A Comparative Assessment of Human and Chimpanzee iPSC-derived Cardiomyocytes with Primary Heart Tissues

Abstract

Comparative genomic studies in primates have the potential to reveal the genetic and mechanistic basis for human specific traits. These studies may also help us better understand inter-species phenotypic differences that are clinically relevant. Unfortunately, the obvious limitation on sample collection and experimentation in humans and non-human apes severely restrict our ability to perform dynamic comparative studies in primates. Induced pluripotent stem cells (iPSCs), and their corresponding differentiated cells, may provide a suitable alternative system for dynamic comparative studies. Yet, to effectively use iPSCs and differentiated cells for comparative studies, one must characterize the extent to which these systems faithfully represent biological processes in primary tissues. To do so, we compared gene expression data from primary adult heart tissue and iPSC-derived cardiomyocytes from multiple human and chimpanzee individuals. We determined that gene expression in cultured cardiomyocytes from both human and chimpanzee is most similar to that of adult hearts compared to other adult tissues. Using a comparative framework, we found that 50% of gene regulatory differences between human and chimpanzee hearts are also observed between species in cultured cardiomyocytes; conversely, inter-species regulatory differences seen in cardiomyocytes are found significantly more often in hearts than in other primary tissues. Our work provides a detailed description of the utility and limitation of differentiated cardiomyocytes as a system for comparative functional genomic studies in primates.

Reorganization of 3D Genome Structure May Contribute to Gene Regulatory Evolution in Primates

“…We previously collected RNA sequencing data from the same human and chimpanzee iPSC lines [61]. We jointly analyzed the Hi-C and RNA-sequencing data to learn how often inter-species differences in 3D genomic contact frequencies are associated with inter-species differences in gene expression.…”

“…We considered our Hi-C data along with gene expression data previously collected from the same cell lines [61] and assessed the extent to which inter-species variation in 3D genome contacts could potentially explain gene expression divergence between species. Previous studies have observed that spatial co-expression of genes is associated with chromatin interaction profiles [26,30,31,41,76].…”

“…Previously, the Gilad lab generated RNA-sequencing expression data on the same iPSC lines from this study (GEO accession GSE110471) [61]. Using these data, we filtered out lowly expressed genes based on a log2 RPKM normalization cutoff of 0.4 that produced a similar proportion of retained genes when using a log2 CPM normalization cutoff of 1.5 (11,946 genes versus 11,292 genes, respectively).…”

“…We obtained the overlap between our gene expression data and our Hi-C data by applying bedtools overlap [91] to the Hi-C loci and the first exon of each gene. Using a curated file of orthologous gene coordinates between humans and chimpanzees [61], we extracted a one-basepair interval at the beginning of each first exon to use as a proxy for transcription start sites (TSSs). Given the large size of our Hi-C loci (10 kb), we anticipated that the overlapping contacts would be representative of the genes' TSSs.…”

Gene expression variability in human and chimpanzee populations share common determinants

“…The 93 availability of suitable samples from chimpanzees has notoriously been a limitation for 94 130 Using a linear model framework (see Methods), we identified 8,880 differentially 131 expressed (DE) genes between species (FDR<0.05; TableS2), including 6,409 DE genes with 132 effect sizes smaller than a twofold change (FigS2). As our comparative sample size is unusually 133 large, we have an opportunity to comment on the robustness of observations that are made using 134 study designs with smaller sample sizes, which are more typical for comparative studies in 135 primates (Khan et al, 2013;Pavlovic et al, 2018;Perry et al, 2012 (Göring et al, 2001;Ioannidis, 2008), which particularly affect under-powered 159 study designs (FigS3E). However, even at this small sample size, the false positive rate 160 associated with the classification of DE genes is well calibrated; when we used an FDR of 5% to 161 classify genes as DE between the species, we empirically estimated a median of 2% (1 -5% 162 IQR) false discoveries based on the gold standard reference.…”