Arabidopsis Transcriptomics Reveals the Role of Lipoxygenase2 (AtLOX2) in Wound-Induced Responses

Abstract

In wounded Arabidopsis thaliana leaves, four 13S-lipoxygenases (AtLOX2, AtLOX3, AtLOX4, AtLOX6) act in a hierarchical manner to contribute to the jasmonate burst. This leads to defense responses with LOX2 playing an important role in plant resistance against caterpillar herb-ivory. In this study, we sought to characterize the impact of AtLOX2 on wound-induced phytohormonal and transcriptional responses to foliar mechanical damage using wildtype (WT) and lox2 mutant plants. Compared with WT, the lox2 mutant had higher constitutive levels of the phytohormone salicylic acid (SA) and enhanced expression of SA-responsive genes. This suggests that AtLOX2 may be involved in the biosynthesis of jasmonates that are involved in the antagonism of SA biosynthesis. As expected, the jasmonate burst in response to wounding was dampened in lox2 plants. Generally, 1 h after wounding, genes linked to jasmonate biosynthesis, jasmonate signaling attenuation and abscisic acid-responsive genes, which are primarily involved in wound sealing and healing, were differentially regulated between WT and lox2 mutants. Twelve h after wounding, WT plants showed stronger expression of genes associated with plant protection against insect herbivory. This study highlights the dynamic nature of jasmonate-responsive gene expression and the contribution of AtLOX2 to this pathway and plant resistance against insects.

Keywords: 13S-lipoxygenase; AtLOX2; jasmonate; transcriptome; wounding.

Figure 1

Foliar constitutive gene expression. (A): Venn diagram of constitutive genes differentially expressed between wildtype (WT) and lox2 Arabidopsis thaliana plants at two time points (1 and 12 h). Differentially expressed genes between the different treatments and times were determined by DESeq2 (p-value (padj) ≤ 0.05 and log2 fold change ≥2 or ≤−2). (B): Gene set enrichment analysis (GSEA) and ridgeline plot highlighting constitutive metabolic pathways differentially expressed between WT and lox2 plants at 1 h. (C): GSEA and ridgeline plot highlighting constitutive metabolic pathways differentially expressed between WT and lox2 plants at 12 h. The ridgeline plot visualizes the distribution of differential enrichment categories identified by GSEA. Significantly differentially expressed pathways in ridgeline plots are indicated in green.

Figure 2

Foliar salicylic acid (SA) levels and SA-related gene expression. Four-week-old Arabidopsis thaliana wildtype (WT) or lox2 plants were left undamaged (U) or wounded (W) with a hole punch on each fully expanded rosette leaf and harvested 1 or 12 h post-damage. (A): Foliar salicylic acid (SA) levels. (B): Constitutive expression of genes involved in SA biosynthesis and SA-responsive gene expression. Phytohormone levels are represented by the mean ± SE. Differences in phytohormone levels were determined by two-factor analysis of variance (2-factor ANOVA) (factors: genotype, treatment) followed by Tukey HSD (Supplemental Table S2). A hashtag (#) indicates genotype differences. Heatmaps visualize constitutive gene expression (wildtype—1 h (WT-1), wildtype—12 h (WT-12), lox2—1 h (lox2-1), lox2—12 h (lox2-12)). Genes: ICS1/SID2/EDS16 (At1g74710), EDS5 (At4g39030), PBS3 (At5g13320), EPS1 (At5g67160), CM1 (At3g29200), CM3 (At1g69370), PAL1 (At2g37040), PAL2 (At3g53260), PAL4 (At3g10340), AIM1 (At4g29010), NIMIN1 (At1g02450), NIMIN2 (At3g25882), WRKY38 (At5g22570), LLP (At5g03350), PR1 (At2g14610), GRX480/ROXY19 (At1g28480), GRXS13 (At1g03850), OPR1 (At1g76680), UGT74F1 (At2g43840), UGT76B1 (At3g11340), BSMT1 (At3g11480).

Figure 3

Foliar jasmonate levels and jasmonate-related gene expression. Four-week-old Arabidopsis thaliana wildtype (WT) or lox2 plants were left undamaged (U) or wounded (W) with a hole punch on each fully expanded rosette leaf and harvested 1 or 12 h post-damage. Jasmonate levels 1 h after wounding: (A): 12-oxo-phytodienoic acid (OPDA), (B): jasmonic acid and (C): 7-jasmonyl-isoleucine (JA-Ile). (D): Foliar wound-induced jasmonate-related gene expression (1 and 12 h). Phytohormone levels are represented by the mean ± SE. Differences in phytohormone levels were determined by two-factor analysis of variance (2-factor ANOVA) (Factors: genotype, treatment) followed by Tukey HSD (Supplemental Table S2). An asterisk (*) indicates wound-induced phytohormone levels and a hashtag (#) represents genotype differences. Wound-induced genes were determined by DESeq2 (p-value (padj) ≤ 0.05 and log2 fold change ≥2 or ≤−2). Heatmaps visualize wound-induced gene expression (wildtype—1 h (WT-1), wildtype—12 h (WT-12), lox2—1 h (lox2-1), lox2—12 h (lox2-12)). Genes: LOX2 (AT3g45140), LOX3 (At1g17420), LOX4 (At1g72520), AOC1 (At3g25760), AOC3 (At3g25780), OPR3 (At2g06050), OPCL1 (At1g20510), At1g20490, FBS1 (At1g61340), MYC2 (At1g32640), JAZ1 (At1g19180), JAZ5 (At1g17380), JAZ7 (At2g34600), JAZ8 (At1g30135), JAZ10 (At5g13220), JAZ13 (At3g22275), JOX2 (At5g05600), JOX3 (At3g55970), JOX4 (At2638240), CYP94B1 (At5g63450), CYP94B3 (At3g48520), CYP94C1 (At2g27690), ST2A (At5g07010), JID1 (At1g06620), ORA47 (At1g74930), ORA59 (At1g6160), VSP1 (At5g24780), THI2.1 (At1g72260), GRX480/ROXY19 (At1g28480).

Figure 3

Foliar jasmonate levels and jasmonate-related gene expression. Four-week-old Arabidopsis thaliana wildtype (WT) or lox2 plants were left undamaged (U) or wounded (W) with a hole punch on each fully expanded rosette leaf and harvested 1 or 12 h post-damage. Jasmonate levels 1 h after wounding: (A): 12-oxo-phytodienoic acid (OPDA), (B): jasmonic acid and (C): 7-jasmonyl-isoleucine (JA-Ile). (D): Foliar wound-induced jasmonate-related gene expression (1 and 12 h). Phytohormone levels are represented by the mean ± SE. Differences in phytohormone levels were determined by two-factor analysis of variance (2-factor ANOVA) (Factors: genotype, treatment) followed by Tukey HSD (Supplemental Table S2). An asterisk (*) indicates wound-induced phytohormone levels and a hashtag (#) represents genotype differences. Wound-induced genes were determined by DESeq2 (p-value (padj) ≤ 0.05 and log2 fold change ≥2 or ≤−2). Heatmaps visualize wound-induced gene expression (wildtype—1 h (WT-1), wildtype—12 h (WT-12), lox2—1 h (lox2-1), lox2—12 h (lox2-12)). Genes: LOX2 (AT3g45140), LOX3 (At1g17420), LOX4 (At1g72520), AOC1 (At3g25760), AOC3 (At3g25780), OPR3 (At2g06050), OPCL1 (At1g20510), At1g20490, FBS1 (At1g61340), MYC2 (At1g32640), JAZ1 (At1g19180), JAZ5 (At1g17380), JAZ7 (At2g34600), JAZ8 (At1g30135), JAZ10 (At5g13220), JAZ13 (At3g22275), JOX2 (At5g05600), JOX3 (At3g55970), JOX4 (At2638240), CYP94B1 (At5g63450), CYP94B3 (At3g48520), CYP94C1 (At2g27690), ST2A (At5g07010), JID1 (At1g06620), ORA47 (At1g74930), ORA59 (At1g6160), VSP1 (At5g24780), THI2.1 (At1g72260), GRX480/ROXY19 (At1g28480).

Figure 4

Wound-induced foliar gene expression. Four-week-old Arabidopsis thaliana wildtype (WT) or lox2 plants were wounded with a hole punch on each fully expanded rosette leaf and harvested at 1 and 12 h. Gene set enrichment analyses (GSEA) and ridgeline plots of WT (1 h) (A,E), WT (12 h) (B,F), lox2 (1 h) (C,G) and lox2 (12 h) (D,H). Wound-induced differentially expressed genes (DEGs) were determined by DESeq2 (p-value (padj) ≤ 0.05 and log2 fold change ≥2 or ≤−2). The ridgeline plot visualizes the distribution of differential enrichment categories identified by GSEA. Significantly differentially expressed pathways in ridgeline plots are indicated in green.

Figure 4

Wound-induced foliar gene expression. Four-week-old Arabidopsis thaliana wildtype (WT) or lox2 plants were wounded with a hole punch on each fully expanded rosette leaf and harvested at 1 and 12 h. Gene set enrichment analyses (GSEA) and ridgeline plots of WT (1 h) (A,E), WT (12 h) (B,F), lox2 (1 h) (C,G) and lox2 (12 h) (D,H). Wound-induced differentially expressed genes (DEGs) were determined by DESeq2 (p-value (padj) ≤ 0.05 and log2 fold change ≥2 or ≤−2). The ridgeline plot visualizes the distribution of differential enrichment categories identified by GSEA. Significantly differentially expressed pathways in ridgeline plots are indicated in green.

Figure 5

Wound-induced foliar gene expression: Patterns. Four-week-old Arabidopsis thaliana wildtype (WT) or lox2 plants were wounded with a hole punch on each fully expanded rosette leaf and harvested at 1 and 12 h. Wound-induced genes fell into six general expression patterns visualized by heatmaps. Early gene expression (peak at 1 h): (A) general-both genotypes, (B) WT and (C) lox2. Later gene expression (peak 12 h): (D) general-both genotypes, (E) WT and (F) lox2. Wound-induced genes were determined by DESeq2 (p-value (padj) ≤ 0.05 and log2 fold change ≥2 or ≤−2) (Supplemental Table S5). Heatmaps visualize gene expression (undamaged-U, wound-W, 1 h-1, 12 h-12).

Figure 5

Wound-induced foliar gene expression: Patterns. Four-week-old Arabidopsis thaliana wildtype (WT) or lox2 plants were wounded with a hole punch on each fully expanded rosette leaf and harvested at 1 and 12 h. Wound-induced genes fell into six general expression patterns visualized by heatmaps. Early gene expression (peak at 1 h): (A) general-both genotypes, (B) WT and (C) lox2. Later gene expression (peak 12 h): (D) general-both genotypes, (E) WT and (F) lox2. Wound-induced genes were determined by DESeq2 (p-value (padj) ≤ 0.05 and log2 fold change ≥2 or ≤−2) (Supplemental Table S5). Heatmaps visualize gene expression (undamaged-U, wound-W, 1 h-1, 12 h-12).

Figure 6

Wound-induced genes involved in polyphenol biosynthesis. Four-week-old Arabidopsis thaliana wildtype (WT) or lox2 plants were wounded with a hole punch on each fully expanded rosette leaf and harvested at 1 or 12 h. (A) Transcriptional regulators and (B) biosynthetic genes. Wound-induced genes were determined by DESeq2 (p-value (padj) ≤ 0.05 and log2 fold change ≥2 or ≤−2). Heatmaps visualize wound-induced gene expression (wildtype—1 h (WT-1), wildtype—12 h (WT-12), lox2—1 h (lox2-1), lox2—12 h (lox2-12)). Genes: TTG2 (At2g37260), TT8 (At4g09820), GL3 (At5g41315), PAP1 (At1g56650), MYB113 (At1g66379), 4CL (At1g20490), TT7 (At5g07990), DFR (At5g42800), LDOX (At4g22880), UF3GT (At5g54060), At4g14090, 3AT1 (At1g03840), 3AT2 (At1g03495), GSTF12 (At5g17220), 5MAT (At3g29590), MYB15 (At3g23250), CAD8 (At4g37990), PRX52 (At5g05340), SST/SCPL9 (At2g23010).

Figure 7

Foliar glucosinolate (GSL) levels and expression of GSL biosynthesis genes. Four-week-old Arabidopsis thaliana wildtype (WT) or lox2 plants were wounded with a hole punch on each fully expanded rosette leaf and harvested at 1 and 12 h. (A) Foliar aliphatic and indolic GSL levels in undamaged (U) and wounded (W) WT and lox2 plants taken 12 h after mechanical damage. (B) GSL pathway illustrating transcriptional activators, biosynthetic enzymes and other GSL-related proteins. Gene expression in (C) undamaged and (D) mechanically damaged arabidopsis rosettes. Glucosinolate levels are represented by the mean ± SE. Differences in GSL levels were determined by a two-factor analysis of variance (2-factor ANOVA) (factors: genotype (WT or lox2), treatment) followed by Tukey HSD (Supplemental Table S3). Heatmaps visualize gene expression (for C: wildtype—1 h (WT-1), wildtype—12 h (WT-12), lox2—1 h (lox2-1), lox2—12 h (lox2-12); for D: wildtype undamaged—WT-U, wildtype wounded—WT-W, lox2 undamaged—lox2-U, lox2 wounded—lox2-W)). Genes: BCAT4 (At3g19710), MAM1 (At5g23010), MAM3 (At5g23020), CYP79F1 (At1g16410), CYP79F2 (At1g16400), CYP83A1 (At4g13770), SUR1 (At2g20610), UGT74B1 (At1g24100), UGT14C1 (At2g31790), SOT18 (At1g74090), SOT17 (At1g18590), FMO_GS-OX1 (At1g65860), FMO_GS-OX5 (At1g12140), APO2 (At5g57930), MYB28 (At5g61420), MYB29 (At5g07690), MYB76 (At5g07700), NPF2.10 (At3g47960), JAL23 (At2g39330), CYP79B2 (At4g39950), CYP79B3 (At2g22330), CYP83B1 (At4g31500), UGT74B1 (At1g24100), SOT16 (At1g74100), CYP81F4 (At4g37410), CYP81F1 (At4g37430), CYP81F2 (At5g57220), CYP81F3 (At4g37400), IGMT1 (At1g21100), MYB34 (At5g60890), MYB51 (At1g18570).

Figure 7

Foliar glucosinolate (GSL) levels and expression of GSL biosynthesis genes. Four-week-old Arabidopsis thaliana wildtype (WT) or lox2 plants were wounded with a hole punch on each fully expanded rosette leaf and harvested at 1 and 12 h. (A) Foliar aliphatic and indolic GSL levels in undamaged (U) and wounded (W) WT and lox2 plants taken 12 h after mechanical damage. (B) GSL pathway illustrating transcriptional activators, biosynthetic enzymes and other GSL-related proteins. Gene expression in (C) undamaged and (D) mechanically damaged arabidopsis rosettes. Glucosinolate levels are represented by the mean ± SE. Differences in GSL levels were determined by a two-factor analysis of variance (2-factor ANOVA) (factors: genotype (WT or lox2), treatment) followed by Tukey HSD (Supplemental Table S3). Heatmaps visualize gene expression (for C: wildtype—1 h (WT-1), wildtype—12 h (WT-12), lox2—1 h (lox2-1), lox2—12 h (lox2-12); for D: wildtype undamaged—WT-U, wildtype wounded—WT-W, lox2 undamaged—lox2-U, lox2 wounded—lox2-W)). Genes: BCAT4 (At3g19710), MAM1 (At5g23010), MAM3 (At5g23020), CYP79F1 (At1g16410), CYP79F2 (At1g16400), CYP83A1 (At4g13770), SUR1 (At2g20610), UGT74B1 (At1g24100), UGT14C1 (At2g31790), SOT18 (At1g74090), SOT17 (At1g18590), FMO_GS-OX1 (At1g65860), FMO_GS-OX5 (At1g12140), APO2 (At5g57930), MYB28 (At5g61420), MYB29 (At5g07690), MYB76 (At5g07700), NPF2.10 (At3g47960), JAL23 (At2g39330), CYP79B2 (At4g39950), CYP79B3 (At2g22330), CYP83B1 (At4g31500), UGT74B1 (At1g24100), SOT16 (At1g74100), CYP81F4 (At4g37410), CYP81F1 (At4g37430), CYP81F2 (At5g57220), CYP81F3 (At4g37400), IGMT1 (At1g21100), MYB34 (At5g60890), MYB51 (At1g18570).

Figure 8

Wound-induced genes involved in plant–insect interactions in arabidopsis WT and lox2 mutants. Foliar gene expression in wounded lox2 mutants, which has a truncated non-functional enzyme, was compared to WT plants. Proteins encoded by early genes (1 h) are shown in blue, whereas later transcript expression (12 h) is depicted in brown. (A). General wound-induced responses found in both genotypes. Jasmonate biosynthesis begins in the chloroplast from 18C membrane lipid-derived precursors, typically α-linolenic acid, to finally form one of the biologically active jasmonates, jasmonoyl-isoleucine (JA-Ile). Numerous genes encoding enzymes in the jasmonate biosynthetic pathway are upregulated in wounded plants. JA-Ile enters the nucleus and forms a bridge between the SCF^COI1 and jasmonate ZIM domain (JAZ) repressors, leading to their degradation through the proteasome. The degradation of JAZ proteins releases MYC2/3/4 transcription factors, leading to jasmonate-responsive gene expression. The abscisic acid (ABA)-PYL6 receptor complex positively interacts with MYC2. Genes encoding JAZ-negative regulators as well as those further metabolizing jasmonic acid (JA) (i.e., JOX, ST2A) to inactive derivatives are also wound-induced. Gibberellin (GA) bound to its receptor GID1 activates a pathway that leads to the proteasome-mediated degradation of negative DELLA growth regulators, such as RGL3. In wounded leaves, the increase in RGL3 expression and genes that encode GA2OX6 and GA2OX8, involved in gibberellin metabolism to inactive products, results in the suppression of plant growth. Numerous genes involved in oxidative stress are wound-induced. For example, the oxidative stress-associated transcription factor RRTF1 and MDAR3, which are part of the Foyer–Halliwell–Asada cycle, a series of interconnected enzymatic reactions to detoxify the reactive oxygen species hydrogen peroxide (H₂O₂). Wound-induced MYB15 leads to the expression of ELI3/CAD8 and PRX52, which contribute to lignin biosynthesis. TPS04 and CYP82G1 are involved in the biosynthesis of volatiles, such as 4,8,12-trimethyltrideca-1,3,7,11-tetraene (TMTT), that are attractive to natural enemies of the herbivorous insect. TI1 and NATA1 contribute to plant resistance against arthropods, whereas MAPKKK21 is a negative regulator of arthropod resistance. (B). LOX2-specific responses. Wound-induced genes expressed at higher levels in WT compared with lox2 plants. Wound-induced expression of genes that encode LOX2 and MYC2 involved in jasmonate biosynthesis and signaling, respectively, is noted. CORI3 is a cysteine lyase involved in cysteine biosynthesis that produces precursors for ethylene, aliphatic glucosinolates (GSLs) and reduced glutathione (GSH) involved in the Foyer–Halliwell–Asada cycle. Also, the expression of the gene encoding DHAR1 in this pathway is wound-induced. Other cellular antioxidants whose biosynthetic pathway is positively regulated in these plants are anthocyanins. Indirect defenses involved in volatile biosynthesis are wound-induced. An increase the genes that encode proteins involved in antinutritive defenses, such as CLH1, ARGAH2 and TIs, as well as MAPKKK17, an important signaling kinase involved in arthropod resistance, is also seen. Abbreviations: ABA: abscisic acid, AOC: allene oxide cyclase, AOS: allene oxide synthase, DHAR1: dehydroascorbate reductase1, GA: gibberellin, GSH: reduced glutathione, GSL: glucosinolate, GSSG: oxidized glutathione, JA: jasmonic acid, JA-Ile: jasmonoyl-isoleucine, JAZ: jasmonate-Zim domain, JOX: jasmonate oxidase, LOX: lipoxygenase, OPDA: 12-oxo-phytodienoic acid, OPR: oxo-phytodienoate reductase, OPCL: OPC-8-CoA ligase, PUFA: polyunsaturated fatty acid, TF: transcription factor, TMTT: 4,8,12-trimethyltrideca-1,3,7,11-tetraene.