Identification of Beta-Blocker Responsive Genes in Human Heart Failure with a Multi-Comparison Pharmacogenomic Analysis

Abstract

Background: Studies with beta-blockers (BB) in heart failure have consistently shown improved myocardial function, but the molecular mechanisms of this improvement remains unclear. The purpose of this study was to identify changes in myocardial gene expression associated with BB use in humans. Methods: We performed microarray analysis on high-quality left ventricular specimens obtained at transplantation from 14 nonfailing hearts (NF), 112 failing hearts not on BB (HF-NBB), and 51 failing hearts on BB (HF-BB). Recognizing that not all beta-blocker associated changes represent normalization, we developed a new analytical strategy in which three separate comparisons are generated. Results: Conventional two-way analysis demonstrated that there were 3140 genes with dysregulation in HF-NBB compared with NF. Similarly, there were 531 genes with differential regulation between HF-NBB and HF-BB. However, the multi-comparison analysis demonstrated that, only 70 genes exhibited a true recovery pattern. Of the genes exhibiting recovery pattern, none had a greater than two-fold-change in expression, and most have a less than 1.5-fold difference in mean transcript abundance. We also employed a separate multi-comparison analysis to explore which genes might exhibit an expression pattern suggestive of persistent pathologic dysregulation. This analysis demonstrates that genes showing a pattern of persistent dysregulation are almost 35 times more abundant than genes showing a recovery despite beta-blocker therapy. When we assessed the effect of heart failure etiology on BB-induced transcriptional changes, multi-comparison analysis revealed about twice as many genes showing recovery in ischemic cardiomyopathy (ICM) group receiving BB compared to dilated cardiomyopathy (DCM). Our analysis also demonstrated that the proportion of genes showing a recovery or persistence pattern varies in a pathway-specific fashion. For example, several sarcomeric proteins and Ca2+ handling proteins appear to normalize with BB therapy, while genes associated with carbohydrate and lipid metabolism show a persistent dysregulation despite BB treatment. Conclusion: These studies provide new insights into the pharmacologic actions of beta blocker therapy and the biology of myocardial recovery. These studies may also help define where post-transcriptional changes may be most important. This approach may also help identify which pathways require adjuvant interventions to promote more comprehensive myocardial recovery with beta-blockers.