FILAMENTOUS FLOWER Is a Direct Target of JAZ3 and Modulates Responses to Jasmonate

Abstract

The plant hormone jasmonate (JA) plays an important role in regulating growth, development, and immunity. Activation of the JA-signaling pathway is based on the hormone-triggered ubiquitination and removal of transcriptional repressors (JASMONATE-ZIM DOMAIN [JAZ] proteins) by an SCF receptor complex (SCF(COI1)/JAZ). This removal allows the rapid activation of transcription factors (TFs) triggering a multitude of downstream responses. Identification of TFs bound by the JAZ proteins is essential to better understand how the JA-signaling pathway modulates and integrates different responses. In this study, we found that the JAZ3 repressor physically interacts with the YABBY (YAB) family transcription factor FILAMENTOUS FLOWER (FIL)/YAB1. In Arabidopsis thaliana, FIL regulates developmental processes such as axial patterning and growth of lateral organs, shoot apical meristem activity, and inflorescence phyllotaxy. Phenotypic analysis of JA-regulated responses in loss- and gain-of-function FIL lines suggested that YABs function as transcriptional activators of JA-triggered responses. Moreover, we show that MYB75, a component of the WD-repeat/bHLH/MYB complex regulating anthocyanin production, is a direct transcriptional target of FIL. We propose that JAZ3 interacts with YABs to attenuate their transcriptional function. Upon perception of JA signal, degradation of JAZ3 by the SCF(COI1) complex releases YABs to activate a subset of JA-regulated genes in leaves leading to anthocyanin accumulation, chlorophyll loss, and reduced bacterial defense.

Figure 1.

JAZ3 Physically Interacts with FIL and YAB3. (A) Y2H assay using all 12 JAZ proteins fused to Gal4 BD (BD-JAZ) and the four AD-YAB fusions. Yeast cells cotransformed with the indicated combinations were selected in medium lacking Leu and Trp (-2) (Supplemental Figure 1A) and grown for 5 d in medium lacking Ade, His, Leu, and Trp without (-4) or with 3-aminotriazole (-4AT). Numbers on top of the panel represent JAZ proteins, from JAZ1 (1) to JAZ12 (12). (B) Immunoblot using anti-GFP antibody of protein extracts from transgenic plants expressing JAZ1-GFP, JAZ3-GFP, JAZ4-GFP, and JAZ9-GFP proteins after PD by recombinant MBP-FIL, MBP-YAB2, MBP-YAB3, and MBP-YAB5 expressed in E. coli. MBP protein was used as negative control. A Coomassie blue-stained gel is shown below the immunoblot as representative of the amount of recombinant MBP-fused protein used in each sample. The first lane of the blot corresponds to 40 μL of the input protein extract used for PD. Asterisks mark the bands corresponding to the indicated protein when several bands are present. (C) Immunoblot using anti-GR antibody of protein extracts from 35S:FIL-GR transgenic plants after PD by recombinant MBP-JAZ proteins expressed in E. coli. Seven days postgermination 35S:FIL-GR seedlings were treated with 20 μM DEX for 2 d to induce FIL-GR accumulation. MBP protein was used as negative control. Col-0 protein extracts were used to discard unspecific antibody cross reactions. A Coomassie blue-stained gel is shown below the immunoblot as representative of the amount of recombinant MBP-fused proteins used in each sample. Forty microliters of the protein extracts used for PD is shown as input. Asterisks mark the band corresponding to the indicated protein when several bands are present. (D) Immunoblot using anti-GFP antibody of protein extracts from 35S:JAZ3-GFP (top panel) and 35S:JAZ3∆Jas-GFP (bottom panel) transgenic plants pulled down by recombinant MBP-FIL proteins expressed in E. coli. MBP protein was used as a negative control and MBP-JAZ3 was used as a positive interaction control. The lowest panel corresponds to a Coomassie blue-stained gel showing the amount of recombinant MBP-fused protein used in each sample. The first lane of the blot correspond to 40 μL of the input protein extract used for PD. Asterisk marks the band corresponding to the indicated protein when several bands appear in the picture. (E) Mapping the FIL binding domain of JAZ3. Y2H assays using FIL fused to the GAL4 Binding Domain (BD-FIL) and full-length JAZ3 or truncated derivatives fused to the GAL4 activation domain (AD-JAZ3). Numbers represent amino acid positions relative to the full-length protein. Yeast cells cotransformed with the indicated combinations were selected in medium lacking Leu and Trp (-2) and grown for 5 d in medium lacking Ade, His, Leu, and Trp without or with 5 mM 3-aminotriazole (-4, -4AT) to select for interactions. AD represents the empty pGADT7 vector.

Figure 2.

YAB Binding Sites Are Enriched among JA-Inducible Promoters. (A) Left: DNA binding motif of YAB1, as described by Franco-Zorrilla et al. (2014). Right: Percentages of JA-inducible genes (0.5-, 1-, and 3-h treatments), whose promoters (1 kb upstream) contain YAB1 binding sites. Blue bars represent the percentage of JA upregulated genes (Up JA) and red bars the percentage of JA downregulated genes (Down JA) at each time point. The green bar (Genome) represents the percentage of promoters in the complete gene set represented in the array that contain the same YAB binding sites. Asterisks indicate statistically significant differences (P value < 0.01, hypergeometric test) between the indicated percentages and the corresponding genome set. (B) Enrichment of JA-regulated genes in a transcriptomic data set corresponding to genes upregulated in response to the activation of the FIL-GR translational fusion with dexamethasone (8 h postactivation). Asterisks indicate significant differences (P value < 0.01, hypergeometric test) between the percentage of genes up- or downregulated by JA (Up by JA or Down by JA) in the FIL-GR data set compared with the proportion of JA-induced or JA-repressed genes in the whole genome (Up by JA in Genome or Down by JA in the Genome). Gene lists corresponding to FIL-GR were obtained from Bonaccorso et al. (2012). (C) Venn diagram representing the overlap between genes upregulated 3 h after JA treatment and genes downregulated in triple fil yab3 yab5 mutants (Sarojam et al., 2010). The proportion of genes differentially expressed in each data set relative to the complete gene set is indicated (Up JA, 2.1%; Down triple, 2.8%), as well as the proportion of overlapping genes (9.3%; random expected percentage: 2.1% × 2.8% = 0.06%).

Figure 3.

Ectopic Expression of FIL Results in Increased Anthocyanin Levels. (A) Anthocyanin content of 7-d-old 35S:FIL-GR transgenic Arabidopsis seedlings grown on MS medium supplemented with 20 μM DEX plus/minus 50 μM JA. Ethanol was the solvent used for JA stock and was added in control treatments. DMSO is the solvent used for DEX stock and was added in control treatment. Bar = 2 mm. (B) Anthocyanin content of 10-d-old 35S:FIL-GR transgenic Arabidopsis seedlings either in Col-0 or mutant jai3-1 backgrounds. Seedlings were grown on Johnson medium for 5 d and then supplemented with 20 μM DEX plus/minus 50 μM JA for five more days. Error bars represent sd. Asterisks indicate statistically significant differences between 35S:FIL-GR in the jai3-1 background compared with 35S:FIL-GR in the presence of JA (P value < 0.01, t test). FW, fresh weight. (C) Anthocyanin content of 10-d-old 35S:FIL-GR transgenic Arabidopsis seedlings either in Col-0 or mutant coi1-1 backgrounds presented as in (B). Arabidopsis jai3-1 and coi1-1 JA-insensitive mutants were used as controls. Error bars represent sd. Asterisks indicate statistically significant differences between 35S:FIL-GR in the coi1-1 background compared with 35S:FIL-GR in the Col-0 background in the presence of JA (P value < 0.01, t test).

Figure 4.

YABs Are Required for JA-Mediated Responses in Aerial Tissues. (A) Anthocyanin content of 10-d-old wild-type and yab mutant Arabidopsis seedlings grown in Johnson medium with or without 50 µM JA. Error bars represent sd. Asterisks indicate statistical significant differences between the corresponding samples and their corresponding wild type (Ler for fil8 yab3 mutants or Coler for fil8 yab2 yab3 yab5) (P value < 0.01, t test). FW, fresh weight. (B) Chlorophyll content of 10-d-old wild-type or yab mutant Arabidopsis seedlings grown in Johnson medium with or without 50 μM JA. JA-insensitive coi1-16 and jai3-1 mutants were used as controls. Error bars represent sd. Asterisks indicate statistically significant differences between samples and their corresponding wild type (Ler for fil8 yab3 mutants or Coler for fil8 yab2 yab3 yab5) (P value < 0.01, t test). (C) Chlorophyll content of 10-d-old 35S:FIL-GR transgenic Arabidopsis seedlings grown on Johnson medium with 20 μM DEX plus/minus 50 μM JA. Arabidopsis 35S:MYC2 and 35S:JAZ3∆Jas were used as JA-hypersensitive and -hyposensitive controls, respectively. FW, fresh weight. Error bars represent sd. Asterisks indicate statistically significant differences between the corresponding samples and Col-0 (P value < 0.01, t test).

Figure 5.

YABs Participate in Plant Defense Responses against P. syringae Infection. (A) Growth of Pto DC3000 on wild-type Arabidopsis Col-0, 35S:FIL-GR, and coi1-30 plants 2 d after spray inoculation. Plants were pretreated with 20 μM DEX for 16 h prior to infection. Asterisks indicate statistically significant differences with the control Col-0 (P value < 0.01, t test). (B) Growth of Pto DC3000 on wild-type Arabidopsis and different yab mutant backgrounds. Bacterial counts are expressed as log (colony-forming units [cfu]/cm²). Error bars indicate se. The results are representative of two independent experiments. fil/+ yab3 and fil/+ yab235 are heterozygous for the fil8 allele and were used instead of fil yab3 and fil yab235 homozygous plants because the extreme phenotype of these plants made it challenging to perform the pathotest assays. Asterisks indicate statistical significant differences with their corresponding control Ler or Coler (P value < 0.01, t test).

Figure 6.

Anthocyanin Accumulation in 35S:FIL-GR Is Dependent on MYB75. (A) Eight-day-old 35S:FIL-GR transgenic Arabidopsis seedlings either in wild-type (Nos) or myb75-1 backgrounds grown on MS medium supplemented with 20 or 20 μM DEX plus 50 μM JA. Bars = 2 mm. (B) Anthocyanin content of 8-d-old 35S:FIL-GR transgenic Arabidopsis seedlings either in wild-type (Nos) or myb75-1 backgrounds. Plants were grown for 8 d in MS medium supplemented with 20 μM DEX plus/minus 50 μM JA. DMSO and ethanol were used as solvents for DEX and JA, respectively, and added in control treatments. FW, fresh weight. Error bars represent se. Asterisks indicate statistically significant differences between the corresponding samples and their reference wild type (Nos) (P value < 0.01, t test). (C) RT-qPCR analysis of MYB75 expression relative to ACTIN in 10-d-old 35S:FIL-GR seedlings following a 4-h exposure to a mock (DMSO/EtOH), DEX (DEX/EtOH), JA (DMSO/JA), or both DEX and JA (DEX/JA) treatment. Asterisks indicate statistically significant differences between the corresponding samples and their control (DMSO/EtOH) (P value < 0.01, t test).

Figure 7.

FIL Is a Transcriptional Activator of MYB75 and JAZ3 Is Able to Repress Its Activity. (A) Schematic representation of the constructs used in transient expression experiments in N. benthamiana. The MYB75 promoter fused to the firefly LUC gene was used as a reporter. FIL and JAZ3∆Jas genes expressed under the CaMV 35S promoter were used as effectors. (B) Transactivation assay using 35S:FIL and 35S:JAZ3∆Jas as effectors in basal conditions (no treatment). Error bars represent sd of 12 replicates. (C) Transactivation assay using 35S:FIL and 35S:JAZ3∆Jas as effectors in inductive (0.5 μM COR) conditions. Error bars represent sd of 12 replicates. (D) EMSA experiment showing that MBP-FIL fusion protein binds to the DNA probe of the MYB757PAP1 promoter containing the putative FIL binding element 1 (FBE1). Biotin-labeled FBE1 probe was incubated with MBP-FIL protein, and the free and bound probe were separated in an acrylamide gel. As indicated, unlabeled probes (FBE1 and G-box) were used as specific and unspecific competitors, respectively. The MBP-FIL-DNA complex is indicated with an arrow. The lower retarded band (higher mobility) likely corresponds to a retarded band by a FIL truncated derivative that lacks sequence specificity. (E) Anthocyanin content of 5-d-old Col-0, pap1-D, fil8 yab3, and pap1-D fil8 yab3 Arabidopsis seedlings grown on MS medium plus/minus 50 μM JA. The wild type, pap1-D, fil8 yab3, and pap1-D fil8 yab3 are all Col-0 × Ler.

Figure 8.

Model for FIL Function in JA Signaling. JA induces the degradation of JAZ proteins via the SCF^COI1 complex leading to the activation of FIL/YAB3 and other TFs (including MYC2, MYC3, MYC4, MYB75, and others). Activation of FIL/YAB3 results in the induction of MYB75 leading to the synergistic formation of active WD-repeat/bHLH/MYB75 complexes and the accumulation of anthocyanin. Moreover, JA-dependent activation of FIL/YAB3 also leads to a chlorophyll breakdown and antagonism of the SA-mediated defense response, most likely through indirect mechanisms (broken arrows). A negative feedback loop arises from JA-induced activation of JAZ expression (dotted arrow). Solid arrows indicate direct physical interactions/regulation.