Calcium channel CNGC19 mediates basal defense signaling to regulate colonization by Piriformospora indica in Arabidopsis roots.

The activation of calcium signaling is a crucial event for perceiving environmental stress. Colonization by Piriformospora indica, a growth-promoting root endosymbiont, activates cytosolic Ca2+ in Arabidopsis roots. In this study, we examined the role and functional relevance of calcium channels responsible for Ca2+ fluxes. Expression profiling revealed that CYCLIC NUCLEOTIDE GATED CHANNEL 19 (CNGC19) is an early-activated gene, induced by unidentified components in P. indica cell-wall extract. Functional analysis showed that loss-of-function of CNGC19 resulted in growth inhibition by P.indica, due to increased colonization and loss of controlled fungal growth. The cngc19 mutant showed reduced elevation of cytosolic Ca2+ in response to P. indica cell-wall extract in comparison to the wild-type. Microbe-associated molecular pattern-triggered immunity was compromised in the cngc19 lines, as evidenced by unaltered callose deposition, reduced cis-(+)-12-oxo-phytodienoic acid, jasmonate, and jasmonoyl isoleucine levels, and down-regulation of jasmonate and other defense-related genes, which contributed to a shift towards a pathogenic response. Loss-of-function of CNGC19 resulted in an inability to modulate indole glucosinolate content during P. indica colonization. CNGC19-mediated basal immunity was dependent on the AtPep receptor, PEPR. CNGC19 was also crucial for P. indica-mediated suppression of AtPep-induced immunity. Our results thus demonstrate that Arabidopsis CNGC19 is an important Ca2+ channel that maintains a robust innate immunity and is crucial for growth-promotion signaling upon colonization by P. indica.

Introduction

Piriformospora indica (syn. Serendipita indica) is a cultivable, root-colonizing endophytic fungus belonging to Sebacinales (Basidiomycota) (Verma ; Weiß ). It colonizes many plant species including Arabidopsis and promotes their growth (Varma ; Peškan‐Berghöfer ; Vadassery ), enhances nutrient uptake (Yadav ; Rani ; Bakshi ; Prasad ), and imparts tolerance to abiotic and biotic stresses to a wide range of its hosts (Waller ; Baltruschat ; Jogawat et al., 2013, 2016; Sun ). It colonizes the root epidermal and cortex cells without penetrating the central cylinder, and displays a biphasic colonization strategy (Deshmukh ; Zuccaro ). However, the establishment of a beneficial plant–microbe interaction is not always harmonious, and rejection of the invading symbiont or control of its colonization can occur due to active plant defense (Vadassery and Oelmüller, 2009). Basal plant defense relies on the recognition of conserved microbial structures called microbe-associated molecular patterns (MAMPs), and is termed MAMP-triggered immunity (MTI) (Millet ). In soil, plant roots perceive MAMPs using specific pattern-recognition receptor (PRR) proteins (Choi and Klessig, 2016). Recognition of MAMPs triggers downstream early plant-defense responses such as elevation of cytosolic calcium (Ca2+cyt) and a burst of reactive oxygen species (ROS), which further activates mitogen-activated protein (MAP) kinase and various phytohormone pathways that stimulate defense-related pathways (Harper and Harmon, 2005; Ranf ; Steinhorst and Kudla, 2013). Similar to plant pathogens, mutualists such as P. indica are also confronted with an effective innate immune system in roots, and the colonization success depends on the evolution of strategies for immunosuppression (Van Wees ; Jacobs ). During colonization by Serendipita indica on Arabidopsis, eATP, which acts as a damage-associated molecular pattern (DAMP) accumulates in the apoplast. Serendipita indica secrets SiE5′NT, an enzymatically active nucleotidase capable of hydrolysing eATP, in the apoplast and thus suppresses immunity (Nizam ). During the early stages of mycorrhiza formation and P. indica colonization, H2O2 is produced and its production declines when a mutualistic interaction is established (Fester and Hause, 2005; Matsuo ). Piriformospora indica also actively represseses ROS accumulation by activating ROS-scavenging genes (Matsuo ). To achieve a harmonious interaction with plants, P. indica also regulates biosynthesis and signaling of several phytohormones such as jasmonic acid (JA), gibberallins (GA), and ethylene (Camehl ; Sun ; Vahabi ; Pan ; Xu ). Piriformospora indica association also alters callose deposition and defense-related metabolites, such as phytoalexins and glucosinolates (GS) (Jacobs ; Lahrmann ). Indole glucosinolates (iGS) are an important part of MTI in plants (Clay ; Böhm ) and they are found to be essential in balancing the beneficial interaction between P. indica and Arabidopsis (Nongbri ; Lahrmann ). In addition, P. indica also suppresses innate immunity upon encountering the flagellin 22 elicitor from bacteria (Jacobs ). An active plant immunity and its suppression is thus critical for controlled P. indica colonization. The early-activated plant defense genes that are responsible for regulating the entry of the symbiont and its subsequent colonization are unknown.

Ca2+ is a universal second messenger, activated very early in signaling cascades upon recognition of both pathogens and symbionts. Rhizobacteria-mediated nodulation and mycorrhiza formation are associated with oscillations in nuclear Ca2+ in host plants. These oscillations upon perception of rhizobia and mycorrhiza activate induction of the common genes that are important for the establishment of the symbioses (Oldroyd, 2013). Colonization by P. indica in Arabidopsis is independent of these common arbuscular mycorrhizal symbiotic genes (Banhara ). However, elevation of Ca2+ is common with other symbiotic interactions, as P. indica cell-wall extract (PiCWE) elevates root Ca2+cyt and is crucial for growth promotion in Arabidopsis (Vadassery ). Using the elevation of Ca2+ as a marker, Johnson identified cellotriose (CT) as the major elicitor in crude PiCWE. It was further confirmed that CT targets a poly(A)-specific ribonuclease in order to modulate plant responses such as elevation of Ca2+cyt, generation of ROS, expression of defense-related genes, phytohormonal signaling, and growth promotion. Elevation of Ca2+cyt requires entry of Ca2+ either across the plasma membrane or from intracellular compartments. In Arabidopsis, ligand-gated channels such as cyclic nucleotide gated channels (CNGCs), glutamate receptor-like channels (GLRs), stretch-activated Ca2+ channels (OSCAs), and the MID1-complementing activity (MCA) families are the four main plasma membrane Ca2+-permeable channels, whilst the slow vacuolar two-pore channel 1 (TPC1) is the key vacuolar channel (Dodd ). The Arabidopsis genome encodes 20 members of the CNGC family, with roles in plant development and a functions related to biotic and biotic stresses (Meena and Vadassery, 2015; DeFalco ). CNGC2, CNGC4, CNGC11, and CNGC12 have been reported to play crucial roles in defense against bacterial and fungal pathogens (Yoshioka ; Ahn, 2007), and we have recently identified a role of the CNGC19 Ca2+ channel in herbivory-induced Ca2+ flux and plant defense against Spodoptera litura (Meena ).

CNGC15 has been identified as critical nuclear channel that generates oscillatory Ca2+ signals during arbuscular mycorrhizal symbiosis with Medicago trancatula roots (Charpentier ). In Lotus japonicus, a mutation in the AtCNGC19 homolog BRUSH is reported to result in impaired infection by nitrogen-fixing rhizobia due to a leaky channel (Chiasson ). The identity of the channel involved in the elevation of Ca2+cyt that is induced by P. indica is not yet known. The P. indica elicitor CT induces expression of GLR Ca2+ channels in Arabidopsis roots; however, (Johnson ) found no functional roles for GLR3.3, GLR2.4, GLR2.5, and TPC1 in this response. Expression levels of CNGCs are altered upon PiCWE treatment in plant roots and these are the only other type of Ca2+ channel known to be involved in the interaction (Vadassery ). We therefore hypothesized that CNGCs might be involved in the generation of elevated Ca2+ in Arabidopsis roots and in the downstream signaling in response to P. indica mutualism. Our results point to a role of Arabidopsis CNGC19 as an important gatekeeper to regulate P. indica colonization.

Materials and methods

Plant and fungal material and conditions

Piriformospora indica (Verma ) was grown and maintained on Kaefer’s medium at 28±2 °C at 110 rpm (Varma ; Hill and Kafer, 2001). For P. indica co-cultivation we used Arabidopsis thaliana wild-type Columbia (Col-0), the T-DNA mutant lines of AtCNGC19 (At3g17690) SALK_129200C (cngc19-2) and SALK_027306 (cngc19-1), which were provided by TAIR (Alonso ), and the pepr1 pepr2 double-mutant line provided by Prof. Gerald Berkowitz (University of Connecticut, USA). Adult plants were grown at 22 °C with a 10/14 h light/dark photoperiod and a light intensity of 150 µmol m−2 s−1 in a growth room (Percival Scientific). For Ca2+ measurements, we used transgenic Col-0 expressing cytosolic apoaequorin, (referred to as WT::aeq; Knight ), and the cngc19 and pepr1 pepr2 mutants transformed with the pMAQ2 vector (referred to as cngc19::aeq and pepr1 pepr2::aeq, respectively. The T2 generation was used.

Plant and fungal interactions in soil and co-cultivation media

For soil experiments, seeds were sown in pots containing soilrite, Irish peat moss, and exfoliated vermiculite (1:1:1, w:w:w) and kept for 2 d at 4 °C in the dark for stratification. The soil was pre-mixed with 1% P. indica mycelia (w/w), and plants were grown for 6 weeks after stratification. Control plants had no P. indica mycelia in the soil. The plants were grown at 22 °C with a 10/14 h light/dark photoperiod and a light intensity of 150 µmol m−2 s−1 in the growth room. Samples were harvested at 42 d post-inoculation (dpi).

For plate experiments, seeds were surface-sterilized, stratified under the conditions described above, and placed on half-strength MS plates supplemented with 1% sucrose and 0.8% agar, and germinated for 7 d. The seedlings were grown at 22 °C with a 10/14 h light/dark photoperiod and a light intensity of 150 µmol m−2 s−1 in the growth room. They were then transferred to 1× PNM medium for co-cultivation with P. indica discs for 14 d (Johnson ) under similar conditions. Samples were harvested at 2, 7, and 14 dpi.

Preparation of P. indica cell-wall extract and application on roots

Piriformospora indica cell-wall extract (PiCWE) was prepared as described by Vadassery . In brief, the mycelia from 14-d-old liquid cultures were homogenized, filtered using nylon membranes, and washed three times with water, twice with chloroform/methanol (1:1), and finally twice with acetone. The mycelial cell wall material obtained was dried at room temperature, suspended in water, and autoclaved for 30 min at 121 °C. It was then filter-sterilized using a 0.22-μM filter, and 50 µl of the resulting extract was used per seedling root for experiments. The active elicitor of PiCWE was recently identified as cellotriose (CT; Johnson ). CT (Sigma, C1167) and cellobiose (Sigma, C7252) were used for some experiments.

Plant treatments and gene expression analysis

Seedlings at 10 d old were treated with 100 μl PiCWE or 10 µM CT by adding to the MS media and were harvested at 0, 15, 30, 45, and 60 min. For the co-cultivation experiment, the seedlings were harvested at 2, 7, and 14 dpi. Each sample consisted of six seedlings and was ground to a fine powder in liquid N2, and total RNA was isolated using TRIzol Reagent (Invitrogen) according to the manufacturer’s protocol. Four replicate samples were used. An additional DNAse (Turbo DNAse, Ambion) treatment was included to eliminate any contaminating DNA. cDNA synthesis was performed using a High Capacity cDNA kit (Applied Biosystems). Gene-specific primers were designed using the NCBI primer design tool (http://www.ncbi.nlm.nih.gov/tools/primer-blast) and are listed in Supplementary Table S1 at JXB online. qRT-PCR was performed in optical 96-well plates on a CFX96 Real-Time PCR Detection System (Bio-Rad) using iTaq universal SYBR green Mix (Bio-Rad). AtActin2 (At3g18780) was used as the endogenous control for normalization of transcripts. The fold-induction values of the target genes were calculated using the ΔΔCT method (Livak and Schmittgen, 2001) and were expressed relative to the mRNA level of the genes in the control seedlings, the values of which were set as 1.

Detection and measurement of P. indica colonization

For tracking of colonization, a green fluorescent protein (GFP)-tagged P. indica strain was utilized (Hilbert ). Roots colonized with tagged P. indica were harvested at 2, 7, and 14 dpi, and were cleaned, mounted, and observed using fluorescence microscopy (Nikon 80i). For confocal microscopy, the roots were treated with propidium iodide and observed under a confocal microscope (Leica TCS M5) at an emission wavelength of 505–530 nm with excitation at 470 nm and digital sectioning of 4–5 µm of root thickness. The relative amount of fungal DNA was determined using real time-qPCR utilizing Arabidopsis Actin2 (At3g18780) and P. indica Tef1 (Bütehorn ). Relative changes in fungal DNA content were calculated using the CT values of PiTef1, which were normalized by the CT values of AtActin2 using the ΔΔCT equation and setting the P. indica DNA content of the control roots as 1 (Vadassery ).

Tissue localization by GUS assays

Transgenics with a fusion of the CNGC19 promoter and β-glucuronidase (GUS) were constructed as previously described by Meena . Arabidopsis seedlings of ProCNGC19::GUS-expressing transgenic plants (T3 generation) were co-cultivated with P. indica, and were carefully harvested at 2 dpi and 7 dpi. They were vacuum-infiltrated with GUS staining solution and incubated in the dark at 37 °C. Tissues were decolored by treating with 70% ethanol at 65 °C and then observed under a light microscope (Nikon 80i).

Measurements of elevation of cytoplasmic Ca2+

WT::aeq, cngc19-2::aeq, and pepr1 pepr2::aeq seedlings (Meena ) were grown on MS medium and roots of 14-d-old seedlings were transferred to a 96-well white plate (ThermoFisher Scientific) containing 5 μM coelenterazine (PJK, Germany) and left in the dark overnight at 21 °C. Bioluminescence counts in the roots were recorded as relative light units (RLU) per second using a microplate luminometer (Luminoscan Ascent, v. 2.6, ThermoFisher Scientific). After a 1-min background reading, PiCWE (50 µl), cellobiose (100 µM), or CT (10 µM) was added and readings were taken for 15 min. Discharge solution (1 M CaCl2 and 10% ethanol) was used for calibrations to estimate the aequorin that remained at the end of the experiment (Vadassery ; Meena ). The luminescence counts obtained were calibrated using the equation presented by Rentel and Knight (2004).

Glucosinolate analysis

For analysis of glucosinolates (GS), samples of plants were harvested at 2, 14, and 42 dpi. At 2 dpi and 14 dpi, 40 whole seedlings per replicate were harvested, whilst at 42 dpi whole rosettes were harvested. Four replicate samples were used at all time points.. The samples were frozen in liquid N2, lyophilized, and ground to a fine powder in TissueLyser II (Qiagen). Total GSs were extracted with 80% methanol solution containing 0.05 mM 4-hydroxybenzylglucosinolate as an internal standard. Extracts were loaded onto DEAE Sephadex A 25 columns and treated with arylsulfatase for desulfation (Sigma-Aldrich). The eluted desulfoglucosinolates were separated using HPLC (Shimadzu CLASS-VP V6.14) on a reversed-phase C-18 column (250×4.6 mm with 0.5 μm internal diameter) with a water–acetonitrile gradient as follows: 0–1.5% acetonitrile from 0–1 min, 1.5–5% acetonitrile from 1–6 min, 5–7% acetonitrile from 6–8 min, 7–21% acetonitrile from 8–18 min, 21–29% acetonitrile from 18–23 min, 29–100% acetonitrile from 23–24 min, 100–1.5% acetonitrile from 24–28 min. This was followed by a washing cycle with a flow of 1 ml min−1 (Vadassery ). Detection was performed using a photodiode array detector and peaks were integrated at 229 nm. The following response factors were used for quantification of individual glucosinolates: aliphatic glucosinolates, 2.0; indole glucosinolates, 0.5; and 2-phenylethyl glucosinolate, 2.0 (Burow ).

Estimation of phytohormones

Phytohormones were quantified as described previously (Vadassery ; Meena ). Seedlings were sampled at 2 dpi and 14 dpi. The samples were harvested, frozen immediately in liquid N2, lyophilized, and ground to a fine powder. Weighed, powdered samples (25 mg) were extracted using 1.5 ml of methanol containing internal standards of 60 ng d6-jasmonic acid (HPC Standards GmbH, Cunnersdorf, Germany), 60 ng salicylic acid-d4 (Santa Cruz Biotechnology), 60 ng abscisic acid-d6 (Toronto Research Chemicals), and 12 ng d6-jasmonic acid-isoleucine conjugate (HPC Standards GmbH). A triple-quadruple LC-MS/MS system was used for phytohormone quantification (SCIEX 6500).

Callose staining, microscopy, and quantification

Seedlings at 2, 7, and 14 dpi were treated with Aniline Blue (0.001%) according to the protocol described by Schenk and Schikora (2015). Colonized and non-colonized seedlings, and control and AtPep1-treated (100 nM) 7 d old seedlings were incubated at room temperature in acetic acid and ethanol (1:3) for decolorization, washed in 150 mM K2HPO4, and stained with Aniline Blue (0.001%) solution. Slides for observing callose deposition were prepared using 50% glycerol under a Nikon 80i microscope at 358 nm excitation and 460 nm emission. Relative callose intensities were calculated by dividing callose pixels and total pixels using the digital photograph analysis software GIMP (Scalschi ).

Phylogenetic analysis

A total of 123 complete CNGC sequences from seven different plants were selected for construction of the phylogenetic tree, namely Arabidopsis thaliana, Glycine max, Medicago truncatula, Solanum lycopersicum, Zea mays, Oryza sativa, and Lotus japonicus. All the amino acid sequences encoding CNGCs were retrieved from previously published reports (Moeder ; Nawaz ; Charpentier ; Chiasson ; Moeder and Yoshioka, 2017) and are listed in Supplementary Table S2. A phylogenetic tree was constructed using the MEGA 10 software (https://www.megasoftware.net/; Kumar ), in which the sequences were aligned by MUSCLE with default parameters. These aligned sequences were used to build the phylogenetic tree using the maximum likelihood (ML) method and the evolutionary distances were computed using a Jones–Taylor–Thornton matrix-based method with 1000 bootstrap replications.

Statistical analysis

Statistical differences between treatments were analysed using two-tailed Student’s t-tests or one-way ANOVA followed by Tukey’s test in SigmaPlot 13.0. Figures were generated using Origin 6.0 (www.originlab.com).

Results

CNGC19 expression is activated by P. indica cell-wall extract

To identify the role of CNGCs in the perception of P. indica by Arabidopsis, we applied either crude cell-wall extract (PiCWE) or its identified active elicitor cellotriose (CT) to Arabidopsis seedlings for 30 min (Fig. 1A). Upon treatment with PiCWE, five CNGCs were found to be induced in Arabidopsis roots, namely CNGC19 (14.3-fold increase), CNGC3 (6.22-fold), CNGC13 (5.9-fold), CNGC10 (5-fold), and CNGC6 (3.7-fold). In response to CT, nine CNGCs were induced, with the highest expression being found for CNGC3 (17.5-fold); however, CNGC19 was not induced by CT. Since it was the highest expressed upon PiCWE treatment and the only transcript specifically induced by PiCWE but not by CT, we hypothesized that CNGC19 was induced by unidentified elicitors. We examined the patterns of CNGC19 expression in wild-type (WT) roots and found that it increased from 10–45 min in response to PiCWE but there was no response to CT (Fig. 1B). We then examined roots of plants co-cultivated with P. indica, and found that expression of CNGC19 was increased by 7-fold at 2 d post-colonization, before returning to the basal level at later time-points (Fig. 1C), thus indicating a potential role in the colonization process. To identify the tissue-specific expression pattern of CNGC19 in Arabidopsis after colonization by P. indica, we used ProCNGC19::GUS. CNGC19 promoter activity was observed in the root primordia and primary vasculature at 2 dpi and 7 dpi (Fig. 1D, E). Interestingly, we also observed a systemic expression of CNGC19 in the leaf vasculature at the same time. The results therefore indicated that CNGC19 is an early-activated gene that is expressed in the vasculature, and it is induced in the roots by unidentified components in PiCWE and systemically in leaves.

Fig. 1.

Expression and localization of CNGC19 in Arabidopsis in response to colonization by P. indica. (A) Expression profiling of CNGCs in 10 d old seedlings after treatment for 30 min with P. indica cell-wall extract (PiCWE) and cellotriose (CT). The heat map represents the fold-change of mRNAs in the treated samples relative to the controls. Data are based on four replicates with six seedling per replicate. Transcripts levels were normalized using Actin2 mRNA. (B) Expression of CNGC19 in 10-d-old seedlings at different times after treatment with either 50 µl PiCWE or 10 µM CT. (C) CNGC19 expression in Arabidopsis seedlings co-cultivated with P. indica on 1× PNM. In (A, B), transcripts levels were normalized using Actin2 mRNA and are presented as the fold-change relative to expression at time zero, which was set as 1. Data are means (±SE) of four replicates, each consisting of six seedlings. (D, E) CNGC19 promoter activity in roots and leaf vasculature upon P. indica colonization at (D) 2 d post-inoculation (dpi) and (E) 7 dpi. Seedlings expressing ProCNGC19::GUS were co-cultivated with P. indica on 1× PNM. Arrows indicate ProCNGC19::GUS expression.

The growth of cngc19 mutants is inhibited by P. indica

To identify the functional role of CNGC19 in the promotion of growth induced by P. indica, we utilized the cngc19-2 and cngc19-1 mutant T-DNA lines (Meena ). In contrast to the WT plants, we observed no growth promotion in the cngc19 mutants in response to P. indica for parameters such as fresh weight (Fig. 2A, B) and root length (Supplementary Fig. S1) when plants were grown in culture medium. Indeed, the mutants showed growth inhibition at 7 dpi and 14 dpi compared to the WT (Fig. 2B). We found similar effects when we repeated the experiment with plants grown in soil, with the growth of the mutants being strongly inhibited at 42 dpi whereas growth promotion was observed in the WT in response to P. indica (Supplementary Fig. S2A–C).

Fig. 2.

Effects of P. indica colonization on Arabidopsis cngc19 mutants. (A) Representative images of the wild-type (WT), and the cngc19-2 and cngc19-1 lines after co-cultivation with P. indica (Pi-treated) for 14 d compared with non-inoculated controls. (B) Fresh weights of the WT, cngc19-2, and cngc19-1 at 2–14 d post-inoculation (dpi) with P. indica compared with non-inoculated controls. Data are means (±SE), n=20 seedlings. Different letters indicate significant differences among the means as determined using one-way ANOVA and a post hoc Tukey test (P≤0.05). (C) Colonization patterns of P. indica on the WT and cngc19-2 as determined by confocal microscopy. GFP-tagged P. indica was visualized at 14 dpi and arrows indicate chlamydospores and hyphae. DIC, differential interference contrast images. (D) Quantification of P. indica colonization in the WT and cngc19-2 mutant grown on plates (14 dpi) and in soil (42 dpi). The relative fungal colonization was calculated by subtracting the CT values of P. indica Tef1 from the CT values of Arabidopsis Actin2. Data are means (±SE) of four replicates, each of which consisted of the combined roots of six seedlings. Significant differences were determined using two-tailed Student’s t-tests (*P≤0.05).

Colonization by P. indica is enhanced in cngc19 roots

We tested the hypothesis that the reduced growth in cngc19 mutants upon P. indica inoculation was due to enhanced colonization. The roots of the WT and cngc19-2 were co-cultivated with GFP-tagged P. indica and were used for microscopic analyses. We observed no differences between P. indica colonization in the WT and cngc19-2 at 2 dpi and 7 dpi (Supplementary Fig. S3); however, at 14 dpi cngc19 roots had increased colonization compared to the WT and clumps of fungal mycelia and spores could be observed (Fig. 2C). This was supported by quantification of the relative fungal DNA content, which increased at 14 dpi (grown in medium) and 42 dpi (grown in soil) in cngc19-2 roots relative to the WT (Fig. 2D). Thus, the loss-of-function of CNGC19 resulted in increased colonization and loss of controlled P. indica growth in the plant–fungal interaction at the post-establishment phase.

The CNGC19 channel is involved in PiCWE-mediated elevation of cytosolic Ca2+

CNGC19 is a plasma membrane-localized Ca2+-permeable channel (Meena ) and we therefore hypothesized that it could be involved in generating the elevation in cytosolic Ca2+ in response to PiCWE. We used the WT::aeq and cngc19::aeq lines to examine intracellular Ca2+cyt upon application of PiCWE and CT to the roots. The substrate of CT, cellobiose (CB), was also used as an unrelated elicitor control. Both CT (Fig. 3A) and cellobiose (Supplementary Fig. S4) induced elevation of Ca2+cyt in the WT::aeq and cngc19 roots at similar levels. When PiCWE was added, the elevation in Ca2+cyt was reduced in the cngc19::aeq line relative to the WT::aeq (Fig. 3B), both in the initial peak and for several minutes thereafter. These results suggest that CNGC19 is a crucial channel that is involved in sensing as yet unidentified elicitors in PiCWE and in activating the elevation of Ca2+cyt.

Fig. 3.

Concentrations of cytosolic calcium (Ca2+cyt) in Arabidopsis in response to treatment with P. indica-related elicitors. Roots of transformed 10-d-old seedlings of the wild-type (WT) and cngc19 expressing cytosolic apoaequorin were treated with (A) cellotriose (10 µM) or (B) P. indica cell-wall extract (50 µl). Data are means (±SE), n=5. The experiment was repeated three times with similar results and the data from one experiment are shown. Water was used as the control and gave background readings in the WT and cngc19. The arrows indicate the time of treatment with the elicitors. Different letters indicate significant differences between the WT and cngc19 during the selected periods enclosed in the boxes, as determined using one-way ANOVA and a post hoc Tukey test (P≤0.001).

Callose deposition in response to P. indica colonization is delayed in cngc19

Increased callose deposition has been reported in Arabidopsis roots colonized by P. indica and indicates the activation of MTI (Jacobs ). In the WT plants, callose deposition was increased at 2 dpi and 7 dpi and then remained unchanged at 14 dpi during the established colonization phase (Fig. 4A, C). In contrast, in the cngc19-2 mutant the callose deposition was unaltered compared to the control at both 2 dpi and 7 dpi, indicating a reduced defense in these plants at these early stages. Callose deposition was not induced in cngc19-2 until 14 dpi. Thus, plant defense was lowered in the cngc19 mutant during the initial stages of the plant–fungal interaction, leading to increased colonization.

Fig. 4.

Patterns of callose deposition in the roots of the Arabidopsis wild-type and the cngc19 mutant in response to colonization by P. indica. (A) Representative images of callose deposition in roots of non-inoculated controls and in roots colonized with P. indica (+Pi) at 2–14 d post-inoculation (dpi) in the wild-type (WT) and the cngc19-2 mutant. (B) Representative images of callose deposition in roots of non-inoculated controls and in roots colonized with P. indica for the WT and cngc19-2 plants with or without AtPep1 treatment (100 nM) applied for 24 h at 2 dpi. (C) Relative callolose intensity of the roots shown in (A) as determined from staining for non-inoculated controls and for roots colonized by P. indica for WT and cngc19-2 seedlings at 2–14 dpi. The intensity was expressed as the number of fluorescent callose-corresponding pixels relative to that of the total number of pixels. Data are is means (±SE) of n=20 seedlings. (D) Relative callose intensity in roots shown in (B). Data are means (±SE) of n=15 seedlings. In (C, D), different letters indicate significant differences among the different treatments, as determined by one-way ANOVA and a post hoc Tukey Test (P≤0.001).

CNGC19-mediated basal immunity upon perception of P. indica is dependent on AtPep-PEPR

Plants roots encounter damage-associated molecular patterns (DAMPs) upon microbial invasion (Boller and Felix, 2009; Albert, 2013). In Arabidopsis, a family of endogenous elicitor peptides referred to as AtPeps acts as DAMPs (Huffaker and Ryan, 2007; Bartels ) and the plasma membrane Pep-receptors PEPR1 and PEPR2 perceive them (Yamaguchi , 2010; Krol ). We tested the possibility that P. indica may suppress AtPep-induced defense, and the role of CNGC19 in such a process. We found that callose deposition induced by application of AtPep1 was suppressed in WT seedlings inoculated with P. indica at 2 dpi (Fig. 4B, D). For the cngc19 mutant, the AtPep1-induced deposition of callose was constitutively lower than that of the WT for all the treatments, and it was not suppressed by colonization by P. indica (Fig. 4D). Thus, the results indicated that CNGC19 was also crucial for P. indica-mediated suppression of DAMP-triggered immunity. The Pep-receptors PEPR1 and PEPR2 have a putative guanyl cyclase domain that generates cyclic nucleotides and they are upstream of CNGC2 (Ma ). In order to determine the role of PEPR1 and PEPR2 in the Arabidopsis–P. indica association, we examined their expression. In WT seedlings, expression of PEPR1 and PEPR1 was found to be induced by PiCWE treatment (up to 3.5-fold; Fig. 5A) and by P. indica colonization (up to 6-fold; Supplementary Fig. S5A). To examine the dependency of CNGC19 activation on PEPR, we measured the expression of CNGC19 in the pepr1 pepr2 background in response to treatment with PiCWE, and found that it was reduced at 45 min and 60 min relative to the WT (Fig. 5B). We then examined the role of PEPRs in P.indica-induced growth promotion. Colonization by P. indica was relatively high in pepr1 pepr2, and instead of growth promotion, inhibition was observed both on plates (Fig. 5C, D) and in soil (Supplementary Fig. S6A, B). However, elevation of Ca2+cyt was unaltered when compared to the WT in pepr1 pepr2::aeq upon treatment with PiCWE (Fig. 5E) and CT (Supplementary Fig. S5B). PEPR signaling contributes to the JA signaling pathway upon herbivory (Klauser ; Meena ). The level of the JA marker VSP2 was found to be reduced in pepr1 pepr2 upon both treatment with PiCWE and colonization by P. indica (Fig. 5E, Supplementary S5C), indicating its function downstream of CNGC19 via jasmonate signaling.

Fig. 5.

The roles of PEPR1 and PEPR2 in the Arabidopsis–P. indica interaction. Expression of (A) PEPR1 and PEPR2 in wild-type (WT) and (B) expression of CNGC19 in the pepr1 pepr2 double-mutant in response to treatment with P. indica cell-wall extract (PiCWE) in 10-d-old seedlings. Transcripts levels were normalized to AtActin2 mRNA and the fold-change in expression is relative to the value at time zero, which was set as 1. Data are means (±SE) of four replicates, each of which consisted of six seedlings. Significant differences were determined using two-tailed Student’s t-test (*P<0.05). (C) Effects of P. indica colonization on the fresh weight of the WT and the pepr1 pepr2 double-mutant in non-inoculated controls in response to colonization by P. indica at 14 d post-inoculation (dpi). Data are means (±SE), n=30. Different letters indicate significant differences among the different treatments, as determined using one-way ANOVA and a post hoc Tukey Test (P≤0.001). (D) Effect of P. indica colonization on fungal colonization in the WT and the pepr1 pepr2 double-mutant. Plants were co-cultivated with or without fungal discs on 1× PNM agar plates and the roots were harvested at 14 dpi. The relative fungal colonization was calculated by subtracting the CT values of P. indica Tef1 from the CT values of Arabidopsis Actin2. Data are means (±SE) of four replicates, with six seedlings per replicate. The significant difference was determined using a two-tailed Student’s t-test (**P≤0.005). (E) Response of cytosolic calcium (Ca2+cyt) to treatment with P. indica cell-wall extract (PiCWE, 50 µl) in roots of transformed 10-d-old seedlings of the WT and pepr1 pepr2 expressing cytosolic apoaequorin. Data are means (±SE), n=5. The experiment was repeated three times with similar results and the data from one experiment are shown. Water was used as the control and gave background readings in the WT and pepr1 pepr2. The arrow indicates the time of treatment with the elicitor. (F) Expression of the defense-related gene VSP2 in response to treatment with PiCWE in roots of transformed 10-d-old seedlings of the WT and pepr1 pepr2 double-mutant expressing cytosolic apoaequorin. Transcripts levels were normalized to AtActin2 mRNA and the fold-change in expression is relative to the value at time zero, which was set as 1. Data are means (±SE), n=4.) Significant differences were determined using two-tailed Student’s t-test (*P≤0.05; **P≤0.005).

Loss-of-function of CNGC19 down-regulates jasmonate biosynthesis upon P. indica colonization

Piriformospora indica activates the JA signaling pathway in Arabidopsis, and this is crucial for colonization and for balancing the beneficial interaction between the two organisms (Stein ; Vahabi ). To identify the role of various phytohormones in P. indica colonization in the cngc19 mutant, we measured their levels at 2 dpi and 14 dpi. The levels of JA and jasmonoyl isoleucine (JA-Ile) were increased in WT plants in response to P. indica at 2 dpi and 14 dpi, and cis-(+)-12-oxo-phytodienoic acid (cis-OPDA) was also strongly increased at 2 dpi (Fig. 6A-C). In contrast, no significant effects of colonization were observed for cngc19. The lack of an effect on the levels of cis-OPDA, JA, and JA-Ile at 2 dpi may have contributed to uncontrolled colonization in cngc19 roots. JA-Ile-OH also showed similar trends (Supplementary Fig. S7A), but no changes in ABA and salicylic acid (SA) were observed in response to P. indica colonization in either genotype at either time-point (Supplementary Fig. S7B, C).

Fig. 6.

Effects of P. indica colonization on levels of phytohormones in seedlings of the Arabidopsis wild-type (WT) and the cngc19 mutant. (A) Jasmonates (JA), (B) the JA bioactive form (+)jasmonoyl isoleucine (JA-ILE), and (C) the JA precursor cis-(+)-12-oxo-phytodienoic acid (cis-OPDA) at 2 d post-inoculation (dpi) and 14 dpi. Data are means (±SE) of three replicates, each of which consisted of 40 seedlings. Different letters indicate significant differences between the WT and cngc19-2 plants at both time-points as determined using one-way ANOVA and a post hoc Tukey Test (P≤0.001).

Phytohormone- and defense-related genes are down-regulated in cngc19 during colonization

Since colonization by P. indica was enhanced in the cngc19 lines, we examined the expression of marker genes of different defense pathways such as those of JA, ROS, and phytoalexin. The JA markers VSP2, PDF1.2, and LOX1 were found to be induced by P. indica in WT plants at both 2 dpi and 14 dpi compared to non-inoculated controls, whereas these genes were found to be down-regulated in cngc19 except LOX1 at 2 dpi (Fig. 7A). The ROS markers RBOHD, RRTF1, and OXI1 were selected based on their known functional roles in the Arabidopsis–P. indica interaction (Camehl ; Matsuo ; Johnson ). RRTF1 was found to be up-regulated in the WT but not in cngc19 (Fig. 7B), RBOHD was up-regulated in both the WT and cngc19 and did not differ between the two, and OXI1 was strongly up-regulated at 2 dpi in the WT but showed no change in cngc19 compared to non-inoculated controls. In contrast, at 14 dpi OXI1 was observed to be strongly up-regulated in cngc19, which may have been due to increased colonization. In addition, the ROS-related genes SOD1, GSTF8, and APX1 were found to be up-regulated in the WT and down-regulated in cngc19 at 2 dpi (Supplementary Fig. S8), and at 14 dpi GR1 and CAT2 were found to be up-regulated in cngc19 but unaltered in the WT. WRKY33 and PAD3, which are related to phytoalexin biosynthesis, were also found to be up-regulated in the WT but not in cngc19 at 2 dpi (Fig. 7C). At 14 dpi, WRKY33 was up-regulated in both the WT and cngc19 and did not differ between the two, whilst PAD3 was also up-regulated in both but had a greater increase in the WT. Thus, the loss-of-function of CNGC19 affected JA-responsive genes, and genes involved in ROS signaling and defense, and was associated with over-colonization by P. indica in cngc19 plants.

Fig. 7.

Effects of P. indica colonization on expression of genes related to phytohormones, reactive oxygen species (ROS), and defense in seedlings of the Arabidopsis wild-type (WT) and the cngc19 mutant. Expression of marker genes for (A) jasmonates, (B) ROS, and (C) defense early 2 d post-inoculation (dpi) and 14 dpi. Transcripts levels were normalized to AtActin2 mRNA and the fold-change in expression is relative to that of the corresponding non-inoculated control, which was set as 1. Data are means (±SE) of four replicates, each of which consisted of six seedlings. Significant differences were determined using two-tailed Student’s t-test: *P≤0.05, ***P<0.0001.

CNGC19-mediated defense signaling in roots acts via indole glucosinolates

CNGC19 loss-of-function results in constitutively reduced levels of aliphatic GSs in Arabidopsis rosettes, which are crucial for defense against herbivory (Meena ). Upon plant–microbe interactions, accumulation of antimicrobial indole glucosinolates (iGSs) and camalexin trigger immunity (Bednarek ; Clay ; Böhm ). The iGS pathway is critical for mutualistic P. indica colonization and for uncompromised plant immunity (Nongbri ; Lahrmann ). We therefore decided to test the effects of mutation in CNGC19 on iGS levels. We found that the constitutive levels of both iGSs and aliphatic GSs in non-inoculated (control) seedlings were the same in the WT and cngc19 at 14 dpi (Fig. 8A, B). At the rosette stage in non-inoculated plants grown in soil (42 dpi), the level of iGSs did not differ between the WT and cngc19 whilst aliphatic GSs were significantly lower in cngc19, as also reported by Meena . We then looked at the effects of P. indica colonization and found that iGSs were increased in the WT at 14 dpi but were not affected in cngc19 (Fig. 8A). The levels of aliphatic GSs also showed a similar trend (Fig. 8B). After a prolonged period of colonization by P. indica (42 dpi, grown in soil), iGS levels were reduced significantly in WT plants but were unaltered in cngc19 (Fig. 8A). Aliphatic GS levels did not change at 42 dpi in either of the genotypes (Fig. 8B). We examined the relative expression of key genes related to the iGS biosynthesis pathway and found that they were generally up-regulated in the WT but not in cngc19 at 2 dpi and 14 dpi (Fig. 8C). The results therefore indicate that CNGC19 plays a crucial role in modulating the content of indole GSs during colonization by P. indica.

Fig. 8.

Effects of P. indica colonization on glucosinolates in seedlings of the Arabidopsis wild-type (WT) and the cngc19 mutant. Levels of (A) indole glucosinolates (iGS) and (B) aliphatic glucosinolates in non-inoculated controls and plants colonized by P. indica grown either on plates (14 d post-inoculation, dpi) or in soil (42 dpi). Data are means (±SE) of three replicates each consisting of 40 seedlings (14 dpi) or five replicates each consisting of one seedling (42 dpi). Different letters indicate significant differences among the different treatments as determined using one-way ANOVA and a post hoc Tukey’s test (P≤0.001). C. Relative expression of iGS biosynthesis pathway genes in seedlings grown on plates at 2 dpi and 14 dpi. Transcripts levels were normalized to AtActin2 mRNA and the fold-change in expression is relative to that of the corresponding non-inoculated control, which was set as 1. Data are means (±SE) of four replicates each consisting of six seedlings. Significant differences were determined using two-tailed Student’s t-test (*P≤0.05, **P≤0.005).

Phylogenetic analysis of CNGC19 indicates it has a distinct role in microbial interactions

Unlike rhizobial nodulation and mycorrhizal symbiosis, P. indica has a broad host range and is a primitive symbiont (Franken ). We constructed a phylogenetic tree in order to understand the relationship between AtCNGC19 and its orthologs in other host plants, many of which form symbiotic interactions. Genome-wide analyses in seven different species have identified distinct CNGCs (Mäser ; Nawaz ; Saand ; Charpentier ). In our unrooted phylogenetic tree, 123 CNGCs clustered into four different groups (Groups I–IV; Fig. 9). Among these, Group IV was further subdivided into IVA and IVB. AtCNGC19 and AtCNGC20 were clustered into Group IVA with orthologs from different legume plants that form symbiotic interactions, such as G. max, M. truncatula, and L. japonicas. AtCNGC19 was clustered with BRUSH from L. japonicas, which plays a crucial role in rhizobial symbiosis by regulating Ca2+ fluxes (Chiasson ). AtCNGC19 also grouped with OsCNGC13 (its ortholog in O. sativa; Moeder and Yoshioka, 2017), which has been shown to be up-regulated by a bacterial pathogen (Nawaz ). Nuclear-localized MtCNGC15 has been shown to be critical in generating oscillations in nuclear Ca2+ during mycorrhizal symbiosis (Charpentier ); however, it was found to be clustered into Group III, which suggests it is evolutionarily divergent from AtCNGC19.

Fig. 9.

Phylogenetic relationships of AtCNGC19 with homologs from other species. Multiple sequence analysis was performed using the MUSCLE software. A maximum-likelihood unrooted tree was constructed using 123 complete amino acid sequences of CNGCs in the MEGA10 software. The evolutionary relationships were analysed with 1000 bootstrap replicates. Each node is labelled with the gene ID and its previously reported name. Genes examined in the current study are highlighted in blue.

Discussion

Plant roots interact with both pathogenic and symbiotic microbes and recognize them as a potential threat by activation of basal MAMP-triggered immunity (MTI). Activation of MTI prevents the establishment of pathogens and regulates colonization by symbionts, and thus the process acts as a gatekeeper (Yu ). Root MTI efficiently restricts penetration and colonization of the mutualist P. indica and prevents over-colonization, with the result that a symbiotic interaction is established (Jacobs ). Recognition of P. indica cell-wall extract (PiCWE) by Arabidopsis roots induces elevation of cytosolic Ca2+ (Ca2+cyt) and is crucial in activating the symbiotic interaction and in promoting growth (Vadassery ; Johnson ). The identity of the ion channels responsible for the influx of Ca2+ and the activation of signaling is currently unknown. In our present study, we identified CNGC19 as an early-activated gene (Fig. 1A) and found that it was induced by unidentified components in PiCWE and not by treatment with cellotriose (CT) (Fig. 1A, B). Cyclic nucleotide gated channels (CNGCs) are altered by treatment with PiCWE and glutamate receptor-like channels (GLRs) are altered by treatment with CT in plant roots (Vadassery ; Johnson ). PiCWE (which contains many elicitors including CT) and CT also differ in their activation of other pathways. Importantly, CT induces a defense pathway comprising ROS accumulation and the expression of its marker gene RBOHD (Johnson ), whereas PiCWE does not activate this pathway (Vadassery ). We found that CNGC19 was crucial for P. indica-induced promotion of growth, as its loss-of-function resulted in increased colonization and the complete loss of growth-promotion phenotype (Fig. 2A, B). Importantly, CNGC19 was found to be critical for the generation of the PiCWE-induced elevation of Ca2+cyt. No other Ca2+ channels have yet been implicated in the growth promotion and elevation of Ca2+cyt that is induced by P. indica. Since PiCWE-induced Ca2+ signals were not completely abolished in the cngc19 mutant (Fig. 3B), they might be controlled by additional genetic interactions between CNGCs and other unknown channels. A leucine‐rich repeat protein mutant, Piriformospora indica-insensitive12 (pii12), has previously been reported to show no promotion of growth when it is associated with P. indica (Shahollari ). Overall, our results indicate that the PiCWE-activated CNGC19 is a critical Ca2+ channel for growth-promotion signaling.

Expression of CNGC19 in Arabidopsis is associated with salinity stress (Kugler ; Oh ). We have previously reported that CNGC19 expressed in the leaf vasculature is crucial for Arabidopsis defense against herbivory by Spodoptera moths by regulating the spread of Ca2+ signals and the levels of jasmonate and aliphatic glucosinolates (Meena ). The AtCNGC19 homolog SlCNGC15 in tomato is induced by both salinity stress and P. indica colonization (Ghorbani ). SlCNGC15 is also associated with disease resistance against the necrotropic fungus Sclerotinia sclerotiorum (Saand ). cngc19 and cngc20 mutants are also more susceptible to infection by Botrytis cinerea (Moeder ). The receptor kinase BAK1/SERK4 phosphorylates the Ca2+-channel complex CNGC20/CNGC19 and has crucial role in pathogen-induced cell death (Yu ). The roles of other CNGCs in plant defense have been demonstrated in many studies. AtCNGC2 and AtCNGC4 are known to regulate Ca2+-induced PAMP-triggered immunity (Chin ; Tian ). They act as a heterotetrameric Ca2+ channel and are phosphorylated and activated by the kinase BIK1 of the pattern-recognition receptor complex, triggering an increase in the concentration of Ca2+cyt (Tian ). CNGC2 and CNGC4, which are also known as dnd1 (defense no death1) and dnd2, respectively, are involved in the hypersensitivity response and DAMP perception during bacterial infection in Arabidopsis (Ahn, 2007). cngc11 and cngc12 are also hypersusceptible to fungal infection (Yoshioka ). In apple, overexpression of MdCNGC1 results in increased susceptibility to fungal infection and in reduced callose deposition when plants are treated with flg22 and chitosan (Zhang ). Interestingly, in our study CNGC2, CNGC4, CNGC11, and CNGC12 were not significantly induced by P. indica (Fig. 1A), indicating that a different set of channels is involved in the interaction. The brush mutant has been isolated in a screen of an ethyl-methanesulfonate-mutated population for plants defective in symbiotic cell development (Maekawa-Yoshikawa ). At 26 °C, brush roots are stunted and infection threads in root hairs do not progress into the cortex, resulting in the formation of non-infected nodules. This has been mapped as a gain-of-function CNGC.IVA mutation and it is orthologous to AtCNGC19 and AtCNGC20, resulting in a leaky tetrameric channel (Chiasson ). Similarly, nuclear-localized CNGC15 in Medicago forms a complex with the potassium-permeable channel DMI1, and is responsible for nuclear Ca2+ release upon mycorrhizal symbiosis (Charpentier ). Our phylogenetic analysis also indicated that CNGCs of diverse plants that clustered in Groups IV and III are involved in both symbiotic and pathogenic interactions (Fig. 9). Thus, activation of Ca2+ channels belonging to the CNGC family seems to be a conserved element in symbiotic interactions.

Increased P. indica colonization in the cngc19 mutants (Fig. 2) indicated that the normal functioning of CNGC19 is crucial for maintaining controlled colonization, and suggests that it has a role in MTI. Plants deposit callose (β1, 3-glucan) into cell walls upon microbial invasion as a part of MTI (Thordal-Christensen, 2003; Nürnberger and Lipka, 2005). Colonization by P. indica is known to induce callose deposition and additional exposure to the elicitor flg22 does not increase callose because of P. indica-mediated suppression of late MTI (Jacobs ). Thus, the suppression of callose deposition is required for progression of P. indica colonization. We found that callose deposition was initially unaltered in cngc19 (Fig. 4A, C), indicating that the plant defense was lowered and hence led to increased colonization and pathogen-like growth. Plants activate robust MAMP perception and subsequent MTI by the action of phytohormones such as salicylic acid, ethylene, and jasmonate to regulate P. indica colonization. It is known that colonization increases JA/JA-Ile levels in co-cultivated plants with the result that plant defense responses are altered and tolerances to pathogenic microbes and root herbivory are improved (Vahabi , 2015; Lahrmann ; Cosme ). It has also been reported that JA signaling is required to suppress late MTI in order to facilitate the progression of P. indica colonization at the late biotrophic stage (Jacobs ). It is known that the jasmonate insensitive1-1 (jin1-1) and jasmonate resistant1-1 (jar1-1) mutants show no promotion of growth upon P. indica colonization (Jacobs ). Loss-of-function of CNGC19 down-regulated jasmonate biosynthesis and JA-responsive genes upon P. indica colonization (Fig. 6 and 7A). This suggests that CNGC19-mediated signaling leads to the activation of JA/JA-Ile signaling (Fig. 10), as has also been observed for defense against Spodoptera herbivory in Arabidopsis leaves (Meena ). In our study, the cngc19 mutants also displayed reduced expression of early and late MTI-related genes involved in phytohormone, ROS, and secondary metabolite pathways (Fig. 7A–C), which contributed to unbalancing the mutualistic relationship between the fungus and host plant (Fig. 10). We also found that H2O2-induced Oxidative Signal Inducible1 (OXI1) kinase was reduced in cngc19 plants at an early stage of colonization (Fig. 7B). The oxi1 mutant together with the agc2-2 (OXI1 kinase homolog) and pdk1.1 pdk1.2 (3-PHOSPHOINOSITIDE-DEPENDENT PROTEIN KINASE1) mutants also have reduced growth upon P. indica colonization (Camehl ). In addition, we also found that genes related to the antioxidant system were altered at early and late stages of P. indica colonization, namely SOD1, GSTF8, GR1, CAT2, and APX1 (Supplementary Fig. S8). Thus, CNGC19 is critical for MTI responses in Arabidopsis roots (Fig. 10).

Fig. 10.

Schematic model of the role of CNGC19 in the interaction between Arabidopsis and P. indica. Upon perception of P. indica cell wall-associated elicitor(s), as yet unknown receptors are activated in the root and lead to activation of CNGC19, which elevates the level of cytosolic calcium. This in turn modulates downstream systemic and defense-related pathways of MAMP-triggered immunity (MTI), such as callose deposition, glucosinolate biosynthesis, and jasmonic acid signaling. AtPep-PEPR signaling functions downstream of CNGC19 via jasmonate signaling. The activation of these pathways synergistically leads to maintenance of the P. indica colonization at a level that provides a mutualistic association between the plant and the fungus.

Damage-associated molecular patterns (DAMPs) are recognized by leucine-rich repeat (LRR)-like receptors, which activate downstream signaling. AtPep1 is perceived as a stronger danger signal by Arabidopsis roots than the MAMP-like bacterial flg22 or chitin, and hence AtPep-PEPR signaling is a major component of surveillance in the roots (Poncini ). We have previously identified that CNGC19 is involved in AtPep1-induced elevation of Ca2+cyt (Meena ). Pep-induced PEPR signaling further intensifies the plant defense response together with MTI (Ross ). Upon addition of AtPep1, P. indica-mediated suppression of callose occurred in the WT but not in the cngc19 mutant (Fig. 4). pepr1 pepr2 double-mutants also showed a growth inhibition phenotype (Fig. 5C), but these genes are not involved in the PiCWE-induced elevation of Ca2+cyt and instead might be acting via the JA pathway (Fig. 5E, F). PEPR signaling maintains basal immunity by regulating JA and SA signaling, locally and systemically (Ross ; Yamada ), which agrees with the down-regulation of the JA-responsive gene VSP2 that we observed upon P. indica colonization in the pepr1pepr2 mutant (Fig. 5). Thus, PEPR signaling works downstream of CNGC19, contributes to the JA pathway, and may interact with unknown receptors and kinases for modulating downstream targets.

It is known that cngc19 mutants are constitutively deficient in aliphatic glucosinolate accumulation and that they hyperaccumulate its precursor, methionine (Meena ). CNGC19 modulates aliphatic glucosinolate biosynthesis in tandem with BRANCHED-CHAIN AMINO ACID TRANSAMINASE4 (BCAT4), which is involved in the chain elongation pathway of metionine-derived glucosinolates (Meena ). However, this phenotype appeared to be age-dependent as it was absent in the seedlings that we studied, and upon P. indica colonization we found that it was the activation of iGS that was important (Fig. 8A). Thus, regulation of glucosinolates by CNGC19 is age- and stimuli-dependent. The indolic glucosinolate pathway plays a major role in the growth restriction of P. indica (Lahrmann ). The loss-of-function mutants cyp79b2/3 and cyp81f2, which are genes involved in iGS biosynthesis, exhibit growth inhibition upon P. indica colonization similar to what we observed in cngc19 (Nongbri ; Lahrmann ). Cytochrome P450 enzymes (CYP79B2, CYP79B3, CYP81F2) and transcription regulators of iGS (MYB51, MYB122, WRKY33) and other iGS-related genes (IGMT1, IGMT2, PAD3, PEN2) have previously been observed to be stimulated during the interaction with P. indica (Jacobs ; Nongbri ; Lahrmann ; Peskan-Berghöfer ). It has also been shown that genes related to the iGS biosynthesis pathway are essential for callose deposition (Clay ). In our study, such genes were found to be down-regulated upon P. indica colonization in cngc19 (Fig. 8), and the levels of iGS were increased at 14 dpi and reduced at 42 dpi in colonized WT plants, but were unaltered in cngc19. The activation of iGS biosynthesis might occur via activation of CNGC19 in the colonized plants. Some other iGS-related mutants (myb34/51/122, pen2-1) have also been shown to have higher levels of P. indica colonization (Jacobs ; Nongbri ; Lahrmann ). In addition, the mutant of a β‐glucosidase (ΔPYK10), which is involved in hydrolysing iGS, has also been observed to have no growth promotion upon P. indica colonization (Sherameti ; Nakano ). All these findings place CNGC19 as an upstream element in iGS activation upon P.indica colonization.

In conclusion, our results indicate that CNGC19 is activated by as yet unidentified elicitors in the cell-wall extract of P.indica. The CNGC19-mediated pathway affects the basal immunity and the levels of phytohormones and glucosinolates in Arabidopsis upon colonization by P.indica, and subsequently affects the growth of the plants. These events do not occur in the cngc19 mutants, and this leads to over-colonization and detrimental effects on plant health and growth (Fig. 10). Thus, CNGC19 plays a central role as a gatekeeper during colonization by P. indica, maintaining a robust innate immunity that ensures that the interaction is mutualistic.

Supplementary data

Supplementary data are available at JXB online.

Table S1. Primers used in this study.

Table S2. Amino acid sequences of CNGCs used for constructing the phylogenetic tree.

Fig. S1. Root lengths of WT and cngc19-2 seedlings co-cultivated with P. indica.

Fig. S2. Growth of WT and cngc19 seedlings co-cultivated with P. indica in soil.

Fig. S3. Colonization patterns of P. indica in WT and cngc19 roots.

Fig. S4. Cytosolic Ca2+ levels in response to treatment with P. indica elicitors.

Fig. S5. Role of PEPR signaling in the Arabidopsis–P. indica interaction.

Fig. S6. Growth effects of P. indica colonization in the pepr1 pepr2 double-mutant.

Fig. S7. Phytohormone levels in WT and cngc19-2 seedlings during P. indica colonization.

Fig. S8. Relative gene expression of several ROS-related genes in WT and cngc19 seedlings after P. indica inoculation.

Click here for additional data file.

Click here for additional data file.