VAPYRIN attenuates defence by repressing PR gene induction and localized lignin accumulation during arbuscular mycorrhizal symbiosis of Petunia hybrida

Summary The intimate association of host and fungus in arbuscular mycorrhizal (AM) symbiosis can potentially trigger induction of host defence mechanisms against the fungus, implying that successful symbiosis requires suppression of defence. We addressed this phenomenon by using AM‐defective vapyrin (vpy) mutants in Petunia hybrida, including a new allele (vpy‐3) with a transposon insertion close to the ATG start codon. We explore whether abortion of fungal infection in vpy mutants is associated with the induction of defence markers, such as cell wall alterations, accumulation of reactive oxygen species (ROS), defence hormones and induction of pathogenesis‐related (PR) genes. We show that vpy mutants exhibit a strong resistance against intracellular colonization, which is associated with the generation of cell wall appositions (papillae) with lignin impregnation at fungal entry sites, while no accumulation of defence hormones, ROS or callose was observed. Systematic analysis of PR gene expression revealed that several PR genes are induced in mycorrhizal roots of the wild‐type, and even more in vpy plants. Some PR genes are induced exclusively in vpy mutants. Our results suggest that VPY is involved in avoiding or suppressing the induction of a cellular defence syndrome that involves localized lignin deposition and PR gene induction.

We next treated colonized roots of all Medicago genotypes with phloroglucinol to reveal lignin deposition. As in petunia, both wild type accessions (A17 and R108) did not accumulate any lignin at the point of fungal infection (Fig. S11a,b). Surprisingly, however, none of the mutants accumulated significant amounts of lignin at infection points ( Fig. S11c-g), including the vpy-2 mutant, while prominent lignin deposition was observed along the vasculature in the stele (Fig. S11h), as in all other genotypes (data not shown). These results indicate that lignin accumulation at AMF infection points is not part of the phenotype of symbiosis mutants in Medicago.

Medicago truncatula growth conditions and treatments
Seeds of Medicago truncatula were scarified with concentrated sulfuric acid for ten minutes, then rinsed five times with sterile water. Subsequently, seeds were surface-sterilised in concentrated Clorox for two minutes, followed by addition of an equal volume of water and an additional minute of sterilization. Then seeds were rinsed with sterile water five times and imbibed at room temp on an orbital shaker for four hours. Then seeds were placed on 0.8 % plant agar in a growth chamber (20 ºC, 18 h light) for germination. Seedlings were transferred to sterilized double plastic jars with perlite in the top and B&D nutrient solution (Broughton & Dilworth, 1971) containing 1mM KNO3 in the bottom jar, connected by a cotton wick. After one month, seedlings were transferred to nurse plant pots with mycorrhizal jive plants that had been inoculated at least one month before with R. irregularis. b-1,3-Glucanase immunostaining Wild type an vpy-3 plants were inoculated with nurse plant inoculum of R. irregularis. Root material was fixed for 2h with 2% (w/v) paraformaldehyde and 1% (v/v) glutaraldehyde at room temperature. Fixed material was dehydrated and embedded in Lowicryl K4M (www.sigmaaldrich.com) as described (Altman et al., 1984) with the following modifications: embedding involved a series of lowicryl diluted with 95% (v/v) ethanol as follows: 1:2 (vol:vol) for 2h, 1:1 for 2h, 2:1 overnight, followed by twice 100% lowicryl for 2h and polymerization at 55°C. Ultrathin sections (70 nm) were blocked with 20 mM NH4Cl, 20 mM lysine and 20 mM glycine. Rabbit antiserum raised against extracellular glucanase (Beffa et al., 1993) was used at a dilution of 1:5 in blocking agent. Goat-anti-rabbit antibodies coupled to 10 nm gold beads (http://www.bbisolutions.com) served as secondary antibodies. Contrasting was performed with 2% (w/v) uranyl acetate (UO2(CH3COO)2) and lead citrate solution prepared according to (Reynolds, 1963). Controls without the primary antibody did not show any labeling. For quantification of immunogold signal, from each treatment six representative pictures of hypodermal cells from two independent experiments were printed out at the same magnification. A 2D grid of small crosses at regular distances (1 cm) printed on a transparent foil was overlaid and the gold particles were counted in the respective cellular compartments relative to the number of crosses within the same compartment.

RNA extraction and quantitative real-time RT-PCR
For gene expression analysis, plants were either inoculated for four weeks with nurse plants (Tables 1,2), or wild type plants were treated with chitin oligosaccharides or with a Penicillium preparation (Thuerig et al., 2005) for 1 and 4h at concentrations of 1 µg/ml and 10 µg/ml, respectively (Table S9). Frozen petunia roots were placed in 2 mL Eppendorf tubes containing a glass bead and were ground using a ball mill. Total RNA was extracted from the powdered roots according to the protocol of the Direct-zol RNA miniprep kit from Zymo Research (https://www.zymoresearch.com), using trizol solution for lysis (38% v/v), saturated phenol (pH 8), 0.8 M guanidine thiocyanate, 0.4 M ammonium thiocyanate, 0.1M Na-acetate pH 5, and 5% (v/v) glycerol. The Direct-zol RNA kit involves a DNAse step to remove genomic DNA during RNA extraction. RNA was used for reverse transcription according to the protocol of the SensiFAST tm cDNA synthesis kit from Bioline. PCR reactions were carried out with 5 µL of 100x diluted cDNA solution, 1 µL of 10 µM forward and reverse primer, 7.5 µL of SensiMix SYBR Hi-ROX (BIO-RAD) and DNAse/RNAse free water up to 15 µL. The reaction cycle was 95°C for 10 min followed by 45 amplification cycles (95°C for 20s, 64°C for 20s, 72°C for 20s). All samples were analyzed in technical duplicates from six independent replicate plants. Actin and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were used as reference genes. Relative expression values were calculated using the delta-Ct method (Pfaffl, 2001). Data are expressed as -fold changes in relative gene expression (AM/control; vpy/wt) in Tables 1, 2, S3, S8, and S9. Expression values in Tables S4, S5, and S10-S12 represent expression values normalized to actin and GAPDH. Gene expression data were statistically analyzed using two-way ANOVA for interaction between genotype and treatment, followed by one way ANOVA as described below.
Statistical analysis qPCR results were tested by two-way ANOVA with host genotype and mycorrhizal status as the two factors of interest. The p-values of the two factors and their interactions are listed in Table S13. All other data, and the derived gene induction ratios from qPCRs, were treated by one-way ANOVA to test for significance of differences among groups (mutants vs. wild type, AM vs. controls) and the corresponding p-values are also listed in Table S13. Statistical analysis of callose accumulation (  Figure S1. Quantitative phenotypic analysis of vpy mutants. Mycorrhizal roots of wild type and vpy mutants were stained with Trypan Blue for quantitative assessment of fungal structures as indicated. Mutants formed infection pegs instead of intracellular coils in the hypodermis, and aborted or malformed structures instead of arbuscules. Shown are the mean + Stdv (n=4). Different letters indicate significant differences (one-way ANOVA).            Columns represent means + Stdv (n=5). There was no significant difference between any of the treatments (one-way ANOVA).

Figure S13. Jasmonic acid levels in mycorrhizal wild type and vpy-3 roots.
(a) Accumulation of free jasmonic acid in wild type P. hybrida (white columns), and vpy-3 mutants (black columns) after 10 days (10d) and 35 days (35d) with or without inoculation with R. irregularis (AM). (b) Accumulation of isoleucine-conjugated jasmonic acid (JA-Ile) in wild type P. hybrida (white columns), and vpy-3 mutants (black columns) after 10 days (10d) and 35 days (35d) with or without inoculation with R. irregularis (AM). Columns represent means + Stdv (n=5). Different letters indicate significant differences (one-way ANOVA).  Same sample as in Fig. 8d. An aborted fungal penetration hypha (degenerating fungus) grown next to a hypodermal cell (upper right) and a cortex cell (lower left). The hypodermal cell has formed a thick cell wall apposition that exhibits strong immunogold signal. hl, hypodermal layer; 1°, primary cell wall; 2°, secondary cell wall. Scale bar, 500 nm. Figure S16. Low-magnification overview picture of the sample shown in Fig. S15. Figure S17. Quantification of b-1,3-glucanase immunogold signal Immuno-gold signal was quantified in six representative TEM images as shown in Fig. S14 and Fig. S15 from two independent experiments. Gold particles were counted in the cell walls of hypodermal cells and expressed as relative density per unit of section area. Shown are the mean + Stdv. Different letters indicate significant differences (one-way ANOVA, n=6)

Table S1. Quantification of callose accumulation in vpy mutants
For each genotype at least 30 infection sites with at least one infected cell were chosen randomly. Presence of callose at infection sites was evaluated as shown in Fig. S6 with strong or weak signal at penetration sites into hypodermal cells. The majority of infection sites did not exhibit any callose signal (negative). The column "callose all" refers to the accumulative cases in which either strong or weak signal was detected. The frequency of callose formation was not different in mutants compared to the wild type (Fisher's Exact Test for count data).  Table S2. Primers used for qRT-PCR of lignin-related genes Sequence identifiers and qRT-PCR primers for genes encoding predicted lignin biosynthetic enzymes. Gene IDs refer to gene names from the Solanaceae Genome Network database (https://solgenomics.net/) using the JBrowse function in the Petunia axillaris genome (https://solgenomics.net/organism/Petunia_axillaris/genome). Table S3. Induction of lignin-related genes in vpy mutants vs. wild type either in the mycorrhizal or non-mycorrhizal context. Induction of lignin biosynthetic genes was determined by qRT-PCR with actin and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as reference genes. Values representfold induction ratios derived by dividing the expression values of mutants by the values from the wild type, either in mycorrhizal (left) or control conditions (right). All expression values represent the average of six biological replicates. Color shading represents induction >2-fold (yellow), >4-fold (orange), and >8-fold (red). Data represent two independent experiments, one with only vpy-1 vs. wild type (asterisks), and one with vpy-2 and vpy-3 vs. wild type. Significant induction ratios are indicated by bold type font (one-way ANOVA, n=6). Table S4. Expression values for lignin-related genes for the experiment on vpy-1 from the genes for which induction ratios were calculated in Table 1 and Table S3. Table S5. Expression values for lignin-related genes for the experiment on vpy-2 and vpy-3 from the genes for which induction ratios were calculated in Table 1 and Table S3. Table S8. Induction of pathogenesis-related (PR) genes in mutants vs. wild type either in the mycorrhizal or non-mycorrhizal context. Induction of PR genes was determined by qRT-PCR with actin and glyceraldehyde-3phosphate dehydrogenase (GAPDH) as reference genes. Values represent -fold induction ratios derived by dividing the expression values of mutants by the values from the wild type, either in mycorrhizal (left) or control conditions (right). All expression values represent the average of six biological replicates. Color shading represents induction >2-fold (yellow), >4-fold (orange), and >8-fold (red). Data represent two independent experiments, one with only vpy-1 vs. wild type (asterisks), and one with vpy-2 and vpy-3 vs. wild type. Significant induction ratios are indicated by bold type font (one-way ANOVA, n=6). Table S9. Induction of pathogenesis-related (PR) genes in wild type plants treated with fungal elicitor preparations. Induction of PR genes was determined by qRT-PCR with actin and glyceraldehyde-3phosphate dehydrogenase (GAPDH) as reference genes. Values represent -fold induction ratios derived by dividing the expression values of roots treated with chitin hydrolysate (Chit) or Penicillium preparation (Pen) for 1h or 4h by the values of control roots. All expression values represent the average of six biological replicates. Color shading represents induction >2-fold (yellow), and >4-fold (orange). Significant induction ratios are indicated by bold type font (one-way ANOVA, n=6). Table S10. Expression values for the experiment on vpy-1 from the genes for which induction ratios were calculated in Table 2 and Table S8. Table S11. Expression values for the experiment on vpy-2 and vpy-3 from the genes for which induction ratios were calculated in Table 2 and Table S8. Table S12. Expression values from the experiment with wild type plants treated with elicitors from genes for which induction ratios were calculated in Table S9.