C. S. Choi1, Y. Imamura Kawasawa3, G. S. Yochum4, L. R. Harris1, S. M. Deiling1, W. A. Koltun1 1Penn State Hershey Medical Center,Division Of Colon & Rectal Surgery, Department Of Surgery, Penn State University College Of Medicine,Hershey, PA, USA 3Penn State University College Of Medicine,Genome Sciences Facility Institute For Personalized Medicine,Hershey, PA, USA 4Penn State University College Of Medicine,Department Of Biochemistry & Molecular Biology,Hershey, PA, USA
Introduction: There is increasing evidence that diverticulitis has a significant genetic predisposition. We sought to identify possible genes involved in diverticulitis by performing RNA-Seq on a group of youthful surgical diverticulitis patients with a positive family history (FD) compared to 1) a group of typical elderly diverticulitis patients (ED) without family history and 2) control patients without evidence of diverticulosis (C).
Methods: The FD group consisted of five surgically managed patients under the age of 50 (mean age 46, 3M/2F) each patient having at least one first-degree relative with diverticulitis. The ED group consisted of five surgically managed patients over age 65 (mean age 72, 3M/2F) without a family history of diverticulitis. Full thickness bowel wall from patients in both FD and ED groups were obtained from both diseased (Ds) and adjacent normal (Nl) sigmoid colon. The C group had single normal samples from five patients (mean age 50, 2M/3F) who required sigmoid resection (endometriosis: 2, polyp/cancer: 3), without diverticulosis. Total RNA was extracted from fresh-frozen tissue specimens, quality confirmed with RNA Integrity Number > 7.0, converted to cDNA and barcoded for next generation sequencing. Pooled libraries were loaded onto an Illumina HiSeq 2500 and run for 50 cycles using the single-read protocol setting. Quality filtered sequencing reads were de-multiplexed and aligned to the human reference genome (hg38) using Tophat. RPKM (Reads Per Kilobase per Million mapped reads) values were calculated after applying GC-content and quantile normalizations using an R package ‘cqn.’ An R package ‘limma’ identified differentially expressed genes between tissue groups based on empirical Bayes moderated t-statistics and associated P-values.
Results: 58 genes were differentially expressed (p<0.015) between FD-Nl vs. ED-Nl and/or C. Based on biological function and eliminating non-coding transcripts and pseudogenes, 3 genes of the same nuclear receptor superfamily (NR4A1, NR4A2, NR4A3) clustered, sharing similar expression profiles. Expression of these genes was significantly decreased in FD-Nl compared to ED-Nl and C groups. In addition, these gene transcripts were reduced in FD-Ds vs. ED-Ds. There was no significant difference between Ds and Nl tissues in either FD or ED.
Conclusion: Expression of NR4A family of nuclear receptors, which are modulators of immunity via the NF-κB pathway and known to be altered in several inflammatory states such as atherosclerosis and psoriasis, was decreased in both diseased and normal sigmoid of FD vs. ED patients. This suggests a possible genetic/molecular mechanism for youthful onset FD.
