SG2-Type R2R3-MYB Transcription Factor MYB15 Controls Defense-Induced Lignification and Basal Immunity in Arabidopsis

Abstract

Lignification of cell wall appositions is a conserved basal defense mechanism in the plant innate immune response. However, the genetic pathway controlling defense-induced lignification remains unknown. Here, we demonstrate the Arabidopsis thaliana SG2-type R2R3-MYB transcription factor MYB15 as a regulator of defense-induced lignification and basal immunity. Loss of MYB15 reduces the content but not the composition of defense-induced lignin, whereas constitutive expression of MYB15 increases lignin content independently of immune activation. Comparative transcriptional and metabolomics analyses implicate MYB15 as necessary for the defense-induced synthesis of guaiacyl lignin and the basal synthesis of the coumarin metabolite scopoletin. MYB15 directly binds to the secondary wall MYB-responsive element consensus sequence, which encompasses the AC elements, to drive lignification. The myb15 and lignin biosynthetic mutants show increased susceptibility to the bacterial pathogen Pseudomonas syringae, consistent with defense-induced lignin having a major role in basal immunity. A scopoletin biosynthetic mutant also shows increased susceptibility independently of immune activation, consistent with a role in preformed defense. Our results support a role for phenylalanine-derived small molecules in preformed and inducible Arabidopsis defense, a role previously dominated by tryptophan-derived small molecules. Understanding the regulatory network linking lignin biosynthesis to plant growth and defense will help lignin engineering efforts to improve the production of biofuels and aromatic industrial products as well as increase disease resistance in energy and agricultural crops.

Figure 1.

MYB15 Is Required for Defense-Induced Lignification. (A) Lignin analysis by phloroglucinol-HCl staining. Seven-day-old seedlings elicited with 1 µM flg22 for 48 h. Samples were stained at the same time, and intensity of purple-red staining correlates with lignin content. (B) Schematic representation of the MYB15 gene (top) and protein (bottom). Gray boxes indicate exons; triangles, positions of T-DNA insertions; thick black line, the corresponding transcript region amplified by qPCR in (D); dashed lines, the corresponding T-DNA insertion sites. (C) qPCR analysis of MYB15, MYB58, and MYB63 transcripts in 9-d-old seedlings elicited with water or 1 µM flg22 for 3 h. Expression values are normalized to that of eIF4A1 housekeeping gene and relative to those of water-treated wild-type sample. Data represent mean ± sd of four replicates of 10 to 15 seedlings each. Different letters denote statistically significant differences (P < 0.05, two-tailed t test). (D) Lignin analysis of two independent DEX:MYB15 lines by phloroglucinol-HCl staining. Seven-day-old seedlings elicited with 20 µM Dex and/or 1 µM flg22 for 72 h. Samples were stained at the same time.

Figure 2.

MYB15 Is Sufficient to Activate flg22-Induced Lignification. (A) to (D) Lignin analysis by AcBr method. Data represent the mean ± sd of six ([A] and [B]), three (C), and eight (D) replicates of 30 to 45 seedlings each. Different letters denote statistically significant differences (P < 0.05, two-tailed t test). Experiments were repeated once or twice, producing similar results. (A) Nine-day-old seedlings elicited with water or 2 µM flg22 for 48 h. (B) and (D) Unelicited 10-d-old seedlings. (C) Nine-day-old seedlings treated with water or 20 µM Dex for 48 h. (E) and (F) Lignin analysis by DFRC method. Nine-day-old seedlings elicited with water or 2 µM flg22 for 48 h. Data represent the mean ± sd of three replicates of 30 to 45 seedlings each. Different letters denote statistically significant differences (P < 0.05, two-tailed t test).

Figure 3.

MYB15 Activates Genes Involved in the Biosynthesis of G-Lignin, Not S-Lignin. (A) Table of monolignol and branch pathway genes transcriptionally activated in response to flg22 elicitation. Nine-day-old wild-type seedlings elicited with water or 1 µM flg22 for 3 h. Expression values are normalized to that of eIF4A1 housekeeping gene and relative to those of water-treated wild-type sample. Data represent mean of four replicates of 10 to 15 seedlings each. P value determined by two-tailed t test. (B) Heat map of relative gene expression levels of monolignol and branch pathway genes in (C) and (D). (C) qPCR analysis of monolignol and branch pathway genes in 9-d-old seedlings elicited with 1 µM flg22 for 3 h. Expression values are normalized to that of eIF4A1 housekeeping gene and relative to those of flg22-treated wild-type sample. Data represent mean ± sd of four replicates of 10 to 15 seedlings each. Asterisks denote statistically significant differences from the wild type (P < 0.05, two-tailed t test; false discovery rate < 0.05). (D) qPCR analysis of monolignol and branch pathway genes in unelicited 9-d-old seedlings. Expression values are normalized to that of eIF4A1 housekeeping gene and relative to those of wild-type sample. Data represent mean ± sd of four replicates of 10 to 15 seedlings each. Asterisks denote statistically significant differences from the wild type (P < 0.05, two-tailed t test; false discovery rate < 0.05). qPCR experiments were repeated once, producing similar results.

Figure 4.

MYB15 Binds to Promoters of Monolignol Pathway Genes. (A) ChIP was performed on 8-d-old myb15-1 seedlings containing DEX:MYB15-myc transgene and pretreated with mock solution (M) or 20 µM Dex (D) for 4 h. PCR analysis was performed on nuclear extracts prior to incubation with myc antibody (input) and after ChIP, using primers flanking regions containing AC-like cis-elements (listed in Supplemental Table 2). (B) Fold enrichment was determined by calculating the ratio of PCR product intensities in ChIP D/M to Input D/M. The mean fold enrichment of two independent immunoprecipitation experiments is reported to the left of the graph. (C) Schematic representations of monolignol pathway gene promoters with the amplified regions represented by black bars. Numbers indicate nucleotide positions upstream of the translational start site.

Figure 5.

MYB15-Dependent Synthesis of G-Lignin Contributes to Basal Resistance against the Virulent Bacterial Pathogen P. syringae. (A), (B), and (D) Pathogen infection assays in 10-d-old seedlings (left) and 5-week-old leaves (right). Bacteria were extracted from untreated plants ([A] and [C]) and from plants treated with 20 µM Dex for 6 h prior to surface-inoculation with Pto DC3000 (B). Data represent the mean ± sd of five ([A], left; [B], left; [D], left), four ([D], right), and three ([A], right; [B], right) replicates. Different letters denote statistically significant differences (P < 0.05, two-tailed t test). FW, fresh weight. Experiments were repeated once or twice, producing similar results. (C) Lignin analysis by phloroglucinol-HCl staining of 9-d-old seedlings elicited with 1 µM flg22 for 48 h.

Figure 6.

MYB15 Increases the Synthesis of Soluble Phenolics Independently of flg22. (A) Table of soluble phenolics detected in 10-d-old wild-type seedlings by LC-MS/MS and whose identities were inferred from accurate m/z measurements of high-resolution mass spectrometry data and confirmed by their absence in biosynthetic mutants f6’h1, fah1-2, and aact1-1. RT, retention time. (B) Heat map of relative abundances of soluble phenolics in 10-d-old seedlings elicited with water or 1 µM flg22 for 3 h. Data representing mean ± sd of five replicates can be found in Supplemental Figure 7. (C) HPLC-FLD analysis of total scopoletin in 9-d-old seedlings, either unelicited (left) or treated with 20 µM Dex for 24 h (right). Data represents the mean ± sd of four replicates of 10 to 15 seedlings each. Different letters denote statistically significant differences (P < 0.05, two-tailed t test). (D) HPLC-DAD analysis of sinapate esters in 9-d-old seedlings, either unelicited (left and center) or treated with 20 µM Dex for 24 h (right). Peak area values are normalized to that of internal standard. Data represents the mean ± sd of four replicates of 10 to 15 seedlings each. Different letters denote statistically significant differences (P < 0.05, two-tailed t test). HPLC experiments were repeated at least once, producing similar results.

Figure 7.

Scopoletin Contributes to Antibacterial Defense Independently of Immune Activation. (A) to (C) Pathogen infection assays in 10-d-old seedlings (A) and 5-week-old leaves ([B] and [C]). Bacteria were extracted from plants surface-inoculated with Pto DC3000 ([A] and [B]) and from leaves preinfiltrated with 0.1 nmol flg22 and 0.05 nmol soluble phenolic per leaf for 24 h prior to infiltration with Pto DC3000 (C). Data represent the mean ± sd of three (A), four (B), six ([C], left), and eight ([C], right) replicates. Different letters denote statistically significant differences (P < 0.05, two-tailed t test). FW, fresh weight. Experiments were repeated once ([A] and [B]) or twice (C), producing similar results. (D) Schematic of monolignol and branch pathways involved in preformed and inducible defenses against a bacterial pathogen (red).

Figure 8.

MYB15 Paralogs MYB13 and MYB14 Do Not Contribute to Defense-Induced Lignification. (A) Table of SG2-type R2R3-MYB genes transcriptionally activated in response to flg22 elicitation. Nine-day-old wild-type seedlings elicited with water or 1 µM flg22 for 3 h. Expression values are normalized to that of eIF4A1 housekeeping gene and relative to those of water-treated wild-type sample. Data represent mean of four replicates of 10 to 15 seedlings each. P value determined by two-tailed t test. (B) qPCR analysis of MYB13 and MYB14 genes in 9-d-old seedlings elicited with water or 1 µM flg22 for 3 h. Expression values are normalized to that of eIF4A1 housekeeping gene and relative to those of water-treated wild-type sample. Data represent mean ± sd of four replicates of 10 to 15 seedlings each. Different letters denote statistically significant differences (P < 0.05, two-tailed t test). (C) Lignin analysis by phloroglucinol-HCl staining. Nine-day-old seedlings elicited with 1 µM flg22 for 48 h. (E) qPCR analysis of PAL1 gene in 9-d-old seedlings elicited with 1 µM flg22 for 3 h. Expression values are normalized to that of eIF4A1 housekeeping gene and relative to those of wild-type sample. Data represent mean ± sd of four replicates of 10 to 15 seedlings each. Different letters denote statistically significant differences (P < 0.05, two-tailed t test).