Phosphorylation of a WRKY transcription factor by MAPKs is required for pollen development and function in Arabidopsis

Abstract

Plant male gametogenesis involves complex and dynamic changes in gene expression. At present, little is known about the transcription factors involved in this process and how their activities are regulated. Here, we show that a pollen-specific transcription factor, WRKY34, and its close homolog, WRKY2, are required for male gametogenesis in Arabidopsis thaliana. When overexpressed using LAT52, a strong pollen-specific promoter, epitope-tagged WRKY34 is temporally phosphorylated by MPK3 and MPK6, two mitogen-activated protein kinases (MAPKs, or MPKs), at early stages in pollen development. During pollen maturation, WRKY34 is dephosphorylated and degraded. Native promoter-driven WRKY34-YFP fusion also follows the same expression pattern at the protein level. WRKY34 functions redundantly with WRKY2 in pollen development, germination, and pollen tube growth. Loss of MPK3/MPK6 phosphorylation sites in WRKY34 compromises the function of WRKY34 in vivo. Epistasis interaction analysis confirmed that MPK6 belongs to the same genetic pathway of WRKY34 and WRKY2. Our study demonstrates the importance of temporal post-translational regulation of WRKY transcription factors in the control of developmental phase transitions in plants.

Figure 1. In vitro phosphorylation of WRKY34 by MPK3 and MPK6.

(A) Putative MAPK-phosphorylation sites in WRKY34 and WRKY33. Bars indicate the position of potential MAPK phosphorylation sites in the protein. Grey boxes indicate WRKY domains. Note that the clusters of phosphorylation sites at N-termini were similar between WRKY34 and WRKY33. (B) In vitro phosphorylation assay of WRKY34 by the activated MPK3 and MPK6 (upper panel). Reactions with various components omitted (-) were used as controls. Recombinant MKK4^DD/MKK5^DD were used to activate MPK3 and MPK6. Myelin basic protein (MBP) was used as control substrate (lower panel). (C) Adjacent sequences of putative MAPK-phosphorylation sites in WRKY34, and the loss-of phosphorylation WRKY34 mutant with all Ser mutated to Ala (WRKY34^SA). (D) Mutation of MAPK-phosphorylation sites greatly reduced the phosphorylation of WRKY34 by MPK3 and MPK6. Phosphorylated WRKY34 was visualized by autoradiography after gel electrophoresis.

Figure 2. Phosphorylation of WRKY34 in vivo during male gametogenesis.

(A) Staging of flowers/buds used for WRKY34 protein analysis. Flower at Stage 13, in which anthesis is about to occur, was designated 0. An open flower right after anthesis was designated +1. Younger flowers/buds were designated using negative numbers. (B) Immunoblot and Phos-tag assays of LAT52 promoter-driven pollen-specific 4myc-WRKY34 protein at different stages of pollen development. The −7 to +1 flowers/buds have pollen at different developmental stages. Black bar indicates flowers or buds containing mature pollen. Dark gray bar indicates buds containing tricellular pollen (TCP). Light gray bar indicates buds containing bicellular pollen (BCP). Levels of 4myc-WRKY34 protein in immunoblot (top panel) and Phos-tag assay (middle panel) were determined using an anti-myc antibody. Protein loading control was confirmed by Coomassie blue staining (bottom panel). (C) Immunoblot (top panel) and Phos-tag assay (middle panel) of pollen-specific 4myc-WRKY34^SA protein at different developmental stages. Protein loading control was confirmed by Coomassie blue staining (bottom panel). Each sample was extracted from the same number of flowers/buds at the corresponding stage, which allows the comparison of WRKY34 protein levels in an equal number of developing/mature pollen grains.

Figure 3. In vivo phosphorylation of WRKY34 is dependent on MPK3 and MPK6.

(A) MPK3 expression in mpk6 and MPK3RNAi mpk6 pollen grains. Total RNAs were isolated from pollen grains. MPK3 transcript levels were determined using quantitative RT-PCR. Error bars = standard derivation. (B) Immunoblot (top panel) and Phos-tag (middle panel) assays of WRKY34 protein at different stages of MPK3RNAi mpk6 P_LAT52:4myc-WRKY34 flower buds. Protein loading control was confirmed by Coomassie blue staining (bottom panel). Each sample was extracted from the same number of flowers/buds at the corresponding stage, which allows the comparison of WRKY34 protein levels in an equal number of developing/mature pollen grains.

Figure 4. Expression and protein localization of WRKY34 and WRKY2.

(A) Quantitative RT-PCR of WRKY34 and WRKY2 transcripts in various tissues. R, roots; St, stems; L, leaves; Sl, seedlings; B, buds; and Of, open flowers. Error bars = standard derivation. (B to I) Expression and localization of WRKY2 promoter-driven WRKY2:YFP fusion protein in pollen. DAPI staining was used to locate nuclei (B to E), and YFP signal reveals the localization of WRKY2:YFP fusion at different developmental stages (F to I). (B and F) UNM stage, no nucleus-localized YFP signal was detected. Vegetative nucleus localized WRKY2:YFP signal was observed at BCP stage (C and G), TCP stage (D and H), and MP stage (E and I). (J to Q) Expression and localization of WRKY34 promoter-driven WRKY34:YFP fusion protein in pollen. (J to M) DAPI staining signal. (N to Q) YFP signal. (J and N) UNM stage, weak signal was observed in microspore nucleus. Vegetative nucleus localized WRKY34:YFP signal was observed at BCP stage (K and O), and TCP stage (L and P). No YFP signal was observed in MP (M and Q). Note that as the vegetative cell expressed genes, the WRKY2 and WRKY34 fusion YFP signals were not detectable in generative or sperm cells. MN, microspore nucleus. VN, vegetative nucleus. GN, generative nucleus/nuclei. SN, sperm cell nuclei. Bar = 50 µm.

Figure 5. Phenotype of wrky2-1 wrky34-1 double mutant pollen.

(A) Diagram of T-DNA insertion alleles of wrky2 and wrky34 mutants. Arrows indicate the positions of RT-qPCR primers. Black bars = untranslated regions (UTRs); gray bars = exons; lines = introns. (B) Quantitative RT-PCR of WRKY2 expression in wild-type, wrky2-1, and wrky2-2 seedlings and pollen grains. Error bars = standard derivation. (C) Normal vegetative growth and development of wrky2-1 wrky34-1 double mutant plants. Five-week-old plants are pictured. (D, E) Alexander staining of wild-type (D) and wrky2-1 wrky34-1 double mutant (E) pollen. Bar = 50 µm. (F, G) Vital staining by FDA of wild type (F) and wrky2-1 wrky34-1 double mutant (G) pollen. Bar = 50 µm. (H, I) Scanning electron microscopy (SEM) of wild type (H) and wrky2-1 wrky34-1 double mutant (I) pollen. Bar = 20 µm. (J, K, L, M) Transmission electron microscopy (TEM) of wild type (J, L) and wrky2-1 wrky34-1 double mutant (K, M) pollen. (J, K) Bar = 5 µm. (L, M) Bar = 1 µm. Arrows in panel K indicate the germination pore with defective intine layer. P, plastid; E, endoplasmic reticulum.

Figure 6. Pollen germination and pollen tube growth are defective in wrky2-1 wrky34-1.

In vitro pollen germination of wild type (A) and wrky2-1 wrky34-1 double mutant (B) pollen. Bars = 50 µm. Aniline blue staining of wild type pistils 8 hours after pollination with wild type (C) and wrky2-1 wrky34-1 double mutant (D) pollen. Bar = 200 µm.

Figure 7. Complementation of wrky2-1 wrky34-1 double mutant pollen phenotypes by WRKY34^WT and WRKY34^SA.

Pollen viability ratio by FDA staining (A), in vitro pollen germination ratio (B), and average pollen tube length (C) in Col-0, wrky2-1 wrky34-1, wrky2-1 wrky34-1 P_WRKY34:WRKY34^WT, and wrky2-1 wrky34-1 P_WRKY34:WRKY34^SA plants. Error bar = standard error. Double asterisks indicate statistically very significant difference from wild-type pollen (p-value<0.01).

Figure 8. Phenotype of mpk6 wrky34-1 and mpk6 wrky2-1 pollen.

Pollen viability ratio by FDA staining (A), in vitro pollen germination ratio (B), and average pollen tube length (C) in Col-0, mpk6, wrky2-1, wrky34-1, mpk6 mpk34-1, and mpk6 wrky2-1 plants. Error bar = standard error. Single asterisks indicate statistically significant difference from wild-type pollen (0.01<p-value<0.05). Double asterisks indicate very significant difference from wild-type pollen (p-value<0.01).