Structural and functional analysis of VQ motif-containing proteins in Arabidopsis as interacting proteins of WRKY transcription factors

Abstract

WRKY transcription factors are encoded by a large gene superfamily with a broad range of roles in plants. Recently, several groups have reported that proteins containing a short VQ (FxxxVQxLTG) motif interact with WRKY proteins. We have recently discovered that two VQ proteins from Arabidopsis (Arabidopsis thaliana), SIGMA FACTOR-INTERACTING PROTEIN1 and SIGMA FACTOR-INTERACTING PROTEIN2, act as coactivators of WRKY33 in plant defense by specifically recognizing the C-terminal WRKY domain and stimulating the DNA-binding activity of WRKY33. In this study, we have analyzed the entire family of 34 structurally divergent VQ proteins from Arabidopsis. Yeast (Saccharomyces cerevisiae) two-hybrid assays showed that Arabidopsis VQ proteins interacted specifically with the C-terminal WRKY domains of group I and the sole WRKY domains of group IIc WRKY proteins. Using site-directed mutagenesis, we identified structural features of these two closely related groups of WRKY domains that are critical for interaction with VQ proteins. Quantitative reverse transcription polymerase chain reaction revealed that expression of a majority of Arabidopsis VQ genes was responsive to pathogen infection and salicylic acid treatment. Functional analysis using both knockout mutants and overexpression lines revealed strong phenotypes in growth, development, and susceptibility to pathogen infection. Altered phenotypes were substantially enhanced through cooverexpression of genes encoding interacting VQ and WRKY proteins. These findings indicate that VQ proteins play an important role in plant growth, development, and response to environmental conditions, most likely by acting as cofactors of group I and IIc WRKY transcription factors.

Figure 1.

Alignment of VQ motif sequences of Arabidopsis proteins. The highly conserved residues in the VQ motif are indicated in red.

Figure 2.

Phylogenetic trees of the VQ proteins from Arabidopsis. The tree was inferred using the neighbor-joining method. Phylogenetic analyses were conducted in MEGA5. Bootstrap values from 1,000 replicates were used to assess the robustness of the tree.

Figure 3.

Interaction of Arabidopsis VQ proteins with WRKY proteins in yeast cells. The Gal4 DNA BD-WRKY domain fusion bait vectors were contransformed with the activation domain (AD)-VQ fusion prey vectors into yeast cells and the transformant cells were assayed for LacZ reporter gene expression. The empty pAD prey vector was used as negative control (−).

Figure 4.

Identification of critical amino acid residues of C-terminal WRKY domain of WRKY33 for interaction with VQ10. A, Sequence comparison of WRKY domains of WRKY33 (W33), WRKY18 (W18), WRKY6 (W6), WRKY51 (W51), WRKY11 (W11), WRKY22 (W22), WRKY38 (W38), which belong to group I, IIa, IIb, IIc, IId, IIe, and III WRKY proteins, respectively. Mutant C-terminal WRKY domains of WRKY33 (W33CT) with D395A, D362A, or R366Q substitution or with an Ala inserted between the two zinc-finger Cys residues (W33CTIns) are also shown. The WRKYGQK sequences and the Cys (C) and His (H) residues involved in zinc coordination are indicated in blue. The two Asp (D) residues and one basic residue in the region preceding the WRKYGQK sequence shared by the W33CT and W51 WRKY domains are in red. Resulted amino acid residues in the mutant W33CT proteins are in green. B, Yeast two-hybrid assays of interactions of VQ10 with wild-type and mutant forms of the C-terminal WRKY domains of WRKY33. pAD-VQ10 fusion vector was cotransformed with various pBD-W33CT constructs into yeast cells. Yeast transformants were analyzed for the LacZ reporter gene expression through assays of β-galactosidase activity on membrane using 5-bromo-4-chloro-3-indoly1-β-d-galactopyranoside (left section) or o-nitrophenyl-β-d-galactopyranose (right section) as substrate.

Figure 5.

Cluster analysis of the expression profiles of Arabidopsis VQ genes. VQ gene expression was determined by qRT-PCR. VQ genes are represented as rows and treatment/time point as columns in the matrix. Red, black, and green elements indicate up-regulated, no change, and down-regulated VQ genes, respectively, in the matrix. Cluster analysis was performed on expression profiles of 34 VQ genes across six transients treatments (P, P. syringae infection; S, SA treatment) and time points (4, 12, and 24 h after pathogen inoculation or SA treatment). The horizontal and vertical dendrograms indicate the degree of similarity between expression profiles for VQ genes and conditions tested, respectively.

Figure 6.

W-box elements in the promoter of Arabidopsis VQ genes. Numbering is from predicted translation start codons.

Figure 7.

Identification and characterization of Arabidopsis vq8-1 mutant. A, The structure of the intronless VQ8 gene. The transposon insertion site of the vq8-1 mutant is indicated. P1 and P2 are the two primers used for identification of homozygous vq8-1 mutant plants. B, Genetic analysis of vq8-1. Col-0 wild type (WT) was crossed with the vq8-1 and the F2 progeny were screened for homozygous vq8 progeny and scored for the absence (−) or presence (+) of its mutant phenotypes. C, The phenotypes of the vq8-1 mutant. The picture was taken at 7 weeks after germination.

Figure 1.

Growth and developmental phenotypes of transgenic VQ-overexpressing Arabidopsis plants. A, Stunted growth of transgenic plants overexpressing VQ17, VQ18, or VQ22. The picture of Col-0 wild type (WT) and two lines (L) of transgenic overexpression plants for each VQ gene was taken 7 weeks after germination. B, Delayed flowering of transgenic plants overexpressing VQ29. The picture of Col-0 wild type (WT) and two lines (L) of transgenic VQ29 overexpression plants was taken 10 weeks after germination. The RNA-blotting analysis of the overexpressed VQ genes in the transgenic lines is shown in Supplemental Figure S2S.

Figure 9.

Altered disease resistance of transgenic VQ-overexpressing Arabidopsis plants. A, Enhanced susceptibility to B. cinerea. Col-0 wild type (WT) and two independent lines (L) of transgenic plants overexpressing VQ5 or VQ20 were spray inoculated with Botrytis and the picture was taken at the third d post inoculation (dpi). B, Enhanced susceptibility to P. syringae. Col-0 wild type (WT) and transgenic plants overexpressing VQ20 (lines 12 and 13) or VQ25 (lines 7 and 8) were inoculated with a virulent strain of P. syringae pv tomato DC3000 (OD₆₀₀= 0.0002 in 10 mm MgCl₂). Samples were taken at 0 and 3 dpi to determine the growth of the bacterial pathogen (top section). According to Duncan’s multiple range test (P= 0.01), means of colony-forming units (cfu) do not differ significantly at 0 dpi if they are indicated with the same lowercase letter and do not differ significantly at 2 dpi if they are indicated with the same uppercase letter. Images from representative inoculated leaves taken at 3 dpi are also shown (bottom section). The RNA-blotting analysis of the overexpressed VQ genes in the transgenic lines is shown in Supplemental Figure S2S.

Figure 10.

Enhanced phenotypes in transgenic plants cooverexpressing interacting VQ10 and WRKY genes. Transgenic plants overexpressing VQ10 (line 1) were crossed with transgenic plants overexpressing WRKY25 (line 1), WRKY26 (line 1), or WRKY33 (line 1). The F1 progeny of the crossing were compared with Col-0 and their parental lines for growth alteration. The images for the plants were taken 6 weeks after germination.