Notch participates in diverse cell fate decisions throughout embryonic development and postnatal life. Members of the NF-κB/Rel family of transcription factors are involved in the regulation of a variety of genes important for immune function. The biological activity of the NF-κB transcription factors is controlled by IκB proteins. Our previous work demonstrated that an intracellular, constitutively active form of human Notch-1/translocation-associated Notch homologue-1 (NotchIC) functions as an IκB molecule with specificity for the NF-κB p50 subunit and physically interacts with NF-κB in T cells. In the current study, we investigated the roles of different domains of NotchIC in the regulation of NF-κB-directed gene expression and NF-κB DNA binding activity. We found that NotchIC localizes to the nucleus and that a region in the N-terminal portion of NotchIC, not the six ankyrin repeats, is responsible for the inhibitory effects of Notch on NF-κB-directed gene expression and NF-κB DNA binding activity. The N-terminal portion of NotchIC inhibited p50 DNA binding and interacted specifically with p50 subunit, not p65 of NF-κB. The interaction between Notch and NF-κB indicates that in addition to its role in the development of the immune system, Notch-1 may also have critical functions in the immune response, inflammation, viral infection, and apoptosis through control of NF-κB-mediated gene expression.
Notch-1 receptors are transmembrane proteins that function in diverse developmental and cell maturation processes to control the choice between alternate cell fates (1, 2). Extracellular domains of Notch proteins (NotchEC)4 contain up to 36 epidermal growth factor-like repeats involved in ligand interaction (3) and three consecutive Lin-12/Notch motifs. The intracellular regions of Notch receptors (NotchIC) contain several functional motifs: ankyrin/CDC10 repeats, RAM, nuclear localization signals (NLS), PEST sequences, and a glutamine-rich domain (OPA). The PEST region may be involved in targeting the protein for proteolytic cleavage. NotchIC contains at least two NLSs, on both sides of the ankyrin repeats. The six ankyrin/CDC10 repeats constitute a putative protein-protein interaction site. The RAM domain (4, 5) is located between the transmembrane region and the ankyrin repeats and is essential for intracellular Notch to bind to transcription factor RBP-Jκ.
Notch receptors are proteolytically processed by a furin-like protease in the trans-Golgi network into the extracellular subunit NotchEC and a transmembrane C-terminal subunit. These two subunits are presented as a functional heterodimer on the cell surface (6, 7). Ligands such as Delta or Jagged/Serrate are thought to induce a second cleavage that requires presenilins (β-amyloid precursor processing-related proteins) at a site within the transmembrane region of Notch (8, 9). This cleavage is suggested to release the intracellular fragment NotchIC, which translocates to the nucleus (9, 10, 11, 12). NotchIC binds to numerous nuclear and cytoplasmic proteins (1, 13). The best-characterized Notch signaling pathway is thought to be mediated by binding to the transcription factors of the C promoter binding factor (CBF)-1/suppressor of hairless/Lag-1 family. The mammalian member of this family is CBF-1, also known as RBP-Jκ. CBF-1 is a DNA binding protein that recognizes the sequence CGTGGGAA. In the absence of Notch, it functions as a repressor that recruits silencing complexes containing various nuclear corepressors and histone deacetylase 1 (14, 15). Interaction with NotchIC appears to displace the silencing complex from CBF-1, converting it into a transcriptional activator (14). The nuclear coactivator ski- interacting protein is recruited to the CBF-1-Notch complex through an interaction with the ankyrin repeats, and it may mediate transcriptional activation by the complex (16). Downstream genes whose expression is up-regulated by Notch activation include the Enhancer of Split group in Drosophila and its mammalian homologues. This group includes several basic helix-loop-helix proteins, such as Hairy/Enhancer of Split-1 and -5 and Enhancer of Split-2 (the homologue of Drosophila groucho) (1).
The mammalian Notch-1 gene is expressed in a broad range of tissues, including the thymus and other lymphoid organs (17, 18, 19). Notch-1 plays multiple nonmutually exclusive roles in T cell development (20, 21). Constitutively active Notch-1 expression in hemopoietic progenitors blocks the development of lymphoid precursor toward the B cell lineage and promotes the formation of immature CD4+CD8+ T cell lymphomas (22). Conversely, inducible deletion of the Notch-1 gene in mice by cre/lox recombination (23) leads to severe thymic atrophy due to a block of the expansion of immature CD4−CD8− T cell precursors. Reduced Notch activity favors the γδ T cell fate over the αβ T cell fate (24). Expression of an activated form of Notch-1 in murine thymocytes under the lck promoter leads to an increase in CD8+ T cells and a parallel decrease in CD4+ T cells (21, 25). More recently, Yasutomo et al. (26) have shown that Notch-1 is necessary for the postcommitment development of CD8+ T cells rather than for the lineage choice between CD4 and CD8.
Truncated forms of Notch receptors that lack all or most of the extracellular region are constitutively active and have transforming activity in vitro and in vivo (1, 27). Such a constitutively active Notch-1 is produced by a 9:7 translocation in about 10% of the cases of human T cell lymphoblastic leukemia (17). This was defined translocation-associated Notch homologue (TAN-1). Overexpression of apparently wild-type Notch-1 is found in cervical carcinomas and various other solid tumors (28, 29). Depending upon the experimental model, Notch-1 activation can delay terminal differentiation (30, 31), modulate cell cycle progression (31), and regulate cell death (32, 33, 34). Aside from CBF-1- or Deltex-mediated regulation of basic helix-loop-helix transcription factors, other molecular targets of Notch-1 have been identified. Notch-1 has been reported to inhibit NF-κB-induced gene expression (35), to regulate the expression of transcription factor peroxisomal proliferator-activated receptors (36), and to induce NF-κB2 (p100) promoter activity (37). The orphan nuclear receptor nur77 (33) and the transcriptional regulator EMB-5 (38) of Caenorhabditis elegans are other putative mediators of Notch effects. These mediators have been shown to bind to the Notch subunit in various experimental systems (13).
NF-κB is the prototype of a family of dimeric transcription factors that play fundamental roles in the immune system (39, 40). The nuclear translocation and DNA binding of Rel-NF-κB complexes are prevented by the IκB family of proteins (40, 41, 42). These include IκB-α, IκB-β, IκB-ε, Bcl-3, ε, and the Drosophila protein cactus. In resting cells, IκB proteins sequester NF-κB in the cytoplasm. Signal-induced activation of IκB kinase (43, 44, 45) by IκB kinase kinase, NAK (46), phosphorylates IκBs and triggers their degradation (47). IκB phosphorylation and degradation lead to transcriptional activation of NF-κB and rapid resynthesis of IκB. Resynthesized IκB translocates to the nucleus, inhibits NF-κB DNA binding, and promotes the transport of NF-κB from the nucleus to the cytoplasm (48, 49).
The NF-κB transcription factors play multiple, crucial roles in the inflammatory and immune responses as well as in the development of the immune and hemopoietic systems. Regulations of cell survival, proliferation, and cytokine production are some of the many functions attributed to these transcription factors (40).
The functional interaction between the Notch signaling network and the NF-κB family is complex. We demonstrated that constitutively active human Notch-1/TAN-1 functions as a novel IκB molecule and regulates NF-κB-mediated gene expression through a direct interaction with the NF-κB p50 subunit. This prevents NF-κB from binding to NF-κB recognition sites in DNA to regulate NF-κB-dependent gene expression (35). Our data have shown for the first time that Notch directly interacted with the nuclear transcription factor NF-κB family member and revealed an important molecular basis for the pivotal role of Notch in the immune system. Other investigators have shown recently that Rel-NF-κB can activate the Notch signaling pathway by inducing the expression of Jagged 1, a ligand for Notch receptors (21), and that NF-κB2 (p100/p52) is a putative target gene of activated Notch-1 via CBF-1/RBP-Jκ (37).
The mechanism of the direct interaction of Notch-1 with p50 and its possible physiological roles vis-à-vis the other Notch signaling pathways remain unclear. By analogy with IκB, the interaction of Notch with NF-κB may have been through the ankyrin repeats of Notch.
In the present study, we have mapped the interaction of Notch/TAN-1 with NF-κB. We found that this interaction is mediated by a domain of Notch-1 that is N-terminal to the ankyrin repeats and appears to be independent of these repeats. Thus, Notch-1 regulates NF-κB through a novel mechanism that does not involve ankyrin repeats interactions. The possible significance of this mechanism is discussed.
Materials and Methods
Notch mutants were amplified by PCR using high fidelity PFU DNA polymerase (Stratagene, La Jolla, CA). Specific primers for each mutant are as follows: N + 6ANK, P1/G2; C + 6ANK, P2/G1; 6ANK, P2/G2; N, P1/G5; P1 (sense), TCG GAT CCA TGG GCT TCA AAG TGT CTG AG; P2 (sense), TCG GAT CCA TGG GGC CTG ATG GCT TCA CCC CG; G1 (antisense), GTG AAT TCG AGG GGG CAC GGA CGG AGA C; G2 (antisense), GTG AAT TCC ATA TGA TCC GTG ATG TC; and G5 (antisense), GTG AAT TCA GGC CCG CGG ACA TTG AC. Amplified PCR products were purified, then digested with BamHI and EcoRI and cloned into pcDNA3 vector for expression in eukaryotic cells. The sequence of each mutant was verified, and each mutant insert was transferred into pGEX1λT vector for expression in bacteria. The reporter plasmid p-6κB-luc (provided by Dr. P. Baeuerle, Micromet, Martinsried, Germany) contains a luciferase gene driven by six reiterated κB sites. Plasmids expressing p50 and p65 have been described previously (50).
Bacterial expression and purification of GST-NotchIC and its mutants
Notch-GST fusion proteins were expressed in Escherichia coli BL21(DE3) and purified as previously described (35). Briefly, after mild sonication, cell debris was removed by centrifugation. The supernatant was loaded onto a Sepharose-glutathione affinity column, washed with solution A (0.5% Triton X-100 in 0.5 M NaCl and 10 mM Tris) and solution B (50 mM Tris (pH 8.0)), then eluted with 10 mM glutathione in 50 mM Tris (pH 8.0). Eluted proteins were aliquoted and stored at −20°C. The molecular mass and purity of each fusion protein were determined by 10% SDS-PAGE, followed by Coomassie blue staining.
Nuclear extracts and EMSA
Jurkat T cells were stimulated with PMA (50 ng/ml; Sigma, St. Louis, MO) for 3 h. Nuclear extracts from stimulated Jurkat T cells were prepared as described previously (35). Purified GST-NotchIC or its derivatives (4 pmol each) were added directly to DNA-protein binding reactions (10 mM Tris (pH 7.5), 10 mM EDTA, 0.5 mM DTT, 50 mM NaCl, 2 μg poly(dI:dC), 5% glycerol, ∼0.5 ng 32P-labeled NF-κB oligonucleotides, and 6 μg nuclear extracts) and kept at room temperature for 30 min. The reaction products were run on 6% nondenaturing polyacrylamide gels with 0.5× Tris-borate-EDTA buffer at room temperature. The gels were dried and exposed to x-ray film at −70°C.
In vitro transcription and translation
Translation of full-length p50 and p65 was performed with in vitro transcription and translation system T7-coupled wheat-germ extract system in the presence of [35S]methionine according to the manufacturer’s protocols (Promega, Madison, WI).
Association of p50 with GST-NotchIC mutants
Purified GST-TAN-1 mutants were loaded on MicroSpin GST purification columns (Amersham, Indianapolis, IN) and incubated for 10 min. Columns were washed twice with PBS. [35S]Methionine-labeled p50 or p65 was added to each column and incubated at 4°C for 2 h. Columns were washed with PBS three times and eluted with 10 mM glutathione in PBS. Eluted samples were run on a 4–12% SDS-PAGE, followed by autoradiography.
Human NTera-2 embryonal carcinoma cells (American Type Culture Collection, Manassas, VA) were grown on Lab-Tek chamber slides (Nalge Nunc International, Naperville, IL) overnight, then transfected with LipofectAMINE (Life Technologies, Gaithersburg, MD) according to the manufacturer’s instructions. After 48 h, cells were permeabilized with 90% methanol at −20°C for 30 min, washed twice with PBS, twice with NGS (50% PBS (pH 7.0), 25% normal rabbit serum, 25% normal goat serum, and 0.1% azide), and then incubated with rabbit polyclonal anti-Notch serum as the first reagent (1/100 dilution) or normal rabbit serum (1/100 dilution) as a control at 37°C for 1 h. Cells were washed twice with NGS (4°C, 15 min/wash), then incubated at 37°C for 1 h with FITC-conjugated F(ab′)2 of polyclonal goat anti-rabbit IgG (Caltag, South San Francisco, CA), followed by three washes at 37°C with NGS (15 min/wash). Cells were observed by epifluorescence microscopy.
Transient transfection and luciferase assay
Ntera-2 cells were cultured in DMEM containing 10% FCS and transfected with LipofectAMINE and the indicated plasmids for 48 h. Cells were lysed, and the lysates were clarified by centrifugation at 10,000 rpm for 5 min. An equal amount of supernatant from each lysate was assayed for luciferase activity using a kit from Promega and a Microtiter Plate Luminometer ML3000 with the Biolinx 2.10 program (Dynatech Laboratories, Chantilly, VA).
Ntera-2 cells were transfected with expression vectors in a six-well plate in the presence of LipofectAMINE. After 48 h, the cells were lysed, and lysates were centrifuged. Supernatants were boiled with sample loading buffer under reducing conditions and then run on a 4–12% SDS-PAGE, followed by electrotransferring to polyvinylidene difluoride membrane. Membranes were blocked with 3% BSA-PBS for 1 h, and then sequentially incubated with rabbit anti-human Notch-1 Ab followed by alkaline phosphatase-conjugated anti-rabbit IgG. Bound Abs were detected in the presence of alkaline phosphatase substrate.
Inhibition of NF-κB binding to DNA by NotchIC and its mutants
We have previously shown that NotchIC inhibited NF-κB binding to DNA (35). To investigate which domain(s) within Notch-1 was responsible for the interaction between NotchIC and NF-κB, we constructed a series of NotchIC mutants, as described in Fig. 1⇓, and cloned into expression vector pcDNA3. Additionally, NotchIC mutants were expressed as GST fusion proteins in E. coli BL21(DE3) and purified by Sepharose-glutathione affinity column as shown in Fig. 2⇓A. GST fusion proteins of NotchIC and its mutants migrated at the predicted molecular masses on 10% SDS-PAGE. The effects of mutant proteins on NF-κB DNA binding were assessed by EMSA. NotchIC completely inhibited NF-κB binding to DNA (Fig. 2⇓B), in agreement with our previous report (35). Interestingly, the six ankyrin repeats by themselves showed no inhibitory effect on NF-κB DNA binding activity. The C-terminal region of NotchIC together with the six ankyrin repeats showed weak inhibitory activity, whereas the N-terminal region of NotchIC together with the six ankyrin repeats abolished NF-κB DNA binding. Interestingly, the N-terminal portion of NotchIC without the six ankyrin repeats also completely prevented NF-κB binding to DNA (Fig. 2⇓C). These data indicate that the N-terminal portion of NotchIC is sufficient for inhibition of NF-κB binding to DNA in vitro, and this activity is independent of the ankyrin repeats.
Direct association of p50 to the N-terminal portion of NotchIC
We previously demonstrated that NotchIC specifically inhibits DNA binding of p50 and directly interacts with p50 subunit of NF-κB but not that of p65 (35). To determine the subunit specificity of NotchIC mutants, EMSA was performed by using in vitro transcript and translated p50 and p65. NotchIC, N + 6ANK, and N effectively inhibited DNA binding of p50 dimers (Fig. 3⇓A), but not that of p65 dimers (data not shown). The mutants without the N-terminal portion of NotchIC, C + 6ANK and 6ANK, had no effect on p50 DNA binding.
Next we examined whether NotchIC mutants interacted directly with p50. To prove such a direct protein-protein interaction, purified GST-NotchIC mutants were tested for precipitation of in vitro translated 35S-labeled p50 or p65 on glutathione-Sepharose beads. Both N and N + 6ANK precipitated p50, but not p65. 6ANK precipitated neither p50 nor p65. These data further confirmed that the N-terminal portion of NotchIC is sufficient for inhibition of NF-κB binding to DNA in vitro, and this activity is dependent on direct interaction of the N-terminal portion of NotchIC with the NF-κB p50 subunit.
Inhibition of NF-κB mediated gene expression by NotchIC and its mutants
After activation, NF-κB migrates to the cell nucleus, where it recognizes specific DNA elements and regulates gene expression. The interaction between NF-κB and inhibitory proteins of the IκB family occurs primarily in the cytoplasm, except for Bcl-3. To identify the cellular compartment where NF-κB interacts with Notch-1, we investigated the subcellular location of transfected NotchIC in NTera-2 cells by immunofluorescence. As shown in Fig. 4⇓, NotchIC immunoreactivity was localized in the nuclei. This suggests that the primary site of interaction with NF-κB may be the nucleus.
We next examined the effect of NotchIC mutants on NF-κB-mediated gene expression. The basal level of luciferase expression directed by six repeated κB elements was very low. When the cells were cotransfected with p50/p65, luciferase activity was dramatically increased. As we previously reported, NotchIC strongly inhibited p50/p65 induced reporter gene expression. Construct N + 6ANK, which contains the N-terminal region and the six ankyrin repeats, and N, which is 109 aa long and contains only the N-terminal region of NotchIC, strongly inhibited p50/p65-directed gene expression (Fig. 5⇓A). In contrast, constructs C + 6ANK (which contains the six ankyrin repeats and the C-terminal region of NotchIC) and 6ANK (which contains only the six ankyrin repeats) had no effect on p50/p65-directed gene expression. To exclude the possibility that NotchIC mutants are not equally expressed, parallel samples were lysed and assayed for NotchIC expression by Western blot with a rabbit polyclonal Ab generated against NotchIC protein. As shown in Fig. 5⇓B, all NotchIC mutants were equally well expressed, and the lack of inhibition of p50/p65-mediated gene expression by C + 6ANK and 6ANK is not because of the absence of NotchIC mutant protein expression. These data indicate that the N-terminal region of NotchIC inhibits NF-κB-mediated gene expression independently of the ankyrin repeats.
We demonstrated that Notch-1 inhibits NF-κB DNA binding and NF-κB-mediated gene expression independently of the ankyrin repeats, through a region N-terminal to these repeats, namely, aa 1773–1881. This 109-aa region of Notch-1 has 98, 98, 77, and 70% identities with the corresponding regions of rat, mouse, Xenopus, and zebrafish Notch, respectively (Fig. 6⇓). It overlaps with the mouse RAM domain (aa 1751–1806), except for the first 11 amino acids (RRQHGQLWFPE, 1751–1761) of the RAM domain. Mutation analysis (4) demonstrated that these 11 amino acids are essential for RBP-Jκ binding to Notch, whereas they are dispensable for NF-κB binding, because the NotchIC used in this study does not contain the corresponding 11 amino acids of human Notch-1. Therefore, we conclude that the domain we identified here is a novel protein-protein interaction domain and designated it the NF-κB binding domain (NBD). NBD and RAM are two partially overlapping, but different, domains. NBD contains one of the NLS. The NLS is highly conserved among these five Notch proteins. The high level of identity among NBDs of these four Notch proteins with that of human Notch-1 suggests that these four Notch proteins may also function as nuclear IκB molecules. NBD of human Notch-1 has 48, 30, 31, and 29% identity with the corresponding region of mouse Notch-3 (accession number Q61982) , mouse Notch-4 (accession number P31695), Drosophila melanogaster (accession number P07207) , and human Notch-4 (accession number AAC32288). The putative NLS seems conserved in these four proteins with three or four basic amino acids. No significant similarity was found between NBD of human Notch-1 with human Notch-2 and Notch-3.
In response to ligand binding, NotchIC is thought to translocate to the nucleus after a presenilin-dependent proteolytic cleavage and to act in the nucleus to regulate the transcription of downstream target genes (10, 51, 52, 53). Two putative NLSs are located on both sides of the ankyrin repeats in NotchIC. All the Notch-1 mutant constructs that show NF-κB inhibition in NF-κB-directed gene expression contain 1) the NBD region and 2) at least one NLS. Thus, a possible interpretation of our data is that the six ankyrin repeats are unable to interfere with NF-κB after transfection due to the lack of an NLS. However, the ankyrin repeats do not interfere with NF-κB DNA binding in EMSA assays. Additionally, the construct C + 6ANK does have an NLS in its C-terminal region, but it does not inhibit NF-κB-mediated gene expression. Thus, the most likely interpretation of our data is that the NBD domain and its NLS are both necessary and sufficient for NF-κB inhibition by NotchIC in our experimental model. Together with our immunofluorescence data showing that transfected NotchIC localizes to the nucleus, these data support a working model in which NotchIC interacts with the p50 subunit of NF-κB in the nucleus through the NBD and blocks the binding of p50/p65 heterodimer to NF-κB sites, thereby interfering with NF-κB-induced transcriptional regulation.
The N-terminal portion of NotchIC contains the high affinity binding site (RAM) required for NotchIC interaction with CBF-1-RBP-Jκ, the putative major effector of Notch signaling (4, 5). Thus, CBF-1 and NF-κB interact with two partially overlapping domains of Notch-1. It is not clear whether these interactions are mutually exclusive. In this case the relative intracellular concentrations and the intracellular compartmentalization of Notch-1, CBF-1, and NF-κB may dictate which complexes are predominantly formed within the nucleus.
The intracellular subunit of Notch receptors appears to contain several functional modules that interact with a variety of intracellular targets (Fig. 7⇓). The ankyrin repeats are involved in binding the C. elegans chromatin-remodeling factor EMB-5 (38) and the cytoplasmic Notch effector Deltex (54, 55), which, in turn, regulate the transcription factor E47 (56). Additionally, activated Notch-1 inhibits the myogenic activity of the myocyte enhancer factor 2C by direct interaction of myocyte enhancer factor 2C with the ankyrin repeat region of Notch (57). Kurooka et al. (5) reported recently that the ankyrin repeats together with C-terminal sequences are required for the interaction with histone acetyltransferases PCAF and GCN5. The same region of Notch-1 appears to be sufficient for its transforming activity (27). Other Notch binding proteins include the orphan nuclear receptor Nur77 (33) and the c-Abl accessory protein Disabled (58). Notch binding proteins that are suggested to be mediators of Notch signaling, and their putative functions are summarized in Fig. 6⇑. It is clear that Notch resides at the center of a complex multipathway signaling network involved in decision processes that affect cell proliferation, differentiation, migration, and apoptosis in multiple cellular and organ systems.
The fact that Notch-1 controls NF-κB-regulated gene expression (35), while NF-κB regulates Notch signaling through inducing the expression of the Notch ligand Jagged-1 on the cell surface (59) suggests that the cross-talk between Notch and NF-κB is bidirectional. The possible biological roles of this cross-talk remain to be elucidated. In principle, this interaction could be used in vivo to amplify the differentiation of two initially equipotent cells into a ligand-expressing signaling cell and a receptor-expressing receiving cell (see Fig. 8⇓). This phenomenon is well known in Drosophila development (1). In this hypothetical model NF-κB activation in a given cell triggers Jagged-1 expression, thus leading this cell to activate Notch in surrounding cells. Jagged-1 activation is known to up-regulate Notch-1 expression in Drosophila (60) and mammalian cells. In turn, Notch-1 up-regulation would inhibit NF-κB in the receiving cells, thus down-regulating NF-κB-inducible gene expression, including Jagged-1 expression. Because individual cells often express both Notch and Delta (61, 62, 63), such a feedback could also be envisioned bidirectionally within the same subset of cells. Notch and Delta may also act in cis as a cell-autonomous mechanism for switching off NF-κB-mediated gene expression after activation.
NF-κB-activating stimuli include cytokines, mitogens, bacteria, and viruses. Recently, Sun et al. reported that protein kinase C-θ is required for TCR-induced activation of NF-κB in mature T lymphocytes (64). NF-κB-responsive genes include chemokines (RANTES, macrophage inflammatory protein-1α, and monocyte chemoattractant protein-1), cytokines (TNF-α, IL-2, and TGF-β2), growth factors (GM-CSF), immunoreceptors (MHC II, CD25, and TCR β-chain), adhesion molecules (VCAM-1 and endothelial-leukocyte adhesion molecule-1), oncogenes (c-myb and c-myc), apoptotic genes (Fas/CD95 and Fas ligand), and viral genes (HIV-1 and CMV) (40, 50, 65, 66, 67). Thus, Notch signaling through its effect on NF-κB could play a role in regulating the expression of these genes and viruses in vivo.
Like NF-κB, Notch receptors are expressed in a wide range of hemopoietic and immune system cells, including human CD34+ hemopoietic stem cells (68, 69, 70). Recent evidence suggests that Notch activation in CD4 T cells may induce peripheral tolerance (71) and cell death in monocytes (72). The possible roles of Notch/NF-κB cross-talk in the regulation of cell fate decisions during hemopoiesis, inflammation, immune responses, viral infection, and apoptosis warrant further investigation.
We thank Elaine Lizzio for technical help.
↵1 L.M. was supported in part by grants from the Illinois Department of Public Health and the National Institutes of Health (1R01CA84065-01).
↵2 Address correspondence and reprint requests to Drs. Jinhai Wang or Ennan Guan, Division of Therapeutic Proteins, Center for Biologics Evaluation and Research, Food and Drug Administration, National Institutes of Health, Building 29B, Room 4E12, HFM-541, 8800 Rockville Pike, Bethesda, MD 20892. E-mail addresses: [email protected] or
↵3 Current address: Cancer Immunology Program, Cardinal Bernardin Cancer Center, Loyola University Chicago, 2160 South First Avenue, Maywood, IL 60153.
4 Abbreviations used in this paper: NotchEC, extracellular domains of Notch proteins; NotchIC, intracellular regions of Notch receptors; NLS, nuclear localization signal; TAN-1, translocation-associated Notch homologue; NBD, NF-κB binding domain; CBF, C promoter binding factor.
- Received August 21, 2000.
- Accepted April 24, 2001.
- Copyright © 2001 by The American Association of Immunologists