1. Top of page

Take-all, caused by the fungus Gaeumannomyces graminis var. tritici, is the most important root disease of wheat worldwide. Many years of intensive research, reflected by the large volume of literature on take-all, has led to a considerable degree of understanding of many aspects of the disease. However, effective and economic control of the disease remains difficult. The application of molecular techniques to study G. graminis and related fungi has resulted in some significant advances, particularly in the development of improved methods for identification and in elucidating the role of the enzyme avenacinase as a pathogenicity determinant in the closely related oat take-all fungus (G. graminis var. avenae). Some progress in identifying other factors that may be involved in determining host range and pathogenicity has been made, despite the difficulties of performing genetic analyses and the lack of a reliable transformation system.


  1. Top of page

Take-all of wheat is caused by the fungus Gaeumannomyces graminis var. tritici (Ggt) and is the most important root disease of wheat worldwide. The same pathogen (Ggt) also affects triticale, barley and rye, but to lesser extents. Yield losses, due to stunting and premature ripening resulting in shrivelled grain, can be large in severely affected crops. Substantial financial losses result from decreases in yield and quality. Using husbandry to minimize the financial impact of the disease imposes restrictions on the cropping systems and rotations that can be employed. Despite considerable progress resulting from decades of intensive research (summarized in Asher and Shipton, 1981; Cook, 2003; Hornby et al., 1998) there is still much that we do not understand about the disease and the fungi that cause it. In the UK, most of Europe and parts of the USA, when consecutive susceptible cereal crops are grown, the disease often rises to a peak over 2–4 years after a break crop, then typically declines in severity (take-all decline). Although there have been reports of a decline in take-all associated with cultivation of continuous wheat in Australia, it may be different from that described above and appears to be less robust than in Europe and the USA.

During the intercrop period, Ggt survives saprophytically on the crop debris and is the main source of inoculum for the following crop. The quantity of viable inoculum remaining in the soil falls sharply in the absence of a susceptible crop and this is usually sufficient to reduce disease severity to an insignificant level in a first susceptible cereal following a 1-year break. Perithecia, which subsequently release ascospores (Fig. 1a,b), are sometimes produced on the cereal stem bases or stubble although this is believed to be of limited importance as a source of inoculum in the field. Ggt can also survive as a pathogen on other grasses and volunteer cereal plants.


Figure 1. Infection of wheat by Gaeumannomyces graminis and Phialophora graminicola. (a) G. graminis var. tritici: perithecia produced on wheat roots in a rotting test, surrounded by released asci; (b) G. graminis var. tritici: ascospores in a single ascus (arrowed, a), with one ascospore separated from the ascus (arrowed, as); (c) G. graminis var. tritici: dark runner hyphae on the surface of a wheat root; (d) a patch of stunted and yellowed wheat plants (foreground), typical of take-all, in the summer; (e) whiteheads of infected plants in a wheat crop with take-all, in late summer; (f) similarities in discoloration of wheat roots, and differences in stem-base blackening of wheat, caused by infection with G. graminis var. tritici (Ggt) (left), G. graminis var. graminis (Ggg) (centre) and P. graminicola (right), in a pathogenicity test; (g) Phialophora sp. (lobed-hyphopodia): lobed hyphopodia on the leaf sheath at the stem base of a wheat plant; (h) Phialophora sp. (lobed-hyphopodia): swollen cells with pores (known as growth cessation structures or vesicles) in a wheat root; (i) P. graminicola: swollen cells with pores (known as growth cessation structures or vesicles) in a wheat root. Images (a), (b), (d), (f), (h) and (i) were kindly provided by Richard Gutteridge (Rothamsted Research).

Download figure to PowerPoint

On autumn-sown winter wheat primary infection of the seminal roots of seedlings occurs with the growth of dark runner hyphae on the root surface (Fig. 1c). Hyaline branches from these runner hyphae penetrate the root by invading the root cortex and then colonize and destroy the vascular tissue. In early spring the root lesions develop further followed by secondary infections resulting from root-to-root contact. In favourable years the disease may be severe by late spring when uneven crop growth can be observed. More typically in early summer, just before ear emergence, the flag leaves of infected plants may begin to roll as a result of drought-stress, and during summer patches of stunted plants with yellowed leaves (Fig. 1d) and whiteheads (Fig. 1e), a symptom of premature ripening, are observed. After harvest infected crop debris remains in the soil and if another susceptible crop is sown the cycle begins again.

The most important cultural practice used to control take-all is crop rotation. Other cultural practices that influence take-all severity include sowing date, tillage practices, the application of fertilizers and grass weed control. Cultivation of non-susceptible break crops, delaying sowing date and grass weed control reduce take-all severity as a result of decreased inoculum. Nitrogen-, phosphate- and manganese-deficiency can all increase susceptibility of a crop to take-all, but the relative importance of these nutrients varies according to geographical location, with nitrogen and phosphate more important in the UK and manganese in the USA and Australia (reviewed in Hornby et al., 1998). Until recently these were the only effective methods available to farmers for controlling take-all. The best control is achieved by rotating susceptible cereal crops with non-susceptible break crops in which grass weeds and cereal volunteers are controlled effectively. Tillage has contributed to reducing take-all severity in the USA (Weller et al., 2002) but the effects of cultural practices involving reduced tillage, such as direct-drilling, are less consistent in the UK than in the USA (Hornby et al., 1998; Yarham, 1981).

Differences between wheat cultivars in their resistance to take-all are generally considered to be too small and too inconsistent to be useful (Cook et al., 1995; Hornby et al., 1998; Weller et al., 2002). There has been relatively little published work concerning resistance of wheat to Ggt during the last 20 years (Eastwood et al., 1994; Penrose, 1985, 1991, 1995; Penrose and Neate, 1994; Rengel et al., 1993, 1994). Rye has greater tolerance of take-all than does wheat and it was hoped that the wheat–rye hybrid triticale might retain this property. However, levels of susceptibility were found to be intermediate, and there is generally no advantage in changing to triticale, except in some situations where severe take-all is expected (Hornby et al., 1998). The grass species Dasapyrum (Haynaldia) and Aegilops also show resistance to take-all, but the prospects for introducing this into wheat seem poor (Linde-Laursen et al., 1973).

Some soils are known to be suppressive to take-all disease. The reasons for this are often uncertain but it may result from effects on the pathogen (growth rate and/or population structure), the host (host susceptibility) or the balance of antagonistic microflora in the soil (Bateman et al., 1997; Hornby, 1983; Rengel, 1997; Rovira and Wildermuth, 1981; Walker, 1975). General suppression increases with increasing microbial biomass in the soil (Weller et al., 2002). Various mechanisms for specific suppression exist (Hornby, 1983; Rovira and Wildermuth, 1981; Walker, 1975). The fastest developing durable mechanism of specific suppression of take-all is take-all decline (TAD), which was first demonstrated experimentally at Rothamsted in the 1960s (Hornby et al., 1998). Walker (1975) defined TAD as ‘suppression developed within a few seasons in the presence of the severely diseased host’. TAD has been reported from many different geographical locations, but although the cause is probably mainly microbial, the microorganisms responsible may be different in different soils. Much research has been aimed at understanding the causes of TAD and at identifying the microorganisms responsible for suppression in TAD or other suppressive soils. Sometimes wheat is grown continuously to exploit take-all decline, but the yields obtained rarely match those of a first cereal after a break. In addition, TAD usually takes 4–6 years to become established over a whole field (Hornby et al., 1998; Weller et al., 2002) and even when TAD is established, take-all can still cause considerable losses if weather and soil conditions are conducive to disease.

A range of different microorganisms have been investigated as potential biological control agents for take-all, many isolated from TAD soils. Particular attention has been paid to fluorescent pseudomonads, which have been demonstrated to provide effective biological control of take-all in pot and field experiments (Chapon et al., 2002; de Souza et al., 2003; Duffy and Weller, 1995, 1996; Ryder et al., 1990; Weller et al., 2002). Production of antibiotics has been demonstrated to provide the mechanism by which many Pseudomonas fluorescens strains protect wheat roots aginst take-all. Phenazine-1-carboxylic acid (PCA) was the first such antibiotic to be identified but production of 2,4-diacetlyphloroglucinol (DAPG) by fluorescent psuedomonads is now believed to be more important in TAD than production of PCA (see reviews by Cook, 2003; Weller et al., 2002). A number of other bacterial species and a few fungal species have also been investigated (Dewan and Sivasithamparam, 1988; Duffy et al., 1996; Kim et al., 1997; Ross et al., 2000; Rovira et al., 1992; Ryder and Rovira, 1993; Sivasithamparam, 1998). Some of the fungi that have been found to be effective are closely related to the take-all fungus, namely avirulent/hypovirulent isolates of G. graminis and Phialophora species (Andrade et al., 1994; Duffy and Weller, 1995, 1996; Mathre et al., 1998; Sivasithamparam, 1975; Wong et al., 1996; Zriba et al., 1999). The accumulated research on potential biological control agents for take-all of wheat demonstrates that screening in vitro, and even in pot experiments, does not necessarily give a reliable indication of effectiveness in the field (Cook et al., 1995; Elsherif and Grossmann, 1994). It is also evident that agents that give control in one geographical location do not necessarily have any beneficial effects elsewhere (e.g. Capper and Higgins, 1993; Hornby et al., 1993). Reports of successful biological control of take-all are mainly from the USA and Australia, with very few from Europe.

Take-all has long attracted the interest of agrochemical companies but until recently there were no fungicides that could be used commercially (Bateman, 1989), and management of take-all relied on husbandry. Recently, however, two fungicides have become available as seed treatments for controlling the disease. Fluquinconazole (a trade name of the seed treatment formulation is Jockey®) is a triazole fungicide with a relatively broad spectrum of activity (Bateman et al., 2003; Löchel et al., 1998). Silthiofam (previously MON65500, trade name Latitude®), by contrast, is specific to Ggt (Beale et al., 1998; Schoeny and Lucas, 1999; Spink et al., 1998). Silthiofam has a novel mode of action, with the mitochondrial adenine nucleotide transporter identified as its probable target (Joseph-Horne et al., 2000). Host defence responses are also enhanced following silthiofam treatment (Huang et al., 2001b).


  1. Top of page

Gaeumannomyces is an ascomycete that was previously classified in the order Diaporthales. However, more recently, links with Magnaporthe species were noted and Gaeumannomyces has now been transferred to the family Magnaporthaceae, which is yet to be assigned to an order (Cannon, 1994). The genus Gaeumannomyces var. Arx and Olivier (emend. Walker) contains seven known species: G. graminis, G. caricis, G. cylindrosporus, G. incrustans, G. medullaris, G. amomi and G. wongoonoo (Bussaban et al., 2001a; Hornby et al., 1998; Wong, 2002). G. graminis and G. cylindrosporus infect cereals and grasses, G. caricis infects sedges, G. incrustans was isolated from turf-grass, G. medullaris from the rush Juncus roemerianus (Kohlmeyer et al., 1995; Landschoot and Jackson, 1989) and, recently, G. amomi from wild ginger (Amomum siamense; Bussaban et al., 2001b) and G. wongoonoo from buffalo grass (Stenophram secundatum; Wong, 2002).

The anamorph of G. medullaris is Trichocladium medullare, whereas the known anamorphs of the other Gaeumannomyces species are all species of Phialophora[the commonly accepted name, used here for convenience, but see reference to Gams (2000) below]. For example, the anamorph of G. cylindrosporus is Phialophora graminicola, and that of G. graminis var. graminis is Phialophora sp. (lobed-hyphopodia) (see below). G. graminis and related Phialophora species found on cereal and grass roots are known collectively as the Gaeumannomyces–Phialophora (G-P) complex, some of which cause superficially similar discoloration of wheat roots (Fig. 1f). The fungi of the G-P complex attack root systems, particularly of Gramineae and Cyperaceae, and most are necrotrophic parasites (Hornby et al., 1998). Phialophora is a highly polymorphic genus containing anamorphs of Discomycetes, Pyrenomycetes and Loculoascomycetes. Gams (2000) has therefore proposed a reclassification of Phialophora spp., which involves segregating the Phialophora anamorphs of Gaeumannomyces species into a new genus Harpophora, which also contains the related fungus Cephalosporium maydis.

Four varieties of G. graminis, which differ in host range, pathogenicity, ascospore size range or hyphopodial structure, have been identified (Table 1). G. graminis varieties tritici (Ggt), avenae (Gga) and maydis (Ggm) have simple hyphopodia, whereas G. graminis var. graminis (Ggg) has characteristically lobed hyphopodia (Fig. 1g), and its anamorph is therefore referred to as Phialophora sp. (lobed-hyphopodia). Ggt and Gga are pathogenic on several cereal crops, whereas Ggg is pathogenic on rice, but is weakly/non-pathogenic on some other cereals and Ggm causes take-all of maize, but only slight infections on some other cereals. Although Ggt and Ggm can occur on some grass species, Gga and Ggg are pathogenic on turf-grasses and Bermuda grass, respectively. Ggt and Gga are highly invasive on wheat roots. Hyaline branches (infection hyphae) from the dark runner hyphae penetrate and invade the root cortex then penetrate the endodermis and invade the stele. The phloem is rapidly destroyed followed by slower colonization and blockage of the xylem with gum-like deposits (Deacon, 1981). While colonization of the internal root tissue within root lesions occurs, growth of the runner hyphae along the surface of wheat roots continues. Further root lesions develop at points of invasion by hyaline branches. Eventually growth of the pathogen may reach the stem base and, under suitable conditions, the culm (beneath the enclosing leaf sheaths) can become completely covered with darkly pigmented mycelium (Cook, 2003). Ggg and P. graminicola are less invasive. Ggg rarely penetrates the steles of seminal root axes of wheat but sometimes invades the steles of root laterals. Colonization of wheat roots by P. graminicola is almost always limited to the outer cortical cells, even in lateral roots (Deacon, 1981). Both Phialophora sp. (lobed-hyphopodia) (Ggg) and P. graminicola form growth cessation structures within the cortex of wheat roots (Fig. 1h,i). Ggt, Gga and Ggg all produce hyphopodia, from branching pigmented hyphae, on infected plant tissue (see above; Skou, 1981), but their role, if any, in the infection process is unclear. It is interesting that the related Magnaporthe grisea (a foliar pathogen that infects leaves via melanized appressoria) can also colonize wheat roots and, when it does so, it produces hyphopodia (Sesma and Osbourn, 2003).

Table 1.  Summary of the major morphological and pathogenicity characteristics of Gaeumannomyces graminis that are used to distinguish between varieties.
G. graminis varietyAbbreviationPathogenicityMean ascospore length (µm)Morphology of hyphopodiaReferences
G. graminis var. triticiGgtCauses take-all of wheat, triticale, barley and rye (in decreasing order of susceptibility) 70–105SimpleDeacon (1981); Walker (1981); Hornby et al. (1998)
G. graminis var. avenaeGgaSimilar to Ggt but also causes take-all of oats and take-all patch of turf-grasses100–130SimpleDeacon (1981); Walker (1981); Hornby et al. (1998)
G. graminis var. graminisGggCauses dieback of Bermuda grass and sheath blight of rice 80–105LobedOu (1972); Walker (1981); McCarty and Lucas (1989); Elliott (1991); Datnoff (1993); Hornby et al. (1998)
G. graminis var. maydisGgmCauses take-all of maize 55–85SimpleYao (1992); Hornby et al. (1998)

Nuclear ribosomal DNA (rDNA) sequences have been used to determine phylogenetic relationships between fungi in the Gaeumannomyces–Phialophora complex. G. incrustans, G. cylindrosporus/P. graminicola and G. graminis formed distinct groups when internal transcribed spacer (ITS) rDNA sequences were analysed using distance-based (Ward and Bateman, 1999) and parsimony-based (Bryan et al., 1995) methods of phylogenetic analysis. Within G. graminis, isolates of Ggt, Gga and Ggm formed distinct groups according to variety, whereas Ggg formed three distinct groups, with the distance-based method (Ward and Bateman, 1999). Using the parsimony-based method isolates of Ggt classified as R (rye-attacking) or N (non-rye-attacking) were grouped separately and Australian oat-attacking isolates, classified as Ggt on the basis of ascospore length (Yeates et al., 1986), were grouped with Gga isolates (Bryan et al., 1995). Phylogenetic analysis of ITS sequences has also confirmed that P. graminicola is the anamorph of G. cylindrosporus (Bryan et al., 1995; Ward and Bateman, 1999), and that P. zeicola and P. radicicola Cain are likely to be the same fungus and the anamorph of Ggm (Ward and Bateman, 1999). A new Phialophora species (typified by isolate GP57), which appears to be the closest known relative to G. graminis and its Phialophora anamorphs, was recently recorded from wheat in the UK (Ward and Bateman, 1999). Ribosomal ITS sequence analysis (E. Ward, unpublished results) indicates that this fungus may be identical to a new Phialophora species identified in Germany (Ulrich et al., 2000). The late wilt pathogen of maize, Cephalosporium maydis, was shown to be closely related to G. graminis but, contrary to earlier suggestions (Gams, 1971, cited in Walker, 1981), it is a distinct taxon (Ward and Bateman, 1999). Sequence analysis has also confirmed links between the Gaeumannomyces–Phialophora complex and Magnaporthe species (Bryan et al., 1995; Bunting et al., 1996; Ward and Bateman, 1999).


  1. Top of page

Classification of G. graminis varieties on the basis of disease symptoms, host range and/or morphological characteristics in culture is difficult, time consuming and laborious. Various semiselective media have proved useful for identification of Ggt, particularly in the USA (Hornby et al., 1998; Juhnke et al., 1984). However, some of these are not suitable for use where avirulent Phialophora spp., which behave similarly to Ggt in culture, are present (Hornby et al., 1998).

Various molecular methods have been used to detect and identify G. graminis at the species and variety level.

Restriction fragment length polymorphism (RFLP) analyses using labelled DNA probes

Several different DNA probes have been used for identification of isolates of G. graminis and related species. A mitochondrial DNA probe from Ggt (pMSU315) hybridized specifically to Gaeumannomyces and Phialophora species with RFLP patterns characteristic of particular species and varieties (Bateman et al., 1992; Henson, 1989, 1992). Ggt, Gga, Ggg and G. cylindrosporus/P. graminicola were conclusively identified by RFLP analysis using a mitochondrial small-subunit rDNA probe (GggMR1/pEG34) from Ggg (Bateman et al., 1997; Ward and Bateman, 1994; Ward and Gray, 1992). RFLP analysis using GggMR1 with EcoRI-digested DNA also identified subgroups of some species/varieties, but even greater discrimination among Ggt isolates is possible using a mitochondrial small-subunit rDNA probe from Ggt (pGgtMS7, J. Freeman, unpublished results). Attempts in our laboratory to develop a PCR assay based on this region have been unsuccessful. Various primers for this region have been exhaustively tested but reliable amplification from all isolates of Ggt has not been achieved (J. Freeman, unpublished results). A PCR product amplified from an isolate of Ggt (L3.4A) with MSU1 and MSU7 (Zhou and Stanosz, 2001) hybridizes to some genomic fragments homologous to pEG34 and pGgtMS7, but not to others. A partial DNA sequence of this fragment is almost identical to sequences within introns 1 and 3 of the mtSSU gene of Cryphonectria parasitica (accession number AF029891). Optional introns are common in fungal mitochondrial genes (Rosewich and Kistler, 2000; Salvo et al., 1998) and it is possible that the variety of RFLP patterns identified by hybridization of pGgtMS7 to EcoRI-digested genomic DNA of Ggt (more than two bands greater than 1.8 kb usually hybridize), and the difficulties amplifying this gene from Ggt by PCR, may be due to the presence of optional introns.

RFLP analysis using a nuclear 26S rDNA probe grouped most G. graminis isolates according to variety and identified subgroups of Ggt (Tan et al., 1994). One RFLP band appeared to be associated with virulence on oats and bentgrass, but the correlation was not absolute. Isolates with the same or similar host preference and pathogenicity were often grouped together by RFLP patterns identified using labelled total DNA of Ggt and a wheat rDNA probe (Bryan et al., 1999; O'Dell et al., 1992).

Polymorphisms in length or restriction patterns of PCR products

PCR product size and/or restriction site polymorphisms in the ITS region and 18S rRNA gene of the nuclear rDNA can be used to identify Geaumannomyces species and varieties of G. graminis (Fouly et al., 1997; Ward and Akrofi, 1994). Size variation in nuclear 18S and 26S rRNA gene PCR products was found to be due to presence/absence and size variation of group I introns (Fouly and Wilkinson, 2000a; Tan and Wong, 1996). Size variation of 18S rRNA gene introns was used to differentiate between isolates of Ggt, Gga, Ggg, G. incrustans, G. cylindrosporus and Ophioceras leptosporum (misnamed G. leptosporus in Fouly et al., 1997). Isolates of Ggg had no introns in the 26S gene, whereas isolates of Ggt and Gga had two or three introns (Tan, 1997). PCR primers were designed that subdivided isolates of Ggt into four groups on the basis of presence/absence of the three introns, and, in combination with restriction digestion, could discriminate between isolates of Ggt and Gga (Tan, 1997). Presence/absence of these introns was used for rapid identification of G. graminis isolates causing take-all patch disease in Bermuda grass (Wong et al., 2000).

Random amplified polymorphic DNA (RAPD) analyses

Various research groups have used RAPD analysis to discriminate between Gaeumannomyces and Phialophora species, varieties and isolates (Augustin et al., 1999; Bryan et al., 1999; Fouly et al., 1996; Ulrich et al., 2000; Wetzel et al., 1996), and to identify subgroups within G. graminis varieties (Augustin et al., 1999; Bryan et al., 1999; Fouly et al., 1996). RAPD analysis also successfully discriminated Gaeumannomyces spp., and some close relatives, from other darkly pigmented fungi commonly found on turf-grass roots (Wetzel et al., 1996). Isolates of Ggg showed more genetic variability than isolates of Ggt and Gga, consistent with results obtained using other molecular methods. Although genetic groupings identified by RAPD analyses often correlate well with identification by other methods, there are often problems in obtaining reproducible results and it is difficult to know whether different results obtained by different laboratories are genuine or result from slight differences in procedure. For example, the OPA-02 band identified by Wetzel et al. (1996) as characteristic of G. cylindrosporus/P. graminicola was not scored in the isolate of this species analysed by Fouly et al. (1996).

Dendrograms produced from analysis of RAPD data grouped G. graminis isolates according to variety (Bryan et al., 1999). Isolates of Ggt classified as rye-adapted (R) or non-rye-adapted (N) formed distinct groups in dendrograms. There was good correlation between these RAPD data and RFLP data and with phylogenetic analysis of nuclear rDNA sequences (see above; Bryan et al., 1995).

Specific PCR assays

Several PCR primer-pairs for specific detection of particular Gaeumannomyces species and varieties of G. graminis have been developed. Most are based on nuclear rDNA sequences. Bryan et al. (1995) reported specific detection of Ggt and Gga using PCR primer pairs pGt1/pGt2 and pGa1/pGa2, respectively, but sequence data presented in the same report suggest that pGt1/pGt2 may amplify DNA from at least one isolate of Ggg (E. Ward and J. Freeman, unpublished results).

Reverse PCR primers GGT-RP and GGA-RP, with NS5 (White et al., 1990) as the forward primer, amplify DNA from Ggt and Gga isolates but not DNA from isolates of Ggg or other Gaeumannomyces species tested (Fouly and Wilkinson, 2000b). Amplification by NS5/GGT-RP can distinguish between isolates of Ggt and isolates of Gga on the basis of PCR product size. A nested PCR assay specific to Gaeumannomyces, Phialophora and Magnaporthe species (Henson, 1992; Schesser et al., 1991) was subsequently simplified to a single-step PCR assay by Ward (1995). Ggt and Gga could be separated from the other varieties and species tested on the basis of PCR product size but PCR products from isolates of Ggm, P. radicicola Cain, P. zeicola and Cephalosporium maydis were all within the range of PCR product size obtained with Ggg (Ward and Bateman, 1999).

PCR primers GG1 and GG2 were designed for diagnosis of take-all patch of turf-grasses (Goodwin et al., 1995). These PCR primers amplified DNA from all isolates of Gga and Ggg tested, but not from the isolates of Ggt, P. graminicola or some other turf-grass pathogens and saprophytes that were tested.

Specific PCR assays for identification of Ggt, Ggg and Gga, based on avenacinase and avenacinase-like gene sequences, have recently been developed (Rachdawong et al., 2002). Variety-specific forward PCR primers were used, singly and together, with a common reverse PCR primer to give amplification products of different sizes from DNA of Ggt, Ggg and Gga isolates. However, some Ggt isolates gave PCR products of the size expected for Ggg isolates and the authors suggest further testing to confirm variety specificity.

Comparisons between different identification methods

A summary of the published molecular methods that can be used for identification of G. graminis is given in Table 2. Although many diverse methods are available, few of these can reliably identify all varieties of G. graminis, and there are many isolates that give atypical results. The difficulties of identification by traditional methods mean that many isolates that have been tested have not been identified to species or variety level, and this can lead to confusion in the literature. Generally, the numbers of isolates tested for each method have been small, or unrepresentative of the wide range of isolates that occur worldwide. There is only one report of molecular testing of the maize take-all fungus, G. graminis var. maydis (Ggm) and its putative anamorphs, P. zeicola and P. radicicola Cain, and this included only one isolate of each (Ward and Bateman, 1999). Several studies indicate that there is more variation in Ggg/Phialophora sp. (lobed-hyphopodia) than the other varieties, and more work is needed to determine the extent and significance of this variation. In addition, methods for identifying the fungi on turf-grasses, for example, may not be applicable on wheat or vice versa. Very little research on genetic diversity within populations of G. graminis has been reported. Harvey et al. (2001) demonstrated that similar cropping history was generally associated with close genetic identity of populations of G. graminis, whereas geographical proximity was not and concluded that populations may consist of series of host-adapted clonal lines. Unless isolated from the same field at the same time and therefore possibly clonally related, isolates of G. graminis are usually vegetatively incompatible with other isolates (Hornby et al., 1998). In view of the difficulties in achieving anastomosis between isolates of G. graminis in the laboratory (Asher, 1981), and of producing hybrid perithecia (see below), it seems unlikely that sexual hybridization is important in field populations of G. graminis.

Table 2.  Summary of features of the various DNA-based methods for the analysis of isolates of G. graminis and related species.
TargetMethodReferencesDiagnostic featuresComments
Nuclear rDNA—ITS regionDNA sequencingWard and Bateman (1999) Bryan et al. (1995)Conclusive identification possible.Useful for phylogenetic analysis. Widely used—many sequences available in databases. Time consuming—not practical for large-scale studies.
 RFLP analysis and size variation of PCR productsWard and Akrofi (1994) Fouly et al. (1997)Identifies isolates to variety.More than one enzyme required to identify all isolates to variety. Some isolates give RFLPs typical of other varieties.
PCR with specific primers pGt1 and pGt2 (Ggt) or pGa1 and pGa2 (Gga)Bryan et al. (1995)Specific identification of Ggt and Gga, respectively.pGt1 and pGt2 may possibly amplify DNA from some isolates of Ggg.
PCR with specific primers GG1 and GG2Goodwin et al. (1995)PCR specific for Ggg and Gga.Used to identify G. graminis isolates pathogenic to turf-grasses (Ggg and Gga) in the US. May not amplify DNA of UK isolates.
Nuclear rDNA—18S rDNASize variation of NS5 and NS6 PCR productsFouly et al. (1997) Fouly and Wilkinson (2000a)Isolates of Ggt and Gga have characteristic PCR product size.Based on very few isolates. Size of PCR products of Ggg isolates varies. Conclusive identification to species/variety not possible.
 PCR with GGT-RP or GGA-RP and NS5Fouly and Wilkinson (2000b)PCR specific for Ggt and Gga.PCR with GGT-RP and NS5 identifies isolates as Ggt or Gga on the basis of PCR product size. PCR with GGA-RP and NS5 can be used to identify isolates pathogenic to cereals (Ggt and Gga).
 RFLP analysis with 18S rDNA probeTan et al. (1994)Size of hybridizing fragment of different varieties varies.Does not identify isolates to species/variety.
Nuclear rDNA—26S rDNARFLP analysis with 26S rDNA probeTan et al. (1994)Can be used to discriminate cereal pathogens (Ggt and Gga) from Ggg. Identifies subgroups of Ggt isolates.Does not discriminate between isolates of Ggt and Gga.
 RFLP analysis and size variation of PCR productsTan et al. (1994) Tan and Wong (1996) Tan (1997) Wong et al. (2000)Identifies isolates of G.graminis to variety.Size variation of PCR products discriminates cereal pathogens (Ggt and Gga) from Ggg. RFLP analysis of PCR products discriminates between Ggt and Gga.
Other repeated DNA sequencesRFLPs of repeated DNA sequences (total Ggt DNA and wheat rDNA probes)O'Dell et al. (1992) Bryan et al. (1999)Can be used to indicate host preference of G. graminis isolates.For a few isolates RFLP results did not correlate with host range.
 Quantification of Ggt and RFLP analysis using probe pG158Herdina et al. (1996) Herdina et al. (1997) Herdina and Roget (2000)Probe is relatively specific for Ggt.Weak hybridization to Gga, even at high stringency.
Mitochondrial DNARFLP analysis with pMSU315Henson (1989) Henson (1992) Bateman et al. (1992) Ward and Bateman (1999)Can be used to discriminate cereal pathogens (Ggt and Gga) from Ggg.Does not discriminate between isolates of Ggt and Gga. Some isolates give atypical RFLP patterns. Does not discriminate between G-P complex isolates from maize.
 RFLP analysis with pEG34Ward and Gray (1992) Ward and Bateman (1994) Bateman et al. (1997) Ward and Bateman (1999)Identifies isolates of G-P complex to species and variety and identifies subgroups of G. graminis varieties.Cannot distinguish Ggm (and P. zeicola, P. radicicola Cain) from some isolates of Ggt.
 Nested PCR with KS1F and KS2R followed by KS4F and KS5RHenson (1989) Schesser et al. (1991) Henson (1992) Elliott et al. (1993) Henson (1993)PCR specific to G-P complex and Magnaporthe spp.Does not identify isolates to species/variety.
 PCR with KS1F and KS2R (including modifications by Ward, 1995)Ward (1995)Specific PCR for G. graminis. Discriminatescereal pathogens (Ggtand Gga) from Ggg.Does not identify isolates of Ggt and Gga to variety.
RandomRAPD analysisFouly et al. (1996) Wetzel et al. (1996) Bryan et al. (1999) Augustin et al. (1999) Ulrich et al. (2000)Can identify isolates to species and variety and identify subgroups within varieties.Generally analysis with more than one RAPD primer is required. Results can be inconsistent.
Avenacinase and avenacinase-like genesPCR analysis with variety-specific (Ggt, Gga and Ggg) forward primers and a common reverse primer (AV3)Rachdawong et al. (2002)Variety-specific PCR.All primers can be included in a single PCR reaction. Based on a single-copy gene, therefore may not be as sensitive as methods using rDNA or mitochondrial targets.Some Ggt isolates gave results typical of Ggg isolates.
Laccase genesDNA sequencingLitvintseva and Henson (2002a, 2002b)Potential alternative to nuclear rDNA sequencing for phylogenetic analyses.Time consuming therefore not practical for large-scale studies.

Very little work has been done to detect G. graminis directly in soil samples or in infected roots by molecular methods. Most of the work has involved isolating and culturing of the fungus. There are no methods available for reliably isolating G. graminis directly from soil, and therefore most isolates are obtained from field-grown plants or by growing susceptible plants in the soil as baits. Isolates are usually made from surface-sterilized lesioned root pieces (Hornby et al., 1998), but this could bias sampling towards more virulent isolates. Choosing an appropriate strategy for sampling fields for the presence of the take-all fungus, and determining its population structure, can be difficult, particularly when the disease is patchy. It is desirable that as much of the area is sampled as is practicable and several different procedures have been used to achieve this (Hornby et al., 1998).


  1. Top of page

The sexual stage in the life cycle of G. graminis involves the production of flask-shaped perithecia. Each of the many asci within the perithecia usually contains eight ascospores, resulting from one meiotic and one mitotic division. Because G. graminis is homothallic (self-fertile), attempts to perform genetic crosses often result in relatively small proportions of hybrid perithecia compared with the numbers that result from selfing of each parental strain (Asher, 1981; Blanch et al., 1981). In intra-isolate crosses, using mutants or antibiotic-resistant transformants, the percentage of hybrid perithecia was generally around 10–25% but sometimes higher frequencies were obtained (Hornby et al., 1998; Musker, 1994; Pilgeram and Henson, 1992). Also, numerous vegetative incompatibility groups exist within G. graminis (Asher, 1981; Jamil and Buck, 1991; Jamil et al., 1984) and this places restrictions on the isolate combinations that can be sucessfully paired. Attempts to make forced heterokaryons by pairing isolates on selective medium produced stable heterokaryons only when mutants of the same isolate were used. Other pairings between different isolates of the same variety or between isolates of Ggt and Gga were unsuccessful (Hornby et al., 1998; Musker, 1994). Further complications arise because the ascospores of G. graminis are loosely coiled together within the ascus (Fig. 1b) and it is not possible to distinguish the first (meiotic) and second (mitotic) generation products (i.e. the tetrad is unordered). This makes genetic analysis more difficult and involves the analysis of random meiotic products (Asher, 1981).

Understanding the genetics of G. graminis has also been hampered by the lack of available mutants. Mutations would be useful, for example in facilitating gene mapping and in providing novel phenotypic markers for use in genetic crosses and population studies. The generation of G. graminis mutants has been hindered by problems in producing sufficient numbers of germinable propagules from uninucleate material; uninucleate conidia are difficult to produce and/or germinate, and ascospores and the vegetative haploid hyphae are multinucleate. Various treatments, such as irradiation with UV or exposure to the chemical mutagens nitrosoguanidine and 4-nitroquinolene oxide, have been used to mutate G. graminis. The most successful approach has been to treat protoplasts with mutagenic agents, although unpredictable regeneration frequencies can be a problem. Despite the difficulties, several mutants have been isolated, including those with resistance to the fungicides benomyl and carboxin, altered morphology or pigmentation and a variety of auxotrophic mutations (Bowyer et al., 1992; Frederick et al., 1999).

Another method for generating mutants is by stable transformation. All published reports of transformation of G. graminis have used PEG-CaCl2-mediated protoplast transformation. Benomyl- and phleomycin-resistant transformants of Ggg and Ggt have been obtained using this method, but the transformation efficiency is very low (1–5 transformants/µg DNA per 107 protoplasts) (Bowyer et al., 1995; Henson et al., 1988; Pilgeram and Henson, 1990), and varies greatly from experiment to experiment (P. Bowyer, personal communication). The stability of the integrated DNA varies. Transgenes in the majority of transformants are mitotically stable, but their stability through meiosis varies with copy number and with different genes (Pilgeram and Henson, 1990, 1992; Pilgeram et al., 1993). For example, benomyl- and phleomycin-resistant transformants of Ggt and Ggg were mitotically stable irrespective of transgene copy number. In most cases, G. graminis transformants with a single copy of the transgene were phenotypically and genotypically stable through meiosis, but, of six transformants with multiple transgene copies, four were both phenotypically and genotypically unstable, and stability did not correlate with either benomyl or phleomycin resistance (Pilgeram and Henson, 1992).

Various mutants have been isolated from G. graminis following transformation and the random insertion of plasmids conferring benomyl- and phleomycin-resistance. These include a nicotinic acid auxotroph of Ggt, a pigment mutant and several Ggg mutants with abnormal/no hyphopodia (Epstein et al., 1994; Pilgeram and Henson, 1992). Targeted gene disruption is a much more effective and precise method of generating mutants provided the requisite gene has been cloned, but we have found only one successful report of its use in G. graminis, namely to produce avenacinase mutants of Gga (Bowyer et al., 1995).

The poor transformation frequencies that are obtained using G. graminis has caused great difficulties understanding gene function in Ggt via gene disruption experiments. As for most fungal species, the technique at present is inadequate for identification of DNA clones from genomic libraries by complementation of mutants. A novel method, involving recombination between linear genomic DNA and a fungal replicating vector (pHELP1), and complementation of Aspergillus nidulans mutants, was used to clone the argB and pyrG genes of Gga (Bowyer et al., 1994). This method relies on a suitable mutant strain being available and on the abilty of the target species gene to complement the mutant gene in A. nidulans. Only two of seven mutants were complemented and the genes responsible for complementation were rescued into Escerichia coli for further analyses.


  1. Top of page


Isolates of the wheat take-all fungus (Ggt) are unable to infect oats, whereas Gga can. Oat roots produce the saponin avenacin, a glycosylated triterpenoid secondary metabolite (Fig. 2), which is a constitutively present antimicrobial agent (phytoanticipin) that can inhibit a wide range of fungi (Osbourn, 1996a,b). Some fungi, including Gga, can detoxify avenacin using the enzyme avenacinase (Fig. 2), whereas others such as Ggt cannot. A series of studies have demonstrated that avenacinase is required for infection of oats by Gga, and that it is involved in host-range determination. The Gga avenacinase gene was cloned using anti-avenacinase antisera to screen cDNA expression libraries of Gga, and this was then used to generate mutants using targeted gene disruption (Bowyer et al., 1995). The avenacinase-minus mutant was no longer pathogenic on oats but was still able to infect wheat. Ggt and some other members of the Gaeumannomyces–Phialophora complex also contain DNA sequences homologous to the avenacinase gene even though they are unable to inactivate avenacin effectively (Osbourn, 1996a; Osbourn et al., 1994a; Rachdawong et al., 2002). The role of these avenacinase-like genes is uncertain but it has been suggested that they might be involved in producing enzymes that detoxify other saponins in hosts other than oats (e.g. grasses). When a wide range of Avena species were screened, Avena longiglumis was the only one found not to produce avenacin, and this was also the only species that was susceptible to Ggt (Osbourn et al., 1994b). Further evidence for the role of avenacin as a resistance determinant was obtained by studies of saponin deficient (sad) mutants of the diploid oat species Avena strigosa, which were isolated using the UV fluorescence properties of avenacin as a preliminary screen (Papadopoulou et al., 1999). These mutants were compromised in their resistance to Ggt as well as Fusarium culmorum and F. avenaceum, but not the leaf-infecting fungi Stagonospora nodorum and S. avenae. The gene encoding the enzyme that catalyses the first committed step in avenacin biosynthesis (β-amyrin synthase, AsbAS1) has recently been cloned and shown to be a novel oxidosqualene cyclase (Haralampidis et al., 2001). It was demonstrated that Sad1 mutants of A. strigosa are mutated in the gene encoding this enzyme. The AsbAS1 gene is highly conserved in oat, but is absent from other cereals such as wheat. Manipulation of these pathways, by genetic engineering or other means, might eventually provide a means of generating resistance to Ggt and other pathogens in wheat, although this could be compromised if there were an increase in the prevalence of Gga, or Ggt isolates that can infect oats (Bryan et al., 1999; Hornby et al., 1998).


Figure 2. Structure of the oat saponin avenacin A-1. Avanacin is produced in the roots of oat plants, but not in wheat or barley roots, and is located in a circle around the periphery of healthy oat roots. The oat-attacking variety of Gaeumannomyces graminis (G. graminis var. avanae (Gga)) produces the saponin-detoxifying enzyme, avenacinase, that detoxifies avanacin by removing the 1[RIGHTWARDS ARROW]2- and 1[RIGHTWARDS ARROW]4-linked terminal d-glucose (glu) sugar molecules.

Download figure to PowerPoint


Various studies have investigated the synthesis and function of melanin in G. graminis varieties. Melanin is a dark polymeric pigment, found in many phytopathogenic fungi, which has a role in the pathogenicity of some fungal species, whereas in other species it protects against environmental stresses such as extremes of temperature and pH, ionizing radiation, oxidative stress, desiccation, metals, plant defence mechanisms and the lytic action of microorganisms in soil (Bell and Wheeler, 1986; Butler and Day, 1998; Butler et al., 2001; Henson et al., 1999; Langfelder et al., 2003).

There are several different pathways for fungal melanin biosynthesis, but the best characterized and most commonly utilized pathway is the dihydroxynaphthalene (DHN)–melanin pathway. Studies with melanin inhibitors have shown that this is also the pathway used by G. graminis (Elliott, 1995; Kelly, 1997). Melanin can, however, be produced from L-dihydroxyphenylalanine (L-DOPA), when added to synthetic media, by G. graminis and this is the basis of a diagnostic selective medium for G. graminis (Juhnke et al., 1984). Extracellular laccases secreted by the fungus are able to convert the L-DOPA to melanin, but there is no evidence that the L-DOPA pathway for melanin synthesis occurs under normal growth conditions.

In certain phytopathogenic fungi, such as Magnaporthe grisea, which use melanized appressoria as their sole means of host surface penetration, melanin is essential for pathogenicity (Henson et al., 1999; Howard and Valent, 1996). In M. grisea, it has been demonstrated that melanin limits cell-wall permeability, allowing osmolyte accumulation and the build up of enormous turgor pressures in the appressoria (Howard et al., 1991; Money, 1995). This process creates the forces required for the penetration peg to break the host cell cuticle and epidermal cell wall. However, it has been proposed that this may not be necessary to overcome the less substantial barriers to infection in the roots; M. grisea mutants unable to synthesize DHN–melanin (alb1 and buf1) were unable to infect wheat leaves but were still able to infect wheat roots (Dufresne and Osbourn, 2001; Osbourn, 2001). G. graminis produces melanized appressorium-like structures called hyphopodia, but these develop from the end of hyphae rather than conidial germ tubes, and their role, whether as a platform for host infection or otherwise, is unclear.

The role of melanin in G. graminis has been investigated in several studies using mutants and inhibitors of melanin synthesis. In Ggg, a mutant (thr) was generated that was unable to synthesize melanin and another (moe) that was constitutively melanized (Frederick et al., 1999). The non-melanized mutant was found to be unaltered in its pathogenicity on rice compared with the wild-type, but the constitutively melanized mutant was actually less pathogenic than the wild-type on rice. The explanation for this phenotype may be that increased amounts of melanin in the hyphopodia limit their ability to secrete the various extracellular lytic enzymes that are required for host tissue degradation and subsequent invasion. Heavily melanized variants of Ggt have also been reported to be non-pathogenic on wheat (Goins et al., 2002). The wild-type and constitutively melanized Ggg mutant were found to be more hydrophobic, and more resistant to lytic enzymes, benomyl, restrictive temperature and UV light than the non-melanized mutant (Frederick et al., 1999). Measurements of hyphopodial turgor were performed on these Ggg strains (Money et al., 1998) and, although it was demonstrated that melanin synthesis was associated with changes in hyphopodial turgor, permeability and cell-wall rigidity, the pressures generated were much lower than in M. grisea.

In Ggt, there is some evidence that melanin is required for pathogenicity because an albino mutant was demonstrated to be non-pathogenic and the DHN melanin synthesis inhibitors tricyclazole and PP389 reduced the pathogenicity of Ggt isolates (Kelly, 1997; Kelly et al., 1997). Earlier work by Elliott (1995) had indicated that inhibitors of DHN–melanin synthesis did not alter pathogenicity, but this study used lower concentrations of the inhibitors. It seems likely that any requirement for melanin in pathogenicity of Ggt may not be for host penetration, but may instead be needed for the production of runner hyphae on roots (Dufresne and Osbourn, 2001; Henson et al., 1999).

The role of melanin in G. graminis therefore remains unclear, particularly in relation to pathogenesis, and may be different in different members of the Gaeumannomyces–Phialophora complex. However, it seems likely that melanin plays a role in protection against environmental stresses, a property that may be particularly relevant in a soil-borne pathogen such as G. graminis that survives saprophytically in the field between susceptible host crops.


G. graminis isolates secrete various laccases that accumulate in their cell walls and extracellularly. This is the basis of a diagnostic semiselective medium for G. graminis (Juhnke et al., 1984), which converts L-DOPA in the medium to melanin. The final polymerization reaction in the DHN–melanin pathway, the oxidation of 1,8-DHN into melanin, is also thought to be mediated by a laccase; extracellular laccase purified from Ggt has been shown to convert 1,8-DHN into polymers (Edens et al., 1999). A monoclonal antibody has been produced for which the target antigen appears to be a Ggt laccase gene and this has been used in immunofluorescence microscopy studies and in diagnostic assays for G. graminis (Thornton et al., 1997). The immunofluorescence studies showed that the antigen was associated with both the melanized runner hyphae and the hyaline mycelia.

Laccases are widely distributed enzymes and individual fungi can produce several different laccase enzymes, each of which may have a variety of roles. This is likely to complicate the determination of their individual functions, and mutants with multiple disruptions may be necessary. One of the main roles of laccases is in melanin synthesis. They may also be involved in degradation of lignin, metabolizing phytotoxic substances and oxidation of humic acids and manganese ions (Litvintseva and Henson, 2002a). The Ggt extracellular laccase will decolorize the lignin-like dye poly B-411, indicating that it may have a role in lignin depolymerization in the plant host (Edens et al., 1999).

Three genes encoding laccase enzymes have recently been cloned from Ggt by screening genomic libraries using a laccase gene probe from Neurospora crassa. Their relative expression patterns were studied using RT-PCR (Litvintseva and Henson, 2002a). LAC1 was transcribed constitutively, but LAC2 was copper inducible and corresponded to the enzyme purified by Edens et al. (1999). All three genes were transcribed in planta, but transcription of LAC3 was found only in planta or in the presence of wheat homogenate.

A full-length LAC1 genomic clone was isolated from Ggg using the Ggt LAC1 gene as a probe, and the LAC2 genomic clone was isolated using PCR with primers based on the Ggt sequence (Litvintseva and Henson, 2002b). The LAC1 and LAC2 sequences were highly conserved between different Ggg and Ggt isolates, but the expression patterns were different. In Ggg, a high level of LAC1 transcription was seen only in the presence of the plant, and copper had no effect on the transcription of LAC2. There was some evidence for the presence of a LAC3 gene in Ggg but, using RT-PCR, no transcription was detected under any of the conditions used. Expression of the LAC1 and LAC2 genes was also monitored in mutants with altered melanin synthesis. The unmelanized (thr) mutant showed similar levels of expression of the two genes to the wild-type, but the over-melanized mutant showed higher levels of LAC1 expression and lower levels of LAC2 expression than the wild-type.

Cell-wall-degrading enzymes

Fungal plant pathogens produce a range of enzymes capable of degrading the components of plant cell walls, and these enzymes may play a role in nutrition and cell penetration (Annis and Goodwin, 1997; Cooper, 1983; Cooper et al., 1988). Various biochemical studies have demonstrated that Ggt can produce cellulases, xylanases and pectinases in infected wheat roots and/or in vitro, but there are discrepancies between some of these reports as to the particular enzymes that are produced by Ggt, and in the correlations between their activities and pathogenicity (reviewed by Hornby et al., 1998; Sivasithamparam and Parker, 1981). Also, in other fungal pathogens, studies with mutants that are defective in individual enzymes have usually failed to show that these enzymes are needed for pathogenicity or virulence. This is probably because most fungi produce multiple forms of many of these enzymes and mutation of a single enzyme does not therefore affect the ability of the fungus to degrade the cell-wall polymer (Annis and Goodwin, 1997). It may therefore prove difficult to demonstrate any links between cell-wall-degrading enzymes and pathogenicity in Ggt even if mutants defective in these enzymes could be isolated.

Cell-wall-degrading enzymes of G. graminis have also been studied indirectly by monitoring their effects on wheat roots (Huang et al., 2001a; Kang et al., 2000). Various cell-wall components were studied in uninoculated and Ggt-infected wheat roots using electron microscopy with enzyme-gold and immunogold labelling techniques (Kang et al., 2000). The labelling densities of cellulose, xylan and pectin in the cell walls of infected roots were significantly reduced compared with healthy wheat roots, indicating that Ggt produces enzymes that degrade these cell-wall components. Comparisons were also made between root tissues infected by Ggt or P. graminicola (Huang et al., 2001a). The results indicated that P. graminicola caused only slight modification of the cortical cell walls compared with Ggt, suggesting that P. graminicola produces less cell-wall-degrading enzymes than Ggt during infection of wheat roots.

Enzymes involved in fatty acid metabolism

A number of recent studies of G. graminis have focused on enzymes involved in the production of oxylipins, biologically active compounds generated by oxidative metabolism of polyunsaturated fatty acids. Two fatty acid dioxygenase enzymes were recently purified and characterized from G. graminis and the resulting peptide sequence information used to clone the genes. These were a lipoxygenase gene that contains manganese at its active site (Hörnsten et al., 2002b; Su and Oliw, 1998), and a linoleate diol synthase (Hörnsten et al., 1999; Su and Oliw, 1996). The roles of these enzymes remain to be elucidated but, in other organisms, oxylipins are thought to have important biological roles in signalling, the degradation of cellular membranes, reproduction and development (Feussner et al., 2001; Herman, 1998; Hörnsten et al., 1999, 2002a; Howe and Schilmiller, 2002). The G. graminis linoleate diol synthase gene was recently shown to be similar to a gene thought to be involved in spore formation in Ustilago maydis (Huber et al., 2002). There is currently much research activity in this area and it is likely that there will be interesting developments in the near future.

Cyclic hydroxamic acids

The first identification of antifungal activity of a hydroxamic acid from a cereal was nearly 50 years ago (Virtanen and Hietala, 1955). Since then, evidence has suggested roles for hydroxamic acids in defence against fungi, bacteria and insects, in the detoxification of herbicides and in allelopathic effects (Niemeyer, 1988). Cyclic hydroxamic acids are found almost exclusively in the Gramineae (Frey et al., 1997). They have a 4-hydroxy-1,4-benzoxazin-3-one structure, and are found constitutively in wheat, rye, triticale, maize and sorghum, but are not present in barley, rice or oats (Morrissey and Osbourn, 1999; Niemeyer, 1988). In planta, hydroxamic acids are sequestered as inactive glucosides, but are hydrolysed after infection or tissue damage to aglucones, e.g. 2,4-dihydroxy-1,4-benzoxazin-3-one (DIBOA) and 2,4-dihydroxy-7-methoxy-1,4-benzoxazin-3-one (DIMBOA), which decompose to BOA and MBOA, respectively (Friebe et al., 1998; Morrissey and Osbourn, 1999). Friebe et al. (1998) studied the ability of G. graminis varieties to degrade and tolerate MBOA and BOA. The results did not reveal any clear correlations with pathogenicity. However, other work has demonstrated correlations between susceptibility of cereals to take-all and hydroxamic acid production (Wilkes et al., 1999). Wheat, rye and triticale, which differ in their susceptibility to Ggt, all contain DIMBOA, whereas only rye and some cultivars of triticale contain DIBOA in addition to DIMBOA. Root extracts from wheat did not inhibit the growth of Ggt, but those of rye and triticale did, and DIBOA and BOA were more potent growth inhibitors of Ggt than DIMBOA and MBOA. The effects of hydroxamic acids on the growth of G. graminis are fungistatic rather than fungicidal (Frey et al., 1997; Wilkes et al., 1999). Wilkes et al. (1999) estimated the concentration of DIBOA in planta to be in the range 0.3–2 mm. DIBOA severely inhibited the growth of Ggt at 0.5–1.0 mm, and thus growth inhibition can occur at concentrations estimated to occur in planta (Wilkes et al., 1999). In cereals, hydroxamic acids are found in all plant parts, but are generally most abundant in the vascular tissues and in young seedlings (Morrissey and Osbourn, 1999; Niemeyer, 1988). In wheat, rye and triticale roots, amounts of hydroxamic acids are greater in the growing root tip and the root stele than behind the root meristem or in the cortex (Niemeyer, 1988; Wilkes et al., 1999), consistent with a role in defence against Ggt infection. A gene cluster encoding the enzymes of the DIBOA biosynthetic pathway has been identified in maize (Frey et al., 1997). Transfer of these genes to wheat may offer an alternative approach to the development of wheat lines less susceptible to take-all.

A PMK1-related MAP kinase

Magnaporthe grisea, although closely related to G. graminis, usually infects the leaves of cereals by means of appressoria formed by conidial germ tubes. This pathogen has recently been demonstrated to cause lesions on wheat and barley roots, but infection of roots does not require the formation of appressoria (Dufresne and Osbourn, 2001). A number of M. grisea mutants were tested for their ability to infect wheat and barley leaves and roots (Dufresne and Osbourn, 2001). Some of the mutants (pmk1) were deficient in MAP (mitogen activated protein) kinase activity. PMK1-related MAP kinases have been shown to be required for virulence of a number of fungi, including the root-infecting fungus Fusarium oxysporum f.sp. lycopersici (Di Pietro et al., 2001; Xu, 2000). The M. grisea pmk1 (MAP kinase deficient) mutants were unable to infect leaves of rice and barley and were also unable to cause lesions on wheat and barley roots. A G. graminis homologue of PMK1 (GMK1) was cloned, using a probe made using degenerate PCR primers designed for amplification of FUS3-related MAP kinases, and expressed in M. grisea pmk1 mutants. All defects of the pmk1 mutants were complemented by the G. graminis gene, and transformants were able to infect leaves and roots, but whether GMK1 has a role in the pathogenicity of G. graminis is still unclear. Until more efficient transformation of G. graminis and other related root-infecting fungi is possible, the complementation of M. grisea mutants offers some prospects for studying genes that may be involved in the pathogenicity of G. graminis and that have previously been identified in other fungal species.


  1. Top of page

Viruses and double-stranded RNAs (dsRNAs) are widespread in G. graminis but, to date, it is not clear what effects they have on their hosts. In several fungi, dsRNA has been shown to be associated with hypovirulence, notably Cryphonectria parasitica, where a cause and effect relationship has been proven (Dawe and Nuss, 2001). In some fungi (e.g. Saccharomyces cerevisiae and Ustilago maydis) dsRNAs have been shown to be associated with toxin production, and in others (e.g. Helminthosporium victoriae and Ophiostoma novo-ulmi) they may cause disease of the fungus (Buck, 1998). Viruses present in G. graminis mycelium were the subject of much research in the 1970s and early 1980s when it was thought that they might be associated with hypovirulence and/or take-all decline (reviewed by Buck, 1986, 1997; Rawlinson and Buck, 1981).

The viruses do not have an extracellular phase and are probably transmitted by vegetative (somatic hyphal) fusion or during mating of isolates. They are sometimes, but not always, eliminated during ascospore formation. G. graminis viruses are isometric and have dsRNA genomes. There is much variation in the properties of these viruses and they have been classified into five groups depending on the physical properties of the particle, the capsid polypeptide serology, and the numbers and sizes of the dsRNAs they contain (Buck, 1986). Similar viruses have been observed in different vegetative compatibility groups of Ggt, the other varieties of G. graminis and P. graminicola (Jamil and Buck, 1991; Jamil et al., 1984; Liang and Chen, 1990; Liang et al., 1989; McGinty et al., 1981). The distribution of G. graminis viruses is complex; Ggt isolates from a single field can have different combinations of several different viruses. This has created difficulties in understanding the relationships among virus, fungus and pathogenicity. The work done to date suggests that viruses in general do not modify pathogenicity, but specific viral RNAs (or combinations of RNAs) may have a role in this. In Rhizoctonia solani, there was also much controversy for many years concerning the effect of dsRNAs on virulence. More recently, however, an explanation for some of the conflicting findings emerged, when distinct dsRNAs were found to be associated with enhanced or diminished virulence (Jian et al., 1997). Further characterization of dsRNAs using molecular techniques could help clarify their role(s), if any, in G. graminis. There appears to have been no research in this area since the early 1990s.

Linear dsDNA plasmids also appear to be common in G. graminis (Henson and Ton That, 1995; Honeyman and Currier, 1986). These range from around 4 kb to 11 kb and are found in the mitochondria. A wide range of G. graminis strains from diverse geographical locations contain these elements but some strains are plasmid-free. Most G. graminis strains analysed had one or two plasmids and the most observed in any one strain was four. The dsDNAs show varying degrees of homology to one another. In other fungi, plasmids have been shown to encode toxins or virulence factors or to be involved in senescence but, to date, it has not been possible to ascribe a function to any of the G. graminis plasmids (Henson and Ton That, 1995).


  1. Top of page

Considerable progress has been made as a result of many years of intensive research on take-all, but there is still much that we do not understand about the disease and the fungi that cause it. Take-all control is hampered by the economics of cereal growing and the need for consecutive cropping, and presents a major challenge in plant pathology. Experimental studies on take-all and effective disease control are slow to perform because of the unpredictability of the disease and its occurrence in patches (epidemics can progress at different rates in different parts of the same field). As for many diseases caused by soil-borne pathogens, the impacts of environmental factors (biological and physical) on disease severity are difficult to measure and predict, and, especially for take-all, can be very different in different parts of the world. Control measures that are effective in the USA, for example, may not be effective in Europe or Australia (Hornby et al., 1998). Much innovative research on potential biological control agents has not led to any widely available, effective commercial products (reviewed in Cook, 2003). However, the recently available fungicides silthiofam and fluquinconazole, applied as seed treatments, offer some promise for controlling the disease, but their effective and economic use depends on accurate targeting of those crops that will benefit from treatment (Bateman et al., 2003). Also, their long-term future could be compromised if resistance/insensitivity or other factors reduce their effectiveness.

Identification of pathogenic and non-pathogenic Gaeumannomyces and Phialophora species has improved with the advent of molecular diagnostic methods although, again, techniques developed in one part of the world may not be universally applicable. There have also been significant advances in elucidating the role of avenacinase in the pathogenicity of the oat take-all fungus. However, progress in understanding other mechanisms determining pathogenicity, and differences between G. graminis varieties, has been hampered by the difficulties of performing genetic analyses with the fungi and the lack of an efficient, stable transformation system. Future advances may be helped by the availability of genome sequence data from other fungal plant pathogens, which could allow the isolation of interesting related genes from G. graminis, and then testing for function in other fungi such as the related Magnaporthe grisea (Osbourn, 2001). Wheat germplasm that is naturally resistant to take-all does not exist. A transgenic plant approach may offer research opportunities. For example, transfer of the genes encoding enzymes involved in biosynthesis of avenacin or cyclic hydroxamic acids to wheat would offer the ability to test whether these compounds could play a role in making wheat lines less susceptible to take-all.

Achieving effective and complete control of take-all is unlikely to be achieved by a single approach, because of the number of interacting factors involved in disease incidence and severity. The use of cultural control measures is likely to remain an important component of any future disease management strategy.


  1. Top of page

Rothamsted Research receives grant-aided support from the Biotechnology and Biological Sciences Research Council of the United Kingdom. J.F. was funded by Defra, HGCA and Aventis through the Sustainable Arable LINK project (LK0907). We thank Richard Gutteridge for providing photographs and Geoff Bateman, John Jenkyn, Richard Gutteridge and Kim Hammond-Kosack for helpful comments on the manuscript.


  1. Top of page
  • Andrade, O.A., Mathre, D.E. and Sands, D.C. (1994) Natural suppression of take-all disease of wheat in Montana soils. Plant Soil, 164, 918.
  • Annis, S.L. and Goodwin, P.H. (1997) Recent advances in the molecular genetics of plant cell wall-degrading enzymes produced by plant pathogenic fungi. Eur. J. Plant Pathol, 103, 114.
  • Asher, M.J.C. (1981) Pathogenic variation. In Biology and Control of Take-All (Asher, M.J.C. and Shipton, P.J., eds). London: Academic Press, pp. 199218.
  • Asher, M.J.C. and Shipton, P.J. (1981) Biology and Control of Take-All. London: Academic Press.
  • Augustin, C., Ulrich, K., Ward, E. and Werner, A. (1999) RAPD-based inter- and intravarietal classification of fungi of the Gaeumannomyces–Phialophora complex. J. Phytopathol, 147, 109117.
  • Bateman, G.L. (1989) Progress in research on the control of take-all in cereals with fungicides—a review. Crop Prot., 8, 7581.
  • Bateman, G.L., Freeman, J., Gutteridge, R.J., Jenkyn, J.F. and Ward, E. (2003) Strategies for Fungicidal Control of Take-All. HGCA project report 309. London: Home-Grown Cereals Authority.
  • Bateman, G.L., Ward, E. and Antoniw, J.F. (1992) Identification of Gaeumannomyces graminis var. tritici and G. graminis var. avenae using a DNA probe and non-molecular methods. Mycol. Res. 96, 737742.
  • Bateman, G.L., Ward, E., Hornby, D. and Gutteridge, R.J. (1997) Comparisons of isolates of the take-all fungus, Gaeumannomyces graminis var. tritici, from different cereal sequences using DNA probes and non-molecular methods. Soil Biol. Biochem. 29, 12251232.
  • Beale, R.E., Phillion, D.P., Headrick, J.M., O'Reilly, P. and Cox, J. (1998) MON65500: a unique fungicide for the control of take-all in wheat. In Proceedings of the 1986 British Crop Protection Conference—Pests and Diseases. Farnham, UK: British Crop Protection Council, pp. 343350.
  • Bell, A.A. and Wheeler, M.H. (1986) Biosynthesis and functions of fungal melanins. Annu. Rev. Phytopathol. 24, 411451.
  • Blanch, P.A., Asher, M.J.C. and Burnett, J.H. (1981) Inheritance of pathogenicity and cultural characters in Gaeumannomyces graminis var. tritici. Trans. Br. Mycol. Soc. 77, 391399.
  • Bowyer, P., Clarke, B.R., Lunness, P., Daniels, M.J. and Osbourn, A.E. (1995) Host-range of a plant-pathogenic fungus determined by a saponin detoxifying enzyme. Science, 267, 371374.
  • Bowyer, P., Musker, R., Osbourn, A.E., Clarke, B., Caten, C. and Daniels, M.J. (1992) Production of mutants of Gaeumannomyces graminis var. tritici and var. avenae by 4-nitroquinolene-oxide treatment of protoplasts. Fungal Genet. Newsletter, 39, 1314.
  • Bowyer, P., Osbourn, A.E. and Daniels, M.J. (1994) An instant gene bank method for heterologous gene cloning—complementation of 2 Aspergillus nidulans mutants with Gaeumannomyces graminis DNA. Mol. Gen. Genet. 242, 448454.
  • Bryan, G.T., Daniels, M.J. and Osbourn, A.E. (1995) Comparison of fungi within the Gaeumannomyces–Phialophora complex by analysis of ribosomal DNA sequences. Appl. Environ. Microbiol. 61, 681689.
  • Bryan, G.T., Labourdette, E., Melton, R.E., Nicholson, P., Daniels, M.J. and Osbourn, A.E. (1999) DNA polymorphism and host range in the take-all fungus, Gaeumannomyces graminis. Mycol. Res. 103, 319327.
  • Buck, K.W. (1986) Viruses of the wheat take-all fungus Gaeumannomyces graminis var. tritici. In Fungal Viruses (Buck, K.W., ed.). Boca Raton, FL: CRC Press, pp. 221236.
  • Buck, K.W. (1997) Viruses of plant pathogenic fungi. In Genetics and Plant Pathogenesis (Day, P.R. and Jellis, G.J., eds). Oxford, UK: Blackwell Scientific Publications, pp. 111126.
  • Buck, K.W. (1998) Molecular variability of viruses of fungi. In Molecular Variability of Fungal Pathogens (Bridge, P.D., Couteaudier, Y. and Clarkson, J.M., eds). Wallingford, UK: CAB International, pp. 5372.
  • Bunting, T.E., Plumley, K.A., Clarke, B.B. and Hillman, B.I. (1996) Identification of Magnaporthe poae by PCR and examination of its relationship to other fungi by analysis of their nuclear rDNA ITS-1 regions. Phytopathology, 86, 398404.
  • Bussaban, B., Lumyong, S., Lumyong, P., Hyde, K.D. and McKenzie, E.H.C. (2001a) Two new species of endophytes (ascomycetes) from Zingiberaceae sporulating in culture. Nova Hedwigia, 73, 487493.
  • Bussaban, B., Lumyong, S., Lumyong, P., McKenzie, E.H.C. and Hyde, K.D. (2001b) Endophytic fungi from Amomum siamense. Can. J. Microbiol. 47, 943948.
  • Butler, M.J. and Day, A.W. (1998) Fungal melanins: a review. Can. J. Microbiol. 44, 11151136.
  • Butler, M.J., Day, A.W., Henson, J.M. and Money, N.P. (2001) Pathogenic properties of fungal melanins. Mycologia, 93, 18.
  • Cannon, P.F. (1994) The newly recognized family Magnaporthaceae and its interrelationships. Systema Ascomycetum, 13, 2542.
  • Capper, A.L. and Higgins, K.P. (1993) Application of Pseudomonas fluorescens isolates to wheat as potential biological control agents against take-all. Plant Pathol. 42, 560567.
  • Chapon, A., Guillerm, A.Y., Delalande, L., Lebreton, L. and Sarniguet, A. (2002) Dominant colonisation of wheat roots by Pseudomonas fluorescens Pf29A and selection of the indigenous microflora in the presence of the take-all fungus. Eur. J. Plant Pathol. 108, 449459.
  • Cook, R.J. (2003) Take-all of wheat. Physiol. Mol. Plant Pathol. 62, 7386.
  • Cook, R.J., Thomashow, L.S., Weller, D.M., Fujimoto, D., Mazzola, M., Bangera, G. and Kim, D. (1995) Molecular mechanisms of defense by rhizobacteria against root disease. Proc. Natl Acad. Sci. USA, 92, 41974201.
  • Cooper, R.M. (1983) The mechanisms and significance of enzymic degradation of host cell walls by parasites. In Biochemical Plant Pathology (Callow, J.A., ed.). Chichester, UK: Wiley, pp. 101137.
  • Cooper, R.M., Longman, D., Campbell, A., Henry, M. and Less, P.E. (1988) Enzymic adaptation of cereal pathogens to the monocotyledonous primary wall. Physiol. Mol. Plant Pathol, 32, 3347.
  • Datnoff, L.E. (1993) Black sheath rot caused by Gaeumannomyces graminis var. graminis on rice in Florida. Plant Dis. 77, 210 .
  • Dawe, A.L. and Nuss, D.L. (2001) Hypoviruses and chestnut blight: exploiting viruses to understand and modulate fungal pathogenesis. Annu. Rev. Genet. 35, 129.
  • Deacon, J.W. (1981) Ecological relationships with other fungi: competitors and hyperparasites. In Biology and Control of Take-All (Asher, M.J.C. and Shipton, P.J., eds). London: Academic Press, pp. 75101.
  • Dewan, M.M. and Sivasithamparam, K. (1988) Idebtity and frequency of occurence of Trichoderma spp. in roots of wheat and rye-grass in Western Australia and their effects on root rot caused by Gaeumannomyces graminis var. tritici. Plant Soil, 109, 93101.
  • Di Pietro, A., García-Maceira, F.I., Méglecz, E. and Roncero, M.I.G. (2001) A MAP kinase of the vascular wilt fungus Fusarium oxysporum is essential for root penetration and pathogenesis. Mol. Microbiol. 39, 11401152.
  • Duffy, B.K., Simon, A. and Weller, D.M. (1996) Combination of Trichoderma koningii with fluorescent pseudomonads for control of take-all on wheat. Phytopathology, 86, 188194.
  • Duffy, B.K. and Weller, D.M. (1995) Use of Gaeumannomyces graminis var. graminis alone and in combination with fluorescent Pseudomonas spp. to suppress take-all of wheat. Plant Dis. 79, 907911.
  • Duffy, B.K. and Weller, D.M. (1996) Biological control of take-all of wheat in the Pacific Northwest of the USA using hypovirulent Gaeumannomyces graminis var. tritici and fluorescent pseudomonads. J. Phytopathol. 144, 585590.
  • Dufresne, M. and Osbourn, A.E. (2001) Definition of tissue-specific and general requirements for plant infection in a phytopathogenic fungus. Mol. Plant–Microbe Interact. 14, 300307.
  • Eastwood, R.F., Kollmorgen, J.F., Hannah, M. and Williams, W.M. (1994) Reaction of somaclonal variants of wheat to the take-all fungus (Gaeumannomyces graminis var. tritici). Plant Pathol. 43, 644650.
  • Edens, W.A., Goins, T.Q., Dooley, D. and Henson, J.M. (1999) Purification and characterization of a secreted laccase of Gaeumannomyces graminis var. tritici. Appl. Environ. Microbiol. 65, 30713074.
  • Elliott, M.L. (1991) Determination of an etiological agent of bermudagrass decline. Phytopathology, 81, 13801384.
  • Elliott, M.L. (1995) Effect of melanin biosynthesis inhibiting compounds on Gaeumannomyces species. Mycologia, 87, 370374.
  • Elliott, M.L., Des Jardin, E.A. and Henson, J.M. (1993) Use of a polymerase chain reaction assay to aid in identification of Gaeumannomyces graminis from different grass hosts. Phytopathology, 83, 414418.
  • Elsherif, M. and Grossmann, F. (1994) Comparative investigations on the antagonistic activity of fluorescent pseudomonads against Gaeumannomyces graminis var. tritici in vitro and in vivo. Microbiol. Res. 149, 371377.
  • Epstein, L., Kaur, S., Goins, T., Kwon, Y.H. and Henson, J.M. (1994) Production of hyphopodia by wild-type and 3 transformants of Gaeumannomyces graminis var. graminis. Mycologia, 86, 7281.
  • Feussner, I., Kuhn, H. and Wasternack, C. (2001) Lipoxygenase-dependent degradation of storage lipids. Trends Plant Sci. 6, 268273.
  • Fouly, H.M. and Wilkinson, H.T. (2000a) A group I intron in the nuclear small subunit ribosomal DNA of Gaeumannomyces graminis. Curr. Microbiol. 40, 291296.
  • Fouly, H.M. and Wilkinson, H.T. (2000b) Detection of Gaeumannomyces graminis varieties using polymerase chain reaction with variety-specific primers. Plant Dis. 84, 947951.
  • Fouly, H.M., Wilkinson, H.T. and Chen, W.D. (1997) Restriction analysis of internal transcribed spacers and the small subunit gene of ribosomal DNA among four Gaeumannomyces species. Mycologia, 89, 590597.
  • Fouly, H.M., Wilkinson, H.T. and Domier, L.L. (1996) Use of random amplified polymorphic DNA (RAPD) for identification of Gaeumannomyces species. Soil Biol. Biochem. 28, 703710.
  • Frederick, B.A., Caesar-Tonthat, T.C., Wheeler, M.H., Sheehan, K.B., Edens, W.A. and Henson, J.M. (1999) Isolation and characterisation of Gaeumannomyces graminis var. graminis melanin mutants. Mycol. Res. 103, 99110.
  • Frey, M., Chomet, P., Glawischnig, E., Stettner, C., Grun, S., Winklmair, A., Eisenreich, W., Bacher, A., Meeley, R.B., Briggs, S.P., Simcox, K. and Gierl, A. (1997) Analysis of a chemical plant defense mechanism in grasses. Science, 277, 696699.
  • Friebe, A., Vilich, V., Hennig, L., Kluge, M. and Sicker, D. (1998) Detoxification of benzoxazolinone allelochemicals from wheat by Gaeumannomyces graminis var. tritici, G. graminis var. graminis, G. graminis var. avenae, and Fusarium culmorum. Appl. Environ. Microbiol. 64, 23862391.
  • Gams, W. (2000) Phialophora and some similar morphologically little-differentiated anamorphs of divergent ascomycetes. Stud. Mycol. 45, 187199 .
  • Goins, T.Q., Edens, W.A. and Henson, J.M. (2002) Heavily-melanized variants of the sexual Gaeumannomyces graminis var. tritici are non-pathogenic and indistinguishable from the asexual, Phialophora state. Mycol. Res. 106, 11791186.
  • Goodwin, P.H., Hsiang, T., Xue, B.G. and Liu, H.W. (1995) Differentiation of Gaeumannomyces graminis from other turf-grass fungi by amplification with primers from ribosomal internal transcribed spacers. Plant Pathol. 44, 384391.
  • Haralampidis, K., Bryan, G., Qi, X., Papadopoulou, K., Bakht, S., Melton, R. and Osbourn, A. (2001) A new class of oxidosqualene cyclases directs synthesis of antimicrobial phytoprotectants in monocots. Proc. Natl Acad. Sci. USA, 98, 1343113436.
  • Harvey, P.R., Langridge, P. and Marshall, D.R. (2001) Genetic drift and host-mediated selection cause genetic differentiation among Gaeumannomyces graminis populations infecting cereals in southern Australia. Mycol. Res. 105, 927935.
  • Henson, J.M. (1989) DNA probe for identification of the take-all fungus, Gaeumannomyces graminis. Appl. Environ. Microbiol. 55, 284288.
  • Henson, J.M. (1992) DNA hybridization and polymerase chain reaction (PCR) tests for identification of Gaeumannomyces, Phialophora and Magnaporthe isolates. Mycol. Res. 96, 629636.
  • Henson, J.M., Blake, N.K. and Pilgeram, A.L. (1988) Transformation of Gaeumannomyces graminis to benomyl resistance. Curr. Genet. 14, 113117.
  • Henson, J.M., Butler, M.J. and Day, A.W. (1999) The dark side of the mycelium: melanins of phytopathogenic fungi. Annu. Rev. Phytopathol. 37, 447471.
  • Henson, J.M. and Ton That, T.C.C. (1995) Mitochondrial plasmids of the Gaeumannomyces–Phialophora complex and their detection by primed in situ fluorescence labeling. Exp. Mycol. 19, 263274.
  • Herdina and Roget, D.K. (2000) Prediction of take-all disease risk in field soils using a rapid and quantitative DNA soil assay. Plant Soil, 227, 8798.
  • Herdina, Harvey, P.H. and OpheI-Keller, K. (1996) Quantification of Gaeumannomyces graminis var. tritici in infected roots and soil using slot-blot hybridization. Mycol. Res. 100, 962970.
  • Herdina, Yang, H.A. and Ophel-Keller, K. (1997) Correlation of take-all disease severity and inoculum level of Gaeumannomyces graminis var. tritici using a slot-blot hybridization assay. Mycol. Res. 101, 13111317.
  • Herman, R.P. (1998) Oxylipin production and action in fungi and related organisms. In Eicosanoids and Related Compounds in Plants and Animals (Rowley, A.F., Kuhn, H. and Schewe, T., eds). London: Portland Press, pp. 115130.
  • Honeyman, A.L. and Currier, T.C. (1986) Isolation and characterization of linear DNA elements from the mitochondria of Gaeumannomyces graminis. Appl. Environ. Microbiol. 52, 924929.
  • Hornby, D. (1983) Suppressive soils. Annu. Rev. Phytopathol. 21, 6585.
  • Hornby, D., Bateman, G.L., Gutteridge, R.J., Lucas, P., Osbourn, A.E., Ward, E. and Yarham, D.J. (1998) Take all Disease of Cereals: a Regional Perspective. Wallingford, UK: CABI International.
  • Hornby, D., Bateman, G.L., Payne, R.W., Brown, M.E., Hendon, D.R. and Campbell, R. (1993) Field-tests of bacteria and soil-applied fungicides as control agents for take-all of winter wheat. Ann. Appl. Biol. 122, 253270.
  • Hörnsten, L., Su, C. and Oliw, E. (2002a) Linoleate diol synthase and PGH synthase—A new gene family of fatty acid heme dioxygenases? Adv. Exp. Med. Biol. 507, 557561.
  • Hörnsten, L., Su, C., Osbourn, A.E., Garosi, P., Hellman, U., Wernstedt, C. and Oliw, E.H. (1999) Cloning of linoleate diol synthase reveals homology with prostaglandin H synthases. J. Biol. Chem. 274, 2821928224.
  • Hörnsten, L., Su, C., Osbourn, A.E., Hellman, U. and Oliw, E.H. (2002b) Cloning of the manganese lipoxygenase gene reveals homology with the lipoxygenase gene family. Eur. J. Biochem. 269, 26902697.
  • Howard, R.J., Ferrari, M.A., Roach, D.H. and Money, N.P. (1991) Penetration of hard substrates by a fungus employing enormous turgor pressure. Proc. Natl Acad. Sci. USA, 88, 1128111284.
  • Howard, R.J. and Valent, B. (1996) Breaking and entering: host penetration by the fungal rice blast pathogen Magnaporthe grisea. Annu. Rev. Microbiol. 50, 491512.
  • Howe, G.A. and Schilmiller, A.L. (2002) Oxylipin metabolism in response to stress. Curr. Opin. Plant Biol. 5, 230236.
  • Huang, L.L., Kang, Z. and Buchenauer, H. (2001a) Comparison of infection of wheat roots by Phialophora graminicola and Gaeumannomyces graminis var. tritici by ultrastructural and cytochemical studies. Z. Pflanzenk. Pflanzens. 108, 593607.
  • Huang, L.L., Kang, Z., Heppner, C. and Buchenauer, H. (2001b) Ultrastructural and immunocytochemical studies on effects of the fungicide Mon65500 (Latitude R.) on colonization of roots of wheat seedlings by Gaeumannomyces graminis var. tritici. Z. Pflanzenk. Pflanzens. 108, 188203.
  • Huber, S., Lottspeich, F. and Kamper, J. (2002) A gene that encodes a product with similarity to dioxygenases is highly expressed in teliospores of Ustilago maydis. Mol. Genet. Genomics, 267, 757771.
  • Jamil, N. and Buck, K.W. (1991) Effect of vegetative incompatibility on double-stranded-RNA and mycovirus transmission in Gaeumannomyces graminis var. tritici. Pak. J. Bot. 23, 160164.
  • Jamil, N., Buck, K.W. and Carlile, M.J. (1984) Sequence relationship between virus double stranded RNA from isolates of Gaeumannomyces graminis in different vegetative incompatibility groups. J. Gen. Virol. 65, 17411747.
  • Jian, J., Lakshman, D.K. and Tavantzis, S.M. (1997) Association of distinct double-stranded RNAs with enhanced or diminished virulence in Rhizoctonia solani infecting potato. Mol. Plant–Microbe Interact. 10, 10021009.
  • Joseph-Horne, T., Heppner, C., Headrick, J. and Hollomon, D.W. (2000) Identification and characterization of the mode of action of MON 65500: a novel inhibitor of ATP export from mitochondria of the wheat ‘take-all’ fungus, Gaeumannomyces graminis var. tritici. Pest. Biochem. Physiol. 67, 168186.
  • Juhnke, M.E., Mathre, D.E. and Sands, D.C. (1984) A selective medium for Gaeumannomyces graminis var. tritici. Plant Dis. 68, 233236.
  • Kang, Z., Huang, L. and Buchenauer, H. (2000) Cytochemistry of cell wall component alterations in wheat roots infected by Gaeumannomyces graminis var. tritici. Z. Pflanzenk. Pflanzens. 107, 337351.
  • Kelly, C. (1997) Genetics of pigmentation and pathogenicity in G. graminis. PhD thesis, University of Birmingham.
  • Kelly, C., Osbourn, A.E. and Caten, C.E. (1997) The genetics of Gaeumannomyces graminis with particular reference to pigment production and pathogenicity. Fungal Genet. Newsletter, 44A, 108.
  • Kim, D.S., Cook, R.J. and Weller, D.M. (1997) Bacillus sp. L324–92 for biological control of three root diseases of wheat grown with reduced tillage. Phytopathology, 87, 551558.
  • Kohlmeyer, J., Volkmannkohlmeyer, B. and Eriksson, O.E. (1995) Fungi on Juncus roemerianus. 4. New marine ascomycetes. Mycologia, 87, 532542.
  • Landschoot, P.J. and Jackson, N. (1989) Gaeumannomyces incrustans sp. nov., a root-infecting hyphopodiate fungus from grass roots in the United-States. Mycol. Res. 93, 5558.
  • Langfelder, K., Streibel, M., Jahn, B., Haase, G. and Brakhage, A.A. (2003) Biosynthesis of fungal melanins and their importance for human pathogenic fungi. Fungal Genet. Biol. 38, 143158.
  • Liang, P.-Y. and Chen, K.-Y. (1990) Homology between Gaeumannomyces graminis virus and other mycoviruses by dot blot hybridization. Chin. J. Virol. 6, 245249.
  • Liang, P.-Y., Chou, S.-M. and Chen, K.-Y. (1989) Properties of double stranded RNA viruses from different strains of Gaeumannomyces graminis in different vegetative compatible groups. Virologica Sinica, 4, 6875.
  • Linde-Laursen, I., Jensen, H.P. and Jørgensen, J.H. (1973) Resistance of Triticale, Aegilops and Haynaldia species to the take-all fungus, Gaeumannomyces graminis. Z. Pflanzenzüchtung, 70, 200213.
  • Litvintseva, A.P. and Henson, J.M. (2002a) Cloning, characterization, and transcription of three laccase genes from Gaeumannomyces graminis var. tritici, the take-all fungus. Appl. Environ. Microbiol. 68, 13051311.
  • Litvintseva, A.P. and Henson, J.M. (2002b) Characterization of two laccase genes of Gaeumannomyces graminis var. graminis and their differential transcription in melanin mutants and wild type. Mycol. Res. 106, 808814.
  • Löchel, A.M., Wenz, M., Russell, P.E., Buschhaus, H., Evans, P.H., Cross, S., Puhl, T. and Bardsley, E. (1998) Root protection using fluquinconazole: a new approach to controlling cereal take-all. Proceedings of the 1998 British Crop Protection Conference—Pests and Diseases, pp. 8996.
  • Mathre, D.E., Johnston, R.H. and Grey, W.E. (1998) Biological control of take-all disease of wheat caused by Gaeumannomyces graminis var. tritici under field conditions using a Phialophora sp. Biocontrol. Sci. Technol. 8, 449457.
  • McCarty, L.B. and Lucas, L.T. (1989) Gaeumannomyces graminis associated with spring dead spot of bermudagrass in the Southeastern United States. Plant Dis. 73, 659661.
  • McGinty, R.M., Buck, K.W. and Rawlinson, C.J. (1981) Virus particles and double-stranded RNA in isolates of Phialophora with lobed hyphopodia, Phialophora graminicola and Gaeumannomyces graminis var. graminis. Phytopathol. Z. 102, 153162.
  • Money, N.P. (1995) Turgor pressure and the mechanics of fungal penetration. Can. J. Bot. 73, S96S102.
  • Money, N.P., Caesar-TonThat, T.C., Frederick, B. and Henson, J.M. (1998) Melanin synthesis is associated with changes in hyphopodial turgor, permeability, and wall rigidity in Gaeumannomyces graminis var. graminis. Fungal Genet. Biol. 24, 240251.
  • Morrissey, J.P. and Osbourn, A.E. (1999) Fungal resistance to plant antibiotics as a mechanism of pathogenesis. Microbiol. Mol. Biol. Rev. 63, 708724.
  • Musker, R. (1994) Genetics and population studies of Gaeumannomyces graminis. PhD thesis, University of Birmingham.
  • Niemeyer, H.M. (1988) Hydroxamic acids (4-hydroxy-1,4-benzoxazin-3-ones), defense chemicals in the Gramineae. Phytochemistry, 27, 33493358.
  • O'Dell, M., Flavell, R.B. and Hollins, T.W. (1992) The classification of isolates of Gaeumannomyces graminis from wheat, rye and oats using restriction-fragment-length-polymorphisms in families of repeated DNA sequences. Plant Pathol. 41, 554562.
  • Osbourn, A.E. (1996a) Saponins and plant defence—A soap story. Trends Plant Sci. 1, 49.
  • Osbourn, A.E. (1996b) Preformed antimicrobial compounds and plant defense against fungal attack. Plant Cell, 8, 18211831.
  • Osbourn, A.E. (2001) Plant mechanisms that give defence against soilborne diseases. Aust. Plant Pathol. 30, 99102.
  • Osbourn, A.E., Bowyer, P., Bryan, G., Lunness, P., Clarke, B. and Daniels, M. (1994a) Detoxification of plant saponins by fungi. In Advances in Molecular Genetics of Plant–Microbe Interactions (Daniels, M.J., Downie, J.A. and Osbourn, A.E., eds). Dordrecht: Kluwer, pp. 215221.
  • Osbourn, A.E., Clarke, B.R., Lunness, P., Scott, P.R. and Daniels, M.J. (1994b) An oat species lacking avenacin is susceptible to infection by Gaeumannomyces graminis var. tritici. Physiol. Mol Plant Pathol, 45, 457467.
  • Ou, S.H. (1972) Rice Diseases. Kew, UK: Commonwealth Mycological Institute.
  • Papadopoulou, K., Melton, R.E., Leggett, M., Daniels, M.J. and Osbourn, A.E. (1999) Compromised disease resistance in saponin-deficient plants. Proc. Natl Acad. Sci. USA, 96, 1292312928.
  • Penrose, L. (1985) Evidence for resistance in wheat cultivars grown in sand culture to the take-all pathogen, Gaeumannomyces graminis var. tritici. Ann. Appl. Biol. 107, 105108.
  • Penrose, L.D.J. (1991) Disease in wheat genotypes naturally infected with Gaeumannomyces graminis var. tritici. Ann. Appl. Biol. 118, 513526.
  • Penrose, L.D.J. (1995) Two wheat genotypes differ in root disease due to Gaeumannomyces graminis without interaction with site. Soil Biol. Biochem. 27, 133138.
  • Penrose, L.D.J. and Neate, S.M. (1994) Resistance to Gaeumannomyces graminis in wheat genotypes grown in field environments and sand culture. Soil Biol. Biochem. 26, 719726.
  • Pilgeram, A.L., Goins, T. and Henson, J.M. (1993) The fate of integrated DNA in Gaeumannomyces graminis transformants. FEMS Microbiol. Lett. 113, 309314.
  • Pilgeram, A.L. and Henson, J.M. (1990) Transformation and cotransformation of Gaeumannomyces graminis to phleomycin resistance. Phytopathology, 80, 11241129.
  • Pilgeram, A.L. and Henson, J.M. (1992) Sexual crosses of the homothallic fungus Gaeuannomyces graminis var. tritici based on use of an auxotroph obtained by transformation. Exp. Mycol. 16, 3543.
  • Rachdawong, S., Cramer, C.L., Grabau, E.A., Stromberg, V.K., Lacy, G.H. and Stromberg, E.L. (2002) Gaeumannomyces graminis vars. avenae, graminis, and tritici identified using PCR amplification of avenacinase-like genes. Plant Dis. 86, 652660.
  • Rawlinson, C.J. and Buck, K.W. (1981) Viruses in Gaeumannomyces graminis and Phialophora. In Biology and Control of Take-All (Asher, M.J.C. and Shipton, P.J., eds). London: Academic Press, pp. 151172.
  • Rengel, Z. (1997) Decreased capacity of take-all fungus to oxidize manganous ions is associated with take-all decline. J. Plant Nutr. 20, 455460.
  • Rengel, Z., Graham, R.D. and Pedler, J.F. (1993) Manganese nutrition and accumulation of phenolics and lignin as related to differential resistance of wheat genotypes to the take-all fungus. Plant Soil, 151, 255263.
  • Rengel, Z., Graham, R.D. and Pedler, J.F. (1994) Time-course of biosynthesis of phenolics and lignin in roots of wheat genotypes differing in manganese efficiency and resistance to take-all fungus. Ann. Bot. 74, 471477.
  • Rosewich, U.L. and Kistler, H.C. (2000) Role of horizontal gene transfer in the evolution of fungi. Annu. Rev. Phytopathol. 38, 325363.
  • Ross, I.L., Alami, Y., Harvey, P.R., Achouak, W. and Ryder, M.H. (2000) Genetic diversity and biological control activity of novel species of closely related pseudomonads isolated from wheat field soils in South Australia. Appl. Environ. Microbiol. 66, 16091616.
  • Rovira, A., Ryder, M. and Harris, A. (1992) Biological control of root diseases with psuedomonads. In Biological Control. of Plant Diseases-Progress and Challenges for the Future (Tjamos, E.S., Papavizas, G.C. and Cook, R.J., eds). New York: Plenum Press, pp. 175184.
  • Rovira, A.D. and Wildermuth, G.B. (1981) The nature and mechanisms of suppression. In Biology and Control of Take-All (Asher, M.J.C. and Shipton, P.J., eds). London: Academic Press, pp. 385415.
  • Ryder, M.H., Brisbane, P.G. and Rovira, A.D. (1990) Mechanisms in the biological control of take-all of wheat by rhizosphere bacteria. In Biological Control of Soil-Borne Plant Pathogens (Hornby, D.H., ed.). Wallingford: CAB International, pp. 123130 .
  • Ryder, M.H. and Rovira, A.D. (1993) Biological control of take-all of glasshouse-grown wheat using strains of Pseudomonas corrugata isolated from wheat field soil. Soil Biol. Biochem. 25, 311320.
  • Salvo, J.L., Rodeghier, B., Rubin, A. and Troischt, T. (1998) Optional introns in mitochondrial DNA of Podospora anserina are the primary source of observed size polymorphisms. Fungal Genet. Biol. 23, 162168.
  • Schesser, K., Luder, A. and Henson, J.M. (1991) Use of polymerase chain reaction to detect the take-all fungus, Gaeumannomyces graminis, in infected wheat plants. Appl. Environ. Microbiol. 57, 553556.
  • Schoeny, A. and Lucas, P. (1999) Modeling of take-all epidemics to evaluate the efficacy of a new seed-treatment fungicide on wheat. Phytopathology, 89, 954961.
  • Sesma, A. and Osbourn, A.E. (2003) Root colonization by Magnaporthe grisea: novel developmental processes associated to a foliar pathogen. In Abstracts of the 11th International Congress on Plant–Microbe Interactions, July 18th−26th 2003, St. Petersburg, Russia, p. 221.
  • Sivasithamparam, K. (1975) Phialophora and Phialophora-like fungi occuring in the root region of wheat. Aust. J. Bot. 23, 193212 .
  • Sivasithamparam, K. (1998) Root cortex—The final frontier for the biocontrol of root-rot with fungal antagonists: a case study on a sterile red fungus. Annu. Rev. Phytopathol. 36, 439452.
  • Sivasithamparam, K. and Parker, C.A. (1981) Physiology and nutrition in culture. In Biology and Control of Take-All (Asher, M.J.C. and Shipton, P.J., eds). London: Academic Press, pp. 125150.
  • Skou, J.P. (1981) Morphology and cytology of the infection process. In Biology and Control of Take-All (Asher, M.J.C. and Shipton, P.J., eds). London: Academic Press, pp. 175197.
  • De Souza, J.T., Weller, D.M. and Raaijmakers, J.M. (2003) Frequency, diversity, and activity of 2,4-diacetylphloroglucinol-producing fluorescent Pseudomonas spp. in Dutch take-all decline soils. Phytopathology, 93, 5463.
  • Spink, J.H., Wade, A.P., Paveley, N.D., Griffin, J.M., Scott, R.K. and Foulkes, M.J. (1998) The effects of a novel seed treatment, MON65500, on take-all severity and crop growth in winter wheat. In Proceedings of the 1998 British Crop Protection Conference—Pests and Diseases, pp. 913920.
  • Su, C. and Oliw, E.H. (1996) Purification and characterization of linoleate 8-dioxygenase from the fungus Gaeumannomyces graminis as a novel hemoprotein. J. Biol. Chem. 271, 1411214118.
  • Su, C. and Oliw, E.H. (1998) Manganese lipoxygenase—Purification and characterization. J. Biol. Chem. 273, 1307213079.
  • Tan, M.K. (1997) Origin and inheritance of group I introns in 26S rRNA genes of Gaeumannomyces graminis. J. Mol. Evol. 44, 637645.
  • Tan, M.K. and Wong, P.T.W. (1996) Group I introns in 26S rRNA genes of Gaeumannomyces graminis as possible indicators of host specificity of G. graminis varieties. Mycol. Res. 100, 337342.
  • Tan, M.K., Wong, P.T.W. and Holley, M.P. (1994) Characterization of nuclear ribosomal DNA (rDNA) in Gaeumannomyces graminis and correlation of rDNA variation with G. graminis varieties. Mycol. Res. 98, 553561.
  • Thornton, C.R., Dewey, F.M. and Gilligan, C.A. (1997) Production and characterization of a monoclonal antibody raised against surface antigens from mycelium of Gaeumannomyces graminis var. tritici: evidence for an extracellular polyphenol oxidase. Phytopathology, 87, 123131.
  • Ulrich, K., Augustin, C. and Werner, A. (2000) Identification and characterization of a new group of root-colonizing fungi within the Gaeumannomyces–Phialophora complex. New Phytol. 145, 127135.
  • Virtanen, A.I. and Hietala, P.K. (1955) 2 (3)-Benzoxazolinone, an anti-Fusarium factor in rye seedlings. Acta Chemica Scand. 9, 15431544.
  • Walker, J.L. (1975) Take-all diseases of Gramineae: a review of recent work. Rev. Plant Pathol. 54, 113144.
  • Walker, J. (1981) Taxonomy of take-all fungi and related genera and species. In Biology and Control of Take-All (Asher, M.J.C. and Shipton, P.J., eds). London: Academic Press, pp. 1574.
  • Ward, E. (1995) Improved polymerase chain reaction (PCR) detection of Gaeumannomyces graminis including a safeguard against false negatives. Eur. J. Plant Pathol. 101, 561566.
  • Ward, E. and Akrofi, A.Y. (1994) Identification of fungi in the Gaeumannomyces–Phialophora complex by RFLPs of PCR-amplified ribosomal DNAs. Mycol. Res. 98, 219224.
  • Ward, E. and Bateman, G.L. (1994) Identification of fungi in the Gaeumannomyces–Phialophora complex associated with take-all of cereals and grasses using DNA probes. In Modern Detection Assays for Plant Pathogenic Fungi: Identification, Detection and Quantification (Schots, A., Dewey, F.M. and Oliver, R., eds). Wallingford, UK: CAB International, pp. 127134.
  • Ward, E. and Bateman, G.L. (1999) Comparison of Gaeumannomyces- and Phialophora-like fungal pathogens from maize and other plants using DNA methods. New Phytol. 141, 323331.
  • Ward, E. and Gray, R. (1992) Generation of a ribosomal probe by PCR and its use in identification of fungi within the Gaeumannomyces–Phialophora complex. Plant Pathol. 41, 730736.
  • Weller, D.M., Raaijmakers, J.M., Gardener, B.B.M. and Thomashow, L.S. (2002) Microbial populations responsible for specific soil suppressiveness to plant pathogens. Annu. Rev. Phytopathol. 40, 309348.
  • Wetzel, H.C., Dernoeden, P.H. and Millner, P.D. (1996) Identification of darkly pigmented fungi associated with turfgrass roots by mycelial characteristics and RAPD-PCR. Plant Dis. 80, 359364.
  • White, T.J., Bruns, T., Lee, S. and Taylor, J. (1990) Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In PCR Protocols. A Guide to Methods and Applications (Innis, M.A., Gelfand, D.H., Sninsky, J.J. and White, T.J., eds). San Diego, CA: Academic Press, pp. 315322.
  • Wilkes, M.A., Marshall, D.R. and Copeland, L. (1999) Hydroxamic acids in cereal roots inhibit the growth of take-all. Soil Biol. Biochem. 31, 18311836.
  • Wong, P.T.W. (2002) Gaeumannomyces wongoonoo sp. nov., the cause of a patch disease of buffalo grass (St Augustine grass). Mycol. Res. 106, 857862.
  • Wong, P.T.W., Mead, J.A. and Holley, M.P. (1996) Enhanced field control of wheat take-all using cold tolerant isolates of Gaeumannomyces graminis var. graminis and Phialophora sp. (lobed hyphopodia). Plant Pathol. 45, 285293.
  • Wong, P.T.W., Tan, M.K. and Beehag, G.W. (2000) Confirmation of take-all patch disease in Tifdwarf hybrid couch grass (bermudagrass) by morphological and DNA methods. Austr. Plant Pathol. 29, 1923.
  • Xu, J.-R. (2000) MAP kinases in fungal pathogens. Fungal Genet. Biol. 31, 137152.
  • Yao, J.M., Wang, Y.C. and Zhu, Y.G. (1992) A new variety of the pathogen of maize take-all. Acta Mycol. Sinica, 11, 99104.
  • Yarham, D.J. (1981) Practical aspects of epidemiology and control. In Biology and Control of Take-All (Asher, M.J.C. and Shipton, P.J., eds). London: Academic Press, pp. 353384.
  • Yeates, J.S., Fang, C.S. and Parker, C.A. (1986) Distribution and importance of oat-attacking isolates of Gaeumannomyces graminis var. tritici in Western Australia. Trans. Br. Mycol. Soc. 86, 145152.
  • Zhou, S.G. and Stanosz, G.R. (2001) Primers for amplification of mt SSU rDNA, and a phylogenetic study of Botryosphaeria and associated anamorphic fungi. Mycol. Res. 105, 10331044.
  • Zriba, N., Sherwood, J.E. and Mathre, D.E. (1999) Characterization and effectiveness of Phialophora spp. isolated from a Montana take-all suppressive soil in controlling take-all disease of wheat. Can J. Plant Pathol. 21, 110118.