Title: Two Cysteines in Plant R2R3 MYB Domains Participate in REDOX-dependent DNA Binding
Abstract: Plant R2R3 MYB domain proteins comprise one of the largest known families of transcription factors. Discrete evolutionary steps have shaped the plant-specific R2R3 MYB family from the broadly distributed R1R2R3 MYB proteins. R1R2R3 MYB domains have a single Cys residue (Cys-130) that needs to be reduced for DNA binding and transcriptional activity. In contrast, most R2R3 MYB domains contain two cysteines, Cys-49 and Cys-53, with Cys-53 at the equivalent position as Cys-130 in R1R2R3 MYB. Using the maize P1 regulator of flavonoid biosynthesis as a typical R2R3 MYB-domain protein, we investigated here the in vitro REDOX requirement for DNA binding by P1. We show that the C53S mutation requires reducing conditions for DNA-binding, whereas C53A binds DNA under oxidizing and reducing conditions. Neither mutation impairs the in vivo regulatory activity of P1. The C49S and C49A mutants bind DNA in vitro irrespective of the REDOX conditions. A C49I mutant, which simulates the MYB domain of c-MYB, binds DNA only under reducing conditions, and its binding is significantly affected by the C53S replacement. It is interesting that under non-reducing conditions, Cys-49 and Cys-53 form a disulfide bond that prevents the R2R3 MYB domain from binding DNA. Together, our results suggest that the evolutionary origin of Cys-49 within the plants has provided R2R3 MYB domains with a regulatory feature not present in animal MYB domains, highlighting fundamental structural and functional differences between similar DNA-binding domains from plants and animals. Plant R2R3 MYB domain proteins comprise one of the largest known families of transcription factors. Discrete evolutionary steps have shaped the plant-specific R2R3 MYB family from the broadly distributed R1R2R3 MYB proteins. R1R2R3 MYB domains have a single Cys residue (Cys-130) that needs to be reduced for DNA binding and transcriptional activity. In contrast, most R2R3 MYB domains contain two cysteines, Cys-49 and Cys-53, with Cys-53 at the equivalent position as Cys-130 in R1R2R3 MYB. Using the maize P1 regulator of flavonoid biosynthesis as a typical R2R3 MYB-domain protein, we investigated here the in vitro REDOX requirement for DNA binding by P1. We show that the C53S mutation requires reducing conditions for DNA-binding, whereas C53A binds DNA under oxidizing and reducing conditions. Neither mutation impairs the in vivo regulatory activity of P1. The C49S and C49A mutants bind DNA in vitro irrespective of the REDOX conditions. A C49I mutant, which simulates the MYB domain of c-MYB, binds DNA only under reducing conditions, and its binding is significantly affected by the C53S replacement. It is interesting that under non-reducing conditions, Cys-49 and Cys-53 form a disulfide bond that prevents the R2R3 MYB domain from binding DNA. Together, our results suggest that the evolutionary origin of Cys-49 within the plants has provided R2R3 MYB domains with a regulatory feature not present in animal MYB domains, highlighting fundamental structural and functional differences between similar DNA-binding domains from plants and animals. MYB DNA-binding domains are formed by one to three or more imperfect repeats (R1, R2, and R3) containing periodic tryptophan residues (1Kanei-Ishii C. Sarai A. Sawazaki T. Nakagoshi H. He D.N. Ogata K. Nishimura Y. Ishii S. J. Biol. Chem. 1990; 265: 19990-19995Abstract Full Text PDF PubMed Google Scholar, 2Ogata K. Hojo H. Aimoto S. Nakai T. Nakamura H. Sarai A. Ishii S. Nishimura Y. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 6428-6432Crossref PubMed Scopus (217) Google Scholar, 3Saikumar P. Murali R. Reddy E.P. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 8452-8456Crossref PubMed Scopus (104) Google Scholar). Each MYB repeat is defined by ∼50 amino acids that form three α-helices. The last two helices of each MYB repeat adopt a helix-turn-helix motif; the third helix of each MYB repeat is involved in the main DNA contacts (4Ogata K. Morikawa S. Nakamura H. Sekikawa A. Inoue T. Kanai H. Sarai A. Ishii S. Nishimura Y. Cell. 1994; 79: 639-648Abstract Full Text PDF PubMed Scopus (439) Google Scholar). The vertebrate Myb genes, which include c-Myb, A-Myb, and B-Myb, encode proteins with MYB domains formed by three MYB repeats (R1R2R3 MYB). In contrast, the majority of plant Myb genes encode proteins with only two MYB repeats most similar to the vertebrate R2 and R3 MYB repeats (R2R3 MYB) (5Braun E.L. Grotewold E. Plant Physiol. 1999; 121: 21-24Crossref PubMed Scopus (69) Google Scholar, 6Lipsick J.S. Oncogene. 1996; 13: 223-235PubMed Google Scholar, 7Rosinski J.A. Atchley W.R. J. Mol. Evol. 1998; 46: 74-83Crossref PubMed Scopus (225) Google Scholar, 8Stracke R. Werber M. Weisshaar B. Curr. Opin. Plant Biol. 2001; 4: 447-456Crossref PubMed Scopus (1534) Google Scholar). Plant R2R3 Myb genes are likely to have originated from an ancestral gene that is represented today by the B-Myb gene in vertebrates (6Lipsick J.S. Oncogene. 1996; 13: 223-235PubMed Google Scholar) and by the small pc-Myb (plant c-Myb) gene family in the plants (5Braun E.L. Grotewold E. Plant Physiol. 1999; 121: 21-24Crossref PubMed Scopus (69) Google Scholar). The evolutionary steps involved in the formation of the plant-specific R2R3 MYB domains from the broadly distributed R1R2R3 MYB domains involved the sequential i) loss of R1 to yield the "atypical R2R3 MYB domains," ii) replacement of the first tryptophan of R3 by a hydrophobic amino acid, and iii) insertion of a Leu residue between the second and third helices of R2, to give the "typical R2R3 MYB" domains (9Dias A.P. Braun E.L. McMullen M.D. Grotewold E. Plant Physiol. 2003; 131: 610-620Crossref PubMed Scopus (129) Google Scholar) (Fig. 1A). The loss of R1 and the replacement of the Trp residue probably had only a moderate effect upon the DNA-binding properties of the MYB domain, based on studies carried out in animal R1R2R3 MYB proteins (4Ogata K. Morikawa S. Nakamura H. Sekikawa A. Inoue T. Kanai H. Sarai A. Ishii S. Nishimura Y. Cell. 1994; 79: 639-648Abstract Full Text PDF PubMed Scopus (439) Google Scholar, 10Dini P.W. Lipsick J.S. Mol. Cell. Biol. 1993; 13: 7334-7348Crossref PubMed Scopus (64) Google Scholar, 11Zargarian L. Le Tilly V. Jamin N. Chaffotte A. Gabrielsen O.S. Toma F. Alpert B. Biochemistry. 1999; 38: 1921-1929Crossref PubMed Scopus (27) Google Scholar). In contrast, the insertion of the Leu residue in v-MYB, an oncogenic form of c-MYB containing only R2 and R3 (12Klempnauer K.H. Gonda T.J. Bishop J.M. Cell. 1982; 31: 453-463Abstract Full Text PDF PubMed Scopus (326) Google Scholar), completely abolished binding to DNA (13Williams C.E. Grotewold E. J. Biol. Chem. 1997; 272: 563-571Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar). An extensive amplification of the R2R3 Myb gene family occurred within the plants 250-400 million years ago (14Rabinowicz P.D. Braun E.L. Wolfe A.D. Bowen B. Grotewold E. Genetics. 1999; 153: 427-444PubMed Google Scholar), which resulted in Arabidopsis thaliana encoding 125 R2R3 Myb genes (8Stracke R. Werber M. Weisshaar B. Curr. Opin. Plant Biol. 2001; 4: 447-456Crossref PubMed Scopus (1534) Google Scholar, 15Riechmann J.L. Ratcliffe O.J. Curr. Opin. Plant Biol. 2000; 3: 423-434Crossref PubMed Scopus (343) Google Scholar) and maize and related monocots encoding more than 200 R2R3 Myb genes (14Rabinowicz P.D. Braun E.L. Wolfe A.D. Bowen B. Grotewold E. Genetics. 1999; 153: 427-444PubMed Google Scholar, 16Jiang C. Gu J. Chopra S. Gu X. Peterson T. Gene. 2004; 326: 13-22Crossref PubMed Scopus (106) Google Scholar). The amplification of the R2R3 Myb gene family in the plants provides a unique opportunity to understand how the evolutionary steps that shaped a family of transcription factors resulted in the functional differences displayed by these regulatory proteins throughout the plant kingdom. A residue that has remained completely conserved in plants, fungi, and animals during the evolution of MYB domains corresponds to a cysteine located in the DNA recognition helix of R2 (Cys-130 in c-MYB corresponding to Cys-53 in Fig. 1B). This cysteine was proposed to serve as a REDOX sensor in vertebrate MYB transcription factors, and mutations of this residue significantly impaired DNA binding and transcriptional activity of c-MYB and v-MYB (17Grasser F.A. LaMontagne K. Whittaker L. Stohr S. Lipsick J.S. Oncogene. 1992; 7: 1005-1009PubMed Google Scholar, 18Guehmann S. Vorbrueggen G. Kalkbrenner F. Moelling K. Nucleic Acids Res. 1992; 20: 2279-2286Crossref PubMed Scopus (107) Google Scholar). The NMR structure of the R2R3 MYB domain of c-MYB indicated that Cys-130, the only cysteine present in c-MYB and related MYB factors from vertebrates, is included in the hydrophobic core, maintaining the three helices of R2 in an unbound conformation at room temperature (4Ogata K. Morikawa S. Nakamura H. Sekikawa A. Inoue T. Kanai H. Sarai A. Ishii S. Nishimura Y. Cell. 1994; 79: 639-648Abstract Full Text PDF PubMed Scopus (439) Google Scholar). Consistent with this model, the replacement of Cys-130 for Ser in c-MYB resulted in the loss of DNA binding (18Guehmann S. Vorbrueggen G. Kalkbrenner F. Moelling K. Nucleic Acids Res. 1992; 20: 2279-2286Crossref PubMed Scopus (107) Google Scholar). It is interesting, however, that during the evolution of R2R3 MYB domains (Fig. 1A), a second cysteine appeared, located four residues N-terminal to the highly conserved Cys-53 (Cys-49 in Fig. 1B). Cys-49 is conserved in typical R2R3 MYB domains, but it is not present in the atypical R2R3 MYB domains or in the MYB domains of pc-MYB proteins (Fig. 1B). We previously investigated the REDOX requirements of the P1 protein for DNA-binding. P1 encodes a typical R2R3 MYB transcriptional regulator of genes encoding biosynthetic enzymes of a branch of the maize flavonoid pathway (19Grotewold E. Drummond B.J. Bowen B. Peterson T. Cell. 1994; 76: 543-553Abstract Full Text PDF PubMed Scopus (535) Google Scholar). P1 activates transcription of the a1 gene by binding to the high and low affinity P1-binding sites present in the a1 promoter (19Grotewold E. Drummond B.J. Bowen B. Peterson T. Cell. 1994; 76: 543-553Abstract Full Text PDF PubMed Scopus (535) Google Scholar, 20Sainz M.B. Grotewold E. Chandler V.L. Plant Cell. 1997; 9: 611-625PubMed Google Scholar). Similar to the DNA-binding activity of c-MYB, P1 requires a strong reducing environment to bind DNA (13Williams C.E. Grotewold E. J. Biol. Chem. 1997; 272: 563-571Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar). As is the case with other R2R3 MYB proteins, P1 contains two Cys residues in the MYB domain (Cys-49 and Cys-53, Fig. 1B) in addition to two cysteines in the short N-terminal region that precedes the MYB domain (21Grotewold E. Athma P. Peterson T. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 4587-4591Crossref PubMed Scopus (204) Google Scholar). In this study, we investigated the participation of the two MYB domain cysteines in the REDOX regulation of the DNA-binding activity of typical R2R3 MYB domains exemplified by the maize P1 transcription factor. We show that the conserved cysteine at position 53 (Cys-53) is not essential for the DNA-binding activity of P1, in sharp departure from what has been established for vertebrate MYB proteins. Instead, Cys-49 plays a much more important role in sensing the REDOX conditions, when Cys-53 is replaced by Ser. However, in the C53A mutant, the DNA-binding activity of P1 becomes insensitive to oxidizing conditions. In vivo, the replacements of Cys-53 to Ser or Ala have no effect on the transcriptional activity of P1. The replacement of Cys-49 by Ser or Ala results in P1 proteins insensitive to the REDOX environment for DNA binding. We also show that Cys-49 and Cys-53 form an intramolecular disulfide bond in non-reducing conditions, providing an evolutionary new opportunity for the modulation of the DNA-binding activity of R2R3 MYB-domain proteins. Together, our studies suggest a novel mechanism for the REDOX control of DNA-binding by R2R3 MYB domains and suggest fundamental structural differences between plant and animal MYB domains. Cloning, Expression, and Purification of Recombinant Proteins—The P1 MYB domain (amino acids 1-118) was previously cloned as a N-terminal poly-histidine fusion (N10His-PMYB) (13Williams C.E. Grotewold E. J. Biol. Chem. 1997; 272: 563-571Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar). For the truncated version of the P1 MYB domain (N6His-PMYBΔ9), the coding region corresponding to amino acids 10-118 was generated by PCR from the full-length cDNA of P1 (21Grotewold E. Athma P. Peterson T. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 4587-4591Crossref PubMed Scopus (204) Google Scholar). The resulting PCR product was then introduced into the pKM260 (22Melcher K. Anal. Biochem. 2000; 277: 109-120Crossref PubMed Scopus (65) Google Scholar) Escherichia coli expression vector via NcoI and BamHI sites to generate N6His-PMYBΔ9. The amino acid sequence of N6His-PMYBΔ9 N-terminal to the MYB-domain sequence shown in Fig. 1 is NH2-MHHHHHHHASENLYFQGAM and the sequence C-terminal to the MYB domain is PAANKARKEAELAATAEQCOO-. All point mutations were generated by site-directed mutagenesis (QuikChange mutagenesis kit; Stratagene) of the PMYBΔ9 coding sequence in the pTAdv (Clontech) cloning vector. The MYB domain was then excised from the pTAdv vector and inserted into pKM260 E. coli expression vector through NcoI and BamHI sites. For expression in E. coli, BL21 (DE3) PlyS cells bearing the corresponding plasmids were grown, induced, and purified essentially as described previously (13Williams C.E. Grotewold E. J. Biol. Chem. 1997; 272: 563-571Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar), with the following modifications. After induction of a 1-L culture, the cells were harvested by centrifugation and stored at -80 °C until further use. The cells were resuspended in 20 ml SB (50 mm sodium phosphate, pH 8.0, 100 mm NaCl, and 100 μg/ml phenylmethylsulfonyl fluoride) and passed twice through a French press. The cell lysate was centrifuged at 14,000 × g for 20 min, and the supernatant was filtered through two layers of Mira-cloth (Calbiochem). The nickel-nitrilotriacetic acid agarose resin (Qiagen) was equilibrated with SB, and 1 ml of 50% slurry was added to the cell lysate supernatant and incubated for 2 h with rocking at 4 °C. The resin was gently recovered by centrifugation, re-suspended in 5 ml of SB, and loaded onto a column. The column was washed five times with 5 column volumes of SB and three times with 5 column volumes of WB (50 mm sodium phosphate, pH 8.0, 300 mm NaCl, 1% Tween 20, 5 mm 2-mercaptoethanol, 10 mm EDTA, and 10% glycerol). The protein was eluted with five washes of 5 column volumes of WB containing 50 mm imidazole. The elutions were then dialyzed against 60 volumes of A-0 buffer (10 mm Tris pH 7.5, 50 mm NaCl, 1 mm DTT, 1The abbreviations used are: DTT, dithiothreitol; APB1, high affinity P1-binding sites from the a1 gene promoter; APB5, a mutant in which the P1-binding sites were destroyed; GUS, β-glucoronidase; DCIA, 7-diethylamino-3-(4′-iodoacetylaminophenyl)-4-methylcoumarin; EMSA, electrophoretic mobility shift assay; WT, wild type. 1 mm EDTA, and 5% glycerol) and stored at -80 °C until further use. Each wash and elution fraction was collected and analyzed by SDS-PAGE, stained with Coomassie Brilliant Blue R-250, and quantified against a lysozyme protein standard. Purified proteins were used for each assay and were judged to be >90% pure by Coomassie Brilliant Blue staining of 15% SDS-PAGE gels. Electrophoretic Mobility Shift Assays—Aliquots of purified proteins were incubated with end-labeled, double-stranded DNA generated by hybridization of complementary oligonucleotides. End labeling of synthetic oligonucleotide probes was carried out using T4-polynucleotide kinase (Invitrogen) in the presence of a 2 molar excess of [γ-32P]ATP (>8,000 Ci/mmol; ICN). The labeled oligonucleotides were then annealed to an equimolar amount of complementary oligonucleotides by heating to 95 °C and cooling to room temperature. A fraction of the double-stranded labeled oligonucleotides was precipitated on glass filters for quantification by scintillation of the radiation incorporated. The probes used contain APB1, the high affinity P1-binding sites from the a1 gene promoter, or APB5, a mutant in which the sites were destroyed (Fig. 2A) (19Grotewold E. Drummond B.J. Bowen B. Peterson T. Cell. 1994; 76: 543-553Abstract Full Text PDF PubMed Scopus (535) Google Scholar). Protein-DNA incubations were performed essentially as described previously (19Grotewold E. Drummond B.J. Bowen B. Peterson T. Cell. 1994; 76: 543-553Abstract Full Text PDF PubMed Scopus (535) Google Scholar) with the following modifications. Approximately 35 ng of protein were incubated in A-0 buffer (10 mm Tris, pH 7.5, 50 mm NaCl, 1 mm EDTA, and 5% glycerol) with 0.8 μg of poly d(I)/d(C), and 1 mm DTT, unless indicated otherwise. Protein-DNA complexes were resolved on a 8% polyacrylamide gels (80:1 acrylamide:bis-acrylamide) in 0.25× Tris-borate/EDTA (22.5 mm Tris-Borate and 0.5 mm EDTA) running buffer at 415 V for 55 min at 4 °C. The gels were then dried onto Whatman paper and subjected to autoradiography at -70 °C overnight. Tryptophan Fluorescence—Fluorescence experiments were carried out in a Horbia Fluoromax-3 Luminescence Spectrometer. An excitation wavelength of 295 nm was used with an emission and excitation slit width of 5 nm. Emission spectra were recorded at 0.5-nm intervals between 310 and 400 nm with a scanning speed of 300 nm/min. Samples were analyzed in a quartz cuvette (1.0 × 1.0 cm) and contained 0.5 μm purified recombinant proteins in 3 ml of A-0 buffer. To determine the fluorescence spectra of the proteins in denaturing conditions, the purified recombinant proteins were incubated with 6 m guanidinium hydrochloride in A-0 buffer at room temperature for 1 h before the fluorescence analysis. Each spectrum was combined as an average of five scans and corrected to the background of the A-0 buffer, with or without guanidinium hydrochloride. DNA Constructs for Transient Expression Experiments—The p35SP1 construct was described previously (23Grotewold E. Sainz M.B. Tagliani L. Hernandez J.M. Bowen B. Chandler V.L. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 13579-13584Crossref PubMed Scopus (267) Google Scholar). Other previously described constructs include pA1Luc, which contains 220 bp of the a1 promoter (19Grotewold E. Drummond B.J. Bowen B. Peterson T. Cell. 1994; 76: 543-553Abstract Full Text PDF PubMed Scopus (535) Google Scholar, 20Sainz M.B. Grotewold E. Chandler V.L. Plant Cell. 1997; 9: 611-625PubMed Google Scholar), p35SBAR (24Grotewold E. Chamberlin M. Snook M. Siame B. Butler L. Swenson J. Maddock S. Clair G.S. Bowen B. Plant Cell. 1998; 10: 721-740Crossref PubMed Scopus (285) Google Scholar), and pUbiGUS (19Grotewold E. Drummond B.J. Bowen B. Peterson T. Cell. 1994; 76: 543-553Abstract Full Text PDF PubMed Scopus (535) Google Scholar). p35SBAR was used for normalizing the concentration of CaMV35S promoter sequences delivered in each bombardment, and the pUbiGUS was used to normalize the efficiency of each bombardment. Mutants in the MYB domain of P1 were generated by PCR from their respective E. coli protein expression vectors and, after sequencing, were used to replace the MYB domain of P1 in the p35SP1 construct by digesting it with BamHI and SnaBI. Microprojectile Bombardment and Gene Expression—Bombardment conditions of suspension maize Black Mexican Sweet cells and transient expression assays for luciferase and GUS were performed essentially as described previously (23Grotewold E. Sainz M.B. Tagliani L. Hernandez J.M. Bowen B. Chandler V.L. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 13579-13584Crossref PubMed Scopus (267) Google Scholar). Bombardments were performed in triplicate, and each experiment was repeated at least twice. The assays for luciferase and GUS and the normalization of the data were performed as described previously (19Grotewold E. Drummond B.J. Bowen B. Peterson T. Cell. 1994; 76: 543-553Abstract Full Text PDF PubMed Scopus (535) Google Scholar). The -fold activation results are expressed as the ratio of arbitrary light units (luciferase) to arbitrary fluorescence units (GUS) of the treatment, with the transcriptional activator divided by the ratio of arbitrary light units (luciferase) to arbitrary fluorescence units (GUS) of the reporter plasmid in the absence of the regulator. Detection of Dimers and Fluorescent Labeling of Recombinant Proteins—Detection of dimer formation was accomplished by incubating 700 ng of recombinant protein on ice for 1 h in the presence of either 10 mm DTT or 30 mm diamide. After incubation in ice, 2× SDS-PAGE loading buffer without reducing agents (100 mm Tris-HCl, pH 6.8, 4% SDS, 0.2% bromphenol blue, and 20% glycerol) was added to each sample and separated by electrophoresis on a 15% non-reducing SDS-polyacrylamide gel. Fluorescent labeling of reduced cysteine residues was accomplished by using the thiol specific modifying agent 7-diethylamino-3-(4′-iodoacetylaminophenyl)-4-methylcoumarin (DCIA; Molecular Probes). Approximately 350 ng of recombinant protein was incubated on ice in either the presence or absence of the reducing agent tris-(2-carboxyethyl) phosphine hydrochloride (Molecular Probes). Samples in reducing conditions (1 mm tris-(2-carboxyethyl) phosphine hydrochloride) were incubated on ice for 30 min. DCIA was then added to all samples to a final concentration of 1 mm and incubated on ice for additional 30 min in the dark. After incubation with DCIA, tris-(2-carboxyethyl) phosphine hydrochloride was added to a final concentration of 3 mm in all samples. Samples were then mixed with 2× non-reducing SDS-PAGE loading buffer and separated on a 15% non-reducing SDS-polyacrylamide gel. After separation, fluorescent imaging of the gel was accomplished using the Bio-RAD Gel Doc 2000 Documentation system. After fluorescent analysis, the gel was stained with Coomassie Brilliant Blue. The MYB Domain of P1 Binds DNA Only When Reduced—We have shown previously that the MYB domain of P1 binds DNA only in the presence of DTT (13Williams C.E. Grotewold E. J. Biol. Chem. 1997; 272: 563-571Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar). The MYB domain of P1 used in those studies (N10His-PMYB) contained four Cys residues, two at positions six and seven before the first helix of R2, and two others at positions 49 and 53 (Fig. 1B). To investigate which Cys residues in P1MYB respond to the reducing conditions, we generated a truncated, polyhistidine-tagged version of P1MYB in which the first nine amino acids of P1 were deleted (N6His-PMYBΔ9), removing the two Cys residues located in the N-terminal extension (21Grotewold E. Athma P. Peterson T. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 4587-4591Crossref PubMed Scopus (204) Google Scholar). To determine whether N6His-PMYBΔ9 bound DNA in a manner similar to that of the previously described N10His-PMYB protein (13Williams C.E. Grotewold E. J. Biol. Chem. 1997; 272: 563-571Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar), we tested total E. coli extracts, expressing equivalent amounts of the two proteins, by electrophoretic mobility shift assays (EMSA), using the APB1 probe (Fig. 2A). APB1 corresponds to the high-affinity P1-binding sites present in the promoter of a1, one of the flavonoid biosynthetic genes regulated by P1 (19Grotewold E. Drummond B.J. Bowen B. Peterson T. Cell. 1994; 76: 543-553Abstract Full Text PDF PubMed Scopus (535) Google Scholar). N6His-PMYBΔ9 binds APB1 as effectively as N10His-PMYB (Fig. 2B), indicating that the first nine amino acids of P1 are dispensable for the DNA-binding activity of P1. The faster mobility of the N6His-PMYBΔ9-APB1 complex reflects the smaller size of N6His-PMYBΔ9, compared with N10His-PMYB. As previously shown for P1 and N10His-PMYB (13Williams C.E. Grotewold E. J. Biol. Chem. 1997; 272: 563-571Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar, 19Grotewold E. Drummond B.J. Bowen B. Peterson T. Cell. 1994; 76: 543-553Abstract Full Text PDF PubMed Scopus (535) Google Scholar), N6His-PMYBΔ9 does not bind to the mutant APB5 DNA probe (Fig. 2D, lane 1) Similar to N10His-PMYB, the DNA-binding activity of N6His-PMYBΔ9 depends on the presence of DTT (Fig. 2C, compare lanes 1 and 4). These results suggest that Cys-49 and/or Cys-53 are responsible for the reducing conditions necessary for P1 to bind DNA. To determine the participation of Cys-53 in the REDOX regulation of DNA binding by P1, we replaced Cys-53 with Ala or Ser in N6His-PMYBΔ9. The corresponding proteins (C53A and C53S respectively) were expressed in E. coli, affinity-purified on a nickel-nitrilotriacetic acid column and analyzed for DNA-binding activity to the APB1 probe by EMSA. In the presence of reducing conditions (1 mm DTT), wild type (WT, N6His-PMYBΔ9) and C53A bound with comparable strengths (Fig. 2C, lanes 1 and 2). In the presence of DTT (Fig. 2C, +DTT) the binding of C53S to APB1 was also comparable with WT (Fig. 2C, lane 3). In the absence of DTT, however (Fig. 2C, -DTT), only C53A bound APB1 (Fig. 2C, lane 5); neither the wild type nor the C53S proteins showed any significant DNA binding. To determine whether the C53A or C53S mutations affected the sequence-specific DNA-binding activity of P1 (13Williams C.E. Grotewold E. J. Biol. Chem. 1997; 272: 563-571Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar, 19Grotewold E. Drummond B.J. Bowen B. Peterson T. Cell. 1994; 76: 543-553Abstract Full Text PDF PubMed Scopus (535) Google Scholar), we tested wild type and the two mutants for binding to the APB5 probe. As shown in Fig. 2D, none of these proteins bound to the mutant sites. Next, we investigated the effect that the addition of the sulfhydryl oxidizing agent diamide had on the DNA-binding activity of these proteins. The DNA-binding activity of the wild type protein (N6His-PMYBΔ9) was abolished by the presence of 3 mm diamide (Fig. 2E, lane 2). The DNA-binding activity of this protein, however, is recovered by the addition of excess DTT (Fig. 2E, lane 3). The C53A protein (Fig. 2E, lanes 4-6) was not significantly affected by the treatment with diamide. Similar to wild type, binding of C53S to APB1 was completely abolished by the presence of diamide (Fig. 2E, lane 8), yet recovered when DTT was added (Fig. 2E, lane 9). To better understand how these mutations affect the ability of the MYB domain of P1 to bind DNA, we made use of the intrinsic fluorescent properties of the five tryptophan residues present in R2R3 MYB domains (Fig. 1B). Using fluorescence spectroscopy, we compared the environments of the tryptophan residues in the wild type (WT, N6His-PMYBΔ9), C53A and C53S proteins. The fluorescence emission spectra for each of the three proteins was determined in 0 mm DTT, 1 mm DTT, and 6 m guanidinium chloride (denaturing conditions), and the wavelength of the maximum emission fluorescence was recorded (Table I). All the denatured proteins had similar emission maxima, between 357 and 359 nm, similar to the maximum of free tryptophan (360 nm) (Table I). Compared with the denatured protein, the emission maximum of the native wild type protein (in the presence of 1 mm DTT) was shifted toward the shorter wavelengths (342 nm), a shift indicative of the tryptophan residues being less exposed to the solvent, probably participating in the formation of a hydrophobic core, as proposed for c-MYB (1Kanei-Ishii C. Sarai A. Sawazaki T. Nakagoshi H. He D.N. Ogata K. Nishimura Y. Ishii S. J. Biol. Chem. 1990; 265: 19990-19995Abstract Full Text PDF PubMed Google Scholar, 2Ogata K. Hojo H. Aimoto S. Nakai T. Nakamura H. Sarai A. Ishii S. Nishimura Y. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 6428-6432Crossref PubMed Scopus (217) Google Scholar, 3Saikumar P. Murali R. Reddy E.P. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 8452-8456Crossref PubMed Scopus (104) Google Scholar, 4Ogata K. Morikawa S. Nakamura H. Sekikawa A. Inoue T. Kanai H. Sarai A. Ishii S. Nishimura Y. Cell. 1994; 79: 639-648Abstract Full Text PDF PubMed Scopus (439) Google Scholar). In the absence of DTT, the wild type protein displayed a more open conformation (Table I, 352 nm), compared with the reduced form (342 nm). C53A had a similar emission maximum in the absence of DTT as the wild type protein in reducing conditions (Table I, compare 343 nm with 342 nm). The addition of 1 mm DTT shifted the emission maximum of C53A slightly further toward the blue (Table I, 338 nm). This result is consistent with the observation that C53A binds efficiently DNA in both reducing and nonreducing conditions (Fig. 2C, lanes 2 and 5). In contrast, the C53S mutant had a similar emission maximum as wild type in non-reducing conditions (Table I, compare 350 nm with 352 nm). In the presence of 1 mm DTT, the maximum of C53S shifted toward the blue to 345 nm, consistent with the ability of this protein to bind DNA in reducing conditions (Fig. 2C, lanes 3 and 6).Table IWavelengthh of maximum fluorescence obtained from tryptophan fluorescence emission spectra of the truncated wild-type P1 MYB domain (WT, N6His-PMYBΔ9) and single cysteine mutants in the presence of either 0 mm DTT, 1 mm DT, or 6 m guanidine HClTreatmentWTC53AC53SC49AC49SC49Inm0 mm DTT3523433503423373501 mm DTT3423383453413363506 m G-HCl358357358358357357 Open table in a new tab The ability of the C53S mutant to bind DNA only in the presence of DTT suggested that Cys-53 is not the only cysteine residue involved in the REDOX-dependent DNA-binding activity of P1. Because C53S has only one additional cysteine residue (Cys-49, Fig. 1B), Cys-49, or t