Background: Multiple studies have identified interleukin (IL)-1 proteins (IL-1α, IL-1β) as playing a key role in intestinal inflammation, including the expansion of IL-1 expressing cells as a hallmark of severe and anti-TNFα non-responsive inflammatory bowel disease (IBD). However, little is known about the presence of bioactive IL-1 proteins within the cell-free gastrointestinal (GI) mucosal environment or whether IL-1-driven programs affect epithelial regeneration in severe IBD with ulcers.
Methods: We established a highly sensitive functional assay (~500-fold greater than ELISA) to assess bioactive extracellular IL-1 protein from previously cryopreserved GI biopsies. We assessed differential cell-free IL-1 bioactivity in a cohort of Crohn’s disease (CD, n=23), ulcerative colitis (UC, n=18) and non-IBD (n=17) patients, with biopsies from paired inflamed and uninflamed intestinal regions per patient (primarily colon but also ileal). We performed RNA-seq, including weighted gene co-expression network analysis (WGCNA), of the biospecimens’ matched cellular compartments. An ulcer-associated gene signature was interrogated in a single-cell (sc)RNA-seq cohort (n=42) of very early onset (VEO)IBD including monogenic disorders, as well as publicly available IBD bulk and scRNA-seq datasets.
Results: Bioactive intestinal mucosal IL-1α and IL-1β corresponded with disease and ulcer severity in CD and UC. The most extreme signals were seen in select CD patients with deep ulceration. Bioactive IL-1α was the predominant contributor to total IL-1 signal in most patients although several with ulcers displayed an IL-1β-predominant signal. Matched RNA-seq analysis demonstrated enhanced IL-1 transcripts, correlating with IL-1 bioactivity. WGCNA revealed a compelling ulcer-specific module with genes enriched in IL-1 signaling, wound repair processes, as well as transcription factors related to stromal cell differentiation. We validated our findings in several published (e.g., RISK cohort of ileal CD) datasets as well as in unpublished independent adult and pediatric RNA-seq datasets from our group. Deconvolution of the ulcer-specific gene module and projection onto a scRNA-seq cohort of VEOIBD patients (including IBD-causative mutations such as IL-10R deficiency characterized by deep ulceration) implicated specific stromal/myeloid populations and IL-1 driven stromal cell responses in ulcer biology.
Conclusion: Mucosal ulceration in IBD is associated with bioactive IL-1 proteins, and transcriptomic evaluation of the same correlates with bioactive signal. An ulcer-specific gene module, validated in other datasets, sheds light on IL-1 biology in intestinal epithelial repair and stromal cell differentiation. Results strongly suggest the IL-1 signaling pathway as being an attractive and precision-based therapeutic target in subsets of IBD patients.

<b>Figure 1. Extent of IL-1 bioactivity reflects disease and ulcer severity. (A) </b>Workflow schematic. <b>(B</b>-<b>C)</b> Anatomical schematic of biopsy locations from patient colonoscopies. <b>(D</b>-<b>E)</b> Detailed bioactive IL-1 protein activity values of combined inflamed IBD biopsies (D) and individual patient representation (E), highlighting peak values in ulceration, especially in CD. <b>(F)</b> Endoscopy images from n=12 inflamed (+/- ulcers) IBD cohort patients with highest mucosa IL-1 bioactivity depicted in (E), organized according to IL-1 levels ranging from 15-70 (upper panel, deep ulcers), 7-15 (middle panel, shallow ulcers) and 3-7 pg/ml (lower panel, no ulcers). Blue arrows indicate representative ulcerated areas.<b> </b><b>(G)</b> Bioactive IL-1 values in the cohort organized by IL-1β and IL-1α contributions to total signal, demonstrating elevation of both in the setting of ulceration. <b>(H</b>-<b>I) </b>Matched RNAseq analysis showing IL1A and IL1B expression in inflamed IBD.

Figure 1. Extent of IL-1 bioactivity reflects disease and ulcer severity. (A) Workflow schematic. (B-C) Anatomical schematic of biopsy locations from patient colonoscopies. (D-E) Detailed bioactive IL-1 protein activity values of combined inflamed IBD biopsies (D) and individual patient representation (E), highlighting peak values in ulceration, especially in CD. (F) Endoscopy images from n=12 inflamed (+/- ulcers) IBD cohort patients with highest mucosa IL-1 bioactivity depicted in (E), organized according to IL-1 levels ranging from 15-70 (upper panel, deep ulcers), 7-15 (middle panel, shallow ulcers) and 3-7 pg/ml (lower panel, no ulcers). Blue arrows indicate representative ulcerated areas. (G) Bioactive IL-1 values in the cohort organized by IL-1β and IL-1α contributions to total signal, demonstrating elevation of both in the setting of ulceration. (H-I) Matched RNAseq analysis showing IL1A and IL1B expression in inflamed IBD.

<b>Figure 2.</b> <b>Transcriptomic analysis reveals ulcer-specific gene module, cell-specific signatures when projected onto VEOIBD and monogenic scRNAseq cohort. </b><b>(A)</b> WGCNA analysis of biopsy-matched RNA-seq identifies gene modules that track with disease including a module (brown module) correlating specifically with ulcerated colon. <b>(B)</b> Gene set variance analysis of brown module within the RISK cohort validates enrichment of this gene set in pediatric ulcerated Crohn's ileum. <b>(C)</b> Gene set variance analysis of brown module across single cell RNA-seq clusters of a separate cohort consisting of VEOIBD patients (including several with monogenic disorders - featured here is a subset (n=15) of all n=42 patients, including a patient with CTLA4 deficiency, STAT3 gain of function, and IL10 receptor deficiency). A specific subset of brown-module enriched clusters, circled, are depleted in several monogenic patients, a finding that ultimately sheds light on unique ulcer-associated biology.

Figure 2. Transcriptomic analysis reveals ulcer-specific gene module, cell-specific signatures when projected onto VEOIBD and monogenic scRNAseq cohort. (A) WGCNA analysis of biopsy-matched RNA-seq identifies gene modules that track with disease including a module (brown module) correlating specifically with ulcerated colon. (B) Gene set variance analysis of brown module within the RISK cohort validates enrichment of this gene set in pediatric ulcerated Crohn's ileum. (C) Gene set variance analysis of brown module across single cell RNA-seq clusters of a separate cohort consisting of VEOIBD patients (including several with monogenic disorders - featured here is a subset (n=15) of all n=42 patients, including a patient with CTLA4 deficiency, STAT3 gain of function, and IL10 receptor deficiency). A specific subset of brown-module enriched clusters, circled, are depleted in several monogenic patients, a finding that ultimately sheds light on unique ulcer-associated biology.

Jared Barends

Appearances