Unstimulated PBMCs, PBMCs activated with anti-CD3/CD28 beads in presence of NETs or LPS-induced NETs were incubated for 7 days followed by (a) flow cytometry to determine the percentage of IL-17+T-cells; (b) ELISA to determine secreted IL-17 protein; and (c) qPCR to determine IL17A, IL17F and RORC mRNA levels. Means are indicated by horizontal lines. Statistics: two-way repeated measures ANOVA of transformed data (a, logit, b, log, c, log₂) with all p-values corrected to maintain experimentwise type I error rate for the tested factor. * indicates p < 0.05, ** indicates p < 0.01. Graphical results for the transformed data and a summary of statistically-significant outcomes are presented in Figure S4.

Figure 5.. NETs enhance the effect of the Act1 D10N variant on global gene expression in CD3/CD28-stimulated PBMCs.

(a) Volcano plot showing log 2 fold-change values for global gene expression determined by RNA-seq for 7 day CD3/CD28-stimulated PBMCs in the presence or the absence of NETs. Significant DEGs (|log₂ FC| ≥ 1, p_adj < 0.05) are highlighted in orange. 840 genes were significantly up-regulated and 503 genes were significantly down-regulated. (b) Box and whisker plots showing average gene expression across all expressed genes belonging to the GO term “cytokine activity”, as well as individual gene expression values for IL17A, IL17F, and RORC. Red bars indicate values obtained for Act1 WT homozygotes and blue bars indicate values obtained for Act1 D10N homozygotes. The top and the bottom of the box represent 25^th and 75^th percentiles respectively and the centerline is the 50^th percentile; upper and lower whiskers extend to the 1.5 fold interquartile range of the 25^th and 75^th percentiles, respectively; and dots represent outliers beyond the 1.5-fold interquartile range. (c) Histograms summarizing the distribution of FC values (Act1 D10N/WT) after CD3/CD28 stimulation in the absence of NETs (upper panel) vs. the presence of NETs (lower panel). Individual gene expression ratio values for IL17A, IL17F, and RORC are indicated above the histograms. Note the extended “tail” of positive gene expression ratios in the presence of NETs.