Biological control of the take-all fungus, Gaeumannomyces graminis, by Phialophora radicicola and similar fungi

Phialophora radicicola is an avirulent fungal root-parasite of grasses and cereals, with runner hyphae like those of Gaeumannomyces graminis. Weakly and non-pathogenic varieties of these fungi control the pathogens, G. graminis vars. tritici and avenue. Biology of these fungi is considered and the evidence for biological control and possible mechanisms reviewed; control is probably widespread in natural plant communities, and host-mediated, perhaps by induction of plant resistance mechanisms.Prospects for application of biological control seem best for P. radicicola var. graminicola established on grass crops, as this is already exploited in British agriculture. New evidence is presented on the effects of grassland factors on this fungus, especially sward composition, age, mineral nutrition and management practices: its population might often be limited by the rate of new root production to replace those with cortices already colonized. Prospects for control by seed inoculation with P. radicicola var. radicicola and G. graminis var. graminis also seem good, but possible dangers of introducing them into cereal cropping are emphasized. The weak pathogens might be used also for indirect control by establishing hyper-parasites or inducing disease suppression (like take-all decline) in soils, but there is no evidence for ‘Phialophora decline’, at least in well-managed grasslands. Finally, different biocontrols of take-all might be combined, and biological with chemical ones for ophiobolus patch disease of turf.P. radicicola var. graminicola has a slight beneficial effect on grass yield, even when the pathogens, G. graminis vars. tritici and arenae arc absent; this probably contributes to its abundance in natural grasslands in Britain. The scale of biological control by this and similar fungi might explain why, in their absence, effective plant resistance to G. graminis is uncommon in the Gramineae.     

Age of wheat and barley roots and infection by gaeumannomyces graminis var tritici

Seminal roots of wheat and barley seedlings were inoculated with G. graminis var tritici on regions 0, 5- and 15-days old, and assessed for intensity and extent of infection after standard times. Wheat roots were most heavily infected on young regions, whereas barley roots were most heavily infected on old ones. The effect of root age in wheat was similar in both unsterile and aseptic conditions, so it could not be ascribed to saprophytic rhizosphere micro-organisms interacting with G. graminis.The contrasting results for wheat and barley are explained by a single hypothesis, based on decreasing host-resistance in the root cortex but increasing resistance at or near the endodermis as the roots age. It is suggested that, under some conditions, even small amounts of non-pathogenic root cortex death can enhance infection by G. graiminis. This interpretation may explain several aspects of take-all and its biological control by other dark mycelial parasites.     

Infection of wheat seminal roots by varieties of Phialophora radicicola and Gaeumannomyces graminis

The growth of isolates of Phialophora radicicola var. radicicola, P. radicicola var. graminicola, Gaeumannomyces graminis var. graminis, G. graminis var. tritici and Leptosphaeria narmari was compared on the coleoptiles and roots of wheat seedlings. Fungal growth was measured as the extent and density of dark runner hyphae. All except P. radicicola var. graminicola grew on coleoptiles and all grew on roots although only G. graminis var. tritici extensively colonized the root stele. Growth rate on roots was positively correlated with that on agar, P. radicicola var. graminicola and L. narmari growing at about half the rate of the other fungi; hyphal density was high for P. radicicola var. graminicola but relatively low for the other fungi. For P. radicicola var. radicicola, P. radicicola var. graminicola and G. graminis var. tritici growing from buried inocula, the extent and density of hyphae up roots towards the seed was similar to that down, but G. graminis var. tritici caused chocolate-brown stelar discoloration up roots only.Root invasion by P. radicicola var. radicicola, P. radicicola var. graminicola and G. graminis var. tritici was described from sections. Each gave a different pattern of hyphae and host response within an inoculum layer, and progressive changes occurred away from the inoculum. Studies of the rate of penetration by each fungus and the rate and pattern of death of cortical cells explained the differences between fungi. G. graminis var. tritici penetrated living cells in advance of other soil micro-organisms, and hence by hyaline hyphae inducing much lignituber formation as a host resistance reaction. P. radicicola var. graminicola penetrated only senescent or dead cells in association with other soil microorganisms, and hence by dark hyphae, inducing little lignituber formation. P. radicicola var. radicicola was intermediate in all these respects. The high hyphal density of P. radicicola var. graminicola was due to the colonization of cortical cells and spaces by dark, clearly visible, rather than hyaline hyphae, which are invisible in unstained roots. Cell death in the outer cortex explained the observed progressive restriction of growth by all fungi to the inner cortex with increasing distance from the inoculum. Spread by G. graminis var. tritici up roots was ectotrophic relative to the stele but down roots hyphae spread rapidly within the stele. Stelar reactions suggested as resistance mechanisms occurred up roots only. Their absence down roots is attributed to infection disrupting stelar transport.     

Hyphal density as a measure of suppression of Gaeumannomyces graminis var. tritici on wheat roots

Runner hyphae of Gaeumannomyces graminis (Sacc.) Arx & Olivier var. tritici Walker on seminal roots of wheat seedlings were photographed and their length measured. As well, their length was estimated using the line-intercept method. The correlation of 0.904 between measured and estimated lengths of hyphae was highly significant. This line intercept method was used to estimate the density (length/unit area) of hyphae on roots of plants growing in the presence and absence of a soil suppressive to G. graminis var. tritici. Estimations were made eight times during 28 days growth at 15°C. In fumigated soil (non-suppressive) inoculated with 0.1% ground oat grain infested with G. graminis var. tritici, the density of hyphae on roots started to increase at five days compared with 15 days when soil there was a 10.8% cover of the root surface after 15 days when the hyphae had reached maximum density. Suppression to G. graminis var. tritici is normally detected by a difference in disease rating of roots at 28 days but this study has shown that suppression can be demonstrated by the difference in the density of hyphae if roots are examined between seven and 19 days.     

Cross-protection against the wheat and oat take-all fungi by Gaeumannomyces graminis var. graminis

Glasshouse experiments have shown that the prior colonisation of wheat roots by Gaeumannomyces graminis var. graminis, a fungus closely related to the wheat and oat take-all fungi but non-pathogenic to temperate cereals, reduced take-all infection along the roots. Cross-protected wheat plants produced grain yields significantly greater than those of unprotected plants but not significantly different to those of healthy wheat plants. A Phialophora-like fungus from grass roots did not confer the same degree of protection. There is some evidence that the cross-protection mechanism may be a specific host response nduced by var. graminis. The possible use of var. graminis in the biological control of take-all is discussed.     

The role of bacteria in the biological control of Gaeumannomyces graminis by suppressive soils

The suppression of Gaeumannomyces graminis var. tritici by certain soils or following certain soil treatments is considered to be an expression of either specific or general antagonism sensu Gerlagh (1968). Specific antagonism is effective in dilutions as high as 1 in 1,000, can be transferred from soil to soil, operates near or on wheat roots, is destroyed by 60°C moist heat for 30 min. or desiccation, is fostered by wheat monoculture but may be lost from a soil by fallow or rotation with certain crops, especially legume hay or pasture crops. Strains of Pseudomonas fluorescens may be involved. General antagonism is a soil property which cannot be transferred and is resistant to 80°C moist heat for 30 min, to methyl bromide and chloropicrin, but not to autoclaving. Take-all control by organic amendments, minimum tillage, or a soil temperature of 28°C may be expressions of increased general antagonism.In much of the southern Australian wheat belt, where take-all can cause heavy crop losses, some general but rarely specific antagonism is apparently operative. Both types of antagonism are probably operative in long-term wheat growing areas of the Pacific Northwest U.S.A. where take-all is virtually nonexistent.     

Wheat-rhizoplane pseudomonads as antagonists of Gaeumannomyces graminis

The role of rhizoplane-inhabiting Pseudomonas spp as inhibitors of take-all on wheat was investigated. Apparent numbers of pseudomonads in wheat rhizoplanes and numbers that were antagonistic in vitro toward Gaeumannomyces graminis var, tritici did not differ when wheat was supplied with NH⁺₄-N or NO^?₃-N. More intense antagonism was expressed by colonies selected from soil treated with NH⁺₄-N than with NO^?₃-N, and from isolation media prepared at pH 5.5 rather than at 7.0. Antagonists were not recovered from methyl bromide-treated soil. Highly antagonistic pseudomonads were recovered from a wheat-monoculture soil which is considered suppressive toward the pathogen in the field, and were not recovered from a “nonsuppressive” soil. Pseudomonad antagonism ratings were inversely correlated with take-all severity in the suppressive soil, but not in the nonsuppressive soil. Pseudomonads were considered to be antagonists of G. graminis on rhizoplanes of wheat in a soil exhibiting the “take-all decline” phenomenon, but the significance of this interaction remains to be determined.     

Soils suppressive to Gaeumannomyces graminis var. tritici: Induction by other fungi

Rhizosphere soils were obtained from wheat plants growing in fumigated soil inoculated with one of the pathogens, Rhizoctonia solani, Gibberella zeae, Fusarium culmorum, Cochliobolus sativus and Pythium irregulare or one of the non-pathogenic fungi, Gaeumannomyces graminis var. graminis (Ggg) or a Phialophora-like fungus (Plf). Using a pot bioassay, these soils were tested for suppression of Gaeumanomyces graminis var. tritici (Ggt) and the fungus involved in the initial induction. G. zeae was the only fungus that induced suppression to Ggt and to itself. Ggg but not Plf induced suppression to Ggt although both fungi induced suppression to themselves.Fungi capable of inducing suppression of take-all have two characteristics in common, they induce suppression of themselves and their saprophytic survival is restricted to organic matter.     

Colonisation of seminal roots of wheat and barley by egg-parasitic nematophagous fungi and their effects on Gaeumannomyces graminis var. tritici and development of root-rot

This study provides evidence that egg-parasitic nematophagous fungi, Pochonia chlamydosporia, Pochonia rubescens and Lecanicillium lecanii, can also reduce root colonisation and root damage by a fungal pathogen. Interactions of nematophagous fungi with the take-all fungus, Gaeumannomyces graminis var. tritici (Ggt), and their influence on severity of the root disease it causes were studied in laboratory and pot experiments. In Petri dish experiments the three nematophagous fungi reduced colonisation of barley roots by Ggt and also reduced necrotic symptoms. On the contrary, root colonisation by nematophagous fungi was unaffected by Ggt. In growth tube experiments, the three nematophagous fungi again reduced Ggt root colonisation and increased effective root length of barley seedlings. This was true for both simultaneous and sequential inoculation of nematophagous fungi versus Ggt. In the pot experiments the inoculum of the tested fungi in soil was applied in the same pot, as a mixture or in layers, or in coupled pots used for wheat grown with a split-root system. The nematophagous fungi P. chlamydosporia (isolate 4624) and L. lecanii (isolate 4629), mixed with Ggt or in split root systems with the pathogen, promoted growth of wheat (i.e. increased shoot weight), although no disease reduction was found. In split root systems, lower levels of peroxidase activity were found in seedlings inoculated with Ggt in combination with the nematophagous isolates 4624 and 4629 than when the take-all fungus was applied alone.Our results show that nematophagous fungi reduce root colonisation by Ggt, root damage and stress induced senescence in Ggt-inoculated plants.     

Factors impacting the activity of 2,4-diacetylphloroglucinol-producing Pseudomonas fluorescens against take-all of wheat

Take-all, caused by Gaeumannomyces graminis var. tritici, is an important soilborne disease of wheat worldwide. Pseudomonas fluorescens producing the antibiotic 2,4-diacetylphloroglucinol (2,4-DAPG) are biocontrol agents of take-all and provide natural suppression of the disease during wheat monoculture known as take-all decline. To identify factors that could contribute to the effectiveness of 2,4-DAPG producers in take-all suppression, P. fluorescens strains Q8r1-96 (genotype D) and Q2-87V1 (genotype B; reduced antibiotic production) were tested against three pathogen isolates differing in sensitivity to 2,4-DAPG (LD5, ARS-A1 and R3-111a-1) and two wheat cultivars (Tara and Buchanan). The antibiotic sensitivity of the pathogen and cultivar significantly affected the level of take-all suppression by Q8r1-96 and Q2-87V1; suppression was greatest with LD5 and Tara. Q8r1-96 suppressed ARS-A1 and R3-111a-1 on Tara but not Buchanan, and Q2-87V1 failed to suppress either pathogen isolate on either cultivar. Q8r1-96 colonized the rhizosphere of Tara and Buchanan grown in soil similarly, but 2,4-DAPG accumulation was higher on the roots of Buchanan than Tara. 2,4-DAPG at 7.5 μg mL⁻¹ reduced the growth of roots of both cultivars, and 10 μg mL⁻¹ caused brown necrosis and tissue collapse of seedling roots and reduced root hair development. The half-life of 2,4-DAPG in the rhizosphere was estimated to be 0.25 days. These results suggest that several interconnected factors including sensitivity of G. graminis var. tritici to 2,4-DAPG, wheat cultivar, fluctuations in populations of 2,4-DAPG producers, and antibiotics accumulation in the rhizosphere will impact the robustness of take-all suppression by P. fluorescens in the field.     

Manganese-oxidizing ability of isolates of the take-all fungus is correlated with virulence

Two successive experiments described here show that the Mn-oxidizing ability of isolates of the take-all fungus Gaeumannomyces graminis var. tritici is positively correlated with the isolate virulence. Isolates of G. graminis var. tritici were screened for virulence on wheat in soil culture pot tests and visually scored for Mn-oxidizing capacity on agar plates. A comparison between experiments showed that the Mn-oxidizing ability of an isolate may be attenuated, as is virulence, with subculturing. Using a solution culture system, we calculated the rate of oxidation of Mn²⁺ by a virulent isolate, Ggt 500, as 4 nmol g^-1 (dry weight hyphae) h^-1. The significance of these results to the epidemiology of the take-all disease of wheat is discussed.     

Suppression of wheat take-all and ophiobolus patch by fluorescent pseudomonads from a fusarium-suppressive soil

Fluorescent pseudomonads isolated from a soil suppressing Fusarium wilt significantly reduced take-all (Gaeumannomyces graminis var. tritici) in wheat and Ophiobolus patch (G. graminis var. avenae) in Agrostis turfgrass. The bacteria were mixed into a conducive soil at a concentration of 10⁷ colony-forming units (cfu)g^?1 soil at sowing. There were significantly fewer (P ? 0.05) diseased wheat roots in the treatments with the bacteria and pathogen than in those with the pathogen alone. Dry weights of the tops of wheat and Agrostis turfgrass were significantly greater (P ? 0.01) in treatments inoculated with the bacteria in the presence of the pathogens compared to controls with the pathogens alone. Dry weights of the tops of plants from treatments inoculated with the bacteria alone were not significantly different to those of healthy wheat non-inoculated with the bacteria, showing that the fluorescent pseudomonads did not stimulate plant growth. At the end of the experiments, the bacterial isolates (genetically-marked with rifampicin resistance) were recovered from wheat roots and rhizosphere soil at concentrations of 10⁵–10⁷cfu g^?1 fresh weight of roots or oven-dried rhizosphere soil.Many of the fluorescent pseudomonads and some non-fluorescent pseudomonads showed in vitro antibiosis on quarter-strength potato dextrose agar (QPDA) against the pathogens. However, there was no correlation between in vitro antibiosis on agar plates and suppression of disease in pot experiments. Further, while some isolates of G. graminis var. tritici and var. avenae were inhibited by certain bacterial isolates, other isolates of the same fungus were not similarly inhibited by the same isolates of bacteria. Most of the fluorescent pseudomonads that produced inhibition zones (>5mm) against G. graminis var. tritici on QPDA did not do so on King's medium B, where fluorescent siderophores were formed. In vitro antibiosis is, therefore, a poor criterion for selecting effective bacterial antagonists of the wheat take-all fungus. All of the fluorescent pseudomonads tested produced siderophores in low-Fe media while a non-fluorescent pseudomonad and the fungal pathogens did not produce siderophores of comparable activity. The addition of 500 μg FeEDTA g^?1 with a lower stability constant did not. The evidence suggests that iron competition at the rhizoplane or in the rhizosphere is one mechanism of suppression.     

Ammonium and nitrate in the rhizosphere of spring barley, hordeum vulgare L., and take-all disease

The incidence and severity of take-all disease, due to Gaeumannomyces graminis (Sacc.) Arx & Olivier var. tritici Walker, was observed on spring barley plants growing in soil in two glasshouse experiments. Soil amendments of NH⁺₄-N significantly increased the number of diseased plants and roots during the first month after germination in comparison with controls unamended with N (P < 0.05). No significant difference in the incidence of take-all disease was detected between more mature barley plants growing in soil amended with either NH⁺₄ or NO^?₃-N and unamended controls. The least take-all disease in 3 month-old barley plants was observed when N was supplied as foliar sprays of urea at 0.5 mg N kg^?1 soil (P < 0.01). There was no significant correlation between the degree of infection and the NH⁺₄-N to NO^?₃-N ratio in the rhizosphere soil     

Mycorrhizal fungi increase biocontrol potential of Pseudomonas fluorescens

Wheat roots are susceptible to colonisation by soil-borne pathogens, such as Gaeumannomyces graminis var. tritici (Ggt), which causes the globally important disease take-all, and mutualistic arbuscular mycorrhizal fungi (AMF). Certain rhizosphere fluorescent Pseudomonas strains have received much attention as potential biocontrol agents given their ability to produce antibiotics, such as 2,4-diacetylphloroglucinol (DAPG), that confer a measure of plant protection. Here we show that Pseudomonas fluorescens only produced DAPG in the presence of soluble carbon from soil containing either Ggt or AMF, and production increased by two orders of magnitude in response to both AMF and Ggt. Encouragement of mycorrhizal colonisation may therefore offer a sustainable strategy for protection against take-all.     

Effects of wheatfield soil on inocula of gaeumannomyces graminis (Sacc.) Arx & Olivier var. tritici J. Walker in relation to take-all decline

Organic debris extracted from wheatfield soils was found to be significantly more infective following 2–3 yr continuous wheat than after the fifth successive crop. Plots with added nitrogen yielded more infective debris than those without. Short periods of contact with decline soil or its suspension, reduced infection of wheat roots by Gaeumannomyces when they were subsequently grown in non-decline soil. Hyphae emerging from previously infected root pieces or culture inocula showed a positive growth response to wheat roots or their exudates. This response was negatively correlated to the distance between root and inoculum. In the presence of decline soil, hyphal response was significantly reduced. This reduction can be related to the decreased infectivity of decline soil and may help to explain take-all decline.     

Dynamics of root colonization by the take-all fungus, Gaeumannomyces graminis

The progressive colonization of wheat seminal roots by Gaeumannomyces graminis var. tritici was monitored following inoculation by single inoculum units. G. graminis grew equally well above and below inoculation sites prior to blockage of the stele but after this growth was favoured up roots, above inoculation sites, rather than down, resulting in an asymmetrical pattern of root colonization. This asymmetrical pattern was common to superficial and cortical runner hyphae. It is suggested that cessation of host assimilate supply to the distal portion of infected roots inhibited further extensive growth of G. graminis. This hypothesis was tested by comparing extents of colonization by G. graminis on seminal roots of wheat with normal, enhanced and diminished assimilate supplies. A diminished assimilate supply to infected roots retarded the extent of pathogen colonization.     

Population dynamics of amoebae in soils suppressive and non-suppressive to wheat take-all

Populations of soil amoebae were estimated throughout two wheat crops and a fallow from wheat rotations suppressive and non-suppressive to take-all. The suppressive pasture-pasture-wheat plot showed higher rhizosphere populations of mycophagous amoebae and higher associations of mycophagous and other amoebae with wheat roots throughout the crop period than the non-suppressive continuous-wheat plot. Populations of mycophagous amoebae were positively and significantly (P = 0.05) correlated with populations of the take-all fungus in both soils, and with soil moisture in the non-suppressive soil. The possible role of mycophagous amoebae in the wheat rhizosphere in suppressive soil is discussed.     

Growth of Gaeumannomyces graminis var. tritici in soil: Effects of temperature and water potential

The effects of temperature and water on the growth of the take-all fungus, Gaeumannomyces graminis var. tritici (Ggt), were examined in two factorial experiments. The first examined the effects of temperature and water potential on the growth of two isolates of Ggt on agar media, using osmotically-adjusted water potentials. The second experiment was concerned with the growth of the Ggt isolates in one sterile and two natural soils at two water regimes in the absence of a living host. Three temperatures (10, 18 and 26°C) were used in these experiments. A third experiment determined growth through soil.Growth was greatest at high temperatures and low water potential in axenic culture, but in unsterile soil growth at different temperatures and water potentials was strongly influenced by competition from the soil biota. The best temperature for growth in unsterile soil was 18°C. Growth at 26°C in unsterile soil was greatly reduced, this being attributed to more intense microbial competition. In sterile soil Ggt grew equally well at 18 and 26°C. At 10°C, both isolates of Ggt grew better in unsterile soil than in sterile soil.Under suitable conditions Ggt grew out readily from infected straw into unsterile soil (up to 5 cm in 10 days) in the absence of a host plant, forming melanized, hyaline and branched hyphae. These hyphae were infectious after dry storage for 5 months in the laboratory. Ggt thus appears to be a more successful soil inhabitant than is widely believed. Our experiments could explain many of the host-based concepts related to field expression of disease.The technique presented here could be of value for testing the suppressiveness or conduciveness of soils by measuring fungal growth in soil.     

Colonization of wheat roots by Gaeumannomyces graminis inhibited by specific soils,microorganisms and ammonium-nitrogen

Lineal extension of Gaeumannomyces graminis var. tritici hyphae along roots of intact wheat plants growing in soils was measured. Hyphal growth rates were lower in soils treated with NH₄⁺-N than with NO₃^?-N. In a soil that is suppressive to the take-all disease, the controlling influence of NH₄⁺-N was eliminated by soil fumigation (methyl bromide), and reintroduced to fumigated soil by additions of 1% nonsterile soil. Effects of fumigation on hyphal growth were absent in a nonsuppressive soil, and in NO₃^?-treatments of the suppressive soil. When inocula of selected groups of wheat rhizoplane microflora were reintroduced into a fumigated or a soil-reinoculated soil via a root-food base, the Pseudomonas spp. consistently appeared more suppressive in NH₄⁺-N treatments than the general bacterial flora, Bacillus spp. spores, streptomycetes, and fungi.     

The effect of tillage practices on distribution,size, infectivity and propagule number of the take-all fungus (Gaeumannomyces graminis var. tritici)

Field trials in 1985 and 1986 were undertaken at a site in the Western Australian wheatbelt to determine the effects of direct-drilling, shallow cultivation and deep-cultivation on the distribution and infectivity of inoculum of the take-all fungus (Gaeumannomyces graminis var. tritici). Soil cores were taken shortly after cultivation and bioassayed to determine infectivity, take-all was assessed (in 1985) and soils were bioassayed post-harvest to determine propagule numbers and infectivity of different inoculum fractions.Inoculum was found to be more infective in the upper 3.3 cm of the direct-drilled plots, and was distributed more evenly to 10 cm in shallow-cultivated and deep-cultivated plots. Take-all was most severe in deep-cultivated plots and least in direct-drilled plots, but post-harvest propagule numbers were greatest after direct-drilled and deep-cultivated treatments and least in shallow-cultivated plots. Most infectious fragments of inoculum were those greater than 0.5 mm, but in 1986 those greater than 5 mm were more infective than any of the others (> 2 mm, > 0.5 mm and < 0.5 mm). Most infectious fragments occurred in the deep-cultivated plots, least in the shallow-cultivated plots and fractions were distributed differently with depth and treatment.