Advertisement
Review Anniversary Review Series: Perspectives on the modern exploration of Experimental Hematology| Volume 42, ISSUE 2, P74-82.e2, February 01, 2014

Heterogeneity and hierarchy of hematopoietic stem cells

  • Hideo Ema
    Correspondence
    Offprint requests to: Hideo Ema, M.D., Department of Cell Differentiation, The Sakaguchi Laboratory of Developmental Biology, Keio University School of Medicine, 35 Shinano-machi, Shinjuku-ku, Tokyo 160-8582 Japan
    Affiliations
    Department of Cell Differentiation, Sakaguchi Laboratories of Developmental Biology, Keio University School of Medicine, Tokyo, Japan
    Search for articles by this author
  • Yohei Morita
    Affiliations
    Leibniz Institute for Age Research, Fritz Lipmann Institute, Jenna, Germany
    Search for articles by this author
  • Toshio Suda
    Affiliations
    Department of Cell Differentiation, Sakaguchi Laboratories of Developmental Biology, Keio University School of Medicine, Tokyo, Japan
    Search for articles by this author
Open ArchivePublished:November 22, 2013DOI:https://doi.org/10.1016/j.exphem.2013.11.004
      Hematopoietic stem cells (HSCs) are a more heterogeneous population than previously thought. Extensive analysis of reconstitution kinetics after transplantation allows a new classifications of HSCs based on lineage balance. Previously unrecognized classes of HSCs, such as myeloid- and lymphoid-biased HSCs, have emerged. However, varying nomenclature has been used to describe these cells, promoting confusion in the field. To establish a common nomenclature, we propose a reclassification of short-, intermediate-, and long-term (ST, IT, and LT) HSCs defined as: ST < 6 months, IT > 6 months, and LT > 12. We observe that myeloid-biased HSCs or α cells overlap with LT-HSCs, whereas lymphoid-biased HSCs or γ/δ cells overlap with ST-HSCs, suggesting that HSC lifespan is linked to cell differentiation. We also suggest that HSC heterogeneity prompts reconsideration of long-term (>4 months) multilineage reconstitution as the gold standard for HSC detection. In this review, we discuss relationships among ST-, IT-, and LT-HSCs relevant to stem cell heterogeneity, hierarchical organization, and differentiation pathways.
      Hematopoietic stem cells (HSCs) are defined as cells with self-renewal and differentiation potential [
      • Till J.E.
      • McCulloch E.A.
      • Siminovitch L.
      A stochastic model of stem cell proliferation, based on the growth of spleen colony-forming cells.
      ]. Accumulated data show that HSCs are a heterogeneous population in multiple aspects, including their degree of self-renewal [
      • Guenechea G.
      • Gan O.I.
      • Dorrell C.
      • Dick J.E.
      Distinct classes of human stem cells that differ in proliferative and self-renewal potential.
      ,
      • Ema H.
      • Sudo K.
      • Seita J.
      • et al.
      Quantification of self-renewal capacity in single hematopoietic stem cells from normal and Lnk-deficient mice.
      ], differentiation manner [
      • Muller-Sieburg C.E.
      • Sieburg H.B.
      • Bernitz J.M.
      • Cattarossi G.
      Stem cell heterogeneity: implications for aging and regenerative medicine.
      ,
      • Copley M.R.
      • Beer P.A.
      • Eaves C.J.
      Hematopoietic stem cell heterogeneity takes center stage.
      ], and lifespan [
      • Osawa M.
      • Hanada K.
      • Hamada H.
      • Nakauchi H.
      Long-term lymphohematopoietic reconstitution by a single CD34-low/negative hematopoietic stem cell.
      ,
      • Morrison S.J.
      • Wandycz A.M.
      • Hemmati H.D.
      • Wright D.E.
      • Weissman I.L.
      Identification of a lineage of multipotent hematopoietic progenitors.
      ,
      • Yang L.
      • Bryder D.
      • Adolfsson J.
      • et al.
      Identification of Lin(-)Sca1(+)kit(+)CD34(+)Flt3- short-term hematopoietic stem cells capable of rapidly reconstituting and rescuing myeloablated transplant recipients.
      ]. Retroviral marking studies indicate that HSCs clonally give rise to all blood lineages and self-renew (a finding that represents definitive proof for the existence of HSCs in mouse bone marrow) [
      • Dick J.E.
      • Magli M.C.
      • Huszar D.
      • Phillips R.A.
      • Bernstein A.
      Introduction of a selectable gene into primitive stem cells capable of long-term reconstitution of the hemopoietic system of W/Wv mice.
      ,
      • Keller G.
      • Paige C.
      • Gilboa E.
      • Wagner E.F.
      Expression of a foreign gene in myeloid and lymphoid cells derived from multipotent haematopoietic precursors.
      ,
      • Lemischka I.R.
      • Raulet D.H.
      • Mulligan R.C.
      Developmental potential and dynamic behavior of hematopoietic stem cells.
      ]. Moreover, marking techniques have been used to demonstrate various patterns of reconstitution kinetics after HSC transplantation. Interestingly, some clones preferentially reconstitute a lymphoid lineage, whereas others preferentially reconstitute a myeloid one [
      • Jordan C.T.
      • Lemischka I.R.
      Clonal and systemic analysis of long-term hematopoiesis in the mouse.
      ].
      Lineage reconstitution kinetics have been examined extensively in mice transplanted with cultured bone marrow cells or with limiting doses of bone marrow cells freshly obtained from adult mice [
      • Muller-Sieburg C.E.
      • Cho R.H.
      • Thoman M.
      • Adkins B.
      • Sieburg H.B.
      Deterministic regulation of hematopoietic stem cell self-renewal and differentiation.
      ,
      • 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.
      ,
      • Cho R.H.
      • Sieburg H.B.
      • Muller-Sieburg C.E.
      A new mechanism for the aging of hematopoietic stem cells: aging changes the clonal composition of the stem cell compartment but not individual stem cells.
      ]. These studies suggest the presence of myeloid-biased HSCs (My-bi HSCs), lymphoid-biased HSCs (Ly-bi HSCs), and balanced HSCs (Bala HSCs). On the other hand, α, β, γ, and δ cells have been defined by others [
      • Dykstra B.
      • Kent D.
      • Bowie M.
      • et al.
      Long-term propagation of distinct hematopoietic differentiation programs in vivo.
      ,
      • Benz C.
      • Copley M.R.
      • Kent D.G.
      • et al.
      Hematopoietic stem cell subtypes expand differentially during development and display distinct lymphopoietic programs.
      ]. The presence of all these HSCs has been verified by single-cell transplantation [
      • Dykstra B.
      • Kent D.
      • Bowie M.
      • et al.
      Long-term propagation of distinct hematopoietic differentiation programs in vivo.
      ,
      • Challen G.A.
      • Boles N.C.
      • Chambers S.M.
      • Goodell M.A.
      Distinct hematopoietic stem cell subtypes are differentially regulated by TGF-beta1.
      ,
      • Morita Y.
      • Ema H.
      • Nakauchi H.
      Heterogeneity and hierarchy within the most primitive hematopoietic stem cell compartment.
      ,
      • Benveniste P.
      • Frelin C.
      • Janmohamed S.
      • et al.
      Intermediate-term hematopoietic stem cells with extended but time-limited reconstitution potential.
      ]. Both types of classification are defined based on myeloid and lymphoid reconstitution ratios, but the criteria used to make these classifications differ fundamentally from one another (discussed later).
      In this study, we propose a third classification, LT-, IT-, and ST-HSCs, based on reconstitution time periods [
      • Yamamoto R.
      • Morita Y.
      • Ooehara J.
      • et al.
      Clonal analysis unveils self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells.
      ]. We then examine the relationship of the three classification systems and discuss how different HSC classes are related to one another in the hematopoietic hierarchy. These comparisons support that HSC lifespan is tightly associated with lineage contribution [
      • Muller-Sieburg C.E.
      • Cho R.H.
      • Thoman M.
      • Adkins B.
      • Sieburg H.B.
      Deterministic regulation of hematopoietic stem cell self-renewal and differentiation.
      ,
      • 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.
      ,
      • Cho R.H.
      • Sieburg H.B.
      • Muller-Sieburg C.E.
      A new mechanism for the aging of hematopoietic stem cells: aging changes the clonal composition of the stem cell compartment but not individual stem cells.
      ,
      • Dykstra B.
      • Kent D.
      • Bowie M.
      • et al.
      Long-term propagation of distinct hematopoietic differentiation programs in vivo.
      ,
      • Benz C.
      • Copley M.R.
      • Kent D.G.
      • et al.
      Hematopoietic stem cell subtypes expand differentially during development and display distinct lymphopoietic programs.
      ].
      In the prevailing bifurcation model [
      • Akashi K.
      • Traver D.
      • Miyamoto T.
      • Weissman I.L.
      A clonogenic common myeloid progenitor that gives rise to all myeloid lineages.
      ], following loss of self-renewal potential HSCs give rise to multipotent progenitors (MPPs), which commit to either myeloid or lymphoid lineages exclusively. According to this model, this is the first step in lineage commitment. However, MPPs or their progenitor equivalents have not been identified at the single-cell level. Other studies suggest that loss of lymphoid differentiation potential could occur as one of the first lineage commitment steps [
      • Kawamoto H.
      • Ohmura K.
      • Fujimoto S.
      • Katsura Y.
      Emergence of T cell progenitors without B cell or myeloid differentiation potential at the earliest stage of hematopoiesis in the murine fetal liver.
      ,
      • Adolfsson J.
      • Månsson R.
      • Buza-Vidas N.
      • et al.
      Identification of Flt3+ lympho-myeloid stem cells lacking erythro-megakaryocytic potential a revised road map for adult blood lineage commitment.
      ]. It is time to consider a more comprehensive differentiation model. We propose a new differentiation model consisting of LT-, IT-, and ST-HSCs.

      HSC classifications

      My-bi, Bala, and Ly-bi HSCs

      Muller-Sieburg et al. [
      • Muller-Sieburg C.E.
      • Cho R.H.
      • Thoman M.
      • Adkins B.
      • Sieburg H.B.
      Deterministic regulation of hematopoietic stem cell self-renewal and differentiation.
      ,
      • 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.
      ] have defined My-bi, Bala, and Ly-bi HSCs based on the ratio of lymphoid to myeloid cells (the L/M ratio). The proportions of lymphoid to myeloid cells are calculated among test-donor-derived cells (Supplementary Figure E1, online only, available at www.exphem.org); thus, (% lymphoid cells) + (% myeloid cells) = 100. In this classification, long-term reconstitution is assessed 20 weeks or more after transplantation [
      • Muller-Sieburg C.E.
      • Cho R.H.
      • Thoman M.
      • Adkins B.
      • Sieburg H.B.
      Deterministic regulation of hematopoietic stem cell self-renewal and differentiation.
      ,
      • 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.
      ,
      • Cho R.H.
      • Sieburg H.B.
      • Muller-Sieburg C.E.
      A new mechanism for the aging of hematopoietic stem cells: aging changes the clonal composition of the stem cell compartment but not individual stem cells.
      ]. Transplanted cells are designated My-bi HSCs when the L/M ratio is less than 3 and Ly-bi HSCs when it exceeds 10. Cells are considered Bala HSCs when the L/M ratio exceeds 3 but is less than 10 [
      • Muller-Sieburg C.E.
      • Cho R.H.
      • Thoman M.
      • Adkins B.
      • Sieburg H.B.
      Deterministic regulation of hematopoietic stem cell self-renewal and differentiation.
      ,
      • 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.
      ,
      • Cho R.H.
      • Sieburg H.B.
      • Muller-Sieburg C.E.
      A new mechanism for the aging of hematopoietic stem cells: aging changes the clonal composition of the stem cell compartment but not individual stem cells.
      ].
      These types of HSCs were detected basically using in vivo limiting dilution analysis [
      • Muller-Sieburg C.E.
      • Cho R.H.
      • Thoman M.
      • Adkins B.
      • Sieburg H.B.
      Deterministic regulation of hematopoietic stem cell self-renewal and differentiation.
      ,
      • 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.
      ,
      • Cho R.H.
      • Sieburg H.B.
      • Muller-Sieburg C.E.
      A new mechanism for the aging of hematopoietic stem cells: aging changes the clonal composition of the stem cell compartment but not individual stem cells.
      ]. Later, a different group reported that Ly-HSCs and My-HSCs are enriched in the upper and lower portions of SP, respectively, and successfully accomplished single-cell reconstitution with these HSCs [
      • Challen G.A.
      • Boles N.C.
      • Chambers S.M.
      • Goodell M.A.
      Distinct hematopoietic stem cell subtypes are differentially regulated by TGF-beta1.
      ]. Platelet-biased HSCs have also been reported as a My-bi subclass potentially residing at the apex of the hematopoietic hierarchy [
      • Sanjuan-Pla A.
      • Macaulay I.C.
      • Jensen C.T.
      • et al.
      Platelet-biased stem cells reside at the apex of the haematopoietic stem-cell hierarchy.
      ].
      Figure 1 shows typical reconstitution patterns seen following transplantation of single My-bi, Bala, and Ly-bi HSCs. My-bi HSCs reconstitute the myeloid lineage after varying latencies, followed by gradual reconstitution of the lymphoid lineage (Fig. 1A). Thus, the myeloid lineage is more significantly reconstituted at early stages of reconstitution. In contrast, Ly-bi HSCs show both myeloid and lymphoid lineage reconstitution from early stages (Fig. 1C). Ly-bi HSCs reconstitute the myeloid lineage to a less extent than the lymphoid lineage. Myeloid reconstitution is often detectable for only a few months, but lymphoid reconstitution can persist relatively longer. Bala HSCs reconstitute the lymphoid lineage soon after the myeloid lineage (Fig. 1B). The proportions of myeloid and lymphoid lineage cells resemble those seen in the peripheral blood of normal mice [
      • Muller-Sieburg C.E.
      • Cho R.H.
      • Thoman M.
      • Adkins B.
      • Sieburg H.B.
      Deterministic regulation of hematopoietic stem cell self-renewal and differentiation.
      ,
      • 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.
      ,
      • Cho R.H.
      • Sieburg H.B.
      • Muller-Sieburg C.E.
      A new mechanism for the aging of hematopoietic stem cells: aging changes the clonal composition of the stem cell compartment but not individual stem cells.
      ]. Many investigators consider Bala HSCs to be typical HSCs, which might account for why the presence of My-bi HSCs and Ly-bi HSCs has been overlooked [
      • Muller-Sieburg C.E.
      • Sieburg H.B.
      • Bernitz J.M.
      • Cattarossi G.
      Stem cell heterogeneity: implications for aging and regenerative medicine.
      ].
      Figure thumbnail gr1
      Figure 1Reconstitution kinetics of My-bi, Bala, and Ly-bi HSCs. Shown are typical reconstitution patterns seen following single-cell transplantation of My-bi (A), Bala (B), and Ly-bi (C) HSCs, based on published data
      [
      • Morita Y.
      • Ema H.
      • Nakauchi H.
      Heterogeneity and hierarchy within the most primitive hematopoietic stem cell compartment.
      ]
      .

      α, β, and γ/δ cells

      Eaves et al. have defined α, β, γ, and δ cells as the percentage of myeloid chimerism relative to that of lymphoid chimerism (the M/L ratio) [
      • Dykstra B.
      • Kent D.
      • Bowie M.
      • et al.
      Long-term propagation of distinct hematopoietic differentiation programs in vivo.
      ,
      • Benz C.
      • Copley M.R.
      • Kent D.G.
      • et al.
      Hematopoietic stem cell subtypes expand differentially during development and display distinct lymphopoietic programs.
      ]. The M/L ratio is not simply the reciprocal of the L/M ratio described by Muller-Sieburg et al. [
      • Muller-Sieburg C.E.
      • Cho R.H.
      • Thoman M.
      • Adkins B.
      • Sieburg H.B.
      Deterministic regulation of hematopoietic stem cell self-renewal and differentiation.
      ,
      • 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.
      ,
      • Cho R.H.
      • Sieburg H.B.
      • Muller-Sieburg C.E.
      A new mechanism for the aging of hematopoietic stem cells: aging changes the clonal composition of the stem cell compartment but not individual stem cells.
      ], because competitor cell contribution is factored in to the M/L ratio. Thus, the percentage of myeloid chimerism is defined as: (% Test cell–derived cells in the granulocyte-macrophage lineage) × 100 / (% Test cell–derived cells + % Competitor cell-derived cells in the granulocyte-macrophage lineage). (See Supplementary Figure E1, online only, available at www.exphem.org.) Percentages of B and T lymphoid cells are defined similarly. The percentage of lymphoid chimerism is defined as (% B lymphoid chimerism) + (% T lymphoid chimerism). In this system, long-term reconstitution is assessed 16 weeks or more after transplantation [
      • Dykstra B.
      • Kent D.
      • Bowie M.
      • et al.
      Long-term propagation of distinct hematopoietic differentiation programs in vivo.
      ,
      • Benz C.
      • Copley M.R.
      • Kent D.G.
      • et al.
      Hematopoietic stem cell subtypes expand differentially during development and display distinct lymphopoietic programs.
      ].
      Single-donor cells are designated α cells when the M/L ratio exceeds 2, and γ or δ cells when it is less than 0.25. When myeloid chimerism exceeds 1%, cells are designated γ cells; when it is less than 1.0%, they are designated δ cells [
      • Dykstra B.
      • Kent D.
      • Bowie M.
      • et al.
      Long-term propagation of distinct hematopoietic differentiation programs in vivo.
      ,
      • Benz C.
      • Copley M.R.
      • Kent D.G.
      • et al.
      Hematopoietic stem cell subtypes expand differentially during development and display distinct lymphopoietic programs.
      ]. Single-donor cells are designated β cells when the M/L ratio exceeds 0.25 but is less than 2. α and β cells are transplantable into secondary recipient mice, but γ and δ cells are not. It has been reported that γ/δ cells are enriched in CD150low/negativeCD34KSL cells, while α and β cells are enriched in CD150high and CD150med CD34KSL cells [
      • Morita Y.
      • Ema H.
      • Nakauchi H.
      Heterogeneity and hierarchy within the most primitive hematopoietic stem cell compartment.
      ,
      • Kent D.G.
      • Copley M.R.
      • Benz C.
      • et al.
      Prospective isolation and molecular characterization of hematopoietic stem cells with durable self-renewal potential.
      ].

      ST-, IT-, and LT-HSCs

      Researchers have long recognized the concept of ST-HSCs and LT-HSCs [
      • Osawa M.
      • Hanada K.
      • Hamada H.
      • Nakauchi H.
      Long-term lymphohematopoietic reconstitution by a single CD34-low/negative hematopoietic stem cell.
      ,
      • Morrison S.J.
      • Wandycz A.M.
      • Hemmati H.D.
      • Wright D.E.
      • Weissman I.L.
      Identification of a lineage of multipotent hematopoietic progenitors.
      ,
      • Yang L.
      • Bryder D.
      • Adolfsson J.
      • et al.
      Identification of Lin(-)Sca1(+)kit(+)CD34(+)Flt3- short-term hematopoietic stem cells capable of rapidly reconstituting and rescuing myeloablated transplant recipients.
      ]. Recently, an intermediate-term, (IT)-HSC, which contributes to reconstitution up to 8 months after transplantation, has been used [
      • Benveniste P.
      • Frelin C.
      • Janmohamed S.
      • et al.
      Intermediate-term hematopoietic stem cells with extended but time-limited reconstitution potential.
      ]. IT-HSCs express integrin α2 (CD49b) among Rho123low, CD34Kit+Sca-1+lineage (CD34-KSL) cells. Given these findings, we propose the use of ST-, IT-, and LT-HSCs as a classification based on reconstitution time period (Fig. 2) [
      • Yamamoto R.
      • Morita Y.
      • Ooehara J.
      • et al.
      Clonal analysis unveils self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells.
      ]. We suggest the following definitions. Only granulocyte reconstitution is considered as a parameter of this classification, not reconstitution of other lineages such as erythrocytes, B cells, and T cells, because granulocytes are extremely short-lived and their reconstitution directly reflects HSC activity [
      • Jordan C.T.
      • Lemischka I.R.
      Clonal and systemic analysis of long-term hematopoiesis in the mouse.
      ,
      • Dykstra B.
      • Kent D.
      • Bowie M.
      • et al.
      Long-term propagation of distinct hematopoietic differentiation programs in vivo.
      ]. ST would be defined as less than 6 months (note that this is much longer than the previous definition of 1–2 months) [
      • Osawa M.
      • Hanada K.
      • Hamada H.
      • Nakauchi H.
      Long-term lymphohematopoietic reconstitution by a single CD34-low/negative hematopoietic stem cell.
      ,
      • Morrison S.J.
      • Wandycz A.M.
      • Hemmati H.D.
      • Wright D.E.
      • Weissman I.L.
      Identification of a lineage of multipotent hematopoietic progenitors.
      ]. Granulocyte reconstitution levels would decrease by 6 months following ST-HSC transplantation. IT would be defined as less than 12 months. Granulocyte reconstitution levels would decrease by 12 months after IT-HSC transplantation. LT would be defined as greater than 12 months. Granulocyte reconstitution would not decrease until 12 months or later after LT-HSC transplantation. Secondary transplantation could be performed within 12 months, such as at 5–6 months after transplantation, but the same criteria would apply [
      • Morita Y.
      • Ema H.
      • Nakauchi H.
      Heterogeneity and hierarchy within the most primitive hematopoietic stem cell compartment.
      ,
      • Yamamoto R.
      • Morita Y.
      • Ooehara J.
      • et al.
      Clonal analysis unveils self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells.
      ]. For example, if granulocyte reconstitution began to decrease by 6 months after secondary transplantation, this HSC would be designated an IT-HSC. Generally, ST-HSCs do not show any reconstitution activity after secondary transplantation, whereas reconstitution levels do not change after secondary transplantation with LT-HSCs [
      • Yamamoto R.
      • Morita Y.
      • Ooehara J.
      • et al.
      Clonal analysis unveils self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells.
      ].
      Figure thumbnail gr2
      Figure 2Roles of ST-, IT-, and LT-HSCs in bone marrow transplantation. After bone marrow transplantation, total chimerism occurs because of several types of repopulating cells. In this model, total chimerism is represented by granulocyte reconstitution. (a) After ST-HSC transplantation, a small single wave is observed. (b) After IT-HSC transplantation, a single, larger wave is observed. (c, d) LT-HSC transplantation produces a sigmoid curve. (d) Reconstitution from latent HSCs has a delayed onset. Column in grey is an example of the time window for analysis.

      Relationship of the three HSC classifications

      My-bi/Bala/Ly-bi and α/β/γ/δ HSC classifications are based on similar concepts. Thus, α, β, and γ/δ cells likely correspond to My-bi HSCs, Bala HSCs, and Ly-bi HSCs, respectively. To assess this correspondence more precisely, we compared these classifications using published data of transplantation with 30 single HSCs [
      • Morita Y.
      • Ema H.
      • Nakauchi H.
      Heterogeneity and hierarchy within the most primitive hematopoietic stem cell compartment.
      ]. Those comparisons are shown by Venn diagram analysis in Figure 3. All possible relationships among classifications are also shown in Supplementary Figure E2 (online only, available at www.exphem.org). Ten α cells were included in a total of 14 My-bi HSCs (Fig. 3A), and four Bala HSCs were included in a total of 7 β cells (Fig. 3B). Ly-bi HSCs overlap primarily with γ cells (Fig. 3C). These data support the idea that both classifications, despite the fact that different criteria are used to define them, identify similar classes of HSCs.
      Figure thumbnail gr3
      Figure 3Comparison of different HSC classifications. Thirty single cells were transplanted and resultant reconstitution data were classified based on three different criteria: 14 My-bi HSCs, four Bala HSC, and 12 Ly-bi HSCs were identified; 10 α cells, 7 β cells, and 13 γ cells were identified; and 8 LT-HSCs, 10 IT-HSC, and 12 ST-HSCs were identified. Venn diagram indicates relationships among (A) My-bi, α, and LT-HSCs cells; (B) Bala, β, and IT-HSCs; and (C) Ly-bi, γ, and ST-HSCs. See for more detail (online only, available at www.exphem.org).
      Interestingly, 7 of 8 LT-HSCs were also classified as My-bi HSCs; 5 of those 8 were classified as α cells and 2 as β cells (Supplementary Figure E2). Most LT-HSCs thus reconstitute a myeloid lineage prior to the lymphoid lineage. Of 10 IT-HSCs, 6 were classified as My-bi HSCs, 3 as Bala HSCs, and 1 as Ly-bi HSC. Of those 10 IT-HSCs, 4 were classified as α, 5 as β cells, and 1 as a γ cell. These data suggest that IT-HSCs exhibit varying reconstitution patterns. Moreover, we found that an almost identical population of cells is identified by Ly-bi HSCs, γ cells, and ST-HSCs (Fig. 3C).

      Relationship between My-bi and Ly-bi HSCs or between α and γ cells

      Whether My-bi HSCs can give rise to Ly-bi HSCs remains controversial [
      • 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.
      ,
      • Dykstra B.
      • Kent D.
      • Bowie M.
      • et al.
      Long-term propagation of distinct hematopoietic differentiation programs in vivo.
      ,
      • Benz C.
      • Copley M.R.
      • Kent D.G.
      • et al.
      Hematopoietic stem cell subtypes expand differentially during development and display distinct lymphopoietic programs.
      ]. In mice, after transplantation of a portion of bone marrow cells from primary into multiple secondary recipients, reconstitution patterns similar to those seen in the primary transplantation are observed, suggesting that My-bi and Ly-bi HSCs use intrinsic differentiation programs and do not undergo interconversion [
      • 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.
      ,
      • Cho R.H.
      • Sieburg H.B.
      • Muller-Sieburg C.E.
      A new mechanism for the aging of hematopoietic stem cells: aging changes the clonal composition of the stem cell compartment but not individual stem cells.
      ].
      It is technically difficult to detect a small number of γ/δ cells in the presence of a large number of α or β cells. To address this issue, investigators have transplanted either a small number of purified cells or single purified cells from the bone marrow of primary recipients into secondary lethally irradiated mice. Interestingly, production of γ/δ cells by α or β cells was detected in some cases [
      • Dykstra B.
      • Kent D.
      • Bowie M.
      • et al.
      Long-term propagation of distinct hematopoietic differentiation programs in vivo.
      ,
      • Benz C.
      • Copley M.R.
      • Kent D.G.
      • et al.
      Hematopoietic stem cell subtypes expand differentially during development and display distinct lymphopoietic programs.
      ], suggesting that some α or β cells can differentiate into a lymphoid lineage via γ/δ cells. Following transplantation of single α or β cells, the entire hematopoietic system can be reconstituted for more than 1 year. When whole bone marrow is reconstituted for a long time, it is difficult to think that a particular population is not reconstituted. Therefore, it is unlikely that only γ/δ cells are missing from bone marrow cells reconstituted with α/β cells.

      Revisiting criteria for HSC detection

      One important question is whether we can detect all HSC classes using criteria commonly used in competitive repopulation.

      Repopulating cells in bone marrow reconstitution

      When a sufficient number of bone marrow cells (e.g., 1 × 106 per mouse) is transplanted into lethally irradiated mice, the peripheral blood, which represents the entire hematopoietic system, is fully reconstituted over time. In addition to stem cells, a variety of progenitors, including those exhibiting radioprotection activity (e.g., colony-forming units in spleen [
      • Till J.E.
      • McCullouch E.A.
      A direct measurement of the radiation sensitivity of normal mouse bone marrow cells.
      ]), are present in transplanted bone marrow, enabling recipients to survive and show complete hematopoietic reconstitution. Figure 2 shows a typical reconstitution pattern after bone marrow transplantation. Results of single-cell transplantation suggest that a large wave of reconstitution occurs via several kinds of repopulating cells with different kinetics [
      • Dykstra B.
      • Kent D.
      • Bowie M.
      • et al.
      Long-term propagation of distinct hematopoietic differentiation programs in vivo.
      ,
      • Morita Y.
      • Ema H.
      • Nakauchi H.
      Heterogeneity and hierarchy within the most primitive hematopoietic stem cell compartment.
      ]. Colony-forming units in spleen play a role in rescuing lethally irradiated mice, likely for the initial 1–3 weeks after transplantation [
      • Yang L.
      • Bryder D.
      • Adolfsson J.
      • et al.
      Identification of Lin(-)Sca1(+)kit(+)CD34(+)Flt3- short-term hematopoietic stem cells capable of rapidly reconstituting and rescuing myeloablated transplant recipients.
      ]. Short-term repopulating cells likely play a role at 3–6 weeks after transplantation [
      • Morrison S.J.
      • Wandycz A.M.
      • Hemmati H.D.
      • Wright D.E.
      • Weissman I.L.
      Identification of a lineage of multipotent hematopoietic progenitors.
      ], and multilineage reconstitution is relayed over months by ST-, IT-, and LT-HSCs. It should be emphasized that all these repopulating cells play essential roles in the process of long-term hematopoietic reconstitution at distinct times.

      Competitive repopulation

      Competitive repopulation, an approach originally developed by Micklem et al. [
      • Micklem H.S.
      • Ford C.E.
      • Evans E.P.
      • Ogden D.A.
      • Papworth D.S.
      Competitive in vivo proliferation of foetal and adult haematopoietic cells in lethally irradiated mice.
      ], is used because cotransplantation of competitor cells ensures long-term survival of lethally irradiated mice regardless of whether test donor cells (or test cells) have radioprotection activity or long-term reconstitution potential. Moreover, competitive repopulation permits quantitation of repopulating activity in test donor cells compared with competitor cells. Both repopulating units (RU) [
      • Harrison D.E.
      • Jordan C.T.
      • Zhong R.K.
      • Astle C.M.
      Primitive hemopoietic stem cells: direct assay of most productive populations by competitive repopulation with simple binomial, correlation and covariance calculations.
      ] and competitive repopulating units (CRU) [
      • Szilvassy S.J.
      • Humphries R.K.
      • Lansdorp P.M.
      • Eaves A.C.
      • Eaves C.J.
      Quantitative assay for totipotent reconstituting hematopoietic stem cells by a competitive repopulation strategy.
      ] have been defined based on competitive repopulation assays. The amount of relative repopulating activity in HSCs can be expressed in RU, while the number of HSCs can be expressed in CRU. The mean activity per stem cell, a number useful to compare HSC qualities, is defined as RU/CRU [
      • Ema H.
      • Nakauchi H.
      Expansion of hematopoietic stem cells in the developing liver of a mouse embryo.
      ].

      The current gold standard for HSC detection

      Long-term (≥16 weeks) multilineage reconstitution is the gold standard to detect HSCs. Specifically, greater than 1% of total chimerism in peripheral leukocytes 4 months after transplantation with detectable myeloid, B lymphoid, and T lymphoid lineage reconstitution is the acceptable criteria to detect HSCs [
      • Dykstra B.
      • Kent D.
      • Bowie M.
      • et al.
      Long-term propagation of distinct hematopoietic differentiation programs in vivo.
      ,
      • Takano H.
      • Ema H.
      • Sudo K.
      • Nakauchi H.
      Asymmetric division and lineage commitment at the level of hematopoietic stem cells: inference from differentiation in daughter cell and granddaughter cell pairs.
      ]. However, we suggest that this standard could be reconsidered given that Ly-bi HSCs cannot be detected by this criterion. All myeloid, B lymphoid, and T lymphoid lineage reconstitution after transplantation might not be detected at 4 months, but instead be detected sequentially, such as myeloid reconstitution at 1–2 months and B and T lymphoid reconstitution at 1–4 months. Latent HSCs, a particular type of LT-HSCs, also cannot be detected as multilineage repopulating cells, because they exhibit only a low level of myeloid reconstitution 4 months after transplantation [
      • Dykstra B.
      • Kent D.
      • Bowie M.
      • et al.
      Long-term propagation of distinct hematopoietic differentiation programs in vivo.
      ,
      • Morita Y.
      • Ema H.
      • Nakauchi H.
      Heterogeneity and hierarchy within the most primitive hematopoietic stem cell compartment.
      ]. However, latent HSCs do exhibit significant reconstitution of all myeloid, B lymphoid, and T lymphoid lineages later or even after secondary transplantation [
      • Morita Y.
      • Ema H.
      • Nakauchi H.
      Heterogeneity and hierarchy within the most primitive hematopoietic stem cell compartment.
      ].

      A new standard to detect all HSC classes

      As noted, data derived from single–time point analysis are not sufficient to detect all ST-, IT-, and LT-HSCs, given that various lineages are reconstituted with different dynamics (Figs. 1 and 2). As illustrated in Figure 2, if recipient mice are analyzed in a narrow time window, some HSC classes may be missed. Ideally, the peripheral blood of recipients should be analyzed as long as animals survive. Ideally, the peripheral blood of recipient mice should be analyzed for as long as animals survive, an approach favored by Harrison [
      • Harrison D.E.
      • Stone M.
      • Astle C.M.
      Effects of transplantation on the primitive immunohematopoietic stem cell.
      ]. However, practical reasons dictate that recipients be analyzed a minimum of three times, for example at 1–2, 4–6, and 8–12 months after transplantation. Alternatively, mice could be analyzed at 1–2 months and then at 4–6 months after primary transplantation, and 4–6 months after secondary transplantation. Of special note is that only granulocyte reconstitution but tri-lineage reconstitution is essential for HSC detection in the new criteria.

      HSC differentiation models

      Bifurcation model (Fig. 4A)

      Weissman's group has identified common lymphoid progenitors (CLPs) and common myeloid progenitors (CMPs) in adult bone marrow [
      • Akashi K.
      • Traver D.
      • Miyamoto T.
      • Weissman I.L.
      A clonogenic common myeloid progenitor that gives rise to all myeloid lineages.
      ,
      • Kondo M.
      • Weissman I.L.
      • Akashi K.
      Identification of clonogenic common lymphoid progenitors in mouse bone marrow.
      ] and fetal liver [
      • Mebius R.E.
      • Miyamoto T.
      • Christensen J.
      • et al.
      The fetal liver counterpart of adult common lymphoid progenitors gives rise to all lymphoid lineages, CD45+CD4+CD3- cells, as well as macrophages.
      ,
      • Traver D.
      • Miyamoto T.
      • Christensen J.
      • Iwasaki-Arai J.
      • Akashi K.
      • Weissman I.L.
      Fetal liver myelopoiesis occurs through distinct, prospectively isolatable progenitor subsets.
      ]. CLPs give rise to B cells, T cells, and natural killer (NK) cells but not granulocytes, macrophages, erythrocytes, or platelets, whereas CMPs give rise to granulocytes, macrophages, erythrocytes, and platelets but not B cells, T cells, or NK cells. Thus, CLPs and CMPs are mutually exclusive populations, suggesting that MPPs are the common progenitors of both. In this model (Fig. 4A), HSCs give rise to MPPs or their equivalent progenitors following loss of self-renewal potential as MPPs maintain all differentiation potentials. This model was proposed over a decade ago [
      • Akashi K.
      • Traver D.
      • Miyamoto T.
      • Weissman I.L.
      A clonogenic common myeloid progenitor that gives rise to all myeloid lineages.
      ], but, at the clonal level, MPPs have not been identified experimentally, and the relationship of MPPs to CLPs or CMPs is yet to be clarified.
      Figure thumbnail gr4
      Figure 4HSC differentiation models. Shown are the (A) bifurcation model
      [
      • Akashi K.
      • Traver D.
      • Miyamoto T.
      • Weissman I.L.
      A clonogenic common myeloid progenitor that gives rise to all myeloid lineages.
      ]
      , (B) myeloid-based model
      [
      • Katsura Y.
      • Kawamoto H.
      Stepwise lineage restriction of progenitors in lympho-myelopoiesis.
      ]
      , and (C) LMPP model
      [
      • Adolfsson J.
      • Månsson R.
      • Buza-Vidas N.
      • et al.
      Identification of Flt3+ lympho-myeloid stem cells lacking erythro-megakaryocytic potential a revised road map for adult blood lineage commitment.
      ]
      . CLP = common lymphoid progenitor; CMP = common myeloid progenitor; GMP = and granulocyte and macrophage progenitor; LMPP = lymphoid-primed multipotent progenitor; MEP = megakaryocyte and erythrocyte progenitor; MPP = multipotent progenitor; MyB = myeloid progenitor with B cell potential; MyBT = myeloid progenitor with B cell and T cell potential; MyE = myeloid progenitor with erythroid potential; MyT = myeloid progenitor with T cell potential.

      Myeloid-based model (Fig. 4B)

      Katsura 's group analyzed fetal liver cells using an in vitro assay they developed known as a multilineage progenitor assay [
      • Kawamoto H.
      • Ohmura K.
      • Katsura Y.
      Direct evidence for the commitment of hematopoietic stem cells to T, B and myeloid lineages in murine fetal liver.
      ]. Fetal thymic organ culture [
      • Jenkinson E.J.
      • Anderson G.
      • Owen J.J.
      Studies on T cell maturation on defined thymic stromal cell populations in vitro.
      ] was modified by adding cytokines (SCF, IL-3, IL-7, and EPO) and providing a higher percentage of oxygen to detect B cell progenitors, granulocytes, macrophages, and erythroblasts, in addition to T cell progenitors [
      • Kawamoto H.
      • Ohmura K.
      • Katsura Y.
      Direct evidence for the commitment of hematopoietic stem cells to T, B and myeloid lineages in murine fetal liver.
      ,
      • Lu M.
      • Kawamoto H.
      • Katsube Y.
      • Ikawa T.
      • Katsura Y.
      The common myelolymphoid progenitor: a key intermediate stage in hemopoiesis generating T and B cells.
      ]. Progenitors such as granulocyte/macrophage/B cell/T cell (MyBT) progenitors, MyB progenitors, and MyT progenitors were detected [
      • Kawamoto H.
      • Ohmura K.
      • Fujimoto S.
      • Katsura Y.
      Emergence of T cell progenitors without B cell or myeloid differentiation potential at the earliest stage of hematopoiesis in the murine fetal liver.
      ,
      • Kawamoto H.
      • Ohmura K.
      • Katsura Y.
      Direct evidence for the commitment of hematopoietic stem cells to T, B and myeloid lineages in murine fetal liver.
      ,
      • Lu M.
      • Kawamoto H.
      • Katsube Y.
      • Ikawa T.
      • Katsura Y.
      The common myelolymphoid progenitor: a key intermediate stage in hemopoiesis generating T and B cells.
      ], whereas BT progenitors were not. Because erythroblast progenitors were not detected among MyBT progenitors, a myeloid-based model (Fig. 4B) was proposed in which HSCs give rise to MyE or MyBT progenitors and MyBT progenitors then give rise to either MyB or MyT progenitors [
      • Lu M.
      • Kawamoto H.
      • Katsube Y.
      • Ikawa T.
      • Katsura Y.
      The common myelolymphoid progenitor: a key intermediate stage in hemopoiesis generating T and B cells.
      ,
      • Katsura Y.
      • Kawamoto H.
      Stepwise lineage restriction of progenitors in lympho-myelopoiesis.
      ]. HSCs were not distinguishable from multipotent progenitors in this system because HSCs cannot be detected by in vitro assays. Megakaryocytes have not been examined in this system. Progenitors in adult bone marrow have not been compared with those in fetal liver by this system.

      LMPP model (Fig. 4C)

      Jacobsen's group identified lymphoid-primed MPPs (LMPPs) that give rise to granulocyte/macrophage and B/T cell lineages but not the megakaryocyte/erythrocyte lineage [
      • Adolfsson J.
      • Månsson R.
      • Buza-Vidas N.
      • et al.
      Identification of Flt3+ lympho-myeloid stem cells lacking erythro-megakaryocytic potential a revised road map for adult blood lineage commitment.
      ,
      • Månsson R.
      • Hultquist A.
      • Luc S.
      • et al.
      Molecular evidence for hierarchical transcriptional lineage priming in fetal and adult stem cells and multipotent progenitors.
      ]. Myeloid potential of single LMPPs was detected with in vitro colony assays, whereas B and T cell potentials were detected with coculture with OP9 and OP9/Delta-like 1 stromal cells plus cytokines, respectively. From these analyses, the authors proposed a model combining elements of the bifurcation and myeloid-based models [
      • Adolfsson J.
      • Månsson R.
      • Buza-Vidas N.
      • et al.
      Identification of Flt3+ lympho-myeloid stem cells lacking erythro-megakaryocytic potential a revised road map for adult blood lineage commitment.
      ]. In this model (Fig. 4C), megakaryocyte/erythrocyte progenitors likely branch from HSCs, resulting in emergence of LMPPs. This first step of HSC differentiation is similar to that proposed in the myeloid-based model. However, lineage tracing studies suggest that LMPPs also differentiate into an ME lineage [
      • Forsberg E.C.
      • Serwold T.
      • Kogan S.
      • Weissman I.L.
      • Passegue E.
      New evidence supporting megakaryocyte-erythrocyte potential of flk2/flt3+ multipotent hematopoietic progenitors.
      ,
      • Boyer S.W.
      • Schroeder A.V.
      • Smith-Berdan S.
      • Forsberg E.C.
      All hematopoietic cells develop from hematopoietic stem cells through Flk2/Flt3-positive progenitor cells.
      ]. More extensive studies are required to clarify ME differentiation pathways. LMPPs reportedly give rise to either granulocyte and macrophage progenitors or CLPs [
      • Adolfsson J.
      • Månsson R.
      • Buza-Vidas N.
      • et al.
      Identification of Flt3+ lympho-myeloid stem cells lacking erythro-megakaryocytic potential a revised road map for adult blood lineage commitment.
      ], a pathway reminiscent of the bifurcation model.

      A new HSC differentiation model

      Investigators in the field assume that MPPs give rise to CMPs, CLPs, MyBT progenitors, or LMPPs [
      • Katsura Y.
      • Kawamoto H.
      Stepwise lineage restriction of progenitors in lympho-myelopoiesis.
      ,
      • Passegue E.
      • Jamieson C.H.
      • Ailles L.E.
      • Weissman I.L.
      Normal and leukemic hematopoiesis: are leukemias a stem cell disorder or a reacquisition of stem cell characteristics?.
      ,
      • Luc S.
      • Buza-Vidas N.
      • Jacobsen S.E.
      Delineating the cellular pathways of hematopoietic lineage commitment.
      ]. However, no clonal study has provided evidence for these pathways, and it is possible that progenitors can be generated without MPPs. The paired daughter cell (PDC) assay is one of the few methods available to address this issue [
      • Takano H.
      • Ema H.
      • Sudo K.
      • Nakauchi H.
      Asymmetric division and lineage commitment at the level of hematopoietic stem cells: inference from differentiation in daughter cell and granddaughter cell pairs.
      ,
      • Suda T.
      • Suda J.
      • Ogawa M.
      Disparate differentiation in mouse hemopoietic colonies derived from paired progenitors.
      ,
      • Ema H.
      • Takano H.
      • Sudo K.
      • Nakauchi H.
      In vitro self-renewal division of hematopoietic stem cells.
      ]. In this approach, after a cultured HSC divides into two daughter cells, each daughter cell is separated by a micromanipulator and transplanted with competitor cells. It was recently reported that the megakaryocyte lineage is one of the first blood cell lineages reconstituted by HSCs [
      • Yamamoto R.
      • Morita Y.
      • Ooehara J.
      • et al.
      Clonal analysis unveils self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells.
      ,
      • Forsberg E.C.
      • Serwold T.
      • Kogan S.
      • Weissman I.L.
      • Passegue E.
      New evidence supporting megakaryocyte-erythrocyte potential of flk2/flt3+ multipotent hematopoietic progenitors.
      ]. Megakaryocyte lineage-specific repopulating cells are often found in highly purified HSC populations [
      • Yamamoto R.
      • Morita Y.
      • Ooehara J.
      • et al.
      Clonal analysis unveils self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells.
      ]. Expression of megakaryocyte markers is sometimes detected in the HSC population [
      • Sanjuan-Pla A.
      • Macaulay I.C.
      • Jensen C.T.
      • et al.
      Platelet-biased stem cells reside at the apex of the haematopoietic stem-cell hierarchy.
      ,
      • Gieger C.
      • Radhakrishnan A.
      • Cvejic A.
      • et al.
      New gene functions in megakaryopoiesis and platelet formation.
      ], suggesting that these cells are developmentally related to HSCs. To address whether they are, PDC assays using single HSCs were performed [
      • Yamamoto R.
      • Morita Y.
      • Ooehara J.
      • et al.
      Clonal analysis unveils self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells.
      ,
      • Takano H.
      • Ema H.
      • Sudo K.
      • Nakauchi H.
      Asymmetric division and lineage commitment at the level of hematopoietic stem cells: inference from differentiation in daughter cell and granddaughter cell pairs.
      ], and emergence of pairs of LT- or ST-HSC and megakaryocyte progenitors, as well as pairs of ST-HSCs and CMPs with repopulating potential (rCMPs)—which differ from previously defined “classical CMPs” [
      • Yamamoto R.
      • Morita Y.
      • Ooehara J.
      • et al.
      Clonal analysis unveils self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells.
      ,
      • Akashi K.
      • Traver D.
      • Miyamoto T.
      • Weissman I.L.
      A clonogenic common myeloid progenitor that gives rise to all myeloid lineages.
      ]—was observed. These data strongly suggest that cells of the myeloid or megakaryocyte lineage are generated directly from HSCs via asymmetric division. Accordingly, the myeloid bypass model has been proposed [
      • Yamamoto R.
      • Morita Y.
      • Ooehara J.
      • et al.
      Clonal analysis unveils self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells.
      ].
      Considering the position of ST-, IT-, and LT-HSCs in the myeloid bypass model, we now propose a new differentiation model (Fig. 5). LT-HSCs give rise to either rCMPs or IT- and ST-HSCs. rCMPs can be replaced by megakaryocyte progenitors with repopulating potential such that megakaryocyte progenitors are generated from HSCs. Recently, the bifurcation model has been revised [
      • Arinobu Y.
      • Mizuno S.
      • Chong Y.
      • et al.
      Reciprocal activation of GATA-1 and PU.1 marks initial specification of hematopoietic stem cells into myeloerythroid and myelolymphoid lineages.
      ]. Now, differences among various models are becoming less apparent.
      Figure thumbnail gr5
      Figure 5New differentiation model. The authors' current understanding of HSC differentiation pathways is illustrated. LT-HSCs give rise to rCMPs and IT- or ST-HSCs. IT-HSCs may represent an intermediate state between LT-HSCs and ST-HSCs. ST-HSCs give rise to myeloid and B cell progenitors (MyB) or myeloid and T cell progenitors (MyT).
      In our new model, ST-HSCs give rise to B or T cell progenitors with myeloid potential (MyB and MyT progenitors). CLP function in lymphopoiesis remains uncertain, because these cells are extremely rare in bone marrow [
      • Inlay M.A.
      • Bhattacharya D.
      • Sahoo D.
      • et al.
      Ly6d marks the earliest stage of B-cell specification and identifies the branchpoint between B-cell and T-cell development.
      ] and have never been detected in single-cell transplantation [
      • Morita Y.
      • Ema H.
      • Nakauchi H.
      Heterogeneity and hierarchy within the most primitive hematopoietic stem cell compartment.
      ,
      • Yamamoto R.
      • Morita Y.
      • Ooehara J.
      • et al.
      Clonal analysis unveils self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells.
      ]. On the other hand, we have detected single MyB or MyT progenitors (Ema, unpublished data, 2004). Thus, lymphoid differentiation pathways based on the Kawamoto and Katsura model [
      • Katsura Y.
      • Kawamoto H.
      Stepwise lineage restriction of progenitors in lympho-myelopoiesis.
      ] are included in this model. Further work is required to define lymphoid differentiation pathways. In particular, a role of MyT progenitors, if any, needs to be clarified.
      Our model predicts that the myeloid compartment is established earlier than the lymphoid compartment. Accordingly, in bone marrow, the myeloid compartment becomes larger than the lymphoid compartment. Similarly, the B lymphoid compartment is larger than the T lymphoid compartment, consistent with the fact that bone marrow is the site of myelopoiesis and B lymphopoiesis but not of T lymphopoiesis. Several candidate thymus-seeding progenitors have been reported [
      • Bhandoola A.
      • von Boehmer H.
      • Petrie H.T.
      • Zuniga-Pflucker J.C.
      Commitment and developmental potential of extrathymic and intrathymic T cell precursors: plenty to choose from.
      ,
      • Benz C.
      • Martins V.C.
      • Radtke F.
      • Bleul C.C.
      The stream of precursors that colonizes the thymus proceeds selectively through the early T lineage precursor stage of T cell development.
      ]. It is now critical to determine whether lineage commitment (lineage restriction) occurs before progenitors migrate into the thymus (i.e., whether ST-HSCs or MyT progenitors as shown in Fig. 5 home to the thymus). To address this issue, circulating HSCs, progenitors, or both must be identified.

      Future challenges

      G0 length in HSCs

      HSCs reportedly enter the cell cycle once every month [
      • Bradford G.B.
      • Williams B.
      • Rossi R.
      • Bertoncello I.
      Quiescence, cycling, and turnover in the primitive hematopoietic stem cell compartment.
      ,
      • Cheshier S.H.
      • Morrison S.J.
      • Liao X.
      • Weissman I.L.
      In vivo proliferation and cell cycle kinetics of long-term self-renewing hematopoietic stem cells.
      ,
      • Sudo K.
      • Ema H.
      • Morita Y.
      • Nakauchi H.
      Age-associated characteristics of murine hematopoietic stem cells.
      ]. Dormant HSCs reportedly enter the cell cycle at 5-month intervals [
      • Wilson A.
      • Laurenti E.
      • Oser G.
      • et al.
      Hematopoietic stem cells reversibly switch from dormancy to self-renewal during homeostasis and repair.
      ]. Thus, an intriguing question is whether the length of G0 differs among LT-, IT-, and ST-HSCs. Dormant HSCs may be present among LT-HSCs and their quiescent state cell-intrinsically regulated. Nevertheless, molecular mechanisms distinguishing LT-HSCs from ST-HSCs remain important to characterize, because they might reveal what controls HSC lifespan. Whether IT-HSCs serve as a transition from LT-HSCs to ST-HSCs should also be determined.

      Ex vivo HSC expansion

      It is difficult to induce in vitro self-renewal in HSCs, possibly because of their heterogeneity [
      • Miller P.H.
      • Knapp D.J.
      • Eaves C.J.
      Heterogeneity in hematopoietic stem cell populations: implications for transplantation.
      ]. If heterogeneity arises from developmental processes, a specific class of HSCs responsible for their in vivo expansion may be useful for ex vivo expansion and manipulation of HSCs as needed. Nevertheless, LT-HSCs can be used to produce a large number of myeloid progenitors in vitro, an approach that can be applied to prevent severe bacterial infection in cancer patients after intensive chemotherapy or irradiation [
      • Nakahata T.
      Ex vivo expansion of human hematopoietic stem cells.
      ].

      In vivo tracking of HSCs

      HSCs have not been marked successfully using an HSC-specific reporter. Even if such tracing were possible, it might remain difficult to track single HSCs in vivo under physiologic circumstances, because their engraftment in nonirradiated mice is detectable only after transplantation of a large number of HSCs [
      • Stewart F.M.
      • Crittenden R.B.
      • Lowry P.A.
      • Pearson-White S.
      • Quesenberry P.J.
      Long-term engraftment of normal and post-5-fluorouracil murine marrow into normal nonmyeloablated mice.
      ,
      • Czechowicz A.
      • Kraft D.
      • Weissman I.L.
      • Bhattacharya D.
      Efficient transplantation via antibody-based clearance of hematopoietic stem cell niches.
      ]. If a small number of donor-derived mature cells in peripheral blood is detectable, one might be able to analyze HSC dynamics in nonirradiated settings. Recently, barcode analysis in conjunction with next-generation sequencing has been applied to in vivo clonal analysis of mouse HSCs after transplantation [
      • Gerrits A.
      • Dykstra B.
      • Kalmykowa O.J.
      • et al.
      Cellular barcoding tool for clonal analysis in the hematopoietic system.
      ,
      • Lu R.
      • Neff N.F.
      • Quake S.R.
      • Weissman I.L.
      Tracking single hematopoietic stem cells in vivo using high-throughput sequencing in conjunction with viral genetic barcoding.
      ]. The barcode, composed of a short random sequence (∼30 nt), is integrated into the HSC genome using retroviral or lentiviral vectors. There remain technical obstacles to this approach, such as unavoidable in vitro manipulation of HSCs for transduction, difficulties in analyzing massive amounts of data from repeated sequencing of multiple blood lineages, and a lack of information about red blood cells or platelets, which lack genomic DNA. However, it may be possible to track rare clones among a large number of normal HSCs using such technology. Moreover, the method should enable the study of human repopulating cells in immunodeficient mice at the clonal level [
      • Cheung A.M.
      • Nguyen L.V.
      • Carles A.
      • et al.
      Analysis of the clonal growth and differentiation dynamics of primitive barcoded human cord blood cells in NSG mice.
      ]. It is important to understand the degree of heterogeneity in human HSCs and the contribution of this heterogeneity to development of the hierarchical organization of human hematopoiesis.

      Acknowledgments

      We thank Connie Eaves for suggesting that we compare differently defined HSC classes, and Aled O'Neill and Keiyo Takubo for critical reading of the manuscript. This work was supported in part by Grants-in-Aid for Scientific Research (A) and (C), Grants-in-Aid for Scientific Research on Innovative Areas in Japan, and the European Union's Seventh Framework Programme ( FP7/2007-2013 ) under grant agreement number 306240 (SyStemAge).

      Supplementary data

      Figure thumbnail fx1
      Supplementary Figure E1Calculation of lymphoid to myeloid (L/M) and M/L ratios. Flow cytometry data are shown. Ly5.1-positive cells were used as test donor cells. Myeloid cells were detected by anti–Mac-1 and anti–Gr-1 antibodies. The L/M ratio was calculated as L1/M1. The M/L ratio was calculated as (% Myeloid chimerism) / (% Lymphoid chimerism). % Myeloid chimerism is defined as 100 M1 / (M1 + M2). % Lymphoid chimerism is defined as 100 L1 / (L1 + L2). These L1, L2, L3 are not actually measured. Instead, 100 B1 / (B1 + B2) and 100 T1 / (T1 + T2) are calculated for % B cells and % T cell, respectively (B1, B2, B3, T1, T1, and T1 are not shown in this figure), and % L is replaced by (% B cells + % T cells)
      [
      • Dykstra B.
      • Kent D.
      • Bowie M.
      • et al.
      Long-term propagation of distinct hematopoietic differentiation programs in vivo.
      ]
      .
      Figure thumbnail fx2
      Supplementary Figure E2Relationships of HSCs classified by three systems. All relationships of My-bi, Bala, and Ly-bi HSCs, α, β, and γ/δ cells, and LT-, IT-, and ST-HSCs are shown. (A) My-bi, Bala, Ly-bi HSCs (denominators) were reclassified as α, β, and γ/δ cells or LT-, IT-, and ST-HSCs (numerators). (B) α, β, and γ/δ cells (denominators) were reclassified as My-bi, Bala, Ly-bi HSCs or LT-, IT-, and ST-HSCs (numerators). (C) LT-, IT-, and ST-HSCs (denominators) were reclassified as My-bi, Bala, Ly-bi HSCs or α, β, and γ/δ cells (numerators).

      References

        • Till J.E.
        • McCulloch E.A.
        • Siminovitch L.
        A stochastic model of stem cell proliferation, based on the growth of spleen colony-forming cells.
        Proc Natl Acad Sci U S A. 1964; 51: 29-36
        • Guenechea G.
        • Gan O.I.
        • Dorrell C.
        • Dick J.E.
        Distinct classes of human stem cells that differ in proliferative and self-renewal potential.
        Nat Immunol. 2001; 2: 75-82
        • Ema H.
        • Sudo K.
        • Seita J.
        • et al.
        Quantification of self-renewal capacity in single hematopoietic stem cells from normal and Lnk-deficient mice.
        Dev Cell. 2005; 8: 907-914
        • Muller-Sieburg C.E.
        • Sieburg H.B.
        • Bernitz J.M.
        • Cattarossi G.
        Stem cell heterogeneity: implications for aging and regenerative medicine.
        Blood. 2012; 119: 3900-3907
        • Copley M.R.
        • Beer P.A.
        • Eaves C.J.
        Hematopoietic stem cell heterogeneity takes center stage.
        Cell Stem Cell. 2012; 10: 690-697
        • Osawa M.
        • Hanada K.
        • Hamada H.
        • Nakauchi H.
        Long-term lymphohematopoietic reconstitution by a single CD34-low/negative hematopoietic stem cell.
        Science. 1996; 273: 242-245
        • Morrison S.J.
        • Wandycz A.M.
        • Hemmati H.D.
        • Wright D.E.
        • Weissman I.L.
        Identification of a lineage of multipotent hematopoietic progenitors.
        Development. 1997; 124: 1929-1939
        • Yang L.
        • Bryder D.
        • Adolfsson J.
        • et al.
        Identification of Lin(-)Sca1(+)kit(+)CD34(+)Flt3- short-term hematopoietic stem cells capable of rapidly reconstituting and rescuing myeloablated transplant recipients.
        Blood. 2005; 105: 2717-2723
        • Dick J.E.
        • Magli M.C.
        • Huszar D.
        • Phillips R.A.
        • Bernstein A.
        Introduction of a selectable gene into primitive stem cells capable of long-term reconstitution of the hemopoietic system of W/Wv mice.
        Cell. 1985; 42: 71-79
        • Keller G.
        • Paige C.
        • Gilboa E.
        • Wagner E.F.
        Expression of a foreign gene in myeloid and lymphoid cells derived from multipotent haematopoietic precursors.
        Nature. 1985; 318: 149-154
        • Lemischka I.R.
        • Raulet D.H.
        • Mulligan R.C.
        Developmental potential and dynamic behavior of hematopoietic stem cells.
        Cell. 1986; 45: 917-927
        • Jordan C.T.
        • Lemischka I.R.
        Clonal and systemic analysis of long-term hematopoiesis in the mouse.
        Genes Dev. 1990; 4: 220-232
        • Muller-Sieburg C.E.
        • Cho R.H.
        • Thoman M.
        • Adkins B.
        • Sieburg H.B.
        Deterministic regulation of hematopoietic stem cell self-renewal and differentiation.
        Blood. 2002; 100: 1302-1309
        • 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; 103: 4111-4118
        • Cho R.H.
        • Sieburg H.B.
        • Muller-Sieburg C.E.
        A new mechanism for the aging of hematopoietic stem cells: aging changes the clonal composition of the stem cell compartment but not individual stem cells.
        Blood. 2008; 111: 5553-5561
        • Dykstra B.
        • Kent D.
        • Bowie M.
        • et al.
        Long-term propagation of distinct hematopoietic differentiation programs in vivo.
        Cell Stem Cell. 2007; 1: 218-229
        • Benz C.
        • Copley M.R.
        • Kent D.G.
        • et al.
        Hematopoietic stem cell subtypes expand differentially during development and display distinct lymphopoietic programs.
        Cell Stem Cell. 2012; 10: 273-283
        • Challen G.A.
        • Boles N.C.
        • Chambers S.M.
        • Goodell M.A.
        Distinct hematopoietic stem cell subtypes are differentially regulated by TGF-beta1.
        Cell Stem Cell. 2010; 6: 265-278
        • Morita Y.
        • Ema H.
        • Nakauchi H.
        Heterogeneity and hierarchy within the most primitive hematopoietic stem cell compartment.
        J Exp Med. 2010; 207: 1173-1182
        • Benveniste P.
        • Frelin C.
        • Janmohamed S.
        • et al.
        Intermediate-term hematopoietic stem cells with extended but time-limited reconstitution potential.
        Cell Stem Cell. 2010; 6: 48-58
        • Yamamoto R.
        • Morita Y.
        • Ooehara J.
        • et al.
        Clonal analysis unveils self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells.
        Cell. 2013; 154: 1112-1126
        • Akashi K.
        • Traver D.
        • Miyamoto T.
        • Weissman I.L.
        A clonogenic common myeloid progenitor that gives rise to all myeloid lineages.
        Nature. 2000; 404: 193-197
        • Kawamoto H.
        • Ohmura K.
        • Fujimoto S.
        • Katsura Y.
        Emergence of T cell progenitors without B cell or myeloid differentiation potential at the earliest stage of hematopoiesis in the murine fetal liver.
        J Immunol. 1999; 162: 2725-2731
        • Adolfsson J.
        • Månsson R.
        • Buza-Vidas N.
        • et al.
        Identification of Flt3+ lympho-myeloid stem cells lacking erythro-megakaryocytic potential a revised road map for adult blood lineage commitment.
        Cell. 2005; 121: 295-306
        • Sanjuan-Pla A.
        • Macaulay I.C.
        • Jensen C.T.
        • et al.
        Platelet-biased stem cells reside at the apex of the haematopoietic stem-cell hierarchy.
        Nature. 2013; 502: 232-236
        • Kent D.G.
        • Copley M.R.
        • Benz C.
        • et al.
        Prospective isolation and molecular characterization of hematopoietic stem cells with durable self-renewal potential.
        Blood. 2009; 113: 6342-6350
        • Till J.E.
        • McCullouch E.A.
        A direct measurement of the radiation sensitivity of normal mouse bone marrow cells.
        Radiat Res. 1961; 14: 213-222
        • Micklem H.S.
        • Ford C.E.
        • Evans E.P.
        • Ogden D.A.
        • Papworth D.S.
        Competitive in vivo proliferation of foetal and adult haematopoietic cells in lethally irradiated mice.
        J Cell Physiol. 1972; 79: 293-298
        • Harrison D.E.
        • Jordan C.T.
        • Zhong R.K.
        • Astle C.M.
        Primitive hemopoietic stem cells: direct assay of most productive populations by competitive repopulation with simple binomial, correlation and covariance calculations.
        Exp Hematol. 1993; 21: 206-219
        • Szilvassy S.J.
        • Humphries R.K.
        • Lansdorp P.M.
        • Eaves A.C.
        • Eaves C.J.
        Quantitative assay for totipotent reconstituting hematopoietic stem cells by a competitive repopulation strategy.
        Proc Natl Acad Sci U S A. 1990; 87: 8736-8740
        • Ema H.
        • Nakauchi H.
        Expansion of hematopoietic stem cells in the developing liver of a mouse embryo.
        Blood. 2000; 95: 2284-2288
        • Takano H.
        • Ema H.
        • Sudo K.
        • Nakauchi H.
        Asymmetric division and lineage commitment at the level of hematopoietic stem cells: inference from differentiation in daughter cell and granddaughter cell pairs.
        J Exp Med. 2004; 199: 295-302
        • Harrison D.E.
        • Stone M.
        • Astle C.M.
        Effects of transplantation on the primitive immunohematopoietic stem cell.
        J Exp Med. 1990; 172: 431-437
        • Kondo M.
        • Weissman I.L.
        • Akashi K.
        Identification of clonogenic common lymphoid progenitors in mouse bone marrow.
        Cell. 1997; : 91661-91672
        • Mebius R.E.
        • Miyamoto T.
        • Christensen J.
        • et al.
        The fetal liver counterpart of adult common lymphoid progenitors gives rise to all lymphoid lineages, CD45+CD4+CD3- cells, as well as macrophages.
        J Immunol. 2001; 166: 6593-6601
        • Traver D.
        • Miyamoto T.
        • Christensen J.
        • Iwasaki-Arai J.
        • Akashi K.
        • Weissman I.L.
        Fetal liver myelopoiesis occurs through distinct, prospectively isolatable progenitor subsets.
        Blood. 2001; 98: 627-635
        • Kawamoto H.
        • Ohmura K.
        • Katsura Y.
        Direct evidence for the commitment of hematopoietic stem cells to T, B and myeloid lineages in murine fetal liver.
        Int Immunol. 1997; 9: 1011-1019
        • Jenkinson E.J.
        • Anderson G.
        • Owen J.J.
        Studies on T cell maturation on defined thymic stromal cell populations in vitro.
        J Exp Med. 1992; 176: 845-853
        • Lu M.
        • Kawamoto H.
        • Katsube Y.
        • Ikawa T.
        • Katsura Y.
        The common myelolymphoid progenitor: a key intermediate stage in hemopoiesis generating T and B cells.
        J Immunol. 2002; 169: 3519-3525
        • Katsura Y.
        • Kawamoto H.
        Stepwise lineage restriction of progenitors in lympho-myelopoiesis.
        Int Rev Immunol. 2001; 20: 1-20
        • Månsson R.
        • Hultquist A.
        • Luc S.
        • et al.
        Molecular evidence for hierarchical transcriptional lineage priming in fetal and adult stem cells and multipotent progenitors.
        Immunity. 2007; 26: 407-419
        • Forsberg E.C.
        • Serwold T.
        • Kogan S.
        • Weissman I.L.
        • Passegue E.
        New evidence supporting megakaryocyte-erythrocyte potential of flk2/flt3+ multipotent hematopoietic progenitors.
        Cell. 2006; 126: 415-426
        • Boyer S.W.
        • Schroeder A.V.
        • Smith-Berdan S.
        • Forsberg E.C.
        All hematopoietic cells develop from hematopoietic stem cells through Flk2/Flt3-positive progenitor cells.
        Cell Stem Cell. 2011; 9: 64-73
        • Passegue E.
        • Jamieson C.H.
        • Ailles L.E.
        • Weissman I.L.
        Normal and leukemic hematopoiesis: are leukemias a stem cell disorder or a reacquisition of stem cell characteristics?.
        Proc Natl Acad Sci U S A. 2003; 100: 11842-11849
        • Luc S.
        • Buza-Vidas N.
        • Jacobsen S.E.
        Delineating the cellular pathways of hematopoietic lineage commitment.
        Semin Immunol. 2008; 20: 213-220
        • Suda T.
        • Suda J.
        • Ogawa M.
        Disparate differentiation in mouse hemopoietic colonies derived from paired progenitors.
        Proc Natl Acad Sci U S A. 1984; 81: 2520-2524
        • Ema H.
        • Takano H.
        • Sudo K.
        • Nakauchi H.
        In vitro self-renewal division of hematopoietic stem cells.
        J Exp Med. 2000; 192: 1281-1288
        • Gieger C.
        • Radhakrishnan A.
        • Cvejic A.
        • et al.
        New gene functions in megakaryopoiesis and platelet formation.
        Nature. 2011; 480: 201-208
        • Arinobu Y.
        • Mizuno S.
        • Chong Y.
        • et al.
        Reciprocal activation of GATA-1 and PU.1 marks initial specification of hematopoietic stem cells into myeloerythroid and myelolymphoid lineages.
        Cell Stem Cell. 2007; 1: 416-427
        • Inlay M.A.
        • Bhattacharya D.
        • Sahoo D.
        • et al.
        Ly6d marks the earliest stage of B-cell specification and identifies the branchpoint between B-cell and T-cell development.
        Genes Dev. 2009; 23: 2376-2381
        • Bhandoola A.
        • von Boehmer H.
        • Petrie H.T.
        • Zuniga-Pflucker J.C.
        Commitment and developmental potential of extrathymic and intrathymic T cell precursors: plenty to choose from.
        Immunity. 2007; 26: 678-689
        • Benz C.
        • Martins V.C.
        • Radtke F.
        • Bleul C.C.
        The stream of precursors that colonizes the thymus proceeds selectively through the early T lineage precursor stage of T cell development.
        J Exp Med. 2008; 205: 1187-1199
        • Bradford G.B.
        • Williams B.
        • Rossi R.
        • Bertoncello I.
        Quiescence, cycling, and turnover in the primitive hematopoietic stem cell compartment.
        Exp Hematol. 1997; 25: 445-453
        • Cheshier S.H.
        • Morrison S.J.
        • Liao X.
        • Weissman I.L.
        In vivo proliferation and cell cycle kinetics of long-term self-renewing hematopoietic stem cells.
        Proc Natl Acad Sci U S A. 1999; 96: 3120-3125
        • Sudo K.
        • Ema H.
        • Morita Y.
        • Nakauchi H.
        Age-associated characteristics of murine hematopoietic stem cells.
        J Exp Med. 2000; 192: 1273-1280
        • Wilson A.
        • Laurenti E.
        • Oser G.
        • et al.
        Hematopoietic stem cells reversibly switch from dormancy to self-renewal during homeostasis and repair.
        Cell. 2008; 135: 1118-1129
        • Miller P.H.
        • Knapp D.J.
        • Eaves C.J.
        Heterogeneity in hematopoietic stem cell populations: implications for transplantation.
        Curr Opin Hematol. 2013; 20: 257-264
        • Nakahata T.
        Ex vivo expansion of human hematopoietic stem cells.
        Int J Hematol. 2001; 73: 6-13
        • Stewart F.M.
        • Crittenden R.B.
        • Lowry P.A.
        • Pearson-White S.
        • Quesenberry P.J.
        Long-term engraftment of normal and post-5-fluorouracil murine marrow into normal nonmyeloablated mice.
        Blood. 1993; 81: 2566-2571
        • Czechowicz A.
        • Kraft D.
        • Weissman I.L.
        • Bhattacharya D.
        Efficient transplantation via antibody-based clearance of hematopoietic stem cell niches.
        Science. 2007; 318: 1296-1299
        • Gerrits A.
        • Dykstra B.
        • Kalmykowa O.J.
        • et al.
        Cellular barcoding tool for clonal analysis in the hematopoietic system.
        Blood. 2010; 115: 2610-2618
        • Lu R.
        • Neff N.F.
        • Quake S.R.
        • Weissman I.L.
        Tracking single hematopoietic stem cells in vivo using high-throughput sequencing in conjunction with viral genetic barcoding.
        Nat Biotechnol. 2011; 29: 928-933
        • Cheung A.M.
        • Nguyen L.V.
        • Carles A.
        • et al.
        Analysis of the clonal growth and differentiation dynamics of primitive barcoded human cord blood cells in NSG mice.
        Blood. 2013; 122: 3129-3137