A systematic analysis of heart transcriptome highlights divergent cardiovascular disease pathways between animal models and humans

Abstract

Animal models have been extensively used in the study of cardiovascular disease (CVD) and have provided important insights into disease pathogenesis and drug development. However, the level of conservation of gene expression patterns of the orthologous genes between human and animal models was unclear. To address this issue, we compared the expression of orthologous genes in human and four models (rhesus, rat, mouse and dog), based on 42 normal heart samples with high quality gene expression data. The results show that the global expression profiles between animal model and human orthologous genes are highly preserved. The phylogenetic tree inferred from the gene expression profiles has similar topology to that of the species tree. However, differentially expressed genes (DEGs) between human and each model were identified and these four gene datasets are enriched with different molecular functions, including hormone–receptor binding and geranyl transferase activity. The 65 overlapped DEGs between four sets are involved in thyroid cancer, proteasome systems, aminoacyl-tRNA biosynthesis and GST (Glycine, Serine and Threonine) metabolism, of which functions are divergent between models and humans. In addition, 46.2% (30/65) of the communal genes have been experimentally proven to be associated with cardiovascular disease. Next, we constructed a co-expression network based on intra- and inter-species variation, to elucidate the altered network organization. It indicates that these DEGs evolved as modules rather than independently. The integrated heart transcriptome data should provide a valuable resource for the in-depth understanding of cardiology and the development of cardiovascular disease models.