From UT-Austin: Cilia gene expression patterns in cancer

Cancer Genomics Proteomics. 2014 Jan-Feb;11(1):13-24.

Cilia gene expression patterns in cancer.

Shpak M1, Goldberg MM, Cowperthwaite MC.

Author information

• 1NeuroTexas Institute, St. David's Healthcare, 1015 E, 32nd St. Suite 404, Austin, TX 78705, U.S.A. shpak.max@gmail.com and Marcus Goldberg, Center for Systems and Synthetic Biology, University of Texas at Austin, 2500 Speedway, Austin TX 78712, U.S.A. E-mail: marcusmgoldberg@gmail.com.

Abstract

Non-motile cilia are thought to be important determinants of the progression of many types of cancers. Our goal was to identify patterns of cilia gene dysregulation in eight cancer types (glioblastoma multiforme, colon adenocarcinoma, breast adenocarcinoma, kidney renal clear cell carcinoma, lungsquamous cell carcinoma, lung adenocarcinoma, rectal adenocarcinoma, and ovarian cancer) profiled by The Cancer Genome Atlas. Among these types, 2.5-19.8% of cilia-associated genes were significantly differentially expressed (versus 5.5-32.4% dysregulation across all genes). In four cancertypes (breast adenocarcinoma, colon adenocarcinoma, glioblastoma multiforme, and ovarian cancer), cilia genes were on average down-regulated (median fold change from -1.53--0.3), in the other four types, cilia genes were up-regulated (fold change=0.86-3.5). Pairwise comparisons betweencancer types revealed varying degrees of similarity in the differentially expressed cilia genes, ranging from 7.1% (ovarian cancer and lung squamous cell carcinoma) to 65.8% (ovarian cancer and rectal adenocarcinoma). Hierarchical clustering and principal components analysis of gene expression identified glioblastoma multiforme, colon adenocarcinoma, breast adenocarcinoma; and kidney renal clear cell carcinoma, lung squamous cell carcinoma, lung adenocarcinoma, rectal adenocarcinoma, and ovarian cancer as sub-classes with similar dysregulation patterns. Our study suggests that genes involved in cilia biosynthesis and function are frequently dysregulated in cancer, and may be useful for identifying and classifying cancertypes.