Haematopoietic stem cells (HSCs) and their subsequent progenitors produce blood cells, but the precise nature and kinetics of this production is a contentious issue. In one model, lymphoid and myeloid production branch after the lymphoid-primed multipotent progenitor (LMPP)1, with both branches subsequently producing dendritic cells2. However, this model is based mainly on in vitro clonal assays and population-based tracking in vivo, which could miss in vivo single-cell complexity3,4,5,6,7. Here we avoid these issues by using a new quantitative version of ‘cellular barcoding’8,9,10 to trace the in vivo fate of hundreds of LMPPs and HSCs at the single-cell level. These data demonstrate that LMPPs are highly heterogeneous in the cell types that they produce, separating into combinations of lymphoid-, myeloid- and dendritic-cell-biased producers. Conversely, although we observe a known lineage bias of some HSCs11,12,13,14, most cellular output is derived from a small number of HSCs that each generates all cell types. Crucially, in vivo analysis of the output of sibling cells derived from single LMPPs shows that they often share a similar fate, suggesting that the fate of these progenitors was imprinted. Furthermore, as this imprinting is also observed for dendritic-cell-biased LMPPs, dendritic cells may be considered a distinct lineage on the basis of separate ancestry. These data suggest a ‘graded commitment’ model of haematopoiesis, in which heritable and diverse lineage imprinting occurs earlier than previously thought.
Access optionsAccess options
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
We thank G. Filion, U. Braunschweig, L. Pagie, M. Hauptmann, P. Lio, R. van der Wath and the NKI Genomics Facility for computational assistance; J. van Heijst, J. Rohr and J. Urbanus for useful discussions; http://www.josharris.com for illustrations in Fig. 1a and Supplementary Fig. 1; NKI cytometry and animal facilities; D. Leone, Z. Ping and M. Lodder for technical assistance; and P. Hodgkin, K. Duffy and J. Coquet for critical reading of the manuscript. This work was supported by ERC grant LIFE-HIS-T and HFSP grant RGP0060/2012, S.H.N. was supported by the National Health and Medical Research Council Australia, Marie Curie Incoming International FP6 Fellowship, and the Leukemia and Lymphoma Society; and L.P. was supported by a Marie Curie Intra European FP7 Fellowship and the Bettencourt Schueller Fondation.
PCA of the LMPPs was used to identify groups of progenitors, and resulted in 3 distinct groups with some 'intermediate' LMPPs between them. Backgating of LMPPs (i.e. a manual selection of LMPPs by their 3-dimensional coordinates on the principal components, and visualization of their corresponding LMPPs in the clustered heatmap) was done using in-software tools for Supplementary Fig. 7.
About this article
Application of single-cell RNA sequencing methodologies in understanding haematopoiesis and immunology
Essays In Biochemistry (2019)