Structural insights into target DNA recognition by R2R3-MYB transcription factors

Abstract

As the largest group of MYB family transcription factors, R2R3-MYB proteins play essential roles during plant growth and development. However, the structural basis underlying how R2R3-MYBs recognize the target DNA remains elusive. Here, we report the crystal structure of Arabidopsis WEREWOLF (WER), an R2R3-MYB protein, in complex with its target DNA. Structural analysis showed that the third α-helices in both the R2 and R3 repeats of WER fit in the major groove of the DNA, specifically recognizing the DNA motif 5'-AACNGC-3'. In combination with mutagenesis, in vitro binding and in vivo luciferase assays, we showed that K55, N106, K109 and N110 are critical for the function of WER. Although L59 of WER is not involved in DNA binding in the structure, ITC analysis suggested that L59 plays an important role in sensing DNA methylation at the fifth position of cytosine (5mC). Like 5mC, methylation at the sixth position of adenine (6mA) in the AAC element also inhibits the interaction between WER and its target DNA. Our study not only unravels the molecular basis of how WER recognizes its target DNA, but also suggests that 5mC and 6mA modifications may block the interaction between R2R3-MYB transcription factors and their target genes.

Figure 1.

DNA motifs recognized by R2R3-MYB proteins in Arabidopsis. (A) Sequences of R2R3-MYB target motifs that have been verified by in vitro and in vivo assays. (B) Sequence alignment of typical R2R3-MYB proteins. The secondary structure of WER was predicted by Phyre² program (http://www.sbg.bio.ic.ac.uk/phyre2). (C) EMSA (Electrophoretic Mobility Shift Assay) assay showing the interaction between WER and its target DNA (5′-AAATTCTCCAACCGCATTTTC-3′, 5′-GAAAATGCGGTTGGAGAATTT-3′). The DNA concentration was fixed at 0.1 μM and the protein concentration was increased from 0 to 0.8 μM. (D) ITC (Isothermal Titration Calorimetry) experiment measuring the binding affinity between wild-type WER and its target DNA (5′-AAATTCTCCAACCGCATTTTC-3′, 5′-GAAAATGCGGTTGGAGAATTT-3′).

Figure 2.

Structure of the WER–DNA complex. (A) The overall folding of WER–DNA complex. DNAs are shown as sticks. In the left and right panels, WER-R2R3 is shown in cartoon and as an electrostatic surface potential map, respectively. The 2F_o-F_c electron density map was contoured at the 1.0 σ level. (B) Close-up view showing the relative orientations of the H3 and H6 helices of WER-R2R3 and DNA. The helices are shown in cartoon-and-stick form. The DNA is shown as spheres. (C–G) Sequence-specific recognition of A10:T13*, A11:T12*, C12:G11*, G14:C9* and C15:G8*, which correspond to the first, second, third, fifth, and sixth base pairs of the DNA, respectively. DNA base pairs and WER residues responsible for base recognition are shown as sticks. The C-atoms of R2 and R3 residues are magenta and cyan, respectively. Water molecules are shown as red spheres. The distances of the direct or water-mediated H-bond interactions are indicated by numbers.

Figure 3.

Verification of sequence-specific interactions. (A) ITC experiments showing the impact of mutation of WER residues involved in dsDNA (5′-AAATTCTCCAACCGCATTTTC-3′, 5′-GAAAATGCGGTTGGAGAATTT-3′) binding. (B) ITC experiments showing the impact of DNA core motif (5′-AACCGC-3′) mutation on WER binding. The detailed sequence of the target DNA is shown on the top of the table. For mutated DNAs, only the mutated base pairs are listed in the table for clarity. (C) Dual luciferase assay of the GL2 promoter activation activity of wild-type and mutated WER proteins. Values are means ± SD of three independent biological replicates. * and *** indicate statistically significant differences between the wild type (WT) and mutant at P < 0.05 and P < 0.001, respectively. (D) ITC experiments showing the binding affinities of WER L59A mutant to the A12:T11*, G12:C11* and T12:A11* mutants of the 5′-AACCGC-3′ motif, respectively.

Figure 4.

R2 repeat is incompatible with DNA 5mC modification. (A) The interactions between L59/E132 and the C5 atom of cytosine in WER–DNA/MsMyb–DNA (PDB code: 1H8A) complexes. (B) ITC analysis showing the impact of 5mC modification on DNA binding by WER and its mutants. (C) Consensus sequence and conservation analysis of the R2 motif involved in DNA recognition of MYB family proteins. Plant and animal MYB proteins are shown in the upper and lower panels, respectively.

Figure 5.

R3 repeat is incompatible with DNA 6mA modification. (A) The interactions between R3 Asn residues and the AA element in WER–DNA and MsMyb–DNA (PDB code: 1H8A) complexes. ITC analysis shows the impact of 6mA modification on DNA binding by (B) WER and (C) MsMyb, respectively. (D) Consensus sequence and conservation analysis of the R3 motif involved in DNA recognition of MYB domain proteins. Plant and animal MYB proteins are shown in the upper and lower panels, respectively. (E) The potential target motif AACNDN (D: A or T or G) recognized by WER can be methylated in vivo by searching the Arabidopsis DNA methylation databases (GSM2807190 for 5mC, GSM2157793 for 6mA).