- Email: [email protected]
Transcription: Activation by cooperating conformations
R616
Dispatch
Transcription: Activation by cooperating conformations Andrew Travers
The cooperative formation of a higher-order complex between transcription factors NFAT and Fos–Jun is accompanied by conformational changes in both DNA and protein. This allows formation of an extended interface between the proteins, while conserving recognition of the core DNA binding sequences. Address: MRC Laboratory of Molecular Biology, Hills Road, Cambridge, CB2 2QH, UK. Current Biology 1998, 8:R616–R618 http://biomednet.com/elecref/09609822008R0616 © Current Biology Publications ISSN 0960-9822
The cooperative assembly of complex nucleoprotein structures is a crucial element in the integration of environmental signals at the level of transcription. Characteristically, closely spaced DNA response elements have a low affinity for the independent binding of target proteins, but cooperative binding can increase these affinities by at least an order of magnitude. In this context, the overall architecture of the assembly must be such as to facilitate mutual interactions between all the participating molecules. When the first structures of multiprotein transcriptional complexes were solved, most notably those involving the TATA-binding protein [1,2] or the yeast MCM1 protein [3], compaction of the complex was seen to involve DNA bending. The role, if any, of protein flexibility in the formation of higher-order complexes has until now been less apparent. A well-studied example of the cooperative induction of transcription is the activation of the interleukin-2 (IL-2)
gene in T cells, which occurs on the presentation of cognate antigen to the T-cell antigen receptor. This presentation results in the activation and translocation to the nucleus of one of the ‘nuclear factors of T cells’ (NFAT). However, activation of transcription of the IL-2 gene requires not only activated NFAT, but also activated Fos–Jun heterodimer. Adjacent binding sites for NFAT and Fos–Jun in the IL-2 gene’s upstream regulatory region are both of low affinity, high occupancy occurring only in the presence of both activated components; in vitro, the DNA-binding domains of the two transcription factors are sufficient for cooperative binding to DNA [4]. These proteins have different DNAbinding domains: that of the Fos–Jun heterodimer is a typical basic–leucine zipper (bZIP) domain, with one end of a continuous α helix forming a coiled-coil with its partner, and the other making sequence-specific contacts in the major groove of DNA; that of NFAT is distantly homologous to the DNA-binding regions of Rel family transcription factors, such as NF-κB, and is consequently known as the ‘Rel homology region’ (RHR). The nature of the interactions between NFAT and the Fos–Jun heterodimer, and the conformational changes on complex formation, have recently been illuminated by the determination of two structures — the solution structure of the RHR of NFAT binding to its cognate site [5], and the crystal structure of a quaternary complex containing the RHR of NFAT and the bZIP moiety of Fos–Jun bound to a composite DNA site [6]. Although it has been apparent for some time that the recognition helices of the bZIP motif are ordered only in the presence of DNA [7],
Figure 1 Changes in the structure of the NFAT DNAbinding domain on interaction with the Fos–Jun heterodimer on DNA. Overlay of the Cα traces from the solution structure of the complex between the DNA-binding domain of NFAT and DNA (blue) with the corresponding portion of the crystal structure (yellow). Two views are shown. (Reproduced with permission from [5].)
Dispatch
Zhou et al. [5] observed that DNA binding by the NFAT RHR also resulted in the ordering of two loops, one of which makes contacts with both DNA and the Fos–Jun bZIP motif in the quaternary complex. The structure of the Fos–Jun component of the quaternary complex differs from that of Fos–Jun bound alone to DNA in two important respects. The first major difference is that, in the ternary complex of Fos–Jun and DNA, the DNA is bent by at most 10° towards the major groove facing the coiled-coil [8], a value largely consistent with measurements of the bend angle by cyclisation kinetics [9]. In the quaternary complex, however, this bend is increased to ~20° in the same direction. One consequence of this distortion is, not to narrow the major groove, but instead to displace the sugar-phosphate backbone so that the adjacent minor groove abutting the NFAT binding site is narrowed. The second major difference between the two complexes lies in the Fos–Jun segments themselves. The coiled-coil interactions are maintained, but the Fos segment flexes at the fork by ~15°. This flexibility between the coiled-coil and the basic regions, which was already apparent from crystallographically distinct structures of the ternary Fos–Jun–DNA complex [8], maintains the local major groove interactions between Fos and DNA, and at the same time shifts the axis of the coiled coil so that it leans away from the perpendicular to the DNA double-helical axis and towards NFAT. The concerted changes in structures of the DNA and protein thus conserve local contacts but alter the overall geometry of the complex. Comparison of the solution structure of the NFAT–DNA complex with the corresponding elements of the quaternary complex shows that the protein structure is again remodelled, but in this case the change is also accompanied by an alteration in the local DNA–protein contacts. In the quaternary complex, the DNA-binding domain of NFAT is reoriented relative to the DNA by a 20–24° rotation towards the Fos–Jun coiled-coil along the long axis of the DNA, and a 10–13° pivot perpendicular to it [5] (Figure 1). Concomitantly, an arginine residue in one of the loops stabilised on DNA shifts from making a bidentate contact with bases T7 and T6′ in the minor groove of the NFAT recognition site to contacting the carbonyl group of base T8, closer to the Fos–Jun contacts. There are also two contacts between the NFAT RHR and DNA that appear unique to the binary complex, yet recognition of the core NFAT site, GGAAA, is conserved between the two complexes. The cooperativity between NFAT and Jun–Fos is presumably driven by formation of the extensive contact surface between the two partners, involving primarily polar interactions. This interaction results in the formation
R617
Figure 2
Fos
Jun
NFAT (RHR-N)
NFAT (RHR-C)
DNA
1
5
10
15
5' T T G G A A A A T T T G T T T C A T A G 3' 3' C C T T T T A A A C A A A G T A T C A A 5' 1'
5'
10'
15'
Molscript representation of the crystal structure of the NFAT–Fos–Jun–DNA complex. The sequence of the DNA is shown at the bottom. (Reproduced with permission from [6].)
of a composite continuous DNA-binding groove for the recognition of 15 base pairs (Figure 2), with the induced bend creating the appropriate alignment of the respective recognition sequences. Although the structural changes in the bound DNA on formation of the quaternary complex are likely to be unfavourable, any energetic penalty for the extra distortion would be small and more than compensated for by the additional protein–protein contacts. Although solving the structure of the quaternary NFAT–Fos–Jun–DNA complex was a formidable crystallographic challenge, the complex is relatively simple compared to the multiprotein complexes such as the ‘enhanceosomes’ that bind to enhancer elements [10]. To build such complex structures, additional inter-protein contact surfaces need to be maintained or established. Conformational remodelling could act in such a way or, as suggested by Zhou et al. [5], the flexibility of the individual constituents could free higher-order complex formation from the strictures of rigid geometrical constraints. A limited flexibility might, however, also confer additional selectivity in partner selection. Cooperative binding should normally fulfill at least two requirements — that the recognition of the core DNA binding sequence is maintained, and that a protein–protein interface is established. The core sequences, which can often be recognised by several members of a transcription factor family,
R618
Current Biology, Vol 8 No 17
establish a spatial constraint which may allow interface formation only between particular partners, and preclude interface formation between others, depending on the permitted flexibility of the proteins involved. There are strong parallels between the structure of the quaternary NFAT–Fos–Jun–DNA complex and that of the ternary complex containing the yeast MATα2 homeodomain protein and the MADS box transcription factor MCM1 bound to DNA [3]. In the latter complex, DNA bending induced by MCM1 brings the two proteins into close proximity, facilitating cooperative binding. Contact between the two proteins also requires a structural reordering, in which the otherwise flexible amino-terminal extension of the MATα2 homeodomain forms a β-hairpin that grips the MCM1 surface by making parallel β-strand hydrogen bonds coupled with hydrophobic packing. Additional conformational flexibility in this case comes from the ability of a sequence of eight amino acids in this aminoterminal extension to exist as either a β strand or an α helix. The more general picture is one in which the assembly of a higher-order protein–DNA structure from its individual component complexes does not follow the principles of Legoland. Structural remodelling of one or more of the participants may be the rule, rather than the exception. Other examples include the distortion of nucleosomal DNA that occurs on binding of linker histone, generating a structurally asymmetric particle, and the twisting of one of the two DNA-binding domains of the TATA-boxbinding protein relative to the other on interaction with DNA and TFIIB [1,2]. References 1. Nikolov DB, Chen H, Halay ED, Usheva AA, Hisatake K, Roeder RG, Burley SK: Crystal structure of a TFIIB-TBP-TATA-element ternary complex. Nature 1995, 377:119-128. 2. Tan S, Hunsiker Y, Sargent DF, Richmond TJ: Crystal structure of a yeast TBP/TFIIA/DNA complex. Nature 1996, 381:127-134. 3. Tan S, Richmond TJ : Crystal structure of the yeast MATa2/MCM1/DNA ternary complex. Nature 1998, 391:660-666. 4. Jain J, McCaffrey PG, Miner Z, Kerppola TK, Lambert JN, Verdine GL, Curran T, Rao A: The T-cell transcription factor NFATp is a substrate for calcineurin and interacts with Fos and Jun. Nature 1993, 365:352-355. 5. Zhou, P, Sun LJ, Dötsch V, Wagner G, Verdine GL: Solution structure of the core NFATC1/DNA complex. Cell 1998, 92:687-696. 6. Chen L, Glover JNM, Hogan PG, Rao A, Harrison SC: Structure of the DNA-binding domains from NFAT, Fos and Jun bound specifically to DNA. Nature 1998, 392:42-48. 7. Ellenberger TE, Brandl CJ, Struhl K, Harrison SC: The GCN5 basic region leucine zipper binds DNA as a dimer of uninterrupted ahelices: crystal structure of the protein-DNA complex. Cell 1992, 71:1223-1237. 8. Glover JN, Harrison SC: Crystal structure of the heterodimeric transcription factor c-Fos-c-Jun bound to DNA. Nature 1995, 373:257-261. 9. Sitlani A, Crothers, DM: DNA-binding domains of Fos and Jun do not induce DNA curvature: an investigation with solution and gel methods. Proc Natl Acad Sci USA 1998, 95:1404-1409. 10. Kim TK, Maniatis T: The mechanism of transcriptional synergy of an in vitro assembled interferon-b enhanceosome. Mol Cell 1998, 1:119-129.
If you found this dispatch interesting, you might also want to read the April 1998 issue of
Current Opinion in Genetics & Development which included the following reviews, edited by C David Allis and Susan M Gasser, on Chromosomes and expression mechanisms: The histone tails of the nucleosome Karolin Luger and Timothy J Richmond Unfolding the mysteries of heterochromatin Lori L Wallrath Imprinting mechanisms in mammals Wolf Reik and Jörn Walter Transcription of chromatin: these are complex times Jennifer A Armstrong and Beverly M Emerson Covalent modifications of histones: expression from chromatin templates James R Davie Dosage compensation in flies and worms: the ups and downs of X-chromosome regulation John C Lucchesi DNA damage checkpoints update: getting molecular Ted Weinert Replication origins in yeast versus metazoa: separation of the haves and have nots David M Gilbert Recombination at work for meiosis Kathleen N Smith and Alain Nicolas Centromeres: proteins, protein complexes, and repeated domains at centromeres of simple eukaryotes Louise Clarke Centromeres: the missing link in the development of human artificial chromosomes Huntington F Willard Telomerase and chromosome end maintenance Joachim Lingner and Thomas R Cech Mechanisms of silencing in Saccharomyces cerevisiae Arthur J Lustig Chromosomal imprinting in plants Robert Martienssen Fragile-X syndrome and myotonic dystrophy: parallels and paradoxes Stephen J Tapscott, Todd R Klesert, RJ Widrow, Reinhard Stöger and Charles D Laird The full text of Current Opinion in Genetics & Development is in the BioMedNet library at http://BioMedNet.com/cbiology/gen
Dispatch
Transcription: Activation by cooperating conformations Andrew Travers
The cooperative formation of a higher-order complex between transcription factors NFAT and Fos–Jun is accompanied by conformational changes in both DNA and protein. This allows formation of an extended interface between the proteins, while conserving recognition of the core DNA binding sequences. Address: MRC Laboratory of Molecular Biology, Hills Road, Cambridge, CB2 2QH, UK. Current Biology 1998, 8:R616–R618 http://biomednet.com/elecref/09609822008R0616 © Current Biology Publications ISSN 0960-9822
The cooperative assembly of complex nucleoprotein structures is a crucial element in the integration of environmental signals at the level of transcription. Characteristically, closely spaced DNA response elements have a low affinity for the independent binding of target proteins, but cooperative binding can increase these affinities by at least an order of magnitude. In this context, the overall architecture of the assembly must be such as to facilitate mutual interactions between all the participating molecules. When the first structures of multiprotein transcriptional complexes were solved, most notably those involving the TATA-binding protein [1,2] or the yeast MCM1 protein [3], compaction of the complex was seen to involve DNA bending. The role, if any, of protein flexibility in the formation of higher-order complexes has until now been less apparent. A well-studied example of the cooperative induction of transcription is the activation of the interleukin-2 (IL-2)
gene in T cells, which occurs on the presentation of cognate antigen to the T-cell antigen receptor. This presentation results in the activation and translocation to the nucleus of one of the ‘nuclear factors of T cells’ (NFAT). However, activation of transcription of the IL-2 gene requires not only activated NFAT, but also activated Fos–Jun heterodimer. Adjacent binding sites for NFAT and Fos–Jun in the IL-2 gene’s upstream regulatory region are both of low affinity, high occupancy occurring only in the presence of both activated components; in vitro, the DNA-binding domains of the two transcription factors are sufficient for cooperative binding to DNA [4]. These proteins have different DNAbinding domains: that of the Fos–Jun heterodimer is a typical basic–leucine zipper (bZIP) domain, with one end of a continuous α helix forming a coiled-coil with its partner, and the other making sequence-specific contacts in the major groove of DNA; that of NFAT is distantly homologous to the DNA-binding regions of Rel family transcription factors, such as NF-κB, and is consequently known as the ‘Rel homology region’ (RHR). The nature of the interactions between NFAT and the Fos–Jun heterodimer, and the conformational changes on complex formation, have recently been illuminated by the determination of two structures — the solution structure of the RHR of NFAT binding to its cognate site [5], and the crystal structure of a quaternary complex containing the RHR of NFAT and the bZIP moiety of Fos–Jun bound to a composite DNA site [6]. Although it has been apparent for some time that the recognition helices of the bZIP motif are ordered only in the presence of DNA [7],
Figure 1 Changes in the structure of the NFAT DNAbinding domain on interaction with the Fos–Jun heterodimer on DNA. Overlay of the Cα traces from the solution structure of the complex between the DNA-binding domain of NFAT and DNA (blue) with the corresponding portion of the crystal structure (yellow). Two views are shown. (Reproduced with permission from [5].)
Dispatch
Zhou et al. [5] observed that DNA binding by the NFAT RHR also resulted in the ordering of two loops, one of which makes contacts with both DNA and the Fos–Jun bZIP motif in the quaternary complex. The structure of the Fos–Jun component of the quaternary complex differs from that of Fos–Jun bound alone to DNA in two important respects. The first major difference is that, in the ternary complex of Fos–Jun and DNA, the DNA is bent by at most 10° towards the major groove facing the coiled-coil [8], a value largely consistent with measurements of the bend angle by cyclisation kinetics [9]. In the quaternary complex, however, this bend is increased to ~20° in the same direction. One consequence of this distortion is, not to narrow the major groove, but instead to displace the sugar-phosphate backbone so that the adjacent minor groove abutting the NFAT binding site is narrowed. The second major difference between the two complexes lies in the Fos–Jun segments themselves. The coiled-coil interactions are maintained, but the Fos segment flexes at the fork by ~15°. This flexibility between the coiled-coil and the basic regions, which was already apparent from crystallographically distinct structures of the ternary Fos–Jun–DNA complex [8], maintains the local major groove interactions between Fos and DNA, and at the same time shifts the axis of the coiled coil so that it leans away from the perpendicular to the DNA double-helical axis and towards NFAT. The concerted changes in structures of the DNA and protein thus conserve local contacts but alter the overall geometry of the complex. Comparison of the solution structure of the NFAT–DNA complex with the corresponding elements of the quaternary complex shows that the protein structure is again remodelled, but in this case the change is also accompanied by an alteration in the local DNA–protein contacts. In the quaternary complex, the DNA-binding domain of NFAT is reoriented relative to the DNA by a 20–24° rotation towards the Fos–Jun coiled-coil along the long axis of the DNA, and a 10–13° pivot perpendicular to it [5] (Figure 1). Concomitantly, an arginine residue in one of the loops stabilised on DNA shifts from making a bidentate contact with bases T7 and T6′ in the minor groove of the NFAT recognition site to contacting the carbonyl group of base T8, closer to the Fos–Jun contacts. There are also two contacts between the NFAT RHR and DNA that appear unique to the binary complex, yet recognition of the core NFAT site, GGAAA, is conserved between the two complexes. The cooperativity between NFAT and Jun–Fos is presumably driven by formation of the extensive contact surface between the two partners, involving primarily polar interactions. This interaction results in the formation
R617
Figure 2
Fos
Jun
NFAT (RHR-N)
NFAT (RHR-C)
DNA
1
5
10
15
5' T T G G A A A A T T T G T T T C A T A G 3' 3' C C T T T T A A A C A A A G T A T C A A 5' 1'
5'
10'
15'
Molscript representation of the crystal structure of the NFAT–Fos–Jun–DNA complex. The sequence of the DNA is shown at the bottom. (Reproduced with permission from [6].)
of a composite continuous DNA-binding groove for the recognition of 15 base pairs (Figure 2), with the induced bend creating the appropriate alignment of the respective recognition sequences. Although the structural changes in the bound DNA on formation of the quaternary complex are likely to be unfavourable, any energetic penalty for the extra distortion would be small and more than compensated for by the additional protein–protein contacts. Although solving the structure of the quaternary NFAT–Fos–Jun–DNA complex was a formidable crystallographic challenge, the complex is relatively simple compared to the multiprotein complexes such as the ‘enhanceosomes’ that bind to enhancer elements [10]. To build such complex structures, additional inter-protein contact surfaces need to be maintained or established. Conformational remodelling could act in such a way or, as suggested by Zhou et al. [5], the flexibility of the individual constituents could free higher-order complex formation from the strictures of rigid geometrical constraints. A limited flexibility might, however, also confer additional selectivity in partner selection. Cooperative binding should normally fulfill at least two requirements — that the recognition of the core DNA binding sequence is maintained, and that a protein–protein interface is established. The core sequences, which can often be recognised by several members of a transcription factor family,
R618
Current Biology, Vol 8 No 17
establish a spatial constraint which may allow interface formation only between particular partners, and preclude interface formation between others, depending on the permitted flexibility of the proteins involved. There are strong parallels between the structure of the quaternary NFAT–Fos–Jun–DNA complex and that of the ternary complex containing the yeast MATα2 homeodomain protein and the MADS box transcription factor MCM1 bound to DNA [3]. In the latter complex, DNA bending induced by MCM1 brings the two proteins into close proximity, facilitating cooperative binding. Contact between the two proteins also requires a structural reordering, in which the otherwise flexible amino-terminal extension of the MATα2 homeodomain forms a β-hairpin that grips the MCM1 surface by making parallel β-strand hydrogen bonds coupled with hydrophobic packing. Additional conformational flexibility in this case comes from the ability of a sequence of eight amino acids in this aminoterminal extension to exist as either a β strand or an α helix. The more general picture is one in which the assembly of a higher-order protein–DNA structure from its individual component complexes does not follow the principles of Legoland. Structural remodelling of one or more of the participants may be the rule, rather than the exception. Other examples include the distortion of nucleosomal DNA that occurs on binding of linker histone, generating a structurally asymmetric particle, and the twisting of one of the two DNA-binding domains of the TATA-boxbinding protein relative to the other on interaction with DNA and TFIIB [1,2]. References 1. Nikolov DB, Chen H, Halay ED, Usheva AA, Hisatake K, Roeder RG, Burley SK: Crystal structure of a TFIIB-TBP-TATA-element ternary complex. Nature 1995, 377:119-128. 2. Tan S, Hunsiker Y, Sargent DF, Richmond TJ: Crystal structure of a yeast TBP/TFIIA/DNA complex. Nature 1996, 381:127-134. 3. Tan S, Richmond TJ : Crystal structure of the yeast MATa2/MCM1/DNA ternary complex. Nature 1998, 391:660-666. 4. Jain J, McCaffrey PG, Miner Z, Kerppola TK, Lambert JN, Verdine GL, Curran T, Rao A: The T-cell transcription factor NFATp is a substrate for calcineurin and interacts with Fos and Jun. Nature 1993, 365:352-355. 5. Zhou, P, Sun LJ, Dötsch V, Wagner G, Verdine GL: Solution structure of the core NFATC1/DNA complex. Cell 1998, 92:687-696. 6. Chen L, Glover JNM, Hogan PG, Rao A, Harrison SC: Structure of the DNA-binding domains from NFAT, Fos and Jun bound specifically to DNA. Nature 1998, 392:42-48. 7. Ellenberger TE, Brandl CJ, Struhl K, Harrison SC: The GCN5 basic region leucine zipper binds DNA as a dimer of uninterrupted ahelices: crystal structure of the protein-DNA complex. Cell 1992, 71:1223-1237. 8. Glover JN, Harrison SC: Crystal structure of the heterodimeric transcription factor c-Fos-c-Jun bound to DNA. Nature 1995, 373:257-261. 9. Sitlani A, Crothers, DM: DNA-binding domains of Fos and Jun do not induce DNA curvature: an investigation with solution and gel methods. Proc Natl Acad Sci USA 1998, 95:1404-1409. 10. Kim TK, Maniatis T: The mechanism of transcriptional synergy of an in vitro assembled interferon-b enhanceosome. Mol Cell 1998, 1:119-129.
If you found this dispatch interesting, you might also want to read the April 1998 issue of
Current Opinion in Genetics & Development which included the following reviews, edited by C David Allis and Susan M Gasser, on Chromosomes and expression mechanisms: The histone tails of the nucleosome Karolin Luger and Timothy J Richmond Unfolding the mysteries of heterochromatin Lori L Wallrath Imprinting mechanisms in mammals Wolf Reik and Jörn Walter Transcription of chromatin: these are complex times Jennifer A Armstrong and Beverly M Emerson Covalent modifications of histones: expression from chromatin templates James R Davie Dosage compensation in flies and worms: the ups and downs of X-chromosome regulation John C Lucchesi DNA damage checkpoints update: getting molecular Ted Weinert Replication origins in yeast versus metazoa: separation of the haves and have nots David M Gilbert Recombination at work for meiosis Kathleen N Smith and Alain Nicolas Centromeres: proteins, protein complexes, and repeated domains at centromeres of simple eukaryotes Louise Clarke Centromeres: the missing link in the development of human artificial chromosomes Huntington F Willard Telomerase and chromosome end maintenance Joachim Lingner and Thomas R Cech Mechanisms of silencing in Saccharomyces cerevisiae Arthur J Lustig Chromosomal imprinting in plants Robert Martienssen Fragile-X syndrome and myotonic dystrophy: parallels and paradoxes Stephen J Tapscott, Todd R Klesert, RJ Widrow, Reinhard Stöger and Charles D Laird The full text of Current Opinion in Genetics & Development is in the BioMedNet library at http://BioMedNet.com/cbiology/gen