Cell type determination for cardiac differentiation occurs soon after seeding of human induced pluripotent stem cells

Abstract

Cardiac directed differentiation of human induced pluripotent stem cells consistently produces a mixed population of cardiomyocytes and non-cardiac cell types even when using very well-characterized protocols. We wondered whether differentiated cell types might result from intrinsic differences in hiPS cells prior to the onset of differentiation. By associating individual differentiated cells that share a common hiPS cell precursor, we were able to test whether expression variability in differentiated cells was pre-determined from the hiPS cell state. Although within a single experiment, differentiated cells that share an hiPS cell progenitor were more transcriptionally similar to each other than to other cells in the differentiated population, when the same hiPS cells were differentiated in parallel, we did not observe high transcriptional similarity across differentiations. Additionally, we found that substantial cell death occurred during differentiation in a manner that suggested that all cells were equally likely to survive or die, suggesting that there was no intrinsic selection bias for cells descended from particular hiPS cell progenitors. These results led us to wonder about how cells grow out spatially during the directed differentiation process. Labeling cells by their expression of a few canonical cell type marker genes, we showed that cells expressing the same marker tended to occur in patches observable by visual inspection, suggesting that cell type determination across multiple cell types, once initiated, is maintained in a cell-autonomous manner for multiple divisions. Altogether, our results show that while there is substantial heterogeneity in the initial hiPS cell population, that heterogeneity is not responsible for heterogeneous outcomes, and that the window during which cell type specification occurs is likely to begin shortly after the seeding of hiPS cells for differentiation.

ClonoCluster: a method for using clonal origin to inform transcriptome clustering

“…In a prototypic clonal barcoding experiment, a population of cells is transfected with random transcribed barcodes such that each initial clone is likely to express unique barcodes. After proliferation, experimenters apply some additional experimental conditions, such as drug treatment or differentiation (Biddy et al, 2018; Emert et al, 2021; Goyal et al, 2021; Jiang et al, 2021; Weinreb et al, 2020). At the chosen endpoint, one can perform single cell RNA sequencing on a pool of individual cells, which is itself composed of some number of clones marked by barcodes.…”

“…We used ClonoCluster to generate hybrid clusters from six different clone barcoded single cell RNA sequencing datasets from our lab and others (Goyal et al, 2021; Jiang et al, 2021; Weinreb et al, 2020) (see Methods and Table 1). We first performed hybrid clustering at stepwise values of α from traditional transcriptome-only clusters (α = 0) to the clone groupings (α = 1).…”

“…“Paclitaxel-treated breast cancer cells” indicates replicates of a monoclonal line, MDA-MB-231-D4, derived from human breast cancer cell line transfected with a barcode library and treated with 1nM paclitaxel, as previously described (Goyal et al, 2021; Shaffer et al, 2020). “Cardio-directed iPSC” indicates a barcode-transfected induced pluripotent stem cell line treated to drive cells towards a cardiomyocyte fate as previously described (Jiang et al, 2021). Count matrices, barcoding methods, and barcode assignments for these samples were generated as described (Goyal et al, 2021; Jiang et al, 2021).…”

“…“Cardio-directed iPSC” indicates a barcode-transfected induced pluripotent stem cell line treated to drive cells towards a cardiomyocyte fate as previously described (Jiang et al, 2021). Count matrices, barcoding methods, and barcode assignments for these samples were generated as described (Goyal et al, 2021; Jiang et al, 2021).…”

“…Figure2: Manipulation of α reveals turnover of cluster markers A. AUC values for the top marker per cluster at zero, low, high, and maximum (α = 1) values of α for multiple samples, an in vitro murine hematopoiesis assay at day 2, a BRAF inhibitor treated clonal melanoma cell line WM989B, and directed differentiation of induced pluripotent stem cells (iPSCs) towards a cardiomyocyte fate, a system where extrinsic determinants of cell fate are expected be dominant over clone barcodes(Jiang et al, 2021). "Global p" indicates the p-value for the non-parametric Kruskal-Wallis test with Bonferroni correction.…”