| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
First published online February 10, 2005; 10.1105/tpc.104.026435 © 2005 American Society of Plant Biologists
Solution Structure of an Arabidopsis WRKY DNA Binding Domain
|
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTS AND DISCUSSIONMETHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTS AND DISCUSSIONMETHODSREFERENCES |
|---|
; Rushton et al., 1995, 1996; de Pater et al., 1996), comprise a large family of plant-specific transcription factors (Eulgem et al., 2000; Riechmann et al., 2000). Most of the WRKY proteins are involved in responses to bacterial or fungal pathogens and a pathogen-related hormone, salicylic acid (Rushton et al., 1996; Eulgem et al., 1999; Chen and Chen, 2000; Du and Chen, 2000). Typically, they mediate signaling triggered by the elicitor molecule encoded by the pathogen and induce rapid apoptosis of cells so as to prevent further invasion. An increasing number of recent reports have indicated that the WRKY proteins are also involved in a variety of other plant-specific reactions, such as senescence (Robatzek and Somssich, 2001), morphogenesis (Johnson et al., 2002), and cold tolerance (Huang and Duman, 2002) and responses to wounding (Hara et al., 2000), drought (Pnueli et al., 2002; Rizhsky et al., 2002; Seki et al., 2002a), heat shock (Rizhsky et al., 2002), high salinity (Seki et al., 2002a), UV radiation (Izaguirre et al., 2003), sugar (Sun et al., 2003), and gibberellin (Zhang et al., 2004). The WRKY proteins are known to mediate signaling through binding to the promoter regions of target genes containing the W-box sequence, (T)TTGACY, where Y is C or T (de Pater et al., 1996; Rushton et al., 1996; Eulgem et al., 1999).
The WRKY proteins share a DNA binding domain of 
60 amino acids in length, which contains an invariant WRKYGQK sequence (after which the domain was named; Rushton et al., 1996) and a CX4–5CX22–23HXH (or CX7CX23HXC in a minority of cases) zinc binding motif. It is known that DNA binding in vitro is abolished by divalent metal chelators (Rushton et al., 1995; de Pater et al., 1996; Hara et al., 2000) and restored by the addition of Zn2+ (Maeo et al., 2001). A mutational experiment revealed that the conserved Cys and His residues in the motif are involved in this zinc-dependent DNA binding activity and that the invariant WRKYGQK sequence is required for DNA binding (Maeo et al., 2001). Many of the WRKY proteins possess two WRKY domains, which are classified into group I, whereas those possessing a single WRKY domain are classified into group II or III. Group III WRKY proteins are typified mainly by having the less common CX7CX23HXC zinc binding motif (Eulgem et al., 2000). It has been shown for Arabidopsis thaliana, parsley (Petroselinum crispum), and sweet potato (Ipomoea batatas) group I WRKY proteins that the C-terminal WRKY domain, but not the N-terminal domain, is responsible for sequence-specific binding to DNA (Ishiguro and Nakamura, 1994; de Pater et al., 1996; Eulgem et al., 1999).
In this study, the three-dimensional structure of the C-terminal WRKY domain of the Arabidopsis WRKY4 protein (WRKY4-C) was determined by NMR spectroscopy. Circular dichroism (CD) and NMR experiments provided evidence for the zinc-dependent folding of the structure. The structure consists of a four-stranded ß-sheet with a zinc binding pocket formed at one end and represents a novel type of zinc and DNA binding domain, although it shows partial similarity to other larger proteins. A DNA titration experiment strongly suggested that the region corresponding to the conserved sequence WRKYGQK is directly involved in DNA binding, which enabled us to propose a structural framework for the DNA recognition mechanism.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTS AND DISCUSSIONMETHODSREFERENCES |
|---|
). Conversely, the final spectrum after the titration (Figure 1A) is more consistent with the spectrum of an unfolded peptide, with a large negative peak at 200 nm (Woody, 1995). Thus, it was shown that EDTA induced unfolding of the WRKY4-C protein by chelating the Zn2+ ion. NMR spectra showed more clearly that the protein in a solution without EDTA possesses a specific tertiary structure, whereas the protein in a solution containing EDTA loses its large chemical shift dispersion, an indication of being largely unfolded (see supplemental data online). Adding Zn2+ ions to the solution of the unfolded protein induced refolding to the native structure as judged by the nearly identical CD spectra (Figure 1A). The titration profile showed complete refolding of the protein at an approximate 1:1 molar ratio of Zn2+ ions to the protein (Figure 1B). These results indicate that binding of a single Zn2+ ion to a WRKY4-C molecule causes the formation of a specific tertiary structure. This is likely to account for the requirement of Zn2+ for DNA binding activity (Rushton et al., 1995; de Pater et al., 1996; Hara et al., 2000; Maeo et al., 2001). These observations are also consistent with the results of mutations of the conserved Cys and His residues, all of which abolished the DNA binding activity (Maeo et al., 2001).
|
) (Figures 2 and 3). The ß1 strand is largely kinked in the middle at the position of Gly418 of the invariant WRKYGQK sequence because the residues on either side (Tyr417 and Gln419) form contacts with two adjacent residues of strand ß2 (Tyr431 and Ser430, respectively) in the antiparallel ß-strand connection. Specifically, Trp414, Arg415, Lys416, Tyr417, Gln419, and Lys420 of strand ß1 are adjacently connected to Cys434, Lys433, Tyr432, Tyr431, Ser430, and Arg429, respectively, of strand ß2 by the formation of interstrand hydrogen bonds or the close proximity of 
-protons, leaving Gly418 with no connecting partner.
|
|
|
atom of His463 and the H
atom of His465 coordinate to the Zn2+ ion, and significant violations were observed when the structure was calculated with theoretical constraints based on alternative assumptions (data not shown). Extensive hydrophobic interactions were observed between Tyr412–His463, Trp414–Cys434, Trp414–Lys443, Trp414–Tyr459, Tyr417–Tyr431, Val422–Asn425, Tyr427–Arg447, Tyr432–Lys443, Lys433–Arg442, and Thr436–His465, which were determined by more than three pairs of side-chain carbon atoms having average distances of <4.5 Å over the 20 structures of the ensemble [e.g., between Tyr412 and His463, 15 carbon atom pairs, such as CB(Tyr412)–CE1(His463), CG(Tyr412)–CE1(His463), CD1(Tyr412)–CD2(His463), CD1(Tyr412)–CE1(His463), and CE1(Tyr412)–CG (His463), have distances <4.5 Å on average]. Most of these residues are the aromatic (Tyr412, Trp414, Tyr431, Tyr432, Tyr459, and His465), aliphatic (Val422), or basic (Lys433, Arg442, Lys443, and Arg447) residues that are highly conserved among the WRKY domains (Figure 2A), indicating that these domains share a common structural architecture. The numerous basic residues involved in the hydrophobic interactions are likely to be suitable for maintaining a single ß-sheet structure. That is, the basic residues possess long hydrophobic side chains with hydrophilic ends, which facilitate hydrophobic packing between residues oriented in similar directions while allowing the hydrophilic ends to be exposed to the solvent. The observation that many Tyr residues, which possess hydrophilic ends, but no Phe residues, which do not, are involved in hydrophobic interactions in WRKY4-C is also consistent with this premise.
It is of particular interest that the bulky side chain of Trp414 of the invariant WRKYGQK sequence, located on strand ß1, has extensive contacts to Cys434 of ß2, Lys443 of ß3, and Tyr459 of ß4 (Figure 4), as listed above. Therefore, Trp414 appears to be of particular importance in stabilizing the structure of the four-stranded ß-sheet. It also makes hydrophobic contacts to the side chains of Tyr412 and His463 (two and three carbon atom pairs with average distances of <4.5 Å were identified for Trp414–Tyr412 and Trp414–His463, respectively). Thus, among the residues contacting Trp414, two (Cys434 and His463) are the Zn2+ coordinating residues. Therefore, Trp414, together with Tyr412, which makes extensive hydrophobic contacts to His463 as described above, is likely to be important in stabilizing the Zn2+ coordination and thereby forming the structural core of the domain (Figure 4). The prominent contribution of Trp414 in stabilizing the structure is enabled by the kink in strand ß1 located at the position of Gly418. As described above, Tyr417 of the WRKYGQK sequence also makes extensive hydrophobic contacts to Tyr431, which likely stabilizes the ß-sheet on the opposite side of Trp414.
|
; Rushton et al., 1996; Eulgem et al., 1999). Using surface plasmon resonance, WRKY4-C was demonstrated to bind to double-stranded DNA containing this consensus sequence (Figures 5A and 5B), with a binding constant of 2.6 x 107 M–1 at low ionic strength (100 mM KCl). Binding was sensitive to the ionic strength and became weaker at higher salt concentrations (Figure 5B). The response ratio of maximal binding of the protein to DNA immobilized on the sensor chip suggested that WRKY4-C binds to DNA with a stoichiometry of 1:1 (Figures 5A and 5B). Nonspecific binding to DNA with an unrelated sequence was also observed, although the binding was much weaker (Figure 5B). Thus, the recognition specificity to the W-box sequence was observed for the domain used in this study.
|
An NMR titration experiment was performed to elucidate the protein–DNA interface (Figure 5C). By adding increasing amounts of DNA, chemical-shift perturbations were observed in heteronuclear single-quantum coherence (HSQC) spectra. The positions of some HSQC cross-peaks changed only slightly or not at all, such that the chemical-shift changes were easy to follow (e.g., Cys434, Thr436, Cys439, Asn464, and Ala469 in Figure 5C). For others, however, the differences were significant enough that the changes could not be followed, in which case analysis of three-dimensional spectra of the protein–DNA complex was necessary for their assignment (e.g., Lys416, Val422, Lys423, Arg429, Arg442, Glu446, Ala448, Ala454, Val456, and Thr457 in Figure 3C). It should be noted that the HSQC cross-peak of Lys416 of the WRKYGQK sequence is extremely downfield shifted, a possible reason for which will be discussed later. These chemical shift changes were completed when the concentration ratio of DNA to protein reached 1.0, which indicated a 1:1 binding stoichiometry of the protein–DNA complex.
By classifying the residues according to their chemical shift differences, it became clear that one side of the structure is largely affected by the binding of DNA (Figure 6). Based on this result, we built a structural model of the complex of WRKY4-C and standard B-DNA using a computational approach (Figure 6). In this model, ß-strand 1 consisting of the invariant WRKYGQK sequence residues deeply enters the major groove of the DNA in such a way that the ß-sheet plane is nearly perpendicular to the DNA axis. Accordingly, the side chains of Arg415, Lys416, Gln419, and Lys420, which are likely the most important residues in the sequence-specific recognition, contact the bases. The backbone atoms of Tyr417 also contact the bases. The basic side chains of Arg413, Lys423, Arg429, and Arg442 and the backbone amide of Lys416 form intermolecular hydrogen bonds to the DNA phosphate groups. The hydrogen bond formation by the Lys416 backbone amide is consistent with the extreme downfield shift of its NMR cross-peak (Figure 5C). In addition, the side chain of Lys433 is at a distance from a DNA phosphate group that would allow the formation of a salt bridge or an indirect hydrogen bond through a water molecule.
|
6-bp region, which is largely consistent with the length of the (T)TTGACY W-box consensus (de Pater et al., 1996; Rushton et al., 1996; Eulgem et al., 1999). The concave curvature of strand ß1 induced by the kink at Gly418 allows the deep entrance of this strand into the DNA groove. Thereby, the three basic residues and the Gln residue of the motif are able to make contacts with the DNA bases, even though the neighboring residues, namely Arg415 and Lys416 or Gln419 and Lys420, are pointing toward the opposite sides of the sheet.
The involvement of the WRKYGQK sequence in DNA binding was demonstrated by a mutational experiment (Maeo et al., 2001). In this experiment, the Trp, Tyr, and two Lys residues were shown to be indispensable for DNA binding. This is consistent with our proposed model, wherein the two Lys residues contact the DNA bases directly. In addition, the Trp and Tyr residues are likely to be necessary for the structural architecture required for DNA binding, as discussed above. In contrast with our proposed model, the results of Maeo et al. (2001) also showed that DNA binding activity was not significantly affected by mutation of the Arg or Gln residues of the sequence, and, although significantly reduced, was not abolished by the mutation of the Gly residue. Our results suggested the involvement of these residues in DNA binding, whether directly or indirectly through forming the structural architecture. The reason for this inconsistency (or partial consistency) and details of the mechanism of sequence-specific DNA recognition by the WRKY domains await a structural determination of the protein complexed with DNA.
Different WRKY Domains
It is known that the C-terminal domain of group I WRKY proteins, but not the N-terminal domain, is responsible for DNA binding (Ishiguro and Nakamura, 1994; de Pater et al., 1996; Eulgem et al., 1999). We investigated the structural basis for this observation by producing a computational model of the WRKY4-N structure based on its sequence homology to WRKY4-C (Figure 7). We also modeled the WRKY4-N/DNA complex, based on the proposed model of the WRKY4-C/DNA shown in Figure 6. The modeled WRKY4-N structure is very similar to WRKY4-C and is energetically favored, which confirms that the structural architecture is shared by the different WRKY domains (Figure 7A). The energy gain upon forming the protein–DNA complex was approximately –200 kcal mol–1 for WRKY4-N, whereas it was approximately –300 kcal mol–1 for WRKY4-C, as estimated with implicit solvent effects in the AMBER force field (Case et al., 2002). This suggests that both domains are capable of DNA binding, although WRKY4-C binds more strongly. The difference in the energy gain of 100 kcal mol–1 is mostly due to differences in electrostatic energies. The surface potential of WRKY4-N (Figure 7B) showed that the domain is still basic, with similar locations of the conserved basic residues to those in WRKY4-C (Figure 3C). However, two conserved acidic residues in the N-terminal WRKY domains, Glu246 and Glu275, are in the equivalent positions to the neutral residues Pro426 and Val456, respectively, of WRKY4-C (Figure 2A) and reduce the basicity of WRKY4-N (Figure 7B). Indeed, the numbers of basic and acidic residues are 12 and 6, respectively, for WRKY4-C, whereas they are 11 and 7, respectively, for WRKY4-N, resulting in the net charges of +6 for WRKY4-C and +4 for WRKY4-N. In addition, Arg413 of WRKY4-C, the side chain of which is positively charged and contacts a DNA phosphate group in the proposed model, is conserved in the C-terminal WRKY domains but not in the N-terminal domains (Figure 2A). These distinctions are likely to be responsible for the predicted differences in the electrostatic energy gain for each domain upon formation of the complex. A more accurate comparison of free energies, including an evaluation of the entropic effects, will require an extensive dynamical calculation encompassing the surrounding water molecules.
|
) are more appropriately classified into group I, even though they possess a single WRKY domain.
Arg413 of WRKY4-C also is not conserved in the group III WRKY domains (Figure 1). In addition, two other important basic residues in WRKY4-C, Lys423 and Arg442, are not conserved in three group III WRKY domains: WRKY30, WRKY46, and WRKY53 (Figure 2A). It has been reported that similar group III WRKY proteins from tobacco (Nicotiana tabacum) are capable of binding to the W-box DNA sequence (Chen and Chen, 2000). Interestingly, for the three group III WRKY domains, basic residues are instead conserved in two different positions, namely, Pro426 and the gap between Thr435 and Thr436 in WRKY4-C (Figure 2A). By examining the proposed model, we suggest that the long side chains of the basic residues in these positions are capable of contacting DNA phosphate groups without a large change in the DNA binding mode proposed in Figure 6. Therefore, these group III WRKY proteins are likely to possess essentially the same DNA binding mode as the group I C-terminal WRKY domains, although this is accomplished at least partly through different residues.
Similarity to Other Structures
The Research Collaboratory for Structural Bioinformatics Protein Data Bank (http://www.rcsb.org/pdb/) was searched for structures similar to that of WRKY4-C using the program DALI (Holm and Sander, 1993). Structures with high similarity scores (Z-scores >4.0) were those of ß-galactosidase (PDB code 1bgl; Z-score 4.6), ferric hydroxamate uptake protein (PDB code 1by5; Z-score 4.2), and clathrin fragment (PDB code 1bpo; Z-score 4.2), which are much larger (500 to 1000 amino acids) than WRKY4-C and possess antiparallel ß-sheets of four strands or more as inseparable parts of their large structural architectures (data not shown). This is also essentially true for another 27 partially similar structures with Z-scores >3.0. These partly similar structures do not bind to zinc or DNA, and so it is evident that WRKY4-C possesses a novel type of zinc and DNA binding structure.
We found two DNA binding domain structures of marginal similarity to WRKY4-C, Arabidopsis NAC (no apical meristem) (PDB code 1ut4; Ernst et al., 2004) (Z-score 2.6) and Drosophila GCM (glia cell missing) (PDB code 1odh; Cohen et al., 2003) (Z-score 2.2), which are transcription factors that have functional similarities to WRKY4-C (Figure 8). Like other partly similar structures, they are significantly larger than WRKY4-C and possess antiparallel ß-sheets of four strands or more as substructures. The NAC structure possesses a six-stranded antiparallel ß-sheet in which the central four strands were aligned to the four-stranded ß-sheet of WRKY4-C using the DALI program (in addition, a short strand of the NAC structure was aligned to a loop region of WRKY4-C) (Figure 8A). It is important to note that ß-strand 3 of NAC, corresponding to ß1 of WRKY4-C, possesses a curvature forming a concave surface, which is induced by the presence of a Gly residue (shown by an arrow in Figure 8A) and one more residue at its C terminus. The indole ring of a Trp residue, located four residues to the N-terminal side of the Gly residue, is oriented inside of the ß-sheet (Figure 8A) and forms extensive hydrophobic interactions. These characteristics are very similar to ß-strand 1 of WRKY4-C (Figure 8C) as described above. In ß-strand 3 of NAC, a WKATGXD[K/R] sequence is conserved, which appears to be partially similar to the WRKYGQK motif in ß-strand 1 of WRKY4-C. In addition, an area of the molecular surface including the side of ß-strand 3 is positively charged and was suggested to be involved in DNA binding (Ernst et al., 2004). Therefore, we propose that the structures of these two plant-specific transcription factor DNA binding domains are related not only in their secondary structure arrangement, but also in the DNA binding mode.
|
) (Figure 8B). This structure possesses a small four-stranded antiparallel ß-sheet, which was aligned to the WRKY4-C ß-sheet using the DALI program. Most importantly, the strand corresponding to ß-strand 1 of WRKY4-C deeply enters the major groove of the DNA in a similar manner proposed for the WRKY/DNA complex (Figure 6). In addition, one of the two zinc ions is coordinated by two Cys and two His residues located in very similar positions to those of the zinc-coordinating Cys and His residues of WRKY4-C (Figures 8B and 8C). Specifically, Cys76 of the GCM domain is located on the ß-strand that was aligned to ß2 of WRKY4-C where Cys434 is located. Cys125 of the GCM domain is located at a loop N-terminal to the ß-strand that was aligned to ß3 of WRKY4-C, at the N terminus of which Cys439 is located. Similarly, the positions of the two His residues, His152 and His154, appear to be equivalent to those of His463 and His465 of WRKY4-C. Furthermore, the two domains are common in that the N atom is used for the zinc coordination in the N-terminal His residues, His152 of the GCM domain and His463 of WRKY4-C, whereas the N atom is used in the C-terminal His residues. Therefore, we suggest that they share a common zinc binding structural motif, although their sequence motifs are different because a subdomain containing another zinc binding site is inserted between Cys76 and Cys125 of the GCM domain.
Plant-Specific Transcription Factors
Despite the structural and functional similarities, no significant sequence similarities were observed for the DNA binding domains of the NAC, GCM, and WRKY proteins. However, considering the high degree of similarity in the Zn binding modes and, likely, DNA binding modes of the GCM and WRKY domains, we suggest the possibility of an evolutionary relationship between these domains. Considering that group I WRKY proteins have been identified recently in primitive eukaryotes, such as a slime mold (Dictyostelium discoideum) and a protist (Giardia lamblia) (Ülker and Somssich, 2004), and that GCM domains so far have been identified only in animals such as insects, fish, and mammals (Cohen et al., 2003), the WRKY domain might be more ancestral than the GCM domain.
It was reported that AtERF1 of the plant-specific AP2/ERF family and Tn916 integrase, a bacterial endonuclease, share a similar DNA binding mode through a three-stranded antiparallel ß-sheet (Allen et al., 1998; Wojciak et al., 1999). Recently, putative endonucleases from bacteria, phages, and a protist possessing DNA binding domains homologous to the plant AP2/ERF domains were reported (Magnani et al., 2004). In addition, the B3 DNA binding domain of the plant-specific transcription factor RAV1 is similar to that of a bacterial restriction enzyme EcoRII in the three-dimensional structure, DNA binding mechanism, and primary sequence (Yamasaki et al., 2004b). Therefore, it is possible that structural frameworks of the DNA binding domains of many plant-specific transcription factors were established before the plant kingdom was isolated from the other eukaryotic kingdoms, after which they gained their specific functions and drastically increased in number, along with the evolution of plant-specific reactions in development and environmental responses.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTS AND DISCUSSIONMETHODSREFERENCES |
|---|
) with the ID code RAFL03-05-E03 (MIPS code: At1g13960). The PCR primers used were designed so that the T7 promoter sequence, ribosome binding site, and oligohistidine tag, as well as the cleavage site for tobacco etch virus protease, are attached at the 5' end and so that the T7 terminator sequence is attached at the 3' end (T. Yabuki, Y. Motoda, M. Saito, N. Matsuda, T. Kigawa, and S. Yokoyama, unpublished results). Consequently, additional amino acids (i.e., GlySerSerGlySerSerGly) derived from the expression vector were attached to the N terminus of the protein. The 13C-, 15N-labeled, and unlabeled proteins were expressed by a large-scale cell-free system developed at RIKEN (Tokyo, Japan) (Kigawa et al., 1999, 2004; Yokoyama et al., 2000). The protein was purified by TALON (Clontech, Palo Alto, CA), HiPrep 26/10 Desalting (Amersham, Piscataway, NJ), and HiTrap SP (Amersham) column chromatography. The buffers used were as follows: 20 mM Tricine-NaOH, pH 8.0, 1 M NaCl, and 5 µM ZnCl2 for sample loading and washing of the TALON column, 20 mM Tris-HCl, pH 7.5, 300 mM NaCl, 500 mM imidazole, and 5 µM ZnCl2 for protein elution from the TALON column, 20 mM Tris-HCl, pH 7.5, 300 mM NaCl, 2 mM 2-mercaptoethanol, 5 µM EDTA, and 10 µM ZnCl2 for HiPrep 26/10 Desalting chromatography, and 20 mM Tris-HCl, pH 7.0, 1 mM DTT, 5 µM EDTA, 10 µM ZnCl2, and 0 to 1 M NaCl for HiTrap SP chromatography. Protein concentration was determined by A280 values with the molar absorption coefficient calculated from the amino acid sequence (Pace et al., 1995). For NMR measurements, proteins of 1.0 mM were dissolved in 20 mM potassium phosphate buffer, pH 6.0, containing 300 mM KCl, 20 µM ZnCl2, 1 mM deuteriated DTT (Isotec, Miamisburg, OH), 0.5 mM sodium 2,2-dimethyl-2-silapentane-5-sulfonate (DSS), and 5% D2O, unless otherwise stated. For CD titration experiments, protein of an initial concentration of 50 µM was dissolved in 20 mM potassium phosphate buffer, pH 6.0, containing 300 mM KCl, and 1 mM DTT.
CD Measurements
CD spectra were recorded on a JASCO J-820 spectropolarimeter (Tokyo, Japan) using a 1-mm path-length cell. Spectra between 195 and 250 nm were obtained using a scanning speed of 20 nm min–1, a response time of 2.0 s, a bandwidth of 1 nm, a spectral resolution of 0.5 nm, and four-scan averaging. Temperature was controlled at 298 K by a Peltier cooling system. Titration experiments were conducted by adding increasing amounts of 1 mM EDTA or 1 mM ZnCl2. Using a nonlinear least-squares method, the obtained titration data were fitted to linear lines with saturation values, assuming that the binding of the Zn2+ ion to the protein was extremely strong.
NMR Measurements and Resonance Assignments
Typical homonuclear and heteronuclear NMR spectra (Wüthrich, 1986; Bax, 1994) were recorded on Bruker DMX-750 (750.13 MHz for 1H and 76.02 MHz for 15N) and DMX-500 (500.13 MHz for 1H, 125.76 MHz for 13C, and 50.68 MHz for 15N) spectrometers (Karlsruhe, Germany) at 298 K, essentially as previously described (Yamasaki et al., 2004a). The backbone and side chain resonance assignments were analyzed as described previously (Yamasaki et al., 2004a). HSQC spectra of samples containing 100% D2O were recorded at 283 K, by which 19 hydrogen bond donors were identified. Using the HMQC-J experiment (Kay and Bax, 1989), 49 3JHNH coupling values were obtained. By analyzing the NOESY, TOCSY, and DQF-COSY spectra, 22 and 6 pairs of Hß and Val H
resonances, respectively, were assigned stereospecifically.
Determination of the Three-Dimensional Structures
The distance constraints derived from the NOESY spectra were classified into four categories, 1.5 to 2.8, 1.5 to 3.5, 2.0 to 4.5, and 2.5 to 6.0 Å, according to the relationship that NOE intensity is inversely proportional to the sixth power of distance, which was calibrated with the average intensities of the intraresidue NOEs of H-H pairs of Phe or Tyr residues. When a pair of NOEs from chemically equivalent protons (e.g., Hß1 and Hß2) without stereospecific assignment was classified into the same distance category, two separate constraints were imposed as if they were stereospecifically assigned. For other NOEs from stereospecifically unassigned protons, the constraints were imposed using the sum-averaged distances from all the equivalent protons (Nilges, 1993). For maintaining a hydrogen bond, a constraint of 1.5 to 2.5 Å was imposed on the distance between the hydrogen and the acceptor oxygen, whereas another constraint, of 2.5 to 3.5 Å, was imposed on the distance between the donor nitrogen and the acceptor oxygen. A force constant of 150 kcal mol–1 Å–2 was used for these distance constraints.
The 
angle constraints were classified into three categories, –120° ± 50°, –65° ± 35°, and –100° ± 70°, corresponding to the 3JN coupling values larger than 8.0 Hz, smaller than 7.0 Hz, and 7.0 to 8.0 Hz, respectively. For stereospecifically assigned residues, 
1 torsion angle constraints classified into three categories, 60° ± 40°, 180° ± 40°, and –60° ± 40°, were imposed. A force constant of 100 to 200 kcal mol–1 rad–2 was used for the dihedral angle constraints.
To maintain the zinc coordination, theoretical constraints were imposed on the Zn–S and Zn–N distances, the S–Zn–S, S–Zn–N, and Cß–S–Zn angles, and the Zn–imidazole ring planes, as described (Lee et al., 1989). The force constants were 150 kcal mol–1 Å–2, 30 kcal mol–1 rad–2, and 500 kcal mol–1 Å–2 for the theoretical distance, angle, and plane positional constraints, respectively.
Random simulated annealing (Nilges et al., 1988) was performed using the program CNS (Brünger et al., 1998). At the initial stage of the structure calculation, NOE distance and dihedral angle constraints were imposed. The hydrogen bond and theoretical constraints were used only at the final stage of the calculation, so that the hydrogen bond acceptors and zinc-coordinating atoms were determined using the structures obtained at an earlier stage. From 50 initial structures, 20 structures with no distance violation larger than 0.2 Å, no torsion angle violation larger than 2°, and the lowest total energies were selected as accepted structures. The minimized mean structures were produced by a protocol for the selection of the accepted structures in the CNS program. The average rms deviations of the coordinates from the unminimized mean structure of the ensemble (Table 1) were calculated using the program MOLMOL (Koradi et al., 1996). The and 
dihedral angles were analyzed using the program Procheck-NMR (Laskowski et al., 1996). The secondary structure elements were also identified by Procheck-NMR.
Surface Plasmon Resonance
Experiments were performed at 298 K using a Biacore X apparatus (Biacore, Uppsala, Sweden). The running buffer was 20 mM potassium phosphate, pH 6.0, containing 100 to 300 mM KCl, and 0.005% Tween 20. A total of 583 and 539 resonance units of two double-stranded DNAs, 5'-bio-CGCCTTTGACCAGCGC-3'/5'-GCGCTGGTCAAAGGCG-3' (the W-box consensus sequence is underlined) and 5'-bio-TCTTTAATTTCTAATATATTTAGAA-3'/5'-TTCTAAATATATTAGAAATTAAAGA-3', respectively ("bio" indicates the 5'-biotinylated strand), were immobilized on the surfaces of Sensor Chip SAs (Biacore) in one (flow cell 2) of the two flow cells, and the other was treated as the control. Solutions containing WRKY4-C, at concentrations of 2 nM to 1 µM, were injected into the flow cells at 20 µL min–1 for 5 min. The equilibrium binding constants were obtained by fitting the equilibrium response values at different protein concentrations to the simple 1:1 binding model using BIAevaluation 3.0 software (Biacore).
NMR Titration Analysis
HSQC spectra of WRKY4-C at the initial concentration of 0.20 mM, dissolved in 20 mM potassium phosphate buffer, pH 6.0, containing 200 mM KCl, 20 µM ZnCl2, 1 mM deuteriated DTT, 0.5 mM 2,2-dimethyl-2-silapentane-5-sulfonate, and 5% D2O, were recorded at 298 K by adding increasing amounts of 4.2 mM of the double-stranded 16mer DNA (5'-CGCCTTTGACCAGCGC-3'/5'-GCGCTGGTCAAAGGCG-3'; the W-box consensus sequence is underlined) dissolved in the same buffer. The concentration of the double-stranded DNA was determined using an extinction coefficient calculated after digestion of the strands with phosphodiesterase I (Worthington, Lakewood, NJ). The assignments of the backbone 1H and 15N resonances of WRKY4-C in the complex with DNA were completed by analyses of heteronuclear three-dimensional spectra recorded on the complex sample at a concentration of 0.8 mM.
Structural Modeling of the WRKY4-C/DNA Complex and WRKY4-N
Starting from an initial model in which the side of the protein containing the residues with largely affected chemical shifts in the DNA titration experiment was oriented to the DNA major groove, the most energetically favored structure was obtained by a careful and systematic search, essentially as described previously (Yamasaki et al., 2004a). The structure model of WRKY4-N was produced on the basis of sequence homology to WRKY4-C, essentially as described (Yamasaki et al., 2004b).
The coordinates of the determined structure have been deposited in the Protein Data Bank under accession ID 1WJ2.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Online version contains Web-only data. 
Article, publication date, and citation information can be found at www.plantcell.org/cgi/doi/10.1105/tpc.104.026435.
Received July 28, 2004; accepted December 4, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTS AND DISCUSSIONMETHODSREFERENCES |
|---|
Bax, A. (1994). Multidimensional nuclear magnetic resonance methods for protein studies. Curr. Opin. Struct. Biol. 4, 738–744.[CrossRef]
Brünger, A.T., et al. (1998). Crystallography & NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr. D Biol. Crystallogr. 54, 905–921.
Case, D.A., et al. (2002). AMBER 7. (San Francisco: University of California).
Chen, C., and Chen, Z. (2000). Isolation and characterization of two pathogen- and salicylic acid-induced genes encoding WRKY DNA-binding proteins from tobacco. Plant Mol. Biol. 42, 387–396.[CrossRef][ISI][Medline]
Cohen, S.X., Moulin, M., Hashemolhosseini, S., Kilian, K., Wegner, M., and Müller, C.W. (2003). Structure of the GCM domain-DNA complex: A DNA-binding domain with a novel fold and mode of target site recognition. EMBO J. 22, 1835–1845.[CrossRef][ISI][Medline]
de Pater, S., Greco, V., Pham, K., Memelink, J., and Kijne, J. (1996). Characterization of a zinc-dependent transcriptional activator from Arabidopsis. Nucleic Acids Res. 24, 4624–4631.
Du, L., and Chen, Z. (2000). Identification of genes encoding receptor-like protein kinases as possible targets of pathogen- and salicylic acid-induced WRKY DNA-binding proteins in Arabidopsis. Plant J. 24, 837–847.[CrossRef][ISI][Medline]
Ernst, H.A., Olsen, A.N., Skriver, K., Larsen, S., and Lo Leggio, L. (2004). Structure of the conserved domain of ANAC, a member of the NAC family of transcription factors. EMBO Rep. 5, 1–7.[CrossRef][ISI][Medline]
Eulgem, T., Rushton, P.J., Robatzek, S., and Somssich, I.E. (2000). The WRKY superfamily of plant transcription factors. Trends Plant Sci. 5, 199–206.[CrossRef][ISI][Medline]
Eulgem, T., Rushton, P.J., Schmelzer, E., Hahlbrock, K., and Somssich, I.E. (1999). Early nuclear events in plant defence signaling: Rapid gene activation by WRKY transcription factors. EMBO J. 18, 4689–4699.[CrossRef][ISI][Medline]
Hara, K., Yagi, M., Kusano, T., and Sano, H. (2000). Rapid systematic accumulation for transcripts encoding a tobacco WRKY transcrption factor upon wounding. Mol. Gen. Genet. 263, 30–37.[CrossRef][ISI][Medline]
Holm, L., and Sander, C. (1993). Protein structure comparison by alignment of distance matrices. J. Mol. Biol. 233, 123–138.[CrossRef][ISI][Medline]
Huang, T., and Duman, J.G. (2002). Cloning and characterization of a thermal hysteresis (antifreeze) protein with DNA-binding activity from winter bittersweet nightshade, Solanum dulcamara. Plant Mol. Biol. 48, 339–350.[CrossRef][ISI][Medline]
Ishiguro, S., and Nakamura, K. (1994). Characterization of a cDNA encoding a novel DNA-binding protein, SPF1, that recognizes SP8 sequences in the 5' upstream regions of genes coding for sporamin and ß-amylase from sweet potato. Mol. Gen. Genet. 244, 563–571.[ISI][Medline]
Izaguirre, M.M., Scopel, A.L., Baldwin, I.T., and Ballare, C.L. (2003). Convergent responses to stress. Solar ultraviolet-B radiation and Manduca sexta herbivory elicit overlapping transcriptional responses in field-grown plants of Nicotiana longiflora. Plant Physiol. 132, 1755–1767.
Johnson, C.S., Kolevski, B., and Smyth, D.R. (2002). TRANSPARENT TESTA GLABRA2, a trichome and seed coat development gene of Arabidopsis, encodes a WRKY transcription factor. Plant Cell 14, 1359–1375.
Kay, L.E., and Bax, A. (1989). New methods for measurement of NH-CH coupling constants in 15N-labelled proteins. J. Magn. Reson. 86, 110–126.
Kigawa, T., Yabuki, T., Matsuda, N., Matsuda, T., Nakajima, R., Tanaka, A., and Yokoyama, S. (2004). Preparation of Escherichia coli cell extract for highly productive cell-free protein expression. J. Struct. Funct. Genomics 5, 63–68.[CrossRef][Medline]
Kigawa, T., Yabuki, T., Yoshida, Y., Tsutsui, M., Ito, Y., Shibata, T., and Yokoyama, S. (1999). Cell-free production and stable-isotope labelling of milligram quantities of proteins. FEBS Lett. 442, 15–19.[CrossRef][ISI][Medline]
Koradi, R., Billeter, M., and Wüthrich, K. (1996). MOLMOL: A program for display and analysis of macromolecular structures. J. Mol. Graph. 14, 51–55.[CrossRef][ISI][Medline]
Kraulis, P.J. (1991). MOLSCRIPT: A program to produce both detailed and schematic plots of protein structures. J. Appl. Crystallogr. 24, 946–950.[CrossRef]
Laskowski, R.A., Rullmann, J.A.C., MacArthur, W.M., Kaptein, R., and Thornton, J.M. (1996). AQUA and PROCHECK-NMR: Programs for checking the quality of protein structures solved by NMR. J. Biomol. NMR 8, 477–486.[ISI][Medline]
Lee, M.S., Gippert, G.P., Soman, K.V., Case, D.A., and Wright, P.E. (1989). Three-dimensional solution structure of a single zinc finger DNA-binding domain. Science 245, 635–637.
Maeo, K., Hayashi, S., Kojima-Suzuki, H., Morikami, A., and Nakamura, K. (2001). Role of conserved residues of the WRKY domain in the DNA-binding to tobacco WRKY family proteins. Biosci. Biotechnol. Biochem. 65, 2428–2436.[CrossRef][Medline]
Magnani, E., Sjölander, K., and Hake, S. (2004). From endonucleases to transcription factors: Evolution of the AP2 DNA binding domain in plants. Plant Cell 16, 2265–2277.
Nilges, M. (1993). A calculation strategy for the structure determination of symmetric dimers by 1H NMR. Proteins Struct. Funct. Genet. 17, 297–309.[CrossRef][ISI][Medline]
Nilges, M., Clore, G.M., and Gronenborn, A.M. (1988). Determination of the three-dimensional structures of proteins from inter-proton distance data by dynamic simulated annealing from a random array of atoms. FEBS Lett. 239, 129–136.[CrossRef][ISI][Medline]
Pace, C.N., Vajdos, F., Fee, L., Grimsley, G., and Gray, T. (1995). How to measure and predict the molar absorption coefficient of a protein. Protein Sci. 4, 2411–2423.[Abstract]
Pnueli, L., Hallak-Herr, E., Rozenberg, M., Cohen, M., Goloubinoff, P., Kaplan, A., and Mittler, R. (2002). Molecular and biochemical mechanisms associated with dormancy and drought tolerance in the desert legume Retama raetam. Plant J. 31, 319–330.[CrossRef][ISI][Medline]
Riechmann, J.L., Heard, J., Martin, G., Reuber, L., Jiang, C.-Z., Keddie, J., Adam, L., Pineda, O., Ratcliffe, O.J., Samaha, R.R., Creelman, R., Pilgrim, M., et al. (2000). Arabidopsis transcription factors: Genome-wide comparative analysis among eukaryotes. Science 290, 2105–2110.
Rizhsky, L., Liang, H., and Mittler, R. (2002). The combined effect of drought stress and heat shock on gene expression in tobacco. Plant Physiol. 130, 1143–1151.
Robatzek, S., and Somssich, I.E. (2001). A new member of the Arabidopsis WRKY transcription factor family, AtWRKY6, is associated with both senescence- and defence-related processes. Plant J. 28, 123–133.[CrossRef][ISI][Medline]
Rushton, P.J., Macdonald, H., Huttly, A.K., Lazarus, C.M., and Hooley, R. (1995). Members of a new family of DNA-binding proteins bind to a conserved cis-element in the promoters of -Amy2 genes. Plant Mol. Biol. 29, 691–702.[CrossRef][ISI][Medline]
Rushton, P.J., Torres, J.T., Parniske, M., Wernert, P., Hahlbrock, K., and Somssich, I.E. (1996). Interaction of elicitor-induced DNA-binding proteins with elicitor response elements in the promoters of parsley PR1 genes. EMBO J. 15, 5690–5700.[ISI][Medline]
Seki, M., et al. (2002a). Monitoring the expression profiles of 7000 Arabidopsis genes under drought, cold and high-salinity stresses using a full-length cDNA microarray. Plant J. 31, 279–292.[CrossRef][ISI][Medline]
Seki, M., et al. (2002b). Functional annotation of a full-length Arabidopsis cDNA collection. Science 296, 141–145.
Sun, C., Palmqvist, S., Olsson, H., Boren, M., Ahlandsberg, S., and Jansson, C. (2003). A novel WRKY transcription factor, SUSIBA2, participates in sugar signaling in barley by binding to the sugar-responsive elements of the iso1 promoter. Plant Cell 15, 2076–2092.
400 Hz at the proton frequency of 750 MHz: residues 416, 417, 419, 430, and 431], yellow [(
