Exocyst subunit Exo70B2 is linked to immune signaling and autophagy

Abstract

During the immune response, activation of the secretory pathway is key to mounting an effective response, while gauging its output is important to maintain cellular homeostasis. The Exo70 subunit of the exocyst functions as a spatiotemporal regulator by mediating numerous interactions with proteins and lipids. However, a molecular understanding of the exocyst regulation remains challenging. We show that, in Arabidopsis thaliana, Exo70B2 behaves as a bona fide exocyst subunit. Conversely, treatment with the salicylic acid (SA) defence hormone analog benzothiadiazole (BTH), or the immunogenic peptide flg22, induced Exo70B2 transport into the vacuole. We reveal that Exo70B2 interacts with AUTOPHAGY-RELATED PROTEIN 8 (ATG8) via two ATG8-interacting motives (AIMs) and its transport into the vacuole is dependent on autophagy. In line with its role in immunity, we discovered that Exo70B2 interacted with and was phosphorylated by the kinase MPK3. Mimicking phosphorylation had a dual impact on Exo70B2: first, by inhibiting localization at sites of active secretion, and second, it increased the interaction with ATG8. Phosphonull variants displayed higher effector-triggered immunity (ETI) and were hypersensitive to BTH, which induce secretion and autophagy. Our results suggest a molecular mechanism by which phosphorylation diverts Exo70B2 from the secretory into the autophagy pathway for its degradation, to dampen secretory activity.

Figure 1

Exo70B2 localizes to the PM and does not transit through the TGN. (A) Confocal laser-scanning microscopy images of 4-day-old homozygous transgenic seedlings expressing GFP-Exo70B2 under the control of the constitutive promoter UBQ10 in the exo70b2-3 background. Shown are epidermal cells of cotyledons (top left) and roots (right). Scale bars represent 20 µm (cotyledons) and 5 µm (roots). (B) UBQ10_pro:GFP-Exo70B2/exo70b2-3 and UBQ10_pro:GFP transgenic seedlings were incubated in 0.8 M mannitol for 40 min before being analyzed by CLSM. In the enlarged cropped region, indicated by the square boxes, GFP-EXO70B2 is visible on the Hechtian strands indicated by arrowheads. Scale bar 20 µm. (C) Super-resolution microscopy images of UBQ10_pro:GFP-Exo70B2/exo70b2-3 transgenic seedling root hairs. Three optical sections are shown from root hair shaft to tip. Scale bar 10 µm, V denotes vacuole. (D) Roots of 4-day-old transgenic seedlings expressing GFP-Exo70B2, GFP-clathrin heavy-chain 1 (CHC1) and free GFP were stained with 5 µM FM4-64 for 5 min and subsequently incubated in 50 µM BFA for 45 min at room temperature. Scale bar 5 µm.

Figure 2

Exo70B2 and exocyst subunits are transported into the vacuole. (A) GFP-EXO70B2 is localized to autophagic body-like compartments after BTH and ConcA treatment. UBQ10_pro:GFP-Exo70B2/exo70b2-3 seedlings were treated overnight with either 0.1% DMSO, 1 µM ConcA, 100 µM BTH, or 1 µM ConcA + 100 µM BTH. Confocal images of the same tissue were taken with identical imaging settings. Scale bars for cotyledon and root tissues are 20 and 10 µm, respectively. (B) Sec6-GFP and Exo70A1-GFP are localized to autophagic-like bodies. Sec6_pro:Sec6-GFP and Exo70A1_pro:Exo70A1-GFP seedlings were treated overnight with 0.1% DMSO, 1 µM ConcA, or 1 µM ConcA + 100 µM BTH and analyzed by CLSM. Scale bar 10 µm. (C) UBQ10_pro:GFP-Exo70B2/exo70b2-3 transgenic seedlings were treated with 0.1% DMSO (control), 100 µM BTH overnight, 1 µM ConcA overnight, and the indicated combinations. Samples from the same experiment were separated into crude fraction with nuclei, ER and chloroplasts, soluble proteins, and microsomal fraction. Total protein was detergent solubilized. Samples were resolved by PAGE and analyzed by IB using anti-GFP antibody. Short exposure (SE), long exposure (LE). Equal loading is shown by coomassie brilliant blue (CBB) staining. Similar results were obtained in three biological replicates.

Figure 3

Exo70B2 is transported into the vacuole by autophagy. (A) GFP-Exo70B2 colocalizes with autophagosome markers RFP-ATG8a and RFP-NBR1. Double transgenic lines carrying UBQ10_pro:GFP-Exo70B2 and UBQ10_pro:RFP-ATG8a or UBQ10_pro:RFP-NBR1 were treated overnight with 1 µM ConcA + 100 µM BTH before CLSM analysis. Scale bar 10 µm. (B) Co-localization analysis for GFP-Exo70B2 with RFP-ATG8a or RFP-NBR1. Data points represent technical replicates of two independent experiments. Boxplots show median and inter quantile range (IQR), outliers (>1.5 times IQR) are shown as circles. (C) Lines expressing UBQ10_pro:GFP-Exo70B2 in Col-0 WT and atg2 mutant background. Transgenic seedlings were treated overnight with 1 µM ConcA + 100 µM BTH before CLSM analysis. Scale bar 10 µm. The insets show the enlarged cropped region (indicated by the squared box). (D) Seedlings expressing UBQ10_pro:GFP-Exo70B2 in Col-0 WT and atg2 mutant background were treated as in (C) and total proteins resolved by SDS-PAGE.

Figure 4

EXO70B2 is phosphorylated at residues S554 and S567. (A) Interaction between Exo70B2 and MPK3 detected by BiFC in Arabidopsis protoplasts. nYFP-MPK3, nYFP-MPK4, nYFP-MPK6, nYFP-MPK8, or nYFP-MPK11 were coexpressed with cYFP-Exo70B2 as indicated. Free RFP was coexpressed to label the cytoplasm and nucleus. Scale bar 50 µm. The experiment was repeated with similar results. (B) Transgenic seedlings carrying UBQ10_pro:GFP-Exo70B2 were treated with 1 µM flg22 for 20 min and subjected to IP with anti-GFP beads. Endogenous CoIPed MPK3 was detected with MPK3-antibodies. Coomassie brilliant blue (CCB), IB (IB). (C) MBP-Exo70B2 pull-down assay using bacterially expressed and purified GST-MPK3 and GST-MPK6 on glutathione agarose beads as baits. (D) GST-Exo70B2 was incubated alone or with activated GST-MPK3, GST-MPK4, and untagged MPK6 (white arrowheads). MPK11 could not be activated by MKK5. Phosphorylation was visualized with ProQ Diamond stain. White arrows indicate autophosphorylated kinases. (E) Diagram depicting Exo70B2, its predicted domains A/B/C, and the localization of phosphorylated sites in Exo70B2 identified by LC–MS/MS from in vitro phosphorylation assays with GST-MPK3 and in vivo from GFP-Exo70B2 immunopurified from transgenic lines.

Figure 5

Ser554 and Ser567 regulate Exo70B2 localization in root hairs. (A) Diagram depicting the identified phosphorylation sites Ser554 and Ser567 (magenta) in the Exo70B2 C-domain, and corresponding amino acids contributing to ES2 binding in Exo70A1 (green) and membrane binding of the baker’s yeast Exo70 (blue). (B) Super-resolution microscopy images of UBQ10_pro:GFP-Exo70B2/exo70b2-3 transgenic seedling root hairs expressing the WT, S554/567A (AA), or S554/567D (DD) proteins. Three optical sections taken from the cortical layer. Inset shows a magnification. Scale bar 5 µm. Numbers show the percentage of root hairs displaying aggregates. The experiment was repeated with similar results. (C) Quantification of punctae in (B) from n ≥ 10 root hairs. (D) PM-enriched protein fractions from GFP-Exo70B2 WT, GFP-Exo70B2 AA, and GFP-Exo70B2 DD in exo70b2-3. Protein samples were obtained from 8-day-old seedlings and analyzed by IB with anti-GFP and anti-PIP2;2. Values represent the band intensity of the PM-enriched fraction relative to GFP-Exo70B2 (WT) calculated from the ratios between the PM to total protein signals. Equal loading shown by coomassie brilliant blue (CBB).

Figure 6

Exo70B2 phosphorylation regulates interaction with ATG8 via C-terminal AIMs. (A) Diagram depicting the localization and sequences of ATG8-interacting motifs (AIMs) highlighted in violet. Phosphorylation sites are highlighted in magenta. (B) BiFC of coexpressed nYFP-ATG8f and either cYFP-Exo70B2 (WT) or mutant variants cYFP-Exo70B2-ΔAIM1 (ΔAIM1), cYFP-Exo70B2-ΔAIM2 (ΔAIM2), cYFP-Exo70B2-ΔAIM1/AIM2 (ΔAIM1/2), or empty pSPYCE. Constructs were transiently coexpressed in Arabidopsis mesophyll protoplasts. mCherry was transformed as a marker for transformation. Boxplots show median and inter quantile range (IQR), outliers (>1.5 times IQR) are shown as circles. Letters indicate statistically significant differences (one-way ANOVA and Tukey’s post hoc test, P < 0.05). Percentages of fluorescence complementation obtained from 30 independent images, each with 25–40 transformed protoplasts. Total protoplasts scored for WT, AIM1, AIM2, AIM1/2 and empty vector were 723, 754, 708, 683 and 542, respectively. The experiment was repeated with similar results.

(C) Pull-down (PD) assay with bacterially expressed and purified recombinant proteins. His-ATG8f was co-purified using MBP-Exo70B2 on amylose agarose beads. Asterisk indicates MBP-Exo70B2. (D) BiFC of coexpressed nYFP-ATG8f and either cYFP-Exo70B2 S554/567A (AA), S554/567D (DD), or empty vector were transiently coexpressed in Arabidopsis mesophyll protoplasts. mCherry was used as a transformation marker. Evaluation of percentage of fluorescence complementation and statistical analysis as in (B). Total protoplasts scored for Exo70B2 WT, AA, DD, and empty vector were 1283, 695, 1325, and 1302, respectively. The experiment was repeated with similar results. (E) Pull-down assay using recombinant His-ATG8f and MBP-Exo70B2 S554/567A (AA) or S554/567D (DD) on amylose agarose beads. Asterisk indicates MBP-Exo70B2. (F) Transgenic seedlings expressing GFP-Exo70B2 WT, S554/567A (AA), and S554/567D (DD) were treated with DMSO, or with 1 µM and 2 µM ConcA for 8 h. Exo70B2 amounts relative to the WT control were obtained by densitometric analysis, and values normalized to CBB; the experiment was repeated three times with similar results.

Figure 7

Plants expressing Exo70B2 phosphonull variant are more sensitive to BTH and resistant to avirulent bacteria. (A) Infection assays with the virulent bacterial pathogen P. syringae pv. tomato DC3000 (Pst) empty vector. Six-week-old plants were spray inoculated with a bacterial suspension of 5 × 10⁸ c.f.u./mL and analyzed 0 and 3 days after inoculation. Data shown as mean ± SD (n = 5). Letters indicate statistically significant differences between c.f.u. in different lines 3 days after inoculation (dai), one-way ANOVA, and Tukey’s post hoc test, P < 0.05. Similar results were obtained in three independent experiments. (B) Infection assays with the avirulent bacterial pathogen P. syringae pv. tomato DC3000 AvrRPS4. Six-week-old plants were syringe-infiltrated with a bacterial suspension OD₆₀₀ 0.001 and analyzed 0 and 3 days after inoculation. Data shown as mean ± SD (n = 5). Letters indicate statistically significant differences (one-way ANOVA and Tukey’s post hoc test, P < 0.05). (C) Primary root lengths of exo70b2-3 transgenic seedlings expressing GFP-Exo70B2 WT, S554/567A (AA), S554/567D (DD), and ΔC-domain (ΔC) measured 7 days after transplanting onto media ±100 µM BTH. Boxplots show median and IQR, outliers (>1.5 times IQR) are shown as circles. Values of a representative experiment are shown. Asterisks indicate statistically significant differences between control and BTH treatment (one-way ANOVA and Tukey’s post hoc test, P < 0.05). The experiment was repeated four times with similar results. See also Supplemental Table S2 for complete statistical analysis of results. (D) Transgenic seedlings described in (A) were treated with 100-µM BTH overnight. Total protein samples were resolved by PAGE and analyzed by IB using anti-PR1 antibodies. Equal loading is shown by CBB staining. The experiment was repeated with similar results.

Figure 8

Working model for Exo70B2 function as a secretion rheostat to maintain homeostasis. Exo70B2 (blue) mediates the secretion of cargo by participating in the tethering of secretory vesicles to the PM. MPK3, and most likely other kinases (?), are activated during the immune response, or other conditions or stresses (?), resulting in Exo70B2(p) phosphorylation. Phosphorylation has a double impact on Exo70B2: (1) It reduces Exo70B2 localization to the PM and (2) it enhances interaction with ATG8. Both effects may additively redirect Exo70B2 from the PM, and inhibit its role in secretion, to be recruited by autophagy, and finally, degraded in the vacuole. Exo70B2 may thus act as a molecular rheostat that gauges secretory output and maintains cellular homeostasis.