Single-cell transcriptome in chronic myeloid leukemia: pseudotime analysis reveals evidence of embryonic and transitional stem cell states

Recent experimental data suggest that the heterogeneity of chronic myeloid leukemia (CML) stem cells may be the result of the development of unique molecular events generating functional consequences in terms of the resistance and persistence of leukemic stem cells. To explore this phenomenon, we designed a single-cell transcriptome assay evaluating simultaneously the expression of 87 genes. Highly purified CD34+ cells from three CML patients at diagnosis were immobilized in microfluidic chips, and the expression of 87 genes was evaluated in each cell. This analysis identified a group of 13 highly connected genes including NANOG, POU5F1, LIN28A, and SOX2, representing on average 8.59% of the cell population analyzed. Bioinformatics analysis with the corrected matrix and t-distributed stochastic neighbor embedding (tSNE) algorithm identified four distinct clusters, and the pseudotime analysis confirmed the presence of seven stem cell states in the four clusters identified. ALOX5 expression was associated with the group of cells expressing the pluripotency markers. In in vitro analyses, two genes that were predicted to undergo similar regulation using pseudotime analysis (ALOX5 and FGFR) were found to be similarly inhibited by ponatinib, an FGFR inhibitor. Finally, in an independent cohort of CML patients, we found that pluripotency gene expression is a common feature of CD34+ CML cells at diagnosis. Overall, these experiments allowed identification of individual CD34+ cells expressing high levels of pluripotency genes at diagnosis, in which a continuum of transitional states were identified using pseudotime analysis. These results suggest that leukemic stem cell persistence in CML needs to be targeted simultaneously rather than using a single pathway. (C) 2020 ISEH - Society for Hematology and Stem Cells. Published by Elsevier Inc. All rights reserved.