Plants are continuously threatened by a plethora of biotic stresses that include encounters with viruses, bacteria, fungi, nematodes and herbivorous animals. Amongst these threats, diseases caused by fungal pathogens have a particular impact, accounting for major damage and yield losses in agriculture. Basic research on such plant diseases is therefore pivotal to advance our understanding of the molecular mechanisms underlying fungal pathogenesis and plant defence. In the long term, these activities may lead to rational strategies for durable disease control that will result in a reduction in fungicide usage. New Phytologist has a long-standing tradition in the publication of original research articles and reviews that address molecular, physiological, evolutionary and environmental aspects of symbiotic and pathogenic plant–fungus interactions. In this Virtual Special Issue we focus on ‘Pathogenic plant–fungus interactions’ and have compiled a selection of recently published New Phytologist papers that include topics on molecular mechanisms of fungal pathogenesis, plant defence signalling and response, as well as analyses of the plant–fungus interface.
Molecular mechanisms of fungal pathogenesis
Fungal pathogens employ diverse routes to colonize host plants. Some fungal species invade host tissues by direct penetration of the walls of epidermal cells, while others enter through natural openings (e.g. stomata) and subsequently invade internal cell types (e.g. mesophyll cells, vascular tissue). The chemical and physical properties of the surface of the host tissue are known to impinge on the differentiation of extracellular fungal infection structures (Zabka et al., 2008). Pathogenic fungi that enter epidermal cells encounter, before the cell wall, the cuticle as the first barrier for entry into the host cell. Cutin-hydolyzing enzymes (cutinases) thus represent a key component of the arsenal of carbohydrate-degrading enzymes encoded by the genomes of many phytopathogenic fungi. Surprisingly, there is considerable sequence diversity and copy-number variation within the cutinase family in filamentous Ascomycetes, with a particular gene-family expansion present in the rice blast fungus, Magnaporthe grisea, which suggests functional diversification of cutinase isoforms in this pathogen (Skamnioti et al., 2008). Aside from host cell entry, suppression of host defences poses another major challenge for pathogenic fungi. Like other microbes that interact with plants, fungal intruders employ secreted effector proteins to manipulate host cells, including the suppression of plant defence responses (reviewed in Dodds et al., 2009). However, plants have evolved extracellular and intracellular immune sensors that perceive the presence of effector proteins, which typically results in a boosted defence response that is often associated with localized hypersensitive cell death (Dodds et al., 2009). Based on historical grounds, the term ‘avirulence protein’ is often used for these ‘recognized’ effectors, while the matching immune sensors were coined resistance (R) proteins. ACE1 is an avirulence gene of M. grisea, encoding a polyketide synthase (PKS) fused to a nonribosomal peptide synthetase (NRPS), which is probably involved in the biosynthesis of a secondary metabolite that is recognized by the Oryza sativa (rice) Pi33 immune receptor. ACE1 is one member of a cluster of 15 genes that are specifically expressed during an early stage of fungal pathogenesis and of which 14 are believed to be involved in the biosynthesis of secondary metabolites (Collemare et al., 2008). On the plant side of this interaction, microarray-based gene-expression profiling has revealed the inventory of genes that exhibit changes in transcript accumulation during Pi33-conditioned resistance in rice against a M. grisea isolate that carries the ACE1 avirulence gene (Vergne et al., 2007). The Leptosphaeria maculans Lmgpi15 gene encodes a protein that is believed to be involved in the transfer of a glycosylphosphatidylinositol (GPI) anchor to membrane proteins. The Lmgpi15 gene product is required for post-invasive growth of the fungal pathogen, demonstrated by the finding that an Lmgpi15 knockout mutant shows reduced growth and abnormal hyphal morphogenesis following host cell penetration (Remy et al., 2008). An unexpected effect on fungal pathogenesis was observed in a Fusarium graminearum deletion mutant lacking the TRI5 gene, which is involved in biosynthesis of the species-specific mycotoxin, deoxynivalenol (DON). While tri5 mutants have reduced pathogenicity on wheat ears they retain full pathogenicity on Arabidopsis thaliana floral tissue, suggesting that DON biosynthesis is dispensable for colonization of the latter species (Cuzick et al., 2008). In contrast to this differential effect, the fungal pathogen appears to exploit ethylene signalling equally in both Arabidopsis and wheat. Genetic and pharmacological interference with ethylene production/perception in these two plants resulted in enhanced disease resistance, suggesting that ethylene signalling is beneficial for pathogenesis of F. graminearum in both monocotyledonous and dicotyledonous host plants (Chen et al., 2009). The genes EDS1 and SGT1 are well-known components of plant defence against various pathogens, including isolate-specific resistance conditioned by some R genes (Austin et al., 2002; Wiermer et al., 2005). Consistently, a mutation in A. thaliana SGT1b results in enhanced disease susceptibility of floral tissues to Fusarium culmorum (Cuzick et al., 2009). Surprisingly, however, the plant pathogenic fungus Botrytis cinerea requires EDS1 and SGT1 for full pathogenicity, which was demonstrated by the finding that silencing of these two genes in Nicotiana benthamiana led to enhanced resistance against the necrotrophic pathogen. It thus seems as if the fungus hijacks these plant-defence components to cause disease (El Oirdi & Bouarab, 2007). The plant cytoskeleton is known to play an important role during antifungal plant defence (reviewed in Schmidt & Panstruga, 2007). However, components of the cytoskeleton are also crucial for fungal pathogenesis. In a topical Tansley review, GeroSteinberg (2007) highlighted the contribution of microtubules as ‘tracks for traffic’ during invasive growth of the corn smut fungus, Ustilago maydis. For this latter fungus, an elegant method for the deletion of multiple genes, based on sequential rounds of site-specific recombination, was recently reported. This technique enables functional analysis based on reverse genetics, even in complex fungal gene families (Khrunyk et al., 2010). Following successful invasion of host tissues, nutrient retrieval from host sources is pivotal for pathogens for further growth and propagation. In vivo NMR analyses of sunflower cotyledons during infection with B. cinerea revealed uptake of hexoses from the plant host and its rapid conversion into mannitol in the fungal pathogen (Dulermo et al., 2009).
Plant defence signalling and execution
Plants need to protect themselves against fungal attack. The plant’s alarm system relies on membrane-bound and cytoplasmic immune receptors that trigger various defence responses, including transcriptional reprogramming, upon stimulation. Members of the family of WRKY transcription factors are important regulators of biotic and abiotic stress responses. Surprisingly, both overexpression and gene silencing of the poplar WRKY23 gene was found to affect successful defence against the rust fungus, Melampsora medusae, possibly via perturbed redox homeostasis and cell wall metabolism (Levee et al., 2009). The polymorphic RPW8 gene, encoding a resistance protein with an unusual domain structure, is a major determinant of powdery mildew resistance in A. thaliana. To unravel whether other genes may contribute to powdery mildew resistance in Arabidopsis, Göllner and co-workers performed genetic analysis on a wide range of powdery mildew-resistant Arabidopsis accessions. They found that broad-spectrum powdery mildew resistance is predominantly either of polygenic origin or conditioned by RPW8 alleles (Göllner et al., 2008). Orthologues of RPW8 also seem to mediate powdery mildew resistance in Arabidopsis lyrata, a close relative of A. thaliana (Jorgensen & Emerson, 2009). Ectopic activation or overexpression of immune sensors can also lead to disease resistance. A recent example of this is the A. thaliana adr2 mutant, in which activation tagging of a cytoplasmic Toll interleukin receptor (TIR) nucleotide-binding site (NBS) leucine-rich repeat (LRR) protein resulted in constitutive activation of salicylic acid-dependent defence responses. As a consequence, the adr2 mutant is more resistant to a range of biotrophic pathogens, but it also exhibited spreading lesions in the absence of pathogen challenge (Aboul-Soud et al., 2009). The host specificity of barley and wheat powdery mildew pathogens remains a fascinating phenomenon: despite the close phylogenetic relatedness of both hosts and parasites, each of these mildew fungi can only colonize their respective host species. Aghnoum and Niks set out to unravel the genetic basis of this unusual host specificity. They identified barley lines with residual susceptibility to the wheat powdery mildew pathogen and performed crosses between these accessions to further increase susceptibility. These lines, termed SusBgt, were used to characterize the infection process in detail, which revealed that defence responses act at various stages to restrict fungal pathogenesis (Aghnoum & Niks, 2010). The role of hydrogen peroxide (H2O2) accumulation during biotic stress responses remains controversial. To study the effect of H2O2 accumulation in the interaction of a plant (wheat, Triticum aestivum) and a hemibiotrophic fungal pathogen (Septoria tritici), Shetty and co-workers infiltrated either catalase or H2O2 at various time-points during fungal pathogenesis into infected wheat leaves. They observed that catalase-conditioned removal of H2O2 rendered leaves more susceptible, whereas H2O2 application made them more resistant (Shetty et al., 2007). Defence execution in plants is often associated with the secretion of antimicrobial cargo (Kwon et al., 2008). A recent study investigated the composition and potential antimicrobial activity of root exudates during the interaction of barley and the soilborne fungal root pathogen, F. graminearum. The authors detected biosynthesis and accumulation of phenolic acids, which in part were shown to inhibit the germination of Fusarium conidia (Lanoue et al., 2010). These results strengthen the notion that secretory processes are pivotal for successful plant defence. Various additional aspects of the cereal–Fusarium interaction were recently highlighted in an informative review about the fungal head blight disease (Walter et al., 2010). Besides secondary metabolites, antimicrobial peptides (defensins) probably contribute to plant immunity. Work by De Coninck and co-workers recently uncovered the role of the Arabidopsis defensin, PDF1.1, in biotic stress responses and identified its potential function in antifungal plant defence (De Coninck et al., 2010).
Analyses of the plant–fungus interface
The cellular interface between fungi and their host plants serves multiple functions. It is the scene of the battle between the attacker and the defender where biochemical warfare is conducted, and it also is the location of nutrient trading (reviewed in O’Connell & Panstruga, 2006). Despite its importance, comparatively little is known about the molecular details of this site. This is in particular true for plant–fungal interactions that predominantly occur inside plant tissues. The problem can be further compounded by disintegration of the plant tissue, for example as a consequence of fungal damage. A recently reported procedure based on cryo-scanning electron microscopy permits three-dimensional in situ visualization of fungal invasion, even in partially damaged plant tissues, so opening up new opportunities for imaging fungal pathogens inside their hosts (Refshauge et al., 2006). Multivesicular bodies are cellular structures containing suites of membrane-bound vesicles. These bodies participate in endomembrane trafficking and in the delivery of cargo to the extracellular space. Qianli An and co-workers used transmission electron microscopy, in combination with cytochemistry, to study the formation and distribution of such multivesicular bodies during compatible and incompatible barley (Hordeum vulgare)–powdery mildew (Blumeria graminis f.sp. hordei) interactions. Their results suggest a contribution of multivesicular bodies in the delivery of building blocks for local cell wall reinforcements and the blocking of plasmodesmata, possibly to contain cell death during the hypersensitive response (An et al., 2006). Nep1-like proteins (NLPs) are produced by an array of unrelated microorganisms and are toxic for dicotyledonous plants. To unravel the molecular basis of its phytotoxicity, Schouten and colleagues heterologously expressed two B. cinerea NLPs and studied their activity upon delivery to protoplasts and cells in suspension. They discovered an association of both fluorophore-tagged proteins with the plasma membrane, the nuclear envelope and the nucleolus. The authors further characterized the associated cell-death process as necrosis or apoptosis, or intermediate forms of both (Schouten et al., 2008).
The array of studies devoted to various aspects of pathogenic plant–fungus interactions recently published in New Phytologist highlight the striking biological diversity of these encounters. Although common mechanistic principles are recognizable, each interaction harbours its own fascinating and surprising molecular details, which are critical for the outcomes of these interactions (‘health’ or ‘disease’). From a strategic perspective, these discoveries justify the ongoing research activities with regard to future food security. However, these insights also represent part of the excitement for those devoted to basic research. With this in mind we, the journal New Phytologist, look forward enthusiastically to new findings and publications in the area of pathogenic plant–fungus interactions.
I thank Anne Osbourn and Holly Slater for helpful suggestions and critical proofreading of the manuscript.