Mechanosensory trichome cells evoke a mechanical stimuli-induced immune response in Arabidopsis thaliana

Abstract

Perception of pathogen-derived ligands by corresponding host receptors is a pivotal strategy in eukaryotic innate immunity. In plants, this is complemented by circadian anticipation of infection timing, promoting basal resistance even in the absence of pathogen threat. Here, we report that trichomes, hair-like structures on the epidermis, directly sense external mechanical forces, including raindrops, to anticipate pathogen infections in Arabidopsis thaliana. Exposure of leaf surfaces to mechanical stimuli initiates the concentric propagation of intercellular calcium waves away from trichomes to induce defence-related genes. Propagating calcium waves enable effective immunity against pathogenic microbes through the CALMODULIN-BINDING TRANSCRIPTION ACTIVATOR 3 (CAMTA3) and mitogen-activated protein kinases. We propose an early layer of plant immunity in which trichomes function as mechanosensory cells that detect potential risks.

Fig. 1. Raindrop- and mechanical stimuli (MS)-induced transcriptome profiles highly correlate with each other.

a Enriched Gene Ontology (GO) categories of 1050 raindrop (10 falling droplets)-induced genes in the wild type (Col-0) 15 min after treatment. The top 10 categories are shown in ascending order of P values (one-sided hypergeometric test). b, c Transcript levels of MS-induced and defense-related genes in 4-week-old Col-0 plants 15 min after being treated with 10 falling droplets (raindrop, b) or 1 brushing (MS, c), determined by RT-qPCR and normalized to UBIQUITIN 5 (UBQ5). Data are presented as mean ± SD. n = 6 plants examined over three independent experiments. Each dot indicates a technical replicate. d Enriched GO categories of 1241 MS (1 brushing)-induced genes in Col-0 15 min after treatment. The top ten categories are shown in ascending order of P values (one-sided hypergeometric test). e Venn diagram of the overlap between transcriptome datasets from raindrop- and MS-induced genes (likelihood ratio test; P < 0.05). f Radar chart of intensity compared with mock (log₂FC) and Pearson correlation coefficient (r = 0.917288423) of 917 raindrop- and MS-induced genes (left). Intensities of major immune regulator genes induced by raindrops and MS in RNA-seq analysis (log₂FC) (likelihood ratio test; P < 0.05) (right).

Fig. 2. Raindrop- and MS-induced mechanosensation triggers defense responses.

a, b Venn diagram (a) and the upset plot (b) between 917 raindrop- and MS-induced genes and transcriptome datasets obtained from salicylic acid (SA), jasmonic acid (JA), and flg22 (PAMP) treatment and Pseudomonas syringae pv. maculicola ES4326 infection (P < 0.05). Overlap with raindrop- and MS-induced genes: SA, 21%, 193/917 genes; JA, 11.8%, 108/917 genes; flg22, 37%, 339/917 genes; Psm ES4326, 25.8%, 237/917 genes; any of the four factors, 58.6%, 537/917 genes. c Accumulation of plant hormones (ng/g fresh weight) SA, JA, JA-isoleucine (JA-Ile), abscisic acid (ABA), and indole-3-acetic acid (IAA) 5 min after treatment with 10 falling droplets (raindrop), 1 brushing (MS), or cutting (wounding). Data are presented as mean ± SD. n = 6 plants examined over three independent experiments. Each dot indicates a biological replicate. Different letters above bars indicate significant differences (one-sided Tukey’s multiple comparison test; P < 0.05). d, e Raindrop (4 droplets)- (d) and MS (4 brushings)-induced (e). MAPK activation in Col-0. Total proteins were extracted from 4-week-old plants treated with raindrops and detected by immunoblot analysis with anti-phospho-p44/42 MAPK antibodies. Relative phosphorylation levels are shown below each blot. Similar results were obtained in four independent experiments. f MS-induced MAPK activation in Col-0, fls2, and bak1–3. Total proteins were extracted from 4-week-old plants after 5 min of MS treatment (1 brushing) and detected by immunoblot analysis with anti-phospho-p44/42 MAPK antibodies. Relative phosphorylation levels are shown below each blot. Similar results were obtained in three independent experiments. g, h Disease progression of Alternaria brassicicola in Col-0 leaves 3 days after inoculation with (+) or without (−) raindrop (10 falling droplets) pretreatment (g) or with (+) or without (−) MS (4 brushings) pretreatment (h). Error bars represent SE. Asterisks indicate significant difference (two-sided Tukey’s t test; ****P < 0.0001). i Growth of Psm ES4326 in Col-0 leaves 2 days after inoculation with (+) or without (−) MS (4 brushings) pretreatment. Error bars represent SE. Asterisks indicate significant difference (two-sided Tukey’s t test; ****P < 0.0001). Cfu colony-forming units. n = 29 (g), (−): n = 13, (+): n = 12 (h), and n = 8 (i) samples examined over three independent experiments. Each dot indicates a biological replicate.

Fig. 3. MS-induced genes are regulated by CAMTA3.

a Promoter analysis of the top 300 (among 917 genes) raindrop- and MS-induced genes in terms of expression levels revealed that the CAMTA-binding CGCG box [CGC(/T)GT] was overrepresented among these genes. b, c Transcript levels of WRKY33 and CBP60g in 4-week-old camta2 camta3 CAMTA3pro:CAMTA3-GFP (CAMTA3) and camta2 camta3 CAMTA3pro:CAMTA3^A855V-GFP [CAMTA3(A855V)] plants 15 min after the plants were treated with 1 falling droplet (b) or brushed 4 times (c), determined by RT-qPCR and normalized to UBQ5 transcript levels in Col-0 Mock. Data are presented as mean ± SD. Asterisks indicate significant difference (one-sided Tukey’s multiple comparison test; **P < 0.01, ***P < 0.001). n = 6 plants examined over three independent experiments. Each dot indicates a technical replicate. d Venn diagram depicting the overlap between genes with CAMTA3-binding sites in their promoters, as determined by ChIP-seq, and raindrop- and MS-induced genes as determined by RNA-seq. A total of 314 genes, shown in red, were identified as CAMTA3-target genes. e The CGCG box was identified as an overrepresented motif among the sequence peaks of 314 genes by MEME-ChIP. f Localization of CAMTA3 on the promoters of the MS-induced genes TCH2, TCH4, CAM2, CBP60g, CML23, and WRKY40, as representative of the 314 genes shown in (d). Blue and red lines indicate CGCGT and CGTGT, respectively. g Growth of Psm ES4326 in camta2 camta3 CAMTA3pro:CAMTA3-GFP (CAMTA3) and camta2 camta3 CAMTA3pro:CAMTA3^A855V-GFP [CAMTA3(A855V)] plants 2 days after inoculation with (+) or without (−) MS (4 brushing) pretreatment. Error bars represent SE. Asterisks indicate a significant difference (one-sided Tukey’s multiple comparison test and two-way ANOVA; ****P < 0.0001). Cfu colony-forming units, NS not significant. CAMTA3 (−, +), CAMTA3(A855V) (−): n = 8, CAMTA3(A855V) (+): n = 6 samples examined over three independent experiments. Each dot indicates a biological replicate.

Fig. 4. Trichomes initiate intercellular calcium waves.

a–c YFP fluorescence in a whole leaf from WRKY33pro:EYFP-NLS (Col-0) with (MS, bottom half) or without brushing (untreated, top half) (a), along with zoomed-in views of brushed (b), and untreated (c) areas. Arrowheads indicate trichomes (b, c). Scale bars, 0.5 mm (a), 0.3 mm (b, c). Similar results were obtained in three independent experiments. d Ca²⁺ imaging using 35Spro:GCaMP3 (Col-0). The leaf surface of a 4-week-old plant was treated with MS by brushing. MS-induced intercellular calcium waves propagated concentrically from trichomes. Scale bar, 1.0 mm. See also Supplementary Movie 1. e Ca²⁺ imaging using 35Spro:GCaMP3 (Col-0). A single trichome from a 2-week-old seedling was flicked with a silver chloride wire. MS-induced intercellular calcium waves propagated concentrically from the trichome (dashed outline). Scale bar, 0.2 mm. See also Supplementary Movie 2. Similar results were obtained in three independent experiments. f [Ca²⁺]_cyt changes at sites indicated by numbers in (e). Similar results were obtained in three independent experiments. g Side view of a trichome whose neck was flicked with a silver chloride wire. MS-induced intercellular Ca²⁺ influx was transiently observed in the trichome base (arrowheads) followed by the formation of circular waves. Scale bar, 0.1 mm. See also Supplementary Movie 3. Similar results were obtained in three independent experiments.

Fig. 5. Trichomes are mechanosensory cells.

a Ca²⁺ imaging using 35Spro:GCaMP3 (Col-0) and 35Spro:GCaMP3 (gl1). Leaf surfaces were exposed to MS by brushing. MS-induced calcium waves were compromised in the gl1 mutant. See also Supplementary Movies 4 and 5. Scale bars, 0.5 mm. b [Ca²⁺]_cyt signature of (a). c Venn diagram of transcriptome datasets for MS-induced genes in Col-0 and gl1 (likelihood ratio test; P < 0.05). NS, not significant. Lower, fold change (FC) (gl1)/FC (Col-0) < 0.5. High, MS (gl1)/Mock (gl1), log₂FC ≥ 1 in gl1 (likelihood ratio test; P < 0.05). d Heatmap of differentially expressed defense-related genes obtained from transcriptome datasets from Col-0 and gl1 plants treated with MS (4 brushings). e Transcript levels of WRKY33, WRKY40, and CBP60g in 4-week-old Col-0 and gl1 plants 15 min after treatment with brushing 4 times, determined using RT-qPCR and normalized to UBQ5. Data are presented as mean ± SD. Asterisks indicate significant difference (one-sided Tukey’s multiple comparison test; ***P < 0.001, ****P < 0.0001). n = 6 plants examined over three independent experiments. Each dot indicates a technical replicate. f MS-induced MAPK activation in Col-0 and gl1. Total proteins were extracted from 4-week-old leaves 5 min after MS (4 brushings) treatment and detected by immunoblot analysis with anti-phospho-p44/42 MAPK antibodies. Relative phosphorylation levels are shown below each blot. Similar results were obtained in three independent experiments. g Growth of Psm ES4326 in Col-0 and gl1 leaves 2 days after inoculation with (+) or without (−) MS (4 brushings) pretreatment. Error bars represent SE. Asterisks indicate a significant difference (one-sided Tukey’s t test and two-way ANOVA; ***P < 0.001). Cfu colony-forming units, NS not significant. n = 8 samples examined over three independent experiments. Each dot indicates a biological replicate. h Disease progression of Alternaria brassicicola in Col-0 and gl1 leaves 3 days after inoculation with (+) or without (−) MS (4 brushings) pretreatment. Error bars represent SE. Asterisks indicate a significant difference (one-sided Tukey’s t test and two-way ANOVA; ****P < 0.0001). NS not significant. Col-0: n = 15, gl1: n = 14 samples examined over three independent experiments. Each dot indicates a biological replicate.

Fig. 6. Mechanosensory trichomes evoke an immune response.

Model showing how trichomes directly sense the mechanical impact of raindrops as an emergency signal in anticipation of possible infections. Mechanosensory trichome cells initiate intercellular calcium waves in response to MS. [Ca²⁺]_cyt initiates the de-repression of Ca²⁺/CaM-dependent CAMTA3 and activates the phosphorylation of MPK3 and MPK6, thereby inducing WRKY-dependent transcription.