DNA bending by Fos-Jun and the orientation of heterodimer binding depend on the sequence of the AP-1 site
1997; Springer Nature; Volume: 16; Issue: 10 Linguagem: Inglês
10.1093/emboj/16.10.2917
ISSN1460-2075
Autores Tópico(s)Advanced biosensing and bioanalysis techniques
ResumoArticle15 May 1997free access DNA bending by Fos–Jun and the orientation of heterodimer binding depend on the sequence of the AP-1 site Nirmala Rajaram Nirmala Rajaram Howard Hughes Medical Institute and Department of Biological Chemistry, University of Michigan Medical School, Ann Arbor, MI, 48109-0650 USA Search for more papers by this author Tom K. Kerppola Corresponding Author Tom K. Kerppola Howard Hughes Medical Institute and Department of Biological Chemistry, University of Michigan Medical School, Ann Arbor, MI, 48109-0650 USA Search for more papers by this author Nirmala Rajaram Nirmala Rajaram Howard Hughes Medical Institute and Department of Biological Chemistry, University of Michigan Medical School, Ann Arbor, MI, 48109-0650 USA Search for more papers by this author Tom K. Kerppola Corresponding Author Tom K. Kerppola Howard Hughes Medical Institute and Department of Biological Chemistry, University of Michigan Medical School, Ann Arbor, MI, 48109-0650 USA Search for more papers by this author Author Information Nirmala Rajaram1 and Tom K. Kerppola 1 1Howard Hughes Medical Institute and Department of Biological Chemistry, University of Michigan Medical School, Ann Arbor, MI, 48109-0650 USA *Corresponding author. E-mail: [email protected] The EMBO Journal (1997)16:2917-2925https://doi.org/10.1093/emboj/16.10.2917 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions Figures & Info Interactions among transcription factors that bind to separate promoter elements depend on distortion of DNA structure and the appropriate orientation of transcription factor binding to allow juxtaposition of complementary structural motifs. We show that Fos and Jun induce distinct DNA bends at different binding sites, and that heterodimers bind to AP-1 sites in a preferred orientation. Sequences on each side of the consensus AP-1 recognition element have independent effects on DNA bending. A single base pair substitution outside the sequences contacted in the X-ray crystal structure alters DNA bending. Substitution of sequences flanking the AP-1 site has converse effects on DNA bending in opposite directions, suggesting that the extent of DNA bending by Fos and Jun is determined in part by the anisotropic bendability of sequences flanking the AP-1 site. DNA bending by Fos and Jun, and the orientation of heterodimer binding are interrelated. Reversal of the orientation of heterodimer binding causes a shift in the direction of DNA bending. The preferred orientation of heterodimer binding is determined both by contacts between a conserved arginine in the basic region of Fos and the central asymmetric guanine as well as the structure of sequences flanking the AP-1 site. Consequently, the structural adaptability of the Fos–Jun–AP1 complex may contribute to its functional versatility at different promoters. Introduction The architecture of the promoter region is critical for the correct assembly of transcription factor complexes and their function in transcription regulation. This architecture is determined by the relative positions and orientations of binding of transcription regulatory proteins and is further elaborated through interactions among these proteins as well as protein-induced changes in DNA structure. Although many transcription regulatory proteins can function when multiple copies of their binding sites are placed upstream of a heterologous transcription initiation site, their activities at such artificial promoters frequently differ from their functions at native promoters (Thanos and Maniatis, 1995). Furthermore, the functions of promoter elements—and by inference the proteins that bind to those elements—are interdependent in transgenic animals (Robertson et al., 1995). Thus, the organization of the regulatory elements in the promoter region and the correct orientation of transcription factor binding to these elements are essential for appropriate transcription regulation in the physiological context. Fos and Jun participate in the selective activation of transcription of different genes in response to distinct extracellular signals. Structural differences between complexes formed at different binding sites, as well as interactions among transcription factors that bind to separate sequence elements, may contribute to the differences in Fos and Jun function at different promoters. Thus, determination of differences in the structures of Fos–Jun complexes at various binding sites as well as differences in their potential interactions with other transcription factors are important for understanding the specificity of their regulatory functions. Fos and Jun regulate gene expression by binding to AP–1 sites that frequently share a conserved heptanucleotide recognition sequence. However, sequences outside of this core recognition element also influence the affinity of Fos and Jun binding (Ryseck and Bravo, 1991; Kerppola and Curran, 1994). The molecular basis of recognition of the heptanucleotide core element has been established through X-ray crystallographic analysis of the minimal leucine zipper dimerization and basic DNA-binding (bZIP) domains of Fos and Jun bound to an AP-1 site (Glover and Harrison, 1995). However, no direct contacts to nucleotides located further than four base pairs from the center of the AP-1 site are detected in the crystal. The molecular basis for the recognition of sequences outside of this core element is therefore unknown. Fos and Jun were originally shown to induce DNA bending by phasing analysis at a consensus AP-1 site (Kerppola and Curran, 1991a,b). Both the bZIP domains as well as regions overlapping the transcription activation domains of Fos and Jun induce DNA bending (Kerppola and Curran, 1997). All of the regions in Fos and Jun that influence DNA bending contain clusters of charged amino acid residues, and DNA bending is reduced in the presence of multivalent cations, suggesting that DNA bending by Fos and Jun is caused at least in part by charge interactions. However, no DNA bending was observed in the X-ray crystal structure of the minimal bZIP regions of Fos and Jun bound to an AP-1 site (Glover and Harrison, 1995). Studies of the sequence dependence of intrinsic DNA bending by gel electrophoresis and X-ray crystallography have also reached diametrically opposite conclusions. Whereas A tracts are the principal source of DNA bending in gel electrophoresis assays (reviewed in Haran et al., 1994), A tracts are always straight in crystals, and bending is frequently observed within G:C-rich sequences (Goodsell et al., 1994 and references therein). Both crystal packing forces (DiGabriele and Steitz, 1993) as well as agents used to promote crystallization (Dlakic et al., 1996) can influence the conformation of DNA in the crystal. The relationship between the variability in DNA structure observed in crystals and the conformational dynamics of DNA in solution is controversial (Goodsell et al., 1994; Haran et al., 1994). The AP-1 site is asymmetric by virtue of the central C:G base pair as well as sequences flanking the heptanucleotide core. Mutational analysis of the AP-1 site as well as UV crosslinking experiments suggest that heterodimers recognize the AP-1 site in an asymmetric manner (Nakabeppu and Nathans, 1989). In contrast, in the X-ray crystal structure, the heterodimer is found to bind to the AP-1 site equally in both orientations (Glover and Harrison, 1995). Furthermore, DNA cleavage studies using Fos and Jun bZIP region peptides coupled to free radical generators at their amino-terminal ends indicated that heterodimer binding to the AP-1 site is orientation-independent (Chen et al., 1995). Thus, although recognition of the core AP-1 sequence is understood on the basis of the X-ray crystal structure, the mechanism of differential recognition of the two half-sites by Fos and Jun remains unknown. We have examined the sequence dependence of DNA bending by Fos and Jun. We have found that sequences flanking the conserved AP-1 recognition element influence DNA bending independently and that base pairs outside of the region contacted in the X-ray crystal structure affect DNA bending. Heterodimers bind to the AP-1 site in a preferred orientation, and the orientation of heterodimer binding and the direction of DNA bending are interdependent. Results To examine DNA bending by Fos and Jun complexes at different AP-1 sites, we constructed probes in which the AP-1 sites were placed adjacent to an intrinsic DNA bend, and the spacing between the AP-1 site and the reference bend was varied over one turn of the DNA helix (Figure 1). DNA bending at the different binding sites was examined by gel electrophoretic phasing analysis. When the protein-induced and intrinsic DNA bends are in phase, they cooperate to increase the overall extent of DNA bending. In contrast, when the two bends are out of phase, they counteract each other and reduce the net DNA bend. Therefore, differences in the relative mobilities of complexes in phasing analysis reflect differences in DNA bending. The DNA bend angle and orientation were calculated by comparison with intrinsic DNA bend standards (Kerppola and Curran, 1997). Since the different AP-1 sites were placed in the same helical phase relationships with the intrinsic DNA bend, comparison of DNA bending at different binding sites does not depend on quantitation of the DNA bend angles, but can be evaluated directly by examination of the relative mobilities of the protein–DNA complexes compared with the probes alone. Figure 1.Analysis of DNA bending at AP-1 sites containing different flanking sequences. (A) Comparison of the structures of complexes containing in phase (above) or out of phase (below) intrinsic and protein-induced DNA bends. These complexes differ by the insertion of five base pairs between the protein binding site and the intrinsic bend, resulting in a difference in net DNA bending and electrophoretic mobility. (B) Sequences of AP-1 sites used to examine the influence of flanking sequences on DNA bending by Fos and Jun. Flanking sequences shared with the M site are single underlined, whereas flanking sequences shared with the X site are double underlined. A central C:G base pair is indicated by a triangle above the sequence, whereas a central G:C base pair is indicated by a triangle below the sequence. Sequence A represents the consensus AP-1 site used in studies of DNA bending by Fos and Jun (Kerppola and Curran, 1991a,b). Sequences B, C and H have been used for analysis of DNA bending by GCN4 (Gartenberg et al., 1990; Paolella et al., 1994). Download figure Download PowerPoint Sequences flanking the AP-1 site influence DNA bending To simplify comparisons between different recognition sequences, we initiated the analysis by using a pseudo-symmetrical binding site (M) in which sequences within nine base pairs from the center of the AP-1 site were palindromic. Bending at this pseudo-symmetrical binding site was compared with bending at a site encompassing the oligonucleotide sequence used in the X-ray crystallographic analysis of DNA bending by Fos and Jun (X) (Figures 2 and 3). All of the complexes examined induced different bends at these binding sites. Figure 2.The sequence of the AP-1 binding site influences DNA bending by Fos and Jun. Chimeric proteins consisting of the transcription activation domains (ovals) of Fos (FA, solid) and Jun (JA, open) fused to the bZIP domains (rounded rectangles) of Fos (FA, solid) and Jun (JD, open) (Kerppola and Curran, 1997) were incubated with phasing analysis probes containing the sites indicated above the lanes (see Figure 1 for descriptions) and the complexes were analyzed on an 8% polyacrylamide gel. Each set of lanes contained probes with 26, 28, 30, 32, 34 and 36 (M, XM, M-6T) or 28, 30, 32, 34, 36 amd 36 (X, XM and X-6G) base pair separations between the centers of the AP–1 site and the reference bend. Download figure Download PowerPoint Figure 3.Quantitation of DNA binding by chimeric Fos and Jun proteins at different binding sites. The relative mobilities of complexes formed by chimeric proteins containing different combinations of the Fos and Jun transcription activation and bZIP domains bound to the AP-1 sites shown above were plotted as a function of the separation between the centers of the AP-1 site and the intrinsic DNA bend. The diagrams on the left indicate the domain organization of the various complexes as discussed in Figure 2 and in Kerppola and Curran (1997). The abscissa is 25 to 39 base pairs for the eight plots on the left, and base pairs 20 to 31 for the three plots on the right, and the ordinate is 0.6 to 1.4. The DNA bend angle and direction are shown in the upper left and upper right corners of each plot. Standard deviations are plotted as vertical bars that are in many cases smaller than the symbols used to plot the data. Multivariate analysis of variance indicated that the differences in DNA bending between all pairs of binding sites were highly significant (P X > XM > MX > M > M-6T. This hierarchy was identical to the ranking of the binding sites based on the G:C content of base pairs flanking the AP-1 site at positions ±6, ±7 and ±8, and was reversed for complexes that bend DNA in the opposite direction. Since Fos–Jun heterodimers are predicted to bend these sequences toward the major groove, this ranking is consistent with the sequence dependence of DNA bending by CAP and nucleosomes (Satchwell et al., 1986; Gartenberg and Crothers, 1988). However, since the base pair at position −6 apparently affected DNA bending independent of other flanking sequences, further studies will be necessary to determine whether the effects of individual base pairs on DNA bending by Fos and Jun are affected by the identities of neighboring base pairs. We have found that Fos and Jun binding to different AP-1 sites results in distinct DNA bends. Moreover, Fos–Jun heterodimers bind to AP-1 sites in a preferred orientation that is determined by contacts to the asymmetric central base pair and by the structure of sequences flanking the AP-1 site. Thus, one function of DNA bending by Fos and Jun may be to control the orientation of heterodimer binding at different regulatory elements both based on the sequence of the recognition site as well as in response to DNA bending by other proteins. It is interesting that Fos–Jun heterodimers can bind to an AP-1 site adjacent to an NFAT site bound by NFATp with a stronger orientation preference than to the AP-1 site in the absence of NFATp (Chen et al., 1995). Therefore, protein-induced changes in DNA structure may mediate both the cooperative binding of transcription factors to overlapping or adjacent binding sites and folding of the promoter region into a conformation compatible with interactions among multiple regulatory proteins and transcription activation. Materials and methods Plasmid construction and protein purification Phasing analysis plasmids pNR412-26..-36, pNR421-28..-38, pNR413-28..-38, pNR431-26..-38, pNR414-26..36, pNR441-28..-38, pNR502-26..-36 and pNR512-26..-36 (where .. indicates a series spaced by two base pairs) containing sites M, X, MX, XM, M-6T, X-6G, H and W respectively were constructed by cloning the oligonucleotides shown in Figure 1 between the XbaI and SalI sites of plasmids pTK401-26 and pTK401-28 (Kerppola and Curran, 1991a). To generate plasmids with different separations between the centers of the AP-1 sites and the intrinsic DNA bends, oligonucleotides of different lengths containing an additional TGAC or TGACTGAC sequence inserted between the AP-1 site and the intrinsic bend were used. The phasing analysis plasmids pTK401-21..-30, pDP-AP1-21..30 and pDP-CRE-21..30 containing sites A, B and C respectively have been described (Kerppola and Curran, 1991a; Paolella et al., 1994). Plasmid vectors for expression of truncated and chimeric Fos and Jun fusion proteins have been described (Kerppola and Curran, 1997). Probes for phasing analysis were prepared by PCR amplification of fragments between 349 and 366 base pairs in size from the various plasmids and phasing analysis of DNA bending by complexes formed by various Fos and Jun complexes was performed as described (Kerppola and Curran, 1997). The DNA bend angle was calculated from the amplitude of the phasing function by using the relative mobilities of DNA fragments containing intrinsic DNA bends as a calibration curve. The direction of DNA bending was calculated from the minima of the phasing function based on the known orientation of DNA bending by A:T tracts. Acknowledgements We thank David Paolella and Alanna Schepartz for providing plasmids pDP-AP1-21..30 and pDP-CRE-21..30, David Leonard, Ronald Diebold and Neelam Taneja for critical comments on the manuscript. References Chen L, Oakley MG, Glover JNM, Jain J, Dervan P, Hogan P, Rao A and Verdiner G (1995) Only one of the two DNA-bound orientations of AP-1 found in solution cooperates with NFATp. Curr Biol, 5, 882–889.CrossrefCASPubMedWeb of Science®Google Scholar Crothers DM and Drak J (1992) Global features of DNA structures by comparative gel electrophoresis. Methods Enzymol, 212, 46–71.CrossrefCASPubMedWeb of Science®Google Scholar DiGabriele AD and Steitz TA (1993) A DNA dodecamer containing an adenine tract crystallizes in a unique lattice and exhibits a new bend. J Mol Biol, 231, 1024–1039.CrossrefCASPubMedWeb of Science®Google Scholar Dlakic M, Park K, Griffith JD, Harvey SC and Harrington RE (1996) The organic crystallizing agent 2-methyl-2,4-pentanediol reduces DNA curvature by means of structural changes in A-tracts. J Biol Chem, 271, 17911–17919.CrossrefCASPubMedWeb of Science®Google Scholar Gartenberg MR and Crothers DM (1988) DNA sequence determinants of CAP-induced bending and protein binding affinity. Nature, 333, 824–829.CrossrefCASPubMedWeb of Science®Google Scholar Gartenberg MR, Ampe C, Steitz TA and Crothers DM (1990) Molecular characterization of the GCN4-DNA complex. Proc Natl Acad Sci USA, 87, 6034–6038.CrossrefCASPubMedWeb of Science®Google Scholar Glover JN and Harrison SC (1995) Crystal structure of the heterodimeric bZIP transcription factor c- Fos-c-Jun bound to DNA. Nature, 373, 257–261.CrossrefCASPubMedWeb of Science®Google Scholar Goodsell DS, Kaczor GM and Dickerson RE (1994) The crystal structure of C-C-A-T-T-A-A-T-G-G. Implications for bending of B-DNA at T-A steps. J Mol Biol, 239, 79–96.CrossrefCASPubMedWeb of Science®Google Scholar Haran TE, Kahn JD and Crothers DM (1994) Sequence elements responsible for DNA curvature. J Mol Biol, 244, 135–143.CrossrefCASPubMedWeb of Science®Google Scholar Kerppola TK (1996) Fos and Jun bend the AP-1 site: Effects of probe geometry on the detection of protein induced DNA bending. Proc Natl Acad Sci USA, 93, 10117–10122.CrossrefCASPubMedWeb of Science®Google Scholar Kerppola TK and Curran T (1991a) Fos-Jun heterodimers and Jun homodimers bend DNA in opposite orientations: Implications for transcription factor cooperativity. Cell, 66, 317–326.CrossrefCASPubMedWeb of Science®Google Scholar Kerppola TK and Curran T (1991b) DNA bending by Fos and Jun: The flexible hinge model. Science, 254, 1210–1214.CrossrefCASPubMedWeb of Science®Google Scholar Kerppola TK and Curran T (1994) A conserved region adjacent to the basic domain is required for recognition of an extended DNA binding site by Maf/Nrl family proteins. Oncogene, 9, 3149–3158.CASPubMedWeb of Science®Google Scholar Kerppola TK and Curran T (1997) The transcription activation of Fos and Jun induce DNA bending through electrostatic interactions. EMBO J, 16, 2917–2926.Wiley Online LibraryPubMedWeb of Science®Google Scholar Leonard DA, Rajaram N and Kerppola TK (1997) Structural basis of DNA bending and oriented heterodimer binding by the bZIP domains of Fos and Jun. Proc Natl Acad Sci USA, 94, 4913–4918.CrossrefCASPubMedWeb of Science®Google Scholar Nakabeppu Y and Nathans D (1989) The basic region of Fos mediates specific DNA binding. EMBO J, 8, 3833–3841.Wiley Online LibraryCASPubMedWeb of Science®Google Scholar Paolella DN, Palmer CR and Schepartz A (1994) DNA targets for certain bZIP proteins distinguished by an intrinsic bend. Science, 264, 1130–1133.CrossrefCASPubMedWeb of Science®Google Scholar Robertson LM, Kerppola TK, Vendrell M, Luk D, Smeyne RJ, Bocchiaro C, Morgan JI and Curran T (1995) Regulation of c–fos expression in transgenic mice requires multiple interdependent transcription control elements. Neuron, 14, 241–252.CrossrefCASPubMedWeb of Science®Google Scholar Ryseck R-P and Bravo R (1991) c-Jun, Jun B, and Jun D differ in their binding affinities to AP-1 and CRE consensus sequences: effect of Fos proteins. Oncogene, 6, 533–542.CASPubMedWeb of Science®Google Scholar Satchwell SC, Drew HR and Travers AA (1986) Sequence periodicities in chicken nucleosome core DNA. J Mol Biol, 191, 659–675.CrossrefCASPubMedWeb of Science®Google Scholar Sitlani A and Crothers DM (1996) Fos and Jun do not bend the AP-1 recognition site. Proc Natl Acad Sci USA, 93, 3248–3252.CrossrefCASPubMedWeb of Science®Google Scholar Starr DB, Matsui W, Thomas JR and Yamamoto K (1996) Intracellular receptors use a common mechanism to interpret signaling information at response elements. Genes Dev, 10, 1271–1283.CrossrefCASPubMedWeb of Science®Google Scholar Thanos D and Maniatis T (1995) Virus induction of human IFN beta gene expression requires the assembly of an enhanceosome. Cell, 83, 1091–1100.CrossrefCASPubMedWeb of Science®Google Scholar Previous ArticleNext Article Volume 16Issue 1015 May 1997In this issue FiguresReferencesRelatedDetailsLoading ...
Referência(s)