Alternative Splicing of Rice WRKY62 and WRKY76 Transcription Factor Genes in Pathogen Defense

Abstract

The WRKY family of transcription factors (TFs) functions as transcriptional activators or repressors in various signaling pathways. In this study, we discovered that OsWRKY62 and OsWRKY76, two genes of the WRKY IIa subfamily, undergo constitutive and inducible alternative splicing. The full-length OsWRKY62.1 and OsWRKY76.1 proteins formed homocomplexes and heterocomplexes, and the heterocomplex dominates in the nuclei when analyzed in Nicotiana benthamiana leaves. Transgenic overexpression of OsWRKY62.1 and OsWRKY76.1 in rice (Oryza sativa) enhanced plant susceptibility to the blast fungus Magnaporthe oryzae and the leaf blight bacterium Xanthomonas oryzae pv oryzae, whereas RNA interference and loss-of-function knockout plants exhibited elevated resistance. The dsOW62/76 and knockout lines of OsWRKY62 and OsWRKY76 also showed greatly increased expression of defense-related genes and the accumulation of phytoalexins. The ratio of full-length versus truncated transcripts changed in dsOW62/76 plants as well as in response to pathogen infection. The short alternative OsWRKY62.2 and OsWRKY76.2 isoforms could interact with each other and with full-length proteins. OsWRKY62.2 showed a reduced repressor activity in planta, and two sequence determinants required for the repressor activity were identified in the amino terminus of OsWRKY62.1. The amino termini of OsWRKY62 and OsWRKY76 splice variants also showed reduced binding to the canonical W box motif. These results not only enhance our understanding of the DNA-binding property, the repressor sequence motifs, and the negative feedback regulation of the IIa subfamily of WRKYs but also provide evidence for alternative splicing of WRKY TFs during the plant defense response.

Figure 1.

Homocomplex and heterocomplex formation of OsWRKY62.1 and OsWRKY76.1. A, Schematic diagrams of OsWRKY62.1 (OW62.1) and OsWRKY76.1 (OW76.1) proteins. Gray and black boxes represent the predicted CC and WRKY DNA-binding domains, respectively (http://smart.embl-heidelberg.de/smart/show_motifs.pl). The numbers are the amino acid positions in the full-length proteins. B, Analyses of OsWRKY62.1 and OsWRKY76.1 interactions in yeast. OW62.1, OW76.1, and their deletion mutants (OW62ΔN, OW62ΔC, OW76ΔN, and OW76ΔC) were fused to the Gal4 DNA-binding domain (BD) and/or activation domain (AD). Yeast cells with serial dilutions (1, 1/10, and 1/100) were incubated in synthetic dropout (SD) medium lacking Leu and Trp (-LW; left) or Leu, Trp, His, and adenine (-LWHA; right) and photographed 3 d after plating. Yeast cells harboring AD-T with BD-53 or BD-Lam vectors were used as the positive or negative control. C, Pull-down assays of OsWRKY62.1 and OsWRKY76.1 interactions. OW62.1 and OW76.1 were purified with their N termini fused with 6×His and their C termini fused with 3×flag or 3×myc tag. Each protein (about 1 μg) with the combinations indicated was incubated at 4°C for 3 h in the immunoprecipitation buffer. The protein complexes were precipitated with EZview Red Anti-c-Myc Affinity Gel, separated on 10% SDS-PAGE gels, and detected with the antibodies indicated. Asterisks indicate the OW76.1 protein. D, BiFC assay of OsWRKY62.1 and OsWRKY76.1 interactions. OW62.1 and OW76.1 were fused in frame with the YFP C-terminal region (YFP^C) or the YFP N-terminal region (YFP^N). The plasmids indicated were introduced into the leaf cells of N. benthamiana through agroinfiltration. Confocal images were taken at 72 h after infiltration. From left to right are YFP images, 4′,6-diamino-phenylindole (DAPI) nuclear staining images, bright-field images (differential interference contrast [DIC]), and merged YFP, DAPI, and DIC images. Bars = 20 μm.

Figure 2.

Expression profiles of the OsWRKY62 and OsWRKY76 genes in response to chitin, flg22, MeJA, and wounding. A to E, Rice seedlings were treated with 0.7 μm flg22, 10 nm chitin, or 100 μm MeJA in 10 mm MES (pH 6) buffer and sampled at the indicated time points. Mechanical wounding was performed on leaves of 3-week-old plants by crushing with a hemostat. F, Transcription levels expressed as the ratio to the level of transcript at 0 h using the rice UBIQUITIN (OsUBQ) gene as an internal standard. Values shown are means ± se of three replicates. Statistically significance changes are indicated by asterisks: *, P < 0.05 and **, P < 0.01. CK, Mock treatment.

Figure 3.

Histochemical analyses of OsWRKY62 and OsWRKY76 expression. OsWRKY62.1 and OsWRKY76.1 promoters were fused with the GUS gene, and the resultant plasmids (Cp-OW62p:Gus and Cp-OW76p:Gus) were transformed into rice calli. The T2 progeny of Cp-OW62p:Gus and Cp-OW76p:Gus were used for GUS activity staining. Seven-day-old seedlings (A), primary and lateral roots of 12-d-old plants (B), stem and the third node counted from the top at the reproductive stage (C), flowers (D), ligule base and leaf sheath at the reproductive stage (E), and immature seed (F) are shown. An image or a group of three images are shown from left to right as Cp-OW76p:Gus, wild-type Zhonghua 17 (ZH17), and Cp-OW62p:Gus. Bars = 1 mm (D), 5 mm (A, B, E, and F), and 1 cm (C).

Figure 4.

OsWRKY62.1 and OsWRKY76.1 play negative roles in resistance against rice blast and bacterial blight pathogens. A, Three-week old transgenic (T2 progeny) and wild-type (ZH17) plants were inoculated with M. oryzae SZ (5 × 10⁵ conidia mL⁻¹) by foliar spraying. Photographs were taken 5 d after inoculation. B, Lesion areas are from inoculated leaves of 10 plants for each line. C, Six-week-old plants were challenged with Xoo isolate J18 using the leaf-clipping method. Photographs were taken 2 weeks after inoculation. D, Lesion lengths are averages of 10 leaves. Values marked with different letters indicate statistically significant differences as analyzed by the SAS software (Duncan’s multiple range test, α = 0.05). Results from a representative experiment are shown. The suffix ox is for OsWRKY62.1- and OsWRKY76.1-overexpressing plants, and the prefix ds is for RNAi lines. Bars = 1 cm (A) and 2 cm (C).

Figure 5.

The OsWRKY62/OsWRKY76 double RNAi plants showed spontaneous cell death and enhanced resistance against rice pathogens. A to D, Phenotypes of wild-type ZH17 (left) and the double RNAi line (dsOW62/76; right). A, Lesions on leaves of 3-week-old dsOW62/76 plants (dd108). B, Detection of cell death by Trypan Blue staining. C, Detection of hydrogen peroxide (H₂O₂) accumulation by 3,3′-diaminobenzidine (DAB) staining in the fourth leaves. D, Dwarfism of the dd108 plant. Leaves are counted from bottom to top. E and G, Transgenic (T2 progeny) and ZH17 plants were inoculated with M. oryzae SZ (E) or Xoo J18 (G), and disease was evaluated as described in Figure 4. F, The amount of fungal mass in inoculated leaves was estimated by qRT-PCR. Values are means ± se of six leaves. H, Lesion lengths of transgenic and control plants. I, Total RNA was isolated from leaves of 3-week-old dd108 and control plants. Gene expression was determined by qRT-PCR using OsUBQ as the reference gene. Transcription levels are shown relative to ZH17 rice leaves. Values shown are means ± se of three replicates. In F, H, and I, Differences between ZH17 and transgenic lines are significant (P < 0.01, Student’s t test). Experiments were repeated three times with similar results. Bars = 1 cm (A–C and E), 2 cm (G), and 10 cm (D).

Figure 6.

Changes in alternative transcript levels of OsWRKY62 and OsWRKY76 in the RNAi plants or plants after pathogen infection or MeJA treatment. A and B, Expression of OsWRKY62 and OsWRKY76 was analyzed by qRT-PCR using total RNA extracted from leaves of 3-week-old plants. Error bars indicate se (n = 3). Values marked with different letters indicate statistically significant differences as analyzed by the SAS software (Duncan’s multiple range test, α = 0.05). C, OsWRKY76.1 to OsWRKY76.3 (OW76.1–OW76.3) and OsWRKY62.1 and OsWRKY62.2 (OW62.1 and OW62.2) were obtained by RT-PCR amplification of complementary DNAs (cDNAs) from dd94 and dd108 plants. D to F, Levels of alternative transcripts OsWRKY62.1 (OW62.1) and OsWRKY62.2 (OW62.2) were analyzed by qRT-PCR in transgenic plants (D), after M. oryzae infection (E), or after MeJA treatment (F). Samples were collected from plants at the time points indicated after inoculation with a virulent (SZ) or avirulent (P131) strain of M. oryzae or after treatment with MeJA (0.1 mM, in 10 mM MES buffer, pH 6). CK, Mock treatment. Transcription levels are shown relative to the level of transcript at 0 h or those in wild-type ZH17 using the OsUBQ gene as an internal standard. Values shown are means ± se of three replicates. Statistically significance changes are indicated by asterisks: *, P < 0.05 and **, P < 0.01 (Student’s t test).

Figure 7.

OsWRKY76 and OsWRKY62 knockout plants exhibit enhanced defense responses. A, Levels of alternative transcripts of OsWRKY76 and OsWRKY62 were determined by RT-PCR analysis. The leaf samples were collected from 8-week-old knockout (KO) and regenerated nontransgenic (NT) plants. B, Transcript levels of defense genes were determined by qRT-PCR using rice UBQ as a reference. Expression levels in the knockout lines are shown relative to that in regenerated nontransgenic rice leaves. Values shown are means ± se of three replicates. The amounts of momilactone A, phytocassanes B, and phytocassanes C were determined by liquid chromatography-tandem mass spectrometry using lidocaine as an internal standard. Data are means ± se of two replicates. *, P < 0.05 and **, P < 0.01 (Student’s t test). C, Detached leaves of 8-week-old plants were inoculated with M. oryzae SZ (1 × 10⁵ spores mL⁻¹). Photographs were taken 6 d post inoculation. Lesion lengths are averages of 10 inoculated sections. Values marked with different letters indicate significant differences as analyzed by the SAS software (Duncan’s multiple range test, α = 0.05). Bar = 0.5 cm.

Figure 8.

Interactions between alternatively spliced transcripts of OsWRKY76 and OsWRKY62. A, Schematic diagrams of OsWRKY62 and OsWRKY76 proteins. The domain is described in Figure 1. The amino acid positions correspond to those in the full-length proteins. B, Interactions among OW76.1, OW76.2, OW62.1, and OW62.2 in yeast two-hybrid assays. OW76.1, OW76.2, OW62.1, and OW62.2 were fused to the Gal4 DNA-binding domain (BD) and/or activation domain (AD). Yeast cells were incubated in SD medium lacking Leu and Trp (-LW; left) or Leu, Trp, His, and adenine (-LWHA; right) and photographed 3 d after plating. Yeast cells harboring AD-T with BD-53 or BD-Lam vector were used as the positive or negative control. C, Inhibition of interaction between OW76.1 and OW62.1 by OW76.2 or OW62.2. Yeast cells were incubated in SD medium lacking Leu and Trp (-LW; left), Leu, Trp, and His (-LWH; middle), or Leu, Trp, His, and Met (-LWHM; right) and photographed 3 d after plating. Yeast cells harboring BD with pAD-OW62.1 were used as a negative control. For β-galactosidase activity assays, three clones of each combination were grown in SD medium lacking Leu and Trp at 30°C to an optical density of 2 at 600 nm. In yeast-three hybrid assays, yeast colonies were grown in SD medium lacking Leu and Trp with (white) or without (black) Met at 30°C. The β-galactosidase activity was measured using O-nitrophenyl-β-d-galactopyranoside as the substrate. Experiments were repeated three times with similar results. Values marked with different letters indicate statistically significant differences (Duncan’s multiple range test, α = 0.05).

Figure 9.

Transcription activities of OsWRKY62 and OsWRKY76 in planta. A, Schematic diagram of effector and reporter plasmids used in the transient expression assays. Box indicate DNA sequences containing the W box (W) or mutated W box (mmW); T-R, terminator of the rice rbcS gene; T-N, terminator of the NOS gene. B, C, and E, An effector gene could be inserted in the position of the effector shown in A. In the case of multiple effectors, the effector gene in the same plasmid with the W box is shown as +w, whereas the additional effector gene was constructed in a separate plasmid without the W box or LUCIFERASE (LUC) gene. The construct was introduced into leaves of N. benthamiana by agroinfiltration. Total proteins were extracted from the leaves 2 d after infiltration. GUS activity was normalized with LUC activity. A slash (/) indicates the presence of effector in the pWGus plasmid, and pWGus is without the effector gene. Data are means ± se of at least five independent experiments. Values marked with different letters indicate statistically significant differences as analyzed by the SAS software (Duncan’s multiple range test, α = 0.05). In E, the wild-type and mutated OsWRKY62 genes were inserted in the position marked as effector shown in A. D, Amino acid sequence of the N-terminal region of OsWRKY62.1. Schematic diagrams of deletion (d1, d2, d3, d18, d29, d35) and poly-Ala substitution (m1, m2, and m3) mutants of OsWRKY62.1 (W62.1) are shown. Dashes represent deleted amino acids. The underlined letters represent the mutation sites.

Figure 10.

Binding of the N termini of OsWRKY76 and OsWRKY62 to the W box element in the promoter of OsWRKY76. A, Nucleotide sequence of 81P1 used in DNA-binding experiments, with the core W box sequences underlined. B, OW76.1N, OW76.2N, OW62.1N, and OW62.2N represent the GST-tagged recombinant proteins of OsWRKY76.1N (amino acids 1–230, C-terminal 97 amino acids deleted), OsWRKY76.2N (amino acids 1–158, C-terminal 97 amino acids deleted), OsWRKYW62.1N (amino acids 1–205, C-terminal 113 amino acids deleted), and OsWRKY62.2N (amino acids 1–166, C-terminal 113 amino acids deleted), respectively. 81P1-B, Biotin-labeled 81P1; 50×, addition of 50-fold unlabeled 81P1; +, presence; −, absence. The reaction mixture was separated on a native PAGE gel and blotted on a nylon membrane. The 81P1-B probe was detected by anti-biotin antibody. C, Loading control indicating the amounts of proteins used in the DNA-binding assay, stained by Coomassie Brilliant Blue. M, Protein molecular weight markers.