the path in fungal plant pathogenicity; many opportunities

THE PATH IN FUNGAL PLANT PATHOGENICITY; MANY OPPORTUNITIES TO OUTWIT THE INTRUDERS? Guus Bakkeren 1 and Scott Gold 2 1 Agriculture & Agri-Food Canada Pacific Agri-Food Research Centre Summerland, BC, Canada V0H 1Z0 2Department of Plant Pathology University of Georgia Athens, GA 30602-7274.

1 Corresponding author: G. Bakkeren, Phone: (250) 494-6368, E-mail: [email protected]

SAMPLE GALLY PROOFS Not final copy

Genetic Engineering, Volume 26, Edited by J. K. Setlow Kluwer Academic / Plenum Publishers, 2004 (pp. 175 - 223 in real print)

mailto:[email protected]

G. BAKKEREN AND S. GOLD

Glossary of the main phytopathogenic fungi discussed: • Alternaria alternata many forma speciales (f. sp.), causing leaf spots and blights on many

parts of many plants. • Blumeria graminis f. sp. hordei (syn. Erysiphe graminis), powdery mildew on barley

(Hordeum vulgare). • Botrytis cinerea (teleomorph Botryotinia fuckeliana), grey mold and blights on many plants. • Cercospora species, leaf spot diseases on many plants such as on soybean (C. kikuchii) and

on tobacco (C. nicotianae). • Cladosporium fulvum, leaf mold on tomato (Lycopersicon esculentum). • Claviceps purpurea, ergot of rye (Secale cereale) and other temperate cereals and grasses. • Cochliobolus heterostrophus, southern corn leaf blight disease; C. sativus, spot blotch on

barley; C. carbonum, leaf spot on maize, C. victoriae, Victoria blight in oats. • Colletotrichum gloeosporioides (teleomorph Glomerella cingulata), anthracnose of

avocado, papaya, etc.; C. lagenarium, anthracnose on cucurbit. • Cryphonectria parasitica, chestnut blight. • Fusarium graminearum (teleomorph Gibberella zeae), scab/head blight of barley, wheat

and corn; F. oxysporum f. sp. lycopersici, tomato vascular wilt; F. solani f.sp. pisi (teleomorph Nectria haematococca), pea foot rot.

• Gaeumannomyces graminis, ‘take all’, root disease of temperate cereals and grasses. • Leptosphaeria maculans, blackleg disease of oilseed rape (canola; Brassica napus). • Magnaporthe grisea (anamorph Pyricularia grisea), blast disease on rice (barley, wheat,

millet, etc). • Mycosphaerella graminicola (anamorph Septoria tritici), wheat blotch. • Nectria haematococca (anamorph Fusarium solani), root rot of pea; N. radicicola, ginseng

root rot. • Ophiostoma ulmi and O. novo-ulmi, Dutch elm disease. • Pyrenopeziza brassicae, light leaf spot disease of oilseed rape (B. napus) and other

brassicaceae. • Puccinia graminis f.sp. tritici, wheat stem or stripe rust; P. triticina (formerly P. recondita

f.sp. tritici), wheat leaf or brown rust. • Rhynchosporium secalis, leaf scald on barley, rye and other grasses. • Sclerotinia sclerotiorum, cottony, stem or soft rots (white mold) on many plants and fruits. • Septoria lycopersici, tomato leaf spot; S. tritici (teleomorph Mycosphaerella graminicola),

wheat leaf blotch. • Uromyces fabae, bean rust (Vicia faba); U. vignae, cowpea rust. • Ustilago maydis, boil smut of corn (Zea mays); U. hordei, covered smut of barley and oats. SUMMARY The number of genes implicated in the infection and disease processes of phytopathogenic fungi is increasing rapidly. Forward genetic approaches have identified mutated genes that affect pathogenicity, host range, virulence and general fitness. Likewise, candidate gene approaches have been used to identify genes of interest based on homology and recently through ‘comparative genomic approaches’ through analysis of large EST databases and whole genome sequences. It is becoming clear that many genes of the fungal genome will be involved in the pathogen-host interaction in its broadest sense, affecting pathogenicity and

176


the disease process in planta. By utilizing the information obtained through these studies, plants may be bred or engineered for effective disease resistance. That is, by trying to disable pathogens by hitting them where it counts. INTRODUCTION In this review we focus on the current understanding of fungal plant pathogenesis at the molecular level. We also briefly discuss realized and potential disease control strategies based on this understanding. Plant pathologists interested in fungi have traditionally focused on microorganisms that actually encompass two kingdoms, including the Protoctistan fungus-like Oomycetes including the major plant pathogenic organisms such as Phytophthora, Pythium, and Peronospora species (“water molds”, the downy mildews and white rusts), as well as the Eumycotan fungi, on which this review will exclusively focus. A recent review addresses similarities in ‘weaponry’ to attack plants between oomycetes and fungi (1). It is undeniable that during the evolution of multicellular organisms (both animals and plants) both pro- and eukaryotic microorganisms must have had ample opportunities to interact with and try to take advantage of, that is, actively weaken, kill, or live ‘in harmony’ with higher life forms. However, it should be reiterated that given this enormous variety of microorganisms, which includes at least 1.5 million fungal species, the outcome “disease” is rather rare in nature with less than 1% of the fungi being pathogenic; otherwise life as we know it would not exist. The few microorganisms that are successful must have evolved the necessary means to recognize susceptible targets, breach physical barriers of hosts and counter and suppress innate and induced defense mechanisms. Pathogenic microbes wage fierce battles with their hosts whose defenses are equally ruthless; this battle is mainly fought by complex molecular interactions and has been shaped over millions of years (2). Initial combinations of pathogenic microbes and their hosts that were successful in both completing their life cycles reached an equilibrium in their respective populations. This was (and still is) a dynamic process involving co-evolution of both host and pathogen. This must have led to ‘boom-bust cycles’ that shaped many populations (3). In agriculture such cycles have been sped up over the last centuries because of selection and breeding practices (2). There are “costs” to the pathogen entering and existing inside a host and conversely, to the host for mounting a defense. These are therefore activities requiring genes induced only when needed. Successful pathogens recognize plants as hosts, that is, each pathogen has the necessary “armor” to establish contact, gain entry and colonize its host plant to cause disease symptoms; the classical definition of a pathogen. To do this, a pathogen must minimize detection by the plant as non-self, avoiding the host’s surveillance mechanisms which when triggered cause the mounting of defenses. In addition, when an eventual plant defense response has been mounted, the pathogen needs to sufficiently counteract and neutralize the hostile environment inside a host. Some strains of a given pathogen species will be better equipped to deal with this. Such a degree of disease caused by the pathogen species on a host population has been defined as virulence and depends on the effects of many, often additive genes. More and more genes are implicated in the interaction between pathogen and host. They range from genes that, when mutated produce non-pathogenic ‘pathogens’ or bring about changes in host range (pathogenicity genes), have an effect on the degree of disease (virulence genes), or even affect the pathogen’s metabolism when in planta (fitness genes). The distinction between such categories of genes is becoming blurred. Judging from recent literature, we are wandering into a semantic quagmire, underlining the complexities of plant-microbe interactions. For example,

177


pathogen mutants found because of their effect on the host, could be affected in growth in vitro on different media and proven to be defective in metabolic or ‘fitness’ genes. However, such effects cannot easily be distinguished in obligate pathogens. Several other recent reviews describe the role fungal pathogenicity genes in the disease process (4, 5). Special issues with obligate pathogens Obligate biotrophs, such as the rust fungi and downy and powdery mildews, develop an intimate and subtle interaction with their host plant because in nature they are incapable of growth on substrates other than living host cells. They produce a common specialized feeding structure within plant cells, the haustorium, which is a nutrient sink and redirects metabolites from the host as has recently been shown for the bean rust system, Uromyces fabae- Vicia faba (6, 7, 8, 9). In doing so, they cause minimal damage and cell death as opposed to ‘necrotrophs’ that kill cells for their contents. Gradations between these ‘extremes’ are found among the hemibiotrophs which are initially benign and grow more damaging by producing toxins as the disease progresses. Obligate biotrophs must have means to avoid recognition and to suppress host defenses over the lifespan of the plant (10). Genes involved in these processes will therefore be pathogenicity genes. There is another class of hemibiotrophs that have different morphologies with different dependencies, such as hemibasidiomycete members of the smut and bunt fungal genera Ustilago and Tilletia. These have haploid cells that grow by budding, are saprobes and can therefore be cultured and manipulated in the laboratory, but when mated become filamentous and are obligate parasitic (11, 12). There are special issues with regard to these fungi; their requirement for mating which induces the morphological switch between haploid growth by budding and the production of the obligate pathogenic, dikaryotic hyphae, implies that all genes involved in this process are by definition pathogenicity genes. We will therefore not discuss these genes here and refer to a large, rapidly expanding body of literature. However, because the haploid yeast-like stage can be manipulated, these fungi are uniquely suitable to the study of genes involved in the biotrophic stage. For example, by manipulating the mating-type genes and engineering of haploid solopathogenic strains, difficulties associated with the effectively diploid genetics of the dikaryon can be bypassed (13, 14). It will be very challenging to find pathogenicity and virulence genes involved in the biotrophic phase. In some systems, genes are beginning to be described, sped up with the generation of genomic resources such as EST databases and whole genome sequencing and their mining by bioinformatics. Such is the case in the bean rust system mentioned above (15, 16, 17, 6, 7), the wheat leaf rust, Puccinia triticina (18, Hu and Bakkeren, manuscript in preparation) and the barley powdery mildew, Blumeria graminis (19, 20, 21). However, verifying their function and role in disease will be even more challenging. A major bottleneck is the difficulty of culturing and manipulating true obligate biotrophs, although some attempts have shown promise (22, 23, 24, 25, 26). In order to make significant progress toward understanding these very important plant pathogens, research efforts need to be increased to establish molecular manipulation techniques and/or develop functional heterologous systems (27). RECOGNITION Fungal spores are deposited passively on plant surfaces by wind, rain, insects, soil-born organisms, or encountered by expanding roots. Additionally, functioning as survival structures,

178


fungal spores may be deposited on the still healthy seeds of future hosts for overwintering, thereby being ideally positioned for subsequent colonization. Under the proper environmental conditions these spores could germinate. However, it is becoming clear that fungal spores need several physical or chemical cues and often contact with a suitable host before they risk their precious genetic material by germinating (reviewed by 28, 29). There is extensive signaling via physical plant surface characteristics and low-molecular weight compounds, such as sugars, amino acids and inorganic salts to break dormancy of conidiospores of saprobes (29). For plant pathogens such as M. grisea and Blumeria graminis, there is proof that surface hydrophobicity plays a role in triggering germination. Inductive substrata often stimulate the release of a proteinaceous extracellular matrix (30, 31) which can facilitate the uptake of small compounds prior to germ tube emergence (32). Even though this matrix is also produced on artificial inductive substrata (and spores can be triggered to germinate on them), the uptake of compounds was not seen before primary germ tube formation. Thus, the spore can receive host-specific cues, possibly anionic cutin monomers, even before germination (32). For the postharvest fruit pathogen, C. gloeosporioides, ethylene produced during ripening and host specific cuticle wax components such as some long-chain fatty alcohols were found to stimulate spore germination (and induce appressorium formation, see below). Interestingly, nonhosts seemed to exude inhibitors of germination (28). Flavonoids have been shown to stimulate the germination of Fusarium solani spores (33) as well as induce several specific pathogenicity genes (34). Urediospores from Puccinia species germinate readily on water surfaces when nonanol is added. The role of Ca2+ signaling in the stimulation of spore germination is controversial; calcium chelators do not affect germination of spores of several plant pathogens but do so for other fungi (reviewed in 29). Even though these processes seem prime targets for interference, their genetic basis has yet to be elucidated. Adhesion to the host surface seems to be an essential step in the disease process to guide germ tube growth and anchoring them to the surface to allow for the enormous forces applied when penetrating. In M. grisea, MPG1, a gene encoding a cysteine-rich hydrophobic protein, is possibly involved in adhesion of appressoria to the rice leaf (see below). Disruption of this gene caused fewer appressoria to form and hence resulted in reduced pathogenicity on rice (35). Recent experimentation has implicated so-called ‘adhesins’, RGD (arginine-glycine-aspartic acid)-tripeptide-containing, extra-cellular matrix (ECM) proteins, which can be recognized by integrins on host cells, in the recognition and interactions between fungal and host cells (36, 37). This has been definitively proven for a handful of fungal proteins in which adhesion to ECM substrates is inhibited by RGD peptides or antibodies against RGD sequences (38). It was recently shown for the cowpea rust Uromyces vignae that such a mechanism is likely involved in its interaction with host and nonhost plants and plays a role in inducing and suppressing cell wall-associated host defense responses (36; see below). GAINING ENTRANCE Softening up the target Plant cell walls consist of a scaffold of cellulose (polysaccharide) microfibrils coated with (and perhaps linked together by) xyloglucans and other hemicelluloses and embedded in pectic polysaccharides. This so-called "type I" wall contrasts with "type II" walls (found in grasses and related species) where the cell walls have relatively low amounts of xyloglucan and pectin but a high proportion of arabinoxylans (39, 40, 41). On the plant-air interface an

179


additional layer, the epidermis consists of complex, long-chain fatty-acid waxes embedded in the biopolymer cutin. Fungi need to breach these physical barriers by enzymatic degradation, brute force or both. The role of cell wall degrading, extracellular lytic enzymes, such as cutinases, pectinases, cellulases and proteases in pathogenicity is still controversial. In general, most pathogenic fungi produce an array of lytic enzymes, often upon contact with host plants, which may contribute to penetration and colonization. For many such fungi, such enzymatic activities have been described in detail but a critical, comprehensive analysis of their role in pathogenicity or virulence through addition or deletion of their genes is missing. For example, a very interesting global analysis of such enzyme activities was presented for the Ustilago maydis-corn interaction and it was shown that different enzyme activities were induced depending on the life cycle stage of the fungus within the tissues targeted but the corresponding genes were not identified (42). Upon close scrutiny of the recently released complete genome sequence of this fungus, such studies could potentially produce a comprehensive understanding of the role of these enzymes in the disease process. Further studies to establish the role of lytic enzymes in U. maydis pathogenicity are necessary. In our laboratory we have identified a gene with high similarity to endo-1,4 β-D-xylanases as upregulated 23-fold in budding haploid cells in comparison with a constitutively filamentous uac1 mutant. Interestingly, preliminary results showed that this gene is also highly expressed in galls (García-Pedrajas and Gold, unpublished). We are further investigating the role of this putative xylanase in pathogenicity in U. maydis.

Certain combinations of lytic enzymes will be most effective depending on the pathosystem and parameters such as mode of penetration and different target tissues. However, because of the sheer variety of lytic activities these fungi are capable of producing, it is likely that disruption of one or even a few of the genes coding for such enzymes will not have a measurable effect on virulence or even pathogenicity. Such genes might be part of a family with other members compensating the loss of one gene, or different lytic enzymes may produce redundant or substituting activities. A role in virulence might be deduced from gene induction and increased production of certain enzyme activities when fungi are grown on complex carbon sources such as xylan, pectin, or purified plant substrates. However, many of these lytic enzymes are also active in ‘saprobic’, in vitro conditions, making distinguishing between functions involved in virulence and in regular fungal metabolism even more difficult. In addition, the production of plant cell wall-degrading enzymes in plant pathogenic fungi is repressed by catabolites such as glucose. From the maize pathogen, Cochliobolus carbonum, from which many cell wall-degrading enzyme genes have been cloned (43), a gene ccSNF1 was cloned. This is a homologue of the yeast SNF1 gene which is required for expression of catabolite-repressed genes when glucose is limiting in this fungus. Together with another C. carbonum gene implicated in glucose repression, CREA, it was shown that these genes are involved in the regulation of transcription of several of the lytic enzyme genes and are required for biochemical processes important in virulence on maize such as the ability to degrade polymers of the plant cell wall (44, 45). Consistent with the above example for C. carbonum, in Fusarium oxysporum a SNF1 homolog, fosnf1, was isolated and mutants shown to have reduced expression of several cell wall-degrading enzyme encoding genes. Compared with wild-type infection, wilt symptoms on Arabidopsis thaliana and Brassica oleracea were considerably delayed with fosnf1 mutants (46).

A controversial role in pathogenicity is exemplified by the cutinases which hydrolyze the major component of plant cuticles (47). A single-copy cutinase gene, Pcb1, from Pyrenopeziza brassicae was found to be essential for pathogenicity. This fungus which causes light leaf spot disease of oilseed rape and other brassicaceae, infects its host by direct cuticle

180


penetration possibly explaining the requirement of this gene (48). Disruption of a cutinase gene, CutA, in Fusarium solani f.sp. pisi had a strong effect on virulence on its host, pea (49). However, disruption of the cutinase A gene of Botrytis cinerea did not have an effect on pathogenicity towards tomato and gerbera (50). Similarly, the disruption of a M. grisea cutinase gene, CUT1, did not have an influence on pathogenicity on rice leaves which strengthened the idea that this fungus penetrates mainly by mechanical force (51). Several fungi have evolved an anticipatory feedback system in which plant cutin monomers as breakdown products produced by spores carrying a small amount of cutinase, can induce fungal cutinase genes through a signal transduction pathway involving transcription factors (reviewed in 28). Cellulases, a large group of enzymes such as cellobiohydrolases, xyloglucan hydrolases and endoglucanases that can catalyze the hydrolysis of beta-(1,4) glycosidic bonds, are produced by many organisms including fungi. The breakdown products are used as nutrients and cellulose induces many of the genes coding for these enzymes, which are under catabolite repression. Because of these saprobic features, the role of these enzymes in pathogenicity is doubtful. The disruption of two cellulase genes from Cochliobolus heterostrophus, EG6 and CBH7, had no effect on virulence even though they were induced strongly during invasive growth in maize leaves (52) and the same was found when disrupting a cellobiohydrolase gene in C. carbonum, CEL1. This mutant, however, still produced cellulase activities in culture (53). A cellulase gene, egl1, has been reported for U. maydis. Expression of egl1 was induced only in the filamentous pathogenic form after mating. In addition, the cellulase was produced mainly at the hyphal tip and was secreted (54), possibly induced by apical meristem tissue (42). Nevertheless, a deletion of this gene did not have any effect on virulence. It is thought that xylanases might play important roles in pathogenicity particularly of cereal-infecting fungi because arabinoxylan accounts for up to 40% of the primary walls in graminaceous crops and grasses (40). However, a role for fungal xylanases in pathogenicity has not been extensively studied, possibly because of the abundance of activities and presence of gene families in many organisms (55). An engineered triple mutant of the maize pathogen C. carbonum in which three xylanase genes were deleted, was still fully pathogenic on maize with regard to lesion size, morphology and rate of lesion development (56). Similarly, a double mutant of M. grisea was not affected in virulence indicating that neither XYL1 nor XYL2 is required for infection on rice (57) even though it was earlier shown to secrete a phytotoxic activity in culture, likely an endoxylanase that killed rice suspension cell cultures (58). Targeted inactivation of xyl4 and of xyl3, xylanase-coding genes of Fusarium oxysporum f. sp. lycopersici, had no detectable effect on virulence on tomato plants, demonstrating that both genes are not essential for pathogenicity (59). In all mutants mentioned, additional, and in some cases residual xylanase activities were found, suggesting extreme redundancy and making an assessment of their role in pathogenicity difficult. Pectin or polygalacturonic acid is a major component of plant cell walls and middle lamellae. Pectin can be degraded by enzymes such as pectate lyase (PL), pectin methylesterase (Pme) and endo- or exo-polygalacturonase (PG). These enzymes are therefore likely attributes in the arsenal of phytopathogenic fungi. Indeed, the disruption of two genes in Nectria haematococca, pelA plus pelD coding for pectate lyases, drastically reduced infection of pea (60). When the B. cinerea pectin methylesterase gene, Bcpme1, was disrupted, virulence was reduced on several of its host plants (61). Similarly, the disruption of the two endoPG genes, cppg1 and cppg2, in Claviceps purpurea nearly eliminates its pathogenic ability on rye (62) and its homologue in B. cinerea, Bcpg1, is required for full virulence on tomato leaves and fruits (63). Interestingly, the endoPG gene, Bcpg1, from B. cinerea apparently can activate grapevine

181


defense reactions unrelated to its enzymatic activity. That is, defense responses can be activated independently or in addition to the known elicitor action of oligogalacturonides (see also below; 64). An interesting set of experiments was done on two closely related pathogens, Alternaria citri which causes citrus black rot by macerating tissues, and Alternaria alternata, which causes brown rot on various hosts by secreting host-specific toxins. The genes for endoPG were deleted and while this enzymatic function proved essential for the macerating pathogen, the mutant did not affect pathogenicity of the brown rot pathogen, illustrating that the need for certain enzymes depends on the mode of infection (65). A gene from Aspergillus flavus, pecA, which produces polygalacturonase P2c, increased invasion and spread in cotton bolls when introduced into a P2c non-producing isolate, and reduced virulence when deleted (66). Several endoPG genes from F. oxysporum f.sp. lycopersici, such as pg1 and pg5, are expressed during early stages of infection and induced when strains sense plant compounds, but their contribution to virulence seems minimal or difficult to assess, probably because of the existence of a PG gene family (67, 68). Similarly, the disruption of a major extracellular endoPG gene of the chestnut blight fungus, Cryphonectria parasitica, did not reduce virulence significantly (69). The maize pathogen, C. carbonum, possesses a large battery of pectic (and other different wall degrading) enzyme genes so it must have come as no surprise that the deletion of some of its endo- and exo-PG genes had only a marginal effect on virulence and did not turn out to be essential for pathogenicity on maize (70, 43). Many fungi seem to be able to produce many PGs and other different wall degrading enzymes that can compensate when several are missing. Although no mutant phenotypes were tested, analysis of two endoPG genes from the bean pathogen, Colletotrichum lindemuthianum, clpg1 and clpg2, were very suggestive for their role in virulence. Although both could be induced in vitro when grown on pectin-containing media, clpg1 was mainly expressed under saprophytic growth conditions and at the beginning of the necrotrophic stage of infection whereas clpg2 showed expression at the onset of the infection (in germinating conidia, appressoria and early penetrating hyphae). The promoter region of clpg2 seemed to harbor elements involved in regulation by pectin and host cues (71, 72). There is some indication that proteases could be involved in the infection process of pathogenic fungi. The bean rust fungus, Uromyces fabae, was shown to produce many extracellular metallo (Ca2+-containing) proteases that could degrade fibrous hydroxyproline-rich proteins known to be structural components of plant cell walls (73). Stagonospora (Septoria) nodorum when grown in liquid culture with wheat cell walls as the sole carbon and nitrogen source, secretes numerous extracellular depolymerases, including a rapidly produced, alkaline trypsin-like protease, SNP1, which participates in the degradation of host cell walls during infection (74). However, deletion of one of at least two aspartic protease genes from B. cinerea, BcAP1, did not result in any discernable loss of virulence on detached tomato leaves (75). In addition to excreting lytic enzymes, pathogenic fungi have probably adapted to optimize infection in another way. It has been shown that several Colletotrichum species are able to locally modulate the pH in the penetration site by secreting ammonia to allow for optimal activity of its pectate lyases (76, 77). Similarly, Sclerotinia sclerotiorum acidifies its ambient environment by producing oxalic acid during plant infection. This has been implicated as a primary determinant of pathogenicity for several fungi. Sclerotial development was favoured by acidic ambient pH conditions and transcripts encoding the endoPG gene pg1 accumulated maximally under acidic culture conditions (78, 79). Based on sequence similarity to A. nidulans PacC, pac1 of S. sclerotiorum may regulate pH-dependent activities required for the appropriate regulation of physiological processes including endoPG gene transcription and therefore full virulence (80). It is becoming clear that the ability of the pathogen to actively

182


increase or decrease its surrounding pH allows it to select the specific virulence factor, out of its vast arsenal, to best fit a particular host (81). The importance of pectic enzymes as virulence factors depends on the pathosystem and mode of infection. This has been demonstrated by experiments in several pathosystems where PG-inhibitor proteins (PGIPs) from plants (sometimes produced experimentally in transgenic hosts) reduce pathogenic effects (82, 83). For example, heterologous expression of pear fruit PGIP in tomato has demonstrated that these inhibit fungal endoPGs and slow the expansion of B. cinerea disease lesions and the associated tissue maceration (84). Even though these kinds of experiments are encouraging, because of the redundancy of many of the lytic enzymes, it is unlikely that targeting their activities will be effective alone in designing novel methods of plant protection. Penetration: bringing in the battering ram To gain entry into host tissues, many fungi produce as another option or in addition to the lytic enzymes discussed above, specialized structures that can exert enormous pressure to gain entry through the epidermis (85) or into cavities covered by stomatal cells such as for the wheat leaf rust, Puccinia triticina (86). Apart from the chemical signals they may perceive as discussed above, they also seem to cue in to physical features of the host surface through thigmotropic sensing and many fungi can be triggered to develop appressoria on artificial substrates (reviewed in 87). Rearrangements in the fungal cytoskeleton and the possible involvement of a channel transducing the membrane stress induced by the leaf topography into an influx of ions, including Ca2+, could trigger differentiation (reviewed in 4). This component of host recognition remains elusive and although many genes could impinge on these processes, not many have been described. However, the M. grisea Pth11 gene encoding a novel plasma membrane protein, has been implicated in proper appressorium development since mutants in this gene fail to correctly respond to inductive substrates (88). Many of the pathogenicity mutants found to date are affected in proper appressorium differentiation. Some of these mutations are in genes in signal transduction pathways such as mitogen activated (MAP) and cAMP dependent protein kinase pathways (discussed below). A novel gene, CBP1, encoding a putative extracellular chitin-binding protein, may play an important role in the hydrophobic surface sensing of M. grisea during appressorium differentiation (89). In a similar category, an interesting M. grisea non-pathogenic mutant, ‘Punchless’, was recently described. This mutant developed a normal appressorium on rice leaves but was unable to initiate a penetration peg. The corresponding gene, PLS1, encodes a tetraspanin-like protein, possibly a component of a membrane signaling complex controlling cell differentiation and motility (90). Several other morphological mutants, such as SMO1 and CON1 that affect proper appressorium development have been described but the corresponding genes and functions are unknown (reviewed in 31). In this fungus the enormous turgor pressure needed in an appressorium to breach the cell wall was found to be generated by high levels of glycerol (91) predicting that several of its metabolic genes will be pathogenicity factors. Also in this fungus, proper metabolism of trehalose, a disaccharide that is commonly found as a storage carbohydrate in eukaryotic cells and is implicated in the response to various environmental stresses, was found to be important during spore germination and appressorium formation (92). Similarly, phospholipid metabolism was implicated in the development of penetration hyphae through study of a M. grisea pde1-insertion mutant (93). The involvement of fatty-acid metabolism in penetration through appressoria was also suggested by studies on the cucurbit

183


anthracnose pathogen, Colletotrichum lagenarium. A pathogenicity mutant in a PEX6 homologue was shown to interfere with proper targeting of certain proteins into the peroxisome thereby affecting several functions thought to be essential for penetration through appressoria and possibly life inside host tissues, such as melanin production and fatty acid beta-oxidation (94). A gene with unknown function, Cap20 from Colletotrichum gloeosporioides, is expressed exclusively during appressorium formation induced by host signals and is required for the formation of appressoria and penetration in avocado and tomato fruits (95). It seems that many genes necessary for proper appressorium development described to date are involved in metabolism leading to the very specialized and exceptional physiological state of these differentiated cell structures. Given this essential stage in the pathogenic development of these fungi, these processes seem very attractive as targets for the design of novel crop protection methods. FUNGAL WALLS AND SURFACES Chitin and glycoproteins Chitin, a beta-1,4-linked polysaccharide of N-acetylglucosamine, is a major structural component of fungal cell walls and chitin synthesis is a process maintained across the fungal kingdom (96). The enzymes responsible for their production are therefore likely essential and not pathogenicity factors in sensu stricto. However, from an insertional mutagenesis screen for pathogenicity mutants, ChsV was cloned showing that F. oxysporum requires a specific class V chitin synthase for pathogenesis. The mutant showed cell wall abnormalities and growth impairment and, interestingly, was also hypersensitive to plant antimicrobial defense compounds such as the tomato phytoanticipin and peroxide. It was suggested that even though the mutant lacked vigour, a cell wall alteration most probably resulted in an inability to protect itself against plant defense mechanisms (97). Similarly, a class I chitin synthase gene, Bcchs1 from B. cinerea contributed to approximately 30% of the wall chitin content and when this gene was disrupted, the mutant was more sensitive to glucanases and exhibited reduced pathogenicity on vine leaves (98). Yet, fungi have multiple classes of chitin synthases that catalyze N-acetylglucosamine polymerization. Of the six chitin synthase genes in U. maydis, only two have been implicated in pathogenicity and their deletion results in less virulent mutants (reviewed in 12). Conceivably, even though there is an obvious redundancy, there might be differential expression at different stages during the life cycle. Cell wall glycoproteins have also been shown to play a protective role against host defenses. The tobacco vascular wilt, Fusarium oxysporum f. sp. nicotianae, is sensitive to the host pathogenesis-related protein (PR-5), osmotin. By overexpressing a heterologous cell wall glycoprotein from Saccharomyces cerevisiae known to protect this fungus from the toxic action of osmotin, in F. oxysporum, disease severity and fungal growth on tobacco were increased. This shows that these fungal cell wall components can increase resistance to plant defense proteins and affect virulence (99). Hydrophobins Fungal hydrophobins have been known for many years. They are usually rather small proteins with eight cysteine residues in a conserved array which contribute to the folding in particular conformations once secreted. They are found, presumably aggregated on the surfaces

184


of cell walls of hyphae and conidia (reviewed in 100, 101). In some of the mushrooms they are thought to create a hydrophobic mantle around hyphae or fruiting bodies helping them with emergence out of watery environments, or in general to mediate interactions between the fungus and the environment. Such “coating” on cells and hyphae could play a role in entering and surviving in host environments through protecting against host enzymes such as chitinases or as “stealth factors” by preventing potential “antigenic” elicitor determinants from being recognized by the host and thereby avoiding eliciting defense reactions. Possible roles for hydrophobins in pathogenicity have been suggested as early as 1994 (102, 103). MPG1 is a small, secreted, cysteine-rich hydrophobin in M. grisea that is abundant during appressorium formation and in early plant infection. When deleted, pathogenicity was severely reduced due to impaired appressorium formation (35). Cerato-ulmin (CU) is a hydrophobin of the Dutch elm disease-causing fungus, Ophiostoma ulmi and O. novo-ulmi. However, molecular-genetic experimentation could not establish a role in pathogenicity but did not exclude a role as a subtle “fitness factor” (104, 105). Cryparin is a hydrophobin with lectin-like properties secreted by the chestnut blight fungus, Cryphonectria parasitica. It binds fungal surfaces and the gene encoding it seems to be regulated via G-protein signaling. G-alpha mutants had pleiotrophic effects including reduced pigmentation and growth, abolishing virulence and reproduction, and decreased cryparin gene transcription. However, the contribution of cryparin to virulence was not established (106). Many fungi analyzed to date have several hydrophobin genes and redundancy could pose problems in proper assessment of a critical role in pathogenesis. For example, the tomato mold Cladosporium fulvum has six hydrophobin genes and the deletion of two of them did not have any effect on pathogenicity (107, 101). Hydrophobins fall into a large group of so-called ‘cysteine-knotted proteins’ which includes a distinct subgroup of pathogen-produced necrosis factors and elicitors of the host defense some of which are avirulence gene products (see below). However, hydrophobins do not seem to act as elicitors of host defense. Melanins Melanins are complex black polymers thought to protect organisms from physical, biological and environmental stress, such as protection from UV radiation and scavenging of “anti-oxidants”, free radicals and other harmful compounds (e.g., microbicidal peptides), providing cell wall strength and preventing leakage of certain metabolites. It protects pathogens inside hostile host environments and the importance of melanin biosynthesis and its regulation in pathogenicity and virulence has been documented for both mammalian and plant pathogens such as M. grisea and C. lagenarium (108, 109, 110, 111, 112, 113, 114, 115). For example, melanins play a role in generating differences in wall permeability by sealing off walls to allow for turgor pressure build-up in appressoria. This has been well-studied in the rice blast M. grisea, where albino, rosy or buff-colored mutants lacking melanin pigment, produce non-functional appressoria (reviewed in 31). The effective fungicide Tricyclazole has been shown to inhibit noncompetitively the active trihydroxynaphthalene reductase which is essential for the biosynthesis of melanin in M. grisea (116). Disruption of the THR1 gene in C. lagenarium generated mutants that produced non-melanized appressoria with very little infectivity (117). Laccases catalyze the single-electron oxidation of phenols or aromatic amines to form different products including melanin and laccase. These genes have been shown to be involved in the biosynthesis and polymerization of melanins in fungi (118 and references therein). The possible roles of melanin and the enzymes involved in its production, in pathogenicity and virulence have also been investigated in some other fungi, such as in the soil-borne, wheat and

185


rice root-infecting fungi, Gaeumannomyces graminis var. tritici and var. graminis. These fungi produce hyphopodia, appressoria-like structures that develop from vegetative hyphae. The involvement of laccase genes in the regulation of melanin production was suggested (119), but melanized hyphopodia or hyphae were not required for plant infection (120). Similarly, some fungi such as Alternaria alternata, produce melanin in their appressoria though it is not required for penetration of their host since melanin-deficient mutants are still pathogenic (121). Evidence indicates that melanin production is an effective survival (and therefore virulence) factor for Cryptococcus neoformans, serving multiple roles in protecting this medically important fungus from host defenses; the ability of melanin to bind proteins and sequester microbicidal peptides thereby abrogating their activity, suggests a protective function (122). The importance of melanin for virulence in other medical pathosystems has been substantiated (reviewed in 115). OF PAMPs AND OTHER ELICITORS; A PATHOGEN’S PERSPECTIVE Just the way I am… The term ‘elicitor’ was coined by the late ‘father of molecular phytopathology’, Dr. Noel Keen, who referred to pathogen-produced compounds that would stimulate (elicit) any of a large number of host defense responses aimed at combating an intruder. This is often achieved via induction of host gene expression and new genes are being discovered regularly. The variety of elicitors is staggering and many reviews have focused on discussing them and how they elicit the host response. For example, by breaching plant cell walls using lytic enzymes (discussed earlier), fungi often contribute to their own detection. Cell wall breakdown products, such as oligogalacturonides released by endo- and exoPGs, are strong elicitors of defense responses in plants (some recent reviews are 123, 124, 125, 126, 127). Because they elicit the host defense, an enormous amount of effort has gone into unraveling the molecular basis of signal perception, transduction and actual production of the (protein) effectors of this defense and resistance response with the hope that such mechanisms could be harnessed for biotechnological crop improvement (128). However, from the point of view of the pathogen, production, secretion and/or presence of such ‘elicitors’ is obviously detrimental and most likely highly balanced during co-evolution of different pathosystems. Elicitors could restrict host range of “would-be pathogens”, bringing about immunity against a whole range of species or genera, the so-called broad host range nonhost resistance. Alternatively, elicitors could be specifically produced by certain species or isolates of a species and cause species- or race-specific resistance, respectively (see below). They need not induce full-fledged immunity but could cause intermediate host defense responses allowing degrees of disease depending on the additional pathogenicity and virulence factors of the pathogen. We could consider two classes of these pathogen ‘elicitors of host defense’. One class covers structural components and serendipitous by-products from pathogen or pathogen-induced activities and as such represents more general elicitors that trigger responses in a wide variety of hosts. A second class of elicitors encompasses products that are generated intentionally, sometimes through induced pathogen gene expression, that serve in the disease process and are by definition pathogenicity or virulence factors. These tend to be more host specific. When searching for fungal mutants altered in pathogenicity or virulence, genes involved in elicitor production are often found although genes in the first class are not pathogenicity or virulence factors in sensu stricto.

186


The first class encompasses the more classical group of small compounds, often break-down products from host cell walls, such as oligosaccharides and (glyco) peptides. These are generated through the action of fungal enzymes as part of a pathogen’s infection strategy and as such are expected to lead to reduced virulence if mutated to avoid detection. In defense, the host also produces lytic enzymes, such as chitinases and glucanases. These enzymes generate fungal cell wall products that are potent elicitors boosting its own defense (123, 125). Even though these molecules are obviously detrimental under some circumstances to the invading pathogen, we should not disregard the possibility that they serve a function in pathogenicity and virulence. That is, that the pathogen has evolved to use them to its own advantage such as has been shown for the cutin monomers that can induce fungal gene expression (see above). Another group in this class has been termed PAMPs, pathogen-associated molecular patterns (a term coined first for mammalian bacterial pathogens). They are mostly conserved microbial products invariant among diverse groups of microorganisms, such as bacterial flagellin, lipopolysaccharides, peptidoglycans, chitin, glucan, lipids and other membrane components such as ergosterol, proteins and glycopeptides. Many PAMPs have been shown to be general elicitors of defense (see reviews 125, 126). Because these are part of the organism’s make-up, in order to become more virulent pathogens, these organisms face the challenging task of introducing subtle changes in these components to try to obscure them from the host surveillance machinery as to avoid recognition but without compromising their function. Alternatively or in addition, they need to specifically suppress the induced defense responses (see below). Exciting active research is underway in this area because these ‘antigenic’ PAMPs, many as ‘epitopes’ on microbial surfaces, are fairly conserved and seem to induce a broad-spectrum of resistance. That is, some of them seem to be at least partially responsible for the phenomenon of nonhost or general resistance describing the immunity of a certain plant, the nonhost, to infection by most microbes. Another defining feature of PAMPs is their absence from hosts as to allow discrimination between self and non-self. The molecular basis underlying this PAMP-induced nonhost resistance is beginning to be understood (126). For example, a portion of flagellin, a main component of flagellae conserved among eubacteria, elicits defense responses in plant species as diverse as Arabidopsis and tomato. The Arabidopsis gene responsible for sensing this component is FLS2, a leucine-rich repeat-transmembrane receptor-kinase. This gene has properties very much like the plant resistance genes involved in cultivar-specific, gene-for-gene resistance (see next section; 129). The second class of elicitors is produced purposely by the pathogen to aid in the disease process. Plant hosts have evolved to recognize these molecules through their surveillance system and utilize their presence to trigger defense. Although they are obviously disadvantageous to the pathogen, they have been maintained in the population because they must bestow a selective advantage overall. Although oomycetes are not considered in this review, a significant research effort has given some insight into a group of so-called ‘elicitins’ which although triggering defense, seem to have a function in the disease process possibly by complexing with lipids (130). In addition, they seem to play a role in host range restriction, such as the avirulence gene inf1 from Phytophthora infestans (131), and might be responsible for nonhost resistance (125 and references therein). Well-known elicitors of host defense are the products of avirulence genes which were identified because of their ability to restrict host range of certain pathogens. Spotlight on avirulence genes; no dual roles but unfortunate tell-tales

187


The term ‘avirulence gene’ is a confusing genetic term that describes the outcome of pathogenicity (inoculation) experiments. When the interaction between a pathogen and its host is incompatible, that is, does not produce disease, the fungal gene responsible for this effect is called the avirulence (Avr) gene. Such a gene “interacts” with a corresponding gene in the host plant, the resistance (R) gene, and both genes need to be present (dominant). This genetic concept, coined ‘the gene-for-gene interaction’ (alternatively, race-cultivar resistance or physiological specialization; 132), represents a monogenic trait and is superimposed on the general pathogenicity. Pathogen isolates of a given species displaying different combinations of Avr and avr alleles, constitute races and these can be tested for their genotypes on differential cultivars of a host species harboring different complements of respective, cognate R and r resistance alleles. Avr genes code for AVR proteins, the ‘elicitors’ that trigger the resistance reaction, often a hypersensitive response (HR) that produces visible necrotic lesions. Avr genes have always attracted interest because they are genetically tractable as single dominant genes, can often be visibly scored in pathogenicity tests, and trigger defenses through the presence of and interaction with R genes. The latter point holds the promise of gaining insight into the defense machinery of the hosts with the possibility of transferring such capabilities to different crops and varieties. Since the cloning of the first Avr gene from a bacterial plant pathogen (133), many have followed (reviewed in 134, 135). Far fewer fungal Avr genes have been isolated due to larger genomes involved and inefficient transformation procedures (reviewed in 136, 125). Several Avr genes were isolated using reverse genetic approaches because it was shown that their products elicited defense reactions, such as Avr9 (137), Avr4 (138), Avr2 (139), Ecp1, Ecp2 (140), Ecp3, Ecp4 (141) of C. fulvum and Nip1 of Rhynchosporium secalis (142). In contrast, PWL2 (143), Avr-Pita (formerly called Avr2-YAMO; 144), Avr1-CO39 (145, 146) and AvrACE1 (147) of M. grisea have been cloned by positional or marker-based approaches. Several Avr genes have been mapped on short genetic intervals in the barley smut, Ustilago hordei (148), M. grisea (149), Leptosphaeria maculans (150) and Blumeria graminis (151, 152). Avr genes have been provisionally mapped in another basidiomycete phytopathogen, the wheat stem rust, Puccinia graminis f.sp. tritici (153). Interestingly, whereas most of the Avr genes restrict infection of host isolates of a certain species (race-cultivar resistance), some of the Avr genes restrict host range in a species-specific manner. The PWL genes of M. grisea governing “avirulence” towards the grass hosts rice and weeping lovegrass represent such factors (reviewed in 125). Such genes could play a role in the nonhost resistance phenomenon in many interactions. Molecular mechanisms underlying the genetic concept and the elicitation of resistance responses are beginning to be understood. An initial ‘receptor-ligand’ model in which the Avr gene product (the ligand) binds to the R gene product (the receptor) has not been validated with some notable exceptions (154, 155). Novel insights have put forward the ‘Guard Hypothesis’ in which a R gene product stands guard over other ‘sensor’ proteins embedded in a membrane-bound complex. An Avr gene product will interact, either directly with the ‘sensor’ or by modifying another substrate which in turn will interact with the sensor. The resulting conformational change(s) and/or loss from the complex is perceived by the guarding R gene product which will initiate a signal transduction cascade involving MAPKinases, leading to defense responses. Many R genes have been isolated and studied and a detailed discussion of their involvement in these responses falls outside the scope of this review. The reader is referred to many excellent recent reviews (e.g., 156, 157, 128, 158, 159, 160). For the scope of this review we would like to focus on the potential roles of these “avirulence” genes in the disease process. One of the intriguing problems is that Avr genes have

188


been cloned from an astonishingly large group of very diverse organisms (bacteria, fungi, oomycetes, viruses, nematodes and even insects) and, while there are groups of related gene sequences or protein configurations (often within pathogen species), there is a staggering diversity overall. Moreover, it has been exceedingly difficult to assign functions to these genes (other than eliciting the host defense). Recent publications have described exciting new experiments shedding light on possible functions of such pathogen-produced elicitors. Most of the new insights, however, come from the work on bacterial Avr gene products. In this field it has become clear that bacteria inject many so-called ‘effector’ proteins into their host cells with diverse roles in pathogenicity and virulence (134, 157, 135). It appears that plants have co-opted a subset of these factors, the elicitors or Avr gene products, as triggers for defense (156). It follows that there is no intended ‘dual role’ for “Avr gene products” from the pathogen’s perspective. Table 1 compiles data on the fungal effector protein genes with elicitor activity (Avr genes). Several lines of evidence suggest roles for effector proteins in the disease process. Upon contact with the host, transcription is turned on for Avr2, Avr4, Avr9, ECP1 and ECP2 from C. fulvum (136), PWL2 (143) and ACE1 from M. grisea. The latter gene is only expressed during penetration in appressoria (147). However, Avr9 expression can also be induced in vitro under nitrogen-limiting growth conditions (169). Significantly, it was also observed that on compatible hosts, virulence of several Avr-containing isolates was often greater than of respective isogenic strains from which the Avr gene had been deleted (see Table 1). None of the Avr genes isolated to date are required for pathogenicity. In addition, when surveying virulent isolates many still harbored a mutated avr allele associated with the recessive genotype. The very fact that such alleles are maintained in the population hints at a likely function in virulence or at least some fitness advantage. Such correlation between Avr genes contributing to virulence and their persistence in the population is not unexpected and has been noted before. This phenomenon and the corollary that cognate host R genes interacting with such persistent factors provide durable resistance, has been experimentally tested (170, 171). The M. grisea proteins, PWL1 and PWL2, have signal peptides suggesting extracellular functions, but when PWL2 was produced in E. coli, no elicitor activity could be detected upon infiltration. A single point mutation apparently was sufficient to abolish “avirulence” function suggesting that this protein is involved in a highly specific interaction (143). Similarly, Avr-PiTa could be active extracellularly. Although a processed 176 amino acid-long effector harbored the elicitor activity, this did not induce an HR when introduced in the apoplasts. It was suggested that the metalloprotease domain is involved in some regulatory function (154). The M. grisea ACE1 effector most closely resembles a polyketide synthase and likely has an enzymatic activity involved in secondary metabolism. ACE1 is only expressed during penetration in appressoria and its product is not found secreted in plant cells. Intriguingly, it is part of a large cluster of genes involved in secondary metabolism, ten of which are co-transcribed with ACE1 in appressoria during penetration. It seems likely that a secondary metabolite is the actual elicitor but its function in virulence is unknown (167). There is a precedent with the avirulence gene avrD from the bacterial pathogen, Pseudomonas syringae pv. tomato, which also codes for an enzyme catalyzing the synthesis of a HR-eliciting metabolite, syringolide (reviewed in 125). The other Avr products mentioned in Table 1 have several features in common: they are relatively small peptides, are secreted in the apoplast and possess an even number of cysteine residues involved in disulphide bridge formation which contributes to the secondary structure of a functional elicitor. Structural similarities of Avr products with so-called ‘cysteine-knotted

189

pathogen gene description vir function citation(s) Cladosporium fulvum

Avr2 78 aa, signal sequence, extracellular, 58 aa elicitor (8 Cys)

+; mutations and partial deletions; maintained in population

139

Avr4 135 aa, signal sequence, extracellular, 86 aa elicitor (8 Cys), chitin binding/protection (”stealth factor”?), heat labile; separated elicitor and virulence functions

+; point mutations but maintained in population

138, 161, 162, 163

Avr9 63 aa, signal sequence, extracellular, 28 aa elicitor (6 Cys); proteinase inhibitor?

-; deleted in avr genotypes 137, 164

Ecp1 96 aa, signal sequence, extracellular, 65 aa elicitor (Cys-rich), suppressor of host defense?

+?, ECP1 produced by all Cf strains

140, 165

Ecp2 165 aa, signal sequence, extracellular, 142 aa elicitor (Cys-rich), suppressor of host defense?

+, ECP2 produced by all Cf strains

140, 165

Ecp4 119 aa, signal sequence, extracellular, 101 aa elicitor (6 Cys)

? 141

Ecp5 115 aa, signal sequence, extracellular, 98 aa elicitor (6 Cys)

? 141

Magnaporthe grisea PWL1 147 aa, signal sequence, extracellular?, species-specific ?; maintained in population 166

PWL2 145 aa, signal sequence, extracellular?, species-specific ?; maintained in population 143

Avr-PiTa 223 aa, signal sequence, extracellular?, 176 aa elicitor, neutral Zn2+ metalloprotease, indirect enzymatic?

?; both deletions and mutations found

144, 154

Avr1-CO39 ? ?; absent from many isolates 146

ACE1 (Avr-IRA7)

polyketide synthase, indirect elicitor through enzymatic action?

? 147, 167

Rhynchosporium secalis

AvrRrs1 (NIP1)

NIP1; 82 aa, signal sequence, extracellular, 60 aa elicitor (10 Cys), hydrophobin?

+; mutations and deletions in population

142, 135, 168

Table 1. Known fungal effector protein genes with at least “avirulence” function (see also 136, 125). Cys = cysteine residue; aa = amino acid; + or – or ? = the gene does or does not increase virulence or has not been tested for this (see text).

proteins’ have been noted before (103). Both ECP1 and ECP2 contribute to C. fulvum virulence and it was suggested that they suppress host defense responses (165). ECP1-deletion strains could still infect and grow in host tissues but conidiospore formation was reduced. The growth in culture or in the plant of ECP2- deletion strains was severely inhibited. ECP4 and ECP5 have not been studied in detail and their functions remain unknown. Avr9 is deleted in virulent isolates and it does not seem to contribute significantly to virulence in susceptible cultivars although a subtle role in pathogen fitness in the population under field conditions cannot be ruled out. AVR9 has a structural motif found in several small proteins such as proteinase inhibitors, ion channel blockers and growth factors but no actual function has been attributed to this first Avr gene to be cloned from fungi (164). Avr2 was recently cloned and has not been studied in detail yet. All recessive avr2 alleles code for truncated proteins suggesting an advantage to its maintenance (139). Avr4 is a most interesting protein which binds to a fungal component (162). It was recently shown that this was chitin (163). All recessive avr4 alleles have point mutations affecting the cysteine residues involved in two of the four disulphide-bridges leading to proteins with a changed conformation that are more protease-sensitive and probably no longer accumulate in the host apoplast and thus elude recognition by the plant. However, these mutated AVR4 protein forms still bind chitin and when bound are not sensitive to proteases. It was suggested that AVR4 functions by binding or masking chitin, thereby protecting the invading fungal mycelium from host chitinase activity. It was shown that virulence and elicitor functions could be separated (163). The role of Rhynchosporium secalis NIP1 in virulence is suggested by its non-specific toxicity in leaf tissues of host and nonhost cereals as well as its resistance gene-independent stimulatory effect on the plant plasma membrane H+-ATPase (135, 168). Because this is a necrosis-inducing pathogen killing host cells for their contents, NIP1 is an example of a toxin (discussed in the next section) the recognition of which certain hosts have co-opted for defense elicitation. PERTURBING THE HOST Secondary metabolites and toxins Fungi synthesize an enormous variety of secondary metabolites and only a subset has phytotoxic activity. These toxins are primarily produced by facultative saprophytic, necrotrophic or hemibiotrophic fungi during their necrotrophic stage. Phytotoxins can play roles in fitness and virulence and occasionally bestow a major advantage on its producer when encountering a host sensitive to the compound. The fungus will then be characterized as a major pathogen and the genes involved in the biosynthetic pathway or the regulation thereof will be revealed as pathogenicity or major virulence factors.

Phytotoxins disable host cellular functions, lead to nutrient leakage and/or host cell death to the advantage of the invading pathogen. Host susceptibility, associated with sensitivity to the toxin, often leads to cell death, a necrosis reminiscent of the HR and experimental data indicate that these toxins may indeed produce the same programmed cell death (PCD) response as seen during a HR brought about by an gene-for-gene (Avr/R-gene) interaction. However, in the latter case, the pathogen-host interaction leads to incompatibility. These results strengthen the hypothesis that these organisms exploit the host resistance response for susceptibility. Phytotoxins can be divided into two classes. General toxins can produce necrotic reactions in a wide variety of plants, even in plants on which the corresponding fungus is not pathogenic, indicating that these general toxins merely “assist” in the disease progression or virulence and


that other factors as discussed in this review are needed. Host-specific toxins (HSTs) are major host range determinants and essential for disease (pathogenicity or virulence). They consist of a group of structurally complex and chemically diverse metabolites produced by plant pathogenic strains of certain fungal species. The genes responsible for the biosynthesis of toxins and sometimes their secretion and/or self-protection, are often found in “selfish clusters” which are hypothesized to be maintained because of the selective advantage they can provide and because horizontal gene transfer is an important mechanism by which these clusters propagate and hence exist (172). Whether host cell death is essential for susceptibility to HST-producing fungi is contentious. For details on the above section the reader is referred to several excellent recent reviews (173, 172, 174, 175, 176).

As mentioned in the introduction, the role of general mycotoxins in the disease process is unclear but they might increase virulence. For example, trichothecene toxins such as deoxynivalenol (DON) seem to contribute to virulence. When the Fusarium graminearum trichodiene synthase (Tri5) gene involved in trichothecene biosynthesis, was deleted, the mutants were still pathogenic but displayed reduced virulence, spreading less in wheat tissues (177, 178). It was suggested that DON conditions the host tissue for colonization, thereby enhancing the spread of F. graminearum on maize. Many of the other genes in the trichothecene gene cluster have been described by comparing the homologues in F. sporotrichioides (179). Their contribution to virulence has not yet been investigated, but is likely to correlate with the production of DON. Fumonisins are a family of polyketide mycotoxins produced by Gibberella moniliformis (anamorph Fusarium verticillioides) and by certain other species of the genus Fusarium but deletion of one of the biosynthetic genes, FUM1, showed that this toxin is not necessary for maize ear rot and ear infection in field tests (180, 181). In this fungus a large gene cluster of 18 contiguous genes that include 5 previously found FUM genes, was recently described. Fifteen of these are cotranscribed and are thought to be involved in the biosynthesis of this complex compound and self-protection through an ABC transporter (see next section). However, deletion of 3 of these had no or only a marginal effect on fumonisin production (182). Oxalic acid is a simple metabolite produced by many organisms. Its phytotoxic activity has been known for years and it has been implicated as a virulence factor in several phytopathogenic fungi (reviewed in 183). It is an essential pathogenicity determinant of Sclerotinia sclerotiorum (184, 185). Its role in pathogenicity is hypothesized to include lowering the pH in the infection site to allow for optimal cell wall degradation by fungal enzymes (discussed earlier; 77, 79), by lowering the pH to weaken the plant, and chelating Ca2+ thereby interfering with defense response signaling (183). The importance of oxalate as a pathogenicity factor is corroborated by the discovery of host oxalate oxidases, germins (g-OXOs) in cereals or germin-like proteins (gl-OXOs, possibly superoxide dismutases) with a wider distribution in the plant kingdom. These oxidases detoxify oxalate to produce peroxide, thereby establishing a link with defense responses to invasion by fungal pathogens (186). Many genera of plant pathogenic fungi produce photosensitizing perylenequinones and other light-activated, non-specific toxins, suggesting that photosensitization is a common contributor to pathogenicity. These ‘photosensitizers’ define a large group of structurally diverse compounds that are activated by visible wavelengths of light and generate activated oxygen species toxic to living cells. Cercosporin is such a non-specific fungal polyketide toxin classified as a photosensitizer and is produced by many Cercospora species. It most likely acts by peroxidation of membrane lipids leading to membrane breakdown, nutrient leakage and cell death (reviewed in 187). Genes involved in the cercosporin biosynthetic pathway and regulation

192


(188, 189), as well as self-protection against the toxin (190, 191, 192) have been identified in C. nicotianae.

Among the many HSTs described, several have been studied in great detail and some or all of their biosynthetic genes isolated. They include small non-ribosomally produced cyclic peptides (HC-toxin and Victorin) and a polyketide (T-toxin) from Cochliobolus species, a polyketide (PM-toxin) from Mycosphaerella zeae-maydis and several complex, highly host range-specific toxins (AAL-, AK-, AF-, ACT- and AM-toxin) from different A. alternata forma speciales (see recent reviews 173, 175). The toxin loci harboring the clustered genes can be very complex and are sometimes located on a dispensable chromosome, such as for the AM- and AK-toxins from A. alternata (193, 194, 195). However, single genes, ToxA and ToxB, encode small, secreted, host selective, ribosomally-produced toxic proteins that function in the plant as the primary determinant of pathogenicity in the Pyrenophora tritici-repentis -wheat interaction (196, 197). Elucidating the role of these HSTs in the disease process is important and might lead to insights into how to counteract the respective pathogens. Although these HSTs are major pathogenicity factors, many of these fungi are still able to infect to some degree whether they are a toxin-producing variety or not. Toxins aid in the progression of the disease process either by killing host cells which most of them do or by suppressing the host response (discussed in the next section). Of course, killing the host cells is a rather resolute way of avoiding a host defense response (175). As mentioned above, the way HSTs seem to kill their host cells is through the induction of the endogenous Programmed Cell Death (PCD) pathway (reviewed in 175). An exception might be the cyclic tetrapeptide HC-toxin from C. carbonum. Recent experiments revealed that HC-toxin targets and inhibits a histone deacetylase (HDAC) in many organisms, likely interfering with the regulation of host genes possibly involved in defense. The fungus’ own HDAC is toxin-insensitive due to mutations (198). In many organisms specialized proteins involved in transport across cell wall membranes have been described. One large class also described in fungi, comprises members of the ATP-binding cassette (ABC) and major facilitator superfamily (MFS) of transporters. In plant pathogenic fungi they mediate the secretion of host-specific toxins to disable the host but also for self-protection by keeping toxin levels inside fungal cells low. In addition, as will be discussed in a later section, they provide protection against plant defense compounds (reviewed in 199, 200). Mutants in genes coding for such transporters are therefore likely to have an effect on pathogenicity and virulence. It was recently shown that the soybean pathogen, Cercospora kikuchii, harbors a cercosporin facilitator protein, CFP. CFP mutants exhibit dramatic reductions in cercosporin production and virulence, and increased sensitivity to the toxin. CFP encodes a cercosporin transporter that contributes resistance to cercosporin by actively exporting cercosporin, thus maintaining low cellular concentrations of the toxin suggesting a role in self-protection (201, 202). Such efflux pump functions were also proposed for two copies of a gene, TOXA, in the HC-toxin cluster of C. carbonum; deletion of one of them had no effect but a double mutant could not be obtained which was interpreted to indicate that they were essential for export and self-protection (203). TOXE in this cluster is a transcriptional regulator with C-terminal ankyrin-like repeats which binds to a 10-base pair motif (‘tox box’) found in the promoter of many of the genes in this cluster. As such it is a pathogenicity factor because it regulates the biosynthesis and most likely self-protection from the toxin (204, 174). A role in the regulation of self-protection was also suggested for the transcription factor, BAP1, related to TOXE, from C. fulvum (205). Phytohormones

193


Plant hormones are involved in the regulation of several stages of plant growth and

development, such as cell division and cell elongation, and tissue differentiation. They can be synthesized not only by plants but also by microorganisms including plant pathogenic fungi. Because of the demonstrated ability of some plant pathogenic fungi to synthesize these plant growth regulators, their potential role in disease development have been investigated. However, the information obtained to date regarding the putative role of phytohormones of fungal origin in the infection process is rather sparse and its interpretation remains controversial. Several plant pathogenic fungi have been shown to produce the auxin indole-3-acetic acid (IAA) in vitro. Fungi belonging to the genus Rhizoctonia produce IAA efficiently from tryptophan (206). Interestingly, further study showed that the amount of IAA produced my R. solani is increased by co-cultivation with rice cells and with rice cell culture filtrates. Yet the potential role of IAA in infection caused by Rhizoctonia species remains unclear (207). There is some indication that auxins facilitate the Oomycete pathogen Pythium group F infections (208); crude culture filtrates or low molecular weight fractions of this pathogen cause morphological alterations in root tips similar to those observed in roots treated with IAA. In addition, chemical analysis has shown that these filtrates contain IAA. A plant pathogen that induces important morphological alterations in its host is U. maydis. The most remarkable symptom induced in infections of maize by U. maydis is the formation of tumors consisting of hypertrophic and hyperplastic plant cells and fungal hyphae. Because of this very specialized interaction it has been hypothesized that phytohormones, either produced by the fungus or induced in the plant, likely play a role in tumor formation. Moulton (209) found that galled tissue contained an auxin in much higher amounts than uninfected tissue. This auxin was later identified to be indole-3-acetic acid (210, 211). It has been shown that U. maydis is able to produce IAA in vitro from tryptophan (211, 212). In recent years, attempts have been made to identify mutant U. maydis strains impaired in their ability to produce IAA in culture, as a first step towards elucidating the putative role of this auxin in pathogenicity. Guevara-Lara et al. (213) have characterized a number of mutants with reduced IAA production in vitro. Unfortunately, the only strain null for IAA production in vitro available to them, named udi-1 (214) was auxotrophic for L-methionine. Because L-methionine auxotrophic strains are compromised for pathogenicity (215, 216) and show reduced in vitro mating reactions (216), the effect of udi-1 mutation on disease severity was confounded in this IAA- strain. Attempts to obtain IAA-/meth+ segregants generated only strains with partial loss of IAA production in vitro. Thus, although inoculation with these strains suggests a role for IAA production in gall formation, these results were not conclusive because even wild-type strains exhibit variability in IAA production in vitro. Another approach that has been used to study the role of IAA in gall formation has been to identify genes predicted to be involved in IAA production, and generate mutant strains deleted for such genes. Because U. maydis is able to produce IAA in vitro from tryptophan, the role of indole-3-acetaldehyde dehydrogenase, Iad1, which was thought to be involved in this pathway, was investigated (217). However, iad1- mutant strains were fully pathogenic and in this mutant background IAA was still produced in vitro from tryptophan, suggesting a different pathway for IAA synthesis.

Other phytohormones produced by plant pathogenic fungi that have been analyzed include gibberellins. Several fungi, such as the rice infecting fungus Gibberella fujikuroi, produce these compounds. And although the biosynthetic pathway of gibberellins including the analysis of all genes involved have been described in detail, still, the role of this phytohormone in disease development remains unclear (218, 181). Cytokinins have also been detected in cultures of some plant pathogenic fungi. In Pyrenophora teres a pathogen of barley, and

194


Dreschslera maydis infecting maize, cytokinins have been detected in culture filtrates using HPLC (219). It was also found, in an assay using detached leaves, that cytokinin levels were higher in infected areas. In these areas an accumulation of metabolites was also observed, leading the authors to postulate a possible role of cytokinins produced by the fungus in generating “metabolic sinks” in these areas to help the proliferation of the fungus. Some information indirectly implicates variation in production of phytohormones by plants in response to fungal pathogens. MAP kinase cascades have been shown to play a role in mediating stress responses in plants to both biotic and abiotic factors. Xiong and Yang (220) found a MAP kinase from rice involved in plant resistance that is induced by the phytohormone abscisic acid. The expression of this MAP kinase was induced by infection with the blast fungus with a much higher level of expression in a resistant interaction than in the susceptible. However, there is no evidence that the induction of this gene after fungal infection is related to the production of abscisic acid. In potato, the gene prp1-1, involved in defense against Phytophthora infestans, has been shown to encode an auxin-responsive glutathione S-transferase (221). Anticipating and countering the host defense Apart from physical barriers against entry of pathogens, host plants defend themselves by means of a smorgasbord of chemical reactions, compounds, anti-microbial peptides and proteins (AMPs) and further physical constraints such as cell wall reinforcement, microbe encasement, up to self-destruction of invaded and surrounding cells (HR) and a plant-wide heightened state of alert and resistance (systemic acquired resistance, SAR, and induced systemic resistance, ISR). Some are quick such as the production of active oxygen species (AOS), some antimicrobial compounds are already made and stored or compartmentalized and just need release or a simple conversion step (phytoanticipins), others need the induced production of (PR) proteins and enzymes that will generate such compounds (phytoalexins) or bring about the other type of reactions mentioned. All need to be triggered by pathogen attack, but some reactions occur rather non-specifically and may protect against a wide range of potential pathogens, whilst others occur upon host recognition of certain type of pathogen “effectors” and elicitors as discussed earlier (recent reviews are 222, 223, 224, 225, 37, 226, 227). One tactic to avoid host defenses already mentioned and likely employed by necrotrophs and hemibiotrophs in their later stages of infection, is the killing of host cells (ironically possibly through induced apoptosis, the HR). This would ensure avoidance of defense responses while paving the way for colonization of the tissues. However, recent literature highlights the fact that plant pathogens have evolved very sophisticated means to avoid, anticipate and/or suppress defense responses. Understanding how they do this would allow us to counteract these offensive moves. As mentioned before, the adhesion to and very early penetration of host cell walls by the obligate biotrophs, the cowpea rust Uromyces vignae and powdery mildew Erysiphe species, induce nonspecific cell wall-associated defense responses via RGD-containing ‘adhesins’ (see back; 36). Communication is through thin plasma membrane-cell wall connections known as ‘Hechtian strands’. This response was seen in both host and nonhost interactions by two Erysiphe species, but only during the incompatible interaction by the cowpea rust on the nonhost, pea; the compatible interaction on its host cowpea, resulted in a reduction of Hechtian strands directly under the penetration site and was correlated with the suppression of host

195


defense responses (36, 37). How these signals are communicated and what genes are involved is not known yet but is actively researched and is of obvious interest. The production of AOS is a very early defense response. It was recently shown that oxalic acid, a pathogenicity factor for Sclerotinia sclerotiorum (see back), actively suppressed this oxidative burst in a tobacco host (228). Many fungi contain enzymes such as superoxide dismutase (SOD), peroxidases, catalase, and perhaps laccases and polyphenol oxidases that can help to remove or inactivate these early-produced AOS (229). A catalase gene of C. gloeosporioides f.sp. malvae is highly expressed during the necrotrophic phase of infection of round-leaved mallow and is thought to play such a role in inactivating AOS (230). However, in several fungi, catalases have been deleted without an obvious reduction in virulence: in Claviceps pupurea on rye (231), C. fulvum on tomato (232), Botrytis cinerea on tomato and bean (233) and all three known catalases in C. heterostrophus on maize (234). Given the probable redundancy in activities that can detoxify AOS (in both gene families and enzyme diversity), this is not surprising. Many fungi, particularly biotrophs, probably employ a more subtle suppression of active, induced host resistance. Studies on the C. fulvum (avirulence) effectors, ECP1 and ECP2 (see above), suggested a role in suppression of the host defense responses (165). Another indication of such a mechanism in fungi came from the study of a small secreted protein encoded by the C. gloeosporioides gene, CgDN3. The basic 54 amino acid mature form, having some homology to a plant wall-associated receptor kinase, seemed to suppress the HR during the biotrophic phase of the infection (235). The obligate biotrophic powdery mildew Blumeria graminis f.sp. tritici, was shown to be able to suppress defense responses in a compatible wheat host (236). Similarly, barley infection by a virulent strain of Blumeria graminis seemed to induce penetration susceptibility and suppression of a race-specific hypersensitive resistance upon subsequent inoculation with an avirulent isolate (237). A recently described rust transferred protein, RTP1p, from the obligate biotroph Uromyces fabae has a nuclear localization sequence (NLS) and is found in the host nucleus where it could serve as a transcriptional effector although the protein reveals no other homologies. This would be the first example of a fungal effector being translocated into the targeted host cell (9). Several similar studies have illustrated that effector molecules from bacterial phytopathogens can perturb host transcription, including of defense genes (238, 239, 240, 241). An intriguing discovery mentioned earlier was the fact that HC-toxin from C. carbonum affects histone deacetylases (HDACs) likely affecting the regulation of host genes, potentially those involved in defense (242). In addition to fungal effectors suppressing the defense reactions of the host, the (enzymatic) actions of such fungal factors could also have indirect effects. For example, cell wall (pectin) degrading actions of the wheat stem rust fungus, P. graminis. f. sp. tritici, releases small oligomers of galacturonic acid which are endogenous suppressors of disease resistance (243). Moreover, it was postulated that enzymatic breakdown products of wheat pectins upon rust invasion differ between resistant versus susceptible hosts, whereby only susceptible plants seem to release suppressor-active pectic oligomers (244). The fungal pathogen must contend with a hostile host environment in which nutrients are limiting or lacking. Many virulence mutants will be affected in genes involved in general metabolism of the fungus or ‘fitness’ within the host environment. This could be the case for mutants in the ornithine decarboxylase gene of Stagonospora (Septoria) nodorum which is required for virulence on wheat (245). Similarly, mutation of an arginine biosynthesis gene, ARG1, caused reduced virulence in F. oxysporum f. sp. melonis (246) and the same was found for the disruption of an alcohol oxidase gene, Aox1, in C. fulvum (247). Other recent

196


experiments suggest that the glyoxylate pathway is essential for disease development by Leptosphaeria maculans, the causal agent of blackleg disease of canola (Brassica napus). Generation of pathogenicity mutants in this fungus led to the cloning of a gene encoding isocitrate lyase, a component of the glyoxylate cycle, which proved to be essential for the successful colonization of B. napus (248). The need for compounds that are limiting in host tissue can interfere with analysis of specific interesting mutants. As discussed earlier in the chapter, the L-methionine auxotrophic U. maydis strain was compromised for pathogenicity when analyzing the udi-1, reduced IAA production mutant (214, 213). In contrast, a U. maydis ammonium permease gene (ump2) induced in budding growth was found to be critical for filamentation in response to limited nitrogen availability but not required for pathogenicity under laboratory conditions (249). Redundancy of ammonium transporters in U. maydis may at least in part be able to compensate for lack of ump2 function. In some cases there will be differences when comparing such mutants for growth in planta and in vitro on defined media, which might give insights into the requirements for growth inside host tissues during specific life cycle stages of the pathogen. Of course, such tests will not be feasible for biotrophs. In addition, some of these fungal genes are specifically induced in planta but it is difficult to clarify the role of genes that confer auxotrophy when mutated in pathogenicity or virulence. Successful pathogens must have the capacity to detoxify antimicrobial host defense compounds or synthesize inhibitors of the enzymes that produce them. Several virulence determinants turned out to be fungal enzymes that could detoxify phytoanticipins or phytoalexins. The ability of Nectria haematococca to detoxify the pea phytoalexin pisatin by demethylation, contributes to its virulence on pea. This activity is encoded by the pda1 gene, a cytochrome P450. It is part of a gene cluster with features reminiscent of bacterial ‘pathogenicity islands’ and sits on a dispensable chromosome. In addition, it was suggested that other pea pathogens having PDA activity, such as Ascochyta pisi, Mycosphaerella pinodes and Phoma pinodella might have independently evolved a specialized cytochrome P450 as a virulence trait for a common host (250, 251). Saponins are preformed antimicrobial glycosylated steroidal and/or steroidal alkaloids found in many plants (224). Detoxification of the oat saponin, avenacin, by avenacinase from Gaeumannomyces graminis, is a prerequisite for this fungus to infect oat roots making this a pathogenicity factor; it also functions as a host range determinant since avenacinase is not required for normal infection of wheat which does not contain saponins (252). The ability to degrade certain compounds might not always be necessary for full virulence or it might serve other functions. Colletotrichum coccodes harbors a gene, beta2-tomatinase, which can detoxify the tomato saponin, α-tomatine, but gene-disruption mutants are as pathogenic as wild type isolates on green tomato fruit, an organ containing high levels of α-tomatine, probably because the fungus has other enzymes that can degrade it. Nevertheless, heterologous expression of the Septoria lycopersici beta-2-tomatinase in Nectria haematococca bestowed on this fungus the ability to infect green tomato fruits (253). In a recent illuminating study, this S. lycopersici α-tomatine-detoxifying enzyme contributed to its virulence on tomato. In addition, the degradation products of the hydrolysis of α-tomatine suppressed induced defense responses in tomato (254). The role in virulence of the ability to degrade other antimicrobial compounds has been studied as well. Even though it was shown that the Fdb1 and Fdb2 genes of Fusarium verticillioides were necessary for the detoxification of the corn preformed antimicrobial compound BOA (2-benzoxazolinone, a cyclic hydroxamic acid which classifies as an phytoanticipin), this did not contribute significantly to the virulence on corn (255).

197


In addition to their roles in toxin secretion discussed above, some fungal ABC and MSF transporters may provide protection against plant defense compounds. Mutants in these transporters therefore will likely also be affected in virulence due to inhibition. From an agronomic point of view, transporters are interesting because they determine sensitivity to fungicides and overexpression of some of these transporters can lead to the development of resistance to chemically-unrelated compounds, a phenomenon described as multidrug resistance (see reviews 199, 200). An ABC transporter gene, PMR5, is involved in multidrug resistance in the phytopathogenic fungus Penicillium digitatum, and deletion mutants of this and another gene, PMR1, increase sensitivity to fungicides and natural phytoalexins (256). This has also been demonstrated for the cloned ABC transporter BcatrB from B. cinerea which proved to be a virulence factor. A deletion mutant was more sensitive to resveratrol, a phytoalexin from grape, although virulence on grape leaves was only slightly reduced (257). More convincingly, the ABC transporter Mgatr4 of the wheat pathogen Mycosphaerella graminicola, was shown to be a virulence factor that affects colonization of substomatal cavities in wheat leaves (258). The potato necrotroph, the dry rot fungus Gibberella pulicaris, harbors an ABC transporter, Gpabc1, which is essential for infection of potato tubers. Deletion mutants could not colonize potato disks, but in addition were hypersensitive to the potato phytoalexins, rishitin and lubimin, thereby suggesting a direct link between pathogenicity and a first-line of defense by being able to secrete plant defense compounds (259). One of the M. grisea pathogenicity mutants able to infect through appressoria but unable to colonize rice and barley tissues, proved to possess a mutated ABC1 transporter. This transporter might not be very selective since the rice phytoalexin sakuranetin and several fungicides, even though they strongly induced the transcription of the gene, did not inhibit growth of the mutant (260). SPORULATION Melanin is an important component of fungal structures. Many fungal spores are heavily melanized for protection (discussed above). Mutants in genes involved in the biosynthesis of melanin are therefore likely to affect the sporulation process or the viability of produced spores. The A. alternata BRM2 gene is involved in melanin biosynthesis and disruption has no affect on pathogenicity, vegetative growth, or the number of conidia produced, but conidial size and septal number are reduced, suggesting that melanin is associated with conidial development. In addition, mutant spores are more sensitive to UV light than wild type ones, demonstrating that melanin confers UV tolerance (121). Several single-gene mutations that control various stages of spore morphogenesis were described in M. grisea; in different combinations these also had an effect on appressorium formation and further growth in the plant and hence on pathogenicity (261). The M. grisea gene ACR1 (Acropetal) encodes a stage-specific negative regulator of conidiation involved in conidiophore architecture. acr1(-) mutants fail to turn off the expression of the hydrophobin encoding gene MPG1 in dormant spores resulting in the production of spores that cannot progress through the disease cycle and are therefore pathogenicity mutants (262). Several genes encoding components of known signaling cascades seem to regulate sporulation processes as well. The histone deacetylase Hda1 from U. maydis is essential for teliospore development (263) and regulates several processes, including the production of spore-specific protein, Ssp1, a factor highly produced in teliospores and likely involved in lipid metabolism. However, deletion of Ssp1 itself had no effect on virulence (264). A search for suppressors of the filamentous phenotype of the U. maydis adr1 mutant (discussed below), identified a gene, hgl1 whose product is a potential target for

198


phosphorylation by PKA (265). Inoculation of plants with compatible combinations of hgl1 mutant strains produced galls, generally larger than those produced by wild-type strains but lacking mature darkly pigmented teliospores. Close examination of these tumors revealed that the fungus is unable to progress from the stage of hyphal fragmentation to teliospore development. Interestingly, haploid cells harboring this mutation become pigmented when cultured in vitro. ECP1 of C. fulvum, an effector (avirulence) protein discussed above, seems to have a function during sporulation. An ECP1-deficient strain emerged from stomata of the lower epidermis, but failed to sporulate as abundantly as did the wild-type strain (165). CHANGES IN VIRULENCE CAUSED BY ENDOGENOUS GENETIC ELEMENTS More than 30 years ago, the cause of hypovirulence in strains of the chestnut blight fungus, Cryphonectria parasitica, was shown to be the presence of a double stranded (ds)RNA virus (reviewed in 266). Phenotypic alterations, in addition to hypovirulence, that distinguished infected colonies from isogenic dsRNA-free virulent strains, included altered colony morphology (as evidenced by a reduced growth rate and a lobed appearance to the general circular shape of colonies), female infertility, reduced asexual sporulation (conidiation), and reduced pigmentation. The pleiotropic nature of these phenotypic changes suggested that one or several regulatory signaling pathways were affected by viral-encoded products. Indeed, further experiments strongly suggested the involvement of heterotrimeric G-proteins encoded by cpg-1 and cpg-2 (267; see further). Such dsRNA mycoviruses have since also been found in other plant pathogenic fungi such as Botrytis cinerea. In this fungus a 33-nm isometric mycovirus with a 6.8-kb dsRNA genome was shown to be responsible for the reduced virulence, lower laccase activity and conidiation rate on bean (268, 269). In F. graminearum the presence of a virus brings about pronounced morphological changes, including reduction in mycelial growth, increased pigmentation, reduced virulence towards wheat, and decreased (60-fold) production of trichothecene mycotoxins (270). Also in Nectria radicicola, the causal fungus of ginseng root rot, several viral dsRNAs resembling plant cryptic viruses were found. However, in this case presence of the 6.0 kb dsRNA was associated with high levels of virulence, sporulation, laccase activity, and pigmentation in this fungus. The 6.0-kb dsRNA-cured strains completely lost virulence-related phenotypes. Interestingly, the up-regulation of virulence may occur through signal-transduction pathways involving cyclic AMP-dependent protein kinase and protein kinase C (271). The infectious transmission of other genetic elements such as plasmids and mutant mitochondrial DNAs causing senescence, can also affect virulence. So-called dysfunctional or “suppressive” mtDNA mutations proliferate rapidly in growing cells and gradually displace functional organelles that contain wild-type mtDNA molecules. Senescence and death ensues if the suppressive mtDNA contains a lethal mutation. Suppressive mtDNA mutations and mitochondrial plasmids can elicit cytoplasmically transmissible “mitochondrial hypovirulence” syndromes in at least some of the phytopathogenic fungi (reviewed in 272). IMPORTANCE OF SIGNAL TRANSDUCTION IN PATHOGENICITY Signaling pathways in plant pathogenic fungi clearly play central roles in environmental sensing, mating processes, morphogenesis and communication with the host. The pathways most well studied and most significant in the fungi, as judged by the frequency with which they are encountered in mutants affected in these processes, are the cAMP dependent (PKA) and

199


mitogen activated protein (MAP) kinase pathways. The importance of signaling involving calcium as a messenger molecule for these processes is clearly a common phenomenon. These pathways tend not to stand alone but rather frequently crosstalk and can be viewed as a web of interconnected pathways, often several of which impact a single phenotypic outcome. A common problem with interpreting these studies is that perturbing one pathway is like cutting a single strand of a spider’s web, consequently deforming many unintended interconnected strands. Several excellent reviews have recently been published that deal with these topics in great detail. cAMP signaling and its interaction with MAP kinase pathways in phytopathogenic fungi was very recently well reviewed (273). Thus here we provide a few well-characterized illustrative examples. In general, mutations inactivating PKA lead to debilitation of a fungus. For example, in U. maydis, mutation in adenylate cyclase, UAC1 (274, 275), the catalytic subunit of PKA, Adr1 (276) or in a G-protein alpha subunit Gpa3, required for activation of adenylate cyclase (277), cause similar consequences, freezing this dimorphic fungus in the filamentous phase and eliminating the ability to colonize maize. Nonetheless, upon close comparison of these mutants they do differ in several subtle phenotypes. For example, the adenylate cyclase mutant is much more invasive on agar than is the PKA catalytic subunit mutant (Gold, unpublished). This indicates that cAMP likely plays a role beyond the function of adr1. In M. grisea an activated cAMP pathway appears to play a major role in transmitting surface recognition signals to generate the morphogenic response of appressorium formation. Inactivated mutants such as adenylate cyclase (Mac1) (278), catalytic subunit of PKA (CpkA) (279) and G-alpha (MagB) (280) all show defects in appressorium formation and plant infection. The hydrophobin Mpg1 is required to sense the surface and induce normal levels of appressorium formation but exogenous cAMP can correct this phenotype indicating that Mpg1 is upstream of cAMP signaling (281). A few additional examples where cAMP signaling has been implicated as a controlling mechanism include the pea flavanoid induced germination response of F. solani spores (33) and the vegetative growth, sporulation and full pathogenicity of B. cinerea (282). As noted above with the various U. maydis mutants, in B. cinerea the mutant phenotypes of the cAMP pathway mutants vary. Mutation in the G-alpha subunit gene, bcg1, has much greater impact on pathogenicity than does a bac mutation indicating that BCG1 has a role beyond regulation of BAC adenylate cyclase. In C. parasitica, infection with a mycovirus (hypovirus) causes a number of phenotypes most interestingly a reduction in virulence (283). One of the primary events associated with hypoviral infection is the reduction of accumulation of the cpg-1 gene encoded G-alpha subunit protein (267). The cpg-1 gene encodes a subunit that instead of being stimulatory to the production of cAMP, likely functions as a inhibitor of adenylate cyclase activity and cAMP content in this fungus. Cosuppression of the gpa-1 gene or hypoviral infection yielded several fold increases in cellular cAMP content (284). Consistently, pharmacological elevation of cAMP levels triggered transcription of a set of genes also induced by hypovirus infection and cpg-1 cosuppression. These results implicate cAMP as a major mediator of the hypovirus infection phenotype. Mutations that activate PKA such as mutations in the regulatory subunit of PKA (a PKA inhibitor protein) or activated alleles of specific G-alpha proteins tend to cause phenotypes of opposite character to those of inactivating mutations. For example, in U. maydis, ubc1 encoding the regulatory subunit of PKA, is an epistatic suppressor to the uac1 mutation (275). Likewise an activated allele of gpa3 generates several phenotypes similar to a ubc1 mutant but is hypostatic to uac1 (277, 285). However, mutations activating the PKA pathway still tend to be detrimental to virulence. Null ubc1 mutants are able to colonize maize leaves but are unable

200


to induce gall formation (286). Mutants with intermediate activation of PKA caused further progression toward the wild type infection (287). In M. grisea, the sum1 mutation (rPKA) restored the growth defects of a mac1 mutant but did not restore full pathogenicity (288). These results indicate that a delicate balance of PKA activation must be maintained for progression through the various phases of the infection cycle. Perturbation in one or another direction, activation or inactivation, leads to detrimental effects on the fungus. Mitogen activated protein kinases have often been encountered in phytopathogenic fungi as required for morphogenesis and for full virulence. This topic has been reviewed in some detail (289). S. cerevisiae has five functional (and partially overlapping) MAP kinase cascades. These are involved in mating, filamentation, cell integrity, high osmotic growth stress response, and ascospore formation. The environmental triggers activating these pathways are commensurate with their function, eg. pheromone for mating. In the genomes of plant pathogenic fungi (e.g., in M. grisea), and filamentous fungi in general for that matter (e.g., in N. crassa), often only three MAP kinase enzymes appear to be present. These fall into three families relative to yeast: pheromone responsive (Fus3/Kss1), osmoregulation (Hog1) and cell integrity (Slt2). Mutations in these MAP kinases tend to cause differential defects in pathogenicity. For example, in M. grisea PMK1 (pheromone response) and MPS1 (cell integrity) are both required for disease on unwounded leaves but mps1 mutants can colonize wounded leaves while pmk1 mutants cannot. In this fungus the third MAP kinase, OSM1, is not required for virulence under laboratory conditions. Pheromone responsive MAP kinase genes have been repeatedly found to be critical for full virulence in the fungi. As noted (289), FUS3/KSS1 homologs from a number of phytopathogenic fungi have been isolated from C. heterostrophus, B. cinerea, F. oxysporum and more recently from Claviceps pupurea (290) and F. graminearum (291). MAP kinase genes from other families have also been shown to be important for virulence in plant pathogenic fungi but tend to be more variable in their effect than the Fus3 family. A F. graminearum mgv1 mutant, a SLT2 family member, was dramatically reduced in virulence although some disease was still encountered (292). Interestingly, this mutant also had a defect in mating in that it became functionally heterothallic. In U. maydis, formation of swollen appressorium-like structures and their production of invading hyphae which penetrate epidermal cells, appear to be distinct steps in the infection process. A mutant strain defective in Kpp6 activity, a b mating-type gene-regulated MAP kinase, has recently been constructed, which is able to produce appressoria but is unable to penetrate plant cells (293). Microscopic observations of plant surfaces after inoculation with compatible strains both carrying an inactivated mutant allele, kpp6T355A,Y357F, showed appressorium formation. However, from the majority of those appressoria only short filaments that failed to penetrate plant cells emerged.

Cross talk between the cAMP and MAP kinase cascade signaling pathways has been well documented (294, 273). In M. grisea, appressorium development involves both cAMP and MAP kinase pathways and mutants affected in either are appressorium-defective. The pmk1 MAP kinase mutant still responds to cAMP for the early stage of germ tube tip deformation during appressorium formation (295). This suggests that the MAP kinase acts downstream of PKA for this function. In U. maydis the interaction of the cAMP and MAP kinase pathways was made evident by the finding that a number of suppressors of the filamentous phenotype of an adenylate cyclase (uac1) mutant were members of the pheromone responsive MAP kinase cascade (296, 297, 298, 299). Also in U. maydis, it has recently been demonstrated that the differential phosphorylation of Prf1 by the pheromone responsive MAPK and the cAMP dependent protein kinase regulates the activity of this transcription factor toward its various

201


promoter targets including the a and b mating-type genes (300). Induction of the a mating-type genes requires the PKA phosphorylation sites while induction of the b genes requires both the PKA and MAPK phosphorylation sites. Role of calcium Calcium, among other signaling compounds, is well documented as playing a role in fungal spore germination (reviewed in 29). In M. grisea, calcium/calmodulin-dependent signaling systems are involved in appressorium formation. Several calcium modulators and calmodulin antagonists inhibited appressorium formation and there was an inhibition of appressorium formation by EGTA, a calcium chelator, which was restored by the addition of exogenous CaCl2. Neomycin, a phospholipase C inhibitor, specifically inhibited appressorium formation suggesting that calcium/calmodulin-dependent signaling is involved in appressorium formation. Interference with Ca2+ signaling did not affect conidium germination in this fungus. In contrast, in the grape pathogen Phyllosticta ampelicida, following contact and attachment to a substratum, pycniospores required Ca2+ to trigger germination (301). After germination Ca2+ was again required for appressorium formation. Similar effects on both germination and appressorium formation were earlier noted in Colletotrichum trifolii (302). Consistent with these observations, antisense expression of the calmodulin gene in C. trifolii blocked appressorium formation and pathogenicity on alfalfa plants (303). In U. maydis, one of the mycovirus encoded killer toxins, KP4, inhibits fungal growth by blocking calcium uptake (304). The U. maydis calcineurin gene ucn1, encoding a Ca2+ calmodulin-dependent protein phosphatase, appears to be involved in the direct or indirect regulation of some PKA substrates (Egan and Gold, unpublished). This is supported by the finding that a ucn1 mutation causes a phenotype reminiscent to activation of the PKA pathway but is hypostatic to a mutation in the catalytic subunit of PKA (adr1). NEW CLASSES OF VIRULENCE AND PATHOGENICITY GENES Global regulators of genes known to contribute to pathogenicity and/or virulence have recently been described. These certainly will help to gain insight into the role the genes they are regulating play. We already mentioned genes that are catabolite-repressed such as many of the lytic enzymes but are expressed on complex media or under nitrogen limiting conditions, mostly simulating in planta conditions. As discussed, the C. carbonum regulators SNF1 and CreA serve these roles (44, 45). In C. fulvum, a nitrogen response factor 1 gene Nrf1, encodes a putative transcription factor with a putative zinc finger DNA-binding domain. This gene regulates the expression of avirulence gene Avr9 known to be induced under nitrogen-limiting in vitro conditions and in planta. However, an Nrf1-deletion mutant did not affect virulence in susceptible cultivars or avirulence on tomato plants containing the functional Cf-9 resistance gene (305). We described the pathogenicity gene MPG1 from M. grisea earlier (35). Two mutants, NPR1 and NPR2 (for nitrogen pathogenicity regulation) seemed impaired in the utilization of many nitrogen sources and for pathogenicity including the regulation of MPG1 (306). In a different class of very intriguing regulators, a GAL4-like, zinc-finger transcription factor, CLTA1, was found in Colletotrichum lindemuthianum. This protein is involved in the switch between biotrophic and necrotrophic phases of the infection process on common bean. The mutant seemed to be halted before the macerating, necrotrophic stage of the infection and in doing so produced only HR-like necrotic lesions (307). A similar effect was obtained with a

202


REMI mutant of the maize anthracnose leaf blight fungus, Colletotrichum graminicola. However, in this case the corresponding CPR1 gene encoded a putative signal peptidase. Interestingly, the REMI insertion was in the 3’-end of the gene and only reduced transcription; a complete deletion of this essential gene could not be obtained. It was suggested that the mutant did not secrete sufficient quantities of degradative enzymes to support the transition to necrotrophic growth (308). In a direct search for U. maydis genes whose expression is changed upon interaction with the maize host, a differential display technique revealed several mig-genes (maize-induced genes). Mig1 was strongly expressed only shortly after entering host tissues up to spore formation and transcription was controlled by both negative and positive promoter elements. The small, highly-charged gene product had a 5’-secretion signal and was shown to be secreted but no other homologies were found to known proteins. Whatever its function, a deletion mutant did not seem to be affected in in vitro growth, mating or pathogenicity (13). Very similar results were found for a Mig2 gene cluster whose members Mig2-1 to Mig2-5, shared homologous N-terminal secretion signals and 3’-parts. The C-terminal parts showed a conserved pattern of 8 cysteine residues (not related to hydrophobins) but reminiscent to known fungal avirulence and oomycete elicitin genes although no homologies were found to known proteins. However, deletion of these genes, Mig2-2 individually or the complete cluster, did not affect virulence under the conditions tested (14). Interestingly, a histone deacetylase, Hda1, was found to be a regulator (repressor) of Mig1 (309) as well as of the endoglucanase gene, egl1, discussed earlier, and the teliospore-specific gene, Ssp1, most likely involved in the metabolism of storage lipids (264). An Hda1-deletion mutant was still pathogenic to some extent but did not allow the maturation to teliospores in corn tissues (263). This was also seen when another regulator of U. maydis, Rum1, was deleted (310). Rum1 represses the endoglucanase gene, egl1, but also a number of other genes known to be regulated by the b mating-type and pathogenicity locus (310). Another U. maydis gene, sql2, with similarities to the yeast guanyl nucleotide exchange factor CDC25 known to act on Ras proteins, seems to regulate filamentous growth and deletion causes a major reduction in virulence on corn plants (311). It will be very illustrative to elucidate how such “global” regulators interfere with the host interaction and the disease process. In a very recent screen using REMI mutagenesis with an enhancer trap linked to a Green Fluorescent Protein reporter, several genes whose expression was affected during the biotrophic phase in host tissue, were discovered. However, upon deleting these, no obvious change in virulence could be detected (312). CPS1, a Cochliobolus heterostrophus gene encodes a protein with two AMP-binding domains which might be an adenylate-forming enzyme involved in biosynthesis of nonribosomal peptides or polyketide-peptide hybrids. A deletion mutant was reduced in virulence reducing lesion size up to 60% and this occurred in both race T and race O of C. heterostrophus on maize, of C. victoriae on oats, and of Gibberella zeae on wheat. The authors indicated that CPS1 belonged to a novel class of genes controlling general virulence functions in plant pathogenic ascomycete fungi (313). Cyclophilins are known to be involved in a wide

variety of cellular processes, including the response to environmental stresses, cell cycle control, the regulation of calcium signaling, and the control of transcriptional repression, and have recently been implicated as virulence factors in fungi. Deletion of the CYP1 gene in M. grisea resulted in mutants that were impaired in plant infection. Appressoria did not develop full turgor and their development showed a very acute sensitivity to cyclosporin A, a known immunosuppressive drug. This sensitivity occurred in a cyclophilin-dependent manner, suggesting a role for calcineurin during regulation (314). In the human pathogenic fungus

203


Cryptococcus neoformans, two cyclophilin A homologs, CPA1 and CPA2, were also found to be virulence factors (315). A recent report introduced another phenomenon: virulence of hybrid species of the wood rotting basidiomycete Heterobasidion annosum, is controlled by their mitochondrial genomes indicating that cooperation between genes of organelles may influence the phenotype of hybrid phytopathogens (316). Importance of organelles in virulence is also suggested by a study in the vascular wilt pathogen F. oxysporum. FOW1 encodes a mitochondrial carrier protein with strong similarity to mitochondrial carrier proteins of yeast and is required specifically for colonization in the plant tissue by this fungus (317). GENOMICS AND OTHER “-OMICS”: NEW GENETIC PARADIGMS Automated sequence technologies and advances in bioinformatics have rendered the task of sequencing entire genomes of organisms with fungal-size genomes within technical and cost reach and are enabling a wide variety of new discoveries including new genes, metabolic pathways and insights into the mechanisms of microbial pathogenesis (318, 319, 320). With these exciting advances, the means to generate sequence and functional information and the tools to use that information to understand the biology of fungi that cause plant disease are now available. Historically, the molecular basis for interactions between plants and microbes was studied using a gene-by-gene approach and naturally occurring host resistance was a major control approach. Now, coupled with the increasing availability of plant host genomic sequences, structural and functional analysis of fungal plant pathogens will increase the speed of identification of genes involved in host-pathogen interaction and will allow genome wide approaches to understanding the role of a gene or pathway in interactions with the plant. Some genes will potentially be useful as sources of pathogen-derived resistance as has already been demonstrated for many viral diseases. Fungal whole genome sequencing projects have been a topic of wide spread interest and heated discussion for several years. However, due primarily to lack of public funding, until recently there was no example of a filamentous fungus with a complete genomic sequence in the public databases. Of course, the budding yeast S. cerevisiae was the first eukaryotic genome fully sequenced and made freely available (321). And this saprophytic fungus remains the paradigm upon which much genomic work is modeled. The 1st release of the Neurospora crassa genome sequence by the Whitehead Institute Center for Genomic Research (WICGR) in February 2001, marked the beginnings of a dramatic change in the availability of fungal genomic data. This was the first product of the WICGR Fungal Genome Initiative (FGI). In June 2002, the 1st assembly of the genome of M. grisea, the first plant pathogenic fungus, was released by the WICGR. Since this time the WICGR-FGI has released the genome assemblies of two additional plant pathogenic fungi, Fusarium graminearum and Ustilago maydis, in 2003. Additional plant pathogenic fungal species on the wish list in the FGI’s second white paper include F. verticillioides, F. solani, F. oxysporum, Mycosphaerella graminicola, Blumeria graminis, Sclerotinia sclerotiorum, Holleya sinecauda, Microbotryum violaceum, Puccinia triticina, and P. striiformis. The FGI is not the only source of publicly released plant pathogenic fungal sequence data, but it has had the most impact by far. In an academic collaboration, TIGR will be sequencing Aspergillus flavus (Gary Payne, personal communication). The US Department of Energy-Joint Genome Institute, in partnership with the USDA, is committed to sequencing two species of the soybean leaf rust, Phakopsora pachyrhizi and P. meibomiae, by the end of 2004. The NSF, also in partnership with the USDA, has launched an initiative that

204


will likely lead to the sequencing of several more plant pathogenic fungal genomes, hopefully including a representative of the cereal rusts. Prior to public release of data, several fungal genomic projects were well advanced in the private sector with or without collaboration with public sector researchers. However, many of these projects have been discontinued as the focus of the companies shifted. For example Paradigm Genetics announced in June, 2000 the mutagenesis of all genes in M. grisea and 70% of those in Septoria tritici, employing Paradigm's proprietary TAG-KO™ technology. In Ashbya gossypii, a pathogen of cotton and citrus fruits, a highly efficient homologous recombination system exists allowing a short sequence PCR based gene disruption methodology similar to that used in S. cerevisiae (322). Thus far, production of mutants in genes of the signal transduction machinery with effects on cell morphogenesis, have been a primary focus (322, 323, 324, 325, 326). At the Torrey Mesa Research Institute in San Diego, CA, C. heterostrophus was the focus of genome sequencing. Projects to delete all the genes of this fungus were underway prior to the closing of the facility in 2003 (Gillian Turgeon, personal communication). Expressed Sequence Tags (ESTs) are another genomic resource that has enjoyed exponential growth for plant pathogenic fungi in recent years. They include databases at COGEME harbouring 36,022 ESTs from 12 different fungi and 2 oomycetes (as of December 2003; http://cogeme.ex.ac.uk/; 327), at GenBank and in several local but public repositories. Several recent contributions were from U. maydis (328), the wheat leaf rust, P. triticina (18, Hu and Bakkeren, in preparation) and the tree pathogen, Heterobasidion annosum (329). Very valuable for comparative analyses are additional genomic resources from related nonpathogenic fungi, such as EST collections from the wood decaying fungus, Schizophyllum commune 330, and two edible mushrooms, Agaricus bisporus 331 and Pleurotus ostreatus 332, and complete genome sequences from the wood rot fungi, Phanerochyte chrysosporium and Coprinus cinereus, the human pathogen and tree saprobe, Cryptococcus neoformans, and ascomycetes A. fumigatus, A. nidulans, C. albicans, Coccidioides posadasii and S. pombe (reviewed in 320). Proteomics is an extension to genomics and is the genome-wide study of proteins. These techniques are beginning to be applied to plant pathogens and in the post-genomic era they will become increasingly important. Proteomics has three main areas: (A) protein micro-characterization for large-scale identification of proteins and their post-translational modifications; (B) 'differential display' proteomics for comparison of protein levels with potential application in a wide range of diseases; and (C) studies of protein-protein interactions using techniques such as mass spectrometry or the yeast two-hybrid system (reviewed in (333). Protein-protein interactions control many cellular activities and are a potential new target for control measures. In yeast, nearly all proteins have been tested for interaction in a pair-wise manner (334). A method, potentially ideally suited for fungi, is the reverse two-hybrid assay (335). This allows positive selection for the identification of the rare molecules that disrupt protein-protein interactions. Green-fluorescent protein (GFP) can also be used as an indicator of protein-protein interaction in living cells. This application exploits a non-destructive spectroscopic method for measuring molecular interactions called fluorescence resonance energy transfer (FRET) that occurs between two differing GFP molecules or between GFP and a secondary fluorophore. Several classes of GFP mutants with appropriate excitation and emission peaks for FRET have been described (336). A FRET-based method for visualizing Ca2+ -mediated protein interactions (337) and cAMP-induced phosphorylation in vivo (338) has been developed. FRET has potential for the study of protein-protein interaction at a genome-

205

http://cogeme.ex.ac.uk/


wide level (339) as an alternative to the yeast two-hybrid system and should be possible in the organism of interest and not limited to interaction in the yeast nucleus. As for global approaches, the biggest challenge will be to assign functions to predicted genes, analyze their transcription and translation and integrate all these basic “-omics” resources including emerging data on changes in metabolism (metabolomics). Systematic gene ‘knock-out’ or ‘knock-down’ approaches with subsequent mutant analysis have been proposed and are being undertaken in several model fungi (340), but an important focus will be on sets of genes/proteins that are involved in pathogenicity and virulence in its broadest sense (318, 341, 320). Current methods include transcript profiling either by using differential display, serial analysis of gene expression (SAGE), as for the barley mildew pathogen Blumeria graminis (19), and EST- or gene microarrays. Such surveys will reveal genes potentially involved in the interaction with plants and will need further functional testing. Developing functional (forward genetic) screens, e.g., by using reporter genes and/or tagged mutations (342, 312), would be advantageous but as a trade-off are likely not global in saturating or comprehensively screening the whole genome. HOW TO EXPLOIT ALL THESE RESOURCES? Apart from all the fascinating insights into the pathogenicity pathways of fungi, how can we try to harness this knowledge to the benefit of agriculture in its broadest sense? There are a myriad of fungal plant diseases that threaten the world food supply, some of them acute and sadly often in countries that are economically less well-off. However, it is dangerous to become complacent about the current state of many ‘western’ agricultural production systems. Over the last 60 years breeders have done a marvellous job at staying abreast of the emergence of many new fungal races, and the chemical industry has similarly achieved tremendous impact. In combination with more recent Integrated Pest Management (IPM) practices and Remote Sensing approaches, agronomists still seem to have the upper hand in the war against (fungal) pathogens. There are nevertheless some clouds on the horizon. Consumers are concerned about pesticide use and want a sustainable agriculture in a clean environment. They are also concerned about biotechnological advances that would necessitate the use of transgenic crops (see below) even though such technologies would be promising. With the enormous volume of travel and commerce around the globe, there is continuous introduction of new fungal species and races that can spread over vast geographic areas leading to potential epidemics. This danger can be aggravated because of changes in current climates, larger monocultures of “elite” germplasm and loss of germplasm resources with available resistances suitable for introgression into commercial crops. A lot of emphasis has been given to boosting host resistance by studying and manipulating the plant genes that are involved in the myriad of defense reactions (343). Ingenious two-component triggers converting race-specific resistance to a more broad-spectrum resistance by driving avirulence genes by a pathogen-inducible promoter (136), or focussing on avirulence genes that are more persistent in the population and might serve essential functions or even dictate host range in combination with their cognate resistance genes, are major areas of investment at the moment. Engineering HR responses in plants seems an obvious choice but it was recently suggested that for some fungi, such as the HST-producing necrotrophs, this might actually lead to enhanced susceptibility (176). Interestingly, some necrotrophic fungi induce (HR-like) cell death in susceptible interactions (discussed above); by expressing animal genes that negatively regulate apoptosis in plants, increased resistance at least against some

206


necrotrophic fungi can be obtained (344). However, understanding these disease processes from the point of the pathogens might give us insight into target genes essential in the life cycle of fungi upon and/or during the infection process (345). Envisaged is the use and application of ‘smart, target-specific fungicides’ or production of transgenic plants expressing such compounds. That is, such compounds need to be very specific and target proteins of which we know the mode of action in the fungal cell, and need to be harmless to other organisms including humans. Ideally, such proteins are essential in the pathogen’s life cycle as to provide durable resistance. Genomic and proteomic approaches will likely reveal such essential genes that could be targeted. Some examples could be: inhibition of ABC transporters to prevent fungi from exporting fungicides or plant phytoalexins or toxins. Targeting metabolite transporters in obligate biotrophic fungi such as amino acid- and sugar transporters which seem to be important in the rust, Uromyces fabae (9). Interference with signal transduction cascades such as the MAP kinase or cAMP pathways which seem to impinge on several crucial steps. Spore germination and appressorium formation (which involve such signaling pathways) seem obvious stages to be blocked because they would prevent any harm to the host as opposed to preventing sporulation which might have allowed systemic spread in the host. CONCLUSIONS This review has outlined that many factors contribute to a microbe actually causing disease and why this is a rare outcome in nature. Nevertheless, many diseases affect our crops and this is exacerbated by enormous monocultures with an increasingly narrow genetic base. The study of plant-microbe interactions brings to light a delicate balance of mind-boggling complexity and creativity through evolution from both microbe and host. Given the diversity among microbes and their fluidity of their genomes through mutation, genetic rearrangements, (para/sexual) hybridization, horizontal gene flow etc., we cannot expect to find a ‘silver bullet’ and win the fight against plant pathogens. However, through biotechnology in combination with classical breeding, diversification of the genetic base, IPM and good agronomic practices, we should be able to push this balance in favor of healthier crops. The speed with which discoveries of microbial genes involved in the interaction with their hosts are made in this ‘-omics era’, holds great promise. We should certainly draw parallels with studies of mammalian (fungal) pathogens. For example, this has been very revealing in the study of bacterial effector proteins and their delivery through specific pili, causing disease in both plants and mammals. In general, targeting pathogens is more prevalent in medical and veterinarian research because manipulating hosts is limited to boosting immune responses. Mammalian fungal pathogens have only recently become major research objects because of an increase in immunocompromised individuals in which normally quiescent fungi cause (lethal) symptoms. Certain plant pathogens have become model systems but the study of many fungi causing major diseases in our crop plants is lagging severely behind or is non-existent. Our next challenge is to find creative ways to harness the accumulating knowledge of the function of all these pathogen genes and turn that into novel ways for crop protection. ACKNOWLEDGEMENTS The authors apologize to colleagues whose pathosystems and data were not discussed or cited; this review was not meant to be comprehensive. We wish to thank Drs. Maria Garcia-Pedrajas and Steven Klosterman for critical reading of the manuscript and comments.

207


REFERENCES 1 Latijnhouwers, M., de Wit, P. J. G. M. and Govers, F. (2003) Trends Microbiol 11, 462-469. 2 Stahl, E. A. and Bishop, J. G. (2000) Curr Opin Plant Biol 3, 299-304. 3 Thrall, P. H. and Burdon, J. J. (2003) Science 299, 1735-7. 4 Knogge, W. (1998) Curr Opin Plant Biol 1, 324-8. 5 Idnurm, A. and Howlett, B. J. (2001) Mol Plant Pathol 2, 241-255. 6 Voegele, R. T., Struck, C., Hahn, M. and Mendgen, K. (2001) Proc Natl Acad Sci U S A 98,

8133-8138. 7 Struck, C., Ernst, M. and Hahn, M. (2002) Mol Plant Pathol 3, 23-30. 8 Mendgen, K. and Hahn, M. (2002) Trends Plant Sci 7, 352-356. 9 Struck, C., Voegele, R. T., Hahn, M. and Mendgen, K. (2003) in Biology of molecular

plant-microbe interactions, Proceedings of the International Society of Molecular Plant-Microbe Interactions, Vol. 4 (Lugtenberg, B., Tikhonovich, I. and Provorov, N., eds.) (in press).

10 Panstruga, R. (2003) Curr Opin Plant Biol 6, 320-6. 11 Kahmann, R., Basse, C. and Feldbrugge, M. (1999) Curr Opin Microbiol 2, 647-50. 12 Martinez-Espinoza, A. D., Garcia-Pedrajas, M. D. and Gold, S. E. (2002) Fungal Genet Biol

35, 1-20. 13 Basse, C. W., Stumpferl, S. and Kahmann, R. (2000) Mol Cell Biol 20, 329-39. 14 Basse, C. W., Kolb, S. and Kahmann, R. (2002) Mol Microbiol 43, 75-93. 15 Hahn, M., Neef, U., Struck, C., Gottfert, M. and Mendgen, K. (1997) Mol Plant Microbe

Interact 10, 438-45. 16 Struck, C., Siebels, C., Rommel, O., Wernitz, M. and Hahn, M. (1998) Mol Plant Microbe

Interact 11, 458-65. 17 Sohn, J., Voegele, R. T., Mendgen, K. and Hahn, M. (2000) Mol Plant Microbe Interact 13,

629-36. 18 Thara, V. K., Fellers, J. P. and Zhou, J. M. (2003) Molecular Plant Pathology 4, 51-56. 19 Thomas, S. W., Rasmussen, S. W., Glaring, M. A., Rouster, J. A., Christiansen, S. K. and

Oliver, R. P. (2001) Fungal Genet Biol 33, 195-211. 20 Zhang, Z. and Gurr, S. J. (2001) Gene 266, 57-65. 21 Vogel, J. P., Raab, T. K., Schiff, C. and Somerville, S. C. (2002) Plant Cell 14, 2095-106. 22 Ruiz-Herrera, J., Leon-Ramirez, C., Cabrera-Ponce, J. L., Martinez-Espinoza, A. D. and

Herrera-Estrella, L. (1999) Mol Gen Genet 262, 468-72. 23 Chaure, P., Gurr, S. J. and Spanu, P. (2000) Nat Biotechnol 18, 205-7. 24 Schillberg, S., Tiburzy, R. and Fischer, R. (2000) Mol Gen Genet 262, 911-5. 25 Arabi, M. I. E. and Jawhar, M. (2002) Phytopath Z 150, 305-307. 26 Fasters, M. K., Daniels, U. and Moerschbacher, B. M. (1993) Physiol Mol Plant Pathol 42,

259-265. 27 Bindslev, L., Kershaw, M. J., Talbot, N. J. and Oliver, R. P. (2001) Mol Plant Microbe

Interact 14, 1368-1375. 28 Kolattukudy, P. E., Rogers, L. M., Li, D. X., Hwang, C. S. and Flaishman, M. A. (1995)

Proc Natl Acad Sci U S A 92, 4080-4087. 29 Osherov, N. and May, G. S. (2001) FEMS Microbiol Lett 199, 153-60. 30 Hamer, J. E., Howard, R. J., Chumley, F. G. and Valent, B. (1988) Science 239, 288-290. 31 Talbot, N. J. (1995) Trends Microbiol 3, 9-16.

208


32 Nielsen, K. A., Nicholson, R. L., Carver, T. L. W., Kunoh, H. and Oliver, R. P. (2000) Physiol Mol Plant Pathol 56, 63-70.

33 Ruan, Y., Kotraiah, V. and Straney, D. C. (1995) Mol Plant Microbe Interact 8, 929-938. 34 Straney, D., Khan, R., Tan, R. and Bagga, S. (2002) Adv Exp Med Biol 505, 9-22. 35 Talbot, N. J., Ebbole, D. J. and Hamer, J. E. (1993) Plant Cell 5, 1575-1590. 36 Mellersh, D. G. and Heath, M. C. (2001) Plant Cell 13, 413-424. 37 Heath, M. C. (2002) Can. J. Plant Pathol. 24, 259-264. 38 Hostetter, M. K. (2000) Curr Opin Microbiol 3, 344-8. 39 Carpita, N. C. and Gibeaut, D. M. (1993) Plant J 3, 1-30. 40 Carpita, N. C. (1996) Annu Rev Plant Physiol Plant Mol Biol 47, 445-476. 41 Pauly, M., Albersheim, P., Darvill, A. and York, W. S. (1999) Plant J 20, 629-639. 42 Cano Canchola, C., Acevedo, L., Ponce Noyola, P., Flores Martinez, A., Flores Carreon, A.

and Leal Morales, C. A. (2000) Fungal Genet Biol 29, 145-151. 43 Scott-Craig, J. S., Cheng, Y. Q., Cervone, F., De Lorenzo, G., Pitkin, J. W. and Walton, J.

D. (1998) Appl Environ Microbiol 64, 1497-503. 44 Tonukari, N. J., Scott-Craig, J. S. and Walton, J. D. (2000) Plant Cell 12, 237-48. 45 Tonukari, N. J., Scott-Craig, J. S. and Walton, J. D. (2003) DNA Seq 14, 103-7. 46 Ospina-Giraldo, M. D., Mullins, E. and Kang, S. (2003) Curr Genet 44, 49-57. 47 Schafer, W. (1993) Trends Microbiol 1, 69-71. 48 Li, D., Ashby, A. M. and Johnstone, K. (2003) Mol Plant Microbe Interact 16, 545-52. 49 Rogers, L. M., Flaishman, M. A. and Kolattukudy, P. E. (1994) Plant Cell 6, 935-45. 50 Kan, J. A. L. V., Klooster, J. W. V. T., Wagemakers, C. A. M., Dees, D. C. T. and Vlugt

Bergmans, C. J. B. V. D. (1997) Mol Plant Microbe Interact 10, 30-38. 51 Sweigard, J. A., Chumley, F. G. and Valent, B. (1992) Mol Gen Genet 232, 183-190. 52 Lev, S. and Horwitz, B. A. (2003) Plant Cell 15, 835-44. 53 Sposato, P., Ahn, J. H. and Walton, J. D. (1995) Mol Plant Microbe Interact 8, 602-9. 54 Schauwecker, F., Wanner, G. and Kahmann, R. (1995) Biol Chem Hoppe-Seyler 376, 617-

25. 55 Sunna, A. and Antranikian, G. (1997) Crit Rev Biotechnol 17, 39-67. 56 Apel Birkhold, P. C. and Walton, J. D. (1996) Appl Environ Microbiol 62, 4129-4135. 57 Wu, S. C., Ham, K. S., Darvill, A. G. and Albersheim, P. (1997) Mol Plant Microbe Interact

10, 700-708. 58 Bucheli, P., Doares, S. H., Albersheim, P. and Darvill, A. (1990) Physiol Mol Plant Pathol

36, 159-173. 59 Gomez Gomez, E., Ruiz Roldan, M. C., Di Pietro, A., Roncero, M. I. G. and Hera, C.

(2002) Fungal Genet Biol 35, 213-222. 60 Rogers, L. M., Kim, Y. K., Guo, W., Gonzalez-Candelas, L., Li, D. and Kolattukudy, P. E.

(2000) Proc Natl Acad Sci U S A 97, 9813-8. 61 Valette-Collet, O., Cimerman, A., Reignault, P., Levis, C. and Boccara, M. (2003) Mol

Plant Microbe Interact 16, 360-7. 62 Oeser, B., Heidrich, P. M., Muller, U., Tudzynski, P. and Tenberge, K. B. (2002) Fungal

Genet Biol 36, 176-86. 63 ten Have, A., Mulder, W., Visser, J. and van Kan, J. A. (1998) Mol Plant Microbe Interact

11, 1009-16. 64 Poinssot, B., Vandelle, E., Bentejac, M., Adrian, M., Levis, C., Brygoo, Y., Garin, J.,

Sicilia, F., Coutos-Thevenot, P. and Pugin, A. (2003) Mol Plant Microbe Interact 16, 553-64.

209


65 Isshiki, A., Akimitsu, K., Yamamoto, M. and Yamamoto, H. (2001) Mol Plant Microbe Interact 14, 749-57.

66 Shieh, M. T., Brown, R. L., Whitehead, M. P., Cary, J. W., Cotty, P. J., Cleveland, T. E. and Dean, R. A. (1997) Appl Environ Microbiol 63, 3548-52.

67 Garcia-Maceira, F. I., Di Pietro, A., Huertas-Gonzalez, M. D., Ruiz-Roldan, M. C. and Roncero, M. I. (2001) Appl Environ Microbiol 67, 2191-6.

68 Di Pietro, A. and Roncero, M. I. (1998) Mol Plant Microbe Interact 11, 91-8. 69 Gao, S., Choi, G. H., Shain, L. and Nuss, D. L. (1996) Appl Environ Microbiol 62, 1984-90. 70 Scott-Craig, J. S., Panaccione, D. G., Cervone, F. and Walton, J. D. (1990) Plant Cell 2,

1191-200. 71 Centis, S., Guillas, I., Sejalon, N., Esquerre Tugaye, M. T. and Dumas, B. (1997) Mol Plant

Microbe Interact 10, 769-775. 72 Herbert, C., Jacquet, C., Borel, C., Esquerre Tugaye, M. T. and Dumas, B. (2002) J Biol

Chem 277, 29125-29131. 73 Rauscher, M., Mendgen, K. and Deising, H. (1995) Exp Mycol 19, 26-34. 74 Carlile, A. J., Bindschedler, L. V., Bailey, A. M., Bowyer, P., Clarkson, J. M. and Cooper,

R. M. (2000) Mol Plant Microbe Interact 13, 538-50. 75 Prins, T. W., Tudzynski, P., Von Tiedemann, A., Tudzynski, B., ten Have, A., Hansen, M.

E., Tenberge, K. B. and van Kan, J. (2000) in Fungal pathology (Kronstad, J. W. ed.) pp. 33-64, Kluwer Academic Publishers, Dordrecht, Boston.

76 Prusky, D., Mcevoy, J. L., Leverentz, B. and Conway, W. S. (2001) Mol Plant Microbe Interact 14, 1105-1113.

77 Drori, N., Kramer-Haimovich, H., Rollins, J., Dinoor, A., Okon, Y., Pines, O. and Prusky, D. (2003) Appl Environ Microbiol 69, 3258-62.

78 Rollins, J. A. and Dickman, M. B. (2001) Appl Environ Microbiol 67, 75-81. 79 Cotton, P., Kasza, Z., Bruel, C., Rascle, C. and Fevre, M. (2003) FEMS Microbiology

Letters 227, 163-169. 80 Rollins, J. A. (2003) Mol Plant Microbe Interact 16, 785-95. 81 Prusky, D. and Yakoby, N. (2003) Mol Plant Pathol 4, 509-516. 82 Lang, C. and Dornenburg, H. (2000) Appl Microbiol Biotechnol 53, 366-75. 83 De Lorenzo, G. and Ferrari, S. (2002) Curr Opin Plant Biol 5, 295-9. 84 Powell, A. L., van Kan, J., ten Have, A., Visser, J., Greve, L. C., Bennett, A. B. and

Labavitch, J. M. (2000) Mol Plant Microbe Interact 13, 942-50. 85 Bechinger, C., Giebel, K. F., Schnell, M., Leiderer, P., Deising, H. B. and Bastmeyer, M.

(1999) Science 285, 1896-9. 86 Read, N. D., Kellock, L. J., Collins, T. J. and Gundlach, A. M. (1997) Planta 202, 163-170. 87 Mendgen, K., Hahn, M. and Deising, H. (1996) Annu Rev Phytopathol 34, 367-386. 88 Dezwaan, T. M., Carroll, A. M., Valent, B. and Sweigard, J. A. (1999) Plant Cell 11, 2013-

2030. 89 Kamakura, T., Yamaguchi, S., Saitoh, K., Teraoka, T. and Yamaguchi, I. (2002) Mol Plant

Microbe Interact 15, 437-444. 90 Clergeot, P. H., Gourgues, M., Cots, J., Laurans, F., Latorse, M. P., Pepin, R., Tharreau, D.,

Notteghem, J. L. and Lebrun, M. H. (2001) Proc Natl Acad Sci U S A 98, 6963-8. 91 De Jong, J. C., McCormack, B. J., Smirnoff, N. and Talbot, N. J. (1997) Nature 389, 244-

245. 92 Foster, A. J., Jenkinson, J. M. and Talbot, N. J. (2003) EMBO J 22, 225-35. 93 Balhadere, P. V. and Talbot, N. J. (2001) Plant Cell 13, 1987-2004.

210


94 Kimura, A., Takano, Y., Furusawa, I. and Okuno, T. (2001) Plant Cell 13, 1945-57. 95 Hwang, C. S., Flaishman, M. A. and Kolattukudy, P. E. (1995) Plant Cell 7, 183-193. 96 Roncero, C. (2002) Curr Genet 41, 367-78. 97 Madrid, M. P., Di Pietro, A. and Roncero, M. I. (2003) Mol Microbiol 47, 257-66. 98 Soulie, M. C., Piffeteau, A., Choquer, M., Boccara, M. and Vidal-Cros, A. (2003) Fungal

Genet Biol 40, 38-46. 99 Narasimhan, M. L., Lee, H., Damsz, B., Singh, N. K., Ibeas, J. I., Matsumoto, T. K.,

Woloshuk, C. P. and Bressan, R. A. (2003) Plant J 36, 390-400. 100 Wessels, J. G. H. (2000) Mycologist 14, 153-159. 101 Whiteford, J. R. and Spanu, P. D. (2002) Mol. Plant Pathol. 3, 391-400. 102 Wessels, J. G. H. (1994) Annu. Rev. Phytopathol. 32, 413-437. 103 Templeton, M. D., Rikkerink, E. H. A. and Beever, R. E. (1994) Mol Plant Microbe

Interact 7, 320-325. 104 Bowden, C. G., Smalley, E., Guries, R. P., Hubbes, M., Temple, B. and Horgen, P. A.

(1996) Mol Plant Microbe Interact 9, 556-564. 105 Temple, B. and Horgen, P. A. (2000) Mycologia 92, 1-9. 106 Segers, G. C. and Nuss, D. L. (2003) Fungal Genet Biol 38, 198-208. 107 Spanu, P. (1997) Gene 193, 89-96. 108 Tsuji, G., Takeda, T., Furusawa, I., Horino, O. and Kubo, Y. (1997) Pestic Biochem

Physiol 57, 211-219. 109 Takano, Y., Kubo, Y., Kawamura, C., Tsuge, T. and Furusawa, I. (1997) Fungal Genet

Biol 21, 131-140. 110 Kim, J. C., Min, J. Y., Kim, H. T., Kim, B. S., Kim, Y. S., Kim, B. T., Yu, S. H.,

Yamaguchi, I. and Cho, K. Y. (1998) Pestic Biochem Physiol 62, 102-112. 111 Henson, J. M., Butler, M. J. and Day, A. W. (1999) Annu Rev Phytopathol 37, 447-471. 112 Tsuji, G., Kenmochi, Y., Takano, Y., Sweigard, J., Farrall, L., Furusawa, I., Horino, O. and

Kubo, Y. (2000) Mol Microbiol 38, 940-954. 113 Jacobson, E. S. (2000) Clin Microbiol Rev 13, 708-17. 114 Butler, M. J., Day, A. W., Henson, J. M. and Money, N. P. (2001) Mycologia 93, 1-8. 115 Langfelder, K., Streibel, M., Jahn, B., Haase, G. and Brakhage, A. A. (2003) Fungal Genet

Biol 38, 143-58. 116 Thompson, J. E., Basarab, G. S., Andersson, A., Lindqvist, Y. and Jordan, D. B. (1997)

Biochemistry 36, 1852-60. 117 Perpetua, N. S., Kubo, Y., Yasuda, N., Takano, Y. and Furusawa, I. (1996) Mol Plant

Microbe Interact 9, 323-329. 118 Nagai, M., Kawata, M., Watanabe, H., Ogawa, M., Saito, K., Takesawa, T., Kanda, K. and

Sato, T. (2003) Microbiology 149, 2455-62. 119 Litvintseva, A. P. and Henson, J. M. (2002) Mycol Res 106, 808-814. 120 Frederick, B. A., Caesar Tonthat, T. C., Wheeler, M. H., Sheehan, K. B., Edens, W. A. and

Henson, J. M. (1999) Mycol Res 103, 99-110. 121 Kawamura, C., Tsujimoto, T. and Tsuge, T. (1999) Mol Plant Microbe Interact 12, 59-63. 122 Ikeda, R., Sugita, T., Jacobson, E. S. and Shinoda, T. (2003) Microbiol Immunol 47, 271-

7. 123 Ebel, J. and Mithofer, A. (1998) Planta 206, 335-348. 124 Scheel, D. (1998) Curr Opin Plant Biol 1, 305-310. 125 Knogge, W. (2002) Avirulence determinants and elicitors. In The Mycota. Agricultural

Applications Vol. XI (Kempken, F. ed.) pp. 289-310, Springer-Verlag, Berlin.

211


126 Nurnberger, T. and Brunner, F. (2002) Curr Opin Plant Biol 5, 318-324. 127 Parker, J. E. (2003) Trends Plant Sci 8, 245-7. 128 Montesano, M., Brader, G. and Palva, E. T. (2003) Mol. Plant Pathol. 4, 73-79. 129 Gomez-Gomez, L. and Boller, T. (2000) Mol Cell 5, 1003-11. 130 Kamoun, S. (2001) Curr Opin Plant Biol 4, 295-300. 131 Kamoun, S., Van West, P., Vleeshouwers, V. G., de Groot, K. E. and Govers, F. (1998)

Plant Cell 10, 1413-26. 132 Flor, H. H. (1971) Annu Rev Phytopathol 9, 275-296. 133 Staskawicz, B., Dahlbeck, D., Keen, N. and Napoli, C. (1987) J Bacteriol 169, 5789-94. 134 White, F. F., Yang, B. and Johnson, L. B. (2000) Curr Opin Plant Biol 3, 291-8. 135 van't Slot, K. A. E. and Knogge, W. (2002) Crit Rev Plant Sci 21, 229-271. 136 Lauge, R. and de Wit, P. J. (1998) Fungal Genet Biol 24, 285-97. 137 van Kan, J. A. L. V., van den Ackerveken, G. F. J. M. and de Wit, P. J. G. M. (1991) Mol

Plant Microbe Interact 4, 52-59. 138 Joosten, M. H. A. J., Cozijnsen, T. J. and de Wit, P. J. G. M. (1994) Nature 367, 384-386. 139 Luderer, R., Takken, F. L., de Wit, P. J. and Joosten, M. H. (2002) Mol Microbiol 45, 875-

84. 140 van den Ackerveken, G. F. J. M., van Kan, J. A. L., Joosten, M. H. A. J., Muisers, J. M.,

Verbakel, H. M. and de Wit, P. J. G. M. (1993) Mol Plant Microbe Interact 6, 210-215. 141 Lauge, R., Goodwin, P. H., de Wit, P. J. and Joosten, M. H. (2000) Plant J 23, 735-45. 142 Rohe, M., Gierlich, A., Hermann, H., Hahn, M., Schmidt, B., Rosahl, S. and Knogge, W.

(1995) EMBO J 14, 4168-4177. 143 Sweigard, J. A., Carroll, A. M., Kang, S., Farrall, L., Chumley, F. G. and Valent, B. (1995)

Plant Cell 7, 1221-1233. 144 Orbach, M. J., Farrall, L., Sweigard, J. A., Chumley, F. G. and Valent, B. (2000) Plant Cell

12, 2019-32. 145 Farman, M. L. and Leong, S. A. (1998) Genetics 150, 1049-1058. 146 Farman, M. L., Eto, Y., Nakao, T., Tosa, Y., Nakayashiki, H., Mayama, S. and Leong, S.

A. (2002) Mol Plant Microbe Interact 15, 6-16. 147 Bohnert, H. U., Fudal, I., Dioh, W., Tharreau, D., Notteghem, J. L. and Lebrun, M. H.

(2001) 10th Int Congress on Molecular Plant-Microbe Interactions. 10–14th July, Abstr. 362.

148 Linning, R., Lin, D., Lee, N., Abdennadher, M., Gaudet, D., Thomas, P., Mills, D., Kronstad, J. W. and Bakkeren, G. (2004) Genetics (in press).

149 Dioh, W., Tharreau, D., Notteghem, J. L., Orbach, M. and Lebrun, M. H. (2000) Mol Plant Microbe Interact 13, 217-227.

150 Attard, A., Gout, L., Gourgues, M., Kuhn, M. L., Schmit, J., Laroche, S., Ansan-Melayah, D., Billault, A., Cattolico, L., Balesdent, M. H. and Rouxel, T. (2002) Mol Plant Microbe Interact 15, 672-82.

151 Caffier, V., Vallavieille Pope, C. D. and Brown, J. K. M. (1996) Phytopathology 86, 1112-1121.

152 Pedersen, C., Rasmussen, S. W. and Giese, H. (2002) Fungal Genet Biol 35, 235-246. 153 Zambino, P. J., Kubelik, A. R. and Szabo, L. J. (2000) Phytopathology 90, 819-826. 154 Jia, Y., McAdams, S. A., Bryan, G. T., Hershey, H. P. and Valent, B. (2000) EMBO J 19,

4004-14. 155 Bogdanove, A. J. and Martin, G. B. (2000) Proc Natl Acad Sci U S A 97, 8836-8840. 156 Dangl, J. L. and Jones, J. D. (2001) Nature 411, 826-33.

212


157 Bonas, U. and Lahaye, T. (2002) Curr Opin Microbiol 5, 44-50. 158 Marathe, R. and Dinesh-Kumar, S. P. (2003) Mol Cell 11, 284-6. 159 Martin, G. B., Bogdanove, A. J. and Sessa, G. (2003) Annu. Rev. Plant Biol. 54, 23-61. 160 Nimchuk, Z., Eulgem, T., Holt, I. B. and Dangl, J. L. (2003) Annu Rev Genet 37, 579-609. 161 Joosten, M. H., Vogelsang, R., Cozijnsen, T. J., Verberne, M. C. and de Wit, P. J. (1997)

Plant Cell 9, 367-79. 162 Westerink, N., Roth, R., Van Den Burg, H. A., de Wit, P. J. and Joosten, M. H. (2002) Mol

Plant Microbe Interact 15, 1219-27. 163 Van Den Burg, H. A., Westerink, N., Francoijs, K. J., Roth, R., Woestenenk, E., Boeren,

S., de Wit, P. J., Joosten, M. H. and Vervoort, J. (2003) J Biol Chem 278, 27340-6. 164 Vervoort, J., Van Den Hooven, H. W., Berg, A., Vossen, P., Vogelsang, R., Joosten, M. H.

and de Wit, P. J. (1997) FEBS Lett 404, 153-8. 165 Lauge, R., Joosten, M. H. A. J., van den Ackerveken, G. F. J. M., van den Broek, H. W. J.

and de Wit, P. J. G. M. (1997) Mol Plant Microbe Interact 10, 725-734. 166 Kang, S., Sweigard, J. A. and Valent, B. (1995) Mol Plant Microbe Interact 8, 939-948. 167 Bohnert, H. U., Fudal, I., Tharreau, D., Notteghem, J. L. and Lebrun, M. H. (2003) 11th

International Congress on Molecular Plant-Microbe Interactions, July 18-26 St.-Petersburg, Russia, Abstract OSY013.

168 van't Slot, K. A., van den Burg, H. A., Kloks, C. P., Hilbers, C. W., Knogge, W. and Papavoine, C. H. (2003) J Biol Chem 278, 45730-6.

169 van den Ackerveken, G. F. J. M., Dunn, R. M., Cozijnsen, A. J., Vossen, J. P. M. J., van den Broek, H. W. J. and de Wit, P. J. G. M. (1994) Mol Gen Genet 243, 277-285.

170 Vera Cruz, C. M., Bai, J., Ona, I., Leung, H., Nelson, R. J., Mew, T. W. and Leach, J. E. (2000) Proc Natl Acad Sci U S A 97, 13500-5.

171 Leach, J. E., Vera Cruz, C. M., Bai, J. and Leung, H. (2001) Annu Rev Phytopathol 39, 187-224.

172 Walton, J. D. (2000) Fungal Genet Biol 30, 167-71. 173 Walton, J. D. (1996) Plant Cell 8, 1723-33. 174 Osbourn, A. E. (2001) Proc Natl Acad Sci U S A 98, 14187-8. 175 Markham, J. E. and Hille, J. (2001) Mol Plant Pathol. 2, 229-239. 176 Wolpert, T. J., Dunkle, L. D. and Ciuffetti, L. M. (2002) Annu Rev Phytopathol 40, 251-

285. 177 Proctor, R. H., Hohn, T. M. and McCormick, S. P. (1995) Mol Plant Microbe Interact 8,

593-601. 178 Harris, L. J., Desjardins, A. E., Plattner, R. D., Nicholson, P., Butler, G., Young, Y. C.,

Weston, G., Proctor, R. H. and Hohn, T. M. (1999) Plant Disease 83, 954-960. 179 Brown, D. W., McCormick, S. P., Alexander, N. J., Proctor, R. H. and Desjardins, A. E.

(2001) Fungal Genet Biol 32, 121-33. 180 Desjardins, A. E., Munkvold, G. P., Plattner, R. D. and Proctor, R. H. (2002) Mol Plant

Microbe Interact 15, 1157-64. 181 Desjardins, A. E. (2003) Annu Rev Phytopathol. 41, 177-198. 182 Proctor, R. H., Brown, D. W., Plattner, R. D. and Desjardins, A. E. (2003) Fungal Genet

Biol 38, 237-49. 183 Dutton, M. V. and Evans, C. S. (1996) Can J Microbiol 42, 881-895. 184 Godoy, G., Steadman, J. R., Dickman, M. B. and Dam, R. (1990) Physiol Mol Plant Pathol

37, 179-191. 185 Zhou, T. and Boland, G. J. (1999) Can J Plant Pathol 21, 93-99.

213


186 Lane, B. G. (2002) IUBMB Life 53, 67-75. 187 Daub, M. E. and Ehrenshaft, M. (2000) Annu Rev Phytopathol 38, 461-490. 188 Chung, K. R., Ehrenshaft, M., Wetzel, D. K. and Daub, M. E. (2003) Mol Genet Genomics

270, 103-113. 189 Chung, K. R., Daub, M. E. and Ehrenshaft, M. (2003) Curr Genet 43, 415-24. 190 Ehrenshaft, M., Jenns, A. E., Chung, K. R. and Daub, M. E. (1998) Mol Cell 1, 603-9. 191 Ehrenshaft, M., Chung, K. R., Jenns, A. E. and Daub, M. E. (1999) Curr Genet 34, 478-85. 192 Chung, K. R., Jenns, A. E., Ehrenshaft, M. and Daub, M. E. (1999) Mol Gen Genet 262,

382-9. 193 Hatta, R., Ito, K., Hosaki, Y., Tanaka, T., Tanaka, A., Yamamoto, M., Akimitsu, K. and

Tsuge, T. (2002) Genetics 161, 59-70. 194 Johnson, R. D., Johnson, L., Itoh, Y., Kodama, M., Otani, H. and Kohmoto, K. (2000) Mol

Plant Microbe Interact 13, 742-53. 195 Johnson, L. J., Johnson, R. D., Akamatsu, H., Salamiah, A., Otani, H., Kohmoto, K. and

Kodama, M. (2001) Curr Genet 40, 65-72. 196 Ciuffetti, L. M., Tuori, R. P. and Gaventa, J. M. (1997) Plant Cell 9, 135-44. 197 Martinez, J. P., Ottum, S. A., Ali, S., Franci, L. J. and Ciuffetti, L. M. (2001) Mol Plant

Microbe Interact 14, 675-7. 198 Baidyaroy, D., Brosch, G., Graessle, S., Trojer, P. and Walton, J. D. (2002) Eukaryot Cell

1, 538-47. 199 Del Sorbo, G., Schoonbeek, H. and De Waard, M. A. (2000) Fungal Genet Biol 30, 1-15. 200 Stergiopoulos, I., Zwiers, L. H. and De Waard, M. A. (2002) Eur. J. of Plant Pathol. 108,

719-734. 201 Callahan, T. M., Rose, M. S., Meade, M. J., Ehrenshaft, M. and Upchurch, R. G. (1999)

Mol Plant Microbe Interact 12, 901-10. 202 Upchurch, R. G., Rose, M. S. and Eweida, M. (2001) FEMS Microbiol Lett 204, 89-93. 203 Pitkin, J. W., Panaccione, D. G. and Walton, J. D. (1996) Microbiology 142 ( Pt 6), 1557-

65. 204 Pedley, K. F. and Walton, J. D. (2001) Proc Natl Acad Sci U S A 98, 14174-9. 205 Bussink, H. J., Clark, A. and Oliver, R. (2001) Eur J Plant Pathol 107, 655-659. 206 Furukawa, T., Koga, J., Adachi, T., Kishi, K. and Syono, K. (1996) Plant Cell Physiol 37,

899-905. 207 Furukawa, T. and Syono, K. (1998) Plant Cell Physiol 39, 43-48. 208 Rey, P., Leucart, S., Desilets, H., Belanger, R. R., Larue, J. P. and Tirilly, Y. (2001) Eur J

Plant Pathol 107, 895-904. 209 Moulton, J. E. (1942) Bot. Gaz. 103, 725-739. 210 Turian, G. and Hamilton, R. H. (1960) Biochim. Biophys. Acta 41, 148-150. 211 Wolf, F. T. (1952) Proc Natl Acad Sci U S A 38, 106-111. 212 Navarre, D. A. (1990) MSc. Thesis, Lousiana State University, Baton Rouge. 213 Guevara-Lara, F., Valverde, M. E. and Paredes-Lopez, O. (2000) World J Microbiol

Biotechnol 16, 481-490. 214 Sosa-Morales, M. E., Guevara-Lara, F., Martinez-Juarez, V. M. and Paredes-Lopez, O.

(1997) Appl Microbiol Biotechnol 48, 726-729. 215 Holliday, R. (1961) Genet. Res. Camb. 2, 204-230. 216 Fischer, J. A., McCann, M. P. and Snetselaar, K. M. (2001) Fungal Genet Biol 34, 21-35. 217 Basse, C. W., Lottspeich, F., Steglich, W. and Kahmann, R. (1996) Eur J Biochem 242,

648-56.

214


218 Tudzynski, B., Mihlan, M., Rojas, M. C., Linnemannstons, P., Gaskin, P. and Hedden, P. (2003) J Biol Chem 278, 28635-43.

219 Angra Sharma, R. and Sharma, D. K. (1999) Mycopathologia 148, 87-95. 220 Xiong, L. and Yang, Y. (2003) Plant Cell 15, 745-59. 221 Hahn, K. and Strittmatter, G. (1994) Eur J Biochem 226, 619-26. 222 Morrissey, J. P. and Osbourn, A. E. (1999) Microbiol. Mol. Biol. Rev. 63, 708–724. 223 Hammerschmidt, R. (1999) Annu Rev Phytopathol 37, 285-306. 224 Osbourn, A. E. (1999) Fungal Genet Biol 26, 163-8. 225 vanEtten, H., Temporini, E. and Wasmann, C. (2001) Physiol. Mol. Plant Pathol. 59, 83-

93. 226 Hammond-Kosack, K. E. and Parker, J. E. (2003) Curr Opin Biotechnol 14, 177-93. 227 Veronese, P., Ruiz, M. T., Coca, M. A., Hernandez-Lopez, A., Lee, H., Ibeas, J. I., Damsz,

B., Pardo, J. M., Hasegawa, P. M., Bressan, R. A. and Narasimhan, M. L. (2003) Plant Physiol 131, 1580-90.

228 Cessna, S. G., Sears, V. E., Dickman, M. B. and Low, P. S. (2000) Plant Cell 12, 2191-200.

229 Mayer, A. M., Staples, R. C. and Gil-Ad, N. L. (2001) Phytochemistry 58, 33-41. 230 Goodwin, P. H., Li, J. and Jin, S. (2001) FEMS Micro Biol Lett 202, 103-107. 231 Garre, V., Muller, U. and Tudzynski, P. (1998) Mol Plant Microbe Interact 11, 772-83. 232 Bussink, H. J. and Oliver, R. (2001) Eur J Biochem 268, 15-24. 233 Schouten, A., ten Berge, K. B., Vermeer, J., Stewart, J., Wagemakers, L., Williamson, B.

and van Kan, J. A. L. (2002) Mol Plant Pathol. 3, 227-238. 234 Robbertse, B., Yoder, O. C., Nguyen, A., Schoch, C. L. and Turgeon, B. G. (2003) Mol.

Plant Microbe Interact. 16, 1013-1021. 235 Stephenson, S. A., Hatfield, J., Rusu, A. G., Maclean, D. J. and Manners, J. M. (2000)

Mol. Plant Microbe Interact. 13, 929-941. 236 Waspi, U., Schweizer, P. and Dudler, R. (2001) Plant Cell 13, 153-61. 237 Lyngkjaer, M. F., Carver, T. L. W. and Zeyen, R. J. (2001) Physiol Mol Plant Pathol 59,

243-256. 238 Lu, M., Tang, X. and Zhou, J. M. (2001) Plant Cell 13, 437-447. 239 Buttner, D. and Bonas, U. (2002) EMBO J 21, 5313-22. 240 Mackey, D., Holt, B. F., Wiig, A. and Dangl, J. L. (2002) Cell 108, 743-54. 241 Abramovitch, R. B., Kim, Y. J., Chen, S., Dickman, M. B. and Martin, G. B. (2003)

EMBO J 22, 60-69. 242 Baidyaroy, D., Brosch, G., Ahn, J. H., Graessle, S., Wegener, S., Tonukari, N. J.,

Caballero, O., Loidl, P. and Walton, J. D. (2001) Plant Cell 13, 1609-24. 243 Moerschbacher, B. M., Mierau, M., Graessner, B., Noll, U. and Mort, A. J. (1999) J Exp

Bot 50, 605-612. 244 Wietholter, N., Graessner, B., Mierau, M., Mort, A. J. and Moerschbacher, B. M. (2003)

Mol Plant Microbe Interact 16, 945-52. 245 Bailey, A., Mueller, E. and Bowyer, P. (2000) J Biol Chem 275, 14242-7. 246 Namiki, F., Matsunaga, M., Okuda, M., Inoue, I., Nishi, K., Fujita, Y. and Tsuge, T. (2001)

Mol Plant Microbe Interact 14, 580-4. 247 Segers, G., Bradshaw, N., Archer, D., Blissett, K. and Oliver, R. P. (2001) Mol Plant

Microbe Interact 14, 367-77. 248 Idnurm, A. and Howlett, B. J. (2002) Eukaryot Cell 1, 719-24.

215


249 Smith, D. G., Garcia-Pedrajas, M. D., Gold, S. E. and Perlin, M. H. (2003) Mol. Microbiol. 50, 259-276.

250 Wasmann, C. C. and vanEtten, H. D. (1996) Mol. Plant Microbe Interact. 9, 793-803. 251 George, H. L. and vanEtten, H. D. (2001) Fungal Genet Biol 33, 37-48. 252 Bowyer, P., Clarke, B. R., Lunness, P., Daniels, M. J. and Osbourn, A. E. (1995) Science

267, 371-4. 253 Sandrock, R. W. and Vanetten, H. D. (2001) Physiol Mol Plant Pathol. 58, 159-171. 254 Bouarab, K., Melton, R., Peart, J., Baulcombe, D. and Osbourn, A. (2002) Nature 418,

889-92. 255 Glenn, A. E., Gold, S. E. and Bacon, C. W. (2002) Mol. Plant Microbe Interact. 15, 91-

101. 256 Nakaune, R., Hamamoto, H., Imada, J., Akutsu, K. and Hibi, T. (2002) Mol Genet

Genomics 267, 179-85. 257 Schoonbeek, H., Del Sorbo, G. and de Waard, M. A. (2001) Mol Plant Microbe Interact

14, 562-71. 258 Stergiopoulos, I., Zwiers, L. H. and de Waard, M. A. (2003) Mol Plant Microbe Interact

16, 689-98. 259 Fleissner, A., Sopalla, C. and Weltring, K. M. (2002) Mol Plant Microbe Interact 15, 102-

8. 260 Urban, M., Bhargava, T. and Hamer, J. E. (1999) EMBO J 18, 512-21. 261 Shi, Z., Christian, D. and Leung, H. (1998) Mol Plant Microbe Interact 11, 199-207. 262 Lau, G. W. and Hamer, J. E. (1998) Fungal Genet Biol 24, 228-239. 263 Reichmann, M., Jamnischek, A., Weinzierl, G., Ladendorf, O., Huber, S., Kahmann, R.

and Kamper, J. (2002) Mol Microbiol 46, 1169-82. 264 Huber, S. M., Lottspeich, F. and Kamper, J. (2002) Mol Genet Genomics 267, 757-71. 265 Durrenberger, F., Laidlaw, R. D. and Kronstad, J. W. (2001) Mol Microbiol 41, 337-48. 266 Dawe, A. L. and Nuss, D. L. (2001) Annu Rev Genet 35, 1-29. 267 Choi, G. H., Chen, B. and Nuss, D. L. (1995) Proc Natl Acad Sci U S A 92, 305-9. 268 Castro, M., Kramer, K., Valdivia, L., Ortiz, S., Benavente, J. and Castillo, A. (1999) FEMS

Microbiol Lett 175, 95-99. 269 Castro, M., Kramer, K., Valdivia, L., Ortiz, S. and Castillo, A. (2003) FEMS Microbiol

Lett 228, 87-91. 270 Chu, Y. M., Jeon, J. J., Yea, S. J., Kim, Y. H., Yun, S. H., Lee, Y. W. and Kim, K. H.

(2002) Appl Environ Microbiol 68, 2529-34. 271 Ahn, I. P. and Lee, Y. H. (2001) Mol. Plant Microbe Interact. 14, 496-507. 272 Bertrand, H. (2000) Annu Rev Phytopathol. 38, 397-422. 273 Lee, N., D'Souza, C. A. and Kronstad, J. W. (2003) Annu Rev Phytopathol. 41, 399-427. 274 Barrett, K. J., Gold, S. E. and Kronstad, J. W. (1993) Mol Plant Microbe Interact 6, 274-

83. 275 Gold, S., Duncan, G., Barrett, K. and Kronstad, J. (1994) Genes Dev 8, 2805-16. 276 Durrenberger, F., Wong, K. and Kronstad, J. W. (1998) Proc Natl Acad Sci U S A 95,

5684-9. 277 Regenfelder, E., Spellig, T., Hartmann, A., Lauenstein, S., Bolker, M. and Kahmann, R.

(1997) EMBO J 16, 1934-42. 278 Choi, W. and Dean, R. A. (1997) Plant Cell 9, 1973-83. 279 Mitchell, T. K. and Dean, R. A. (1995) Plant Cell 7, 1869-78. 280 Liu, S. and Dean, R. A. (1997) Mol Plant Microbe Interact 10, 1075-86.

216


281 Hamer, J. E. and Talbot, N. J. (1998) Curr Opin Microbiol 1, 693-7. 282 Klimpel, A., Gronover, C. S., Williamson, B., Stewart, J. A. and Tudzynski, B. (2002)

Molecular Plant Pathology 3, 439-450. 283 Choi, G. H. and Nuss, D. L. (1992) Science 257, 800-3. 284 Chen, B. S., Gao, S. J., Choi, G. H. and Nuss, D. L. (1996) Proc. Natl. Acad. Sci. USA 93,

7996-8000. 285 Kruger, J., Loubradou, G., Regenfelder, E., Hartmann, A. and Kahmann, R. (1998) Mol

Gen Genet 260, 193-8. 286 Gold, S. E., Brogdon, S. M., Mayorga, M. E. and Kronstad, J. W. (1997) Plant Cell 9,

1585-94. 287 Kruger, J., Loubradou, G., Wanner, G., Regenfelder, E., Feldbrugge, M. and Kahmann, R.

(2000) Mol. Plant Microbe Interact. 13, 1034-1040. 288 Adachi, K. and Hamer, J. E. (1998) Plant Cell 10, 1361-74. 289 Xu, J. R. (2000) Fungal Genet Biol 31, 137-52. 290 Mey, G., Oeser, B., Lebrun, M. H. and Tudzynski, P. (2002) Mol Plant Microbe Interact

15, 303-12. 291 Jenczmionka, N. J., Maier, F. J., Losch, A. P. and Schafer, W. (2003) Curr Genet 43, 87-

95. 292 Hou, Z., Xue, C., Peng, Y., Katan, T., Kistler, H. C. and Xu, J. R. (2002) Mol Plant

Microbe Interact 15, 1119-27. 293 Brachmann, A., Schirawski, J., Muller, P. and Kahmann, R. (2003) EMBO J 22, 2199-210. 294 D'Souza, C. A. and Heitman, J. (2001) FEMS Microbiol Rev 25, 349-64. 295 Xu, J. R. and Hamer, J. E. (1996) Genes Dev 10, 2696-706. 296 Mayorga, M. E. and Gold, S. E. (1998) Fungal Genet Biol. 24, 364-376. 297 Mayorga, M. E. and Gold, S. E. (1999) Mol Microbiol 34, 485-97. 298 Andrews, D. L., Egan, J. D., Mayorga, M. E. and Gold, S. E. (2000) Mol Plant Microbe

Interact 13, 781-6. 299 Mayorga, M. E. and Gold, S. E. (2001) Mol Microbiol 41, 1365-79. 300 Kaffarnik, F., Muller, P., Leibundgut, M., Kahmann, R. and Feldbrugge, M. (2003) EMBO

J 22, 5817-26. 301 Shaw, B. D. and Hoch, H. C. (2000) Fungal Genet Biol 31, 43-53. 302 Warwar, V. and Dickman, M. B. (1996) Appl Environ Microbiol 62, 74-79. 303 Warwar, V., Oved, S. and Dickman, M. B. (2000) FEMS Microbiol Lett 191, 213-9. 304 Gage, M. J., Bruenn, J., Fischer, M., Sanders, D. and Smith, T. J. (2001) Mol. Microbiol.

41, 775-785. 305 Perez Garcia, A., Snoeijers, S. S., Joosten, M. H. A. J., Goosen, T. and de Wit, P. J. G. M.

(2001) Mol. Plant Microbe Interact. 14, 316-325. 306 Lau, G. and Hamer, J. E. (1996) Plant Cell 8, 771-781. 307 Dufresne, M., Perfect, S., Pellier, A. L., Bailey, J. A. and Langin, T. (2000) Plant Cell 12,

1579-90. 308 Thon, M. R., Nuckles, E. M., Takach, J. E. and Vaillancourt, L. J. (2002) Mol Plant

Microbe Interact 15, 120-8. 309 Torreblanca, J., Stumpferl, S. and Basse, C. W. (2003) Fungal Genet Biol 38, 22-32. 310 Quadbeck-Seeger, C., Wanner, G., Huber, S., Kahmann, R. and Kamper, J. (2000) Mol

Microbiol 38, 154-66. 311 Muller, P., Katzenberger, J. D., Loubradou, G. and Kahmann, R. (2003) Eukaryot Cell 2,

609-17.

217


312 Aichinger, C., Hansson, K., Eichhorn, H., Lessing, F., Mannhaupt, G., Mewes, W. and Kahmann, R. (2003) Mol. Genet. Genomics 270, 303-314.

313 Lu, S. W., Kroken, S., Lee, B. N., Robbertse, B., Churchill, A. C., Yoder, O. C. and Turgeon, B. G. (2003) Proc Natl Acad Sci U S A 100, 5980-5.

314 Viaud, M. C., Balhadere, P. V. and Talbot, N. J. (2002) Plant Cell 14, 917-30. 315 Wang, P., Cardenas, M. E., Cox, G. M., Perfect, J. R. and Heitman, J. (2001) EMBO Rep

2, 511-8. 316 Olson, A. and Stenlid, J. (2001) Nature 411, 438. 317 Inoue, I., Namiki, F. and Tsuge, T. (2002) Plant Cell 14, 1869-83. 318 Yoder, O. C. and Turgeon, B. G. (2001) Curr Opin Plant Biol 4, 315-21. 319 Leach, J. E., Gold, S., Tolin, S. and Eversole, K. (2003) Phytopathol. 93, 524-527. 320 Yarden, O., Ebbole, D. J., Freeman, S., Rodriguez, R. J. and Dickman, M. B. (2003) Mol

Plant Microbe Interact 16, 859-66. 321 Goffeau, A. (1998) Pathol Biol (Paris) 46, 96-7. 322 Wendland, J., Ayad-Durieux, Y., Knechtle, P., Rebischung, C. and Philippsen, P. (2000)

Gene 242, 381-91. 323 Alberti-Segui, C., Dietrich, F., Altmann-Johl, R., Hoepfner, D. and Philippsen, P. (2001) J

Cell Sci 114, 975-86. 324 Wendland, J. and Philippsen, P. (2002) Fungal Genet Biol 37, 81-8. 325 Wendland, J. (2003) EMBO Rep 4, 200-4. 326 Knechtle, P., Dietrich, F. and Philippsen, P. (2003) Mol Biol Cell 14, 4140-4154. 327 Soanes, D. M., Skinner, W., Keon, J., Hargreaves, J. and Talbot, N. J. (2002) Mol. Plant

Microbe Interact. 15, 421-427. 328 Sacadura, N. T. and Saville, B. J. (2003) Fungal Genet Biol 40, 47-64. 329 Karlsson, M., Olson, A. and Stenlid, J. (2003) Fungal Genet Biol 39, 51-9. 330 Guettler, S., Jackson, E. N., Lucchese, S. A., Honaas, L., Green, A., Hittinger, C. T., Tian,

Y., Lilly, W. W. and Gathman, A. C. (2003) Fungal Genet Biol 39, 191-8. 331 Ospina Giraldo, M. D., Collopy, P. D., Romaine, C. P. and Royse, D. J. (2000) Exp Mycol

29, 81-94. 332 Lee, S. H., Kim, B. G., Kim, K. J., Lee, J. S., Yun, D. W., Hahn, J. H., Kim, G. H., Lee, K.

H., Suh, D. S., Kwon, S. T., Lee, C. S. and Yoo, Y. B. (2002) Fungal Genet Biol 35, 115-34.

333 Pandey, A. and Mann, M. (2000) Nature 405, 837-46. 334 Uetz, P., Giot, L., Cagney, G., Mansfield, T. A., Judson, R. S., Knight, J. R., Lockshon, D.,

Narayan, V., Srinivasan, M., Pochart, P., Qureshi-Emili, A., Li, Y., Godwin, B., Conover, D., Kalbfleisch, T., Vijayadamodar, G., Yang, M., Johnston, M., Fields, S. and Rothberg, J. M. (2000) Nature 403, 623-7.

335 Vidal, M. and Endoh, H. (1999) Trends Biotechnol 17, 374-81. 336 Tsien, R. Y. (1998) Annu Rev Biochem 67, 509-44. 337 Miyawaki, A., Llopis, J., Heim, R., Mccaffery, J. M., Adams, J. A., Ikura, M. and Tsien, R.

Y. (1997) Nature 388, 882-7. 338 Nagai, Y., Miyazaki, M., Aoki, R., Zama, T., Inouye, S., Hirose, K., Iino, M. and

Hagiwara, M. (2000) Nature Biotechnology 18, 313-316. 339 De Angelis, D. A. (1999) Physiol Genomics 1, 93-9. 340 Hamer, L., Adachi, K., Montenegro-Chamorro, M. V., Tanzer, M. M., Mahanty, S. K., Lo,

C., Tarpey, R. W., Skalchunes, A. R., Heiniger, R. W., Frank, S. A., Darveaux, B. A.,

218


Lampe, D. J., Slater, T. M., Ramamurthy, L., Dezwaan, T. M., Nelson, G. H., Shuster, J. R., Woessner, J. and Hamer, J. E. (2001) Proc Natl Acad Sci U S A 98, 5110-5.

341 Sweigard, J. A. and Ebbole, D. J. (2001) Curr Opin Microbiol 4, 387-92. 342 Kahmann, R. and Basse, C. (2001) Curr Opin Microbiol 4, 374-80. 343 Stuiver, M. H. and Custers, J. H. (2001) Nature 411, 865-8. 344 Dickman, M. B., Park, Y. K., Oltersdorf, T., Li, W., Clemente, T. and French, R. (2001)

Proc Natl Acad Sci U S A 98, 6957-62. 345 Punja, Z. K. (2001) Can J Plant Pathol 23, 216-235.

219

the path in fungal plant pathogenicity; many opportunities

Documents