Cell Lineage Affiliation During Hematopoiesis

Abstract

By the mid-1960s, hematopoietic stem cells (HSCs) were well described. They generate perhaps the most complex array of functionally mature cells in an adult organism. HSCs and their descendants have been studied extensively, and findings have provided principles that have been applied to the development of many cell systems. However, there are uncertainties about the process of HSC development. They center around when and how HSCs become affiliated with a single-cell lineage. A longstanding view is that this occurs late in development and stepwise via a series of committed oligopotent progenitor cells, which eventually give rise to unipotent progenitors. A very different view is that lineage affiliation can occur as early as within HSCs, and the development of these cells to a mature end cell is then a continuous process. A key consideration is the extent to which lineage-affiliated HSCs self-renew to make a major contribution to hematopoiesis. This review examines the above aspects in relation to our understanding of hematopoiesis.

Full text

Academic Editor: Irmgard Tegeder
Received: 24 February 2025
Revised: 30 March 2025
Accepted: 2 April 2025
Published: 3 April 2025
Citation: Brown, G. Cell Lineage
Affiliation During Hematopoiesis. Int.
J. Mol. Sci. 2025,26, 3346. https://
doi.org/10.3390/ijms26073346
Copyright: © 2025 by the author.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license
(https://creativecommons.org/
licenses/by/4.0/).
Review
Cell Lineage Affiliation During Hematopoiesis
Geoffrey Brown
Department of Biomedical Sciences, School of Infection, Inflammation, and Immunology, College of Medicine and
Health, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK; g.brown@bham.ac.uk;
Tel.: +44-(0)121-414-4082
Abstract: By the mid-1960s, hematopoietic stem cells (HSCs) were well described. They
generate perhaps the most complex array of functionally mature cells in an adult organism.
HSCs and their descendants have been studied extensively, and findings have provided
principles that have been applied to the development of many cell systems. However, there
are uncertainties about the process of HSC development. They center around when and
how HSCs become affiliated with a single-cell lineage. A longstanding view is that this
occurs late in development and stepwise via a series of committed oligopotent progenitor
cells, which eventually give rise to unipotent progenitors. A very different view is that
lineage affiliation can occur as early as within HSCs, and the development of these cells to
a mature end cell is then a continuous process. A key consideration is the extent to which
lineage-affiliated HSCs self-renew to make a major contribution to hematopoiesis. This
review examines the above aspects in relation to our understanding of hematopoiesis.
Keywords: hematopoiesis; hematopoietic stem cells; lineage affiliation; differentiation;
self-renewal
1. Introduction
The stem cell concept dates to the 19th century and HSCs and their descendants have
been extensively studied due to their ease of accessibility. Hematopoietic stem cells (HSCs),
which are rare cells in the bone marrow, ensure that blood and immune cells are produced
each day throughout life to replenish worn-out or short-lived mature cells. For example,
around 10
11
granulocytes are produced each day in humans [
1
]. The ‘feed’ from HSCs into
hematopoiesis is low due to their scarcity, which is coupled with a low level of amplification
(see below). To meet the large and continuous demand for mature cells, HSCs self-renew
giving rise to identical daughter cells that have not differentiated so that there is a supply
of HSCs for the lifespan of an organism.
For many years, hematopoiesis has been seen as a three-tier system with each com-
partment comprising cells with different attributes, namely self-renewal, the capacity to
expand cell numbers from the use of a single HSC, and end-cell functionality (Figure 1).
Self-renewing HSCs are the first tier. Offspring that are selected for differentiation give
rise to the second tier of cells, which serves to amplify the production of mature cells by
means of two cell populations. Hematopoietic progenitor cells (HPCs) divide a consider-
able number of times (perhaps 20 to 30 cycles); as few as twenty cycles would result in an
amplification of 10
6
times. As these cells divide, they are not thought to give rise to identical
offspring and instead age stepwise, via a series of progenitor cells with differing lineage
potentials, to eventually give rise to cells that are committed to a single lineage, namely
unipotent HPCs. In keeping HPCs, which includes cells that are multipotent progenitors
(MPPs), do not engraft a mouse to sustain hematopoiesis long-term. Uni-potent HPCs give
Int. J. Mol. Sci. 2025,26, 3346 https://doi.org/10.3390/ijms26073346

Int. J. Mol. Sci. 2025,26, 3346 2 of 15
rise to the morphologically recognizable precursors, for example, myeloblasts, which are
also committed to generating their blood cell type. Myeloblasts divide three to five times to
yield the non-dividing metamyelocytes that give rise to mature neutrophils. The third tier
consists of functionally mature cells, which do not typically divide, other than lymphocytes.
Int. J. Mol. Sci. 2025, 26, x FOR PEER REVIEW 2 of 15
namely unipotent HPCs. In keeping HPCs, which includes cells that are multipotent pro-
genitors (MPPs), do not engraft a mouse to sustain hematopoiesis long-term. Uni-potent
HPCs give rise to the morphologically recognizable precursors, for example, myeloblasts,
which are also commied to generating their blood cell type. Myeloblasts divide three to
five times to yield the non-dividing metamyelocytes that give rise to mature neutrophils.
The third tier consists of functionally mature cells, which do not typically divide, other
than lymphocytes.
Figure 1. Compartmentalization of hematopoiesis. Self-renewing hematopoietic stem cells (HSCs)
give rise to the second tier of cells, which amplifies the number of cells produced by an HSC. This
tier includes the hematopoietic progenitor cells (HPCs) and the recognizable precursor cells to each
cell type. HPCs divide a substantial number of times to give rise ultimately to the unipotent HPCs.
Precursor cells divide to a lesser extent as they mature to give rise to the third tier of functional end
cells. Expansion is shown by the triangle with the lines providing an indication of the number of
cell cycles. Meg, megakaryocytes; Ery, erythrocytes; Bas, basophils; Eos, eosinophils; Neut, neutro-
phils; Mon, monocytes; DC, dendritic cells; ILC, innate lymphoid cells.
From HSCs to functional end cells is a steady progression (Figure 1) with controver-
sies centering around whether cells undergo stepwise commitment or development is
continuous and at what stage(s) there is self-maintenance of cells. In canonical models of
hematopoiesis, HSCs are a homogeneous population of multipotent cells that self-renew.
HPCs are a mixture of multipotent progenitor cells (MPPs), cells with differing lineage
potentials, and uni-potent cells. It has become increasingly clear that mouse and human
HSCs are a mixture of cells whereby lineage affiliation can occur much earlier than
thought. Lineage primed/affiliated HSCs have been identified from gene expression stud-
ies. HSCs that are biased towards stable and long-term reconstituting of some blood cell
types, and just one cell type, have been identified from transplantation studies. These find-
ings have challenged whether (1) self-renewal is an aribute that is unique to HSCs, and
(2) lineage affiliation is entirely the domain of MPPs and their offspring. If HSCs can de-
termine their lineage fate directly, there is no longer the need to describe cells as HSCs
versus HPCs. This review examines the evidence to support or otherwise different views
on hematopoiesis.
2. HSCs Are Rare and Quiescent Cells That Self-Renew
A precise figure for the frequency of HSCs in human bone marrow is difficult because
marrow aspiration is inconsistent. From surface marker analyses, HSCs are found within
the CD34+ population of cells, which is heterogeneous, and ~0.7% of the total marrow cells
in young adults [2]. The frequency of HSCs within CD34+ cells is ~5% (range 2–8%) [3].
Figure 1. Compartmentalization of hematopoiesis. Self-renewing hematopoietic stem cells (HSCs)
give rise to the second tier of cells, which amplifies the number of cells produced by an HSC. This
tier includes the hematopoietic progenitor cells (HPCs) and the recognizable precursor cells to each
cell type. HPCs divide a substantial number of times to give rise ultimately to the unipotent HPCs.
Precursor cells divide to a lesser extent as they mature to give rise to the third tier of functional end
cells. Expansion is shown by the triangle with the lines providing an indication of the number of cell
cycles. Meg, megakaryocytes; Ery, erythrocytes; Bas, basophils; Eos, eosinophils; Neut, neutrophils;
Mon, monocytes; DC, dendritic cells; ILC, innate lymphoid cells.
From HSCs to functional end cells is a steady progression (Figure 1) with controver-
sies centering around whether cells undergo stepwise commitment or development is
continuous and at what stage(s) there is self-maintenance of cells. In canonical models of
hematopoiesis, HSCs are a homogeneous population of multipotent cells that self-renew.
HPCs are a mixture of multipotent progenitor cells (MPPs), cells with differing lineage
potentials, and uni-potent cells. It has become increasingly clear that mouse and human
HSCs are a mixture of cells whereby lineage affiliation can occur much earlier than thought.
Lineage primed/affiliated HSCs have been identified from gene expression studies. HSCs
that are biased towards stable and long-term reconstituting of some blood cell types, and
just one cell type, have been identified from transplantation studies. These findings have
challenged whether (1) self-renewal is an attribute that is unique to HSCs, and (2) lineage
affiliation is entirely the domain of MPPs and their offspring. If HSCs can determine their
lineage fate directly, there is no longer the need to describe cells as HSCs versus HPCs. This
review examines the evidence to support or otherwise different views on hematopoiesis.
2. HSCs Are Rare and Quiescent Cells That Self-Renew
A precise figure for the frequency of HSCs in human bone marrow is difficult because
marrow aspiration is inconsistent. From surface marker analyses, HSCs are found within
the CD34
+
population of cells, which is heterogeneous, and ~0.7% of the total marrow cells
in young adults [
2
]. The frequency of HSCs within CD34
+
cells is ~5% (range 2–8%) [
3
].
The use of the above percentages gives rise to a frequency of ~0.03% for HSCs in bone
marrow. A value of ~11,000 cells has been calculated for the reserve of human HSCs
in adolescent marrow [
4
]. The frequency of mouse HSCs in bone marrow is 1 in 10
4
as
determined by transplantation and long-term repopulation studies [
5
]. A frequency of

Int. J. Mol. Sci. 2025,26, 3346 3 of 15
2–5 HSCs
per 10
5
total bone marrow cells was obtained for adult mouse bone marrow from
a study that looked at HSCs that were most productive and that had a durable self-renewal
potential [
6
,
7
]. The frequency of mouse HSCs is higher at ~0.01% of nucleated cells based
on the expression of stem cell-affiliated surface markers, with ~5000 HSCs obtained from a
mouse [8]. Both human and mouse HSCs are largely quiescent. The percentage of human
HSCs that are cycling (non-G0) is around 10 (range 5–15) [
3
] and they replicate on average
around once every 40 weeks (range 25 to 50). Mouse HSCs replicate on average once every
2.5 weeks [4].
HSCs are often mistaken as immortal cells. Transplantation studies showed that HSCs
generate erythrocytes and lymphocytes throughout the lifespan of a mouse leading to the
extensive lifespan of HSCs [
9
]. Blood cell reconstitution has been repeated for three to
four mouse lifespans in serial transplantation studies [
10
]. To account for this self-renewal,
a longstanding view is that HSCs reside in a specialized perivascular niche in the bone
marrow. The niche provides signals that are crucial to self-renewal and is created partly by
mesenchymal cells and endothelial cells. However, it is much more complex— there is no
singular niche cell, and other influences include, for example, non-myelinating Schwann
cells, osteoblasts, macrophages, and the blood reviewed in [
11
]. Even so, whether HSCs
are truly immortal is a futile argument because self-renewal has not been demonstrated
for tens of thousands of divisions and each HSC probably divides just some two to three
hundred times in four mouse lifespans (divides ~ every 2.5 weeks). Accumulated genetic
defects from endless cell divisions, and/or as HSCs age, may decrease the capacity for
repair and, in turn, the fidelity of self-renewal [12].
3. Are HSCs Homogenous or a Mixture of Cells?
3.1. The HSC Compartment Is a Homogeneous Population of Multipotent Cells
In a ground-breaking series of experiments in the 1960s, Till and McCulloch used
sublethal irradiation to induce clonal markers in bone marrow cells. When these cells
were transplanted into a mouse, they gave rise to visible clones of marked cells in the
spleen (spleen colony forming units, CFU-S) that contained a mixture of granulocytes,
macrophages, erythrocytes, and megakaryocytes. Some CFU-S cells had replicated them-
selves because transferring CFU-S cells to an irradiated host produced more CFU-S, with
the above mixture of cells, and some spleens contained cells with the potential to make lym-
phocytes [
13
–
15
]. Till and McCulloch proposed that these multipotent and self-renewing
cells are HSCs. Subsequently, mouse HSCs were purified as cells that lacked the lineage-
specific markers CD4 (T cells), CD8 (T cells), B22 (B cells), Gr-1 (granulocytes), Mac-1
(macrophages) and TER119 (erythrocytes), abbreviated as Lin
−
, and cells that expressed
Sca-1
+
(a stem cell marker) and c-kit
+
(a stem cell growth factor receptor). The most primi-
tive HSCs were found within Lin
−
, Sca-1
+
, c-kit
+
(LSK) cells that were low to negative for
expression of CD34 (a stem/progenitor cell marker). In 1996, the injection of a single LSK,
CD34low/−cell
into a mouse led to the long-term reconstitution of myeloid and lymphoid
cells in 21% of recipients [
16
]. A more efficient method for the transplantation of LSK,
CD34
−
mouse HSCs was provided by culturing sorted individual cells in mouse c-kit
ligand and either mouse interleukin-11 or human recombinant G-CSF. Transplantation of
clones of <15 cells showed high-level multilineage reconstitution [
17
]. Clearly, multipotent
HSCs that self-renew reside at the top of the hematopoietic hierarchy.
A longstanding view is that self-renewal is unique to HSCs. From transplantation
studies, HSCs that repopulate all cell lineages have been classified as long-term HSCs
(LT-HSCs), intermediate-term HSCs (IT-LSCs), and short-term HSCs (ST-HSCs) [
18
]. These
cells are phenotypically HSCs, and the level and duration of reconstitution varies for the
three sub-types. LT-HSCs sustain hematopoiesis for the lifespan of a mouse. ST-HSCs

Int. J. Mol. Sci. 2025,26, 3346 4 of 15
generate all blood cells but for only several weeks, and in this case do not repopulate
with absolute efficiency [
19
]. When ST-HSCs are secondary transplanted they generally
do not show any constitution, whereas reconstitution levels do not change for secondary
transplantation of LT-HSCs. As mice age, the number of HSCs increases but their ability to
repopulate a mouse decreases [
20
]. The capacity for self-renewal is, therefore, a variable
trait within HSCs, and Hoxb5 plays a role in imparting this functional heterogeneity. For
Hoxb5-negative HSCs, exogenous Hoxb5 expression conferred protection against the loss
of self-renewal capacity and partially altered the fate of ST-HSCs to that of LT-HSCs [21].
Are HSCs the sole hematopoietic cells with the capacity to self-renew? Erythroid
progenitors from the bone marrow of adult mice expand 10
2
-to 10
5
-fold when grown in
the presence of erythropoietin, in keeping with their role to expand cell numbers
in vivo
.
Culture of cells obtained from E9.5 mouse yolk sac in serum-free conditions generated
immature erythroblasts, which expanded 10
10
- to 10
30
-fold, and at best 10
64
-fold. The
cells that were extensively renewed were definitive erythroid (Ter119
low
), c-kit
high
, and
eventually matured into enucleated erythrocytes [
22
]. The following studies provided clear
evidence that loss of self-renewal is not fundamental to the affiliation of HSCs to a cell
lineage(s). Mouse HSCs are more stringently defined than human cells and often as LSK
CD150
+
CD48
−
CD34
−
cells. LSK, CD150
+
, CD48
−
, CD34
−
sorted cells were cultured
as single cells in the presence of stem cell factor and thrombopoietin, which is the major
cytokine to megakaryocyte development. Daughter pairs were micromanipulated, and
each cell was transplanted into lethally irradiated mice to examine their reconstitution
nature. For some pairs, both daughter cells reconstituted hematopoiesis whereby the
single cell had undergone a symmetric cell division to give rise to HSCs (combinations of a
LT-HSC, IT-HSC and ST-HSC). Distinct patterns were observed for some pairs whereby
the single cell had undergone an asymmetric cell division to give rise to either an HSC
(LT-HSC or ST-HSC) and a megakaryocyte-restricted repopulating cell or an HSC (ST-HSC)
and a common myeloid-restricted (megakaryocyte, erythroid, and myeloid) repopulating
cell. HSCs had generated lineage-restricted cells directly that were able to self-renew
to a considerable extent [
18
]. Altogether, the above findings support the view that self-
maintenance is not restricted to HSCs. This, in turn, blurs any clear distinction between
cells that we compartmentalize as HSCs versus HPCs.
3.2. HSCs Are a Complex Mixture of Cells
Early studies used single-cell RT-PCR to examine HSC heterogeneity regarding the
expression of cytokine receptors by mouse multipotent FDCP-mix A4 cells. Self-renewal
of these cells was maintained by culturing in interleukin 3. There was variable low-level
expression of mRNAs for the receptors for the lineage-affiliated cytokines erythropoietin,
granulocyte-colony stimulating factor, granulocyte/macrophage colony-stimulating factor,
and macrophage colony-stimulating factor (M-CSF). There was a high degree of cell het-
erogeneity regarding co-expression. Therefore, the expression of lineage-affiliated genes
occurs in the self-renewal state with the investigators concluding that promiscuous lineage
priming occurs prior to HSC lineage commitment [
23
]. From single-cell RT-PCR studies of
mouse HSCs, ~12% of LT-HSC (LSK CD150
+
CD48
−
, CD34
−
) and ~20% of ST-HSC (LSK
CD150
+
CD48
−
CD34
+
) expressed mRNA for the erythropoietin receptor, and ~19% of
LT-HSC and ~24% of ST-HSC expressed the receptor for macrophage colony-stimulating
factor (M-CSFR) at their surface. The fms-like tyrosine kinase receptor (Flt3) was expressed
at the cell surface by ~5% of LT-HSC and ~8% of ST-HSC. Co-expression of the mRNAs
encoding Flt3 and the erythropoietin receptor was rarely seen whereas co-expression of Flt3
and M-CSFR was observed by 1 to 3% of the M-CSFR positive cells. LT-HSC and ST-HSCs
are, therefore, a heterogeneous population of cells [24].

Int. J. Mol. Sci. 2025,26, 3346 5 of 15
A low level/priming of gene expression may reflect either a clear inclination to adopt
a developmental pathway or be insignificant if such relates to noise within gene expression.
Regarding inclination and physiological importance, it is important to note that growth
factors for some of the above receptors can direct the fate of HSCs/HPCs. Erythropoietin
has been shown to guide HSCs toward an erythroid fate [
25
], and M-CSF has been shown
to instruct granulocyte/macrophage HPCs to adopt a macrophage fate [
26
]. Additionally,
the expression of the receptors for M-CSF and granulocyte colony-stimulating factor is
autoregulated, which might serve to enhance low-level expression [27].
Evidence for lineage biases within HSCs has also been provided from transplanta-
tion studies. Adolphson and colleagues examined heterogeneity within LSK HSCs by
using Flt3 as a surface marker. LSK KSCs that expressed a Flt3 at a high level had a
high proliferative potential and sustained granulocyte, monocyte, and B and T cell pro-
duction and LSK HSCs that did not express Flt3 failed to produce significant erythroid
and megakaryocytic progeny. Downregulation of genes that encoded regulators of ery-
throid and megakaryocyte development accompanied distinct lineage restriction. The
Flt3
high
cells were termed lymphoid-myeloid HSCs, and the investigators proposed a re-
vised road map for hematopoiesis [
28
]. Muller-Sieburg and colleagues obtained clones
of HSCs from mouse bone marrow by limiting dilution on a stromal cell (S17cells) and
examined their lineage potentials by transplantation into lethally irradiated mice. From
this approach, they described HSCs with an extensive self-renewal capacity that were
myeloid biased. These cells showed an impaired responsiveness to interleukin-7, a cy-
tokine that is critical for B cell, T cell, and innate lymphoid cell generation [
29
]. In aged
mice, there is a skewing of HSCs towards myeloid development [
30
]. This may be at-
tributable to a genetic change, but other investigators have argued that a change to the
cells’ environment may contribute [31]. Montecino-Rodriguez and colleagues described a
lymphoid-biased mouse HSC (LSK CD150
low
CD135
−
) that efficiently generates lymphoid
progeny and is maintained with age [
32
]. The expression of CD201 has been associated
with lymphoid-biased HSCs [
33
]. Sanjuan and colleagues identified a megakaryocyte-
primed subset of mouse HSCs (LSK CD150
+
CD48
−
CD34
−
) by virtue of expression of
the platelet-associated von Willebrand factor, which is involved in platelet aggregation.
The primed cells showed a propensity for short- and long-term platelet reconstitution,
and thrombopoietin, a cytokine that primarily regulates megakaryocyte development, was
required for their maintenance. The investigators proposed that megakaryocyte-biased
HSCs reside at the top of hematopoiesis [34].
Myeloid-biased HSCs have been described for human HSCs, and as for mouse HSCs,
there is HSC skewing towards myeloid development and a decline in B cell lymphopoiesis
in old age [
3
]. As seen for mouse platelet-biased HSCs, a subset of human adult bone
marrow HSCs (CD34
+
CD38
−/dim
cells) was identified that expresses the receptor for
thrombopoietin [
35
]. To investigate heterogeneity within primitive CD34
+
HSCs from
human cord blood, cells were CD49f
+
index-sorted, lentivirus barcoded, and pools of
transduced cells were injected into irradiated mice to examine their lineage potentials. Few
clones (8% at 30–38 weeks) yielded cells of many lineages, and, strikingly, 30% produced B
cells
±
CD34
+
cells [
36
]. Therefore, both human and mouse HSCs can be organized into
subsets, and we are yet to learn about more subsets.
4. What Are the Lineage Options That Are Available to HSCs?
4.1. HSCs Choose to Develop Towards Either Myeloid or Lymphoid Cells
In 1997, Kondo and colleagues described a clonogenic cell that was present in adult
mouse bone marrow (Lin
−
, Sca-1
lo
, c-Kit
lo
, IL-7R
+
, Thy-1
−
) that had a rapid reconstitution
capacity
in vivo
that was restricted to B cells, T cells, and natural killer cells. The cell

Int. J. Mol. Sci. 2025,26, 3346 6 of 15
lacked myeloid differentiation potential
in vitro
, and is, therefore, a common lymphoid
progenitor (CLP) [
37
]. A mouse cell that gave rise to either megakaryocyte/erythrocyte
or granulocyte/macrophage progenitors, the common myeloid progenitor (CMP), was
described in 2000 [
38
]. A feature of canonical models is the presence of a CLP and a CMP
whereby HSCs/MPPs make an immediate choice to differentiate towards either lymphoid
or myeloid cells.
4.2. A Continuum View of the Options That Are Available to HSCs
In 2009, a new model for hematopoiesis provided a very different view regarding
HSC developmental options [
39
]. All end options were shown to be available as a con-
tinuum of fate choice. The existence of lymphoid-myeloid HSCs contradicts a strict lym-
phoid/myeloid dichotomy. Moreover, macrophages and B cells are closely related function-
ally as they are both antigen-processing cells. Comparison of the phosphoproteins that are
expressed by human precursor macrophage and B cell lines revealed similar patterns [
40
].
Other workers derived a macrophage cell line from a pro-B lymphoma line (ABLS8.1) [
41
]
and observed macrophage lineage switching of mouse early pre-B lymphoid cells [
42
].
Lymphoproliferative disorders in interleukin-7 transgenic mice were an expansion of im-
mature B cells that also have macrophage potential [
43
]. In 1992, a bipotent precursor of
B cells and macrophages was identified in mouse fetal liver [
44
] and later in adult bone
marrow [45].
A further proposal was that HSCs can affiliate directly to a cell lineage. This is
supported by mouse and human HSCs that selectively express the receptors for cytokines
that are associated with the development of a cell lineage, including for erythropoietin,
M-CSF, and thrombopoietin (see above). Knapp and colleagues concluded that multiple
directional paths are available to primitive human cord blood cells from analyzing the
molecular features of single cells and the presence of unipotent B cell clones (see above).
Mouse bone marrow cells were sorted as LSK cells, and single-cell RNA-seq was used to
examine their transcriptional landscape. A detailed map for the clustering of single-cell
profiles revealed 19 subpopulations that were transcriptionally homogeneous. There was
an absence of progenitors with a mixed lineage and the investigators concluded that there
was transcriptional priming towards seven different fates, namely neutrophil, basophil,
eosinophil, monocyte, dendritic cell, erythroid, and megakaryocyte [
46
]. Whilst the sorted
cells were myeloid HPCs, as described, the findings concur with the direct availability of
multiple fates to a developing cell.
However, HSCs have been described as able to self-maintain and are lymphoid/myeloid
or lymphoid or myeloid biased. Perhaps this leans towards having to just extend self-
maintenance to oligopotent HPCs. In other words, perhaps cells that are at whatever stage
of lineage restriction can self-maintain. Recent findings from tracking of HSC differentia-
tion over time, the use of single-cell RNA sequencing, and computational analysis have
revealed a complex dynamic to the nature of HSCs and HPCs. These studies showed
that self-renewal varies across the HSC/HPC landscape and extends to lineage-specific
patterns of self-renewal. Regarding differentiation trajectories, cell fate and subsequent
differentiation existed crucially in specific experimental conditions. For example, cells that
are described as CMPs rarely show a combined megakaryocyte, erythrocyte, granulocyte,
and monocyte output, are primed towards a specific lineage and rarely behave as mul-
tipotent cells. CMPs move under strong differentiation conditions towards a particular
fate rather than exploring other fate options [
47
] and rarely showed the above-combined
output in the early studies that described CMPs [
38
]. A final consideration that is often
overlooked regarding blood cell production is the extent to which cell pools that are embry-
onic born contribute to hematopoiesis during early and adult life. Multipotent embryonic

Int. J. Mol. Sci. 2025,26, 3346 7 of 15
hematopoietic progenitors that arise soon after endothelial-to-hematopoietic transition
predominantly drive hematopoiesis in the young adult. Their contribution to blood cell
production decreases over time but they have a lifelong contribution as a predominant
source of lymphocytes [
48
]. It has also been shown that pools of HSCs and HPCs that are
established during embryogenesis self-renew in parallel over life and contribute to blood
cell production [49].
As follows, there is a need to look at findings for HSCs as they develop toward an
end cell type. Is there a canonical hematopoietic cell lineage tree with a discrete set of
irreversible differentiation states that are defined by a series of progenitors with differing
lineage potentials? Alternatively, do HSCs start their development from a broad landscape
of a continuum of options, are the trajectories broad, and do the controls on lineage
development drive choices between adjacent cell fates?
5. HSC/HPC Developmental Progression
5.1. A Canonical Cell Lineage Tree for Developmental Progression
Figure 2shows a simplified hematopoietic cell lineage tree to illustrate the principle
of stepwise restriction of lineage options via a binary decision-making process. Canonical
models depict a tree topology [
50
], and there is a plethora of models regarding the routes
that are followed by developing MPPs [
39
]. In all branching models, HPCs follow a
preferred route to stepwise restrict their development towards a particular fate. The routes
were mapped largely from the types of HPCs that were observed when bone marrow cells
were dispersed in a semi-solid medium whereby individual colony-forming unit (CFU)
cells were observed to give rise to colonies that contained various types of cells [
51
]. Like
CMPs, some cells gave rise to colonies containing granulocytes, erythrocytes, monocytes,
and megakaryocytes and were termed colony-forming unit (CFU)-GEMM. Other cells
gave rise to colonies that contained just two types of cells, for example, granulocytes and
macrophages, and were termed CFU-GM. Stepwise commitment that was irreversible was
concluded by placing the types of CFUs observed in an order so that they progressively
reduced their fate options.
Int. J. Mol. Sci. 2025, 26, x FOR PEER REVIEW 7 of 15
multipotent cells. CMPs move under strong differentiation conditions towards a particu-
lar fate rather than exploring other fate options [47] and rarely showed the above-com-
bined output in the early studies that described CMPs [38]. A final consideration that is
often overlooked regarding blood cell production is the extent to which cell pools that are
embryonic born contribute to hematopoiesis during early and adult life. Multipotent em-
bryonic hematopoietic progenitors that arise soon after endothelial-to-hematopoietic tran-
sition predominantly drive hematopoiesis in the young adult. Their contribution to blood
cell production decreases over time but they have a lifelong contribution as a predominant
source of lymphocytes [48]. It has also been shown that pools of HSCs and HPCs that are
established during embryogenesis self-renew in parallel over life and contribute to blood
cell production [49].
As follows, there is a need to look at findings for HSCs as they develop toward an
end cell type. Is there a canonical hematopoietic cell lineage tree with a discrete set of
irreversible differentiation states that are defined by a series of progenitors with differing
lineage potentials? Alternatively, do HSCs start their development from a broad land-
scape of a continuum of options, are the trajectories broad, and do the controls on lineage
development drive choices between adjacent cell fates?
5. HSC/HPC Developmental Progression
5.1. A Canonical Cell Lineage Tree for Developmental Progression
Figure 2 shows a simplified hematopoietic cell lineage tree to illustrate the principle
of stepwise restriction of lineage options via a binary decision-making process. Canonical
models depict a tree topology [50], and there is a plethora of models regarding the routes
that are followed by developing MPPs [39]. In all branching models, HPCs follow a pre-
ferred route to stepwise restrict their development towards a particular fate. The routes
were mapped largely from the types of HPCs that were observed when bone marrow cells
were dispersed in a semi-solid medium whereby individual colony-forming unit (CFU)
cells were observed to give rise to colonies that contained various types of cells [51]. Like
CMPs, some cells gave rise to colonies containing granulocytes, erythrocytes, monocytes,
and megakaryocytes and were termed colony-forming unit (CFU)-GEMM. Other cells
gave rise to colonies that contained just two types of cells, for example, granulocytes and
macrophages, and were termed CFU-GM. Stepwise commitment that was irreversible
was concluded by placing the types of CFUs observed in an order so that they progres-
sively reduced their fate options.
Figure 2. Stepwise restriction of lineage options. Hematopoietic stem cells (HSC) that are undergo-
ing differentiation follow a preferred route to stepwise restrict their lineage options. The choice that
Figure 2. Stepwise restriction of lineage options. Hematopoietic stem cells (HSC) that are under-
going differentiation follow a preferred route to stepwise restrict their lineage options. The choice
that an HSC/MPP makes is to develop towards either myeloid or lymphoid cells. Subsequently,
hematopoietic progenitors that have different cohorts of options define the model. MPP, multipotent
progenitor cell; CMP, common myeloid progenitor cell, CLP, common lymphoid progenitor cell;
MEP, megakaryocyte/erythroid progenitor cell; GMP, granulocyte/monocyte progenitor cells, Meg,
megakaryocytes; Ery, erythrocytes; Neut, neutrophils, Mon, monocytes; ILC, innate lymphoid cells.

Int. J. Mol. Sci. 2025,26, 3346 8 of 15
For some canonical models, topologies show various pathways to an end cell type [
39
].
Early evidence to support the use of different trajectories by HPCs was provided by
examining the transcription profiles of dendritic cells that had been derived
in vitro
from
cells that had been purified as CLPs and CMPs and the finding that the dendritic cell
profiles were identical. The investigators concluded that the program for dendritic cells can
operate independently of the myeloid and lymphoid pathways [
52
]. Many investigators
have brought to attention trans-differentiation of one cell type into another, but this has not
been demonstrated convincingly at a clonal level for marked hematopoietic cells.
Like HSCs, HPC decision-making in canonical models is a binary process. There is
strong support of the view that cellular controls are bi-stable with cells switching from one
steady state to an alternative steady state as driven by an external event or a change to
an internal process. Binary tree-like maps invariably describe the development of entire
organisms, for example, the newly hatched roundworm Caenorhabditis elegans [
53
], and
various tissues, for example, neuronal-crest-derived cells [
54
]. A recent and complex gene
regulatory/neural network-based analysis arrived at a binary-fork transition map for
Caenorhabditis elegans and hematopoiesis, with decisions being driven by a combination
of signals [55].
However, might HSCs be multi-stable and be able to process more than two exclusive
options? Multi-stability exists for positive feedback systems and signaling pathways [
56
,
57
]
and metabolic networks [
58
]. The transcription factors GATA1, GATA2, and PU.1 play
key roles in the development of HSCs and HPCs. One option versus another is easier to
model, and investigators first arrived at bi-stable models for a framework for the actions
of the transcription factors. Bi-stable models were then embedded to achieve a tri-stable
model, which was further modeled to encompass four mutually exclusive stable states. The
findings from their modeling fitted experimental data [
59
]. We cannot, therefore, exclude
that HSCs can process complex information regarding how they make a choice of lineage.
5.2. HPC Progression Is Gradual and Versatile
The continuum model in Figure 3shows that all end options are available directly to
HSCs and that development to an end cell type is then a gradual and continuous process.
There are no arrows to show a preferred route because a series of HPCs with different
set potentials does not underlie progression. Having selected a cell lineage, HSCs and
their offspring can still step sideways to an alternative, albeit closely related fate. In other
words, developing cells are versatile and tend to switch most readily to a path that leads to
their pairwise neighbor fate, with alternative fates having remained latent or clandestine.
Support to versatility has been provided from RNA sequencing of more than 1600 mouse
cells to examine the trajectories of HSCs and HPCs as these cells developed along the
erythroid, granulocyte/macrophage, and lymphoid pathways. The construction of a single-
cell resolution map for differentiation led investigators to conclude that trajectories are
broad, which would allow cells to sidestep to the left or right of a chosen pathway [60].
In the continuum model, there are close relationships between the pathways for each
cell lineage. They are like those shown in canonical models, which is not too surprising
because placing neighbors in both models took into consideration the cohorts of options
that were revealed by CFU cells and bi-potent normal and cell line cells [
61
]. Investigators
arrived at a radius plot identical to that shown in Figure 3from mathematical modeling
of the networks that control the dynamics of blood formation and state transitions [
62
].
The placement of cell lineages in Figure 3also took into consideration the shared usage of
transcription factors [
39
]. Microarray profiling of gene expression, including the expression
of transcription factors, within HPCs and mature cell types arrived at a similar pattern

Int. J. Mol. Sci. 2025,26, 3346 9 of 15
of near neighbors. They were erythrocytes, megakaryocytes, granulocytes/monocytes,
dendritic cells, B cells, natural killer cells, and T cells [63].
Int. J. Mol. Sci. 2025, 26, x FOR PEER REVIEW 9 of 15
Figure 3. A continuum model for hematopoiesis. Lineage-affiliation is initiated as early as within
HSCs whereby each either self-renews or makes a choice from a continuum of all end cell options
that are directly available. An HSC then differentiates in a continuous manner. There are close rela-
tionships between developmental pathways, and HSCs having ‘chosen’ an affiliation can still
change their mind to step sideways to an adjacent pathway. Meg, megakaryocytes; Ery, erythro-
cytes; Bas/mast, basophils/mast cells; Eos, eosinophils; Neut, neutrophils; Mon, monocytes; DC,
dendritic cells; ILC, innate lymphoid cells.
In the continuum model, there are close relationships between the pathways for each
cell lineage. They are like those shown in canonical models, which is not too surprising
because placing neighbors in both models took into consideration the cohorts of options
that were revealed by CFU cells and bi-potent normal and cell line cells [61]. Investigators
arrived at a radius plot identical to that shown in Figure 3 from mathematical modeling
of the networks that control the dynamics of blood formation and state transitions [62].
The placement of cell lineages in Figure 3 also took into consideration the shared usage of
transcription factors [39]. Microarray profiling of gene expression, including the expres-
sion of transcription factors, within HPCs and mature cell types arrived at a similar pat-
tern of near neighbors. They were erythrocytes, megakaryocytes, granulocytes/mono-
cytes, dendritic cells, B cells, natural killer cells, and T cells [63].
Is the progression of HSCs towards uni-potent HPCs a continuous process? Investi-
gators captured the transcriptional landscapes of multipotent (LSK) and oligopotent (Lin−,
Sca−, Kit+) cells as they developed and the relationships between pathways by clonal bar-
coding of cells. These cells gave rise to nine cell types, namely megakaryocytes, erythro-
cytes, basophils, mast cells, eosinophils, neutrophils, monocytes, dendritic cells, and lym-
phoid precursors when cultured in conditions for cell growth and muti-lineage differen-
tiation. Single-cell RNA sequencing was undertaken for cells that were sampled immedi-
ately and later. The transcriptional states of the least differentiated HPCs did not match a
discrete hierarchy of oligopotent and intermediate HPCs and instead supported the view
that developing cells lie along a continuum of states. Some clones exhibited uni-lineage
differentiation and others multi-lineage differentiation. The transcriptional landscapes ob-
tained for the cells that had appeared by 6 days in culture were used to construct a map
for how multipotent HPCs had veered towards pathways. The near-neighbor pathways
observed were for megakaryocyte, erythrocyte, mast cell, basophil, eosinophil, neutro-
phil, monocyte, migratory dendritic cell, plasmacytoid dendritic cell, and lymphoid de-
velopment, like those shown in Figure 3. A continuum landscape was also seen for cells
that had developed in vivo. This was revealed by barcoding LT-HSCs/ST-HSCs (Lin−, Sca-
high, kit+), culturing for two days, and then transplanting into irradiated mice. The near-
Figure 3. A continuum model for hematopoiesis. Lineage-affiliation is initiated as early as within
HSCs whereby each either self-renews or makes a choice from a continuum of all end cell options
that are directly available. An HSC then differentiates in a continuous manner. There are close
relationships between developmental pathways, and HSCs having ‘chosen’ an affiliation can still
change their mind to step sideways to an adjacent pathway. Meg, megakaryocytes; Ery, erythrocytes;
Bas/mast, basophils/mast cells; Eos, eosinophils; Neut, neutrophils; Mon, monocytes; DC, dendritic
cells; ILC, innate lymphoid cells.
Is the progression of HSCs towards uni-potent HPCs a continuous process? Inves-
tigators captured the transcriptional landscapes of multipotent (LSK) and oligopotent
(Lin
−
, Sca
−
, Kit
+
) cells as they developed and the relationships between pathways by
clonal barcoding of cells. These cells gave rise to nine cell types, namely megakaryocytes,
erythrocytes, basophils, mast cells, eosinophils, neutrophils, monocytes, dendritic cells,
and lymphoid precursors when cultured in conditions for cell growth and muti-lineage
differentiation. Single-cell RNA sequencing was undertaken for cells that were sampled im-
mediately and later. The transcriptional states of the least differentiated HPCs did not match
a discrete hierarchy of oligopotent and intermediate HPCs and instead supported the view
that developing cells lie along a continuum of states. Some clones exhibited uni-lineage
differentiation and others multi-lineage differentiation. The transcriptional landscapes
obtained for the cells that had appeared by 6 days in culture were used to construct a map
for how multipotent HPCs had veered towards pathways. The near-neighbor pathways
observed were for megakaryocyte, erythrocyte, mast cell, basophil, eosinophil, neutrophil,
monocyte, migratory dendritic cell, plasmacytoid dendritic cell, and lymphoid develop-
ment, like those shown in Figure 3. A continuum landscape was also seen for cells that had
developed
in vivo
. This was revealed by barcoding LT-HSCs/ST-HSCs (Lin
−
, Sca
high
, kit
+
),
culturing for two days, and then transplanting into irradiated mice. The near-neighbor
pathways observed were for development towards erythrocytes, basophils, neutrophils,
monocytes, dendritic cells, B cells, and T cells [
64
]. From findings for the lineage contribu-
tion of mouse HSCs and HPCs from transplantation studies, investigators concluded that
native hematopoiesis is not entirely established by hierarchical HPC differentiation. The
model they proposed has an initial broad landscape with pathways eventually emerging for
megakaryocyte, erythrocyte, myeloid/granulocyte, myeloid/monocyte, and lymphocyte
development [65].
Findings for human HPCs support the view that their development is a continuous
process. This was revealed by constructing the developmental trajectories for the immedi-

Int. J. Mol. Sci. 2025,26, 3346 10 of 15
ate progeny of HSCs (Lin
−
, CD34
+
, CD38
−
) and more differentiated HPCs (Lin
−
, CD34
+
,
CD38
+
). Findings from single-cell RNA sequencing were combined with findings for single-
cell culture differentiation outcomes. For the immediate progeny of HSCs (Lin
−
, CD34
+
,
CD38
−
), there was an absence of clusters, and instead, these cells were a continuously con-
nected entity. Clusters that conformed to distinct HPCs for each of the major hematopoietic
cell types were observed for the more mature Lin
−
, CD34
+
, and CD38
+
cells. From these
findings, the investigators concluded that early hematopoiesis is best represented by a
cellular continuum of low-primed undifferentiated (CLOUD)-HSCs/HPCs. They proposed
that these cells gradually acquire lineage priming in multiple directions. A graphical sum-
mary of the findings showed that the erythroid, megakaryocyte, eosinophil/basophil/mast
cell, neutrophil, monocyte/dendritic cell, and B cell pathways were near neighbors [66].
In 1957, Waddington proposed a metaphorical and epigenetic landscape for the de-
velopment of primitive embryonic cells. He proposed that these cells develop from a
broad upland, like continuum models. Developing cells then coursed downwards through
bifurcating valleys with the hills to the valleys offering little chance of sideways escape
to emerge as an alternative differentiated cell [
67
]. In this case, the likelihood of lateral
transition by HPCs and reprogramming may become more restricted as development
proceeds by virtue of end-cell-type programs becoming increasingly stable. HSCs and/or
MPPs might, therefore, be more likely to be instructed to veer toward an alternative op-
tion(s) during their early development. The ligand to FLt3 has been proposed to have
an instructive role at an early stage of HSC development. Instruction was dependent on
ligand concentration because if the signal strength exceeded a certain threshold, level cells
veered toward lymphoid and myeloid fates at the expense of megakaryocyte and erythroid
development [
68
,
69
]. When FLt3 was overexpressed in megakaryocyte/erythroid HPCs,
they veered toward granulocytes and macrophages [
70
]. A recent study has described the
cellularity of human bone marrow relating to Flt3 ligand deficiency and concurred that if
early HPCs do not receive sufficient Flt3 ligand stimulation they preferentially differentiate
toward megakaryocyte and erythrocytes. The investigators concluded that the Flt3 ligand
governs the development of partially overlapping lineages [71].
6. Can Findings Be Reconciled to a Consensus Model?
A view for many years has been that the only long-term self-renewing cells in the
hematopoietic system are HSCs. The term ‘immortality’ has been too loosely applied
to these cells because this has not been tested in a strict sense. Now there is also good
evidence to support extending the capacity for self-maintenance to multi-lineage biased
and even lineage affiliated cells whereby these cells can make a substantial contribution
to hematopoiesis to replenish an end cell type(s). They may do so to a greater extent
during emergency hematopoiesis as opposed to steady state. Vector integration sites were
used as markers of clonal identity in studies that used lentivirus HSCs for gene therapy
of metachromatic leukodystrophy, Wiskott–Aldrich syndrome, and β-thalassemia. For all
conditions, 50% of the transplanted clones showed a multi-lineage output. The remain-
ing clones showed a preferential lineage output, which was myeloid for metachromatic
leukodystrophy, lymphoid for Wiskott–Aldrich syndrome, and erythroid for
β
-thalassemia.
From these findings, lineage-committed HSCs/HPCs, including uni-lineage cells, had
persisted for several years post-transplantation [
72
]. Again, the capacity for self-renewal
does not draw a clear line between cells that have been ring-fenced for many years as HSCs
versus HPCs.
In early studies, the lineage potentials of HPCs were determined from the diverse
combinations of mature cells that were seen when bone marrow cells were dispersed in
semi-solid medium to give rise to CFUs and transferred to an irradiated mice to give rise to

Int. J. Mol. Sci. 2025,26, 3346 11 of 15
spleen CFUs. Findings from these studies provided the backbone for arriving at canonical
models. The placement of lineage pathways close to one another is quite similar in canonical
models and the continuum model. Therefore, there are clearly close relationships between
cell lineages. Various oligopotent progenitors can be reconciled with a continuum model
because HSCs/HPCs developmental progression is a versatile process. They were mapped
to the continuum model by assuming that the different cell types generated by single cells
in the CFU assay relate to limited transitions to another pathway as colony cell numbers
increase [73].
More recent studies of the lineage potentials of HSCs and HPCs have examined the
molecular feature of single HPCs/HSCs globally, made use of clonal bar-coding to track
their development, and sub-fractionated HSCs for transplantation. Findings have led to
a much more complex topology and to the developmental progression of HSCs/HPCs.
Features of continuum models are that multiple pathways arise directly from HSCs, the
developmental progression of cells is a continuous process, and cells that have chosen an
option can still adopt another pathway. This versatility questions whether irreversible states
of commitment exist. It has been known for many years that cells that are well on their way
to becoming T cells can generate other cell types. Early T cell progenitor (double negative
(DN)) cells express neither CD4 nor CD8. Kit
hi
DN1 and DN2 cells can give rise to natural
killer and myeloid cells [
74
–
76
]. Mature cells can change their functional characteristics
as seen in the interconversion of CD4+ T effector cells. T helper 2 cells can give rise to
follicular T helper cells [
77
] and regulatory T cells can convert to pro-inflammatory T helper
17 cells [78].
7. Concluding Remarks
Some 75 years have elapsed since the description of HSCs. Presently, we might
conclude that drawing a hierarchical model for a series of hierarchical progenitors that
have strict lineage options is an oversimplification of the complex ability and adaptability
of hematopoiesis. We might expect versatility/flexibility because hematopoiesis must
meet the demands of both steady state and emergency circumstances. However, we are
still learning how to combine aspects of various models to arrive at a consensus as to the
process of hematopoiesis. This is important to what the difference is between the behavior
of normal and leukemia stem cells. Leukemia stem cells are rare cells that sustain leukemia,
and like HSCs, they generate a hierarchy of cells. The different acute leukemias most likely
arise from the transformation of an HSC. The self-renewal ability of HSCs places them
most at risk of transformation regarding the generation of a cell that can sustain leukemia.
The leukemias are categorized according to the cellular types that are present, namely
myeloblastic, granulocytic, erythroblastic, monocytic, lymphoblastic, etc. The existence of
lineage-affiliated HSCs may explain why a transformed HSC is restricted to generating
cells of just one cell type. An alternative view relates to each of the leukemias having a
signature oncogene, and there is good evidence to support the view that the oncogene
instructs cell lineage [
79
–
82
]. Hence, a resolve to the precise nature of hematopoiesis is
inextricably tied to efforts to unravel the nature of leukemia.
Funding: G.B. is funded by UK Research and Innovation (UKRI) under the UK government’s Horizon
Europe funding guarantee EP/Y030818/1 and is an associate partner to eRaDicate.
Conflicts of Interest: The author declares no conflicts of interest.

Int. J. Mol. Sci. 2025,26, 3346 12 of 15
References
1. Manz, M.G.; Boettcher, S. Emergency granulopoiesis. Nat. Rev. Immunol. 2014,14, 302–314.
2.
Macedo, A.; Orfao, A.; Ciudad, J.; Gonzalez, M.; Vidriales, B.; Lopez-Berges, C.; Martinez, A.; Landolf, C.; Canizo, C.; San Miguel,
J.F. Phenotypic analysis of CD34 subpopulations in normal human bone marrow and its application for the detection of minimal
residual disease. Leukemia 1995,9, 1896–1901.
3.
Pang, W.W.; Price, E.A.; School, D.; Beerman, I.; Maloney, W.J.; Rossi, D.J.; Schrier, S.L.; Weissman, I.L. Human bone marrow
hematopoietic stem cells are increased in frequency and myeloid-biased with age. Proc. Natl. Acad. Sci. USA 2011,108,
20012–20017. [CrossRef] [PubMed]
4.
Catlin, S.N.; Busque, L.; Gale, R.E.; Guttorp, P.; Abkowitz, J.L. The replication rate of human hematopoietic stem cells
in vivo
.
Blood 2011,117, 4460–4466. [PubMed]
5.
Szilvassy, S.J.; Humphries, R.K.; Landsdorp, P.M.; Eaves, A.C.; Eaves, C.J. Quantitative assay for totipotent reconstitution
hematopoietic stem cells by a competitive repopulation strategy. Proc. Natl. Acad. Sci. USA 1990,87, 8736–8740. [PubMed]
6.
Harrison, D.E.; Jordon, C.T.; Zhang, R.H.; Astle, C.M. Primitive hematopoietic stem cells: Direct assay of most productive
populations by competitive repopulation with simple binomial correlation and covariance calculations. Exp. Hematol. 1993,21,
206–219.
7.
Kent, D.G.; Copley, M.R.; Benz, C.; Wohner, S.; Dykstra, B.J.; Ma, E.; Cheyne, J.; Zhao, Y.; Bowie, M.B.; Zhao, Y.; et al. Prospective
isolation and molecular characterisation of haematopoietic stem cells with durable self-renewal potential. Blood 2009,113,
6342–6350.
8.
Challen, G.A.; Boles, N.; Kuan-Yin, K.; Goodall, M.A. Mouse hematopoietic stem cell identification and analysis. Cytometry A
2009,75, 14–24.
9.
Harrison, D.E.; Astle, C.M.; Lerner, C. Number and continuous proliferative pattern of transplanted immune hematopoietic stem
cells. Proc. Natl. Acad. Sci. USA 1988,89, 822–826.
10.
Harrison, D.E.; Astle, C.M. Loss of stem cell repopulating ability upon transplantation. Effects of donor age, cell number, and
transplantation procedure. J. Exp. Med. 1982,156, 1767–1779.
11. Morrison, S.J.; Scadden, D.T. The bone marrow niche for haematopoietic stem cells. Nature 2014,505, 327–334.
12.
Sieburg, H.B.; Cattarossi, G.; Muller-Sieburg, C.E. Lifespan differences in hematopoietic stem cells are due to imperfect repair and
unstable mean reversion. PLoS Comput. Biol. 2013,9, e1003006.
13.
Till, J.E.; McCulloch, E.A. A direct measurement of the radiosensitivity of normal bone marrow cells. Radiat. Res. 1961,14, 213–222.
[PubMed]
14.
Wu, A.M.; Till, J.E.; Simmonovitch, L.; McCulloch, E.A. A cytological study of the capacity for differentiation of normal
colony-forming cells. J. Cell Physiol. 1967,69, 177–184. [CrossRef] [PubMed]
15.
Wu, A.M.; Till, J.E.; Simmonovitch, L.; McCulloch, E.A. Cytological evidence for a relationship between normal hematopoietic
colony-forming cells and cells of the lymphoid system. J. Exp. Med. 1968,127, 455–464.
16.
Osawa, M.; Hanada, K.-I.; Hamada, H.; Nakauchi, H. Long-term lymphohematopoietic reconstitution by a single CD34-
low/negative hematopoietic stem cell. Science 1996,273, 242–245.
17.
Abe, T.; Masuya, M.; Ogawa, M. An efficient method for single hematopoietic stem cell (HSC) engraftment in mice based on cell
cycle dormancy of HSCs. Exp. Hematol. 2010,38, 603–608. [CrossRef] [PubMed]
18.
Yamamoto, R.; Morita, Y.; Ooehara, J.; Hamanaka, S.; Onodera, M.; Rudolph, K.L.; Ema, H.; Nakauchi, H. Clonal analysis unveils
self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells. Cell 2013,154, 1112–1126.
19.
Harrison, D.E.; Zhang, R.K. The same exhaustible multilineage precursor produces both myeloid and lymphoid cells as early as
3-4 weeks after marrow transplantation. Proc. Natl. Acad. Sci. USA 1992,89, 10134–10138.
20.
Chambers, S.M.; Shaw, C.A.; Gatza, C.; Fisk, C.J.; Donehower, L.A.; Goodell, M.A. Aging hematopoietic stem cells decline in
function and exhibit epigenetic dysregulation. PLoS Biol. 2007,5, e201.
21.
Sakamaki, T.; Kao, K.S.; Nishi, K.; Chen, J.Y.; Sadaoke, M.; Takaori-Kondo, A.; Weissman, I.L.; Miyanishi, M. Hoxb5 defines the
heterogeneity of self-renewal capacity in the haematopoietic stem cell compartment. Biochem. Biophys. Res. Commun. 2021,539,
34–41. [PubMed]
22.
England, S.J.; McGrath, K.E.; Frame, J.M.; Palis, J. Immature erythroblasts with extensive ex vivo self-renewal capacity emerge
from the early mammalian fetus. Blood 2010,117, 2708–2717. [PubMed]
23.
Hu, M.; Krause, D.; Greaves, M.; Sharkis, S.; Dexter, M.; Heyworth, C.; Enver, T. Multilineage gene expression precedes
commitment in the hematopoietic system. Genes Dev. 1997,11, 774–785. [PubMed]
24.
Mooney, C.J.; Cunningham, A.; Tsapogas, P.; Toellner, K.-M.; Brown, G. Selective expression of Flt3 within the mouse hematopoietic
stem cell compartment. Int. J. Mol. Sci. 2017,18, 1037. [CrossRef]
25.
Grover, R.; Mancini, I.C.; Moore, S.; Mead, A.J.; Atkinson, D.; Rasmussan, K.D.; O’Carrol, D.D.; Jacobsen, S.E.W.; Nerlov, C.
Erythropoietin guides multipotent progenitors towards an erythroid fate. J. Exp. Med. 2014,211, 181–188.

Int. J. Mol. Sci. 2025,26, 3346 13 of 15
26.
Rieger, M.A.; Hoppe, P.S.; Smeijkal, B.M.; Eitelhuber, A.C.; Schroeder, T. Haematopoietic cytokines can instruct lineage choice.
Science 2009,325, 217–218.
27. Brown, G.; Ceredig, R. Modelling the hematopoietic landscape. Front. Cell Dev. Biol. 2019,7, 104.
28.
Adolfsson, J.; Mansson, R.; Buza-Vidas, N.; Huitquist, A.; Liuba, K.; Jensen, C.T.; Bryder, D.; Yang, L.; Berge, O.-J.; Thoren, L.A.M.;
et al. Identification of Flt3+ lymphoid-myeloid stem cell lacking erythro-megakaryocytic potential: A revised road map for adult
blood lineage commitment. Cell 2005,121, 295–306.
29.
Muller-Sieburg, C.E.; Cho, R.H.; Karlsson, L.; Huang, J.-F.; Sieburg, H.B. Myeloid biased hematopoietic stem cells have extensive
self-renewal capacity but generate diminished lymphoid progeny with impaired IL-7 responsiveness. Blood 2004,104, 4111–4118.
30. Gieger, H.; de Hann, G.; Florian, M. The aging haematopoietic stem cell compartment. Nat. Rev. Immunol. 2013,13, 376–389.
31.
Dorshkind, K.; Hofer, T.; Montecino-Rodriguez, E.; Poili, D.D.; Rodewald, H.-R. Do haematopoietic stem cells age? Nat. Rev.
Immunol. 2020,20, 196–202.
32.
Montecino-Rodriguez, E.; Kong, Y.; Casero, D.; Rouoult, A.; Dorshkind, K.; Poili, P.D. Lymphoid-biased HSC are maintained with
age and efficiently generate lymphoid progeny. Stem Cell Rep. 2019,12, 584–598.
33.
Vanickova, K.; Milosevic, M.; Bas, I.R.; Burocziove, M.; Yokota, A.; Danek, P.; Grusanovic, S.; Chilinski, M.; Plewczynski, D.;
Rohlena, J.; et al. Hematopoietic stem cells undergo a lymphoid to myeloid switch in early stages of emergency granulopoiesis.
EMBO J. 2023,42, e113527. [PubMed]
34.
Sanjuan-Pla, A.; Macaulay, I.C.; Jensen, C.T.; Woll, P.S.; Luis, T.C.; Mead, A.; Moore, S.; Corella, C.; Matsuoka, S.; Jones, T.B.; et al.
Platelet-biased stem cells reside at the apex of the haematopoietic stem-cell hierarchy. Nature 2013,502, 232–236. [PubMed]
35.
Ninos, J.M.; Jefferis, L.C.; Cogle, C.R.; Kerr, W.G. The thrombopoietin receptor, c-Mpl, is a selective surface marker for human
hematopoietic stem cells. J. Trans. Med. 2006,4, 9.
36.
Knapp, D.J.H.F.; Mammond, C.A.; Hul, T.; van Loenhout, M.T.J.; Wang, F.; Aghaeepour, N.; Miller, P.H.; Moksa, M.; Rabu, G.M.;
Beer, P.A.; et al. Single-cell analysis identifies a CD33C subset of human cord blood cells with high regenerative potential. Nat.
Cell Biol. 2018,20, 710–720.
37.
Kondo, M.; Weissman, I.L.; Akashi, K. Identification of clonogenic common lymphoid progenitors in mouse bone marrow. Cell
1997,91, 661–672.
38.
Akashi, K.; Trover, D.; Miyamoto, Y.; Weissman, I.L. A clonogenic common myeloid progenitor that gives rise to all myeloid
lineages. Nature 2000,404, 193–197.
39.
Ceredig, R.; Rolink, A.G.; Brown, G. Models of haematopoiesis: Seeing the wood for the trees. Nat. Rev. Immunol. 2009,9, 293–299.
40.
Wong, A.K.-Y.; Bunce, C.M.; Lord, J.M.; Salt, J.; Brown, G. Evidence that precursor cells of monocytes and B lymphocytes are
closely related. Exp. Hematol. 1989,17, 968–973.
41.
Boyd, A.W.; Schrader, J.W. Derivation of macrophage-like lines from the pre-B lymphoma ABLS 8.1 using 5-azacytidine. Nature
1982,297, 691–693. [PubMed]
42.
Borzillo, G.V.; Ashmun, R.A.; Sherr, C.J. Macrophage lineage switching of murine early pre-B lymphoid cells expressing
transduced fms genes. Mol. Cell. Biol. 1990,10, 2703–2714.
43.
Fisher, A.G.; Burdet, C.; Bunce, C.; Merkenschlager, M.; Ceredig, R. Lymphoproliferative disorders in IL-7 transgenic mice:
Expansion of immature B cells which retain macrophage potential. Int. Immunol. 1995,7, 415–423.
44.
Cumano, A.; Paige, C.J.; Iscove, N.N.; Brady, G. Bipotential precursors of B cells and macrophages in murine fetal liver. Nature
1992,356, 612–615. [PubMed]
45.
Montecino-Rodriguez, E.; Leathers, H.; Dorshkind, K. Bipotential B-macrophage progenitors are present in adult bone marrow.
Nat. Immunol. 2001,2, 83–88. [PubMed]
46.
Paul, F.; Arkin, Y.; Giladi, A.; Jaitin, D.A.; Kenigsberg, E.; Keren-Shaul, H.; Winter, D.; Lara-Astiaso, D.; Gury, M.; Weiner, A.; et al.
Transcriptional heterogeneity and lineage commitment in myeloid progenitors. Cell 2015,163, 1663–1677.
47.
Kucinski, I.; Campos, J.; Barile, M.; Severi, F.; Bohin, N.; Mareira, P.N.; Allen, L.; Lawson, H.; Haltalli, M.L.R.; Kinston, S.J.; et al. A
time and single-cell resolved model of murine bone marrow hematopoiesis. Cell Stem Cell 2024,31, 244–259.
48.
Patel, S.H.; Christodoulou, C.; Weinreb, C.; Yu, Q.; Lummertz da Rocha, E.; Pepe-Mooney, B.J.; Bowling, S.; Li, L.; Osario, F.G.;
Doley, G.Q.; et al. Lifelong multilineage contribution by embryonic-born blood progenitors. Nature 2021,606, 747–753.
49.
Kapadia, C.D.; Willams, N.; Dawson, K.J.; Watson, C.; Yousefzadeh, M.J.; Le, D.; Nyamondo, K.; Kodavali, S.; Cogan, A.;
Waldvogel, S.; et al. Clonal dynamics and somatic evolution of haematopoiesis in mouse. bioRxiv 2025. [CrossRef]
50.
Weissman, I.L.; Anderson, D.J.; Gage, F. Stem and progenitor cells: Origins, phenotypes, lineage commitments and transdifferen-
tiations. Annu. Rev. Cell Dev. Biol. 2001,17, 387–403.
51.
Bradley, T.R.; Metcalf, D. The growth of mouse bone marrow cells
in vitro
.Aust. J. Exp. Biol. Med. Sci. 1966,44, 287–299.
[CrossRef] [PubMed]
52.
Ishikawa, F.; Niiro, H.; Iino, T.; Yashida, S.; Saito, N.; Onohara, S.; Miyamoto, T.; Minagawa, H.; Fujii, S.-I.; Shultz, L.D.; et al. The
developmental program of human dendritic cells is operated independently of conventional myeloid and lymphoid pathways.
Blood 2007,110, 3591–3660. [CrossRef]

Int. J. Mol. Sci. 2025,26, 3346 14 of 15
53.
Sulston, J.E.; Schierenberg, E.; White, J.G.; Thomson, J.N. The embryonic cell lineage of the nematode caenorhabditis elegans. Dev.
Biol. 1983,100, 64–119. [CrossRef]
54.
Westin, J.A. Motile and social behaviour of neural crest cells. In Cell Behavior; Bellairs, R., Curtis, A., Dunn, G., Eds.; Cambridge
University Press: London, UK, 1981; pp. 429–470.
55.
Paczko, M.; Vordos, D.; Szabo, P.; Jekely, G.; Szathmary, E.; Szilagyi, A. A neural network-based framework for cell-fate decisions
and development. Commun. Biol. 2024,7, 323. [CrossRef]
56.
Angeli, D.; Ferrell, J.E.; Sontag, E.D. Detection of multistability, bifurcations, and hysteresis in a large class of biological positive
feedback systems. Proc. Natl. Acad. Sci. USA 2004,101, 1822–1827. [CrossRef] [PubMed]
57.
Harrington, H.A.; Feliu, E.; Wiuf, C.; Stumpf, M.P. Cellular compartments cause multistability and allow cells to process more
information. Biophys. J. 2013,104, 1824–1831. [CrossRef] [PubMed]
58.
Craciun, G.; Tang, Y.; Feinberg, M. Understanding bistability in complex enzyme-driven reaction networks. Proc. Natl. Acad. Sci.
USA 2006,103, 8697–8702. [CrossRef]
59.
Wu, S.; Zhou, T.; Tian, T. A robust method for designing multistable systems by embedding bistable subsystems. Syst. Biol. Appl.
2022,8, 10. [CrossRef]
60.
Nestorowa, S.; Hamey, F.K.; Pijuan Sala, B.; Diamanti, E.; Shepherd, M.; Laurenti, E.; Wilson, N.K.; Kent, D.G.; Gottgens, B. A
single-cell resolution map of mouse hematopoietic stem and progenitor cell differentiation. Blood 2016,128, e20–e31. [CrossRef]
61.
Brown, G.; Hughes, P.J.; Michell, R.H.; Rolink, A.G.; Ceredig, R. The sequential determination model of hematopoiesis. Trends
Immunol. 2007,28, 442–448. [CrossRef]
62.
Grishechkin, A.; Mulcherjee, A.; Karin, O. Hierarchical control of state transitions in dense associated memories. arXiv 2025,
arXiv:2412.111336V2.
63.
Novershtern, N.; Subramanian, A.; Lawton, L.N.; Mak, R.H.; Haining, W.N.; McConkey, M.E.; Habib, N.; Yosef, N.; Chang, C.Y.;
Shay, T.; et al. Densely Interconnected Transcriptional Circuits Control Cell States in Human Hematopoiesis. Cell 2011,144,
296–309.
64.
Weinreb, C.; Rodriguez-Fraticelli, A.; Camargo, F.D.; Klein, A.M. Lineage tracing on transcriptional landscapes links state to fate
during differentiation. Science 2020,367, aaw3381.
65.
Gao, S.; Zhang, Y.; Liu, F. Revisiting the lineage contribution of hematopoietic stem and progenitor cells. Development 2023,150,
dev201609. [PubMed]
66.
Velten, L.; Haas, S.F.; Raffel, S.; Blaszkiewicz, S.; Islam, S.; Hennig, B.P.; Hirche, C.; Lutz, C.; Buss, E.C.; Nowak, D.; et al. Human
haematopoietic stem cell lineage commitment is a continuous process. Nat. Cell Biol. 2017,19, 271–281.
67.
Waddington, C.H. The Strategy of the Genes: A Discussion of Some Aspects of Theoretical Biology; Allen & Unwin: London, UK, 1957.
68.
Tsapogas, P.; Swee, L.K.; Nusser, A.; Nuber, N.; Kreuzaler, M.; Capoferri, G.; Rolink, H.; Ceredig, R.; Rolink, A.
In vivo
evidence
for an instructive role of fms-like tyrosine kinase-3 (FLT3) ligand in hematopoietic development. Haematologica 2014,99, 638–646.
[PubMed]
69.
Tsapogas, P.; Mooney, C.J.; Brown, G.; Rolink, A. The cytokine Flt3-ligand in normal and malignant hematopoiesis. Int. J. Mol. Sci.
2017,18, 1115. [CrossRef]
70.
Onai, N.; Obata-Onai, A.; Tussiwand, R.; Lanzavecchia, A.; Manz, M.G. Activation of the FLT3 signal transduction cascade
rescues and enhances type 1 interferon-producing and dendritic cell development. J. Exp. Med. 2006,203, 227–238.
71.
Momenilandi, M.; Levy, R.; Sobrino, S.; Li, J.; Lagresle-Peyrou, C.; Esmaelzadeh, H.; Fayard, A.; Le Floc’h, C.; Guerin, A.; Mina,
E.D.; et al. FLT3 governs the development of partially overlapping hematopoietic lineages in humans and mice. Cell 2024,187,
2817–2837.
72.
Calabria, A.; Spinozzi, G.; Casana, D.; Buscaroli, E.; Benedicenti, F.; Pais, G.; Gazzo, F.; Scala, S.; Lidonnici, M.R.; Scaramuzza, S.;
et al. Long-term lineage commitment in haematopoietic stem cell gene therapy. Nature 2024,636, 162–171. [CrossRef]
73.
Brown, G.; Ceredig, R.; Tsapogas, P. The making of hematopoiesis: Developmental ancestry and environmental nurture. Int. J.
Mol. Sci. 2018,19, 2122. [CrossRef] [PubMed]
74.
Porritt, H.E.; Rumfelt, L.L.; Tabrizifard, S.; Smitt, T.M.; Zuniga-Pflucker, J.C.; Petrie, H.T. Heterogeneity among DN1 prothy-
mocytes reveals multiple progenitors with different capacities to generate T cell and non-T cell lineages. Immunity 2004,20,
735–745.
75.
Lehar, S.M.; Dooley, J.; Farr, A.G.; Bevan, M.J. Notch ligands Delta 1 and Jagged1 transmit distinct signals to T-cell precursors.
Blood 2005,105, 1440–1447. [CrossRef]
76.
Balciunaite, G.; Ceredig, R.; Rolink, A.G. The earliest subpopulation of mouse thymocytes contains potent T, significant
macrophage, and natural killer cell but no B-lymphocyte potential. Blood 2005,105, 1930–1936. [PubMed]
77.
Glatman Zaretsky, A.; Taylor, J.J.; King, I.L.; Marshall, F.A.; Mohrs, M.; Pearce, E.J. T follicular helper cells differentiate from th2
cells in response to helminth antigens. J. Exp. Med. 2009,206, 991–999.

Int. J. Mol. Sci. 2025,26, 3346 15 of 15
78.
Singh, K.; Gatzka, M.; Peters, T.; Borkner, L.; Hainzl, A.; Wang, H.; Sindrilaru, A.; Scharffetter-Kochanek, K. Reduced cd18 levels
drive regulatory t cell conversion into th17 cells in the cd18hypo pl/j mouse model of psoriasis. J. Immunol. 2013,190, 2544–2553.
[PubMed]
79.
García-Ramírez, I.; Bhatia, S.; Rodríguez-Hernández, G.; González-Herrero, I.; Walter, C.; González de Tena-Dávila, S.; Parvin, S.;
Haas, O.; Woessmann, W.; Stanulla, M.; et al. Lmo2 expression defines tumor cell identity during T-cell leukemogenesis. EMBO J.
2018,37, e98783. [PubMed]
80.
Martín-Lorenzo, A.; Auer, F.; Chan, L.N.; García-Ramírez, I.; González-Herrero, I.; Rodríguez-Hernández, G.; Bartenhagen, C.;
Dugas, M.; Gombert, M.; Ginzel, S.; et al. Loss of Pax5 Exploits Sca1-BCR-ABLp190 Susceptibility to Confer the Metabolic Shift
Essential for pB-ALL. Cancer Res. 2018,78, 2669–2679.
81.
Perez-Caro, M.; Cobaleda, C.; Gonzalez-Herrero, I.; Vicente-Duenas, C.; Bermejo-Rodriguez, C.; Sanchez-Beato, M.; Orfao,
A.; Pintado, B.; Flores, T.; Sánchez-Martín, M.; et al. Cancer induction by restriction of oncogene expression to the stem cell
compartment. EMBO J. 2009,28, 8–20.
82.
Brown, G.; Sanchez, L.; Sanchez-Garcia, I. Lineage decision-making within normal hematopoietic and leukaemic stem cells. Int. J.
Mol. Sci. 2020,21, 2247.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.