Interdependence of plasma membrane nanoscale dynamics of a kinase and its cognate substrate underlies Arabidopsis response to viral infection

Abstract

Plant viruses represent a risk to agricultural production and as only a few treatments exist, it is urgent to identify resistance mechanisms and factors. In plant immunity, plasma membrane (PM)-localized proteins play an essential role in sensing the extracellular threat presented by bacteria, fungi, or herbivores. Viruses are intracellular pathogens and as such the role of the plant PM in detection and resistance against viruses is often overlooked. We investigated the role of the partially PM-bound Calcium-dependent protein kinase 3 (CPK3) in viral infection and we discovered that it displayed a specific ability to hamper viral propagation over CPK isoforms that are involved in immune response to extracellular pathogens. More and more evidence supports that the lateral organization of PM proteins and lipids underlies signal transduction in plants. We showed here that CPK3 diffusion in the PM is reduced upon activation as well as upon viral infection and that such immobilization depended on its substrate, Remorin (REM1.2), a scaffold protein. Furthermore, we discovered that the viral infection induced a CPK3-dependent increase of REM1.2 PM diffusion. Such interdependence was also observable regarding viral propagation. This study unveils a complex relationship between a kinase and its substrate that contrasts with the commonly described co-stabilisation upon activation while it proposes a PM-based mechanism involved in decreased sensitivity to viral infection in plants.

Keywords: A. thaliana; kinases; lipid; membranes; nanodomains; plant biology; remorin; virus.

Figure 1.. Arabidopsis thaliana calcium-dependent protein kinase 3 (CPK3) is specifically involved in the restriction of plantago asiatica mosaic virus (PlAMV) cell-to-cell movement.

(A) Representative images of PlAMV-GFP infection foci at 5 dpi in the different mutant backgrounds. Scale bar = 500 µm. (B) Box plots of the mean area of PlAMV-GFP infection foci 5 d after infection in CPK multiple mutant lines, normalized to the mean area measured in Col-0. Three independent biological repeats were performed, with at least 47 foci per experiment and per genotype. Significant differences were revealed using a One-Way ANOVA followed by a Tukey’s multiple comparison test. Letters are used to discriminate between statistically different conditions (p<0.05). (C) Box plots of the mean area of PlAMV-GFP infection foci in cpk3-1 and cpk3-2 single mutants and in CPK3 over-expressing lines (Pro35S:CPK3-HA #8.2 and Pro35S:CPK3-HA #16.2), normalized to the mean area measured in Col-0. Three independent biological repeats were performed, with at least 56 foci per experiment and per genotype. Significant differences were revealed using a one-way ANOVA followed by a Tukey’s multiple comparison test. Letters are used to discriminate between statistically different conditions (p<0.05). (D) Representative images of A. thaliana plants infected with PlAMV-GFP and imaged with a CCD Camera from 10 to 17 dpi. The region of interest used for measurement of pixel intensity is circled with a white dotted line. Multicolored scale is used to enhance contrast and ranges from blue (low intensity) to red (high intensity). Scale bar = 4 cm. (E) Box plots of the mean cumulated intensity measured in infected leaves in Col-0, cpk3-2, and Pro35S:CPK3-HA #16.2 during systemic viral propagation. Two independent experiments were conducted. Statistical differences could be observed between the genotypes and time-points using a Kruskal-Wallis followed by a Dunn’s multiple comparison test (p<0.05). For clarity, only the results of the statistical test of the comparison of the different time-points (10, 14, and 17 dpi) within a genotype are displayed and are color-coded depending on the genotype. (F) Kinase activity of CPK3 dead variant. Recombinant proteins GST-CPK3 WT and K107M were incubated with REM1.2^1-118-6His in kinase reaction buffer in the presence of EGTA (-) or 100 µM free Ca²⁺ (+). Radioactivity is detected on dried gel (upper panel). The protein amount is monitored by Coomassie staining (lower panel). (G) Box plots of the mean area of PlAMV-GFP infection foci in cpk3-2 complemented lines cpk3-2/Pro35S:CPK3-myc and cpk3−2/Pro35S:CPK3^K107M-myc. Three independent biological repeats were performed, with at least 51 foci per experiment and per genotype. Significant differences were revealed using a one-way ANOVA followed by a Tukey’s multiple comparison test. Letters are used to discriminate between statistically different conditions (p<0.0001).

Figure 1—figure supplement 1.. Viral propagation experimental design.

Schematic representation of the method used for local and systemic plantago asiatica mosaic virus (PlAMV) propagation assays. Shorty, 3-wk-old A. thaliana leaves are infiltrated with PlAMV-GFP. For local propagation, the inoculated leaves are cut out at 5 dpi to observe individual foci using a macroscope. For systemic propagation, the whole plant is imaged with a CCD camera at 10, 14, and 17 dpi to follow the evolution of PlAMV propagation.

Figure 1—figure supplement 2.. Specificity of calcium-dependent protein kinase 3 (CPK) kinase activity towards REM1.2 in vitro.

Left panel: GST-CPKs recombinant proteins were incubated with REM1.2^1-118-6His or histone as substrates, in kinase reaction buffer in the presence of Ca²⁺. The radioactivity is detected on dried gel and the protein amount is monitored by Coomassie staining. Right panel: the radioactive signals were quantified with ImageQuant and the ratio between REM1.2 and histone phosphorylations is shown in a boxplot (n=4). Different letters indicate statistical differences (p<0.05, ncompar test from R).

Figure 1—figure supplement 3.. Calcium-dependent protein kinase 3 (CPK3) protein levels in knock-out and overexpressing lines.

Western blot of CPK3 on seedlings of cpk3-1 and cpk3-2 mutant lines along with Pro35S:CPK3-HA #8.2 and Pro35S:CPK3-HA #16.2 overexpressing lines. CPK3 was detected with anti-CPK3 antibody. Protein loading was monitored by Coomassie staining.

Figure 2.. Calcium-dependent protein kinase 3 (CPK3) diffusion decreases upon plantago asiatica mosaic virus (PlAMV) infection in Arabidopsis thaliana.

(A) Confocal images of the secant view of A. thaliana epidermal cells of the cpk3-2/CPK3-mRFP1.2 or cpk3-2/ProUbi10:CPK3^G2A-mRFP1.2. Scale bar = 5 µm. (B) Box plots of the mean area of PlAMV-GFP infection foci in cpk3-2 and complemented lines cpk3-2/ProUbi10:CPK3-mRFP1.2 and cpk3-2/ProUbi10:CPK3^G2A-mRFP1.2. Three independent biological repeats were performed, with at least 32 foci per experiment and per genotype. Significant differences were revealed using a one-way ANOVA followed by a Tukey’s multiple comparison test. Letters are used to discriminate between statistically different conditions (p<0.0001). (C) Confocal images of the surface view of A. thaliana epidermal cells of the Col-0/ProUbi10:CPK3-mRFP1.2, infiltrated either with free GFP (‘mock’) or PlAMV-GFP. Scale bar = 5 µm. (D) Representative trajectories of Col-0/CPK3-mEOS3.2 infiltrated either with free GFP (‘mock’) or PlAMV-GFP; Scale bar = 2 µm. (E) Distribution of the diffusion coefficient (D), represented as log(D) for Col-0/ProUbi10:CPK3-mEOS3.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. Data were acquired from at least 8086 trajectories obtained in at least 16 cells over the course of three independent experiments. (F) Box plot of the mean peak value extracted from the Gaussian fit of log(D) distribution. Significant difference was revealed using a Mann-Whitney test. *p<0.05. (G) Mean square displacement (MSD) over time of Col-0/ProUbi10:CPK3-mEOS3.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. Representative trajectories extracted from (D) illustrate each curve. Scale bar = 1 µm. Data were acquired from at least 16 cells over the course of three independent experiments. (H) Voronoi tessellation illustration of Col-0/ProUbi10:CPK3-mEOS3.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. ND are circled in red. Scale bar = 2 µm. (I) Distribution of the ND diameter of Col-0/ProUbi10:CPK3-mEOS3.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. (J) Box plot representing the mean peak value of nanodomains (ND) diameter extracted from the Gaussian fit of (I). No significant differences were revealed using a Mann-Whitney test. (K) Boxplot of the proportion of Col-0/ProUbi10:CPK3-mEOS3.2 detections found in ND 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. No significant differences were revealed using a Mann-Whitney test.

Figure 2—figure supplement 1.. Calcium-dependent protein kinase 3 (CPK3) transcript and protein levels are not modified upon viral infection.

(A) Transcript levels of CPK3 in leaves infiltrated with either free GFP or PlAMV-GFP and harvested 7 d after infection. Three independent experiments were performed with three independent plants for each condition per experiment. Transcript levels were normalized to actin transcripts. No significant difference could be deciphered using a Mann-Whitney test. (B) Left panel: CPK3 protein levels were determined by quantification of western-blot signal of Col-0 plants infiltrated with free GFP (‘mock’) or PlAMV-GFP blotted with an anti-CPK3-specific antibody. Three independent assays were performed. No difference could be deciphered using a Mann-Whitney test. Right panel: representative western blot. Protein loading was monitored by stain-free staining.

Figure 2—figure supplement 2.. Calcium-dependent protein kinase 3 (CPK3) does not accumulate at plasmodesmata during plantago asiatica mosaic virus (PlAMV) infection.

Confocal images of the secant view of A. thaliana epidermal cells of the cpk3-2/CPK3-mRFP1.2 stained with aniline blue 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. Scale bar = 5 µm.

Figure 3.. Plantago asiatica mosaic virus (PlAMV)-induced activation of calcium-dependent protein kinase 3 (CPK3) in N. benthamiana induces its confinement and clustering in plasma membrane (PM) domains.

(A) Confocal images of the surface view of N. benthamiana epidermal cells transiently expressing ProUbi10:CPK3-mRFP1.2, ProUbi10:CPK3-mRFP1.2+PlAMV GFP or ProUbi10:CPK3^CA-mRFP1.2. Scale bar = 5 µm. (B) Representative trajectories of ProUbi10:CPK3-mEOS3.2, ProUbi10:CPK3.2-mEOS3.2+PlAMV and ProUbi10:CPK3^CA-mEOS3.2. Scale bar = 2 µm. (C) Distribution of the diffusion coefficient (D), represented as log(D) for ProUbi10:CPK3-mEOS3.2, ProUbi10:CPK3-mEOS3.2+PlAMV GFP and ProUbi10:CPK3^CA-mEOS3.2. Data were acquired from at least 6144 trajectories obtained in at least 15 cells over the course of three independent experiments. (D) Box plots of the fraction of immobile proteins (log(D)p<0.005). (E) Mean square displacement (MSD) over time of fast (circle) or slow (triangle) diffusing ProUbi10:CPK3-mEOS3.2, ProUbi10:CPK3-mEOS3.2+PlAMV GFP, ProUbi10:CPK3^CA-mEOS3.2. Standard error is displayed from three independent experiments. (F) Voronoï tessellation illustration of ProUbi10:CPK3-mEOS3.2, ProUbi10:CPK3-mEOS3.2+PlAMV GFP and ProUbi10:CPK3^CA-mEOS3.2. ND are circled in red. Scale bar = 2 µm. (G) Distribution of the ND diameter of ProUbi10:CPK3-mEOS3.2, ProUbi10:CPK3-mEOS3.2+PlAMV GFP and ProUbi10:CPK3^CA-mEOS3.2. (H) Box plot representing the mean peak value of ND diameter extracted from the Gaussian fit of (G). No significant differences were revealed using a Kruskal-Wallis followed by a Dunn’s multiple comparison test.(I) Boxplot of the proportion of detections found in ND of ProUbi10:CPK3-mEOS3.2, ProUbi10:CPK3-mEOS3.2+PlAMV GFP and ProUbi10:CPK3^CA-mEOS3.2. Significant differences were revealed using a Kruskal-Wallis followed by a Dunn’s multiple comparison test. Letters are used to discriminate between statistically different conditions (p<0.005). (J) Left: Confocal images of the surface view of N. benthamiana epidermal cells transiently expressing ProUbi10:CPK3^CA-mRFP1.2 and infiltrated with either DMSO or 50 µg/mL fenpropimorph. Scale bar = 5 µm; Right: Box plot of the mean spatial clustering index (SCI) of CPK3^CA. At least three experiments were performed, with at least 10 cells per experiment; statistical significance was determined using a Student t-test, ****:p<0.0001. (K) Left: Confocal images of the surface view of N. benthamiana epidermal cells transiently co-expressing ProUbi10:CPK3^CA-mRFP1.2 with active or dead SAC1, mutated for its phosphatase activity. Scale bar = 5 µm; Right: Box plot of the mean SCI of CPK3^CA. At least three experiments were performed, with at least 10 cells per experiment; statistical significance was determined using a Student t-test, ****p<0.0001.

Figure 3—figure supplement 1.. Calcium-dependent protein kinase 3 (CPK3) and CPK3^CA display a similar subcellular localization.

Confocal images of the secant view of N. benthamiana epidermal cells transiently expressing ProUbi10:CPK3-mRFP1.2 or ProUbi10:CPK3^CA-mRFP1.2. Scale bar = 5 µm.

Figure 3—figure supplement 2.. Calcium-dependent protein kinase 3 (CPK3) and CPK3CA display a similar subcellular localization.

Figure 4.. Group 1 REMORIN (REM) hampers plantago asiatica mosaic virus (PlAMV)-GFP cell-to-cell propagation and REM1.2 diffusion increases upon infection.

(A) Box plots of the mean area of PlAMV-GFP infection foci in rem1.2, rem1.3, rem1.4 single mutants along with rem1.2 rem1.3 double mutant and rem1.2 rem1.3 rem1.4 triple mutant. Three independent biological repeats were performed, with at least 36 foci per experiment and per genotype. Significant differences were revealed using a one-way ANOVA followed by a Tukey’s multiple comparison test. Letters are used to discriminate between statistically different conditions (p<0.05). (B) Box plots of the mean cumulated intensity measured in infected leaves in Col-0 and rem1.2 rem1.3 rem1.4 during systemic viral propagation. Two independent experiments were conducted. Statistical significance of the difference between Col-0 and rem1.2 rem1.3 rem1.4 at each time point was assessed using a Mann-Whitney test. *p<0.05, **p<0.01. (C) Representative images of A. thaliana plants infected with PlAMV-GFP and imaged with a CCD Camera from 10 to 17 dpi. Systemic leaves are circled with a white dotted line. Multicolored scale is used to enhance contrast and ranges from blue (low intensity) to red (high intensity). Scale bar = 4 cm. (D) Confocal images of the surface view of A. thaliana epidermal cells of the Col-0/ProUbi10:mRFP1.2-REM1.2, infiltrated either with free GFP (‘mock’) or PlAMV-GFP. Scale bar = 5 µm (E) Representative trajectories of Col-0/ProUbi10:mEOS3.2-REM1.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. Scale bar = 2 µm. (F) Distribution of the diffusion coefficient (D), represented as log(D) for Col-0/ProUbi10:mEOS3.2-REM1.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. Data were acquired from at least 28638 trajectories obtained from at least 16 cells over the course of three independent experiments. (G) Box plot of the mean peak value extracted from the Gaussian fit of log(D) distribution. Significant difference was revealed using a Mann-Whitney test. *p<0.05. (H) Mean square displacement (MSD) over time of Col-0/ProUbi10:mEOS3.2-REM1.2 infiltrated either with free GFP (‘mock’) or PlAMV-GFP. Representative trajectories extracted from (E) illustrate each curve. Scale bar = 1 µm. (I) Voronoi tessellation illustration of Col-0/ProUbi10:mEOS3.2-REM1.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. ND are circled in red. Scale bar = 2 µm (J) Distribution of the nanodomains (ND) diameter of Col-0/ProUbi10:mEOS3.2-REM1.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. (K) Box plot representing the mean peak value of ND diameter extracted from the Gaussian fit of (J). No significant difference was revealed using a Mann-Whitney test. (L) Boxplot of the proportion of Col-0/ProUbi10:mEOS3.2-REM1.2 detections found in ND 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. No significant difference was revealed using a Mann-Whitney test.

Figure 4—figure supplement 1.. Group 1 REMORIN (REM) single, double, and triple knock-out lines.

(A) Positions of the T-DNA insertions of the single mutants used in this study for viral propagation and crossing. For rem1.2, the T-DNA is inserted in an exon, for rem1.3 in the promoter, and for rem1.4 in an intron. Boxes represent exons while black lines represent introns. (B) Western blots of Col-0, rem1.2, rem1.3, rem1.4 single mutants, rem1.2 rem1.3 double mutant and rem1.2 rem1.3 rem1.4 triple mutant, with antibodies raised against REM1.2/1.3 (upper panel) or specifically against REM1.3 (middle panel). Protein loading was monitored by stain-free staining.

Figure 4—figure supplement 2.. Group 1 REMORINs (REMs) single and multiple knock-out lines do not display any obvious developmental phenotype.

(A) 7-d-old seedlings of Col-0, rem1.2, rem1.3 and rem1.4 single mutants. Neither of the single mutants showed developmental defects. All seedlings were grown on the same plate. Scale bar = 2 cm. (B) 7-d-old seedlings of Col-0, rem1.2 rem1.3 double and rem1.2 rem1.3 rem1.4 triple mutants. rem1.2 rem1.3 nor rem1.2 rem1.3 rem1.4 showed obvious developmental defects. All seedlings were grown on the same plate. Scale bar = 2 cm. (C) 3-wk-old plants of Col-0, rem1.2, rem1.3 and rem1.4 single mutants. Plants were grown together in the same conditions. Scale bar = 5 cm. (D) 3-wk-old plants of Col-0, rem1.2 rem1.3 double and rem1.2 rem1.3 rem1.4 triple mutants. Plants were grown together in the same conditions. Scale bar = 5 cm.

Figure 4—figure supplement 3.. Group 1 REMORINs (REMs) transcript and protein levels are not modified during a plantago asiatica mosaic virus (PlAMV) infection.

(A) Transcript levels of REM1.2, REM1.3, and REM1.4 in leaves infiltrated with either free GFP (dark gray) or PlAMV-GFP (light gray) and harvested 7 d after infection. Three independent experiments were performed, with three independent plants for each condition per experiment. Transcript levels were normalized to housekeeping genes. Statistical significance was determined using a Mann-Whitney test. No significant difference could be deciphered. (B) Left panel: REM1.2 protein levels were determined by quantification of western-blot signals from three independent assays, performed with antibodies raised against REM1.2/1.3. No significant difference could be deciphered using a Mann-Whitney test. Right panel: representative western blot of REM1.2/1.3. Protein loading was monitored by stain-free staining.

Figure 4—figure supplement 4.. REM1.2 does not accumulate at plasmodesmata during plantago asiatica mosaic virus (PlAMV) infection.

Confocal images of the secant view of A. thaliana epidermal cells of the rem1.2/mRFP1.2-REM1.2 stained with aniline blue 5 d after infiltration with free RFP (‘mock’) or PlAMV-mCherry Scale bar = 5 µm.

Figure 5.. Group 1 REMORINs (REMs) and calcium-dependent protein kinase 3 (CPK3) are in the same functional pathway and regulate each other plasma membrane (PM) diffusion upon plantago asiatica mosaic virus (PlAMV) infection.

(A) Box plots of the mean area of PlAMV-GFP infection foci in cpk3-2, rem1.2 rem1.3 rem1.4 triple mutant and rem1.2 rem1.3 rem1.4/cpk3 #1 and #2 quadruple mutants. Three independent biological repeats were performed, with at least 23 foci per experiment and per genotype. Significant differences were revealed using a one-way ANOVA followed by a Tukey’s multiple comparison test. Letters are used to discriminate between statistically different conditions (p<0.0001). (B) Representative images of PlAMV-GFP infection foci at 5 dpi in the different mutant backgrounds. Scale bar = 500 µm.(C) Representative trajectories of cpk3-2/ProUbi10:mEOS3.2-REM1.2 5 d after infiltration with either free GFP (‘mock’) or PlAMV-GFP. Scale bar = 2 µm. (D) Distribution of the diffusion coefficient (D), represented as log(D) for cpk3-2/ProUbi10:mEOS3.2-REM1.2 5 d after infiltration with either free GFP (‘mock’) or PlAMV-GFP. Data were acquired from at least 20462 trajectories obtained from at least 11 cells over the course of three independent experiments. (E) Box plot representing the mean peak value extracted from the Gaussian fit of the distribution of the diffusion coefficient (D) represented in (D). No significant difference was revealed using a Mann-Whitney test. (F) Mean square displacement (MSD) over time of cpk3-2/ProUbi10:mEOS3.2-REM1.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. (G) Representative trajectories of rem1.2 rem1.3 rem1.4/ProUbi10:CPK3-mEOS3.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. Scale bar = 2 µm. (H) Distribution of the diffusion coefficient (D), represented as log(D) for rem1.2 rem1.3 rem1.4/ProUbi10:CPK3-mEOS3.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. Data were acquired from at least 11724 trajectories obtained from at least 10 cells over the course of three independent experiments. (I) Box plot representing the mean peak value extracted from the Gaussian fit of the distribution of the diffusion coefficient (D) represented in (H). No significant difference was revealed using a Mann-Whitney test. (J) Mean square displacement (MSD) over time of rem1.2 rem1.3 rem1.4/ProUbi10:CPK3-mEOS3.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP.

Figure 5—figure supplement 1.. CRISPR-mediated knock-out of cpk3 in rem1.2 rem1.3 rem1.4 background.

(A) Sequences of wild-type calcium-dependent protein kinase 3 (CPK3) along with the CRISPR-generated mutants in rem1.2 rem1.3 rem1.4 background. In rem1.2 rem1.3 rem1.4 cpk3 #1, an insertion of one adenosine at position 209 bp induced a frameshift leading to an early stop codon at position 234–236 bp (235–237 bp after insertion). In rem1.2 rem1.3 rem1.4/cpk3 #2, a deletion of 121 nucleotides at position 88–208 bp induced a frameshift leading to an early stop codon at position 284–286 bp (163–165 bp after deletion). As a result, the CRISPR mutations led to a truncated CPK3 protein of 78 aa and 54 aa, in rem1.2 rem1.3 rem1.4 cpk3 #1 and rem1.2 rem1.3 rem1.4 cpk3 #2, respectively. (B) Schematic representation of the amino-acid sequence of wild-type and truncated CPK3. The mutations in the CRISPR-generated lines resulted in proteins deprived of the kinase domain, the junction domain, and the calcium-binding domain. (C) Western blots of Col-0 WT and the two CRISPR lines rem1.2 rem1.3 rem1.4 cpk3 #1 and rem1.2 rem1.3 rem1.4 cpk3 #2 using anti-REM1.2/1.3 and anti-CPK3 antibodies. Protein loading was monitored by stain-free staining.

Figure 5—figure supplement 2.. rem1.2 rem1.3 rem1.4 cpk3 quadruple knock-out lines do not display any obvious developmental phenotype.

(A) In vitro growth of 7-d-old seedlings of Col-0, rem1.2 rem1.3 rem1.4 cpk3 #1 and rem1.2 rem1.3 rem1.4 cpk3 #2. All seedlings were grown on the same plate. Scale bar = 2 cm. (B) Growth on soil of 3-wk-old plants of Col-0, rem1.2 rem1.3 rem1.4 cpk3 #1 and rem1.2 rem1.3 rem1.4 cpk3 #2. Plants were grown together under the same conditions. Scale bar = 5 cm.

Figure 5—figure supplement 3.. calcium-dependent protein kinase 3 (CPK3) and group1 REMORIN (REM) role in each other plasma membrane (PM) nano-organization during plantago asiatica mosaic virus (PlAMV) infection.

(A) Voronoi tessellation illustration of cpk3-2/ProUbi10:mEOS3.2-REM1.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. Nanodomains (ND) are circled in red. Scale bar = 2 µm. (B) Distribution of the ND diameter of cpk3-2/ProUbi10:mEOS3.2-REM1.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. (C) Box plot representing the mean peak value extracted from the Gaussian fit of the panel B. No significant difference was revealed using a Mann-Whitney test. (D) Box plot of the proportion of cpk3-2/ProUbi10:mEOS3.2-REM1.2 detections found in ND 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. No significant difference was revealed using a Mann-Whitney test. (E) Voronoi tessellation illustration of rem1.2 rem1.3 rem1.4/ProUbi10:CPK3-mEOS3.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. ND are circled in red. Scale bar = 2 µm. (F) Distribution of the ND diameter of rem1.2 rem1.3 rem1.4/ProUbi10:CPK3-mEOS3.2 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. (G) Box plot representing the mean peak value extracted from the Gaussian fit of the panel F. No significant difference was revealed using a Mann-Whitney test. (H) Boxplot of the proportion of rem1.2 rem1.3 rem1.4/ProUbi10:CPK3-mEOS3.2 detections found in ND 5 d after infiltration with free GFP (‘mock’) or PlAMV-GFP. Significant difference was revealed using a Mann-Whitney test, **p<0.001.

Figure 5—figure supplement 4.. REM1.2 diffusion is increased upon co-expression with CPK3^CA but not with CPK3^CA-G2A.

(A) Representative trajectories of ProUbi10:mEOS3.2-REM1.2 2 d after infiltration or co-infiltration with ProUbi10:CPK3^CA-mRFP1.2 or ProUbi10:CPK3^CA-G2A-mRFP1.2 in N. benthamiana leaves. Scale bar = 2 µm (B) Box plot of the mean peak value extracted from the Gaussian fit of log(D) distribution. Significant difference was revealed using a Mann-Whitney test. Different letters indicate statistical differences (p<0.01).

Figure 5—figure supplement 5.. Plantago asiatica mosaic virus (PlAMV) infection does not modify calcium-dependent protein kinase 3 (CPK3) and group1 REMORIN (REM) colocalization coefficient.

(A) Box plot of Pearson’s correlation coefficient between ProREM1.2:YFP-REM1.2 and ProUbi10:mTagBFP2-CPK3 5 d after infiltration of Col-0/ProREM1.2:YFP-REM1.2/ProUbi10:mTagBFP2-CPK3 with free RFP (‘mock’) or PlAMV-RFP. No significant difference could be deciphered using a Student t-test. The colocalization analysis was repeated upon 90° rotation of the mTagBFP2 channel (‘randomized’). (B) Representative confocal images of the surface view of A. thaliana epidermal cells of Col-0 /ProREM1.2:YFP-REM1.2/ProUbi10:mTagBFP2-CPK3. Scale bar = 5 µm.