Flagellin peptide flg22 gains access to long-distance trafficking in Arabidopsis via its receptor, FLS2

Abstract

Diverse pathogen-derived molecules, such as bacterial flagellin and its conserved peptide flg22, are recognized in plants via plasma membrane receptors and induce both local and systemic immune responses. The fate of such ligands was unknown: whether and by what mechanism(s) they enter plant cells and whether they are transported to distal tissues. We used biologically active fluorophore and radiolabeled peptides to establish that flg22 moves to distal organs with the closest vascular connections. Remarkably, entry into the plant cell via endocytosis together with the FLS2 receptor is needed for delivery to vascular tissue and long-distance transport of flg22. This contrasts with known routes of long distance transport of other non-cell-permeant molecules in plants, which require membrane-localized transporters for entry to vascular tissue. Thus, a plasma membrane receptor acts as a transporter to enable access of its ligand to distal trafficking routes.

Keywords: Arabidopsis; FLS2; Pseudomonas syringae; flagellin; flg22; long-distance traffic; peptide; receptor-mediated endocytosis; transport..

Fig. 1.

flg22 transport in plants requires FLS2 receptor. flg22 is detected in vascular regions. (A) FAM–flg22 (Ff22) was detected on the adaxial side of a detached Col leaf 1 h after drop application of 10 μM peptide to the abaxial side. Migration of FAM–scrambled peptide (Fscr) in Col and FAM–flg22 in fls2 and bak1 was significantly lower. FAM fluorescence was quantified in epifluorescence microscopy images of adaxial (distal) leaf side; data combined from three experiments are shown as percentage of FAM–flg22 signal in Col. Background fluorescence (mock treatment without peptides) is shown by a dashed line. Letters indicate significant difference (P<0.01, ANOVA/Tukey’s test, n≥5). (B, C) FAM–flg22 transport to distal leaves is less efficient in the fls2 mutant. FAM–flg22 or FAM–Y-flg22 (5–10 μM) was infiltrated into one leaf of soil-grown plants, and fluorescence was quantified after 16 h in epifluorescence microscopy images of distal orthostichous leaves (such as in C); data combined from three experiments are shown as percentage of signal in Col. *P<0.02 (t-test, n=12). Background fluorescence is shown by dashed line. (D) Two hours after infiltration of 10 μM FAM–flg22 or FAM–Y-flg22 into a part of Col leaf, FAM fluorescence was detected inside vascular cells outside the infiltrated area. FAM–flg22 in fls2 and FAM–scrambled peptide in Col were rarely detectable outside infiltration area and only between the cells. A composite of confocal microscopy image with differential interference contrast (DIC), green FAM fluorescence and red chloroplast (chl) autofluorescence is shown. Bar: 20 μm. (E) TAMRA–flg22 is detected predominantly in vascular region of Col. Five micomolar TAMRA–flg22 was applied on a filter paper disc placed on the adaxial side of a leaf of in vitro-grown Arabidopsis plants and confocal images of treated tissue (adaxial side underneath filter paper) were obtained 1 d later, after washing the leaf with water. The distinct vein pattern in Col was observed in 4 out of 7 experiments. Bar: 50 μm.

Fig. 2.

flg22 long-distance movement depends on receptors and internalization. Five micromolar TAMRA–flg22 (Tf22) with or without inhibitors was applied on a filter paper disc placed on one leaf of in vitro-grown Arabidopsis, and orthostichous distal tissues (leaves and flowers) were imaged by confocal microscopy after 1 d. TAMRA–flg22 was transported more efficiently to distal tissue in Col than in fls2 and bak1 or in Col treated with inhibitors. (A) Representative confocal microscopy images of distal tissues 1 d after peptide application. Bar: 50 μm. (B) TAMRA fluorescence was quantified in images such as in (A) from at least four experiments (except BDM treatment in two experiments) and presented as percentage of fluorescence in distal tissue of Col. Letters indicate significant difference (P<0.0002, n≥7, ANOVA/Tukey’s test).

Fig. 3.

flg22 movement is specific and FLS2-dependent. Labeled peptide is not degraded in plants. (A, B) ¹²⁵I-Y-flg22 is detected 3 h after application in distal orthostichous leaves of Col but not fls2. Scrambled peptide (scr) is not mobile. A 1 μl droplet of 0.5 μM radiolabeled peptide was applied to the abaxial side of one leaf (arrow) of plants grown in soil and imaged by autoradiography after 3 h (A). The highest signal was detected in orthostichous leaves of flg22-treated Col. Percentage of radioactivity in distal leaves was quantified in six plants from four experiments (B). Letters indicate significant difference (P<0.05, ANOVA). (C) Transport of control Na¹²⁵I is not affected by FLS2. A 1 μl droplet of Na¹²⁵I (1 × 10⁶ cpm) was applied to the abaxial side of one leaf (as in A) and plants were incubated for 3 h. ¹²⁵I signal was quantified in distal leaves of five plants from three experiments (P=0.7, t-test). (D, E) flg22 is not degraded and is detected in distal parts of droplet-treated Col. Five microliters of 3 μM ¹²⁵I-Y-flg22 was applied to Col plants (as in A) and after 3 h the treated leaf was removed and cut in two parts (dashed line, D). Proteins extracted from both parts of the treated leaf and from distal orthostichous leaves were separated by SDS-PAGE and detected by autoradiography (E). Parts of the same gel (same exposure) are shown in (E). T, extract from treated leaf segment (0.13% of sample); D, extract from distal region of the treated leaf (40% of sample); DL, extract from distal orthostichous leaves (80% of sample); –, control ¹²⁵I-Y-flg22 (not extracted from plants). Unmodified ¹²⁵I-Y-flg22 (~2 kDa, red arrow) and larger band (~12 kDa) were detected in extracts from treated and distal tissue of five plants in two experiments.

Fig. 4.

flg22 is internalized together with FLS2. Leaf discs of N. benthamiana (A) and Arabidopsis Ws expressing FLS2–GFP (B, E) or control Ws (C) were floated on 2 μM TAMRA–flg22 (Tf22) for the indicated time, washed in water and imaged by confocal microscopy. Images in red (TAMRA) and green (GFP) channels, and their composite with or without DIC image are shown. Graphs show fluorescence profiles along yellow line. Bar: 20 μm in (A, E) and 10 μm in (B, C). These experiments were repeated three to four times. (A) TAMRA–flg22 colocalizes at plasma membrane with FLS2–GFP after several minutes to 1 h incubation. See additional images in Supplementary Figs S5 and S6. (B) TAMRA–flg22 is internalized and colocalizes with FLS2–GFP in vesicles (line 2, two vesicles along the line marked with an arrowhead and an arrow) and MVBs (line 1) after 1–2 h incubation. See additional images in Supplementary Fig. S7. (C) TAMRA–flg22 does not enter wild-type Ws cells that lack FLS2 after 1–2 h incubation. See additional images in Supplementary Figs S6 and S7. (D) TAMRA–flg22 colocalizes with FLS2–GFP after floating Ws/FLS2–GFP leaf discs on peptide solution for 1–2 h. Fluorescent foci were counted in 16 picture areas (20 μm × 20 μm) for each genotype. Total number of foci: Ws/FLS2–GFP: 152 green (FLS2–GFP), 161 red (Tf22), 143 overlap; Ws: 2 green, 5 red, 0 overlap. Error bars, SE, *P<0.0001 (t-test, n=16). (E) After overnight incubation, TAMRA–flg22 fills the cells, whereas FLS2–GFP is mostly detected at the plasma membrane.

Fig. 5.

flg22 undergoes endocytosis and accumulates in vacuoles. Col leaf discs were vacuum infiltrated with fluorophore-labeled flg22 and a dye prior to floating on flg22/dye solution for the indicated time. Confocal microscopy images in red (TAMRA, FM4-64) and green (FAM, BCECF) channels and a composite of fluorescence channels with or without DIC are shown. Graphs show fluorescence profiles along yellow line. Bar: 10 μm in (A–C, E, F) and 20 μm in (G). These experiments were repeated twice. (A) FAM–flg22 (Ff22, 5–10 μM) colocalizes in vesicles with plasma membrane/endosome dye FM4-64 (6–8 μM) after 1 h incubation. (B) Later (after 1–2 h), FAM–flg22 colocalizes with FM4-64 in endosomes and MVBs (arrowhead). Not all endosomes and MVBs contain FAM–flg22 (arrow). (C) FAM–flg22 is not detected inside fls2 cells after 1–2 h incubation. (D) FAM–flg22 colocalizes with endocytic dye FM4-64 in Col leaf discs after 1–2 h incubation. Fluorescent foci were counted in 18 picture areas (20 μm × 20 μm) for each genotype. Total number of foci: Col: 217 green (Ff22), 325 red (FM4-64), 212 overlap; fls2: 10 green, 289 red, 2 overlap. Error bars, SE, *P<0.0001 (t-test, n=18). (E, F) TAMRA–flg22 (Tf22, 5 μM) colocalizes with the vacuolar dye BCECF (40 μM) in MVBs (E) and prevacuolar compartments (F) after 2–4 h incubation. TAMRA–flg22 is also found in endosomes (narrow red peaks in fluorescence profile) moving to MVBs (E). (G) After overnight incubation, TAMRA–flg22 (1 μM) colocalizes with BCECF (20 μM) in central vacuoles.

Fig. 6.

Accumulation of flg22 in plant cells requires FLS2 and endocytosis. Arabidopsis leaf discs were floated on TAMRA–flg22 (Tf22) with and without an endocytosis inhibitor for 1–2 h (A, B; 5 μM TAMRA–flg22) or overnight (C, D; 1–2 μM TAMRA–flg22) and washed in water. Fifty micromolar MG132 or 50 mM BDM was added simultaneously with TAMRA–flg22. Fluorescence was quantified in confocal microscopy images such as in (A, C); data were combined from at least two experiments for each genotype/treatment and shown as percentage of TAMRA–flg22 signal in Col. Letters indicate significant difference (ANOVA/Tukey’s test). Background fluorescence in the absence of TAMRA–flg22 was negligible. Bar: 10 μm in (A) and 20 μm in (C). (A, B) Initial (up to 1.5 h) uptake/binding of TAMRA–flg22 is reduced in fls2, but not highly affected in bak1 or in Col treated with inhibitors. P<0.05, n≥6. (C, D) Overnight accumulation of TAMRA–flg22 in plant cells is inhibited in fls2 and bak1 and in Col treated with inhibitors. P<0.0002, n≥7.

Fig. 7.

Defense activation does not increase peptide accumulation. Only the active peptide accumulates in Col cells. Defense activation does not make cells permeable to inactive peptide or dye. Leaf discs were floated on indicated mixed peptide solution overnight; 1 μM TAMRA–flg22 (T-flg22), 2 μM FAM–flg22 (F-flg22), 2 μM FAM–scrambled peptide (F-scr), 2 μM flg22 and 2 μM fluorescein were used. (A) Confocal microscopy images in red (TAMRA) and green (FAM, fluorescein) channels and a composite of fluorescence channels with or without DIC are shown. (B) TAMRA and FAM fluorescence was quantified in confocal microscopy images of the same field. Data from at least three experiments per treatment were combined and presented as percentage of fluorescence in leaf discs treated with active peptide alone. Letters show significant difference (P<0.05, ANOVA/Tukey’s test, n≥8). Dashed line is background fluorescence (mock treatment without peptides). Mixed active peptides show lower fluorescence than single peptide in each channel due to competition of TAMRA–flg22 (T-f22) and FAM–flg22 (F-f22). f22 is flg22.

Fig. 8.

Defense activation does not increase peptide transport. Only active TAMRA–flg22 (5 μM) moves to distal tissue in Col when mixed with FAM–scrambled peptide (10 μM). Peptide transport was not observed in fls2 and bak1. Mixed peptides were applied on a filter paper disc placed on one leaf of in vitro-grown plants. (A) Confocal microscopy images of distal orthostichous leaves in red and green channels and a composite of fluorescence channels with DIC are shown. (B) TAMRA and FAM fluorescence was quantified in red and green channels in the same confocal microscopy images of distal orthostichous tissues 1 d after application. Letters indicate significant difference (calculated separately for TAMRA and FAM; P<0.05, ANOVA/Tukey’s test, n=6).