Analysis of Archived Human Heart Specimens Uncovers Cardiomyopathy-Specific Genomic Signatures

Direct evaluation of primary human tissue is unparalleled for translational investigation of organ-specific physiology and disease. Aside from readily accessible (e.g. skin) and commonly biopsied (e.g. tumor) tissues, most translational evaluation remains challenging. Regulatory restrictions associated with organ procurement and extensive tissue fixation of autopsy specimens present additional hurdles. Consequently, government-sponsored organ procurement efforts (e.g. National Disease Research Exchange and GTEx) have improved access to difficult-to-obtain human tissues, but still contain relatively few specimens. Alternatively, RNA and DNA amplification is typically required to obtain transcriptomic and epigenomic information from archived specimens. Here we describe a highly optimized, versatile, and straightforward procedure to simultaneously purify RNA and DNA from fresh, fresh frozen, or fixed human heart tissue. To demonstrate the utility of our method, we generated high-quality RNA-seq and ATAC-seq datasets from primary heart tissue derived from rejected donor hearts and autopsy specimens. Next, we obtained high-resolution chromatin accessibility maps from three separate forms of human cardiomyopathy: ischemic, dilated, and hypertrophic. Detailed analysis of epigenomic data uncovered cardiomyopathy-specific genomic signatures. Gene Ontology and motif enrichment analyses identified relevant biological terms and transcription factors to suggest mechanistic differences of each cardiomyopathy subtype. Furthermore, the chromatin accessibility maps also guided retrospective categorization of difficult-to-diagnose cardiomyopathy subtype in additional patients. Additional chromatin accessibility maps unbiasedly classified pre-left ventricular assist devices (LVAD) and remodeled (post-LVAD implantation) hearts from same individual (N=2). Cardiovascular Genome Wide Association Studies (GWAS) variants landing on subtype specific chromatin signatures correlated to disease specific impact on gene regulation. As an effort towards furture diagnostic application, we established a simple ATAC qPCR methodology to detect disease subtypes from limited tissue. Taken together, our new methodology utilizing stable epigenomic features provide a robust pipeline for translational discovery and drive disease etiology using archived specimens over conventional RNA-seq or microarray-based disease prediction approaches.