Answers are grounded in this paper and cite their sources.
Ethylene-responsive element-binding proteins (EREBPs)have novel DNA-binding domains (ERF domains), which are widely conserved in plants, and interact specifically with sequences containing AGCCGCC motifs (GCC box). Deletion experiments show that some flanking region at the N terminus of the conserved 59-amino acid ERF domain is required for stable binding to the GCC box. Three ERF domain-containing fragments of EREBP2, EREBP4, and AtERF1 from tobacco and Arabidopsis, bind to the sequence containing the GCC box with a high binding affinity in the pm range. The high affinity binding is conferred by a monomeric ERF domain fragment, and DNA truncation experiments show that only 11-base pair DNA containing the GCC box is sufficient for stable ERF domain interaction. Systematic DNA mutation analyses demonstrate that the specific amino acid contacts are confined within the 6-base pair GCCGCC region of the GCC box, and the first G, the fourth G, and the sixth C exhibit highest binding specificity common in all three ERF domain-containing fragments studied. Other bases within the GCC box exhibit modulated binding specificity varying from protein to protein, implying that these positions are important for differential binding by different EREBPs. The conserved N-terminal half is likely responsible for formation of a stable complex with the GCC box and the divergent C-terminal half for modulating the specificity. Ethylene-responsive element-binding proteins (EREBPs)have novel DNA-binding domains (ERF domains), which are widely conserved in plants, and interact specifically with sequences containing AGCCGCC motifs (GCC box). Deletion experiments show that some flanking region at the N terminus of the conserved 59-amino acid ERF domain is required for stable binding to the GCC box. Three ERF domain-containing fragments of EREBP2, EREBP4, and AtERF1 from tobacco and Arabidopsis, bind to the sequence containing the GCC box with a high binding affinity in the pm range. The high affinity binding is conferred by a monomeric ERF domain fragment, and DNA truncation experiments show that only 11-base pair DNA containing the GCC box is sufficient for stable ERF domain interaction. Systematic DNA mutation analyses demonstrate that the specific amino acid contacts are confined within the 6-base pair GCCGCC region of the GCC box, and the first G, the fourth G, and the sixth C exhibit highest binding specificity common in all three ERF domain-containing fragments studied. Other bases within the GCC box exhibit modulated binding specificity varying from protein to protein, implying that these positions are important for differential binding by different EREBPs. The conserved N-terminal half is likely responsible for formation of a stable complex with the GCC box and the divergent C-terminal half for modulating the specificity. pathogenesis-related ethylene-responsive element ERE-binding protein ERE-binding factor base pair(s) amino acid(s) electrophoresis mobility shift binding assay N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine polyacrylamide gel electro- phoresis. The plant hormone ethylene regulates many complex processes of stress responses and developmental adaptations in higher plants (1Abeles F.B. Morgan P.W. Saltveit Jr., M.E. Ethylene in Plant Biology. 2nd Ed. Academic Press, Inc., New York1992Google Scholar). Stimulation of expression of pathogenesis-related (PR)1 genes, such as class I basic chitinases and β-1,3-glucanases, is among the consequences of ethylene involvement (2Ohme-Takagi M. Shinshi H. Plant Mol. Biol. 1990; 15: 941-946Crossref PubMed Scopus (105) Google Scholar). Analysis of the promoter regions of ethylene-inducible PR genes has led to the identification of an ethylene-responsive element (ERE) (3Sato F. Kitajima S. Koyama T. Yamada Y. Plant Cell Physiol. 1996; 37: 249-255Crossref PubMed Scopus (81) Google Scholar, 4Ohme-Takagi M. Shinshi H. Plant Cell. 1995; 7: 173-182Crossref PubMed Scopus (946) Google Scholar), which contains a conserved AGCCGCC motif (2Ohme-Takagi M. Shinshi H. Plant Mol. Biol. 1990; 15: 941-946Crossref PubMed Scopus (105) Google Scholar, 5Eyal Y. Meller Y. Lev-Yadun S. Fluhr R. Plant J. 1993; 4: 225-234Crossref PubMed Scopus (136) Google Scholar, 6Hart C.M. Nagy F. Meins Jr., F. Plamt Mol. Biol. 1993; 21: 121-131Crossref PubMed Scopus (91) Google Scholar, 7Zhou J. Tang X. Martin G.B. EMBO J. 1997; 16: 3217-3218Google Scholar) designated as the GCC box or occasionally the PR box (4Ohme-Takagi M. Shinshi H. Plant Cell. 1995; 7: 173-182Crossref PubMed Scopus (946) Google Scholar, 7Zhou J. Tang X. Martin G.B. EMBO J. 1997; 16: 3217-3218Google Scholar). Four different cDNAs encoding the GCC box-specific binding factors (EREBPs) have been isolated from tobacco (4Ohme-Takagi M. Shinshi H. Plant Cell. 1995; 7: 173-182Crossref PubMed Scopus (946) Google Scholar). The deduced amino acid sequences of EREBPs neither show homology to known DNA-binding proteins and transcription factors nor do they contain basic leucine zipper or zinc finger motifs, suggesting that EREBPs are a group of novel DNA-binding proteins. EREBPs contain highly conserved 58–59 sequences of residues (designated as ERF domains in this paper, see Fig. 1) within which GCC box-specific binding activity was identified (4Ohme-Takagi M. Shinshi H. Plant Cell. 1995; 7: 173-182Crossref PubMed Scopus (946) Google Scholar). Recently, it has been reported that ERF domains exist in numerous products of regulatory genes and cDNAs of unknown function from various plants (7Zhou J. Tang X. Martin G.B. EMBO J. 1997; 16: 3217-3218Google Scholar, 8Okamuro J.K. Caster B. Villarroel R. Van Montagu M. Jofuku K.D. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 7076-7081Crossref PubMed Scopus (518) Google Scholar), but they are not found in mammalian, fungi, or yeast (9Weigel D. Plant Cell. 1995; 7: 388-389Crossref PubMed Scopus (155) Google Scholar). Whereas ERF domains are highly conserved, EREBPs exhibit some divergence. The ERF domain also has been argued to be closely related to the AP2 domain of Arabidopsis (8Okamuro J.K. Caster B. Villarroel R. Van Montagu M. Jofuku K.D. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 7076-7081Crossref PubMed Scopus (518) Google Scholar, 10Stockinger E.J. Gilmour S.J. Thomashow M.F. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 1035-1040Crossref PubMed Scopus (1449) Google Scholar) although the DNA-binding activity of AP2 has not been proven nor has its target sequence been identified. Thus, it has been suggested that these transcription factors belong to two subdivisions (7Zhou J. Tang X. Martin G.B. EMBO J. 1997; 16: 3217-3218Google Scholar, 9Weigel D. Plant Cell. 1995; 7: 388-389Crossref PubMed Scopus (155) Google Scholar, 11Büttner M. Singh K.B. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 5961-5966Crossref PubMed Scopus (298) Google Scholar), with fewer amino acids being identical in all members of the extended family. Moreover, AP2 contains two DNA-binding domains whereas EREBPs have only one (9Weigel D. Plant Cell. 1995; 7: 388-389Crossref PubMed Scopus (155) Google Scholar). Among the proteins containing the ERF domain, EREBPs (4Ohme-Takagi M. Shinshi H. Plant Cell. 1995; 7: 173-182Crossref PubMed Scopus (946) Google Scholar), AtEBP (11Büttner M. Singh K.B. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 5961-5966Crossref PubMed Scopus (298) Google Scholar), Pti4–6 (7Zhou J. Tang X. Martin G.B. EMBO J. 1997; 16: 3217-3218Google Scholar), and AtERFs (Arabidopsis ethylene-responsive transcription factor) 2M. Ohta, M. Ohme-Takagi, and H. Shinshi, unpublished data. have been shown to have GCC box-specific binding activity. However, their DNA recognition mechanism and biological function are not yet understood. It is also unclear whether each ERF domain exhibits the same binding specificity with the GCC box. Attempts in this report were made to establish the GCC box-binding specificity of ERF domain by single-base substitution of the GCC box and quantitative electrophoretic mobility shift assay (EMSA). The GCC box binding affinities of three ERF domain-containing fragments, from tobacco (EREBP2 and EREBP4) and Arabidopsis (AtERF1), were examined, and the stoichiometry of their DNA interactions were established. Truncation analyses of the ERF domain fragment have revealed a minimal region for stable GCC box binding, and systematic mutation and truncation assays identified crucial bases within the GCC box responsible for specific protein-DNA interaction. The coding region of ERF domain fragments from EREBP2 (aa 88–164), EREBP4 (aa 122–222), and AtERF1 (aa 138–213), designated as EBP2-F, EBP4-F, and AtERF1-F, respectively, in this paper, were prepared by PCR, cloned into pET3C plasmid (Novagen), and expressed in BL21(DE3)pLysS Escherichia coli cells. After freeze-thaw and subsequent gentle sonication, the cell lysate was clarified and applied to heparin columns (Pros HE, PerSeptive Biosystems). Fractions containing ERF domain fragments were filtrated through Superdex 75 (Amersham Pharmacia Biotech) and further purified by cation exchange chromatography (Resource S, Amersham Pharmacia Biotech). The purified ERF domain fragments were concentrated and dialyzed against water. For preparation of the fusion proteins with maltose-binding protein, the coding region of EBP4-F or tobacco TGA1a (12Katagiri F. Lam E. Chua N.H. Nature. 1989; 340: 727-730Crossref PubMed Scopus (334) Google Scholar) were cloned into pMAL-2 plasmid (New England Biolabs Inc.), expressed in BL21 E. coli cells, and purified using amylose columns according to manufacturer instructions. Protein concentrations were measured using bicinchoninic acid (BCA) protein assay reagent kit (Pierce) and further confirmed by the methods of Gill and von Hippel (13Gill S.C. von Hippel P.H. Anal. Biochem. 1989; 182: 319-326Crossref PubMed Scopus (5083) Google Scholar). A 16-bp GCC box fragment (5′-CATAAGAGCCGCCACT-3′) of the 5′ upstream region of the tobacco Gln2 gene was prepared by synthesizing both strands (2Ohme-Takagi M. Shinshi H. Plant Mol. Biol. 1990; 15: 941-946Crossref PubMed Scopus (105) Google Scholar). After purification, the fragments were suspended in 1 × STE buffer, and the complementary strands were annealed. The resulting double strand oligonucleotides were labeled with [γ-32P]ATP (NEN Life Science Products) by T4 polynucleotide kinase (Toyobo) and purified through a Sephadex G-50 column before binding experiments. Mutant and truncated sequences (see Table I) were made in the same way.Table ITruncational and mutational binding assays for GCC boxnameGCC box segquencesKdWt16CATAAGAGCCGCCACT70 ± 58 pmMmgtattcAGCCGCCtga52 ± 17 pmMdCATAAGAtCCtCCACT>1.0 μmT14TAAGAGCCGCCACT44 ± 17 pmT12AGAGCCGCCACT81 ± 20 pmT12aGAGCCGCCACTG71 ± 30 pmT12bAGCCGCCACTGA>1.0 μmT12cAAGAGCCGCCAC>1.0 μmT12dTAAGAGCCGCCA>1.0 μmT11GAGCCGCCACT73 ± 28 pmA 16-mer wild type GCC box sequence (Wt16) contains a core region of -AGCCGCC- (highlighted and underlined). Quantitative EMSA was performed by using Wt16 or different mutants (M) and truncates (T) and EBP2-F at various concentrations. The numbers after T indicate the length of the oligos. The data for K d values are represented as the mean of three replicates ± S.D. K d values for nonspecific binding is about 0.1–1.0 μm or higher. Open table in a new tab A 16-mer wild type GCC box sequence (Wt16) contains a core region of -AGCCGCC- (highlighted and underlined). Quantitative EMSA was performed by using Wt16 or different mutants (M) and truncates (T) and EBP2-F at various concentrations. The numbers after T indicate the length of the oligos. The data for K d values are represented as the mean of three replicates ± S.D. K d values for nonspecific binding is about 0.1–1.0 μm or higher. Equilibrium dissociation constants (K d) of ERF domain-DNA binding were measured by quantitative EMSA. Labeled DNA (at a concentration of 10–100-fold lower than the K d) was mixed with a gradient concentration of ERF domain fragments in 10 μl of binding buffer (25 mm HEPES-KOH buffer at pH 7.5, containing 40 mmKCl, 0.1 mm EDTA, 0.1 mg/ml bovine serum albumin, and 10% glycerol). After incubation at room temperature for 15 min, the samples were run on an 8% nondenaturing PAGE, and the resulting gel was dried and visualized on a FUJIX BAS-2000 system (Fuji). Complexed and free DNA bands were scanned, their photodensity was measured, and K d was determined from binding titration analysis. The binding free energy change (ΔΔG) was calculated as: ΔΔG = RT ln [K d(mutant)/K d (wild-type)], where R is the gas constant and T is absolute temperature. Positive ΔΔG represents a decreased binding affinity, negative ΔΔG corresponds to an increased binding affinity, and zero ΔΔG means no effect on binding. A 10-fold decrease in binding affinity increases ΔΔG by about 1.3 kcal/mol. For protein cross-linking experiments, the purified proteins were incubated with glutaraldehyde in total volume of 4 μl in DNA-binding buffer (see above). After a 10-min incubation at room temperature, cross-linked products were separated by standard Tris-glycine, pH 8.3, discontinuous SDS-PAGE, or for smaller polypeptide, in the Tris-Tricine system (14Schägger H. von Jagow G. Anal. Biochem. 1987; 166: 368-379Crossref PubMed Scopus (10527) Google Scholar). Derivatives of a fragment of EREBP2 (EBP2-F), containing an ERF domain and 10-aa N-terminal and 8-aa C-terminal flanking regions, were used to identify the minimal sequence for GCC box binding. The different truncates of this fragment were subjected to EMSA for examination of the binding activity to the 16-bp GCC box sequence (Wt16, see Table I). Fig. 2 shows that EBP2-F exhibits a high binding activity to the GCC box. Deletion of 10 aa from the N terminus (Ec, aa 98–164) or both N- and C-terminal flanking regions (Enc, aa 98–156) leads to either a dramatic decrease or complete abolition of binding activity with the GCC box. Elimination of the C-terminal flanking region (En, aa 88–156) causes only a negligible decrease in binding affinity in comparison with EBP2-F. This indicates that the short stretch of N-terminal flanking region is necessary for binding of the ERF domains to the GCC box. To measure the equilibrium binding of the EREBPs, labeled wild type 16-bp GCC box sequence was subjected to EMSA for interaction with EBP2-F, EBP4-F, or AtERF1-F. Upon titration analysis, the equilibrium dissociation constant (K d) was determined under the given binding condition. All three fragments exhibit a high binding affinity with K d in a range from about 1 to 120 pm. The highest binding was found with EBP4-F (K d = 5.4 ± 4.3 pm) and the lowest was with AtERF1-F (K d = 167 ± 40 pm), whereas the K d for EBP2-F was intermediate at about 70 ± 58 pm. A similarK d was also observed for the binding between full-length EREBP2 and the GCC box. The high binding activity observed above led to consideration of whether a dimer or even oligomers of ERF domain fragments form in free solution or under DNA-binding condition. The possibility of dimerization of ERF domain was investigated by EMSA using two ERF domain fragments of different sizes. A maltose-binding protein fused to EBP4-F (Mal-EBP4-F, 53 kDa) was mixed with EBP4-F (11 kDa) in varying concentration before binding to the GCC box. The results shown in Fig. 3 illustrate that a mixture of these two molecules (lanes 8–10) at equivalent concentrations shift the DNA into two bands corresponding to the individual proteins (lanes 2–7), and no extra band was observed. When the Mal-EBP4-F concentration is kept constant at 1.0 nm while varying EBP4-F from 10 pm to 0.1 μm(lane 11–13), the DNA shifts reflect an intercompetitive behavior of the proteins. These results indicate the absence of an hetero-complex under the given binding condition. To further examine the possibility of an intrinsic homo-dimer present in binding buffer or DNA-induced dimerization, EBP2-F with or without interaction of the GCC box was cross-linked with increasing concentrations of glutaraldehyde and then subjected to standard SDS-PAGE. In a control experiment (Fig. 4 A), the bZip protein of tobacco TGA1a, known to be a dimer in solution (12Katagiri F. Lam E. Chua N.H. Nature. 1989; 340: 727-730Crossref PubMed Scopus (334) Google Scholar), was examined to confirm that glutaraldehyde cross-linking can be used to detect oligomerization. Using DNA-binding conditions in the presence or absence of the GCC box, no bands corresponding to higher forms of EBP2-F (8.3 kDa) were observed regardless of the concentration of glutaraldehyde or the presence of GCC box sequence (Fig. 4, B and C). This suggests that EBP2-F exists in binding solution as a monomer and that no dimerization is induced upon binding with GCC box. To determine the minimal GCC box motif required for stable binding by EREBPs, were made at the of the GCC box as in Table and the binding assays with EBP2-F were Table I shows that the 16-bp GCC box can be by to and at the 5′ terminus without effect on binding the terminus the binding and This indicates that the DNA fragment is the minimal GCC box sequence by ERF domain fragments and that flanking sequences 1 at 5′ and 3 at terminus of the GCC box are necessary for the stable binding of EREBPs. Table I also shows that of the region of the GCC box at both the 5′ and has no effect on binding However, double substitution of with T at positions and within the GCC box the binding. This suggests that the GCC is responsible for specific protein binding. To further the ERF domain binding specificity of each individual base pair within the GCC box, EMSA with the three ERF domain fragments was performed with GCC box mutants systematic single-base base in the GCC box from to was with the three sequence see Fig. and free energy (ΔΔG) were from quantitative titration analysis. The results in Fig. show that for all three ERF domain fragments, base at and the decrease in binding affinity, that these three positions highly specific interactions with EREBPs. The of and in the binding specificity with EREBPs with (4Ohme-Takagi M. Shinshi H. Plant Cell. 1995; 7: 173-182Crossref PubMed Scopus (946) Google Scholar, 7Zhou J. Tang X. Martin G.B. EMBO J. 1997; 16: 3217-3218Google Scholar), but also to be an important base within the GCC box. For all three ERF domain fragments substitution of A at of the GCC box with bases shows effect on binding for AtERF1-F, of to causes an decrease in binding In the ERF domain fragments are by base at positions than and For of to and to T EBP2-F binding whereas substitution of to either or T has effect on the binding affinity of specific contacts between the EREBPs and the GCCGCC region of the GCC box, but modulated interactions also with the various proteins of the ERF family. with of EBP2-F (see Fig. have revealed that its which the ERF domain a short N-terminal is necessary for GCC box binding. This is with a of and Shinshi (4Ohme-Takagi M. Shinshi H. Plant Cell. 1995; 7: 173-182Crossref PubMed Scopus (946) Google Scholar). However, the ERF domain not binding and the N-terminal flanking to be crucial for binding with the GCC box. It is unclear the N-terminal amino acids specifically interact with the GCC box. is no homology at N-terminal flanking region among the of ERF domain-containing it not be responsible for the specific GCC box but an of the ERF domain for DNA binding. The present also shows that ERF domain fragments from tobacco and Arabidopsis bind to 16-bp oligonucleotides containing the GCC box from the 5′ upstream region of the tobacco PR All three fragments exhibit high binding affinity to the GCC box with K d in the pm range. a high binding constant is to that of such as the (K d = Y. A. Proc. Natl. Acad. Sci. U. S. A. 1989; PubMed Scopus Google Scholar) and (K d = 0.1 H. S. J. Mol. Biol. PubMed Scopus Google Scholar), and transcription factors such as (K d = J. T. M. S. A. Proc. Natl. Acad. Sci. U. S. A. 1993; PubMed Scopus Google Scholar). All of the DNA-binding proteins are either or contain with their DNA of in a However, gel shift binding assays for dimer formation revealed that each ERF domain fragments interact as with the GCC box (see Fig. the results from protein cross-linking experiments (see Fig. and gel of EBP2-F not that the ERF domain fragment is a monomer in and the GCC box not dimerization upon protein binding. truncation assays (see Table I) that the GCC box 1 at the 5′ and 3 at the flanking regions are sufficient for the binding activity of EREBPs. To this binding between a monomeric protein domain and such a short DNA sequence has not been affinity be by the base in the GCC box, the of ERF domain fragments = of through of the ERF domains or DNA such as DNA In the present results that a novel of DNA binding is in interactions of ERF domains with the GCC box. Quantitative EMSA with systematic single-base of the GCC box that the contacts with the ERF domain fragments at and whereas is comparison (see Fig. 1) of the conserved ERF domain among the three ERF domain fragments shows that their N-terminal are This region to with the GCC box binding specificity of all three ERF domain fragments observed in this implying that the N-terminal half is responsible for the specific recognition of the GCC box. (8Okamuro J.K. Caster B. Villarroel R. Van Montagu M. Jofuku K.D. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 7076-7081Crossref PubMed Scopus (518) Google Scholar) that an N-terminal region (designated the of which be than function in DNA binding of the of basic amino acids at this conserved In many proteins with an in Fig. 1) with N-terminal half in their ERF domain have been shown to GCC box binding activity. the and (see Fig. 1) have sequence homology in their ERF suggesting that they a target In has been shown to bind a sequence of E.J. Gilmour S.J. Thomashow M.F. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 1035-1040Crossref PubMed Scopus (1449) Google Scholar) than the The from the systematic mutation show some of GCC box binding specificity at the positions such as and among the three ERF domain fragments (see Fig. This is related to of the sequences of the C-terminal of the three ERF domain fragments a region with high this region not interact with the GCC box its sequence of the ERF the binding specificity with the GCC box. The also interact with regulatory factors this region is common in the conserved ERF domain and interactions (9Weigel D. Plant Cell. 1995; 7: 388-389Crossref PubMed Scopus (155) Google Scholar, K.D. Van Montagu M. J.K. Plant Cell. PubMed Scopus Google Scholar). The in GCC box binding specificity found in this important in the expression of PR genes by different ERF domain-containing proteins. for at the of this
No takes yet. Share an insight, caveat, or question.
Hao et al. (1998) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: