Abstract
We analyzed the requirement of specific defense pathways for powdery mildew (Golovinomyces orontii) resistance induced by the basidiomycete Piriformospora indica in Arabidopsis. Piriformospora indica root colonization reduced G. orontii conidia in wild-type (Col-0), npr1-3 (nonexpressor of PR genes 1-3) and NahG plants, but not in the npr1-1 null mutant. Therefore, cytoplasmic but not nuclear localization of NPR1 is required for P. indica-induced resistance. Two jasmonate signaling mutants were non-responsive to P. indica, and jasmonic acid-responsive vegetative storage protein expression was primed and thus elevated in response to powdery mildew, suggesting that P. indica confers resistance reminiscent of induced systemic resistance (ISR).
The root endophytic basidiomycete Piriformospora indica belongs to the recently defined order Sebacinales (Weiss et al. 2004). Species of this order form a novel type of mutualistic mycorrhizal symbiosis with a broad spectrum of host plants, such as barley, maize, Arabidopsis, tomato and tobacco (Varma et al. 1999, Shahollari et al. 2005, Deshmukh et al. 2006). In barley, the fungus induces resistance to root diseases and to the powdery mildew disease caused by Blumeria graminis f. sp. hordei (Waller et al. 2005). Plant colonization of P. indica is restricted to the root cortex, suggesting that resistance to leaf pathogens conferred by P. indica requires systemic signals. In Arabidopsis thaliana two mechanisms of systemic disease resistance are well studied: systemic acquired resistance (SAR) induced by necrotizing pathogens and a range of synthetic chemicals; and induced systemic resistance (ISR), induced primarily by non-pathogenic rhizobacteria (Pieterse et al. 1996, Ryals et al. 1996, Kogel and Langen 2005). We used Arabidopsis mutants to investigate whether these defense pathways are involved in P. indica-mediated resistance to powdery mildew caused by Golovinomyces orontii.
Microscopic inspection of roots inoculated with P. indica detected hyphae on the root surface, intercellularly aligned to root cell walls, and, albeit less frequently, intracellularly (Fig. 1A). Hyphae were confined mostly to the rhizodermis, and newly formed chlamydospores could be observed from 10 to 14 days post-infection (dpi) (Fig. 1B) on the root surface and inside rhizodermal cells (Fig. 1C, D). Interestingly, intracellular hyphal growth was not associated with increased autofluorescence of adjacent host cell walls under excitation light (365 nm) indicating that, similar to barley (Deshmukh and Kogel 2007), the fungus does not induce strong defense reactions in Arabidopsis roots.
Colonization of Arabidopsis roots with Piriformospora indica and its effect on the development of the leaf pathogen Golovinomyces orontii. Arabidopsis root sections colonized by P. indica. (A) Fungal hyphae stained with WGA-AF 488 in roots 7 dpi. Intercellular hyphae are closely aligned to rhizodermal cell walls; protruding short hyphal branches are characteristic for intracellular penetration attempts. The image was recorded with a confocal microscope (maximum projection of 12 optical sections). (B–D) Chlamydospore formation in Arabidopsis roots. Staining with 0.01% acid fuchsine–lactic acid visualizes chlamydospores 21 dpi. (B) Bright-field image. (C) and (D) were obtained with a confocal microscope, representing the same root section in the transmission channel (C), and in the fluorescence channel (D) detecting chlamydospores stained with fuchsine–lactic acid (maximum projection of 30 optical sections). New chlamydospores are formed inside rhizodermal cells, on the root surface (B), and in root hairs (C and D). Bars in A, B, C and D represent 25 µm. (E, F) Development of G. orontii on Arabidopsis leaves of control plants (E) and P. indica-colonized plants (F). Fourteen days after root inoculation with P. indica, leaves were inoculated with conidia of the powdery mildew fungus G. orontii. Five days after challenge inoculation, leaves were cleared with ethanol and fungal structures stained with blue ink. E and F show a fraction of a mycelium derived from successful development of a single conidium. Reduced numbers of rod-shaped conidiophores are visible in P. indica-colonized plants as compared with control plants. Bars in E and F represent 100 µm.
To assess the influence of P. indica on plant morphology, root and shoot growth was analyzed. By 14 dpi, the main root length of P. indica-colonized plants was reduced to 86.5% (±2.6 SD) of that of non-colonized plants. The number of lateral roots was 80.5% (±12.6 SD), the average length of lateral roots was 107.6% (±12.1 SD) and the total root length was 84.4% (±6.1 SD). However, only main root length was affected significantly (Students t-test P < 0.05). In contrast to barley, P. indica-colonized Arabidopsis did not show visible differences in shoot development, or an increase in shoot weight in experiments reported here. However, Arabidopsis biomass is increased by P. indica using specific culture conditions (Shahollari et al. 2005), which could be reproduced in several laboratories, including ours.
To test if P. indica is capable of inducing resistance in Arabidopsis to a leaf pathogen, we compared development of the powdery mildew fungus G. orontii on leaves of both wild-type plants and Arabidopsis mutants compromised in jasmonic acid (JA) and salicylic acid (SA) signaling. We found a strong systemic resistance response mediated by P. indica on Arabidopsis (Col-0): the number of conidiophores formed per mycelium was reduced at 5 dpi (compare Fig. 1E and F). The degree of resistance was quantified by determining the amount of G. orontii conidia per mg of leaf fresh weight: by 10 dpi, the number of conidia in P. indica-colonized plants was only 47% of that recorded for non-colonized wild-type plants (Fig. 2). Upon P. indica colonization, NahG plants (which are unable to accumulate SA; Gaffney et al. 1993) and the mutant nonexpressor of PRgenes1-3 (npr1-3; Cao et al. 1994), showed reduced amounts of powdery mildew conidia of 46 and 45%, indicating that the SA-dependent pathway and nuclear localization of NPR1, abolished in npr1-3, are not required for P. indica-mediated resistance (Fig. 2). In contrast, the mutants jasmonate resistant 1-1 (jar1-1), jasmonate insensitive 1 (jin1) and npr1-1 were fully compromised in P. indica-mediated powdery mildew resistance (Fig. 2), indicating that the systemic resistance response was independent of salicylate signaling, but required an operative jasmonate defense pathway. For jar1-1, a reduced sensitivity to JA and methyl jasmonate (MeJA) which results in several defective JA-mediated responses, including ISR, was shown (Staswick et al. 1992, Pieterse et al. 1998, Staswick et al. 1998). In addition, our experiments show the requirement for JIN1 (AtMYC2). This transcriptional regulator activates a subset of JA-regulated genes, specifically wound-induced genes, while other JA-regulated genes are repressed (Lorenzo et al. 2004). Recently, the requirement of AtMYC2 for ISR mediated by Pseudomonas fluorescens WCS417r, which results in priming of JA-regulated genes, was shown (Pozo et al. 2008).
Genotype-dependent enhanced resistance against powdery mildew after colonization with P. indica. Fourteen days after root inoculation with P. indica or mock inoculation (control), leaves were inoculated with conidia of the powdery mildew fungus Golovinomyces orontii. Ten days post-inoculation, leaves were detached, and the amount of conidia per 10 mg of leaf fresh weight determined for at least five individually treated plants. Values are means and were set to 1 for P. indica non-inoculated Col-0 plants to enable comparison of experiments. Open columns depict non-inoculated, filled columns depict P. indica-inoculated plants of the indicated genotypes, with bars indicating standard errors. Similar results were obtained in three independent experiments.
We also checked the extent of root colonization in different defense pathway mutants by determining P. indica DNA relative to plant DNA using quantitative PCR (Q-PCR). In NahG, npr1-1 and npr1-3 plants, we found a significantly (Students t-test P < 0.05) higher relative amount of fungal DNA (53, 55 and 35% increase, respectively, compared with Col-0). In contrast, relative amounts of fungal DNA were not significantly different in jin-1 and jar1-1 (43 and 0% increase). Despite these quantitative differences, a microscopic inspection revealed no qualitative differences in the tested genotypes. Enhanced colonization of npr1-1, npr1-3 and NahG plants could be explained by a defective SA defense pathway that restricts fungal development in wild-type plants. On the contrary, ethylene (ET) signaling was required for P. indica colonization of the roots, as ethylene resistant 1-1 (etr1-1) and ethylene insensitive 2-1 (ein2-1) (Bleecker et al. 1988, Guzman and Ecker 1990) were less colonized. This finding made it impossible to determine experimentally the role of ET signaling for P. indica-induced systemic resistance. It was previously shown that ET is required for some, but not all, bacterial strains inducing ISR (Iavicoli et al. 2003).
To clarify further the induction of signaling pathways, we analyzed expression of SA-, JA- and ET-responsive genes in leaves 14 d after root inoculation with P. indica. JA-responsive vegetative storage protein (VSP), plant defensin 1.2 (PDF1.2) and lipoxygenase 2 (LOX2) mRNA levels were slightly, but not significantly, lower in P. indica-colonized plants, whereas expression of SA-responsive pathogenesis-related 1 (PR1) and PR5, and ET-responsive ethylene response factor 1 (ERF1) were unaffected (Fig. 3). Upon inoculation of leaves with G. orontii, transcripts of PR1 were induced at 3 dpi, and, to a much higher degree at 6 dpi. PR5, ERF1 and PDF1.2 were induced by G. orontii at 6 dpi. In contrast to the other genes, a significant 8-fold higher expression of VSP was detected 3 d after powdery mildew challenge in P. indica-colonized as compared with non-colonized plants (Fig. 3). This result is consistent with the observed loss of P. indica-mediated resistance in jin1, which is defective in JA-induced VSP expression (Berger et al. 1996). Enhanced VSP expression thus supports a role for JA in a primed response associated with powdery mildew resistance induced by P. indica. In the absence of a pathogen challenge, in contrast, mRNA levels of SA-, JA- and ET-responsive genes were indistinguishable between P. indica-colonized and non-colonized plants (Fig. 3). These data are consistent with the notion that ISR is accompanied by weak systemic up- or down-regulation of transcripts without challenge, while a subset of JA-regulated defense genes, such as VSP, are more strongly expressed upon pathogen challenge (Van Wees et al. 1999, Cartieaux et al. 2003, Verhagen et al. 2004).
Expression of hormone-responsive genes in leaves of Arabidopsis plants colonized with P. indica relative to non-colonized plants 0, 3 and 6 d after powdery mildew challenge. The mRNA levels of SA-responsive PR1 (A) and PR5 (B), JA-responsive VSP (C), PDF1.2 (D) and LOX2 (E), and ET-responsive ERF1 (F) were analyzed in leaves at 0, 3 and 6 dpi by quantitative RT–PCR. Expression levels were calculated relative to the constitutively present ubiquitin 5 mRNA. Relative expression values for control plants (open columns) and P. indica-inoculated plants (filled columns) were calibrated to the control 3 dpi time point set to 100. Values shown represent average values from three independent experiments, with error bars depicting standard errors. For PR1b, no error bars are given at time point 0 dpi, as transcript levels were not detectable in two of three experiments.
An induced ISR mechanism is also in accordance with the P. indica response of the two NPR1 mutants: npr1-3 is lacking a functional nuclear localization signal and, as the npr1-3 mutant is not impaired in P. indica-mediated resistance, a nuclear localization of NPR1 is not required. However, npr1-3 has been shown to retain a cytosolic function regulating expression of some JA-dependent genes (Spoel et al. 2003), and we observed that P. indica-induced priming of VSP at 3 dpi is present in npr1-3. This indicates that JA-mediated priming responses induced by P. indica are still intact in the npr1-3 mutant. Consistently, the null mutant npr1-1 is impaired in cytosolic NPR1 function, in ISR and in P. indica-mediated resistance to G. orontii.
Analysis of gene expression in leaves of P. indica-colonized barley showed—similarly to the Arabidopsis results presented here and to ISR—no induction of JA or SA marker genes in the absence of a pathogen challenge (Waller et al. 2005, Waller et al. 2008). Recent results indicated a P. indica-dependent priming of NPR1-regulated genes in barley (Molitor et al. unpublished), suggesting that the fungus triggers common signaling pathways in mono- and dicotyledonous plants (Kogel and Langen 2005).
In conclusion, we suggest that P. indica-induced resistance requires jasmonate signaling, and is associated with priming of JA-regulated defense genes after powdery mildew challenge, while it is independent of salicylate-based mechanisms. Furthermore, our results indicate that the fungus requires only cytosolic but not nuclear localization of NPR1 to induce systemic resistance. The P. indica–Arabidopsis interaction therefore has the potential to become a model system for mechanistic investigations of induced resistance and plant–microbe symbiosis.
Materials and Methods
Arabidopsis thaliana seeds were incubated at 4°C for 48 h. After 14 d growth on agar plates, roots were either mock-inoculated or inoculated with 5 × 105 ml−1P. indica chlamydospores and plants were transferred to pots with sand/potting soil. Piriformospora indica was propagated as described (Waller et al. 2005). Golovinomyces orontii (syn. Erysiphe cichoracearum USC1) inoculum was spray-inoculated at a density of 4–6 conidia mm−2. Microscopy was performed as described in Waller et al. (2005) and Deshmukh et al. (2006).
Quantitative reverse transcription–PCR conditions, oligonucleotide primers and further methodological details are described in the Supplementary material.
Funding
Grant FOR666-B2 by Deutsche Forschungsgemeinschaft (DFG).
Acknowledgments
We are grateful to Yves Marco for providing seeds, and Ralph Panstruga, MPI Köln, for providing Golovinomyces orontii. We thank Magali Bourdeau (Université de Nice Sophia Antipolis) for help in preparing microscopic images.
References
Abbreviations:
- dpi
days post-inoculation
- ERF1
ethylene response factor 1
- ET
ethylene
- ISR
induced systemic resistance
- JA
jasmonic acid
- LOX2
lipoxygenase 2
- MeJA
methyl jasmonate
- PDF1.2
plant defensin 1.2
- PR
pathogenesis related
- Q-PCR
quantitative PCR
- SA
salicylic acid
- SAR
systemic acquired resistance
- VSP
vegetative storage protein.
Author notes
2These authors contributed equally to this work.
