Transcriptional profiling of intestinal biopsies has been performed on patients with ulcerative colitis to better understand the gene expression profile underlying the active disease state and guide therapeutic advances. In prior studies, variability introduced by the extent of inflammation, anatomic site of biopsy, and medications were not explored. We sought to more globally understand the variability of colonic tissue from patients with ulcerative colitis and guide future clinical study design in regards to tissue gene expression.Methods:
We performed transcriptional profiling on 13 subjects, including pediatric and adult patients with ulcerative colitis from 2 hospital sites. We collected 6 biopsies within a single anatomic area of macroscopically inflamed tissue, and 4 biopsies from macroscopically normal appearing tissue. Isolated RNA was used for gene expression analysis utilizing Affymetrix Human Primeview microarrays. Genes were called as significant with a corrected P-value < 0.05 and fold-change ≥±2. Ingenuity Pathway Analysis was used to assess over-representation of gene ontology and biological pathways within the data. Canonical pathways, molecular networks, and upstream regulators were assessed for significant enrichment and directionality of effect utilizing a P-value < 0.05 (right tailed Fisher's exact test) and z-score > ±2. Additionally, mRNAseq was also performed on a subset of the same biopsies and differential analysis was assessed utilizing the EdgeR algorithm to compare affected versus unaffected samples and the overlap between these 2 methods. Last, we modeled the minimum number of biopsies required to reliably detect gene expression across different subject numbers utilizing the Primeview data.Results:
Transcriptional profiles of samples collected co-clustered independently of the hospital collection site, age, sex, or colonic location, and parallels prior gene expression findings. A small set of genes not previously described where identified and contains interesting hits for future studies. Ingenuity Pathway Analysis revealed that top canonical pathways were granulocyte and agranulocyte adhesion and diapedesis. Top molecular networks most significantly enriched were cellular movement of the hematological system, closely followed by the humoral immune response and cellular movement of the immune system. Top predicted upstream regulators included: lipopolysaccharide, tumor necrosis factor, and transforming growth factor beta 1. Transcriptional profiling by mRNAseq overlapped significantly with the microarrays, revealing clear separation by disease status. Our modeling analysis reveals the minimal number of biopsies and patients per cohort to yield reliable and reproducible results in smaller clinical studies where reliable gene expression analysis is desired (with 3 paired biopsies—inflamed and uninflamed, only 4 subjects are required to reach statistical significance for 2000 probes; with 2 paired biopsies, 6 subjects are required; with 1 paired biopsy, 12 individuals are required).Conclusions:
Key findings include concordance with previously published gene expression studies, further expansion of potential new disease related genes, and similarity among UC subjects in different age groups. We also establish for the first time to our knowledge a guide to systematically and reliably identifying numbers of biopsies and subjects to enable dependable expression profiling to guide future work in randomized control studies on patients with ulcerative colitis.