Map of Differential Transcript Expression in the Normal Human Large Intestine
Total Page:16
File Type:pdf, Size:1020Kb
Physiol Genomics 33: 50–64, 2008. First published December 4, 2007; doi:10.1152/physiolgenomics.00185.2006. Map of differential transcript expression in the normal human large intestine Lawrence C. LaPointe,1,2,3 Robert Dunne,2 Glenn S. Brown,3 Daniel L. Worthley,1 Peter L. Molloy,3 David Wattchow,4 and Graeme P. Young1 1Department of Medicine, Flinders University of South Australia, Adelaide, South Australia; 2Preventative Health National Research Flagship, CSIRO Mathematical and Information Sciences, Sydney; 3Preventative Health National Research Flagship, CSIRO Molecular and Health Technologies, Sydney, New South Wales; and 4Department of Surgery, Flinders University of South Australia, Adelaide, South Australia, Australia Submitted 22 August 2006; accepted in final form 27 November 2007 LaPointe LC, Dunne R, Brown GS, Worthley DL, Molloy PL, neoplastic or nonneoplastic setting (6). Epidemiologic studies Wattchow D, Young GP. Map of differential transcript expression in the of colorectal adenocarcinoma suggest support for variable normal human large intestine. Physiol Genomics 33: 50–64, 2008. First incidence, histopathology, and prognosis between proximal published December 4, 2007; doi:10.1152/physiolgenomics.00185.2006.— and distal tumors (8, 9, 18, 19). Thus an understanding of While there is considerable research related to using differential gene location-specific variation could provide valuable insight into expression to predict disease phenotype classification, e.g., neoplastic tissue from nonneoplastic controls, there is little understanding of the those diseases that have characteristic distribution patterns range of expression in normal tissues. Understanding patterns of gene along the colorectum, including colorectal cancer (7, 12, 22). expression in nonneoplastic tissue, including regional anatomic ex- The large intestine is often divided for clinical convenience pression changes within an organ, is vital to understanding gene into six anatomical regions starting from the terminal region of expression changes in diseased tissue. To explore the gene expression the ileum: the cecum, the ascending colon, the transverse change along the proximal-distal axis of the large intestine, we colon, the descending colon, the sigmoid colon, and the rec- analyzed microarray data in 184 normal human specimens using tum. Alternatively, these segments may be grouped to divide univariate and multivariate techniques. We found 219 probe sets that the large intestine into a two-region model comprising the were differentially expressed between the proximal and distal colo- proximal and distal large intestine. The proximal (“right”) rectal regions and 115 probe sets that were differentially expressed between the terminal segments, i.e., the cecum and rectum. We did not region is generally taken to include the cecum, ascending observe any probe sets that were statistically different between any colon, and the transverse colon, while the distal (“left”) region two contiguous colorectal segments. The dominant expression pattern includes the splenic flexure, the descending colon, the sigmoid (65 probe sets) follows a dichotomous expression pattern consistent colon, and the rectum. This division is supported by the distinct with the midgut-hindgut embryonic origins of the gut while a second embryonic ontogenesis of these regions whose junction is pattern (50 probe sets) depicts a gradual change in transcript levels two-thirds along the transverse colon and also by the distinct from the cecum to the rectum. While the dichotomous pattern includes arterial supply to each region. While the proximal large intes- roughly equal numbers of probe sets that are elevated proximally and tine develops from the embryonic midgut and is supplied by distally, nearly all probe sets that show a gradual change demonstrate the superior mesenteric artery, the distal large intestine forms increasing expression levels moving from proximal to distal segments. from the embryonic hindgut and is supplied by the inferior These patterns describe an expression map of individual transcript variation as well as multigene expression patterns along the large mesenteric artery (3). A comprehensive of review of proximal/ intestine. This is the first gene expression map of an entire human distal differences are provided in Ref. 29. organ. The longitudinal nature of the large intestine along the proximal-distal axis provides a relatively unique opportunity colorectal gene expression for constructing a whole organ map of gene expression. Pre- vious research suggests that there is a clear distinction between THE ADVENT OF GENE EXPRESSION profiling has led to an improved the gene expression patterns of proximal colonic tissues and understanding of intestinal mucosa development. For example, distal colorectal tissues (7, 25, 33). While these findings the regulation of transcription factors involved in producing support a broad model of gene expression difference, there and maintaining the radial-axis balance from the crypt base to have been no studies to explore the detailed nature of expres- the lumen and those giving rise to epithelial cell differentiation sion gradients of such genes. Given the interesting embryology is now better understood as a result of microarray gene expres- related to the midgut and hindgut junction near the splenic sion analysis (43, 48). Similarly, understanding of the devel- flexure during embryogenesis, the question is raised: Do dif- opmentally programmed genetic events within the embryonic ferentially expressed genes exhibit an abrupt expression schism gut has improved, especially those molecular control mecha- between the midgut- and hindgut-derived tissues or does ex- nisms responsible for regional epithelium differences between pression follow a gentle gradient along the proximal-distal the small intestine and colon (17, 42). On the other hand, little axis? is known about the proximal-distal gene expression variation To explore this question, this work investigates the gene along the longitudinal axis of the colorectum in either the expression patterns observed along the proximal-distal axis of the large intestine. By exploring these patterns in nonneoplastic tissues we aim to improve understanding of gene expression Article published online before print. See web site for date of publication variation in healthy normal adults without the added complex- (http://physiolgenomics.physiology.org). Address for reprint requests and other correspondence: L. C. LaPointe, 11 ity of neoplasia-related gene expression changes. We have Julius Ave., Riverside Life Sciences Bldg., North Ryde, NSW 2113 Australia built expression profile “maps” that identify individual genes (e-mail: larry.lapointe@flinders.edu.au). whose expression appears to be location dependent, and we 50 1094-8341/08 $8.00 Copyright © 2008 the American Physiological Society GENE EXPRESSION MAP OF THE LARGE INTESTINE 51 have described the nature of multigene expression variance nient volumes and stored Ϫ80°C. Total RNA was extracted from longitudinally along the colon. We apply linear models to these tissue lysates using the Promega SV Total RNA system according to maps to compare the embryology-consistent proximal vs. distal manufacturer’s instructions and integrity was assessed visually by gel two-region model with a more gradual model based on con- electrophoresis. tinuously variable expression between the cecum proximally To measure relative expression of mRNA transcripts, tissue RNA samples were analyzed using Affymetrix HG U133 Plus 2.0 Gene- and rectum distally. Such gene expression maps of the normal Chips (Affymetrix, Santa Clara, CA) according to the manufacturer’s adult colon will provide a foundation for improved understand- protocols (2). Biotin-labeled cRNA was prepared using 5 g (1.0 ing of gene expression variation in both the normal and g/l) total RNA (ϳ1 g mRNA) with the “One-Cycle cDNA” kit diseased state. [incorporating a T7-oligo(dT) primer] and the GeneChip IVT labeling kit. In vitro transcribed cRNA was fragmented (20 g) and analyzed MATERIALS AND METHODS for quality control purposes by spectrophotometry and gel electro- phoresis prior to hybridization. Finally, an hybridization cocktail was Gene Expression Data prepared with 15 g of cRNA (0.5 g/l) and hybridized to HG U133 The data for this study are generated using oligonucleotide mi- Plus 2.0 microarrays for 16 h at 45°C in an Affymetrix Hybridization croarrays hybridized to labeled cRNA synthesized from poly-A Chamber 640. Each cRNA sample was spiked with standard prokary- mRNA transcripts isolated from colorectal tissue specimens. otic hybridization controls for quality monitoring. “Discovery” data set. Gene expression and clinical descriptions for Hybridized microarrays were stained with streptavidin phyco- 184 colorectal tissue specimens were purchased from GeneLogic erythrin and washed with a solution containing biotinylated anti- (Gaithersburg, MD). Individual tissue microarray data were selected streptavidin antibodies using the Affymetrix Fluidics Station 450. with the following characteristics: nonneoplastic colorectal mucosa Finally, the stained and washed microarrays were scanned with the free of nonmucosa contaminating tissue (confirmed by histology) Affymetrix Scanner 3000. from otherwise healthy tissue specimen (i.e., no evidence of inflam- The Affymetrix software package was used to transform raw mation or other disease at specimen site) with an anatomically microarray image files to digitized format.