Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

An essential role for the VASt domain of the Arabidopsis VAD1 protein in the regulation of defense and cell death in response to pathogens

  • Mehdi Khafif ,

    Contributed equally to this work with: Mehdi Khafif, Claudine Balagué

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft

    Affiliation LIPM, Université de Toulouse, INRA, CNRS, Castanet-Tolosan, France

  • Claudine Balagué ,

    Contributed equally to this work with: Mehdi Khafif, Claudine Balagué

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – review & editing

    Affiliation LIPM, Université de Toulouse, INRA, CNRS, Castanet-Tolosan, France

  • Carine Huard-Chauveau,

    Roles Conceptualization

    Affiliation LIPM, Université de Toulouse, INRA, CNRS, Castanet-Tolosan, France

  • Dominique Roby

    Roles Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing – original draft, Writing – review & editing

    dominique.roby@toulouse.inra.fr

    Affiliation LIPM, Université de Toulouse, INRA, CNRS, Castanet-Tolosan, France

    ORCID http://orcid.org/0000-0001-8414-1799

An essential role for the VASt domain of the Arabidopsis VAD1 protein in the regulation of defense and cell death in response to pathogens

  • Mehdi Khafif, 
  • Claudine Balagué, 
  • Carine Huard-Chauveau, 
  • Dominique Roby
PLOS
x

Abstract

Several regulators of programmed cell death (PCD) have been identified in plants which encode proteins with putative lipid-binding domains. Among them, VAD1 (Vascular Associated Death) contains a novel protein domain called VASt (VAD1 analog StAR-related lipid transfer) still uncharacterized. The Arabidopsis mutant vad1-1 has been shown to exhibit a lesion mimic phenotype with light-conditional appearance of propagative hypersensitive response-like lesions along the vascular system, associated with defense gene expression and increased resistance to Pseudomonas strains. To test the potential of ectopic expression of VAD1 to influence HR cell death and to elucidate the role of the VASt domain in this function, we performed a structure-function analysis of VAD1 by transient over-expression in Nicotiana benthamiana and by complementation of the mutant vad1-1. We found that (i) overexpression of VAD1 controls negatively the HR cell death and defense expression either transiently in Nicotiana benthamania or in Arabidopsis plants in response to avirulent strains of Pseudomonas syringae, (ii) VAD1 is expressed in multiple subcellular compartments, including the nucleus, and (iii) while the GRAM domain does not modify neither the subcellular localization of VAD1 nor its immunorepressor activity, the domain VASt plays an essential role in both processes. In conclusion, VAD1 acts as a negative regulator of cell death associated with the plant immune response and the VASt domain of this unknown protein plays an essential role in this function, opening the way for the functional analysis of VASt-containing proteins and the characterization of novel mechanisms regulating PCD.

Introduction

Programmed cell death (PCD) occurs in plants in a variety of cell types during development and to face environmental stresses such as pathogens or other stress signals [1, 2]. The Hypersensitive Response (HR) is one of the best characterized PCD, which is localized at the site of attempted pathogen invasion. In many plant-pathogen interactions, HR is triggered upon pathogen recognition. The plant immune system contains several layers [3]. The basal layer of defense, referred to as PAMP Triggered Immunity (PTI), is activated by the presence of pattern recognition receptors (PRRs) which recognize Microbe–or Pathogen-Associated Molecular Patterns (MAMPs or PAMPs). PRRs are plasma membrane receptors, that activate rapid and effective immune responses upon PAMP recognition [4]. However, pathogenic microorganisms can suppress PTI by producing effector proteins, thus leading to plant susceptibility to pathogen multiplication. However, a second and more robust layer of defense response, termed Effector Triggered Immunity (ETI) occurs upon recognition of pathogen effectors by a second type of plant immune receptors (R proteins). In many cases, ETI leads to the hypersensitive cell death, leading to the confinement of the pathogens at the attempted infection site.

PCD is highly regulated at the genetic level as the control of this cell death programme is essential for the development of multicellular organisms. In plants, its fundamental role in response to pathogens has been revealed by the fact that many pathogens have evolved different strategies to suppress the HR. The molecular mechanisms leading to cell death after pathogen attack during ETI are not fully elucidated. However, some regulators or executioners of HR have been identified, such as BI-1 (BAX INHIBITOR 1, [5, 6], caspases and metacaspases [7] or MYB30 [8]. Characterization of mutants displaying spontaneous HR-like cell death on leaves, the so-called lesion mimic mutants (LMMs) has also been frequently reported, leading to the identification of genes encoding inhibitors of plant cell death, frequently associated with immunity [9, 10]. Demonstration of their implication in immunity control has been made in several cases, either by effector targeting [11, 12], by genetic interaction with immunity pathway components [13, 14, 15] or by suppressor identification [10]. In summary, numerous autoimmune or LMMs and their suppressors represent genes (potentially) involved in the control of the HR, and should be characterized more in depth [16].

We previously reported that a LMM loss-of-function mutant of the gene VASCULAR ASSOCIATED DEATH1 (VAD1) exhibits constitutive immune responses and HR-like spontaneous lesions [14]. The vad1 phenotypes are suppressed by mutations affecting immunity pathways such as eds1 and npr1, and by mutations affecting hormonal pathways (salicylic acid and ethylene pathways), known to control the immune response [17, 14]. In contrast to most LMM mutants, vad1 is affected in the control of cell death propagation instead of its initiation [14]. VAD1 encodes a protein of unknown function containing a GRAM domain predicted to bind lipids [18, 19] and a VASt (VAD1 Analog of StAR-related lipid transfer) domain. This domain which belongs to the Bet v1-like superfamily [5], has no molecular function assigned yet. Interestingly, among many uncharacterized proteins containing both a VASt and a GRAM domain, the yeast YSP2 (Yeast Suicide protein 2) gene encodes a cell death regulator [20]. These findings open new perspectives for the functional analysis of VAD1. It should be noted that a significant proportion of plant LMM mutants are affected in genes associated with lipid synthesis or homeostasis [10]. It is the case of acd5 (Accelerated Cell Death 5) mutant which is deficient in a ceramide kinase [21], erh1 (Enhancing RPW8- mediated HR-like cell death) which lacks the inositol-phosphorylceramide synthase 2 (IPCS2)[22], acd11 (Accelerated Cell Death 11) which is affected in the ceramide-1-phosphate transfer protein (C1P)[23], or loh1 (LAG One Homolog) affected in a ceramide synthase [24]. This reveals that the lipid metabolism plays a major role in promoting PCD. In this context, it is interesting to note that the vad1 phenotypes are dependent on the PATATIN-LIKE PROTEIN 2 (PLP2), a pathogen induced patatin like lipid acyl hydrolase [25].

In this study, a functional analysis of VAD1 was performed. Our results show that (i) overexpression of VAD1 controls negatively the HR cell death and defense expression either transiently in Nicotiana benthamania or in Arabidopsis plants in response to avirulent strains of Pseudomonas syringae, (ii) while the GRAM domain does not modify neither the subcellular localization of VAD1 nor its immunorepressor activity, the domain VASt plays an essential role in both processes.

Materials and methods

Plant material

As previously described, mutant and wild-type seeds were sterilized and sown on Murashige and Skoog plates [26], supplemented with the appropriate antibiotic under a light period of 16 h (71 μmol m−2 s−1) at 21°C. Arabidopsis mutant (vad1-1) and other transgenic lines used in this study (35S::RFP::VAD1, 35S::RFP::ΔGRAM, 35S::RFP::ΔVASt) are in the WS-4 or vad1-1 background and were grown respectively in presence of kanamycin 50 μg mL−1 for the vad1-1 mutant or hygromycin 25 μg mL−1 for the transgenic lines.

7 to 10 days after sowing, seedlings were transplanted to Jiffy pots and grown in a growth chamber under a light period of 9 h (192 μmol m−2 s−1) at 21°C and 40% to 70% relative humidity and4- to 5-week old plants were used for most experiments. Under these growth conditions, lesions appear on the vad1-1 mutant 21 days after transplanting.

Bacterial growth and plant infection

For transient expression of proteins in N. benthamiana, bacterial cultures of Agrobacterium tumefaciens (strain C58C1) were grown in YEB medium supplemented with the appropriate antibiotics: rifampicin 50 μg mL−1, tetracyclin 50 μg mL−1 and spectinomycin 50 μg mL−1 for strains carrying the different VAD1 constructs. Bacterial cultures of Agrobacterium tumefaciens (strain GV3101) (MtSYMREM1) were grown in YEB medium supplemented with rifampicin 50 μg mL−1, carbenicillin 50 μg mL−1, and gentamycin 50 μg mL−1. After centrifugation, cells were resuspended in induction buffer (10 mM MgCl2, 10 mM Mes, pH 5.6, and 200 μM acetosyringone) to an OD600 of 0.5. Leaves of 4-week-old N. benthamiana grown at 22°C were inoculated and then placed in a growth chamber. Five hours after, the same leaves were inoculated with an A. tumefaciens strain carrying the avirulent gene (AvrRpt2 or AvrRpt0). One or two days after A. tumefaciens infiltration, leaf discs were harvested and processed, or frozen immediately in liquid nitrogen and stored at –80°C.

Transient expression of proteins in Arabidopsis seedlings was performed as previously described [12]. Briefly, A. tumefaciens overnight cultures were collected and resuspended at an OD600 of 1 in 2 mL of Arabidopsis liquid medium containing 200 μM acetosyringone. Seedlings were vacuum infiltrated (10 mm Hg) twice for 1 min with the A. tumefaciens solution. Excess infiltration medium was subsequently removed and the plates transferred to a culture room (22°C, long day conditions, 150 μmol m−2 s−1 light intensity) for 3 days.

The avirulent Pseudomonas syringae pv. tomato strains were grown at 28°C on King's B medium supplemented with antibiotics (rifampicin 50 μg mL−1, kanamycin 50 μg mL−1 and tetracyclin 10 μg mL−1) (DC3000/AvrRpm1). Four- or 5-week-old plants were used for bacterial inoculation. After 12 h at high humidity, plants were grown under light-promoting lesions under the following conditions: 9h light/15h dark and 90% relative humidity [14]. Plants were infiltrated with a bacterial suspension of 5 × 106 colony forming units (CFU) mL−1 (DC3000/avrRpm1) for visualization of HR phenotypes, 1 × 107 CFU mL−1 for gene expression analysis and 2 × 107 CFU mL−1 for runaway cell death experiments. Determination of in planta bacterial growth was performed as previously described [14].

RNA extraction and quantitative real-time PCR (qRT-PCR)

Total RNA extraction was performed from leaves with the Nucleospin RNA kit (Macherey-Nagel). cDNA was synthesized with Superscript III reverse transcriptase (Invitrogen Carlsbad, CA) using 1 μg of total RNA. Quantitative RT-PCR was performed using gene-specific primers (S1 Table), on a Light Cycler 480 II apparatus (Roche Diagnostics, Meylan, France) using Roche reagents. SYBR green I. Five pmol of each primer and 1 μL of a five-fold dilution of reverse transcriptase reaction product were used in a final reaction volume of 10 μL with PCR cycling conditions as follows: 9 min at 95°C, followed by 45 cycles of 5 s at 95°C, 10 s at 65°C and 20 s at 72°C. At least three independent biological assays were used for Quantitative PCR reactions. All reactions were checked for their dissociation curves. ΔCT between each gene and the internal controls (At5g16260, for Arabidopsis thaliana experiments and At5g15710 for N. benthamiana experiments [27] was then calculated for each sample, and expression levels for each transcript were calculated as 2–ΔCT and expressed in arbitrary units.

Quantification of cell death using electrolyte leakage

Three leaf discs (6 mm diameter) were harvested 24 hours post-inoculation (hpi) with the indicated strains, washed, and incubated at room temperature in 10 mL of distilled water. Leaf discs were then transferred into 5mL of fresh distilled water and incubated 1 hour at room temperature before measuring conductivity. Two to five independent experiments were performed with six plants (three leaves per plant). For the runaway cell death monitoring in Arabidopsis lines, 3 leaf discs (0.4 mm diameter) were used and the leaf discs of 2 plants were pooled into one vial containing 3.5mL of fresh distilled water. Two independent experiments were performed with 9–15 plants.

Protein extraction and western blotting

Total protein was extracted from leaves harvested 24h after treatment (transient expression experiments) or 2–3 weeks after sawing (stable expression experiments) using the Laemmli buffer [28] and separated on SDS-PAGE. For CFP tagged proteins, blots were incubated with rabbit anti-GFP (AMS Biotechnology, [1:5,000]) antibody and a goat anti-rabbit IgG HRP-coupled secondary antibody (Millipore [1:10,000]). For RFP-tagged proteins, blots were incubated with rabbit anti-RFP HRP-coupled antibody (1:5,000) goat anti-rabbit secondary antibody (Millipore [1:10,000]). Proteins were visualized using the Immobilon kit (Millipore).

Gene cloning and Arabidopsis transformation

Unless otherwise indicated, plasmids used in this study were constructed using the Gateway technology (GW; Invitrogen) following the instructions of the manufacturer. To generate the VAD promoter::VAD1::GFP construct, the VAD1 genomic sequence including 604 nucleotides upstream the initiation site was amplified using pfx DNA polymerase (Invitrogen) and the T7I23 BAC clone as DNA template. The PCR product was then digested with BamHI and NotI restriction enzymes to be cloned into the pENTR4 vector and then by a BP reaction into pMDC110 (Invitrogen). To generate the 35S::RFP::VAD1 constructs, the open reading frame (ORF) of VAD1 was cloned into the binary vector, downstream of the CaMV 35S and the RFP tag. PCR products flanked by the attB sites were cloned into the pDONR207 vectors (Invitrogen) via a BP reaction to create the corresponding entry clones with attL sites. Inserts cloned into the entry clones were subsequently recombined into the destination vector pH7WGR2, obtained from Plant System Biology (Ghent University; http://gateway.psb.ugent.be/) via an LR reaction to create the expression constructs. All the cloned PCR products were confirmed by sequencing. Arabidopsis vad1-1 lines were transformed with the different constructs by the floral dipping method [29]. T1 seedlings were screened on MS plate supplemented with hygromycin B 25 μg mL−1. Homozygous transformant lines were confirmed by segregation experiments using T3 seedlings. All the primers used for cloning are listed in S1 Table.

Fluorescence microscopy

Fluorescence images were acquired using a Leica SP8 confocal microscope equipped with a water immersion objective lens (× 25, numerical aperture 1.20; PL APO). GFP fluorescence was excited with the 488 nm ray line of the argon laser and recorded in one of the confocal channels in the 505 to 530 nm emission range. RFP fluorescence was excited with the 561 nm ray line of the He-Ne laser and detected in the range between 595 and 620 nm. Images were acquired in the sequential mode using Leica LCS software and analyzed using the ImageJ software. A set of fluorescent protein fusion constructs that mark different sub-cellular compartments has been used, including AtMYB30 (At3g28910, nucleus) [11], AtERD2 (At1g29330), Golgi-cis/ER network or AtWAK2 fused to HDEL retention peptide, ER) [30, 31], γ-TIP (tonoplast intrinsic protein) (At2g36830, tonoplast membrane) [16], AtRANBP1A (At1g07140, cytoplasm/nuclear export) [32] and PMA4, (X66737, plasma membrane) [33].

Statistics

Each experiment was independently conducted 2–4 times with similar results. The results are shown as means ± standard deviation. For Q-RT-PCR, samples consisted of pools of leaf discs collected from 5 plants for bacterial internal growth quantification or 9–15 plants for cell death quantification. Data were analyzed by Student's t-test or Kruskal and Wallis one-way analysis of variance followed by post-hoc tests for pairwise multiple comparisons (Nemenyi's test with Tukey-Dist approximation for independent samples) or nonparametric multiple comparison procedures for unbalanced one-way factorial designs [34]. All these analyses were performed using the R statistical software system (http://www.cran.r-project.org, Nparcomp R package version 2.6, PMCMR R package).

Results

Transient expression of VAD1 leads to HR cell death and defense suppression

We previously demonstrated that a loss of function mutation of VAD1 was associated with appearance of propagative HR-like lesions along the vascular system and increased resistance to virulent and avirulent pathogens [14]. We also found that the vad1 phenotypes were dependent on ETI signaling components such as NDR1 and EDS1 [14]. To determine whether VAD1 acts as a negative regulator of immunity and has a direct role in suppressing cell death, we analyzed its overexpression effect on PCD. First we tested the effect of overexpression of VAD1 by Agrobacterium-mediated transient expression of a 35S::RFP::VAD1 fusion in leaf epidermal cells of Nicotiana benthamiana (N. benthamiana) leaves, on the HR induced by the expression of the Pseudomonas effector AvrRpt2 [35]. As shown in Fig 1A, expression of VAD1 has a negative effect on the development of the AvrRpt2-triggered HR, whereas the expression of another unrelated protein (MtSYMREM1, [36]) does not affect HR progress. These visual observations were confirmed by ion leakage measurements in leaf disk assays 24h after plant agro-infiltration with the different strains (Fig 1B). Interestingly, suppression of cell death correlated with decreased defense gene expression. Indeed, PR1a expression was drastically reduced when VAD1 was co-expressed with AvrRpt2 (Fig 1C). We additionally tested the effect of VAD1 overexpression on HR induced by another avirulence gene AvrPto [35] (S1 Fig), with similar results. Therefore, overexpression of VAD1 in N. benthamiana greatly decreased HR induced by two different effector proteins.

thumbnail
Fig 1. VAD1 overexpression leads to suppression of HR cell death and defense in response to bacterial avirulence effectors or avirulent pathogens.

(A) VAD1 delays and reduces HR induced by AvrRpt2 after agroinfiltration of N. benthamiana leaves. AvrRpt2 was agroinfiltrated on leaves (marked by white circles) to induce HR and observations were made 24 hours post-infiltration. Except for the control (AvrRpt2), all areas were co-agroinfiltrated with 35S::RFP::VAD1, or 35S::SYMREM1::CFP (a control protein). (B) Quantification of cell death by measuring electrolyte leakage 24 h after agroinfiltration of N. benthamiana leaves with the indicated strains (OD 0.5 for constructs, 0.3 for AvrRpt2). Data are expressed relative to AvrRpt2 data. Statistically significant differences were determined using t-test (* indicates P < 0.05). (C) Expression analysis of the PR1a defense gene in N. benthamiana leaves after agroinfiltration with the indicated strains (OD 0.5 for constructs and OD 0.3 for AvrRpt2). Mean and SE values were calculated from the results of a representative out of 3 independent experiments with 3 individual plants (four leaves/plant). Statistically significant differences were determined using t-test (* indicates P < 0.05). (D) VAD1 delays and reduces HR induced by Pst DC3000 harboring AvrRpm1. The bacterial strain was inoculated on Arabidopsis leaves to induce HR (indicated by the tissue collapse at the inoculation site). Leaves were shown 30 hours post-inoculation with DC3000 AvrRpm1. Except for the control (AvrRpm1), all areas were co-agroinfiltrated with 35S::RFP::VAD1, or 35S::SYMREM1::CFP (a control protein). (E) Quantification of cell death by measuring electrolyte leakage 48h after inoculation of leaves with Pst DC3000 AvrRpm1 (1x107 CFU.mL-1). Data are expressed relative to AvrRpm1 data 9h after sampling. Statistically significant differences were determined using t-test (* indicates P < 0.05). (F) Expression analysis of the PR1 defense gene in A. thaliana leaves after inoculation with Pst DC3000 AvrRpm1 (1x107 CFU.mL-1). Mean and SE values were calculated from the results of a representative out of 3 independent experiments with 3 individual plants (four leaves/plant). Statistically significant differences were determined using t-test (* indicates P < 0.05).

https://doi.org/10.1371/journal.pone.0179782.g001

We subsequently analyzed the effect of overexpression of VAD1 on PCD induced by Pseudomonas syringae pv. tomato (Pst) DC3000 carrying the avirulent effector AvrRpm1 in Arabidopsis [37]. At 12 hours post-inoculation, a strong HR was observed on all leaves inoculated with Pst DC3000 AvrRpm1 alone, while HR was delayed and reduced when 35S::VAD1::GFP was simultaneously agroinfiltrated (Fig 1D). These observations were confirmed by ion leakage measurements and PR1 gene expression (Fig 1E and 1F).

Taken together, these data show that overexpression of VAD1 in Arabidopsis and N. benthamiana greatly reduced HR induced respectively, either by two different avirulence effectors, or by an avirulent strain of Pst.

VAD1 is localized in multiple subcellular compartments

To monitor the subcellular localization of the VAD1 protein, a C-terminal RFP fusion construct under the control of the 35S promoter was expressed in the leaf epidermal cells of Nicotiana benthamiana leaves by agroinfiltration. Using a Leica SP8 laser confocal microscope, VAD1 was observed in the Golgi apparatus, the cytoplasm, the nucleus, the tonoplast and the plasma membrane (Fig 2A, left panels). This multiple localization was confirmed by co-expression with different subcellular markers (Fig 2A, central panels) [11, 30, 38, 32, 33]. Similar observations were done by transient expression of the same constructs in Arabidopsis thaliana leaves [39] (S2A Fig). 35S::RFP::VAD1 was not observed in mitochondria, peroxysomes (data not shown) and chloroplasts (Fig 2A). These observations were also confirmed using a C-terminal GFP fusion construct under the control of the VAD1 promoter (S2C Fig), with the exception of the nucleus where expression was not, or rarely, observed.

thumbnail
Fig 2. VAD1 localizes in multiple subcellular compartments and a VASt domain deletion excludes VAD1 from the nucleus.

(A) Transient co-expression by agroinfiltration of N. benthamiana leaves of 35S::RFP::VAD1 construct (left panels), with subcellular markers (central panels, ERD2–GFP for the Golgi, γ-TIP–GFP for the tonoplast, GFP for the cytoplasm, MYB30-GFP for the nucleus, PMA4–GFP for the plasma membrane). Co-localization of VAD1 with the different subcellular markers is shown in the merge panel (right). Confocal images were observed two days after agroinfiltration. Scale bars = 20μM. (B) Transient co-expression by agroinfiltration of N. benthamiana leaves of 35S::RFP::ΔVASt construct (left panels), with subcellular markers (central panels, ERD2–GFP labels the Golgi, γ-TIP–GFP labels the tonoplast, GFP labels the cytoplasm, MYB30-GFP labels the nucleus, PMA4–GFP labels the plasma membrane (Plasma m.). Co-localization of VAD1 with the different subcellular markers is shown in the merge panel (right). Localization of 35S::RFP::ΔVASt is not observed in the nucleus. Confocal images were observed two days after agroinfiltration. Bars = 20 μM.

https://doi.org/10.1371/journal.pone.0179782.g002

The VASt domain of VAD1 is necessary for VAD1 nuclear localization and suppression of HR

We previously reported that VAD1 harbors several conserved domains: a GRAM putative lipid binding domain at the N-terminal region (position 116–182), a VASt (VAD1 analog to START) domain (position 257–449), a transmembrane domain and a coiled-coil domain (Fig 3A)[5]. To determine if a particular VAD1 domain is necessary for its effects on HR and defense, deleted versions of VAD1 lacking either the GRAM domain or the VASt domain were used (Fig 3A). First we tested the effect of overexpression of these constructs by Agrobacterium-mediated transient expression in leaf epidermal cells of N. benthamiana leaves, on the HR induced by the expression of the Pseudomonas effector AvrRpt2, as described before. While the VAD1-ΔGRAM deletion did not significantly alter the negative effect of VAD1 on the development of the AvrRpt2-triggered HR, the VAD1-ΔVASt deletion strongly reduced this effect (Fig 3B). These visual observations were confirmed by ion leakage measurements in leaf disk assays 24h after plant agro-infiltration with the different strains (Fig 3C). PR1a and HSR203J (a well known marker of HR cell death) [40, 41] gene expression was also correlated with the visual observations, showing a similar level when AvrRpt2 alone, or AvrRpt2 + VAD1-ΔVASt were co-expressed (Fig 3D and 3E). Similar results were observed when we tested the effect of the same constructs on HR induced by another avirulence gene AvrPto [42], S1 Fig). Interestingly, the localization of VAD1 in multiple subcellular compartments was not affected by the GRAM or VASt domain deletions, excepted in one case: the nuclear localization of VAD1 was abolished by the ΔVASt deletion (Fig 2B and S2B Fig). Finally, expression of proteins fused to RFP was confirmed by Western blot in all cases (S3 Fig), showing that subcellular detection of RFP fluorescence in different compartments is due to localization of VAD1 and not passive diffusion of truncated RFP. In addition, while the proteins cannot be detected in presence of AvrRpt2, probably as a result of protein degradation [43, 44], they can be detected in the case of AvrPto, suggesting that even if they are present at low levels and/or transiently before degradation by AvrRpt2, they induce the observed effects on HR.

thumbnail
Fig 3. Structure-function analysis of VAD1 effects on HR cell death and defense in response to bacterial avirulence effectors.

(A) Schematic representation of VAD1 and constructs. GRAM (Glucosyltransferases, Rab-like GTPase Activators, Myotubularins) domain; VASt (VAD1 analog of START) domain; TM, transmembrane helix; CC: Coiled-coil. Residue number corresponds to amino acid position of VAD1 domains. (B) Observation of HR induced by AvrRpt2 after agroinfiltration of N. benthamiana leaves alone or co-expressed with the constructs 35S::VAD1, 35S::ΔGRAM::VAD1, 35S::ΔVASt::VAD1 or 35S::SYMREM1. Observations were made 24h post-infiltration. (C) Quantification of cell death by measuring electrolyte leakage 24h after agroinfiltration of N. benthamiana leaves with the indicated strains (OD 0.3). Data are expressed relative to AvrRpt2 data at 1 h after sampling. Statistically significant differences were determined using Kruskal and Wallis one-way analysis of variance (letters indicate P < 0.05). (D and E) Expression analysis of the PR1a defense gene and HSR203J gene in N. benthamiana leaves after agroinfiltration with the indicated strains (OD 0.3) Data are presented as means ± SE (n ≥ 12), and error bars represent standard error. Letters indicate a significant difference between tested construct and control at P<0.05 by Kruskal and Wallis one-way analysis of variance test.

https://doi.org/10.1371/journal.pone.0179782.g003

Taken together, these data show that (i) VAD1 mediated effects on HR cell death and defense expression is not dependent on the presence of the GRAM domain, (ii) the VASt domain is essential for VAD1 activity as an HR cell death repressor, and for its nuclear localization.

VAD1 overexpressing plants exhibit increased susceptibility to a virulent strain of the bacterial pathogen Pseudomonas syringae

To get a deeper understanding of the potential of VAD1 at the entire plant level, we produced transgenic Arabidopsis in the vad1-1 mutant background expressing a green-fluorescing protein fusion of VAD1 under control of the cauliflower mosaic virus CaMV35S promoter and the deleted versions of VAD1 lacking either the GRAM domain or the VASt domain, already used (Fig 3A). Transgene expression was confirmed in all generations by western blots (S3C Fig). Transgenic 35S::VAD1 and 35S::ΔGRAM::VAD1 plants developed similarly to the wild type in T0 and following generations (Fig 4A). However a cell death phenotype similar to the one observed typically for vad1-1, was observed in the case of the transgenic 35S::ΔVASt::VAD1 (Fig 4A), confirming that the VASt domain is essential for VAD1 activity as an HR cell death repressor. In addition, plant development was affected, the transgenic 35S::ΔVASt::VAD1 plants showing a smaller rosette than the wild type and presenting vad1-1 leaf morphology in relation to lesion appearance [14]. These data were quantitatively confirmed by measuring the percentage of leaves with lesions in the different lines between 17 and 31 days post-transplanting (Table 1). In response to inoculation with Pst DC3000 AvrRpm1, a typical HR was observed in the vad1-1 plants complemented with 35S::VAD1 and 35S::ΔGRAM::VAD1 (Fig 4B and 4C). In contrast, in the mutant vad1-1 and the transgenic 35S::ΔVASt::VAD1, the leaves that display propagative lesions along the whole primary vein showed spreading cell death (runaway cell death, RCD), as previously shown for vad1-1 [14](Fig 4B and 4C). These visual observations were confirmed by ion leakage measurements in leaf disk assays 24h after inoculation with Pst DC3000 AvrRpm1 (Fig 4D).

thumbnail
Fig 4. Overexpression of VAD1, but not VAD1 lacking the VASt domain, leads to suppression of cell death and defense phenotypes in vad1-1 during plant development and in response to bacterial avirulent pathogens.

(A) Four-week old vad1-1 plants and transgenic vad1-1 plants containing the 35S::RFP::ΔVASt construct show typical cell death lesions (marked by white arrows), while transgenic vad1-1 plants containing the 35S::RFP::VAD1 construct or the 35S::RFP::ΔGRAM construct show a wild type phenotype. (B-C) Five-week old plants have been inoculated with suspensions (2.106 colony-forming units (CFU/mL) of Pst strain DC3000 expressing AvrRpm1. Wild type and transgenic 35S::RFP::VAD1 and 35S::RFP::ΔGRAM plants showed typical HR lesions, while vad1-1 and transgenic plants containing the construct 35S::RFP::ΔVASt construct showed run away cell death. These phenotypes were observed 3 days post-inoculation (B) or 6 days post-inoculation (C). (D) Quantification of cell death by measuring electrolyte leakage 24 (grey bars) and 48h (black bars) after infiltration of leaves with Pst strain DC3000 AvrRpm1 (2x107 colony-forming units (CFU/mL). Data are expressed relative to WS-4 1h after sampling. One transgenic line is shown per construct, out of 2–3 analyzed with similar results. Statistically significant differences were determined using Kruskal and Wallis one-way analysis of variance followed by nonparametric multiple comparison (* indicates P < 0.05).

https://doi.org/10.1371/journal.pone.0179782.g004

thumbnail
Table 1. Percentage of leaves with lesions in the different plant lines, at several times (17, 21, 24 and 28 days post-transplanting) of plant development, under normal growth conditions.

https://doi.org/10.1371/journal.pone.0179782.t001

We next examined the response of the same transgenic lines to a virulent strain of Pst (strain DC3000)(Fig 5A). A typical chlorosis was observed in the case of the wild type and the transgenic vad1-1 plants containing the 35S::RFP::VAD1 construct or the 35S::RFP::ΔGRAM construct, while very faint or no symptoms were observed in the mutant vad1-1, as previously reported [14]. The same weak phenotype was observed for the transgenic vad1-1 plants containing the 35S::RFP::ΔVASt construct. These observations were confirmed by evaluation of in planta bacterial growth (Fig 5B). No significant differences in resistance could be found between the vad1-1 mutant and the two transgenic lines expressing the 35S::RFP::ΔVASt construct, indicating that this construct is unable to complement the mutant phenotype. Interestingly, in the transgenic lines expressing the 35S::RFP::VAD1 construct or the 35S::RFP::ΔGRAM construct, a significantly increased susceptibility was observed as compared to the wild type. Therefore, overexpression of VAD1 leads to increased susceptibility to virulent strains of Pst.

thumbnail
Fig 5. Overexpression of VAD1, but not VAD1 lacking the VASt domain, leads to enhanced susceptibility in vad1-1 in response to the virulent strain Pst DC3000.

(A) Four-week old wild type and transgenic vad1-1 plants containing the 35S::RFP::VAD1 construct or the 35S::RFP::ΔGRAM construct show a typical chlorosis 72h after inoculation with suspensions (5x106 colony-forming units (CFU/mL) of Pst strain DC3000. vad1-1 plants and transgenic vad1-1 plants containing the 35S::RFP::ΔVASt construct do not show (or very faint) symptoms. (B) Growth of Pst DC3000 in the different lines indicated. Leaves of four-week old plants were inoculated with a bacterial suspension (2x105 CFU/mL) of Pst strain DC3000 and bacterial growth was measured 0 (grey bars) and 3 (black bars) days after inoculation. Data were collected from four independent experiments with five individual plants (four leaves/plant) per point. Statistical differences using Kruskal and Wallis one-way analysis of variance followed by the Conover's-test for multiple comparisons analysis of variance (P value < 0.05) are indicated by letters.

https://doi.org/10.1371/journal.pone.0179782.g005

Discussion

The mechanistic regulation of PCD is a fundamental question in plant biology. Our understanding of this biological process in response to environmental stresses including pathogens, came for a large part from the identification of mutants displaying spontaneous HR-like cell death, the LMMs [1, 10]. Indeed, a considerable number of reports link these LMM phenotypes to ETI components (R genes and signaling pathways)[16]. However, the functions of these negative regulators of immunity and cell death remain poorly understood.

Here we demonstrate that overexpression of VAD1 controls negatively the HR cell death and defense expression either transiently in N. benthamania or in Arabidopsis plants in response to avirulent strains of Pseudomonas syringae. Bax-Inhibitor-1 (BI-1), a conserved cell death regulator in yeast, animals and plants, was also shown to suppress elicitor-induced cell death [45]. Importantly, its overexpression conferred reduced susceptibility to the necrotrophic pathogen Fusarium graminearum and decreased resistance to the biotrophic parasite Blumeria graminis f.sp. hordei, suggesting that it can modulate cell death in interaction with plant pathogens of different lifestyles [46]. Interestingly, another protein, structurally related to VAD1 and which has been recently identified [47], is required for regulation of cell death and basal resistance to Botrytis cinerea. This C2 GRAM domain-containing protein is an interactor of the Arabidopsis BAG6 protein (BAGP1, for BAG-Associated GRAM Protein). BAGP1 is required for BAG6 processing and this cleavage is required for resistance to Botrytis. BAGP1 is one of the rare examples among the numerous and ubiquitous C2 GRAM domain containing proteins, for which a function is known: regulation of autophagic cell death and resistance to Botrytis, functions similar to those identified to VAD1. In this context, our data support that over-expression of VAD1 weakens defense and resistance to the biotrophic bacterial pathogen Pst DC3000 (Fig 5). This is in good agreement with the autoimmune phenotype observed in the mutant. This is also in line with VAD1 gene expression pattern during plant–pathogen interactions [14]. Indeed, VAD1 is triggered early in response to inoculation with avirulent strains of Xanthomonas and Pseudomonas, and then rapidly repressed to reach the background level observed in healthy plants. Altogether these results demonstrate that VAD1 controls HR cell death kinetics and intensity and defense. However, VAD1 is involved in the control of cell death propagation and not its initiation, in contrast to most LMM genes characterized to date [14]. Our observations (Fig 4) of the run-away cell death phenotype in the mutant line in one hand, and of highly limited HR cell death in the overexpressor line, are clearly in favor of this hypothesis.

Further insights into the roles of VAD1 in the control of plant defense and cell death control should be gained from examination of its biochemical function. VAD1 encodes a protein harboring a GRAM domain predicted to bind lipids [14] and a VASt domain related to Bet v1-like domains known to bind hydrophobic ligands such as phytohormones, lipids and polyketides [5]. In the present study, molecular dissection of the VAD1 gene demonstrated that the VASt domain is necessary for the inhibition of pathogen-induced cell death in N. benthamiana and in Arabidopsis thaliana. ΔVASt-VAD1 (lacking the internal 181 amino acid residues of the VASt domain of VAD1) completely abolished the cell death–suppression ability. On the other hand, ΔGRAM-VAD1, in which the N-terminal 65 amino acids of VAD1 were deleted, maintained its activity. Considering that VASt and GRAM domains are (i) associated in nearly all the eukaryotic VASt containing proteins, (ii) are both supposed to be involved in lipid/hydrophobic ligand [5], it is rather surprising to find that only the VASt domain is essential for cell death suppressor activity. In this context, VAD1 is clearly different from the mammalian CERT protein, with which VAD1 shares a common domain structure involving a PH superfamily domain (PH and GRAM respectively) and a Bet v1-like superfamily domain (START and VASt respectively)[5]. Cooperation between the PH and START domains in CERT is critical for its function as a ceramide transport protein [48]. This suggests that the two VAD1 domains (GRAM and VASt) are probably involved in different functions and/or behave independently.

Another interesting feature of the VASt domain is related to the subcellular localization of VAD1. Analysis of the plant cellular localization of VAD-1 indicated that the localization of VAD1 was in multiple subcellular compartments and was not affected by the GRAM or VASt domain deletions, excepted in one case: the nuclear localization of VAD1 was abolished by the ΔVASt deletion (Fig 2). While deletion of the VASt domain leads to abolition of both nuclear localization and cell death suppressor activity, it is tempting to hypothesize that the VASt domain is crucial for cell death activity through its nuclear localization. However, the fact that when the analysis of the VAD-1 cellular localization was performed with a C-terminal GFP fusion construct under the control of the VAD1 promoter (instead of a construct 35S::RFP::VAD1), a much lower level of expression was observed in all compartments, especially in the nucleus (almost undetectable), makes this hypothesis difficult to verify. Other approaches would be necessary to clarify this point. Subcellular localization of plant cell death regulator proteins has been investigated only in a few cases. For example, the C-terminal region of BnBI-1 is located on the cytosolic face of the ER [49], and supposed to play an important role in cell death regulation, through a potential role at the level of ER membrane homeostasis in relation to oxidative stress. However, the precise mechanism of cell death suppression by BI-1 remains unknown. Another Arabidopsis cell death regulator, EDR2, encodes a protein with a structural organization close to that of VAD1 and consisting of a PH domain, START domain and a plant-specific (DUF1336) domain of unknown function [50, 51]. EDR2 localizes to the plasma membrane, endosomes and endoplasmic reticulum [51] and its PH domain (close to GRAM domain) has been shown to bind specifically to phosphatidylinositol- 4-phosphate in vitro. However, the function of the START or the DUF1336 domains has not been characterized, either biochemically or in relation to subcellular protein localization. Finally, among numerous proteins containing both a GRAM and a VASt domain, Ysp2p, one of six StART-like proteins in yeast, has two StART-like domains which have been recently demonstrated both to bind sterol [20]. Interestingly, localization of YSP2 in the ER at contact sites with the plasma membrane, has been shown to be essential for its function, potentially involved in the rate of transfer of sterol from PM to ER [52]. However, the specific role of the START domains has not been addressed genetically. To our knowledge, the present work constitutes the first structure-function analysis performed on such cell death related proteins containing both a GRAM and a VASt domain. Although the exact function of the VASt domain of VAD1 remains to be discovered, its essential role in VAD1 cell death regulation has been genetically demonstrated.

Now, an important step in unraveling the role of VAD1 in restricting cell death in plant-pathogen interactions, would be to determine the ligand(s) (lipid, hormone or other molecules) bound by the VASt domain, which has been demonstrated here to be essential for the cell death regulatory function of VAD1. Another powerful, but more indirect approach, to understand VAD1-mediated cell death and defense responses, would be to identify additional components in the VAD1 signaling pathway, by conducting a vad1 suppressor screen.

Supporting information

S1 Fig. VAD1 overexpression leads to reduction of HR cell death and defense in response to bacterial avirulence effectors.

Observation of HR induced by AvrPto after agroinfiltration of N. benthamiana leaves alone or co-expressed with the constructs 35S::RFP::VAD1, 35S::RFP::ΔGRAM, 35S::RFP::ΔVASt. Observations were made 56h post-inoculation. (B) Quantification of cell death by measuring electrolyte leakage 48h after agroinfiltration of N. benthamiana leaves with the indicated strains (OD 0.5). Data are expressed relative to AvrPto data at 1h after sampling. Statistically significant differences were determined using Kruskal and Wallis one-way analysis of variance followed by nonparametric multiple comparison (* indicates P < 0.05).

https://doi.org/10.1371/journal.pone.0179782.s001

(TIF)

S2 Fig. VAD1 localizes in multiple subcellular compartments and a VASt domain deletion excludes VAD1 from the nucleus in Arabidopsis thaliana.

Transient co-expression by agroinfiltration of Arabidopsis thaliana leaves of 35S::RFP::VAD1 construct (A), 35S::RFP::ΔVASt construct (B) or pVAD::VAD1::GFP (C), the construct alone (left panels), with subcellular markers (central panels, ERD2–GFP labels the golgi/endoplasmic reticulum, MYB30-GFP labels the nucleus). Co-localization of VAD1 with the different subcellular markers is shown in the merge panel (right). Confocal images were observed two days after agroinfiltration. Scale bars = 20 μM.

https://doi.org/10.1371/journal.pone.0179782.s002

(TIF)

S3 Fig. VAD1 protein detection in protein extracts from Nicotiana leaves after transient expression or in Arabidopsis transgenic lines.

Transient expression of VAD1 in presence or not of bacterial avirulence effectors: N. benthamiana leaves were agroinfiltrated with the constructs 35S::RFP::VAD1, 35S::RFP::ΔGRAM, 35S::ΔVASt::VAD1, 35S::SYMREM1 alone, or co-expressed with (A) AvrRpt2 (OD 0.3) or (B) AvrPto (OD 0.5). Leaves were harvested 24h post-inoculation. Blots of total leaf protein extracts were incubated with RFP antibody or GFP antibody. The full-length RFP-VAD1, SymREM1-CFP protein or truncated VAD1 proteins VAD1-ΔGRAM and VAD1-ΔVAST fused to the RFP are predicted to be respectively 94 kDa, 60 kDa, 86 kDa and 73 kDa. Western blot shows VAD1 protein detection (VAD) and VAD1 derivatives (ΔGRAM, ΔVASt). VAD1 and derivatives are not detectable in the presence of AvrRpt2. Ponceau S staining confirms equal loading. These experiments were repeated three times with similar results. (C) Immunoblot analysis of total protein extracts from 2–3 week old T3 transgenic plants, separated by SDS-PAGE and probed with α-RFP antibody. Transgenic lines were generated in the WS-4 vad1 background. Ponceau S staining of the membrane confirms equal loading. Arrows on the left indicate protein positions.

https://doi.org/10.1371/journal.pone.0179782.s003

(EPS)

Acknowledgments

Authors would like to thank Dr B. Lefebvre for providing SymREM constructs and antiRFP antibodies. This work was performed on the Toulouse Plant Microbe Phenotyping platform (TPMP) (https://www.fraib.fr/fraib_eng/Technological-platforms/Plant-microbe-phenotyping). This work was supported by the French Laboratory of Excellence project "TULIP" (ANR-10-LABX-41; ANR-11-IDEX-0002-02).

References

  1. 1. Williams B, Dickman M (2008) Plant programmed cell death: can’t live with it; can’t live without it. Mol. Plant Pathol. 9: 531–544. pmid:18705866
  2. 2. Van Durme M, Nowack MK (2016) Mechanisms of developmentally controlled cell death in plants. Curr Opin Plant Biol 29: 29–37. pmid:26658336
  3. 3. Jones JD, Dangl JL (2006) The plant immune system. Nature 16; 444(7117):323–9. pmid:17108957
  4. 4. Macho A.P. and Zipfel C (2014) Plant PRRs and the activation of innate immune signaling. Mol. Cell, 54, 263–272. pmid:24766890
  5. 5. Khafif M, Cottret L, Balagué C, Raffaele S (2014) Identification and phylogenetic analyses of VASt, an uncharacterized protein domain associated with lipid-binding domains in Eukaryotes. BMC Bioinformatics 2014, 15:222. pmid:24965341
  6. 6. Watanabe N, Lam E (2009) Bax Inhibitor-1, a conserved cell death suppressor, is a key molecular switch downstream from a variety of biotic and abiotic stress signals in plants. Int J Mol Sci 10: 3149–3167. pmid:19742129
  7. 7. Coll NS, Epple P, and Dangl JL (2011) Programmed cell death in the plant immune system. Cell Death and Diffrentiation 18: 1247–1256.
  8. 8. Vailleau F, Daniel X, Tronchet M, Montillet JL, Triantaphylides C, and Roby D (2002) A R2R3-MYB gene, AtMYB30, acts as a positive regulator of the hypersensitive cell death program in plants in response to pathogen attack. Proc Natl Acad Sci USA 99: 10179–10184. pmid:12119395
  9. 9. Lorrain S, Vailleau F, Balagué C, and Roby D (2003) Lesion mimic mutants: keys for deciphering cell death and defense pathways in plants? Trends in Plant Science 8: 263–271. pmid:12818660
  10. 10. Bruggeman Q, Raynaud C, Benhamed M, and Delarue M (2015) To die or not to die? Lessons from lesion mimic mutants. Fontiers in Plant Science 6: 1–22.
  11. 11. Cannone J, Marino D, Jauneau A, Pouzet C, Brière C, Roby D, and Rivas S (2011) The Xanthomonas Type III Eeffector XopD targets the Arabidopsis transcription factor MYB30 to suppress plant defense. Plant Cell 23, 3498–3511. pmid:21917550
  12. 12. Mackey D, Holt BF, Wiig A, and Dangl JL (2002) RIN4 interacts with Pseudomonas syringae type III effector molecules and is required for RPM1-mediated resistance in Arabidopsis. Cell 108: 743–754. pmid:11955429
  13. 13. Aviv DH, Rustérucci C, Holt BF 3rd, Dietrich RA, Parker JE, Dangl JL (2002) Runaway cell death, but not basal disease resistance, in lsd1 is SA- and NIM1/NPR1-dependent. Plant J. 29(3) 381–91. pmid:11844114
  14. 14. Lorrain S, Lin B, Auriac MC, Kroj T, Saindrenan P, Nicole M, et al. (2004). Vascular associated death 1, a novel GRAM domain-containing protein, is a regulator of cell death and defense responses in vascular tissues. The Plant Cell 16, 2217–2232. pmid:15269331
  15. 15. Simanshu DK, Zhai X, Munch D, Hofius D, Markham JE, Bielawski J, Bielawska A, Malinina L, Molotkovsky JG, Mundy JW, Patel DJ, Brown RE (2014) Arabidopsis accelerated cell death 11, ACD11, is a ceramide-1-phosphate transfer protein and intermediary regulator of phytoceramide levels. Cell Rep. 6(2):388–99. pmid:24412362
  16. 16. Rodriguez E, El Ghoul H, Mundy J, Petersen M (2016) Making sense of plant autoimmunity and negative regulators. FEBS J. 283, 1365–1391.
  17. 17. Bouchez O, Huard C, Lorrain S, Roby D, Balagué C (2007) Ethylene is one of the key elements for cell death and defense response control in the Arabidopsis lesion mimic mutant vad1. Plant Physiol. 145(2):465–77. pmid:17720753
  18. 18. Berger P, Schaffitzel C, Berger I, Ban N, and Suter U (2003) Membrane association of myotubularin-related protein 2 is mediated by a pleckstrin homology-GRAM domain and a coiled-coil dimerization module. Proc. Natl. Acad. Sci. U. S. A. 100, 12177–12182. pmid:14530412
  19. 19. Tsujita K, Itoh T, Ijuin T, Yamamoto A, Shisheva A, Laporte J, Takenawa T (2004) Myotubularin regulates the function of the late endosome through the GRAM domain-phosphatidylinositol 3,5-bisphosphate interaction. J. Biol. Chem. 279, (14) 13817–13824. pmid:14722070
  20. 20. Sokolov S, Knorre D, Smirnova E, Markova O, Pozniakovsky A, Skulachev V, Severin F (2006) Ysp2 mediates death of yeast induced by amiodarone or intracellular acidification. Biochim Biophys Acta Bioenerg 1757(9):1366–1370.
  21. 21. Liang H, Yao N, Song JT, Luo S, Lu H, Greenberg JT (2003) Ceramides modulate programmed cell death in plants. Genes Dev. 17, 2636–2641. pmid:14563678
  22. 22. Bi FC, Liu Z, Wu JX, Liang H, Xi XL, Fang C, et al. (2014) Loss of ceramide kinase in arabidopsis impairs defenses and promotes ceramide accumulation and mitochondrial H2O2 bursts. Plant Cell. 26, 3449–3467. pmid:25149397
  23. 23. Brodersen P, Petersen M, Pike HM, Olszak B, Skov S, Odum N, et al. (2002). Knockout of Arabidopsis accelerated-cell-death 11 encoding a sphingosine transfer protein causes activation of programmed cell death and defense. Genes Dev. 16, 490–502. pmid:11850411
  24. 24. Ternes P, Feussner K, Werner S, Lerche J, Iven T,Heilmann I, et al. (2011). Disruption of the ceramide synthase LOH1 causes spontaneous cell death in Arabidopsis thaliana. New Phytol. 192, 841–854. pmid:21883234
  25. 25. La Camera S, Balague C, Gobel C, Geoffroy P, Legrand M, Feussner I, et al. (2009).The Arabidopsis patatin-like protein 2 (PLP2) plays an essential role in cell death execution and differentially affects biosynthesis of oxylipins and resistance to pathogens. Mol. Plant Microbe Interact. 22, 469–481. pmid:19271961
  26. 26. Balagué C, Lin B, Alcon C, Flottes G, Malmstrom S, Kohler C, et al. (2003). HLM1, an essential signaling component in the hypersensitive response, is a member of the cyclic nucleotide-gated channel ion channel family. The Plant Cell 15, 365–379. pmid:12566578
  27. 27. Liu D, Shi L, Han C, Yu J, Li D, Zhang Y. (2012) Validation of reference genes for gene expression studies in virus-infected Nicotiana benthamiana using quantitative real-time PCR. PLoS One 7(9):e46451. pmid:23029521
  28. 28. Laemmli UK (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227 (5259): 680–685. pmid:5432063
  29. 29. Clough SJ, Bent AF. (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16(6):735–43. pmid:10069079
  30. 30. Boevink P, Oparka K, Santa Cruz S, Martin B, Betteridge A, Hawes C (1998) Stacks on tracks: the plant Golgi apparatus traffics on an actin/ER network. Plant J. 15(3):441–7. pmid:9750355
  31. 31. Nelson BK, Cai X, Nebenführ A (2007) A multicolored set of in vivo organelle markers for co-localization studies in Arabidopsis and other plants. The Plant Journal 51, 1126–1136. pmid:17666025
  32. 32. Haasen D1, Köhler C, Neuhaus G, Merkle T (1999) Nuclear export of proteins in plants: AtXPO1 is the export receptor for leucine-rich nuclear export signals in Arabidopsis thaliana. Plant J. 20(6):695–705. pmid:10652141
  33. 33. Lefebvre B, Batoko H, Duby G, Boutry M (2004). Targeting of a Nicotiana plumbaginifolia H+-ATPase to the plasma membrane is not by default and requires cytosolic structural determinants. The Plant Cell 16(7), 1040–4651.
  34. 34. Gao X, Alvo M, Chen J, Li L (2008) Nonparametric multiple comparison procedures for unbalanced two-way layouts. J. Statistical Planning and Inference, 138 (12), 3674–3686.
  35. 35. Mudgett MB, Staskawicz BJ. (1998) Protein signaling via type III secretion pathways in phytopathogenic bacteria. Curr Opin Microbiol. 1(1):109–14. pmid:10438234
  36. 36. Lefebvre B, Timmers T, Mbengue M, Moreau S, Hervé C, Tóth K, et al. (2010). A remorin protein interacts with symbiotic receptors and regulates bacterial infection. Proc. Natl. Acad. Sci. USA 107: 2343–2348. pmid:20133878
  37. 37. Yang H, Yang S, Li Y, and Hua J (2007) The Arabidopsis BAP1 and BAP2 genes are general inhibitors of programmed cell death. Plant Physiol. 145: 135–146 pmid:17631528
  38. 38. Saito C, Ueda T, Abe H, Wada Y, Kuroiwa T, Hisada A, Furuya M, Nakano A (2002). A complex and mobile structure forms a distinct subregion within the continuous vacuolar membrane in young cotyledons of Arabidopsis. The Plant Journal, 29(3), 245–255. pmid:11844103
  39. 39. Marion J, Bach L, Bellec Y, Meyer C, Gissot L, Faure JD (2008) Systematic analysis of protein subcellular localization and interaction using high-throughput transient transformation of Arabidopsis seedlings. Plant J. 56: 169–179 pmid:18643979
  40. 40. Buell CR, Somerville SC (1995) Expression of defense-related and putative signaling genes during tolerant and susceptible interactions of Arabidopsis with Xanthomonas campestris pv. campestris. Mol. Plant Microbe Interact. 8:435–443.
  41. 41. Pontier D, Tronchet M, Rogowsky P, Lam E, Roby D (1998) Activation of hsr203, a plant gene expressed during incompatible plant-pathogen interactions, is correlated with programmed cell death. Mol Plant Microbe Interact. 11(6):544–54. pmid:9612953
  42. 42. Ronald PC, Salmeron JM, Carland FM, Staskawicz BJ (1992) The cloned avirulence gene avrPto induces disease resistance in tomato cultivars containing the Pto resistance gene. J. Bacteriol. 174:1604–11. pmid:1537802
  43. 43. Axtell MJ1, Chisholm ST, Dahlbeck D, Staskawicz BJ (2003) Genetic and molecular evidence that the Pseudomonas syringae type III effector protein AvrRpt2 is a cysteine protease. Mol Microbiol. 49(6):1537–46. pmid:12950919
  44. 44. Chisholm ST, Dahlbeck D, Krishnamurthy N, Day B, Sjolander K, Staskawicz BJ (2005) Molecular characterization of proteolytic cleavage sites of the Pseudomonas syringae effector AvrRpt2. Proc Natl Acad Sci U S A. 8; 102(6):2087–92. pmid:15684089
  45. 45. Matsumura H, Nirasawa S, Kiba A, Urasaki N, Saitoh H, Ito M et al. (2003) Overexpression of Bax inhibitor suppresses the fungal elicitor-induced cell death in rice (Oryza sativa L.) cells. Plant J 2003; 33: 425–434. pmid:12581301
  46. 46. Babaeizad V, Imani J, Kogel KH, Eichmann R, Huckelhoven R. (2008) Over-expression of the cell death regulator BAX inhibitor-1 in barley confers reduced or enhanced susceptibility to distinct fungal pathogens. Theor Appl Genet 2008; 118: 455–463. pmid:18956174
  47. 47. Li Y, Kabbage M, Liu W, Dickman MB (2016) Aspartyl protease-mediated cleavage of BAG6 is necessary for autophagy and fungal resistance in plants. The Plant Cell, 28: 233–247. pmid:26739014
  48. 48. Hanada K, Kumagai K, Tomishige N, Kawano M. (2007) CERT and intracellular trafficking of ceramide. Biochim Biophys Acta 1771(6):644–53. pmid:17314061
  49. 49. Bolduc N, Ouellet M, Pitre F, Brisson LF (2003) Molecular characterization of two plant BI-1 homologues which suppress Bax-induced apoptosis in human 293 cells. Planta 216: 377–386. pmid:12520328
  50. 50. Tang D, Ade J, Frye CA, Innes RW (2005) Regulation of plant defense responses in Arabidopsis by EDR2, a PH and START domain-containing protein. Plant J 44(2):245–257 pmid:16212604
  51. 51. Vorwerk S, Schiff C, Santamaria M, Koh S, Nishimura M, Vogel J, Somerville C, Somerville S (2007) EDR2 negatively regulates salicylic acid-based defenses and cell death during powdery mildew infections of Arabidopsis thaliana. BMC Plant Biology 7:35 pmid:17612410
  52. 52. Gatta AT, Wong LH, Sere YY, Calderon-Noren DM, Cockcroft S, Menon AK, Levine TP (2015) A new family of StART domain proteins at membrane contact sites has a role in ER-PM sterol transport. eLife 4:e07253. pmid:26001273