From Hematopoietic Stem Cells to Platelets: Unifying Differentiation Pathways Identified by Lineage Tracing Mouse Models

Abstract

Platelets are the terminal progeny of megakaryocytes, primarily produced in the bone marrow, and play critical roles in blood homeostasis, clotting, and wound healing. Traditionally, megakaryocytes and platelets are thought to arise from multipotent hematopoietic stem cells (HSCs) via multiple discrete progenitor populations with successive, lineage-restricting differentiation steps. However, this view has recently been challenged by studies suggesting that (1) some HSC clones are biased and/or restricted to the platelet lineage, (2) not all platelet generation follows the "canonical" megakaryocytic differentiation path of hematopoiesis, and (3) platelet output is the default program of steady-state hematopoiesis. Here, we specifically investigate the evidence that in vivo lineage tracing studies provide for the route(s) of platelet generation and investigate the involvement of various intermediate progenitor cell populations. We further identify the challenges that need to be overcome that are required to determine the presence, role, and kinetics of these possible alternate pathways.

Keywords: hematopoietic stem cell (HSC); lineage tracing; megakaryocyte progenitor (MkP); megakaryopoiesis; platelet; thrombopoiesis; transplantation.

Figure 1

Classical hematopoietic tree. Self-renewing, multipotent HSCs reside at the apex of the hematopoietic hierarchy. The differentiation to MPPs results in the loss of self-renewal yet maintains multipotency. Successive differentiation then occurs, with downstream progenitor pools becoming progressively more lineage-restricted. Classically, platelets arise by the differentiation of MPPs into CMPs, MEPs, and MkPs, which mature into megakaryocytes that ultimately generate platelets.

Figure 2

Routes of platelet generation revealed by lineage tracing. Two major lineage tracing methods have been primarily employed to interrogate the route(s) of platelet specification. (A) Single and bulk cell transplantation and (B) in situ labeling have suggested the possibility of multiple alternative paths of megakaryopoiesis involving the differential use of progenitor cell states. Solid lines indicate “classical” paths, whereas dashed lines represent new and/or expanded differentiation steps elucidated by the studies discussed here. Other cell lineages are omitted for visual clarity.

Figure 3

Different experimental strategies require unique interpretations. The experimental approach utilized in lineage tracing studies (transplantation or in situ labeling) necessitates specific interpretation of results and understanding of potential caveats. Single HSC transplants allow an assessment of cellular output from an individual HSC clone, yet readout may be underestimated due to the potential for relatively low-donor-derived chimerism that could be below the method of detection employed. Conversely, bulk HSC transplants significantly improve the detection of donor-derived cells, yet the output of individual HSCs is not possible to assess but is rather the average response of all HSCs transplanted. In situ label induction that uniquely labels each HSC clone allows for the simultaneous assessment of each clone yet may suffer from the same limitations as single-cell HSC transplants (i.e., underestimation of individual HSC contribution due to limits of detection). Similar to bulk HSC transplants, genetic labeling strategies label only a subset of heterogenic HSCs whose measured output is the average of all labeled cells. Both transplantation and in situ labeling also have the potential to skew resulting data and interpretation due to the cell surface phenotype or type of label induction employed. For example, fluorescent genetic reporters may be detectable in platelets, but genetic barcoding approaches are undetectable due to lack of genetic material. If only some heterogeneous HSC clones are assessed, then results can only be understood based on the phenotypic or transcriptomic profile used experimentally [20,68,70,80].

Figure 4

Proposed unification of adult steady-state platelet generation from HSCs, as determined by lineage tracing. Combining the available lineage tracing data, we propose an expanded and unified view of megakaryopoiesis. The phenotypic HSC pool comprises heterogeneous populations likely to be ordered into various sub-hierarchies and may also possess varying degrees of lineage priming, bias, and/or restriction. HSCs then transition to MPPs, including MPP2, which may be a primary subset involved in platelet formation. Importantly, the transition out of the LT-HSC cell state must be accompanied by gene expression of Flk2, which is incompatible with a direct HSC-to-platelet path. The “classical” CMP > MEP > MkP differentiation progression may then occur, or specific myeloid progenitor cell states may be bypassed. All possible pathways converge upon the obligate MkP cell state, the maturation of which into megakaryocytes results in eventual platelet production.

Figure 5

The biological significance of seemingly redundant platelet specification pathways. The collective findings of the studies reviewed herein reveal continual and consistent megakaryocyte lineage generation and platelet production by the hematopoietic system (see also Figure 4). We posit that platelet and erythrocyte production are the default fates of hematopoiesis, with the many shared intermediate progenitor cell states acquiring and/or shifting their differentiation potential to other specific lineages as physiological demand requires. The biological significance of such parallel and redundant paths is to ensure hemostasis and temper the effects of perturbation with respect to platelet output.