Portal de Periódicos da CAPES

IL-7Ralpha and E47: independent pathways required for development of multipotent lymphoid progenitors

2002; Springer Nature; Volume: 21; Issue: 1 Linguagem: Inglês

10.1093/emboj/21.1.103

1460-2075

Barbara L. Kee, Gretchen Bain, Cornelis Murre,

Immune Response and Inflammation

Article15 January 2002free access IL-7Rα and E47: independent pathways required for development of multipotent lymphoid progenitors Barbara L. Kee Corresponding Author Barbara L. Kee Present address: University of Chicago, Department of Pathology and Committee on Immunology, 5841 S. Maryland Avenue, MC 1089, Chicago, IL, 60637 USA Search for more papers by this author Gretchen Bain Gretchen Bain Department of Biology, University of California San Diego, La Jolla, CA, 92093 USA Search for more papers by this author Cornelis Murre Cornelis Murre Department of Biology, University of California San Diego, La Jolla, CA, 92093 USA Search for more papers by this author Barbara L. Kee Corresponding Author Barbara L. Kee Present address: University of Chicago, Department of Pathology and Committee on Immunology, 5841 S. Maryland Avenue, MC 1089, Chicago, IL, 60637 USA Search for more papers by this author Gretchen Bain Gretchen Bain Department of Biology, University of California San Diego, La Jolla, CA, 92093 USA Search for more papers by this author Cornelis Murre Cornelis Murre Department of Biology, University of California San Diego, La Jolla, CA, 92093 USA Search for more papers by this author Author Information Barbara L. Kee 2, Gretchen Bain1 and Cornelis Murre1 1Department of Biology, University of California San Diego, La Jolla, CA, 92093 USA 2Present address: University of Chicago, Department of Pathology and Committee on Immunology, 5841 S. Maryland Avenue, MC 1089, Chicago, IL, 60637 USA *Corresponding author. University of Chicago, Department of Pathology and Committee on Immunology, 5841 S. Maryland Avenue, MC 1089, IL 60637, USA, E-mail: [email protected] The EMBO Journal (2002)21:103-113https://doi.org/10.1093/emboj/21.1.103 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Mice that lack the transcription factors encoded by the E2A gene or the receptor for interleukin 7 (IL-7R) have severe overlapping defects in lymphocyte development. Here, we show that E2A proteins are required for the survival of early T-lineage cells; however, they function through a pathway that is distinct from the survival pathway initiated by IL-7R signaling. While E2A proteins are required to suppress caspase 3 activation, ectopic expression of the anti-apoptotic protein Bcl-2 is not sufficient to overcome the lymphopoietic defects observed in the absence of E2A. Remarkably, mice that lack both IL-7Rα and E47 display a synergistic decrease in the number of T-cell, NK-cell and multipotent progenitors in the thymus, indicating that these distinct survival pathways converge to promote the development of multipotent lymphoid progenitors. Introduction The mature cells of the hematopoietic system develop from multipotent stem cells through the progressive restriction in their developmental potential (reviewed in Morrison et al., 1997). Multiparameter flow cytometry, combined with in vitro culture systems that support hematopoietic differentiation, have been used to identify an array of multipotent progenitor cell types with lymphopoietic activity. These populations include bipotent B cell/macrophage (Cumano et al., 1992; Montecino-Rodriguez et al., 2001), T cell/natural killer (NK) cell (Rodewald et al., 1992; Carlyle et al., 1997) and common lymphoid progenitors (CLPs), which have B, T, NK and dendritic cell potential (Galy et al., 1995; Kondo et al., 1997b). At the present time, very little is known about the factors regulating the development of these multipotent progenitors or their differentiation into lineage-restricted progenitors. The CLP population can be identified, in part, by expression of receptors for growth factors and cytokines that may regulate their growth, survival and differentiation (Kondo et al., 1997b; Tudor et al., 2000). CLPs express the receptor for interleukin (IL)-7, a cytokine that promotes the proliferation and survival of B- and T-lymphocyte progenitors (Namen et al., 1988; Goodwin et al., 1990; Park et al., 1990; Corcoran et al., 1996; Kim et al., 1998). IL-7 is an essential cytokine, as demonstrated by the block in early B- and T-cell development observed in mice that lack IL-7, or its receptor (IL-7R), or in mice administered a neutralizing antibody against IL-7 (Morrissey et al., 1991; Grabstein et al., 1993; Peschon et al., 1994; von Freeden-Jeffrey et al., 1995). IL-7R is composed of two proteins: the α chain and the γ chain (γc), which is a common component of the receptors for IL-2, IL-4, IL-7, IL-9 and IL-15 (Goodwin et al., 1990; Kondo et al., 1993; Noguchi et al., 1993; Russell et al., 1993). Both IL-7Rα−/− and γc−/− mice have severe defects in lymphocyte development (Peschon et al., 1994; DiSanto et al., 1995; He and Malek, 1996; Maki et al., 1996). Total thymocyte numbers are reduced in both strains of mice, and there are few CD4−CD8− (double-negative, DN) cells that express CD25, a marker associated with commitment to the T-cell lineage (Godfrey et al., 1993). IL-7Rα−/− mice have normal numbers of NK cells and CLPs, but lack γδ T cells due to an inability to express the T-cell receptor (TCR) γ and δ loci (He and Malek, 1996; Maki et al., 1996; Kondo et al., 1997a; Perumal et al., 1997; Ye et al., 1999). IL-7 has also been implicated in the rearrangement of the TCR α and β loci and the immunoglobulin heavy (IgH) chain locus (Muegge et al., 1993; Corcoran et al., 1996, 1998). However, the major function of IL-7 in T-cell development is to promote cell survival by controlling the ratio of anti-apoptotic to pro-apoptotic members of the Bcl-2 family (Akashi et al., 1997; Kondo et al., 1997a; Maraskovsky et al., 1997). Expression of the anti-apoptotic protein Bcl-2 in DN thymocytes is IL-7 dependent and IL-7 inducible, and ectopic expression of Bcl-2 is sufficient to overcome the defects in T-cell development in IL-7Rα−/− and γc−/− mice (Akashi et al., 1997; Kondo et al., 1997a; Maraskovsky et al., 1997; von Freeden-Jeffry et al., 1997; Kim et al., 1998). Surprisingly however, IL-7 is not required for the development of CLPs and other multipotent lymphoid progenitors (MLPs) even though the receptor is expressed on these cells (Kee and Paige, 1996; Kondo et al., 1997a). IL-7R functions, in part, through activation of the Janus family kinases Jak1 and Jak3, which activate the signal transducers and activators of transcription (STAT) factors STAT1 and STAT5 (reviewed in Leonard and O'Shea, 1998). In addition, IL-7R signaling leads to the activation of numerous proteins involved in signal transduction, such as PI3-kinase, pim-1 and the src related kinases (Leonard and O'Shea, 1998). It has also been hypothesized that IL-7R signaling may regulate the E2A transcription factors since the phenotypes of E2A−/− and IL-7Rα−/− mice are similar (Bain and Murre, 1998; Kee et al., 2000). Alternatively, IL-7R signaling and E2A proteins may function in distinct pathways that are required for the development of the same cell types. The E2A gene encodes two E-box binding transcription factors (E-proteins): E12 and E47. These proteins contain a basic domain, which is required for binding to DNA, followed by a helix–loop–helix (HLH) protein dimerization domain (Murre et al., 1994). E12 and E47 are identical except for the use of alternatively spliced exons encoding the bHLH domains. However, the bHLH domains of these proteins are >80% identical and both proteins bind to the same DNA sequence, although E47 has a higher affinity for this sequence than E12 (Sun and Baltimore, 1991). In extracts of B-lineage cells, the predominant E-box binding activity is composed of E47 homodimers (Bain et al., 1993; Shen and Kadesch, 1995), whereas in T-cell extracts, E-box binding activity consists of heterodimers of E47 and another E-protein, HEB (Sawada and Littman, 1993). E2A−/− and E47−/− mice, which lack E47 and have <10% the normal level of E12, lack B-lineage cells and have a number of defects in T-cell development (Bain et al., 1994, 1997a; Zhuang et al., 1994). Thymocyte numbers are decreased 3- to 5-fold in these mice and they exhibit a partial developmental arrest at the transition from CD44+CD25− to the CD44+CD25+ stage (Bain et al., 1997a). As in IL-7Rα−/− mice, E2A−/− mice have decreased numbers of CD4+CD8+ (double-positive, DP) thymocytes compared with wild-type (WT) animals. This phenotype is attributed to both decreased survival of DP cells and increased maturation of DP cells (Bain et al., 1999a). The absence of E2A proteins also leads to alterations in the timing of rearrangement of TCR γ and δ loci, and the absence of some γδ T-cell populations (Bain et al., 1999b). However, unlike IL-7Rα−/− or γc−/− mice, E2A−/− mice have a high incidence of thymic lymphoma, suggesting additional roles for E2A proteins in the regulation of thymocyte homeostasis (Bain et al., 1997a; Yan et al., 1997). The phenotypic similarities between IL-7Rα−/− and E47−/− mice prompted us to examine whether these proteins might function in a common pathway to regulate early lymphocyte development. Herein we demonstrate that DN thymocytes from E47−/− mice show an increased rate of apoptosis and activation of caspase 3 compared with WT DN thymocytes. However, unlike IL-7R−/− mice, Bcl-2 expression does not rescue the thymic defects in E47−/− mice. Therefore, E2A proteins may promote survival through a mechanism that is distinct from IL-7R signaling. In support of this hypothesis we found no evidence for a compound phenotype in mice that are heterozygous for mutations in both IL-7Rα and E47. In contrast, mice that lack both IL-7Rα and E2A activity lack thymic T-lineage progenitors and NK cells as well as B lymphocytes. Taken together, these finding indicate that IL-7Rα and E47 function in distinct pathways that are required for the proper development of MLPs. Our data provide important insights into the mechanisms controlling lymphocyte survival and demonstrate an essential role for IL-7Rα and E2A proteins in MLPs. Results Bcl-2 does not rescue B- or T-cell development in E47−/− mice The similar phenotypes in B- and T-cell development in IL-7Rα−/− and E47−/− mice led to the hypothesis that these proteins may function in the same signaling pathway (Bain and Murre, 1998; Kee et al., 2000). Alternatively, IL-7R signaling and E47 could function independently to promote the development of the same cell types. IL-7R signaling is required for the expression of the anti-apoptotic protein Bcl-2 in DN thymocytes, and Bcl-2 can rescue T-cell development in IL-7Rα−/− and γc−/− mice (Akashi et al., 1997; Kondo et al., 1997a; Maraskovsky et al., 1997). IL-7 also maintains Bcl-2 expression in pro-B lymphocytes, but may have additional functions in either pro-B-cell survival or differentiation (Kondo et al., 1997a; Maraskovsky et al., 1998). To determine whether the lymphopoietic phenotype observed in E47−/− mice is due to loss of a Bcl-2-mediated survival pathway, we tested the ability of the H2K-Bcl-2 transgene to overcome the defects in B- and T-cell development in E47−/− mice. It has been reported that the Bcl-2 transgene driven by the H2K promoter is expressed in hematopoietic stem cells (HSC) as well as multipotent and committed lymphoid progenitors (Kondo et al., 1997a). We found that this Bcl-2 transgene is expressed at similar levels in the majority of lineage-negative (Lin−) bone marrow (BM) cells and DN thymocytes from WT, E47−/− and IL-7Rα−/− mice, indicating that the transgene is expressed at high levels early during lymphopoiesis (Figure 1A). As reported previously, the H2K-Bcl-2 transgene caused a slight increase in thymocyte number in IL-7Rα+/− mice and nearly restored thymocyte numbers in IL-7Rα−/− mice to the level observed in IL-7Rα+/− mice (Akashi et al., 1997) (Figure 1B). Expression of the Bcl-2 transgene also caused a 1.6-fold increase in the number of B220+CD43− lymphocytes in the BM of E47+/− mice (3.2 × 106 versus 5 × 106 cells per femur, respectively) (Figure 1D). This increase was due, primarily, to an increase in the B220hi mature B-cell population, as observed by Maraskovsky et al. (1998). The Bcl-2 transgene did not, however, rescue the development of B220+CD43− cells in E47−/− mice (Figure 1D). In addition, no CD19+ cells were found in E47−/− or E47−/−;H2K-Bcl-2 BM (data not shown). Therefore, as observed in IL-7Rα−/− mice, ectopic expression of Bcl-2 does not rescue the development of pro-, pre- or mature B lymphocytes in E47−/− mice. Figure 1.Forced expression of Bcl-2 fails to rescue B- or T-cell development in E47−/− mice. (A) Flow cytometric analysis of human Bcl-2 expression in lineage-negative (Lin−) BM and DN thymocytes from E47−/− (shaded histogram), WT (dark line) or IL-7Rα−/− (stippled line) Bcl-2 transgenic mice, or E47−/− non-transgenic animals (light line). (B) Graphic representation of the number of cells in the thymus of IL-7R+/− and IL-7Rα−/− mice or (C) E47+/− and E47−/− mice with or without the H2K-Bcl-2 transgene. The mean number of thymocytes and the standard deviation of 3–8 animals analyzed between 5 and 8 weeks of age is shown. (D) FACS analysis of B220 and CD43 expression on BM cells isolated from mice of the genotype indicated. The boxed area represents cells with a pre-B, immature B or mature B-cell phenotype. The percentage of cells within the boxed area is indicated for each plot. FACS analysis of CD4 and CD8 expression on total thymocytes (E), and CD44 and CD25 expression on DN thymocytes (F), from mice with the genotypes indicated. The percentage of cells in each quadrant or each region is indicated in the upper right of each plot. Download figure Download PowerPoint In E47+/− thymus, the Bcl-2 transgene caused a 1.9-fold increase in total cell numbers (Figure 1C). In contrast, ectopic expression of Bcl-2 in the thymus of E47−/− mice led to a modest, but not significant, increase in cell numbers (Figure 1C). Both E47−/− and E47−/−;H2K-Bcl-2 transgenic mice had a similar number of DP thymocytes (1.6 × 107 versus 1.9 × 107) and DN thymocytes (6 × 105 versus 7 × 105 cells) (Figure 1E). The slight decrease in the percentage of DN thymocytes, and increase in the percentage of CD4+ thymocytes, in the E47−/−;H2K-Bcl-2 transgenic mouse shown in Figure 1E is within the range observed for E47−/− mice. Moreover, the partial arrest in T-cell development at the DN CD44+CD25lo stage was observed consistently in both E47−/− and E47−/−;H2K-Bcl-2 transgenic mice (Figure 1F). Therefore, ectopic expression of Bcl-2 did not significantly affect the total number of thymocytes in E47−/− mice and is not sufficient to rescue the lymphopoietic defects observed in these animals. Spontaneous apoptosis of T-cell progenitors in the absence of E2A proteins E2A activity is required for survival of pro-B lymphocytes (Kee et al., 2001). However, the inability of Bcl-2 to rescue T-cell development in E47−/− mice led us to question whether E2A proteins are required for survival of DN thymocyte progenitors. To address this question we examined the sensitivity of DN CD25+ thymocytes from WT and E47−/− mice to spontaneous apoptosis during in vitro culture. At the time of isolation, there were already ∼2-fold more annexin V+ cells in the DN CD25+ population from E47−/− thymus than from WT thymus, indicating an increase in apoptosis in this population in vivo (Figure 2A, upper panels). After in vitro culture for 24 h, 25–70% of E47−/− DN thymocytes bound annexin V, whereas only 2–6% of WT thymocytes were annexin V+ (Figure 2A, lower panels). In addition, by 24 h, there were significantly more cells with active caspase 3, an executioner of the apoptotic process, in E47−/− thymocyte cultures (26%) compared with WT thymocyte cultures (11%) (Figure 2B). While the overall degree of cell death varied in individual experiments, this pattern of sensitivity was observed consistently. Analysis of the number of viable DN thymocytes 24 and 48 h after the initiation of culture further indicated that E47−/− DN thymocytes were dying more rapidly than their WT counterparts (Figure 2C). Similar to our previous observations in pro-B lymphocytes, addition of IL-7 to the culture media had only a minor effect on the viability of E47−/− DN thymocytes, although it promoted the survival of WT DN thymocytes (Figure 2C). Taken together, our data indicate that E2A proteins are required to suppress activation of caspase 3 and prevent apoptosis in DN CD25+ thymocytes even in the presence of IL-7. Figure 2.Spontaneous apoptosis of E47−/− DN CD25+ thymocytes in vitro. (A) Annexin V staining of DN thymocytes from WT (left panels) and E47−/− (right panels) mice at the time of isolation (upper panels) or after 24 h of in vitro culture (lower panels). (B) Intracellular staining for active caspase 3 in DN thymocytes from WT (left panel) or E47−/− mice after 24 h of in vitro culture. The DN thymocytes shown in (A) and (B) are from different animals. (C) Graphic representation of the relative number of WT (circles) or E47−/− (squares) DN thymocytes after culture for the time indicated in media alone (open) or IL-7 (filled). The mean ± SD from three independent wells is shown. Download figure Download PowerPoint Reduced splenic T-cell numbers in IL-7Rα−/−;E47−/− mice Our data indicate that E2A proteins promote lymphocyte survival through a Bcl-2-independent mechanism or possibly downstream of Bcl-2-related proteins. Given that Bcl-2 is the major target of the IL-7 survival pathway, we questioned whether E2A proteins were functioning in the same pathway as IL-7R signaling or through an alternative pathway. To address this question we created mice that lack both IL-7Rα and E47 (IL-7Rα−/−;E47−/− mice), and examined the effect of this double mutation on T-cell development. We predicted that if these proteins function in the same pathway, the phenotype of IL-7Rα−/−;E47−/− mice would be similar to that of either single-gene-deficient animal since in the absence of one gene the function or expression of the other would be compromised. In addition, mice that are heterozygous for both mutations might display a phenotype that is more severe than that of either single-heterozygous animal, as seen with other proteins that function in a common pathway, or regulate overlapping targets (Floss et al., 1996; Dunn et al., 1997; O'Riordan and Grosschedl, 1999). However, if these proteins function in distinct pathways, the double-heterozygous animals would not be predicted to display a defect beyond that of either single-heterozygous animal, but we might observe novel defects in the absence of both proteins. We isolated the spleen from mice of all genotypes generated from IL-7Rα+/−;E47+/− × IL-7Rα+/−;E47+/− crosses, and counted the number of viable cells. We found no difference in the phenotype or number of cells from WT thymus compared with any heterozygous combination, and therefore each of these genotypes was considered WT for this analysis. Similarly, the phenotype and number of thymocytes from IL-7Rα−/− mice were identical to IL-7Rα−/−;E47+/− mice and E47−/− mice were identical to IL-7R+/−;E47−/− mice. Therefore, these animals were considered IL-7Rα−/− or E47−/− for the analysis presented here. The average number of splenocytes in WT, IL-7Rα−/− or E47−/− mice was 4.6 × 107, 1.6 × 107 and 107, respectively (Figure 3A). The spleens from IL- 7Rα−/−;E47−/− mice were visibly smaller than those of their littermates and contained on average only 6.5 × 106 cells (Figure 3A). In our analysis, an average of 2 × 107 T cells were found in the spleen of WT animals. IL-7Rα−/− spleen contained an average of 6 × 106 T cells, representing a 3.3-fold decrease (Figure 3). E47−/− spleen contained an average of 7 × 106 T cells, a 2.8-fold decrease compared with WT spleen, although the percentage of splenocytes that are T lymphocytes is greatly increased owing to the absence of B lymphocytes in these animals (Bain et al., 1994, 1997a; Zhuang et al., 1994). Remarkably, 1% of WT and those with <1% of WT (Peschon et al., 1994; Maraskovsky et al., 1998). IL-7Rα−/− mice with 1%, lacking CD4- and CD8-expressing T-cell populations. However, this distinction has not been observed in our colony and no IL-7Rα−/− mice have been identified that lack DP or single-positive (SP) thymocytes, even in the few cases where thymocyte cellularity was 104-fold) decrease in thymocyte numbers in IL-7Rα−/−;E47−/− mice compared with IL-7Rα−/− and E47−/− mice. Figure 4.Thymic hypocellularity and impaired T-cell development in IL-7Rα−/−;E47−/− mice. (A) Total thymocyte numbers are shown for WT, IL-7Rα−/−, E47−/− and IL-7Rα−/−;E47−/− mice. Each point represents the thymus from one mouse analyzed between 4 and 8 weeks of age. (B) Forward (FSC) and side light scatter (SSC) of cells isolated from mice of each of the indicated genotypes. Ten thousand cells within the indicated gate were collected for each genotype with the exception of the IL-7Rα−/−;E47−/− thymocytes, for which half of the total cells were acquired. (C) FACS analysis of CD4 and CD8 expression on thymocytes isolated from mice of each of the indicated genotypes. The percentage of cells in each quadrant is indicated in the upper right of each plot. (D) The total number of DN, DP, SP thymocytes from each of the indicated genotypes was determined from the total number of cells and the percentage of cells with the given phenotype. Each point represents the total number of cells in one thymus. The average number of thymocytes is indicated by the horizontal bar. Download figure Download PowerPoint To determine which thymic populations are altered in IL-7Rα−/−;E47−/− mice we examined the expression of CD4 and CD8 by flow cytometry. Remarkably, >95% of the cells in the IL-7Rα−/−;E47−/− thymus lacked CD4 and CD8 (Figure 4C). However, this represents <1 × 104 cells, whereas there are ∼3.5 × 106, 1 × 105 or 4 × 106 DN thymocytes in WT, IL-7Rα−/− or E47−/− thymus, respectively (Figure 4D). The decline in thymocyte numbers is even more severe when the absolute number of DP thymocytes is considered. WT, IL-7Rα−/− or E47−/− mice have an average of 1.2 × 108, 2.8 × 107 or 2 × 107 DP thymocytes, respectively (Figure 4D). In contrast, IL-7Rα−/−;E47−/− mice have an average of only 1.1 × 103 DP thymocytes, representing a 1 × 105-, 2.5 × 104- or 1.8 × 104-fold decrease compared with WT, IL-7Rα−/− or E47−/− mice, respectively (Figure 4D). SP thymocytes were also significantly reduced in IL-7Rα−/−;E47−/− compared with WT, IL-7Rα−/− or E47−/− mice. Therefore, the absence of IL-7Rα and E47 leads to a severe reduction in all thymocyte subsets. Absence of T-lineage progenitors in the thymus of IL-7Rα−/−;E47−/− mice In order to determine the stage at which developmental arrest occurs during T-cell development in the absence of IL-7Rα and E47, we examined DN thymocytes from IL-7Rα−/−;E47−/− mice for expression of CD44, CD25 and Thy1. The most immature cells in WT thymus lack CD4 and CD8, express high levels of CD44 and low levels of Thy1, and lack CD25 (Figure 5A) (Godfrey et al., 1993). These cells acquire CD25 as they commit to the αβ T-cell lineage, and subsequently downregulate CD44 and then CD25 before they progress to the DP stage. In the absence of IL-7Rα, there is a reduced number of DN CD25+ cells compared with WT thymus (2 × 104 versus 8.7 × 106 cells), but the majority (85.5%) of IL-7Rα−/− DN cells express Thy1, indicating that they are lymphocyte progenitors (Figure 5A and B). In the absence of E47, thymocyte development is arrested at the CD44+CD25lo to CD44+CD25+ stage, consistent with a role for E2A proteins in commitment of MLPs to the T lineage (Bain et al., 1997a). However, the majority of E47−/− DN thymocytes also express Thy1, demonstrating their T-lineage potential (Figure 5A and B). In contrast, <70 DN cells in the thymus of IL-7Rα−/−;E47−/− mice express CD44, CD25 or Thy1, indicating that these cells are not lymphocyte progenitors (Figure 5A and B). Therefore, in the absence of IL-7Rα and E47, T-cell development is arrested prior to the development of the first thymic immigrants with lymphopoietic activity. Figure 5.Absence of lymphocyte progenitors and NK cells in the thymus of IL-7Rα−/−;E47−/− mice. The DN population of thymocytes from mice of each of the indicated genotypes was analyzed by flow cytometry for the expression of CD44 and CD25 (A), Thy1 (B) and CD44 and DX5 (C). The percentage of cells in each quadrant is indicated in the upper right of each plot. The percentage of Thy1-positive cells is also noted. The plots in (C) are from different mice than those shown in (A) and (B). Download figure Download PowerPoint Absence of NK cells in IL-7Rα−/−;E47−/− thymus The earliest thymic immigrants are thought to be cells that have not yet undergone definitive lineage commitment, and include cells which can give rise to T cells as well as B lymphocytes and NK cells (Galy et al., 1995; Carlyle et al., 1997; Kondo et al., 1997b). Our analysis indicates that the absence of IL-7Rα and E47 inhibits the development of this MLP population. If these progenitors fail to develop, the number of NK cells in the thymus should be reduced. Alternatively, if this population develops normally, the few cells that remain in the IL-7Rα−/−;E47−/− thymus could be NK cells. In order to identify NK cells, we examined the DN population of IL-7Rα−/−;E47−/− thymus for cells that express CD44 and DX5, a pan-NK cell marker. A small population of CD44+DX5+ cells could be detected among WT DN thymocytes, representing, in the experiment shown, ∼104 cells (Figure 5C). The number of CD44+DX5+ cells was somewhat variable in IL-7Rα−/− and E47−/− mice; however, these thymi invariably had an increase in the percentage of CD44+DX5+ cells compared with WT DN thymocytes (Figure 5C). In some cases, the absolute number of DX5+ cells in E47−/− thymus was increased 3- to 4-fold. In the example shown, there were 1.3 × 104 DX5+ cells in the IL-7Rα−/− DN population and 2.4 × 104 DX5+ cells in the E47−/− DN population (Figure 5C and data not shown). In contrast, the IL-7Rα−/−;E47−/− DN population contained <102 DX5+ cells (Figure 5C). Therefore, the absence of both IL-7Rα and E47 affects the development of progenitors of T and NK cells. Bcl-2 can rescue the development of MLPs in IL-7Rα−/−;E47−/− mice Our data indicate that IL-7Rα and E47 function in distinct pathways that converge to promote the survival of MLPs. Therefore, either IL-7R signaling or E47 activity can promote MLP survival. If IL-7R signaling is required to increase the expression of pro-apoptotic Bcl-2-related proteins in MLPs we would predict that Bcl-2 should rescue the development of MLPs in IL-7Rα−/−;E47−/− mice. In order to test this prediction we created IL-7Rα−/−;E47−/− mice that also express the H2K-Bcl-2 transgene. Since this transgene is detected at high levels in Lin− BM cells we expect that it is expressed in MLPs and their BM-derived precursors (Figure 1A). We found that the IL-7Rα−/−;E47−/−;H2K-Bcl-2 transgenic mice had significantly more thymocytes than their