Interleukin-22 regulates neutrophil recruitment in ulcerative colitis and is associated with resistance to ustekinumab therapy

Abstract

The function of interleukin-22 (IL-22) in intestinal barrier homeostasis remains controversial. Here, we map the transcriptional landscape regulated by IL-22 in human colonic epithelial organoids and evaluate the biological, functional and clinical significance of the IL-22 mediated pathways in ulcerative colitis (UC). We show that IL-22 regulated pro-inflammatory pathways are involved in microbial recognition, cancer and immune cell chemotaxis; most prominently those involving CXCR2⁺ neutrophils. IL-22-mediated transcriptional regulation of CXC-family neutrophil-active chemokine expression is highly conserved across species, is dependent on STAT3 signaling, and is functionally and pathologically important in the recruitment of CXCR2⁺ neutrophils into colonic tissue. In UC patients, the magnitude of enrichment of the IL-22 regulated transcripts in colonic biopsies correlates with colonic neutrophil infiltration and is enriched in non-responders to ustekinumab therapy. Our data provide further insights into the biology of IL-22 in human disease and highlight its function in the regulation of pathogenic immune pathways, including neutrophil chemotaxis. The transcriptional networks regulated by IL-22 are functionally and clinically important in UC, impacting patient trajectories and responsiveness to biological intervention.

Fig. 1. Clinical significance of the IL-22 responsive transcriptional network in ulcerative colitis.

a Experimental schema of IL-22 stimulation of colonic organoids (n = 4 biological replicates, IL-22 concentration: 10 ng/ml, duration: 24 h). b Volcano plot demonstrating fold change and false discovery rate (FDR) of differentially expressed genes in human colonoids treated with recombinant IL-22. c Pathway analysis of DEG regulated by IL-22 (top 10, hallmark gene sets as defined in MSigDB, NES: normalized enrichment score). d Expression of the top 50 upregulated transcripts regulated by IL-22 in colonic mucosa separates healthy controls (blue, n = 11), patients with endoscopically inactive UC (green, n = 23) and active UC (red, n = 74) (principal component analysis, reposited dataset: GSE50971). e Enrichment of the IL-22 regulated transcriptional program (gene set variation analysis, gene set: top 50 upregulated genes) in the reposited dataset GSE50971 (healthy control: control, iUC-inactive UC, aUC-active UC, Kruskal–Wallis test, ****p < 0.0001). f Validation of the enrichment for the IL-22 regulated transcriptional program in biopsies taken from healthy controls (control, n = 18) and patients with UC (n = 550) participating in UNIFI trial (Mann–Whitney test, two-sided test, ***p < 0.001). g, h, i Association of the IL-22 enrichment score with clinical outcomes in UNIFI. Clinical remission (defined as a total Mayo score of ≤2 and no subscore >1) and deep remission [which required both clinical remission and mucosal healing defined as histologic improvement (neutrophil infiltration in <5% of crypts, no crypt destruction, and no erosions, ulcerations, or granulation tissue) and endoscopic improvement] at week 8 in UC patients enrolled in the UNIFI clinical trial program stratified according to IL-22 enrichment score in baseline biopsies sampled immediately prior to initiation of ustekinumab (n = 358) or placebo (n = 184). Blue bar shows response rate in placebo-treated UC patients. Gray bar shows response for all patients treated with ustekinumab. Red bars show response for all patients treated with ustekinumab stratified based on the extent of the IL-22 transcriptional program’s activation (ES). Patient numbers and statistical analysis shown in Supplementary Fig. 1. Source data for b, c, d, e, f, h and i are provided as a Source Data file.

Fig. 2. Biological pathways and upstream regulators associated with enrichment of the IL-22 regulated transcriptional program.

a Pathway enrichment analysis (IPA) of the differentially expressed genes in patients enrolled in UNIFI with high enrichment for the IL-22 transcriptional program (right-tailed Fisher’s exact test), b genes belonging to “Migration of cells” pathway that are upregulated in patients with high enrichment of the IL-22 transcriptional program, as identified using IPA downstream analysis. Genes have been further subdivided into functional categories. Symbols denote type of protein, color denotes regulation and color of line denotes relationship, c regulator effects analysis performed on the top 3 predicted upstream activators by IPA in the patients with high enrichment for the IL22 transcriptional program, d normalized expression intensity (to healthy controls, n = 18) of known IL-22 regulators stratified by the IL-22 transcriptional program enrichment (UNIFI, UC n = 550, Kruskal–Wallis test with Dunn multiple comparisons test, *p < 0.05, **p < 0.01, ***p < 0.001). Source data for a and d are provided as a Source Data file.

Fig. 3. Causal network analysis identifies induction of neutrophil-active chemokines as a key biological activity of IL-22 in the colonic epithelium.

a Top 10 pathways enriched in transcriptional changes regulated by IL-22 in human colonoids (n = 4) as identified by IPA, b Circos plot showcasing the shared differentially expressed transcripts regulated by the different cytokines in human colonic organoids (purple lines connecting same genes across DEG lists), c Venn diagrams of shared canonical pathways identified in IPA between IL-22 and other pro-inflammatory cytokines, d clique of neutrophil attracting chemokines regulated by IL-22 identified by protein–protein interaction (PPI) network analysis (STRING), colors depict fold changes [log_e]. e Rulation of transcripts coding for chemokines by IL-22 and other cytokines in human colonoids, f cumulative effect of IL-22 and IL-17A co-treatment in the expression of neutrophil-attracting chemokines (FDR: false discovery rate). Source data for a, e and f are provided as a Source Data file.

Fig. 4. The IL-22 regulated transcriptome correlates with colonic neutrophil accumulation which is associated with resistance to ustekinumab therapy in ulcerative colitis.

a Relative expression of neutrophil attracting chemokines in sigmoid biopsies of UC patients participating to the UNIFI study (n = 550) and non-IBD controls (n = 18) (Mann–Whitney test, two-tailed, no multiple testing correction applied), b relative expression of the neutrophil attracting chemokines in the colonic mucosa of healthy controls (HC), UC patients with inactive and active disease (GSE50971), c non-parametric correlation (Spearman two-tailed) between the enrichment score for the IL-22 transcriptional program and the chemokine gene set (CXCL1, CXCL2, CXCL3, CXCL5, CXCL6, CXCL8), r = 0.54, 95% CI (0.48, 0.60), d IL-22 enrichment scores stratified by the neutrophil subscore of the Geboes histology score in the epithelium and lamina propria in colonic biopsies sampled from the UNITI cohort prior to starting ustekinumab. A higher subscore reflects an increase in neutrophil infiltration (for lamina propria 0: no increase, 1: mild but unequivocal increase, 2: moderate increase, 3: marked increase; for epithelium 0: none, 1: <5% crypts involved, 2: <50% crypts involved, 3: >50% involved) (Kruskal–Wallis test with Dunn’s multiple comparisons test, ****p < 0.0001, ***p < 0.001, line denotes median, n = 550) e bioinformatically computed proportion of neutrophils (Cibersort) in mucosal biopsies of UC patients participating in the UNIFI trial correlates positively (Spearman correlation, two-tailed) with the IL-22 enrichment score (single sample gene set enrichment analysis-ssGSEA), f bioinformatically computed proportion of neutrophils (Cibersort) in mucosal biopsies of UC patients participating in the UNIFI trial stratified by response (mucosal healing, week 8) to ustekinumab and healthy controls (HC, n = 18). UC: UC patients n = 358, R responders, NR non-responders, *p < 0.0001 for NR vs. R and for UC vs. HC, Mann–Whitney test, two-tailed, box & whiskers representing interquartile range and overall range, median value depicted with horizontal line, g non-parametric correlation (Spearman, two-tailed) between the enrichment score for the IL-22 transcriptional program and the enrichment score for the gene set defining UC1 and UC2 populations by Czarnewski et al. (r = 0.79), 95% CI (0.77,0.83), (FDR false discovery rate). Source data for a, c, d–g are provided as a Source Data file.

Fig. 5. IL-22 mediated remodeling of the colonic epithelial transcriptome is conserved across species at gene and pathway level.

a Non-parametric (Spearman, two-tailed) correlation of transcripts regulated by IL-22 in human and mouse colonoids (FC fold change) [r = 0.65, (0.60, 0.71), p < 0.0001], b non-parametric (Spearman, two tailed) correlation of pathways regulated by IL-22 in human and mouse colonoids (z-scores derived from Ingenuity Pathway Analysis- IPA) [r = 0.79, (0.64, 0.88), p < 0.0001], c effects of IL-22, IL-17A and their combination to the regulation of the neutrophil attracting chemokines Cxcl1 and Cxcl5 expression in mouse colonoids, (n = 12, Kruskal–Wallis with Dunn’s multiple comparisons test, **p < 0.01, ****p < 0.0001, median denoted with black line), d effects of IL-22, IL-17A and their combination to the production of the neutrophil attracting chemokines Cxcl1 and Cxcl5 in mouse colonoids (n = 6, paired Wilcoxon, two tailed test, all comparisons against control, *p = 0.31, median denoted with black line), e enrichment of the IL-22 transcriptional program (top 50 upregulated genes by IL-22 in mouse colonoids) in preclinical models of colitis (single sample GSEA), including the T-cell transfer model (induced by adoptive transfer of naive CD4⁺ T cells to Rag2^−/− mice), the spontaneous, microbiota-dependent models of colitis developing in Il10^−/−, and the Tbx21^−/− Rag2^−/− Ulcerative Colitis (TRUC) mice, colitis occurring following administration of DSS administration in drinking water, or rectal administration of DNBS, and innate-immune-mediated colitis occurring in Rag1^−/− mice following administration of agonistic anti-CD40 antibodies (n = 3 biological replicates per group, Mann–Whitney test, one-tailed, for all comparisons *p = 0.05, besides TRUC vs Rag2^−/− *p = 0.03, line representing median), f relative expression of neutrophil attracting chemokines in colonic tissue of different mouse models of colitis. (FDR: false discovery rate). Source data for a–d are provided as a Source Data file.

Fig. 6. IL-22 is a functionally important regulator of neutrophil recruitment in chronic colitis.

a Volcano plot demonstrating fold change and P value of differentially expressed genes in TRUC mice vs controls (Rag2^−/−)(n = 6, biological replicates), b effects of IL-22 neutralization to the expression of neutrophil attracting chemokines Cxcl1 and Cxcl5 (n = 36 biological replicates, Kruskal–Wallis with Dunn’s multiple comparison test for TRUC + anti-IL22 and TRUC IL22^−/− vs TRUC, *p < 0.05, ****p < 0.0001 and Mann–Whitney two tailed test for TRUC IL22^−/− + IL-22 vs. TRUC IL22^−/−, ***p < 0.001, line denotes median), c flow cytometry plot, showing the relative frequency of neutrophils in colonic tissue of TRUC mice treated with a monoclonal antibody blocking IL-22, d absolute number of neutrophils in mouse colonic tissue of TRUC mice and TRUC mice with IL-22 neutralization ((n = 36, biological replicates, Kruskal–Wallis with Dunn’s multiple comparison test for TRUC + anti-IL22 and TRUC IL22^−/− vs TRUC, *p < 0.05, ****p < 0.0001 and Mann–Whitney two tailed test for TRUC IL22^−/− + IL-22 vs. TRUC IL22^−/−, ***p < 0.001, lines denote median)), e effects of IL-22 neutralization on the severity of TRUC colitis (n = 36, biological replicates, Kruskal–Wallis with Dunn’s multiple comparison test for TRUC + anti-IL22 and TRUC IL22^−/− vs TRUC, ****p < 0.0001 and Mann–Whitney two tailed test for TRUC IL22^−/− + IL-22 vs. TRUC IL22^−/−, ***p < 0.001, lines denote median), f schematic representation of the co-culturing experiment using ILC3 isolated from Il22^−/− and control mice and cultured with colonoids, g relative expression of Cxcl1 and Cxcl5 in colonoids co-cultured with ILC3 derived from Il22^+/+ and Il22^−/− mice (n = 3, biological replicates per group, Mann–Whitney, one-tailed test, line denotes median). h Colon mass and i colitis score in TRUC mice treated with anti-CXCR2 (n = 4) or control antibody (n = 14, biological replicates, Mann–Whitney, two-tailed test). Neutralizing anti-IL-22 mAb (clone IL22-01, 200 μg per mouse) were administered intraperitoneally (ip) every 3–4 days. Recombinant IL-22 (rIL-22, 100 μg per mouse) were administered ip at days 0, 4, 8 and 12. Source data for a, b, d, e, g–i are provided as a Source Data file.

Fig. 7. IL-22 mediated induction of neutrophil-active chemokines in colonic epithelial cells is abrogated by JAK inhibition.

a Representative images of immunostaining for IL22RA1, pSTAT3 and MAP3K8 in the colon of a healthy control, UC patient without active inflammation and a UC patients with active inflammation, b, c relative expression of Cxcl1 and Cxcl5 in IL-22 treated colonoids generated from Villin-Cre x Stat3^fl/fl mice (Stat3^ΔIEL), Map3k8^−/− mice and controls, (n = 12 biological replicates, Mann–Whitney test, two-tailed *p < 0.05, **p < 0.01). d Effects of STAT3 genetic disruption in key gene transcripts regulated by IL-22 (GSE15955) e effects of a JAK inhibitor (tofacitinib) in CXCL1 and CXCL5 expression in human colonic organoids (n = 3, biological replicates, ANOVA with Tukey’s multiple comparison test, *p < 0.05, ***p < 0.0001). Source data for b–d and are provided as a Source Data file.