Ileal Paneth Cell Phenotype is a Cellular Biomarker for Pouch Complications in Ulcerative Colitis

Abstract

Background & aims: Biomarkers that integrate genetic and environmental factors and predict outcome in complex immune diseases such as inflammatory bowel disease (IBD; including Crohn's disease [CD] and ulcerative colitis [UC]) are needed. We showed that morphologic patterns of ileal Paneth cells (Paneth cell phenotype [PCP]; a surrogate for PC function) is one such cellular biomarker for CD. Given the shared features between CD and UC, we hypothesized that PCP is also associated with molecular/genetic features and outcome in UC. Because PC density is highest in the ileum, we further hypothesized that PCP predicts outcome in UC subjects who underwent total colectomy and ileal pouch-anal anastomosis (IPAA).

Methods: Uninflamed ileal resection margins from UC subjects with colectomy and IPAA were used for PCP and transcriptomic analyses. PCP was defined using defensin 5 immunofluorescence. Genotyping was performed using Immunochip. UC transcriptomic and genotype associations of PCP were incorporated with data from CD subjects to identify common IBD-related pathways and genes that regulate PCP.

Results: The prevalence of abnormal ileal PCP was 27%, comparable to that seen in CD. Combined analysis of UC and CD subjects showed that abnormal PCP was associated with transcriptomic pathways of secretory granule maturation and polymorphisms in innate immunity genes. Abnormal ileal PCP at the time of colectomy was also associated with pouch complications including de novo CD in the pouch and time to first episode of pouchitis.

Conclusions: Ileal PCP is biologically and clinically relevant in UC and can be used as a biomarker in IBD.

Keywords: de novo Crohn’s disease; ileal pouch-anal anastomosis; pouchitis.

Graphical Abstract

Abnormal Paneth cell phenotype in ulcerative colitis subjects is associated with poor outcome, genetics, and ileal transcriptomic changes.

Figure 1

Type I Paneth cell phenotype was prevalent in UC subjects with IPAA. [A] Flowchart of UC subjects stratified by clinical outcome after colectomy and IPAA. [B] Diagram of Paneth cell phenotype classification under human defensin 5 immunofluorescence. Each Paneth cell can be classified into normal or one of the five abnormal categories: disordered, diminished, diffuse, excluded, and enlarged. Diagram was prepared in BioRender. [C] Representative photomicrographs of Paneth cells in specimens from UC subjects with Type I and II Paneth cell phenotypes. Type I Paneth cell phenotype was defined as ≥ 20% of total Paneth cells showing abnormal morphology patterns. Type II Paneth cell phenotype was defined as < 20% of total Paneth cells showing morphological defects. Green: human defensin 5; blue: Hoechst stain. Scale bars: 10 µm. Paneth cells are outlined in white. Crypt unit is outlined by dashed line in white. [D] Comparison of Paneth cell phenotype distribution between UC total colectomy and IPAA samples [n = 459] and CD resection samples [n = 420]. Both cohorts showed similar prevalence of type I Paneth cell phenotype [p > 0.05 by Fisher’s exact test]. UC, ulcerative colitis; IPAA, ileal pouch-anal anastomosis; CD, Crohn’s disease.

Figure 2

Transcriptomic themes associated with Paneth cells phenotypes in IBD subjects. [A] Transcriptomic themes that were positively associated with percentages of normal Paneth cells in UC samples included themes associated with secretory granule maturation, spliceosome, ubiquitination, and epigenetic regulations. [B] Overlapping themes between CD and UC that positively correlated with the percentages of normal Paneth cells. IBD, inflammatory bowel disease; UC, ulcerative colitis; CD, Crohn’s disease.

Figure 3

Genetic determinants of Paneth cell phenotype in IBD. [A] Flowchart demonstrating genes from both hypothesis-driven and hypothesis-free approaches were included in analysis. [B] Genes associated with Paneth cell phenotype were involved in the following molecular pathways: innate immunity signalling [autophagosome, inflammasome complex, antigen processing and presentation], Parkinsonism and neural development, aryl hydrocarbon receptor [AhR] signalling, mitochondria function, Rho/Rac cytoskeletal reorganisation, and ulcerative colitis. Analysis was performed using CompBio. IBD, inflammatory bowel disease.

Figure 4

Type I Paneth cell phenotype was associated with higher risk in developing de novo CD and shorter time to first episode of pouchitis. [A] Subjects who developed de novo CD more commonly exhibited Type I Paneth cell phenotype [p = 0.034 by Fisher’s exact test]. [B] Both Paneth cell phenotype and age at surgery were significantly associated with the risk for de novo CD by multivariable logistic regression. OR, odds ratio; CI, confidence interval. [C] Receiver operating characteristic [ROC] curve of the model built with Paneth cell phenotype and age for predicting the development of de novo CD in ileal pouch. AUC: area under the ROC curve. [D] Among UC subjects who did not develop de novo CD, those with Type I Paneth cell phenotype [n = 63] showed shorter time to first episode of pouchitis compared with those with Type II Paneth cell phenotype [n = 213] [p = 0.02 by Kaplan–Meier analysis using log-rank test]. UC, ulcerative colitis; CD, Crohn’s disease.