The OXI1 kinase pathway mediates Piriformospora indica-induced growth promotion in Arabidopsis

Abstract

Piriformospora indica is an endophytic fungus that colonizes roots of many plant species and promotes growth and resistance to certain plant pathogens. Despite its potential use in agriculture, little is known on the molecular basis of this beneficial plant-fungal interaction. In a genetic screen for plants, which do not show a P. indica- induced growth response, we isolated an Arabidopsis mutant in the OXI1 (Oxidative Signal Inducible1) gene. OXI1 has been characterized as a protein kinase which plays a role in pathogen response and is regulated by H₂O₂ and PDK1 (3-PHOSPHOINOSITIDE-DEPENDENT PROTEIN KINASE1). A genetic analysis showed that double mutants of the two closely related PDK1.1 and PDK1.2 genes are defective in the growth response to P. indica. While OXI1 and PDK1 gene expression is upregulated in P. indica-colonized roots, defense genes are downregulated, indicating that the fungus suppresses plant defense reactions. PDK1 is activated by phosphatidic acid (PA) and P. indica triggers PA synthesis in Arabidopsis plants. Under beneficial co-cultivation conditions, H₂O₂ formation is even reduced by the fungus. Importantly, phospholipase D (PLD)α1 or PLDδ mutants, which are impaired in PA synthesis do not show growth promotion in response to fungal infection. These data establish that the P. indica-stimulated growth response is mediated by a pathway consisting of the PLD-PDK1-OXI1 cascade.

Figure 1. P. indica-mediated increase in fresh weight (%) of wild-type (WT) and mutant plants.

Data are based on 5–9 independent experiments with 10 plants per treatment. Bars represent SEs (significant difference to WT; * p<0.05; ** p<0.001).

Figure 2. Characterization of PDK1 mutants.

(A) PDK1.1, PDK1.2 and GAPC2 transcript amounts were determined by RT-PCR with gene-specific primer pairs. The pdk1.1 pdk1.2 line does not contain pdk1 transcripts. Two independent pdk1.1 lines (a and b), pdk1.2 and pdk1.1 pdk1.2 were analysed. (B) Siliques and (C) phenotypes of pdk1.1 pdk1.2 and wild-type plants are shown.

Figure 3. Plant PA levels increase in response to treatment with P. indica exudate.

Five-days old seedlings were ³²P_i-labelled overnight and then treated with P. indica exudates. (A) Time series of plant PA amounts induced by 50 µl P. indica exudates. (B) Dose response curve of plant PA production in response to different amounts of P. indica exudate. Lipids were extracted, analysed by thin layer chromatography and PA levels were quantified by phosphoimaging. ³²P-PA control levels were ∼1.5% of the total ³²P-labelled lipids. The values represent: radioactivity [+P. indica extract/+buffer]. Bars represent SEs, based on 3 independent experiments. Bars marked with an asterisk are significantly different compared to wild type (p<0.05).

Figure 4. P. indica-induced increases in fresh weight (%) of wild-type and pld mutants.

Data are based on at least three independent experiments with 10 plants per treatment. SEs are shown. Bars marked with an asterisk are significantly different compared to wild type (p<0.05).

Figure 5. Expression levels of PDK1, OXI1 and AGC2-2 genes and defense genes in colonized wild-type or mutant roots relative to uncolonized control plants.

Panel A shows PDK1, OXI1 and AGC2-2 expression levels and panel B shows expression levels of several defense genes. RNA was extracted from roots and real-time PCR analyses were performed with the housekeeping gene UBQ5 as control. Calculations were performed according to . Bars show the mean out of at least four independent experiments with SEs. Data are presented on a log scale.

Figure 6. Expression levels of PDK1, OXI1 and AGC2-2 genes and defense genes after treatment with a high dosis of P. indica (for details, cf. Methods and Materials and Figure S4A,B in Text S1).

Panel A shows PDK1, OXI1 and AGC2-2 expression levels and panel B shows expression levels of several defense genes. RNA was extracted from roots and real-time PCR analyses were performed with the housekeeping gene GAPC2 as control. Calculations were performed according to . Bars show fold-induction of RNA values from wild type and agc mutant roots 7 days after co-cultivation on a fungal lawn relative to the RNA levels from seedlings grown in the absence of the fungus. Bars show the mean out of at least three independent experiments with SEs. Data are presented on a log scale.

Figure 7. Proposed model describing the role of PLD, PA, AGC and MAP kinases in the beneficial interaction between P. indica and Arabidopsis.

For MPK, see .