P063 Representative and comprehensive analysis of colonic and ileac biopsies from IBD patients by gene expression profiling using the straightforward, fast, and affordable novel application Whole Transcriptome AmpliSeq on the Ion Torrent NGS platform

• Published in: Journal of Crohn's and colitis Vol. 13; no. Supplement_1; pp. S117 - S118
• Main Authors: Raulf, F, Roth, L, Delucis-Bronn, C, Begrich, A, Wieczorek, G, Picard, D, Rush, J, Beglinger, C, Hruz, P
• Format: Journal Article
• Language: English
• Published: US Oxford University Press 25-01-2019
• ISSN: 1873-9946; 1876-4479
• DOI: 10.1093/ecco-jcc/jjy222.187

Abstract
Background
Whole transcriptome (WT) AmpliSeq analysis of 20803 genes offers significant advantages compared both to hybridisation-based genechip analysis (specificity and sensitivity), and to RNA-Seq (costs and reproducibility), while data show good correlation to both platforms. We applied WT AmpliSeq for gene expression profiling (GEP) of colonic and ileac biopsies from patients with active ulcerative colitis (UC) and Crohn’s disease (CD) to create a new representative dataset fostering disease understanding, and helping to prioritise new targets and biomarkers.
Methods
Mucosal biopsies from inflamed and non-inflamed areas of patients with CD and UC as well as control subjects were immediately immersed in RNAlater. RNA was extracted by RNeasy with DNase digestion (Qiagen), quantified by UV spectrophotometry, and quality controlled by Bioanalyzer (Agilent). 10 ng total RNA were subjected to ultrahigh-multiplexed RT-PCR using the AmpliSeq Transcriptome Human Gene Expression kit (Thermo Fisher). Sixteen to 18 barcoded samples were sequenced on a 540 chip by an Ion GeneStudio S5 System (Thermo Fisher). The resulting Excel matrix of normalised reads [RPM = reads per million mapped reads] was analysed by GeneSpring (Agilent) and QOE (Qlucore).
Results
172 biopsies were analysed from 36 CD, 26 UC, and 6 non-IBD patients. Biopsies from inflamed and non-inflamed areas were included for each IBD patient. Biopsies yielded total RNAs of varying quality (RIN 3.4–9.9). Due to small amplicon design, no differences could be noted between extremes. Unbiased PCA analysis revealed no gender bias and no effect of age. UC samples separated reasonably well into involved and non-involved biopsies while CD samples were much more mixed, reflecting possibly a more systemic instead of localised inflammation.
Gene signatures of active inflamed CD and UC (few exclusive genes), and a large common overlap in comparison to normal controls could be identified, in line with previous findings by genechips on a lack of major differences between CD and UC inflamed biopsy transcriptomes. Statistically significant differentially expressed genes between UC and CD with highest levels in inflamed UC could be identified, eg, the T-cell-specific CD3D (IHC confirmed), the plasma cell marker SDC1 (CD138), and the chemokine CXCL13, the B cell-attracting chemokine 1.
Conclusions
WT AmpliSeq can perform GEP of large numbers of mucosal biopsies enabling further research of IBD pathomechanisms and of new biomarker signatures for better patient stratification.