Quantitative and molecular differences distinguish adult human medullary and extramedullary haematopoietic stem and progenitor cell landscapes

Abstract

In adults, the bone marrow (BM) is the main site of haematopoietic stem and progenitor cells (HSPCs) maintenance and differentiation. It is known that other anatomical sites can contribute significantly to blood production under stress conditions. However limited tissue availability restricts our knowledge on the cellular, molecular and functional composition of extramedullary HSPC pools in humans at steady state or under stress. Here we describe the landscape of human HSPC differentiation across the three major haematopoietic anatomical sites: BM, spleen and peripheral blood (PB), using matched tissues isolated from the same individuals. Single cell RNA-seq of 30,000 HSPCs and 700 phenotypic haematopoietic stem cells and multipotent progenitors (HSC/MPP) demonstrates significantly different dynamics of haematopoiesis between BM and extramedullary tissues. Lineage-committed progenitors of spleen and PB do not actively divide, whereas BM is the primary site of progenitor proliferation. The balance of differentiation in spleen and PB is skewed towards the lymphoid and erythroid lineages, whereas in BM it is tilted towards megakaryocytic and myeloid progenitors. Extramedullary tissues also harbour a molecularly defined subset of HSC/MPP not found in the BM, which is marked by a specific acto-myosin cytoskeletal signature and transcriptional priming for division and lineage differentiation. Collectively, our findings define a unique cellular and molecular structure of the haematopoietic landscape in extramedullary organs, positioned for rapid lineage-primed demand-adapted haematopoiesis. These data also provide a framework for better understanding of human extramedullary haematopoiesis in health and disease.

The cellular immune response to COVID-19 deciphered by single cell multi-omics across three UK centres

“…2J ). Unlike the bone marrow, which contains rapidly cycling progenitors, the peripheral blood is not thought to constitute a site for haematopoiesis 34 , consistent with the low number of CD34 + cells expressing a cell cycle signature, which was furthermore restricted to genes associated with S-phase ( Fig. 2I ).…”

Single-cell multi-omics analysis of the immune response in COVID-19

“…2j). As peripheral blood is not a site for hematopoiesis 28 , this finding likely reflects COVID-19-mediated pertur bation of normal homeostatic functioning of the bone marrow HSPC compartment.…”

“…Division of Virology, Department of Pathology, University of Cambridge, Addenbrooke's Hospital, Cambridge, UK 27. Department of Biomedicine, University and University Hospital Basel, Basel, Switzerland 28. Botnar Research Centre for Child Health (BRCCH), University Basel & ETH Zurich, Basel, Switzerland 29.…”

Mapping the biogenesis of forward programmed megakaryocytes from induced pluripotent stem cells

“…In-vivo 10x Genomics scRNA-seq integration An in-vivo 10x Genomics scRNA-seq dataset of human haematopoietic stem and progenitor cells from the bone marrow, peripheral blood and spleen of two donors was used to assess similarities between in-vivo haematopoietic intermediates and in-vitro megakaryocyte differentiation using our optimised protocol. We used as starting point the count matrix with raw expression levels and metadata with annotated cell type identities as presented in Mende et al, 2020. Based on inspection of distributions of sequencing depth, number of detected features, proportion of UMIs assigned to mitochondria and ribosomal genes (MT% and RP%, respectively), cells with >1,000 features, <10% MT UMIs and >20% RP UMIs were retained for further analysis, totalling 13,099 cells for donor 1 and 17,350 cells for donor 2.…”

“…It does not tell us, however, whether we generate cells which correspond to their in vivo counterparts. To investigate this, we compared our dataset to a dataset (Mende et al, 2020) containing 30,873 10x Genomics sequenced CD19 -CD34 + human haematopoietic stem and progenitor cells (HSPCs) from the bone marrow, peripheral blood and spleen of two donors. A random forest model was trained on the cell types from the in vivo data, and deployed to predict the cell type identities of the time series data (further details available in the methods section).…”