Control of Blumeria graminis f.sp. hordei by treatment with mycelial extracts from cultured fungi

Authors


*To whom correspondence should be addressed.†E-mail: dbc@kvl.dk

Abstract

High levels of protection against infection from Blumeria graminis f.sp. hordei in the susceptible barley cv. Pallas were obtained by pretreatment with mycelial extracts or culture filtrates from seven different fungi (Bipolaris oryzae, B. sorokiniana, Drechslera teres f. maculata, Fusarium culmorum, Trichoderma harzianum, Pythium ultimum and Rhizopus stolonifer). Three mycelial extracts from the taxonomically different fungi B. oryzae, P. ultimum and R. stolonifer were selected for detailed study. In general the number of colonies formed was reduced by 70–98% compared with controls. Furthermore, the few colonies that developed on treated leaves were generally smaller and showed reduced spore production. Protection was limited to the area of the leaves treated with mycelial extracts and a systemic effect could not be detected. No differences in the protection level were observed when treatment was conducted between 2 h and 3 days before inoculation. It is suggested that components in the mycelial extracts interacted directly with B. graminis and that this antifungal effect was responsible for the observed protection. However, indirect effects of the mycelial extracts mediated by locally induced resistance cannot be precluded.

Introduction

When not controlled, barley powdery mildew is a devastating disease that may cause severe yield losses in barley (Hordeum vulgare) production. Barley powdery mildew is caused by the obligate biotrophic fungus Blumeria graminis f.sp. hordei (syn. Erysiphe graminis f.sp. hordei). Control of B. graminis in barley is mainly achieved by fungicide treatment, but it is a problem to obtain effective control as the fungal population has the potential to develop tolerance against most used fungicides (De Waard et al., 1993). Alternatively, control can be obtained by the use of plant resistance. However, most known race-specific resistance genes currently used are matched by virulence in the fungal population, resulting in reduced efficiency in the field (Jørgensen, 1993). This, together with growing public concern about food quality, scepticism about the use of genetically modified organisms, and the impact of agricultural production on the environment, indicates the need for other methods of plant protection. These should involve integrated disease management systems, including the use of more durable forms of resistance and biological control methods. Biopesticides, based on natural components extracted from microorganisms, offer one means of biological control. Potentially, biopesticides are likely to be highly biodegradable without any adverse impact on, for example, the ground water. Compared with living biological control agents, use of a biopesticide is less sensitive to environmental conditions and the potential risks of introducing organisms to natural ecosystems is eliminated.

Only a few investigations have considered the approach of using components from microorganisms as biopesticides (Schönbeck et al., 1980, 1982; Schönbeck & Dehne, 1986; Reglinski et al., 1990, 1994a,b). Reglinski et al. (1990, 1994a) showed that yeast cell wall extracts were able to protect barley against B. graminis f.sp. hordei in the field. Protection was suggested to involve activation of host plant resistance against fungal penetration attempts, based on the ability of the extracts to induce phenylalanine ammonia lyase (PAL) accumulation and enhance papilla formation (Reglinski et al., 1994b). In wheat, Schönbeck et al. (1982) obtained protection against B. graminis f.sp. tritici by treating plants with culture filtrates from Bacillus subtilis. Here, protection was suggested to involve activation of host plant resistance after B. graminis penetration (Stenzel et al., 1985).

Many purified microbial compounds such as polysaccharides, fatty acids, peptides, enzymes and glycoproteins have been identified as general activators of host defence responses in many different plant species (reviewed by Dixon, 1986; Ebel, 1986). Furthermore, compounds such as chitin and chitosan from fungal cell walls have been shown both to act as elicitors of defence responses and to possess antimicrobial effects (Allan & Hadwiger, 1979; Hadwiger & Beckman, 1980; Barber et al., 1989).

In the current investigation, the barley powdery mildew system was chosen to investigate the disease-controlling effect of fungal mycelial extracts applied to barley leaves. Fungi belonging to different taxonomic groups, and hence with different cell wall compositions (Bartnicki-Garcia, 1968), were chosen as sources for the mycelial extracts. A method for producing mycelial extracts is reported, together with a bioassay suitable for testing their ability to protect barley against powdery mildew. For convenience the term fungal is used throughout although P. ultimum is a member of the Oomycota.

Materials and methods

Mycelial extracts

Isolates of the following seven fungal species were used as sources for mycelial extracts: Bipolaris oryzae (isolate NN007964), Bipolaris sorokiniana (isolate CP 1623), Drechslera teres f. maculata (isolate CP 2051), Fusarium culmorum (isolate no. 5), Trichoderma harzianum (isolate T3), Pythium ultimum (isolate Hb2), and Rhizopus stolonifer (isolate 1521).

The fungi were grown for 7 days on potato dextrose agar plates (PDA, Difco, Detroit, MI, USA) at 20°C in darkness and then cultured in potato dextrose broth (PDB, Difco) in 150 mL conical flasks on an orbital shaker at 125 r.p.m. for 5 days at 20°C in light. Mycelia were harvested by filtration on four layers of cheesecloth, rinsed with sterile distilled water and then weighed. It was necessary to add sterile water (1 mL g−1 for the B. oryzae extract, 2 mL g−1 for the P. ultimum extract, and 3 mL g−1 for the R. stolonifer mycelial extract) to homogenize the mycelia with an Ultraturrax (T25, Janke & Kunkel, IKA Labortechnik, Staufen, Germany). Further breakdown of the mycelia was achieved by treatment with 2 µg Novozyme (Novo Nordisk A/S, Bagsværd, Denmark) per mL mycelial extract overnight on a magnetic stirrer at 20°C. The mycelial extracts were then filtrated through four layers of cheesecloth, and centrifuged at 15 000 g for 10 min before sterile filtration (0·45 µm, Sartorius, Minisart, Germany). The preparations were stored at −20°C until use. In one experiment mycelial extracts from B. oryzae, P. ultimum and R. stolonifer were autoclaved before use. Before the mycelial extracts were used in the bioassay, Tween 20 surfactant was added to give a final concentration of 0·1%. The growth media remaining following removal of fungal mycelia were designated culture filtrates. These culture filtrates were also tested in the bioassay after sterile filtration and addition of Tween 20 (see above).

Plants and pathogens

Barley cv. Pallas or cv. Golden Promise, both susceptible to the B. graminis isolates employed, were used as test plants. Seeds were sown in K-soil (Svalöf Weibull AB, Hammenhög, Sweden) and maintained at 20°C/15°C in a growth chamber with cycles of 16 h light (∼200 µmol m−2 s−1) and 8 h darkness until they reached the desired leaf stage. First, second and fourth leaves were tested.

Barley powdery mildew [B. graminis f.sp. hordei (syn. E. graminis f.sp. hordei)] isolate A6 was maintained on the barley line Pallas P-01 and B. graminis isolate C-15 was maintained on the near-isogenic barley line Pallas P-02 in separate growth chambers, under the same conditions as mentioned above. To secure fresh inoculum for experimentation, old spores were removed by shaking the mildewed plants 24 h before inoculation.

Bioassay

Possible effects of mycelial extracts and culture filtrates were tested in a bioassay. Leaves of test plants, similar in size, were fixed horizontally to a Perspex plate with unbleached cotton strings, abaxial side up (Fig. 1) (Thordal-Christensen & Smedegaard-Petersen, 1988). Mycelial extracts and culture filtrates were gently applied with a cotton stick to the leaf area between the strings. Leaves were then left to dry normally for 24 h before inoculation. However, even at earlier inoculation times, leaves were always dry when inoculated. Inoculations were performed by placing test plants in an inoculation tower and shaking heavily mildewed plants above the open end of the tower to give an inoculation density of 5–10 spores per mm2. After inoculation, plants were incubated under the same conditions as mentioned above.

Figure 1.

 First leaves of barley cv. Pallas treated with Bipolaris oryzae mycelial extract, water, Pythium ultimum mycelial extract, untreated, and Rhizopus stolonifer mycelial extract 24 h before inoculation with Blumeria graminis f.sp. hordei isolate A6. Note the local effect of mycelial extract treatments (only the leaf area between the two cotton strings has been treated). Photograph taken 7 days after inoculation.

Two controls were included in each experiment: untreated leaves and leaves treated with 0·1% Tween 20 solution (designated water-treated). Novozyme was not added to the water-treated control, because preliminary experiments had shown that 2 µg mL−1 of this enzyme did not significantly change the numbers of B. graminis colonies compared with water-treated and untreated controls. The water treatment without Novozyme was chosen as the standard reference.

Assessment of disease was made 7 days after inoculation by counting the number of colonies on 2·5 cm of the middle part of the treated leaf area. In some cases, leaves with colonies were assessed visually and classified according to a scale from 0 to 4 as follows: type 0, no colonies present; 1, very sparse mycelial growth and sporulation; 2, sparse mycelial growth and sporulation; 3, nearly normal mycelial growth, but reduced sporulation; and 4, fully developed heavily sporulating colonies.

Experimental designs

Test of mycelial extracts and culture filtrates

The effect of mycelial extracts and culture filtrates from all seven fungi was tested on first leaves (10 days old) of cv. Pallas inoculated with B. graminis isolate A6, according to the bioassay. Mycelial extracts from B. oryzae, P. ultimum and R. stolonifer were further tested on second (14 days old) and fourth leaves (30 days old) on both barley cultivars, and against both B. graminis isolates.

Test of the persistence of mycelial extract treatments

Persistence of treatments was tested by application of mycelial extracts to all test plants of cv. Pallas at the same time. Individual inoculations with B. graminis isolate A6 were performed at 2, 12, 24, 48 and 72 h after treatment. A time lag > 3 days between extract application and inoculation could not be tested because leaf senescence preceded symptom appearance.

Test of systemic effect of mycelial extract treatment

Possible systemic effects of mycelia extract treatments were tested on first leaves of cv. Pallas. Mycelial extracts were applied as described above. Seven days after treatment of first leaves, second leaves were inoculated with B. graminis isolate A6. Assessment of number of colonies on the second leaf was made 7 days after inoculation. This experiment was repeated twice. In another experiment, second leaves were treated with mycelial extracts and third leaves were inoculated with B. graminis isolate A6 48 h after treatment.

Test of possible antifungal effects of the mycelial extracts

To determine whether mycelial extracts had direct antifungal effects on powdery mildew development, the mycelial extracts were removed from the leaves before inoculation. This was done by treating leaves of cv. Pallas as described for the bioassay and then, just before inoculation with B. graminis isolate A6, the mycelial extract residues left on the leaf surface were removed by gently wiping the leaves with wet cotton as described by (Cho & Smedegaard-Petersen, 1986).

All experiments were repeated at least twice, except for the one with fourth leaves, which was only performed once.

Statistical analysis

Number of colonies recorded in the experiments is a discrete variable assumed to follow a Poisson distribution. Therefore, data were analysed by logistic regression (proc genmod) by PC-SAS (release 6·12; SAS Institute, Cary, NC, USA) corrected for overdispersion (Collett, 1991) when present.

Results

Test of mycelial extracts and culture filtrates

In no case did leaf treatment with any mycelial extract or culture filtrate cause any visible phytotoxic effects. However, in all cases, treatment with mycelial extracts or culture filtrates from B. oryzae, B. sorokiniana, D. teres, F. culmorum, P. ultimum, R. stolonifer and T. harzianum significantly reduced (P < 0·001) the number of B. graminis isolate A6 colonies on first leaves of cv. Pallas. The average reduction in number of colonies compared with the water-treated control corresponded to percentage reductions of 69·5 and 98·5% for treatments with mycelial extracts and between 68·5 and 93·8% for treatments with culture filtrates (Fig. 2).

Figure 2.

 Average number of Blumeria graminis f.sp. hordei isolate A6 colonies in two independent experiments on first leaves of barley cv. Pallas treated with different mycelial extracts or culture filtrates 24 h before inoculation. Untreated and water-treated leaves were included as controls. Statistical comparisons are made with the water-treated control: ns, nonsignificant difference; ***significant difference with P ≤ 0·001.

After the initial test, mycelial extracts from B. oryzae, P. ultimum and R. stolonifer were selected for further detailed investigation because they represent different taxonomic groups and hence exhibit different cell wall compositions (Bartnicki-Garcia, 1968). Treatment of first, second and fourth leaves of cv. Pallas with the three selected mycelial extracts gave high protection against B. graminis when tested in the bioassay (Tables 1 and 2). On first leaves (Table 1) the protection was tested against two B. graminis isolates (A6 and C15). The protection level, measured as reduction in the number of B. graminis colonies, was above 90% for isolate A6, and above 80% for isolate C15. The protective effects of mycelial extracts were also evident on second leaves, although this effect appeared somewhat less marked and more variable than on first leaves (Table 2). On fourth leaves, the protection was 100% for the B. oryzae and P. ultimum mycelial extracts, and for R. stolonifer extract-treated leaves, the number of B. graminis colonies was reduced by 97%. Autoclaving the mycelial extracts did not influence their ability to reduce numbers of powdery mildew colonies compared with nonautoclaved mycelial extracts (data not shown).

Table 1.   Average number of Blumeria graminis f.sp. hordei isolate A6 and C15 colonies on first leaves of barley cv. Pallas pretreated with mycelial extracts or water 24 h before inoculation. An untreated control was included for comparison
   Mycelial extract
TreatmentUntreatedWater-treatedBipolaris oryzaePythium ultimumRhizopus stolonifer
  1. Statistical comparisons are made with the water-treated control: ns, nonsignificant difference; ***significant difference with P ≤ 0·001; *significant difference with P ≤ 0·05.

  2. Values in parentheses are percentage reductions compared with the water-treated control.

Isolate A672·3 ns66·23·6*** (95)6·7*** (90)5·7*** (91)
Isolate C1549·6*36·76·9*** (81)4·0*** (89)2·0*** (95)
Table 2.   Average number of Blumeria graminis f.sp. hordei isolate A6 colonies on second leaves (14-day-old plants) and fourth leaves (4-week-old plants) of barley cv. Pallas pretreated with mycelial extracts or water 24 h before inoculation. An untreated control was included for comparison
   Mycelial extract
TreatmentUntreatedWater-treatedBipolaris oryzaePythium ultimumRhizopus stolonifer
  1. Statistical comparisons are made with the water-treated control: ns, nonsignificant difference; ***significant difference with P ≤ 0·001; **significant difference with P ≤ 0·01; *significant difference with P ≤ 0·05.

  2. Values in parentheses are percentage reductions compared with the water-treated control.

Second leaf17·8 ns18·84·3*** (77)10·6** (44)8·0* (57)
Fourth leaf15·9 ns16·20·0 (100)0·0 (100)0·5*** (97)

On second leaves of cv. Golden Promise, treatments resulted in average reductions in colony number by 89, 82 and 87% for B. oryzae, P. ultimum and R. stolonifer mycelial extract-treated leaves, respectively (Fig. 3). Besides a reduction in actual number, the B. graminis colonies that developed on mycelial extract-treated leaves were often smaller, with lower spore production than on control leaves. On the 0–4 scale, B. graminis colonies which developed on mycelial extract-treated leaves were typically type 1 or 2, compared with 3 or 4 on control leaves (Fig. 4).

Figure 3.

 Average number of Blumeria graminis f.sp. hordei isolate A6 colonies on second leaves of barley cv. Golden Promise treated with mycelial extracts or water 24 h before inoculation. An untreated control was included for comparison. Statistical comparisons are made with the water-treated control: ns, nonsignificant difference; ***significant difference with P ≤ 0·001.

Figure 4.

 (a) Distribution of first leaves of barley cv. Pallas classified according to Blumeria graminis f.sp. hordei isolate A6 colony morphology types after treatment with mycelial extracts 24 h before inoculation. Untreated and water-treated leaves were included as controls. Each leaf was classified according to a scale from 0 to 4 for colony morphology: type 0, no colonies present; 1, very sparse mycelial growth and sporulation; 2, sparse mycelial growth and sporulation; 3, nearly normal mycelial growth, but reduced sporulation; and 4, fully developed heavily sporulating colonies. (b) Percentage reduction in number of colonies compared with the water-treated control in the associated bioassay.

Test of the persistence of mycelial extract treatments

Inoculation of mycelial extract-treated first leaves of cv. Pallas at different times after application of mycelial extracts showed that the protection against B. graminis isolate A6 was achieved almost immediately after treatment and lasted for at least 3 days (Fig. 5).

Figure 5.

 Percentage reduction in number of Blumeria graminis f.sp. hordei isolate A6 colonies on first leaves of barley cv. Pallas treated with mycelial extracts compared with the water-treated control. Inoculations were performed 2, 12, 24, 48 or 72 h after treatment. An untreated control was included for comparison. Statistical comparisons were made with the water-treated control within each time point: ns = nonsignificant difference; ***significant difference with P ≤ 0·001.

Test of systemic effect of mycelial extract treatment

The protection obtained by the mycelial extract treatments was strictly localized to the treated area of the leaf. Spread of the protection was not observed acropetally or basipetally (see Fig. 1), or transversely in the leaf (data not shown). No consistent systemic effect of the treatments was observed in cv. Pallas when first or second leaves were treated with mycelial extracts and the leaves formed above were inoculated with B. graminis isolate A6 (Table 3).

Table 3.   Average number of Blumeria graminis f.sp. hordei isolate A6 colonies on upper leaves of barley cv. Pallas when lower leaves were pretreated with mycelial extracts or water before inoculation. An untreated control was included for comparison
   Mycelial extract
TreatmentUntreatedWater-treatedBipolaris oryzaePythium ultimumRhizopus stolonifer
  • a

    First leaves of barley cv. Pallas treated with mycelial extracts and second leaves inoculated with B. graminis f.sp. hordei 7 days later.

  • b

    Second leaves of barley cv. Pallas treated with mycelial extracts and third leaves inoculated with B. graminis f.sp. hordei 48 h later.

  • c

    No reduction in number of colonies. Values in parentheses are percentage reductions compared with the water-treated control.

  • Statistical comparisons are made with the water-treated control: ns, nonsignificant difference; **significant difference with P ≤ 0·01.

Experiment 1a51·6 ns60·435·7** (41)56·7 ns (6)51·3 ns (15)
Experiment 2a33·3 ns38·936·6 ns (6)37·0 ns (5)29·7 ns (24)
Experiment 3b24·9 ns23·624·9 ns (–)c31·4 ns (–)23·2 ns (2)

Test of possible antifungal effects of the mycelial extracts

The possible effect of removing superficial residues of the mycelial extract treatment by wiping with wet cotton was tested by wiping leaves 24 h after treatment and just before B. graminis inoculation. When leaves were wiped with wet cotton, the numbers of B. graminis colonies formed were significantly higher than on nonwiped treated leaves (Fig. 6). However, wiping the leaves still resulted in a significant reduction in numbers of colonies compared with the water-treated control.

Figure 6.

 Average number of Blumeria graminis f.sp. hordei isolate A6 colonies on first leaves of cv. Pallas in an ordinary bioassay (black bars) compared with that on leaves wiped with wet cotton before inoculation (hatched bars). An untreated control was included for comparison. Statistical comparisons are made with the water-treated control within the ordinary bioassay (black bars) and the wiped bioassay (hatched bars): ns, nonsignificant difference; ***significant difference with P ≤ 0·001.

Discussion

Treatment of barley with mycelial extracts and culture filtrates from seven different fungi resulted in effective protection against B. graminis f.sp. hordei infection. The high protection was similar to levels of protection normally obtained using conventional B. graminis control in the field. Extracts from microorganisms have previously been reported to reduce infection by B. graminis in cereals (Schönbeck et al., 1982; Schönbeck & Dehne, 1986; Reglinski et al., 1994b). Reglinski et al. (1994b) showed that cell wall extracts from S. cerevisiae were able to reduce the number of B. graminis f.sp. hordei colonies on detached leaves by up to 95%, and in field experiments a 40% reduction in disease severity was obtained. In wheat, a culture filtrate from B. subtilis reduced the number and size of B. graminis f.sp. tritici colonies both in glasshouse and field experiments (Schönbeck et al., 1982; Schönbeck & Dehne, 1986). The level of protection in a commercial wheat field was reported to be > 90% following five applications of the culture filtrate (Schönbeck et al., 1982).

To help application of the mycelial extracts to leaves, Tween 20 was added as a surfactant. Therefore, two types of controls were included in the present investigation: untreated plants and plants treated with water to which 0·1% Tween 20 was added. The water-treated leaves often showed a slight, though nonsignificant, reduction in the number of B. graminis colonies compared with untreated leaves. This might indicate an effect of the application method or the surfactant. In order to avoid the influence of the application method/surfactant, possible effects of mycelial extracts were always compared with the water-treated control. However, this meant that the percentage of overall disease reduction when compared with untreated leaves was higher than the 70–98% reduction in number of colonies obtained when compared with water-treated controls. This has to be taken into account when comparing the effect obtained using the tested mycelial extracts with other similar studies such as that of Reglinski et al. (1994b). Regarding the culture filtrates, it cannot be discounted that the culture media itself may have contributed to the protective effect, but this remains to be determined.

The effect of the mycelial extracts persisted for at least 3 days before inoculation and was expressed as early as 2 h after treatment. Cho & Smedegaard-Petersen (1986) have previously reported rapid induction of resistance in barley against B. graminis. They found that a 30-min induction period with either a virulent or avirulent B. graminis isolate was enough to induce significant resistance against a later B. graminis attack.

When colonies did develop on mycelial extract-treated leaves, they appeared smaller and produced fewer spores than colonies on untreated or water-treated leaves. The consequence of reduced spore production was not measured in detail, but it should reduce secondary infections and hence reduce the overall disease incidence in the field. Reduced colony size and spore production following treatment of wheat leaves with culture filtrate from B. subtilis were also reported by Stenzel et al. (1985). Here, spore production from B. graminis f.sp. tritici was reduced by ∼70% compared with the control.

The mycelial extract treatment seemed to interact directly with B. graminis because the protection efficiency decreased when superficial residues from the treatment were removed by wiping treated leaves with wet cotton before inoculation. The method does not allow a quantification of the direct effect, as it was difficult to confirm that all of the mycelial extracts were removed; the possible influence of wounding was also difficult to assess. However, other protection mechanisms may also be involved.

Another protection mechanism might be induced resistance. In the barley–B. graminis f.sp. hordei system, disease reductions suggested to be due to local induced resistance have been demonstrated with biological agents ranging from living pathogens and saprophytes to various exudates and microbial components. Examples include compatible and incompatible races of B. graminis f.sp. hordei (Cho & Smedegaard-Petersen, 1986; Thordal-Christensen & Smedegaard-Petersen, 1988), Bipolaris maydis (Jørgensen et al., 1996), Cladosporium macrocarpum (Gregersen & Smedegaard, 1989), culture filtrate from B. subtilis (Kehlenbeck & Schönbeck, 1995), yeast cell wall extracts (Reglinski et al., 1990), and compounds from conidia of B. graminis (Kristensen & Smedegaard-Petersen, 1997). Conversely, there are only two reports of systemic induced resistance in barley against B. graminis (Hwang & Heitefuss, 1982; Fujiwara et al., 1989). In general there are relatively few reports of systemic induction of resistance in monocots following induction with either biological (Hwang & Heitefuss, 1982; Fujiwara et al., 1989; Smith & Metraux, 1991; Kumar et al., 1993; Manandhar et al., 1998a) or chemical inducers (Cartwright et al., 1980; Walters et al., 1993; Kogel et al., 1994; Görlach et al., 1996; Jin et al., 1997; Schweizer et al., 1997; Manandhar et al., 1998b; Morris et al., 1998) and they have seldom been confirmed by independent laboratories. Hence, in monocots, induced resistance seems to be a mainly local phenomenon.

The protection obtained following treatment with mycelial extracts gave only local protection that did not extend beyond the treated area, either within the leaf or from leaf to leaf. So if induced resistance is involved in the protection mechanism, it is local in its nature. However, to determine whether local induced resistance is involved in the protection mechanism exerted by mycelial extracts, it should be demonstrated that these can activate the plant's own defence system. Such an investigation should include histological studies to determine when fungal growth is arrested and molecular analysis to show whether known resistance mechanisms are actually activated.

In conclusion, it is suggested that the mycelial extracts protect barley against B. graminis f.sp. hordei by a direct fungitoxic effect. However, it is a possibility that local induced resistance is also involved in the protection mechanism.

Acknowledgements

Annette Jensen is gratefully acknowledged for skilful technical assistance. We are indebted to Drs Inge MB Knudsen and Helge Green for providing the F. culmorum and T. harzianum isolates, and to Prof. Lene Lange for stimulating and supporting this project in its initiation, and for providing the B. oryzae isolate and Novozyme. The work was financed by the Ministry of Environment and Energy, the Danish Environmental Protection Agency.

Footnotes

  1. †E-mail: dbc@kvl.dk

Ancillary