- Email: [email protected]
From the Yolk Sac to the Spleen
Immunity, Vol. 18, 587–589, May, 2003, Copyright 2003 by Cell Press
From the Yolk Sac to the Spleen: New Roles for Notch in Regulating Hematopoiesis Ivan Maillard,1,2 Yiping He,2,3 and Warren S. Pear2,3 Division of Hematology-Oncology 2 Abramson Family Cancer Research Institute 3 Department of Pathology and Laboratory Medicine Institute for Medicine and Engineering University of Pennsylvania 611 BRB II/III, 421 Curie Boulevard Philadelphia, Pennsylvania 19104
1
Although Notch receptors are widely expressed during hematopoiesis, their roles outside of the T cell lineage are not well characterized. Two reports in this issue of Immunity show that Notch1 and Notch2, respectively, are required to generate the earliest embryonic hematopoietic stem cells and splenic marginal zone B cells. Thus, different Notch receptors have specific and nonoverlapping functions that influence multiple hematopoietic lineages at various stages of development. Notch signaling is an evolutionarily conserved pathway controlling many developmental cell fate decisions. In hematopoiesis, Notch is required for T cell commitment and has been postulated to influence multiple other events, including hematopoietic stem cell (HSC) homeostasis and marginal zone B cell (MZB) development. Two reports in this issue of Immunity, Kumano et al. and Saito et al., provide new insights into the function of Notch signaling in HSCs and MZB cells. In early hematopoiesis, Notch1 appears critical for the generation of HSCs, whereas Notch2 is required for MZB cell development in the spleen. Mice with a targeted disruption of Notch1 develop multiple defects, including abnormal somitogenesis and impaired vascular development, a defect that may directly contribute to embryonic lethality before E11.5 (Swiatek et al., 1994; Conlon et al., 1995). Conditional inactivation of Notch1 avoids such lethality, providing the opportunity to characterize the role of Notch1 in specific organ systems (for example, see Radtke et al., 1999). This approach has revealed the critical role of Notch1 in the T/B lineage decision during lymphoid development, while most other aspects of hematopoiesis were unaffected, including the ability to competitively repopulate mixed bone marrow chimeras (Radtke et al., 1999). Similar findings were observed in mice after deletion of a floxed allele of CSL/RBP-Jk (CSL), a transcription factor that mediates most of the well-characterized effects of Notch (Han et al., 2002). In this issue of Immunity, Kumano et al. closely evaluate early hematopoiesis in Notch1 and Notch2 knockout embryos. Blood cell formation starts with primitive hematopoiesis in the extraembryonic yolk sac (YS), followed by definitive hematopoiesis in the embryo itself (reviewed by Orkin and Zon, 2002; Kondo et al., 2003). *Correspondence: [email protected]
Minireview
The first multipotent hematopoietic progenitors are found in the YS around E8.5 and thereafter in the paraaortic splanchnopleura (P-Sp) or aorta/gonad/mesonephros (AGM) region, where they differentiate from ventral mesoderm before colonizing the liver, and ultimately the bone marrow. Although the existence of true hemangioblasts remains debatable, early hematopoietic and endothelial progenitors are closely related in both the YS and the P-Sp region, where cells with shared hematopoietic and vascular markers have been identified. In their study, Kumano et al. use P-Sp explant cultures as an assay for early hematopoietic differentiation, together with transfer of progenitors to suitably conditioned recipients to assess stem cell activity. Characterization of Notch receptor expression in P-Sp shows that Notch1 and Notch4 but not Notch2 are coexpressed with the endothelial marker Flk-1, while hematopoietic cells emerging from the culture express Notch1 and Notch2 but not Notch4. Major differences between Notch1 knockout and wild-type P-Sp cultures were observed, while Notch2 knockout embryos were not different from controls. In the absence of Notch1, P-Sp explants had markedly impaired vascular network formation, consistent with a role of Notch1 in angiogenesis, while vascular beds reflective of vasculogenesis were preserved. Strikingly, sharply reduced numbers of hematopoietic colony forming cells (CFCs) were recovered from Notch1 knockout P-Sp cultures, indicating impairment of definitive hematopoiesis. Interestingly, close to normal numbers of CFCs were obtained from Notch1 knockout YS, suggesting that primitive hematopoiesis is at least partially preserved. However, cells recovered from both YS and P-Sp of Notch1 knockout embryos showed no detectable reconstitution of conditioned newborn recipients, suggesting that Notch1 is crucial for generating true stem cells. The defect was cell autonomous since mixed knockout and wild-type cultures did not rescue the phenotype. The data correlated well with experiments using ␥-secretase inhibitors (GSI) to inhibit Notch in P-Sp cultures, with rescue by a constitutive form of Notch. Interestingly, no impairment in the generation of hematopoietic cells was observed when the P-Sp were obtained from E10.5 embryos, a time point when hematopoietic progenitors with stem cell potential are already present. Altogether, these results suggest that Notch1 is critical for the genesis of HSCs, while not being required once their pool has been established. This situation bears striking similarities to the role of other genes in early hematopoietic and vascular development. For example, mice with a targeted inactivation of SCL/tal-1 have no primitive and definitive hematopoiesis, as well as angiogenic defects (reviewed in Orkin and Zon, 2002). However, conditional inactivation of SCL/tal-1 does not result in detectable impairment of HSC function at later time points (Mikkola et al., 2003). In the case of Notch, these results contrast with previous data suggesting a role of Notch in HSC self-renewal (Stier et al., 2002; Kunisato et al., 2003; reviewed in Ohishi et al., 2003). However, these reports involved hyperactivation of the
Immunity 588
Figure 1. Multiple Roles for Notch in Hematopoiesis (A) Notch signaling plays distinct roles at several stages of hematopoietic development. During early hematopoiesis in the embryo, Notch1 is critical for the generation of cells with a hematopoietic stem cell potential from presumed hemangioblasts. It also plays a critical role later during vascular development, beyond vasculogenesis, although the mechanisms are unknown. During lymphopoiesis, Notch1 signaling is required for T lineage specification and its absence allows the development of the B lineage. While it is widely accepted that there is a CLP population that gives rise to the lymphoid lineage, a recent report identified a CLP-independent progenitor population (ETP) with a high propensity for T cell development (Allman et al., 2003). The role of Notch signaling is less clear in subsequent stages of T cell development and function. During splenic B cell development, Notch2 signaling is required for the generation of MZB cells. HSC, hematopoietic stem cell; CMP, common myeloid progenitor; CLP, common lymphoid progenitor; ETP, early T cell progenitor; T, T cells; B, B cells; NK, nature killer cells; Tr. B, transitional B cells; Fo, follicular B cells; and MZ, marginal zone B cells. (B) Structure of the four mammalian Notch receptors (reviewed in Allman et al., 2002). The Notch1 TAD is stronger than the Notch2 TAD, when analyzed by reporter assays, while this domain is absent from Notch3 and Notch4. All four Notch receptors show similarity in their extracellular domains. ECN, extracellular Notch; ICN, intracellular Notch; EGFR, EGF repeats; LNR, LIN Notch repeats; ANK, ankyrin repeats; NCR, Notch cytokine response region; TAD, transactivation domain; P, PEST domain; and M, membrane.
Notch pathway either through ligand-mediated stimulation or through the expression of constitutive forms of Notch. Whether physiological levels of Notch signaling are important in stem cell maintenance and activity remains to be rigorously tested, although the wealth of data from knockout mice so far do not suggest a major defect (Radtke et al., 1999; Han et al. 2002). In contrast, HSC self-renewal may be influenced by other morphogens such as Sonic Hedgehog and Wnt, as shown re-
cently by experiments including loss of function approaches (Bhardwaj et al., 2001; Reya et al., 2003). The precise molecular relationship between Notch and these pathways in HSC biology has not been defined, although there is suggestive evidence that they may lie in a hierarchy (Reya et al., 2003). In terms of molecular mechanisms, it will be important to characterize the precise interrelationships between Notch and other molecules known to be involved in the generation of early hematopoietic progenitors, such as the transcription factors SCL/Tal-1, GATA-2, LMO2, AML1, HoxB4, and the VEGF receptor Flk-1 (Orkin and Zon 2002; Kondo et al., 2003). Kumano et al. show that the expression of many of these factors is downregulated in the absence of Notch1, suggesting that Notch may function as a master switch in the early hematopoietic differentiation program. This may have special relevance since Notch is a transmembrane receptor able to provide critical environmental signals to the early stem cells. Where these signals come from, and what are the relevant Notch ligands and ligand-expressing cells during early hematopoietic differentiation will be another important question to address. The mechanisms by which Notch affects vascular development also need to be better understood. In contrast to the essential role of Notch1 in early HSC and T cell development, Saito et al. show that Notch2 is critical for MZB development. A dramatic decrease in MZB has been reported in conditional CSL knockout mice (Tanigaki et al., 2002), strongly suggesting involvement of the Notch pathway. Because CSL is a transcriptional repressor in the absence of Notch, it was unclear whether the defect resulted from loss of repression or absence of Notch signals. Indirect evidence supporting the latter was that mice lacking MINT, a putative Notch negative regulator, showed an increase in MZB cells (Kuroda et al., 2003). Significantly, this manuscript also showed that Notch2 was highly expressed in the splenic MZ. Saito et al., using conditional inactivation of Notch2, definitively show that Notch2 is required for MZB development through cell autonomous mechanisms. The level of Notch2 expression appears important, as heterozygous mice have an ⵑ50% decrease in MZB cells, in contrast to the Notch1-mediated T/B lineage decision where haploinsufficiency has not been observed. Since Notch2 is highly expressed in B lymphocytes, whereas Notch1 is expressed at very low levels, the differential effect of Notch1 and Notch2 may be related to their expression pattern, as perhaps in T cell development where Notch1 expression is high and Notch2 expression is low. MZB cells are thought to be important in T cell-independent immune responses and share some characteristics with B1 B cells (Martin and Kearney, 2002). They are located in the spleen MZ and interspersed amongst macrophages and follicular dendritic cells. Although integrin signaling is required to retain MZB cells in the MZ (Lu and Cyster, 2002), their development is poorly understood. One hypothesis is that upon leaving the BM, B cells give rise to transitional T1 B cells that then form T2 B cells and may be the precursors to both follicular (Fo) and MZB cells (Loder et al., 1999). In conditional CSL KO mice, both a decrease in MZ cells and an increase in Fo B cells were observed, while other B
Minireview 589
cell subsets were unaffected. This suggested that MZ and Fo B cells arise from a common precursor. In contrast, Saito et al. find that T2 B cells are also decreased. This finding is largely derived from immunophenotypic analysis of CD21 expression in combination with other markers in control and knockout mice. However, the authors also show that CD21 expression decreases in the knockout mice, which complicates the assessment of the decline in the T2 population. If there is a true loss in the T2 population, it is surprising that the MZ but not the Fo B cell pool is affected. As recent data suggest that MZB cells can also arise from mature lymph node recirculating B cells (Vinuesa et al., 2003), the exact precursor/product relationship of these B cell subpopulations awaits further clarification. In any case, Saito et al. provide a possible hint toward the identification of a MZ precursor by showing that a subpopulation of CD9⫹ CD1dhigh CD21high CD23⫹ non-MZB cells also declines in the absence of Notch2. Many pathways influence MZB cell development (reviewed in Martin and Kearney, 2002). These include pathways that affect BCR signaling (such as CD19, CD22, and Aiolos), cell survival (e.g., Baff), cell migration/retention (e.g., integrin and Pyk2-mediated chemokine receptor signaling), and NFB signaling. The relation of Notch to these pathways is unknown. The mechanisms of Notch2 activation in MZB cells and the downstream signals are poorly understood. Two members of the Delta family of Notch ligands, Dll1 and 4, are expressed in the follicular dendritic cells (FDC) found in the MZ (Tanigaki et al., 2002). In this scenario, the MZ precursor would arrive in the MZ where Delta signals derived from FDCs would trigger maturation into MZB cells. Another potential ligand is Jagged1, which is highly expressed in the splenic MZ, although the identity of the Jagged1-expressing cells has not been shown (Bash et al., 1999). Regarding downstream mediators of Notch2 signaling in MZB development, Saito et al. show that Deltex1 is highly expressed in MZB cells. Deltex is a Notch modifier whose precise function is not known. In some contexts, it augments Notch signaling, whereas in others it is inhibitory. For example, transgenic expression of Deltex1 in HSCs blocks T cell development, suggesting that it antagonizes Notch1, whereas Drosophila Deltex acts as a positive Notch modulator (Izon et al., 2002, and references therein). Therefore, it is difficult to predict the role of Deltex1 in MZB development, and additional studies are required to determine if Deltex1 and/or other Notch transcriptional targets are important. In summary, these two manuscripts show that Notch1 and Notch2 play nonoverlapping roles in HSC and MZB development. It remains to be determined whether Notch signaling is controlled at the level of ligand interaction or regulatory molecules, such as Numb or Fringe. In addition, very little is known about the transcriptional regulation of the Notch receptor genes themselves, whose expression is likely to provide a significant level of control. Furthermore, since all four mammalian Notch receptors bind CSL, the mechanisms by which different receptors lead to different outcomes is not known. Understanding the details of Notch signaling in specific contexts, as revealed by these two manuscripts, will provide the basis for investigating these unanswered questions.
Selected Reading Allman, D., Aster, J.C., and Pear, W.S. (2002). Immunol. Rev. 187, 75–86. Allman, D., Sambandam, A., Kim, S., Miller, J.P., Pagan, A., Well, D., Meraz, A., and Bhandoola, A. (2003). Nat. Immunol. 4, 168–174. Bash, J., Zong, W.X., Banga, S., Rivera, A., Ballard, D.W., Ron, Y., and Galinas, C. (1999). EMBO J. 18, 2803–2811. Bhardwaj, G., Murdoch, B., Wu, D., Baker, D.P., Williams, K.P., Chadwick, K., Ling, L.E., Karanu, F.N., and Bhatia, M. (2001). Nat. Immunol. 2, 172–180. Conlon, R.A., Reaume, A.G., and Rossant, J. (1995). Development 121, 1533–1545. Han, H., Tanigaki, K., Yamamoto, N., Kuroda, K., Yoshimoto, M., Nakahata, T., Ikuta, K., and Honjo, T. (2002). Int. Immunol. 14, 637–645. Izon, D.J., Aster, J.C., He, Y., Weng, A., Karnell, F.G., Patriub, V., Xu, L., Bakkour, S., Rodriguez, C., Allman, D., et al. (2002). Immunity 16, 231–243. Kondo, M., Wagers, A.J., Manz, M.G., Prohaska, S.S., Scherer, D.C., Beilhack, G.F., Shizuru, J.A., and Weissman, I.L. (2003). Annu. Rev. Immunol. 21, 759–806. Kumano, K., Chiba, S., Kunisato, A., Sata, M., Saito, T., NakagamiYamaguchi, E., Yamaguchi, T., Masuda, S., Shimizu, K., Takahashi, T., et al. (2003). Immunity 18, this issue, 699–711. Kunisato, A., Chiba, S., Nakagami-Yamaguchi, E., Kumano, K., Saito, T., Masuda, S., Yamaguchi, T., Osawa, M., Kageyama, R., Nakauchi, H., et al. (2003). Blood 101, 1777–1783. Kuroda, K., Han, H., Tani, S., Tanigaki, K., Tun, T., Furukawa, T., Taniguchi, Y., Kurooka, H., Hamada, Y., Toyokuni, S., et al. (2003). Immunity 18, 301–312. Loder, F., Mutschler, B., Ray, R.J., Paige, C.J., Sideras, P., Torres, R., Lamers, M.C., and Carsetti, R. (1999). J. Exp. Med. 190, 75–89. Lu, T.T., and Cyster, J.G. (2002). Science 297, 409–412. Martin, F., and Kearney, J.F. (2002). Nat. Rev. Immunol. 2, 323–335. Mikkola, H.K., Klintman, J., Yang, H., Hock, H., Schlaeger, T.M., Fujiwara, Y., and Orkin, S.H. (2003). Nature 421, 547–551. Ohishi, K., Katayama, N., Shiku, H., Varnum-Finney, B., and Bernstein, I.D. (2003). Semin. Cell Dev. Biol. 14, 143–150. Orkin, S.H., and Zon, L.I. (2002). Nat. Immunol. 3, 323–328. Radtke, F., Wilson, A., Stark, G., Bauer, M., van Meerwijk, J., MacDonald, H.R., and Aguet, M. (1999). Immunity 10, 547–558. Reya, T., Duncan, A.W., Ailles, L., Domen, J., Scherer, D., Willert, K., Hintz, L., Nusse, R., and Weissman, I.L. (2003). Nature, in press. Saito, T., Chiba, S., Ichikawa, M., Kunisato, A., Asai, T., Shimizu, K., Yamaguchi, T., Yamamoto, G., Seo, S., Kumano, K., et al. (2003). Immunity 18, this issue, 675–685. Stier, S., Cheng, T., Dombkowski, D., Carlesso, N., and Scadden, D.T. (2002). Blood 99, 2369–2378. Swiatek, P.J., Lindsell, C.E., del Amo, F.F., Weinmaster, G., and Gridley, T. (1994). Genes Dev. 8, 707–719. Tanigaki, K., Han, H., Yamamoto, N., Tashiro, K., Ikegawa, M., Kuroda, K., Suzuki, A., Nakano, T., and Honjo, T. (2002). Nat. Immunol. 3, 443–450. Vinuesa, C.G., Sze, D.M., Cook, M.C., Toellner, K.M., Klaus, G.G., Ball, J., and MacLennan, I.C. (2003). Eur. J. Immunol. 33, 297–305.
From the Yolk Sac to the Spleen: New Roles for Notch in Regulating Hematopoiesis Ivan Maillard,1,2 Yiping He,2,3 and Warren S. Pear2,3 Division of Hematology-Oncology 2 Abramson Family Cancer Research Institute 3 Department of Pathology and Laboratory Medicine Institute for Medicine and Engineering University of Pennsylvania 611 BRB II/III, 421 Curie Boulevard Philadelphia, Pennsylvania 19104
1
Although Notch receptors are widely expressed during hematopoiesis, their roles outside of the T cell lineage are not well characterized. Two reports in this issue of Immunity show that Notch1 and Notch2, respectively, are required to generate the earliest embryonic hematopoietic stem cells and splenic marginal zone B cells. Thus, different Notch receptors have specific and nonoverlapping functions that influence multiple hematopoietic lineages at various stages of development. Notch signaling is an evolutionarily conserved pathway controlling many developmental cell fate decisions. In hematopoiesis, Notch is required for T cell commitment and has been postulated to influence multiple other events, including hematopoietic stem cell (HSC) homeostasis and marginal zone B cell (MZB) development. Two reports in this issue of Immunity, Kumano et al. and Saito et al., provide new insights into the function of Notch signaling in HSCs and MZB cells. In early hematopoiesis, Notch1 appears critical for the generation of HSCs, whereas Notch2 is required for MZB cell development in the spleen. Mice with a targeted disruption of Notch1 develop multiple defects, including abnormal somitogenesis and impaired vascular development, a defect that may directly contribute to embryonic lethality before E11.5 (Swiatek et al., 1994; Conlon et al., 1995). Conditional inactivation of Notch1 avoids such lethality, providing the opportunity to characterize the role of Notch1 in specific organ systems (for example, see Radtke et al., 1999). This approach has revealed the critical role of Notch1 in the T/B lineage decision during lymphoid development, while most other aspects of hematopoiesis were unaffected, including the ability to competitively repopulate mixed bone marrow chimeras (Radtke et al., 1999). Similar findings were observed in mice after deletion of a floxed allele of CSL/RBP-Jk (CSL), a transcription factor that mediates most of the well-characterized effects of Notch (Han et al., 2002). In this issue of Immunity, Kumano et al. closely evaluate early hematopoiesis in Notch1 and Notch2 knockout embryos. Blood cell formation starts with primitive hematopoiesis in the extraembryonic yolk sac (YS), followed by definitive hematopoiesis in the embryo itself (reviewed by Orkin and Zon, 2002; Kondo et al., 2003). *Correspondence: [email protected]
Minireview
The first multipotent hematopoietic progenitors are found in the YS around E8.5 and thereafter in the paraaortic splanchnopleura (P-Sp) or aorta/gonad/mesonephros (AGM) region, where they differentiate from ventral mesoderm before colonizing the liver, and ultimately the bone marrow. Although the existence of true hemangioblasts remains debatable, early hematopoietic and endothelial progenitors are closely related in both the YS and the P-Sp region, where cells with shared hematopoietic and vascular markers have been identified. In their study, Kumano et al. use P-Sp explant cultures as an assay for early hematopoietic differentiation, together with transfer of progenitors to suitably conditioned recipients to assess stem cell activity. Characterization of Notch receptor expression in P-Sp shows that Notch1 and Notch4 but not Notch2 are coexpressed with the endothelial marker Flk-1, while hematopoietic cells emerging from the culture express Notch1 and Notch2 but not Notch4. Major differences between Notch1 knockout and wild-type P-Sp cultures were observed, while Notch2 knockout embryos were not different from controls. In the absence of Notch1, P-Sp explants had markedly impaired vascular network formation, consistent with a role of Notch1 in angiogenesis, while vascular beds reflective of vasculogenesis were preserved. Strikingly, sharply reduced numbers of hematopoietic colony forming cells (CFCs) were recovered from Notch1 knockout P-Sp cultures, indicating impairment of definitive hematopoiesis. Interestingly, close to normal numbers of CFCs were obtained from Notch1 knockout YS, suggesting that primitive hematopoiesis is at least partially preserved. However, cells recovered from both YS and P-Sp of Notch1 knockout embryos showed no detectable reconstitution of conditioned newborn recipients, suggesting that Notch1 is crucial for generating true stem cells. The defect was cell autonomous since mixed knockout and wild-type cultures did not rescue the phenotype. The data correlated well with experiments using ␥-secretase inhibitors (GSI) to inhibit Notch in P-Sp cultures, with rescue by a constitutive form of Notch. Interestingly, no impairment in the generation of hematopoietic cells was observed when the P-Sp were obtained from E10.5 embryos, a time point when hematopoietic progenitors with stem cell potential are already present. Altogether, these results suggest that Notch1 is critical for the genesis of HSCs, while not being required once their pool has been established. This situation bears striking similarities to the role of other genes in early hematopoietic and vascular development. For example, mice with a targeted inactivation of SCL/tal-1 have no primitive and definitive hematopoiesis, as well as angiogenic defects (reviewed in Orkin and Zon, 2002). However, conditional inactivation of SCL/tal-1 does not result in detectable impairment of HSC function at later time points (Mikkola et al., 2003). In the case of Notch, these results contrast with previous data suggesting a role of Notch in HSC self-renewal (Stier et al., 2002; Kunisato et al., 2003; reviewed in Ohishi et al., 2003). However, these reports involved hyperactivation of the
Immunity 588
Figure 1. Multiple Roles for Notch in Hematopoiesis (A) Notch signaling plays distinct roles at several stages of hematopoietic development. During early hematopoiesis in the embryo, Notch1 is critical for the generation of cells with a hematopoietic stem cell potential from presumed hemangioblasts. It also plays a critical role later during vascular development, beyond vasculogenesis, although the mechanisms are unknown. During lymphopoiesis, Notch1 signaling is required for T lineage specification and its absence allows the development of the B lineage. While it is widely accepted that there is a CLP population that gives rise to the lymphoid lineage, a recent report identified a CLP-independent progenitor population (ETP) with a high propensity for T cell development (Allman et al., 2003). The role of Notch signaling is less clear in subsequent stages of T cell development and function. During splenic B cell development, Notch2 signaling is required for the generation of MZB cells. HSC, hematopoietic stem cell; CMP, common myeloid progenitor; CLP, common lymphoid progenitor; ETP, early T cell progenitor; T, T cells; B, B cells; NK, nature killer cells; Tr. B, transitional B cells; Fo, follicular B cells; and MZ, marginal zone B cells. (B) Structure of the four mammalian Notch receptors (reviewed in Allman et al., 2002). The Notch1 TAD is stronger than the Notch2 TAD, when analyzed by reporter assays, while this domain is absent from Notch3 and Notch4. All four Notch receptors show similarity in their extracellular domains. ECN, extracellular Notch; ICN, intracellular Notch; EGFR, EGF repeats; LNR, LIN Notch repeats; ANK, ankyrin repeats; NCR, Notch cytokine response region; TAD, transactivation domain; P, PEST domain; and M, membrane.
Notch pathway either through ligand-mediated stimulation or through the expression of constitutive forms of Notch. Whether physiological levels of Notch signaling are important in stem cell maintenance and activity remains to be rigorously tested, although the wealth of data from knockout mice so far do not suggest a major defect (Radtke et al., 1999; Han et al. 2002). In contrast, HSC self-renewal may be influenced by other morphogens such as Sonic Hedgehog and Wnt, as shown re-
cently by experiments including loss of function approaches (Bhardwaj et al., 2001; Reya et al., 2003). The precise molecular relationship between Notch and these pathways in HSC biology has not been defined, although there is suggestive evidence that they may lie in a hierarchy (Reya et al., 2003). In terms of molecular mechanisms, it will be important to characterize the precise interrelationships between Notch and other molecules known to be involved in the generation of early hematopoietic progenitors, such as the transcription factors SCL/Tal-1, GATA-2, LMO2, AML1, HoxB4, and the VEGF receptor Flk-1 (Orkin and Zon 2002; Kondo et al., 2003). Kumano et al. show that the expression of many of these factors is downregulated in the absence of Notch1, suggesting that Notch may function as a master switch in the early hematopoietic differentiation program. This may have special relevance since Notch is a transmembrane receptor able to provide critical environmental signals to the early stem cells. Where these signals come from, and what are the relevant Notch ligands and ligand-expressing cells during early hematopoietic differentiation will be another important question to address. The mechanisms by which Notch affects vascular development also need to be better understood. In contrast to the essential role of Notch1 in early HSC and T cell development, Saito et al. show that Notch2 is critical for MZB development. A dramatic decrease in MZB has been reported in conditional CSL knockout mice (Tanigaki et al., 2002), strongly suggesting involvement of the Notch pathway. Because CSL is a transcriptional repressor in the absence of Notch, it was unclear whether the defect resulted from loss of repression or absence of Notch signals. Indirect evidence supporting the latter was that mice lacking MINT, a putative Notch negative regulator, showed an increase in MZB cells (Kuroda et al., 2003). Significantly, this manuscript also showed that Notch2 was highly expressed in the splenic MZ. Saito et al., using conditional inactivation of Notch2, definitively show that Notch2 is required for MZB development through cell autonomous mechanisms. The level of Notch2 expression appears important, as heterozygous mice have an ⵑ50% decrease in MZB cells, in contrast to the Notch1-mediated T/B lineage decision where haploinsufficiency has not been observed. Since Notch2 is highly expressed in B lymphocytes, whereas Notch1 is expressed at very low levels, the differential effect of Notch1 and Notch2 may be related to their expression pattern, as perhaps in T cell development where Notch1 expression is high and Notch2 expression is low. MZB cells are thought to be important in T cell-independent immune responses and share some characteristics with B1 B cells (Martin and Kearney, 2002). They are located in the spleen MZ and interspersed amongst macrophages and follicular dendritic cells. Although integrin signaling is required to retain MZB cells in the MZ (Lu and Cyster, 2002), their development is poorly understood. One hypothesis is that upon leaving the BM, B cells give rise to transitional T1 B cells that then form T2 B cells and may be the precursors to both follicular (Fo) and MZB cells (Loder et al., 1999). In conditional CSL KO mice, both a decrease in MZ cells and an increase in Fo B cells were observed, while other B
Minireview 589
cell subsets were unaffected. This suggested that MZ and Fo B cells arise from a common precursor. In contrast, Saito et al. find that T2 B cells are also decreased. This finding is largely derived from immunophenotypic analysis of CD21 expression in combination with other markers in control and knockout mice. However, the authors also show that CD21 expression decreases in the knockout mice, which complicates the assessment of the decline in the T2 population. If there is a true loss in the T2 population, it is surprising that the MZ but not the Fo B cell pool is affected. As recent data suggest that MZB cells can also arise from mature lymph node recirculating B cells (Vinuesa et al., 2003), the exact precursor/product relationship of these B cell subpopulations awaits further clarification. In any case, Saito et al. provide a possible hint toward the identification of a MZ precursor by showing that a subpopulation of CD9⫹ CD1dhigh CD21high CD23⫹ non-MZB cells also declines in the absence of Notch2. Many pathways influence MZB cell development (reviewed in Martin and Kearney, 2002). These include pathways that affect BCR signaling (such as CD19, CD22, and Aiolos), cell survival (e.g., Baff), cell migration/retention (e.g., integrin and Pyk2-mediated chemokine receptor signaling), and NFB signaling. The relation of Notch to these pathways is unknown. The mechanisms of Notch2 activation in MZB cells and the downstream signals are poorly understood. Two members of the Delta family of Notch ligands, Dll1 and 4, are expressed in the follicular dendritic cells (FDC) found in the MZ (Tanigaki et al., 2002). In this scenario, the MZ precursor would arrive in the MZ where Delta signals derived from FDCs would trigger maturation into MZB cells. Another potential ligand is Jagged1, which is highly expressed in the splenic MZ, although the identity of the Jagged1-expressing cells has not been shown (Bash et al., 1999). Regarding downstream mediators of Notch2 signaling in MZB development, Saito et al. show that Deltex1 is highly expressed in MZB cells. Deltex is a Notch modifier whose precise function is not known. In some contexts, it augments Notch signaling, whereas in others it is inhibitory. For example, transgenic expression of Deltex1 in HSCs blocks T cell development, suggesting that it antagonizes Notch1, whereas Drosophila Deltex acts as a positive Notch modulator (Izon et al., 2002, and references therein). Therefore, it is difficult to predict the role of Deltex1 in MZB development, and additional studies are required to determine if Deltex1 and/or other Notch transcriptional targets are important. In summary, these two manuscripts show that Notch1 and Notch2 play nonoverlapping roles in HSC and MZB development. It remains to be determined whether Notch signaling is controlled at the level of ligand interaction or regulatory molecules, such as Numb or Fringe. In addition, very little is known about the transcriptional regulation of the Notch receptor genes themselves, whose expression is likely to provide a significant level of control. Furthermore, since all four mammalian Notch receptors bind CSL, the mechanisms by which different receptors lead to different outcomes is not known. Understanding the details of Notch signaling in specific contexts, as revealed by these two manuscripts, will provide the basis for investigating these unanswered questions.
Selected Reading Allman, D., Aster, J.C., and Pear, W.S. (2002). Immunol. Rev. 187, 75–86. Allman, D., Sambandam, A., Kim, S., Miller, J.P., Pagan, A., Well, D., Meraz, A., and Bhandoola, A. (2003). Nat. Immunol. 4, 168–174. Bash, J., Zong, W.X., Banga, S., Rivera, A., Ballard, D.W., Ron, Y., and Galinas, C. (1999). EMBO J. 18, 2803–2811. Bhardwaj, G., Murdoch, B., Wu, D., Baker, D.P., Williams, K.P., Chadwick, K., Ling, L.E., Karanu, F.N., and Bhatia, M. (2001). Nat. Immunol. 2, 172–180. Conlon, R.A., Reaume, A.G., and Rossant, J. (1995). Development 121, 1533–1545. Han, H., Tanigaki, K., Yamamoto, N., Kuroda, K., Yoshimoto, M., Nakahata, T., Ikuta, K., and Honjo, T. (2002). Int. Immunol. 14, 637–645. Izon, D.J., Aster, J.C., He, Y., Weng, A., Karnell, F.G., Patriub, V., Xu, L., Bakkour, S., Rodriguez, C., Allman, D., et al. (2002). Immunity 16, 231–243. Kondo, M., Wagers, A.J., Manz, M.G., Prohaska, S.S., Scherer, D.C., Beilhack, G.F., Shizuru, J.A., and Weissman, I.L. (2003). Annu. Rev. Immunol. 21, 759–806. Kumano, K., Chiba, S., Kunisato, A., Sata, M., Saito, T., NakagamiYamaguchi, E., Yamaguchi, T., Masuda, S., Shimizu, K., Takahashi, T., et al. (2003). Immunity 18, this issue, 699–711. Kunisato, A., Chiba, S., Nakagami-Yamaguchi, E., Kumano, K., Saito, T., Masuda, S., Yamaguchi, T., Osawa, M., Kageyama, R., Nakauchi, H., et al. (2003). Blood 101, 1777–1783. Kuroda, K., Han, H., Tani, S., Tanigaki, K., Tun, T., Furukawa, T., Taniguchi, Y., Kurooka, H., Hamada, Y., Toyokuni, S., et al. (2003). Immunity 18, 301–312. Loder, F., Mutschler, B., Ray, R.J., Paige, C.J., Sideras, P., Torres, R., Lamers, M.C., and Carsetti, R. (1999). J. Exp. Med. 190, 75–89. Lu, T.T., and Cyster, J.G. (2002). Science 297, 409–412. Martin, F., and Kearney, J.F. (2002). Nat. Rev. Immunol. 2, 323–335. Mikkola, H.K., Klintman, J., Yang, H., Hock, H., Schlaeger, T.M., Fujiwara, Y., and Orkin, S.H. (2003). Nature 421, 547–551. Ohishi, K., Katayama, N., Shiku, H., Varnum-Finney, B., and Bernstein, I.D. (2003). Semin. Cell Dev. Biol. 14, 143–150. Orkin, S.H., and Zon, L.I. (2002). Nat. Immunol. 3, 323–328. Radtke, F., Wilson, A., Stark, G., Bauer, M., van Meerwijk, J., MacDonald, H.R., and Aguet, M. (1999). Immunity 10, 547–558. Reya, T., Duncan, A.W., Ailles, L., Domen, J., Scherer, D., Willert, K., Hintz, L., Nusse, R., and Weissman, I.L. (2003). Nature, in press. Saito, T., Chiba, S., Ichikawa, M., Kunisato, A., Asai, T., Shimizu, K., Yamaguchi, T., Yamamoto, G., Seo, S., Kumano, K., et al. (2003). Immunity 18, this issue, 675–685. Stier, S., Cheng, T., Dombkowski, D., Carlesso, N., and Scadden, D.T. (2002). Blood 99, 2369–2378. Swiatek, P.J., Lindsell, C.E., del Amo, F.F., Weinmaster, G., and Gridley, T. (1994). Genes Dev. 8, 707–719. Tanigaki, K., Han, H., Yamamoto, N., Tashiro, K., Ikegawa, M., Kuroda, K., Suzuki, A., Nakano, T., and Honjo, T. (2002). Nat. Immunol. 3, 443–450. Vinuesa, C.G., Sze, D.M., Cook, M.C., Toellner, K.M., Klaus, G.G., Ball, J., and MacLennan, I.C. (2003). Eur. J. Immunol. 33, 297–305.