Abstract
We exploited CITE-seq on over 250,000 progenitors, machine learning, and flow cytometry on millions of Kit+ progenitors to provide linkage between single cell transcriptional clusters, single cell surface protein detection and real-world flow cytometry. First, we generated an InfinityFlow object from millions of progenitors using spectral cytometry. CITE seq ADT values and spectral flow cytometric values were integrated to allow transfer of clusters from single cell RNA seq into the Infinity object. Hematopoietic stem cell and progenitor clusters delineated by the expression of discrete developmental gene expression programs could now be visualized within the flow cytometry data. This enabled the development of automated unsupervised gating strategies for the isolation and functional validation of transcriptionally defined developmental stages of hematopoietic stem and progenitor populations. For example, cells with specific gene expression programs along the developmental trajectory of neutrophil specification and commitment can be isolated to purity with a combination of novel and previously-defined cell surface markers; as evidenced by sorting and recapture for scRNA seq. Moreover, we identified the precursor of the neutrophil-monocytic bipotential progenitor. This multilineage progenitor is transcriptionally reprogrammed by Th2 cytokines (generated by Nippostrongylus brasiliensis infection) to produce the successive waves of neutrophils, basophils and eosinophils critically required to expel worms. Thus, worm infection stimulates Th2 responses that call to the marrow to specify granulocyte production. In summary, this work supports the hypothesis that hematopoiesis follows a step-wise transition through discrete developmentally-relevant gene expression programs, which can now be sorted with fluorescence activated cell sorting and functionally characterized.
Disclosures
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
