Inhibition of Pathogenicity of the Rice Blast Fungus by Saccharomyces cerevisiae α-Factor

See allHide authors and affiliations

Science  16 May 1997:
Vol. 276, Issue 5315, pp. 1116-1118
DOI: 10.1126/science.276.5315.1116


Magnaporthe grisea is a fungal pathogen with two mating types, MAT1-1 and MAT1-2, that forms a specialized cell necessary for pathogenesis, the appressorium.Saccharomyces cerevisiae α-factor pheromone blocked appressorium formation in a mating type–specific manner and protected plants from infection by MAT1-2 strains. Experiments with α-factor analogs suggest that the observed activity is due to a specific interaction of α-factor with an M. grisea receptor. Culture filtrates of a MAT1-1 strain contained an activity that inhibited appressorium formation of mating type MAT1-2 strains. These findings provide evidence that a pheromone response pathway exists in M. grisea that can be exploited for plant protection.

The heterothallic ascomyceteMagnaporthe grisea is a pathogen of a wide variety of grasses but is best known as the causal agent of rice blast disease. The costs of controlling disease with fungicides and the difficulty in breeding durable and effective resistance have led to intense interest in understanding the mechanisms governing the pathogenicity of this fungus (1). Conidia of M. grisea attach to the plant host with an adhesive that is released from the tip of the conidium upon hydration (2). After germination, the fungus responds to contact with the host surface by producing an appressorium, a specialized cell that uses turgor pressure to aid in penetration of the host cell (3).

The mating behavior of M. grisea is determined by the mating-type locus, which contains either MAT1-1 orMAT1-2 DNA. One parent of each of the two mating types participates in a sexual cross. The mating-type loci of filamentous ascomycetes are thought to encode master regulators that control the expression of mating type–specific genes, such as pheromones and pheromone receptors (4). In Saccharomyces cerevisiae, α-factor and a-factor pheromones are produced by strains with MATα and MAT amating types. Each pheromone is recognized by a corresponding heterotrimeric GTP-binding protein–coupled receptor expressed in the opposite mating type (5).

Appressorium formation of mating type MAT1-2 strains ofM. grisea is inhibited when conidia are germinated in the presence of 2% yeast extract. However, 2% peptone and 2% tryptone do not inhibit appressorium formation (6). We found that appressorium formation of MAT1-1 strains was not inhibited by yeast extract to the same degree as MAT1-2 strains (Table1). Yeast extract contained an unidentified factor that could be partially purified by an organic extraction procedure designed for purification of small peptides (7). This fraction inhibited appressorium formation in MAT1-2 strains, and the active component appears to be a polypeptide. We found 91 ± 7% appressorium formation of strain 4091-5-8 (MAT1-2) with proteinase K–treated extract and 1 ± 1% appressorium formation with untreated extract (8). The mating type–specific effect of yeast extract on appressorium formation suggested that M. grisea might respond to a pheromone by suppressing infection-related development.

Table 1

Mating type–specific inhibition of appressorium formation in M. grisea. Conidia were incubated in 50 mM potassium phosphate buffer (pH 6.5) (control), 2% yeast extract, α-factor pheromone in 50 mM potassium phosphate buffer (pH 6.5), or extracts of strain CP987 culture filtrate. Appressorium formation assays were performed as described (6). In each of three experiments, a minimum of 300 conidia were counted. The average number of conidia producing at least one appressorium is reported. Variation in appressorium formation between experiments in controls was ±10% or less; appressorium formation in MAT1-1 strains in yeast extract or CP987 extracts varied by ±20% or less; forMAT1-1 strains in α-factor, variation was ±10% or less; for MAT1-2 strains in yeast extract, α-factor, or CP987, extract variation was ±2% or less.

View this table:

The α-factor pheromone of S. cerevisiae has activity in closely related yeast species (9). We tested the effect of synthetic S. cerevisiae α-factor on M. griseaand found that appressorium formation was inhibited inMAT1-2 strains (Table 1 and Fig. 1, A and B) but not in MAT1-1 strains (Table 1 and Fig. 1C). The concentration of α-factor needed to cause >95% inhibition of appressorium formation of all MAT1-2 strains tested was 300 μM. This is 104-fold higher than the concentration of α-factor required to induce morphological changes in S. cerevisiae (10). Because it does not contain α-factor (7), we conclude that yeast extract contains a distinct peptide that has a specific biological effect on M. grisea.

Figure 1

Mating type–specific inhibition of appressorium formation. Conidial suspensions (1 × 104conidia per ml) were inoculated onto Teflon film (DuPont). Conidia are indicated by arrowheads. Appressoria are spherical cells indicated by arrows. (A) Germlings of strain 4091-5-8 (MAT1-2) in 50 mM phosphate buffer. Addition of 300 μM α-factor inhibits appressorium formation of strain 4091-5-8 (B) but not strain CP987 (MAT1-1) (C) or strain 4091-5-8 germinated in the presence of 300 μM α-factor and 10 mM cAMP (D). Extract of strain CP987 culture filtrate blocks appressorium formation in strain 4091-5-8 (E) but not in strain CP987 (F). Samples were incubated for 16 hours and the Teflon sheet was placed on a microscope slide before examination by bright-field microscopy. Bar, 30 μm.

We tested segregation of mating type and sensitivity to α-factor by examining the progeny of a cross between strains CP987 and 4091-5-8. In the 31 progeny examined, inhibition of appressorium formation by α-factor cosegregated with the MAT1-2 mating type. As a further test that the mating type–specific response was tightly linked to the mating-type locus, we examined strains in which mating types had been switched by transformation with cosmid clones containing the mating-type loci (11). Strain CP2738 was derived from strain 4091-5-8 (MAT1-2) by transformation with MAT1-1mating-type DNA. The genome of CP2738 does not containMAT1-2 DNA, but does contain MAT1-1 DNA (11). Likewise, CP2735 is derived from the MAT1-1strain 4136-4-3. These strains are switched in mating behavior (11) and exhibit a corresponding switch in the response to α-factor (Table 1).

Appressorium formation is induced by cyclic adenosine 3′,5′-monophosphate (cAMP) (12). The inhibitory activity of α-factor toward MAT1-2 strains can be overridden by cAMP. For strain 4091-5-8, we found 79 ± 10% appressorium formation with α-factor and 10 mM cAMP, and 0.7 ± 0.3% appressorium formation with α-factor alone. Although, the appressoria often appeared to be immature or delayed in formation (Fig. 1D), this suggests that α-factor acts at an early stage in the pathway that signals appressorium formation and that the block can be reversed or bypassed by cAMP.

We examined a series of α-factor analogs to test the specificity of α-factor in inhibiting appressorium formation (Table2). Analogs that retained high activity toward S. cerevisiae also retained the ability to inhibit appressorium formation. No effect on growth or appressorium formation of either mating type of M. grisea was observed with thea-factor pheromone of S. cerevisiae(13).

Table 2

Biological activity of α-factor analogs. Appressorium assays were performed as described in the legend of Table1. α-Factor is H2N-WHWLQLKPGQPMY-COOH (17). Asparagine substituted for glutamine at position 5 is designated [Asn5]. The removal of tryptophan at position 1 is designated des-Trp1. Other substitutions are norleucine (Nle), cyclohexylalanine (Cha), and 3,4-dehydro-l-proline (DHP). The previously reported (10) concentration of analog required to induce mating projections in 50% of yeast a cells (bar1-1) is shown for comparison.

View this table:

To determine whether an activity could be identified from M. grisea that inhibited appressorium formation, we prepared extracts from culture filtrates of both mating types (14). The extract from the MAT1-1 strain inhibited appressorium formation in MAT1-2 strains but had little effect onMAT1-1 strains (Table 1 and Fig. 1, E and F). The extract derived from strain 4091-5-8 (MAT1-2) did not significantly inhibit appressorium formation in any strain.

To determine whether the appressorium-inhibiting activity ofMAT1-1 culture filtrates might be a polypeptide, we tested the sensitivity of the extract to protease digestion. Because chymotrypsin recognizes bulky hydrophobic residues in polypeptides, we used chymotrypsin linked to acrylic beads (Sigma) to examine whether the appressorium-inhibiting activity of MAT1-1 culture filtrates might be a polypeptide. Chymotrypsin eliminated the activity in MAT1-1 culture filtrates (90 ± 4% appressorium formation for strain 4091-5-8), but heat-inactivated chymotrypsin did not affect the activity (0.5 ± 0.1% appressorium formation). Chymotrypsin treatment of α-factor destroyed activity toward strain 4091-5-8 and the ability to arrest growth in S. cerevisiaeRC629; activity was retained when heat-treated protease was used.

As with the artificial substrates, we noted that appressorium formation by strain 4091-5-8 (MAT1-2) was inhibited when conidia germinated on the surface of barley leaves in the presence of 300 μM α-factor. To determine whether inhibition of appressorium formation could be correlated with reduced pathogenicity, we inoculated barley plants with conidia of strain 4091-5-8 and the corresponding switched mating-type strain CP2738 in the presence or absence of α-factor with norleucine substituted at position 12 (15). Significant protection against the MAT1-2 strain 4091-5-8 was observed (Fig. 2). Mean lesion densities for strain 4091-5-8 were 50 lesions per 12.5-cm leaf segment for the control and 2.3 lesions per 12.5-cm leaf segment for the α-factor–treated plants (t = 11.16, P < 0.0001, df = 22). α-Factor treatment did not significantly protect plants against strain CP2738 (MAT1-1) (t = 0.82, P = 0.43, df = 22). A similar result was obtained with the second switched mating-type pair. Strain 4136-4-3 (MAT1-1) was unaffected by α-factor, but infection of plants with strain CP2735 (MAT1-2) was significantly reduced by α-factor treatment.

Figure 2

Effect of α-factor pheromone on M. grisea pathogenicity. Fourteen-day-old barley seedlings were spray inoculated with 3 ml of 1% gelatin in 50 mM potassium phosphate buffer (pH 6.5) (A), 1% gelatin containing 104conidia per ml of strain 4091-5-8 (B), strain 4091-5-8 with 300 μM α-factor (C), 1% gelatin containing 104 conidia per ml of strain CP2738 (D), and strain CP2739 with 300 μM α-factor (E). The conidial suspensions (2 × 104 conidia per ml) were mixed with equal volumes of 1% gelatin in 50 mM potassium phosphate (pH 6.5) or 1% gelatin in 50 mM potassium phosphate containing 600 μM [Nle12] α-factor to give a final suspension containing 1 × 104 conidia per ml.

The ability of α-factor to block appressorium formation in M. grisea can be explained if α-factor is similar in sequence to anM. grisea pheromone and can interact with a corresponding receptor. In this model, binding of pheromone to receptor interferes with signaling of appressorium development. Because cAMP induces appressorium formation in the presence of α-factor, pheromone–receptor interaction may inhibit appressorium formation by preventing cAMP accumulation. The correlation of inhibition of appressorium formation by α-factor analogs with biological activity in yeast argues for a specific recognition of α-factor byMAT1-2 strains. Structural and functional similarities of peptide hormones have been recognized previously. α-Factor has limited sequence similarity to gonadotropin-releasing hormone and stimulates the release of luteinizing hormone from cultured gonadotrophs, although α-factor is less active than the authentic hormone by a factor of 104 (16). α-Factor pheromones of three Saccharomyces species that differ from each other at four of 13 positions have similar interspecies activity in causing growth arrest (9).

Our results suggest that stimulation of the pheromone receptor ofM. grisea blocks pathogenic development. Field isolates ofM. grisea that infect rice are usually found to be infertile, although the cause of this infertility is not known. Thus, examination of the activity of α-factor against typicalMAT1-2 field isolates is needed to evaluate the use of pheromones for disease control. However, our results demonstrate the potential for fungal pheromones to control plant disease. Presumably, the authentic mating factors from M. grisea have biological activity at low concentrations, as is observed in the pheromone response in S. cerevisiae. As peptides, fungal pheromones are amenable to chemical modification to optimize their activity and to control their persistence in the environment.

  • * To whom correspondence should be addressed. E-mail: dje0282{at}


View Abstract

Navigate This Article