Single Cell Resolution of Glucocorticoid Effects on Erythroid Progenitor Cells

To generate the billions of new erythrocytes required on a daily basis, erythroid progenitor cells must exponentially increase in number before undergoing terminal differentiation. A limited number of cell divisions occur during erythropoietin (EPO)-regulated erythroid terminal differentiation, but the principal regulation of erythroid transit-amplification occurs earlier in erythropoiesis between the burst forming unit-erythroid (BFU-E) and colony forming unit-erythroid (CFU-E) stages of development. The importance of this EPO-independent early erythropoietic process is highlighted in Diamond-Blackfan Anemia (DBA). DBA is characterized by pure red cell aplasia, loss of BFU-E and CFU-E progenitors in the bone marrow, and severe anemia despite high circulating EPO levels. The only known effective medical therapy for DBA also provides insight into the regulation of erythroid transit-amplification. In patients responsive to glucocorticoid treatment, there are increased numbers of BFU-E and CFU-E progenitors in the bone marrow, and ex vivo culture studies indicate that the synthetic glucocorticoid dexamethasone (Dex) predominantly increases proliferative capacity of BFU-Es, with minimal effect on proliferative capacity of CFU-Es. These findings led to a prevailing model that glucocorticoids increase BFU-E proliferative capacity by stimulating several rounds of self-renewal cell divisions. However, a limitation of this model is that there is no mechanistic explanation for how BFU-Es regulate the fate choice of undergoing a self-renewal cell division versus a differentiation cell division in the presence or absence of glucocorticoids. In this study, we address this question by examining progression of early erythroid progenitor development at single cell resolution, and subsequently elucidate the true mechanistic nature of glucocorticoid-induced erythroid progenitor proliferative capacity amplification. By performing single cell transcriptome profiling (scRNAseq) of primary-isolated mouse fetal liver BFU-Es, CFU-Es, and their developmental intermediates, we identify a continuum of transcriptomic states during erythroid transit-amplification when performing principle component analysis (PCA) on transcriptomes of individual cells. We show that ex vivo culture of primary-isolated BFU-Es in serum free media supplemented with stem cell factor, insulin-like growth factor 1, and EPO results in developmental progression along the transcriptome continuum when performing scRNAseq and PCA on cultured BFU-Es. The addition of Dex into this culture system does not result in self-renewal of BFU-Es at the transcriptome level, but rather still results in developmental progression, albeit to less of a degree per cell division than BFU-Es cultured without Dex. We additionally show that the continuum of transcriptome states in erythroid transit-amplification is reflective of a continuum of functional states, with developmental progression characterized by decreasing proliferative capacity and decreasing glucocorticoid-responsiveness. Lastly, through manual separation of daughter cells resulting from a BFU-E cell division, we demonstrate that BFU-E cell division is a symmetric process at the transcriptome level, both with and without the addition of Dex. Our results clarify the nature of how glucocorticoids amplify BFU-E proliferative capacity. As opposed to stimulating a finite number of BFU-E self-renewal cell divisions, glucocorticoids decrease the extent of progression through the erythroid transit-amplifying developmental continuum per cell division. Thus, a decreased rate of progression through the developmental continuum is associated with an increased number of transit-amplifying cell divisions prior to terminal differentiation. These findings are important not only for the rational development of glucocorticoid-alternatives for treating DBA, but also for all bone marrow failure syndromes characterized by progenitor cell hypoplasia.