Structural Requirements for Assembly of the CSL·Intracellular Notch1·Mastermind-like 1 Transcriptional Activation Complex

Ligand binding by Notch receptors triggers a series of proteolytic cleavages that liberate the intracellular portion of Notch (ICN) from the cell membrane, permitting it to translocate to the nucleus. Nuclear ICN binds to a highly conserved DNA-binding transcription factor called CSL (also known as RBP-Jκ, CBF1, Suppressor of Hairless, and Lag-1) and recruits Mastermind-like transcriptional co-activators to form a transcriptional activation complex. Using bioinformatics tools, we identified a Rel homology region (RHR) within CSL that was used as a guide to determine the minimal protein requirements for ternary complex formation. The RHR of CSL contains both the N- and C-terminal β-sheet domains (RHR-n and RHR-c) of typical Rel transcription factors, as judged by circular dichroism spectra. Binding of monomeric CSL to DNA requires the entire RHR of CSL and an additional 125-residue N-terminal sequence, whereas binding to ICN requires only the RHR-n domain. Although the RAM (RBP-Jκ (recombination-signal-sequence-binding protein for Jκ genes)-associated molecule) domain of ICN is flexible and relatively unstructured as an isolated polypeptide in solution, it associates stably with CSL on DNA. Recruitment of Mastermind-like 1 (MAML1) to CSL·ICN complexes on DNA requires inclusion of the ankyrin repeat domain of ICN, and N- and C-terminal sequences of CSL extending beyond the DNA-binding region. The requirement for cooperative assembly of the MAML1·ICN·CSL·DNA complex suggests that a primary function of ICN is to render CSL competent for MAML loading. On the basis of our results, we present a working structural model for the organization of the MAML1·ICN·CSL·DNA complex. Ligand binding by Notch receptors triggers a series of proteolytic cleavages that liberate the intracellular portion of Notch (ICN) from the cell membrane, permitting it to translocate to the nucleus. Nuclear ICN binds to a highly conserved DNA-binding transcription factor called CSL (also known as RBP-Jκ, CBF1, Suppressor of Hairless, and Lag-1) and recruits Mastermind-like transcriptional co-activators to form a transcriptional activation complex. Using bioinformatics tools, we identified a Rel homology region (RHR) within CSL that was used as a guide to determine the minimal protein requirements for ternary complex formation. The RHR of CSL contains both the N- and C-terminal β-sheet domains (RHR-n and RHR-c) of typical Rel transcription factors, as judged by circular dichroism spectra. Binding of monomeric CSL to DNA requires the entire RHR of CSL and an additional 125-residue N-terminal sequence, whereas binding to ICN requires only the RHR-n domain. Although the RAM (RBP-Jκ (recombination-signal-sequence-binding protein for Jκ genes)-associated molecule) domain of ICN is flexible and relatively unstructured as an isolated polypeptide in solution, it associates stably with CSL on DNA. Recruitment of Mastermind-like 1 (MAML1) to CSL·ICN complexes on DNA requires inclusion of the ankyrin repeat domain of ICN, and N- and C-terminal sequences of CSL extending beyond the DNA-binding region. The requirement for cooperative assembly of the MAML1·ICN·CSL·DNA complex suggests that a primary function of ICN is to render CSL competent for MAML loading. On the basis of our results, we present a working structural model for the organization of the MAML1·ICN·CSL·DNA complex. Notch proteins act as receptors in an evolutionarily conserved signaling pathway that controls differentiation and proliferation in response to ligands expressed on neighboring cells (reviewed in Refs. 1Artavanis-Tsakonas S. Rand M.D. Lake R.J. Science. 1999; 284: 770-776Crossref PubMed Scopus (4875) Google Scholar and 2Nam Y. Aster J.C. Blacklow S.C. Curr. Opin. Chem. Biol. 2002; 6: 501-509Crossref PubMed Scopus (62) Google Scholar). Ligand binding to the extracellular region of Notch receptors causes activation of a signal by triggering a series of proteolytic cleavages that release the intracellular portion of Notch (ICN) 1The abbreviations used are: ICN, intracellular portion of Notch; ANK, ankyrin; RAM, RBP-J-associating molecule; MAML, mammalian Mastermind-like; RHR, Rel homology region; RHR-n and -c, RHR N- and C-terminal β-sheet domains; IVT, in vitro transcription/translation; GST, glutathione S-transferase; TRX, thioredoxin; TEV, tobacco etch virus; DTT, dithiothreitol; Ni-NTA, nickel-nitrilotriacetic acid; EMSA, electrophoretic mobility shift assay; 3D-PSSM, three-dimensional position-specific scoring matrix; NFAT, nuclear factor of activated T cells. 1The abbreviations used are: ICN, intracellular portion of Notch; ANK, ankyrin; RAM, RBP-J-associating molecule; MAML, mammalian Mastermind-like; RHR, Rel homology region; RHR-n and -c, RHR N- and C-terminal β-sheet domains; IVT, in vitro transcription/translation; GST, glutathione S-transferase; TRX, thioredoxin; TEV, tobacco etch virus; DTT, dithiothreitol; Ni-NTA, nickel-nitrilotriacetic acid; EMSA, electrophoretic mobility shift assay; 3D-PSSM, three-dimensional position-specific scoring matrix; NFAT, nuclear factor of activated T cells. from the plasma membrane. ICN then migrates to the nucleus where it activates the transcription of target genes. ICNs are modular polypeptides comprised of an N-terminal RAM domain of ∼110 amino acids, a set of seven iterated ankyrin repeats (3Zweifel M.E. Barrick D. Biochemistry. 2001; 40: 14344-14356Crossref PubMed Scopus (54) Google Scholar, 4Zweifel M.E. Barrick D. Biochemistry. 2001; 40: 14357-14367Crossref PubMed Scopus (88) Google Scholar), and less conserved C-terminal regions that include a transcriptional activation domain and a far C-terminal PEST sequence. The ankyrin repeat (ANK) domain is the most highly conserved region of ICN and is essential for all known Notch functions (1Artavanis-Tsakonas S. Rand M.D. Lake R.J. Science. 1999; 284: 770-776Crossref PubMed Scopus (4875) Google Scholar). The primary target of ICN in the nucleus is the highly conserved DNA-binding transcription factor, CSL (also known as RBP-Jκ, CBF1, Suppressor of Hairless (Su)H), and Lag-1) (5Fortini M.E. Artavanis-Tsakonas S. Cell. 1994; 79: 273-282Abstract Full Text PDF PubMed Scopus (459) Google Scholar, 6Jarriault S. Brou C. Logeat F. Schroeter E.H. Kopan R. Israel A. Nature. 1995; 377: 355-358Crossref PubMed Scopus (1214) Google Scholar, 7Christensen S. Kodoyianni V. Bosenberg M. Friedman L. Kimble J. Development. 1996; 122: 1373-1383PubMed Google Scholar). Previous analysis of CSL proteins using primary sequence alignment tools has failed to detect similarity to other known transcription factors. In the absence of ICN, CSL represses transcription by virtue of interactions with a number of co-repressors, including SMRT/N-CoR (silencing mediator for retinoic acid and thyroid hormone receptors/nuclear repressor co-repressor), CIR (CBF1-interacting co-repressor), and KyoT2 (8Hsieh J.J. Zhou S. Chen L. Young D.B. Hayward S.D. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 23-28Crossref PubMed Scopus (251) Google Scholar, 9Kao H.Y. Ordentlich P. Koyano-Nakagawa N. Tang Z. Downes M. Kintner C.R. Evans R.M. Kadesch T. Genes Dev. 1998; 12: 2269-2277Crossref PubMed Scopus (487) Google Scholar, 10Taniguchi Y. Furukawa T. Tun T. Han H. Honjo T. Mol. Cell. Biol. 1998; 18: 644-654Crossref PubMed Scopus (160) Google Scholar). Binding of ICNs not only displaces co-repressors from CSL proteins but also recruits transcriptional co-activators to CSL·ICN complexes on DNA. Several studies indicate that Mastermind-like polypeptides such as Lag-3 in worms (11Petcherski A.G. Kimble J. Nature. 2000; 405: 364-368Crossref PubMed Scopus (156) Google Scholar) and mammalian Mastermind-like-1 (MAML1) are core components of a higher order transcriptional activation complex on DNA (12Wu L. Aster J.C. Blacklow S.C. Lake R. Artavanis-Tsakonas S. Griffin J.D. Nat. Genet. 2000; 26: 484-489Crossref PubMed Scopus (461) Google Scholar). The existence of a CSL·ICN·Mastermind signaling complex was foreshadowed by genetic screens in the fly revealing that mutations in each of these genes cause similar “neurogenic” loss-of-function phenotypes (13Schweisguth F. Posakony J.W. Cell. 1992; 69: 1199-1212Abstract Full Text PDF PubMed Scopus (229) Google Scholar, 14Yedvobnick B. Smoller D. Young P. Mills D. Genetics. 1988; 118: 483-497Crossref PubMed Google Scholar, 15Artavanis-Tsakonas S. Muskavitch M.A. Yedvobnick B. Proc. Natl. Acad. Sci. U. S. A. 1983; 80: 1977-1981Crossref PubMed Scopus (143) Google Scholar). Although the level of ICN in the nucleus of most cells undergoing physiologic Notch signaling is too low to be detected readily, certain T cell leukemias associated with chromosomal translocations involving the human NOTCH1 gene express nuclear ICN1-like polypeptides at high levels (16Aster J. Pear W. Hasserjian R. Erba H. Davi F. Luo B. Scott M. Baltimore D. Sklar J. Cold Spring Harbor Symp. Quant. Biol. 1994; 59: 125-136Crossref PubMed Scopus (82) Google Scholar). Capobianco's group (17Jeffries S. Robbins D.J. Capobianco A.J. Mol. Cell. Biol. 2002; 22: 3927-3941Crossref PubMed Scopus (130) Google Scholar) exploited a cell line derived from a NOTCH1 leukemia, SUPT-1, to demonstrate the of high nuclear complexes and also that the of of on cells that of of of Notch Y. Pear Griffin J.D. Blacklow S.C. Aster J.C. Mol. Cell. Biol. PubMed Scopus Google Scholar). studies and genetic and that is an essential within a protein assembly for the transcription of Notch target genes. and other of the MAML are polypeptides of amino acid to be of low structural that function as for of additional co-activators the transcription The N-terminal region of and polypeptides contains a sequence that associates with ICN and CSL (12Wu L. Aster J.C. Blacklow S.C. Lake R. Artavanis-Tsakonas S. Griffin J.D. Nat. Genet. 2000; 26: 484-489Crossref PubMed Scopus (461) Google Scholar, A.G. Kimble J. Curr. Biol. 2000; Full Text Full Text PDF PubMed Scopus Google Scholar). expressed as a the N-terminal region of is a of Notch that C-terminal sequences are also for function (12Wu L. Aster J.C. Blacklow S.C. Lake R. Artavanis-Tsakonas S. Griffin J.D. Nat. Genet. 2000; 26: 484-489Crossref PubMed Scopus (461) Google Scholar). studies indicate these C-terminal sequences the transcriptional as as other that are for activation of transcription from in vitro Kintner C. Genes Dev. 2002; PubMed Scopus Google Scholar, U. Mol. Cell. Biol. 2002; 22: PubMed Scopus Google Scholar). present is the of CSL and higher order complexes with ICN and using bioinformatics tools, we the domains of a Rel homology region (RHR) within that CSL is a of the Rel homology of transcription factors. these domains as a we the regions of CSL for the of and for assembly of ternary complexes on that both ICN and CSL be present to studies the of CSL as a of the Rel homology of transcription factors. The requirement for cooperative assembly of the ternary complex on with the CSL suggests that a primary function of is to render CSL competent for loading. DNA CSL are derived from of human CSL used to polypeptides using in vitro by and by using for to the RHR RHR-n and in the a tobacco etch at the and the CSL sequence. the polypeptides and (ANK) of human in of and sequences in a of the using the polypeptide was a similar sequence and a in a of human was of a and for (12Wu L. Aster J.C. Blacklow S.C. Lake R. Artavanis-Tsakonas S. Griffin J.D. Nat. Genet. 2000; 26: 484-489Crossref PubMed Scopus (461) Google Scholar), and CSL J.C. Hasserjian Sklar J. J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar) protein was for in by the of by and by in 1 1 and proteins to with and and with of at for was by on using a of and 1 In a RHR and to by on a in DTT, and 1 polypeptides expressed and was from and by to was to and in proteins then by on using a of and 1 and by on a in DTT, and 1 of was as for CSL and ICN was from inclusion by in and by on a the protein was in and and with polypeptide was by using a of in and 1 was acid and was in with and sequence using for all protein sequences was using the B. C. J. Mol. Biol. PubMed Scopus Google Scholar, B. C. 1994; PubMed Scopus Google Scholar). The structural homology of CSL proteins CSL and Suppressor of to Rel proteins of known three-dimensional was detected using the D. C. A. A. D. R.M. B. L. M. 1999; PubMed Scopus Google Scholar, R.M. J. Mol. Biol. 2000; PubMed Scopus Google Scholar). dichroism on an with a to H. Biochemistry. 6: PubMed Scopus Google Scholar). of CSL RAM and in and CSL RHR was in the with of CSL RHR, RAM, ANK, and at in a as the of using a The CSL was with an of of each polypeptide in a in the The signal was at using a a signal and a of 1 from to The of and The for was by the of the of the signal with to on a using an and determine the of and of and of and in and DTT, the and 1 and by with was also used to the at and by with a by to a model using and of for and for A.J. J.C. in and Science. of Scholar). mobility shift using a as (12Wu L. Aster J.C. Blacklow S.C. Lake R. Artavanis-Tsakonas S. Griffin J.D. Nat. Genet. 2000; 26: 484-489Crossref PubMed Scopus (461) Google Scholar). Binding at for 1 with with DTT, and with proteins with binding for 1 at by in the proteins by for in in proteins by of the Rel of of human CSL a protein of amino of CSL include DNA binding T. Y. N. Furukawa T. Honjo T. M. 1994; 22: PubMed Scopus Google Scholar), with ICN polypeptides (5Fortini M.E. Artavanis-Tsakonas S. Cell. 1994; 79: 273-282Abstract Full Text PDF PubMed Scopus (459) Google Scholar), and of higher order complexes with MAML polypeptides (11Petcherski A.G. Kimble J. Nature. 2000; 405: 364-368Crossref PubMed Scopus (156) Google Scholar, L. Aster J.C. Blacklow S.C. Lake R. Artavanis-Tsakonas S. Griffin J.D. Nat. Genet. 2000; 26: 484-489Crossref PubMed Scopus (461) Google Scholar). the structural basis for these we used a bioinformatics to guide the of CSL primary sequence homology CSL proteins and other proteins in the was not we for of homology to protein domains of known using the sequences with the protein using sequence and D. C. A. A. D. R.M. B. L. M. 1999; PubMed Scopus Google Scholar, R.M. J. Mol. Biol. 2000; PubMed Scopus Google Scholar). the RHR of as the for human CSL is the of also an RHR the of with a level of sequences other proteins as such as of and that CSL proteins are to the Rel homology of transcription and The RHR, a conserved region proteins of the Rel S. 1998; PubMed Scopus Google Scholar), of of M. S.C. Nature. 1995; PubMed Scopus Google Scholar, S. Nature. 1995; PubMed Scopus Google Scholar) RHR-n and (also known as a transcription factor by to to and of The are also identified in both CSL proteins by of and but the RHR-n are the of RHR-n domains are highly the a alignment of the CSL RHR-n domain is not the of the domain for the of an RHR-n domain. the RHR region of CSL contains domains in the structural are we the of the RHR and domains by that the RHR region of as as the isolated contains a of and the the are with the of a RHR domain within CSL The RHR domain with additional to guide of the RHR domain on the basis of primary sequence homology and CSL and a protein that binds the DNA sequence but not ICNs S. Y. H. T. U. Honjo T. Mol. Cell. Biol. PubMed Google Scholar), at C-terminal to the RHR of the of a of by B. C. J. Mol. Biol. PubMed Scopus Google Scholar, B. C. 1994; PubMed Scopus Google Scholar) to The C-terminal of a low sequence with for and within regions by to by the results, we CSL with at and and C-terminal at and CSL to DNA as a Using these we the requirements for binding of CSL to DNA using an of the of CSL not with DNA but with at to DNA that the RHR region is C-terminal of CSL to not with DNA but to the DNA CSL the RHR and an additional N-terminal of amino acids, are both and to and of RAM, ANK, and of RAM, ANK, and of RAM and at in physiologic of the RAM region was from the the and spectra. of the signal at for and in physiologic and of and at in physiologic of the of protein as a function of the of the at (ANK) The line a of (ANK) to the is other Rel homology transcription to to DNA as M. S.C. Nature. 1995; PubMed Scopus Google Scholar, S. Nature. 1995; PubMed Scopus Google Scholar, J.C. C. A. Chen L. Nat. Biol. 2002; PubMed Scopus Google Scholar), we the of in complexes by in the of a of CSL protein of and a C-terminal form of CSL binds and the complexes in the both CSL are to an electrophoretic mobility shift CSL and are not that CSL binds DNA as a as for CSL using nuclear Y. Honjo T. M. 1994; 22: PubMed Scopus Google Scholar). of RAM, ANK, and tools, and with F. M. C. P. U. J. 2001; PubMed Scopus Google Scholar), the of the domain of in the that of human a series of seven ankyrin repeats a by studies of the domain of Notch (3Zweifel M.E. Barrick D. Biochemistry. 2001; 40: 14344-14356Crossref PubMed Scopus (54) Google Scholar, 4Zweifel M.E. Barrick D. Biochemistry. 2001; 40: 14357-14367Crossref PubMed Scopus (88) Google Scholar). The RAM has sequence homology to protein of known we highly RAM, ANK, and for studies to the and of the ankyrin repeat domain of human and to determine the RAM region contains a in in judged by circular the polypeptide contains an within the for an ankyrin repeat domain a with a of that not at of to at and is monomeric at the used in the In the RAM region is as judged by inclusion of the RAM region in the of not a of in the RAM the the signal from the RAM region in the of the of The for of is the as for of ANK, that the of RAM not the of the RAM both as an isolated polypeptide and in the of the a flexible polypeptide is monomeric and not at to at as judged by The RHR-n of CSL with we the region of CSL for binding of Previous mammalian of CSL with ICN J.J. T. P. Hayward S.D. Mol. Cell. Biol. 1996; PubMed Scopus Google Scholar), suggests that is the RHR-n domain expressed in associates with that binding is we from and on glutathione with RAM, and both and RAM, that these complexes are in the absence of additional protein DNA In not form a complex with the of the isolated RHR-n domain for is too to detect in the absence of other only for the regions of ICN for assembly of ternary complexes on DNA using an electrophoretic mobility shift In these we CSL and a human polypeptide and that the minimal region to form complexes with ICN and CSL Y. Pear Griffin J.D. Blacklow S.C. Aster J.C. Mol. Cell. Biol. PubMed Scopus Google Scholar). and RAM form complexes with CSL on DNA and but not Although the not form complexes with CSL and DNA the of the to the complex of with CSL and DNA causes an additional of a ternary complex also a complex with and DNA a with from other (17Jeffries S. Robbins D.J. Capobianco A.J. Mol. Cell. Biol. 2002; 22: 3927-3941Crossref PubMed Scopus (130) Google Scholar, Kintner C. Genes Dev. 2002; PubMed Scopus Google Scholar). is for of to the as a is not of not to the requirement of for of and the of CSL and ICN we binds stably to in the absence of other factors. not that of the polypeptide a complex requires the of both ICN and then identified the regions of CSL to binding of in the of and DNA. Although binds DNA and complexes with it to ternary complexes with studies that sequences both N- and C-terminal to the DNA-binding region of CSL are complex is the N- and C-terminal are ICN, and with and with DNA is to Notch signaling activates The primary of these studies was to the regions of ICN, and for of complexes with and with DNA. also has for the associated with of of ICN in complexes of ICN with CSL and in the nucleus. for the ICN The of and bioinformatics used in these studies a model for the of the transcriptional activation complex The primary of the model are the CSL transcription factor is to proteins in the Rel homology of transcription and ICN and CSL to the The that CSL proteins are to the Rel homology of transcription from analysis of the CSL and amino acid sequences using the The Rel homology of DNA-binding proteins a of transcription including the and proteins such as H. Biochemistry. 6: PubMed Scopus Google Scholar). Rel homology regions in these transcription amino and of domains (RHR-n and RHR-c) by a sequence. On the basis of the the RHR-n domain of human CSL and the domain CSL the RHR by the that CSL contains both RHR-n and structural the RHR as as the isolated contains a of and our studies that a only the isolated RHR-n domain the high binding for In the C-terminal of the region of CSL for high binding of DNA with the C-terminal of the domain. studies using CSL are with that the DNA-binding of CSL is a Y. Honjo T. M. 1994; 22: PubMed Scopus Google Scholar). Binding of DNA by CSL requires a region of the protein the entire RHR domain and an additional the requirement that the entire RHR in DNA binding is by the other Rel homology transcription factors, a of structural in DNA binding In high of complexes RHR transcription and both the RHR-n and domains in the DNA binding Although the M. S.C. Nature. 1995; PubMed Scopus Google Scholar, S. Nature. 1995; PubMed Scopus Google Scholar) and J.C. C. A. Chen L. Nat. Biol. 2002; PubMed Scopus Google Scholar) proteins DNA as with the primary by the the RHR domain of binds to DNA as a in with the high binding to DNA L. A. S.C. Nature. 1998; PubMed Scopus Google Scholar). additional with the the N-terminal region of CSL the for and for additional and to high DNA binding DNA as a CSL most the also to DNA as with the to form high The RHR-n of CSL to ICN is the binding of ICN to complexes is to be the for activation of transcription of genes. Previous studies the existence of a CSL binding within the RAM and a binding in the domain H. Y. H. S. T. S. Honjo T. Development. PubMed Google Scholar). RAM, ANK, and for these polypeptides using such analysis of these domains from ICN proteins has studies and studies of the repeat region of Notch (3Zweifel M.E. Barrick D. Biochemistry. 2001; 40: 14344-14356Crossref PubMed Scopus (54) Google Scholar, 4Zweifel M.E. Barrick D. Biochemistry. 2001; 40: 14357-14367Crossref PubMed Scopus (88) Google Scholar). dichroism demonstrate that from human a similar of and with the ankyrin repeat region of by that the RAM both as an isolated polypeptide and in the of the is a flexible polypeptide both and monomeric at that is to a in physiologic Notch using RAM, ANK, and from ICN in demonstrate binding of the RHR-n domain of CSL to RAM and of RAM, and by also that the of for high binding of CSL within the RAM region. these the that the RAM region of and the RHR-n domain of CSL binding and RAM as an isolated polypeptide in the of The of the RAM domain has as it is not for transcriptional activation in J.C. Hasserjian Sklar J. J. Biol. Chem. Full Text Full Text PDF PubMed Scopus Google Scholar), assembly of ICN complexes with CSL in the of in cells (17Jeffries S. Robbins D.J. Capobianco A.J. Mol. Cell. Biol. 2002; 22: 3927-3941Crossref PubMed Scopus (130) Google Scholar), J.C. L. V. J.C. Pear Mol. Cell. Biol. 2000; PubMed Scopus Google Scholar). of these on high level of ICN, the CSL binding within for assembly of transcriptional activation In of high of complexes in the of and in indicate that the RAM region is not for of complexes in cells with levels of ICN (17Jeffries S. Robbins D.J. Capobianco A.J. Mol. Cell. Biol. 2002; 22: 3927-3941Crossref PubMed Scopus (130) Google Scholar). levels of ICNs Notch activation are the of these the high RAM and RHR-n on The complex and CSL both and with the complex and the only of a complex and RHR proteins for high structural is T. D.B. S. Cell. 1998; Full Text Full Text PDF PubMed Scopus Google Scholar, D.B. Chen Nature. 1998; PubMed Scopus Google Scholar). In the the the ankyrin repeats and both domains of the CSL binds to DNA as a it not present as an for binding to the domain of the domain of ICN and the RHR region of CSL to binding only of the in the complex be present and of and ICN the binding of RAM for the RHR-n domain as an to complexes of ICN with Recruitment of and the N- and of the that of with the complex of the protein as a transcriptional In a with CSL only in the of ICN A.G. Kimble J. Curr. Biol. 2000; Full Text Full Text PDF PubMed Scopus Google Scholar). In with CSL from cells only ICN is also present (17Jeffries S. Robbins D.J. Capobianco A.J. Mol. Cell. Biol. 2002; 22: 3927-3941Crossref PubMed Scopus (130) Google Scholar). that the domain of recruits to the complex and that high of the domain for not to function as a that binds to with high we not detect and using highly that not with CSL ICN is present in our suggests that both and CSL to a high binding for for a model in in to ICN, is in to complexes on DNA from our by that CSL proteins at the to form complexes with and DNA. The of to also with to form complexes in cells with proteins not for CSL and act to are the N- and C-terminal regions of CSL are to only in the of ANK, both and the CSL to that the of the CSL N- and is to a structural of high and is less the domain is as an isolated domain in a high of the complex the of the protein components in the transcriptional activation complex. both CSL and to an to activation of Notch target genes and to that genes are Notch to CSL in the absence of activated target genes of CSL to be by co-repressors to and not form a complex of ICN with DNA-binding proteins other CSL not cause of gene suggests that not only as an for transcription of Notch target genes but also the complex for Kintner C. Genes Dev. 2002; PubMed Scopus Google Scholar). the of Notch activation with to also a for of the and of by activated for