The cause of chronic inflammation manifestations such as inflammatory bowel disease (IBD) is still not yet fully understood. Evidence however points to the convergence of environmental factors, genetic factors, intestinal microbiota and the immune response at the intestinal epithelial surface that consequently leads to a breakdown of barrier function in IBD. An increasing number of theories implicate a dysfunctional intestinal epithelial surface in the pathogenesis of IBD and progression to complicated disease, which contributes significantly to the IBD burden. As yet, the significance of the intestinal epithelial surface in IBD manifestation and progression has not been explored using a systems biology approach, that compares the mucosa of healthy patients to those with IBD. In the context of dyregulation of the epithelial barrier, there is a scarcity of publications that explore the contribution of proteins- the effectors of the cells and RNA message in the pathways affected by CD and in particular the localization of inflammatory response. We aim to evaluate the functional pathways of chronic inflammation at the intestinal epithelial surface using a combination of transcriptomic and proteomic analyses on intestinal epithelial tissue samples from paired inflamed and non-inflammed Crohn’s disease (CD) and healthy control subjects. Concordance of several biological pathways from both data sets was found to be altered in CD patients’ epithelia when compared to healthy controls. This information could be helpful in identifying novel therapeutic targets that aim to restore barrier function at the intestinal epithelial surface and to guide therapy.

Collection of samples

Pinch biopsies from Crohns disease patients undergoing colonoscopy at Concord Repatriation General Hospital were collected from inflamed Terminal ileum portion of bowel and the closest macroscopically normal mucosa section. Two pinches were collected from both the inflamed (I) and non-inflamed (NI) portions of bowel. Pinch biopsies were immediately placed in 150μl of RNAlater (ThermoFisher Scientific, Melbourne, Australia) and stored at -80°C until use. Bowel was graded as either I or NI by the gastroenterologist performing the colonoscopy based on the SES-CD. In addition, samples were also collected from healthy control patients undergoing colonoscopy for unrelated conditions.

Extraction of total RNA

7–20mg of frozen bowel tissue was added to a chilled and washed 2ml Eppendorf tube. Lysis buffer from the Qiagen RNAeasy Mini Kit (Melbourne, Victoria, Australia) was added along with 6mg of 1.0mm zirconium beads previously washed with IPGphor detergent and nitric acid, and the tissue was homogenised for 40 seconds in a MiniBeadbeater (Biospec Products, Bartlesville, USA) then rested for 2 minutes, for 8 cycles carried out in a cold room at 4°C. Tubes were spun for 3 minutes at 1500 x g. The lysate was transferred to a new RNA/DNA free 1.5ml microcentrifuge tube. Total RNA was extracted using the Qiagen RNAeasy Mini Kit according to the manufacturer’s instructions. RNA was eluted with nuclease-free water supplies with the kit. The quality of the RNA was checked by testing 260/280 ratio was >1.8, 260/230 ratio was between 1.8–2.2 and the RNA integrity number was >8. Patient samples consisted of 5 non-inflamed controls and 44 inflamed Crohn's disease terminal ileum (TI) tissue samples biopsied from an inflamed and non-inflamed region of the TI patient paired.

Analysis of total RNA and RNA-Seq bioinformatics analysis

Analysis of total RNA from 15 samples was performed by the Ramaciotti Centre for Genomics using the NovaSeq platform to obtain 100 bp paired end reads. Counts were quantified using salmon (v 0.9.1) and the R package txtimport using the Ensembl annotation file GRCh38.87.fa. A total of 17,556 genes remained after filtering out lowly expressed genes. Differential expression analysis was performed using the edgeR and limma (voom) R packages and the treat function to select for genes with a logFC of at least log2(1.2).

Analysis of CD tissue proteome

Global proteomics was carried out on protein extracted from 7–20mg of frozen bowel tissue in lysis buffer (0.1M DTT, 0.1M CHAPS, 9M Urea, 3mM Tris) with protease inhibitor and 6mg of 1.0mm zirconium beads using a cell shearing method. One hundred μg of protein was digested in a 1:100 ratio of protein to trypsin at room temperature overnight.

Proteomic Mass spectrometry was carried out using a QExactive Plus (Thermo Electron, Bremen, Germany). Briefly, the sample, 1.5 μg (2.0 μL from 10μL), was loaded onto a micro C18 pre-column (300 μm × 5 mm, Dionex) with H₂O:CH₃CN (98:2, 0.1% TFA) at 10 μL min^-1. After washing, the pre-column was switched (Valco 10 port valve, Dionex) into line with a fritless nanocolumn (75 μm i.d × 20 cm) containing reverse phase C18 media (1.9 μm, 120 A, Dr. Maisch HPLC GmbH). Peptides were eluted using a linear gradient of H₂O:CH₃CN (98:2, 0.1% formic acid) to H₂O:CH₃CN (64:36, 0.1% formic acid) at 250 nL min−1 over 60 min. The QExactive mass spectrometer was run in DDA mode where a high voltage of 2000 V was applied to a low volume union and the column (45 ◦C) positioned 0.5 cm from the heated capillary (275°C). A survey scan 350–1750 m/z was acquired in the Orbitrap (resolution 70,000 at 200 m/z) with an accumulation target of 106 ions, lock mass enabled and up to the 10 most abundant ions (AGC target set to 105, minimum AGC target set to 1.5 × 104) with charge states ≥ +2 and ≤ +6 sequentially isolated and fragmented.