Improving Human Cardiac Organoid Design Using Transcriptomics

Cardiovascular disease (CVD) is the leading cause of death worldwide. To this end, human cardiac organoids (hCOs) have been developed for improved organotypic CVD modeling over conventional in vivo animal models. Utilizing human cells, hCOs hold great promise to bridge key gaps in CVD research pertaining to human-specific conditions. hCOs are multicellular 3D models which resemble heart structure and function. Varying hCOs fabrication techniques leads to functional and phenotypic differences. To investigate heterogeneity across hCO platforms, we performed a transcriptomic analysis utilizing bulk RNA-sequencing from four previously published unique hCO studies. We further compared selected hCOs to 2D and 3D hiPSC-derived cardiomyocytes (hiPSC-CMs), as well as fetal and adult human myocardium bulk RNA-sequencing samples. Upon investigation utilizing Principal Component Analysis (PCA), K-means clustering analysis of key genes, and further downstream analyses such as Gene Set Enrichment (GSEA), Gene Set Variation (GSVA), and GO term enrichment, we found that hCO fabrication method influences maturity and cellular heterogeneity across models. Thus, we propose that adjustment of fabrication method will result in an hCO with a defined maturity and transcriptomic profile to facilitate its specified applications, in turn maximizing its modeling potential. Total study utilizes 35 raw FASTQ files from across 7 studies. GSE91383 (n = 3), GSE181397 (n = 3), GSE153185 (n = 9), GSE209997 (n = 6), GSE93841 (n = 4), GSE113871 (n = 3), GSE62913 (n = 7) were reanalyzed for the purposes of this study. Raw FASTQ files were selected from studies of interest, adaptors were trimmed using Trimmomatic, QC'ed, and mapped to hg38 using RNASTAR. FeatureCounts was used to generate counts files, which were normalized and analyzed using the R package "DESeq2".