Gene expression profiles in end-stage human idiopathic dilated cardiomyopathy: altered expression of apoptotic and cytoskeletal genes Christina K. Yung, a, * Victoria L. Halperin, b Gordon F. Tomaselli, b and Raimond L. Winslow a,b a Whitaker Biomedical Engineering Institute, Department of Biomedical Engineering, The Johns Hopkins University School of Medicine & Whiting School of Engineering, Baltimore, MD 21218, USA b Institute of Molecular Cardiobiology, Division of Cardiology, Department of Medicine, The Johns Hopkins University School of Medicine, Baltimore, MD 21205, USA Received 22 April 2003; accepted 8 August 2003 Abstract Dilated cardiomyopathy is now the leading cause of cardiovascular morbidity and mortality. While the molecular basis of this disease remains uncertain, evidence is emerging that gene expression profiles of left ventricular myocardium isolated from failing versus nonfailing patients differ dramatically. In this study, we use high-density oligonucleotide microarrays with f 22,000 probes to characterize differences in the expression profiles further. To facilitate interpretation of experimental data, we evaluate algorithms for normalization of hybridization data and for computation of gene expression indices using a control spike-in data set. We then use these methods to identify statistically significant changes in the expression levels of genes not previously implicated in the molecular phenotype of heart failure. These regulated genes take part in diverse cellular processes, including transcription, apoptosis, sarcomeric and cytoskeletal function, remodeling of the extracellular matrix, membrane transport, and metabolism. D 2003 Elsevier Inc. All rights reserved. Keywords: Heart failure; Dilated cardiomyopathy; Cardiovascular genomics; Gene expression; Oligonucleotide microarray Dilated cardiomyopathy (DCM) is a common cause of congestive heart failure (CHF). Heart failure affects 4.8 million people in the United States and is the leading cause of cardiovascular morbidity and mortality. According to the Framingham Heart Study, approximately one in five people are at risk for CHF [1]. DCM is characterized by left ventricular (LV) dilatation, systolic and often diastolic dysfunction, reduced cardiac output, and significantly in- creased risk of sudden cardiac death [2–6]. Factors produc- ing biomechanical stress such as ischemia, pressure (e.g., hypertension) or volume (valvular regurgitation) overload, myocardial inflammation or infiltration, and inherited muta- tions can initiate dilatation of the ventricle and development of DCM [2–5]. Initial changes in gene expression help to maintain cardiac output and tissue perfusion but ultimately prove to be maladaptive, leading to loss of myocardium via apoptosis and necrosis, adverse ventricular remodeling, interstitial hyperplasia, and progressive reduction of force development and impairment of ventricular relaxation. This in turn increases biomechanical stress for the remaining viable myocardium, leading to progression of the heart failure phenotype [3,4]. Several studies have used cDNA microarrays [7–10] and high-density oligonucleotide microarrays [11–13] to profile global changes of gene expression in human heart failure. These studies have faced a number of challenges including: (a) having limited representation of the genome on the array (ranging from 2000 to 12,000 transcripts), (b) acquiring sufficient number of biological samples (from two to eight patients in each condition sampled in these studies) and technical replicates, (c) whether to pool sam- ples for hybridizations, and (d) application of appropriate normalization and statistical algorithms for inference of regulated genes. In this study, we use the Affymetrix GeneChip HG- U133A oligonucleotide array with >22,000 transcripts to identify genes that are differentially regulated in end-stage idiopathic DCM. Hybridizations are performed in duplicate for every sample. Sample sizes of the control and DCM groups are comparable to the previous studies (six failing 0888-7543/$ - see front matter D 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.ygeno.2003.08.007 * Corresponding author. Fax: +1-410-516-5294. E-mail address: cyung@bme.jhu.edu (C.K. Yung). www.elsevier.com/locate/ygeno Genomics 83 (2004) 281 – 297