Independent Signaling Pathways Regulate Cellular Turgor during Hyperosmotic Stress and Appressorium-Mediated Plant Infection by Magnaporthe grisea

Correspondence to: Nicholas J. Talbot, N.J.Talbot{at}exeter.ac.uk (E-mail), 44-1392-264668 (fax)

	ABSTRACT

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION METHODS REFERENCES

The phytopathogenic fungus Magnaporthe grisea elaborates a specialized infection cell called an appressorium with which it mechanically ruptures the plant cuticle. To generate mechanical force, appressoria produce enormous hydrostatic turgor by accumulating molar concentrations of glycerol. To investigate the genetic control of cellular turgor, we analyzed the response of M. grisea to hyperosmotic stress. During acute and chronic hyperosmotic stress adaptation, M. grisea accumulates arabitol as its major compatible solute in addition to smaller quantities of glycerol. A mitogen-activated protein kinase–encoding gene OSM1 was isolated from M. grisea and shown to encode a functional homolog of HIGH-OSMOLARITY GLYCEROL1 (HOG1), which encodes a mitogen-activated protein kinase that regulates cellular turgor in yeast. A null mutation of OSM1 was generated in M. grisea by targeted gene replacement, and the resulting mutants were sensitive to osmotic stress and showed morphological defects when grown under hyperosmotic conditions. M. grisea osm1 mutants showed a dramatically reduced ability to accumulate arabitol in the mycelium. Surprisingly, glycerol accumulation and turgor generation in appressoria were unaltered by the osm1 null mutation, and the mutants were fully pathogenic. This result indicates that independent signal transduction pathways regulate cellular turgor during hyperosmotic stress and appressorium-mediated plant infection. Consistent with this, exposure of M. grisea appressoria to external hyperosmotic stress induced OSM1-dependent production of arabitol.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION METHODS REFERENCES

Cellular turgor in eukaryotic cells is regulated in response to the environment or as a component of developmental programs. The most apparent stimulus affecting cellular turgor is hyperosmotic stress, which cells contend with by generating high concentrations of compatible solutes to maintain cellular turgor and thus prevent water loss. Adjustment of cellular turgor is, however, also vital for the generation of the mechanical force that plants and microorganisms need for growth and proliferation, particularly in terrestrial environments. The importance of cellular turgor to microorganisms is exemplified by the plant pathogenic fungus Magnaporthe grisea, which causes a serious disease of rice (Talbot 1995 ; Howard and Valent 1996 ). M. grisea elaborates specialized cells called appressoria to breach the plant cuticle. M. grisea appressoria are dome-shaped cells that penetrate rice cuticles by generating mechanical force. This is achieved by producing enormous turgor pressure within the appressorium estimated to be as high as 8 mPa (Howard et al. 1991 ; Money and Howard 1996 ).

Appressorium turgor is focused on a small area at the base of the appressorium, allowing a penetration peg to rupture the underlying cuticle. Precisely how turgor is translated into mechanical force remains unclear, but it probably involves localized cell wall dissolution at the base of the appressorium and reorientation of the cytoskeleton to allow polarization of the penetration peg (Bourett and Howard 1990 , Bourett and Howard 1992 ). The infection process may be accelerated by the action of extracellular enzymes, but the fact that M. grisea appressoria can penetrate inert plastic surfaces shows that mechanical force is the primary means of infection (Howard et al. 1991 ). Recently, cellular turgor in appressoria of M. grisea was shown to be generated by accumulation of high concentrations of intracellular glycerol (de Jong et al. 1997 ). Glycerol concentrations in developing appressoria rise to as high as 3.2 M during turgor generation. Glycerol is maintained in the appressorium by means of the heavily melanin-pigmented cell wall, which provides a glycerol-impermeable layer (de Jong et al. 1997 ; Money 1997 ). Nonmelanized mutants of M. grisea lack this wall layer and are nonpathogenic and unable to generate turgor (Howard and Ferrari 1989 ; Chumley and Valent 1990 ; de Jong et al. 1997 ). In this study, we set out to investigate the control of cellular turgor in M. grisea by characterizing the response of the fungus to external osmotic stress and comparing the mechanism of this response to that of pressure generation in appressoria.

Cellular turgor in eukaryotes appears to be controlled by a conserved mitogen-activated protein kinase (MAPK) signaling system called the HOG (for high-osmolarity glycerol) pathway, which was first discovered in the budding yeast Saccharomyces cerevisiae (Brewster et al. 1993 ). In S. cerevisiae, the HOG pathway is composed of the SSK2-, SSK22-, and STE11-encoded MAP kinase kinase kinases (MAPKKKs), the PBS2-encoded MAP kinase kinase (MAPKK), and the HOG1 MAPK (Gustin et al. 1998 ). The HOG pathway can be activated in two ways. First, a membrane-bound osmosensor encoded by SHO1 can act on the MAPKKK Ste11p and the MAPKK Pbs2p, triggering phosphorylation of Hog1p (Maeda et al. 1995 ). In the second pathway, the SSK2- and SSK22-encoded MAPKKKs are activated by a three-component histidine–aspartate kinase complex, related to the two- and three-component sensors common in prokaryotes, encoded by SLN1, YPD1, and SSK1 (Maeda et al. 1994 ; Posas et al. 1996 ; Posas and Saito 1998 ). Stimulation of the HOG pathway in budding yeast leads to production of glycerol by transcriptional activation of GPD1, which encodes NADH-dependent glycerol-3-phosphate dehydrogenase, and HOR1, which encodes glycerol-3-phosphatase (Albertyn et al. 1994 ; Hirayama et al. 1995 ; Norbeck et al. 1996 ). The resulting accumulation of glycerol in yeast cells maintains cellular turgor even after exposure to severe hyperosmotic stress. Elements of the HOG pathway have also been found in mammals, Drosophila, and plants (Han et al. 1994 , Han et al. 1998 ; Popping et al. 1996 ) and shown either to be required for hyperosmotic adaptation (Sheikh Hamad et al. 1998 ) or to be capable of complementing corresponding yeast mutants (Kumar et al. 1995 ; Han et al. 1998 ).

In this study, we have characterized the cellular response of M. grisea to hyperosmotic stress. We show that M. grisea accumulates mannitol and trehalose as storage carbohydrates during normal growth and development. Arabitol, however, accumulates as the principal compatible solute during the response to hyperosmotic stress. We also present evidence that arabitol accumulation and resistance to hyperosmotic stress are controlled by OSM1, an osmosensory MAPK-encoding gene that is functionally homologous to S. cerevisiae HOG1. Remarkably, although cellular turgor is controlled by OSM1 during the response to external osmotic shock, appressorium turgor is unaltered in osm1 mutants. This suggests that plant pathogenic fungi have evolved specific signaling pathways for appressorium-mediated plant infection that operate independently of the conserved eukaryotic regulatory mechanism for modulation of cellular turgor.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION METHODS REFERENCES

Cellular Turgor Generation in the Rice Blast Fungus M. grisea
Hyperosmotic stress results in the production of compatible solutes, which cells use to prevent water loss and regulate their cellular turgor. In fungi, compatible solutes are predominantly polyols, including glycerol, which is generated in S. cerevisiae, and erythritol, which is used by Aspergillus nidulans (Blomberg and Adler 1992 ). In contrast, plants tend to accumulate proline or glycine betaine, whereas mammalian cells accumulate sorbitol, betaine, or ionic solutes (Lewis and Smith 1967 ; Blomberg and Adler 1992 ). We investigated compatible solute generation by M. grisea during both chronic and acute hyperosmotic stress. First, the nature of compatible solutes accumulating during hyperosmotic stress was determined by using gas–liquid chromatography. During normal growth and development, M. grisea accumulates trehalose and mannitol as its major storage carbohydrates in mycelium, as shown in Figure 1A. During hyperosmotic stress, which was imposed by incubating fungal mycelium in 0.4 M NaCl, levels of mannitol remained broadly constant, whereas trehalose levels fell. Arabitol, which is present only in trace amounts during growth of M. grisea under isoosmotic conditions, accumulated rapidly in response to hyperosmotic stress, as shown in Figure 1B. Smaller amounts of glycerol were also present in mycelial extracts and increased during osmotic stress. Arabitol was found to accumulate in direct proportion to external osmolarity (data not shown) and was the major solute accumulating under severe hyperosmotic conditions. Gas–liquid chromatography and ¹³C nuclear magnetic resonance analyses failed to identify any amino acids or other noncarbohydrate solutes accumulating in osmotically stressed mycelium (data not shown). Based on these observations, we conclude that M. grisea responds to hyperosmotic stress by accumulating arabitol as its principal osmolyte, with lower concentrations of glycerol also present in stressed mycelium.

View larger version (19K): [in this window] [in a new window]   Figure 1. Compatible Solute Production in M. grisea during Hyperosmotic Stress and Appressorium Turgor Generation. Compatible solute accumulation was assayed in M. grisea mycelium from the wild-type strain Guy11 after hyperosmotic stress. The mycelium was grown for 48 hr in complete medium (CM) before being transferred to CM or CM plus 0.4 M NaCl for an additional 24 hr. Carbohydrates were extracted and quantified by gas–liquid chromatography. Error bars indicate standard deviation from three repeats of the experiment. (A) Major storage carbohydrates accumulating in M. grisea mycelium during growth in isoosmotic conditions (CM). 1, mannitol; 2, trehalose; 3, glycerol; 4, arabitol. (B) Major storage carbohydrates accumulating in M. grisea mycelium after acute hyperosmotic stress (CM + 0.4 M NaCl). M. grisea accumulates mannitol and trehalose as its main storage carbohydrates during normal growth and development. Arabitol accumulates as the principal compatible solute during hyperosmotic stress. 1, mannitol; 2, trehalose; 3, glycerol; 4, arabitol. (C) Cytorrhysis (cell collapse) assay to determine the concentration of glycerol present in appressoria during maturation. Appressoria were allowed to form for 48 hr on the hydrophobic side of Gelbond membranes. Glycerol was added to a final concentration of 1 to 5 M. The number of appressoria is proportional to the mean concentration (conc.) of glycerol within the cells.

Appressoria of M. grisea accumulate large amounts of glycerol during turgor generation and plant infection. Glycerol has previously been measured in bulk appressorial extracts by use of an enzymatic assay (de Jong et al. 1997 ). The internal concentration of glycerol in these small cells (M. grisea appressoria are 7 to 8 µm in diameter) can be estimated more readily, however, using a cytorrhysis (cell collapse) assay (Howard et al. 1991 ; de Jong et al. 1997 ). Appressoria were allowed to form in water drops on the hydrophobic side of Gelbond membranes. Glycerol was then carefully added to the drop to give a final concentration of between 1 and 5 M glycerol. When the external glycerol concentration exceeds the intracellular glycerol concentration, appressoria collapse (de Jong et al. 1997 ), allowing an estimate of appressorial turgor to be made (Figure 1C). A concentration of 3.2 M glycerol was sufficient to collapse 50% of the mature appressoria, indicating that the mean glycerol concentration within appressoria (allowing for the presence of other cellular constituents) was within the 2 to 4 M range.

Identification of the OSM1 Gene Encoding a MAPK from M. grisea
A candidate gene encoding the stress-activated M. grisea Hog1/p38 MAPK was isolated by polymerase chain reaction (PCR) amplification with primers designed from conserved regions of the S. cerevisiae amino acid sequence (see Methods). The resulting PCR product was then used to probe a M. grisea genomic library and a cDNA library derived from M. grisea conidia. A genomic clone was isolated, and two genomic subclones were sequenced. Sequencing of 4 kb of DNA identified an open reading frame of 1071 nucleotides putatively encoding a 357–amino acid protein, as shown in Figure 2. The open reading frame was divided by nine potential introns, confirmed by sequencing a full-length cDNA clone (data not shown). The predicted amino acid sequence showed 80.3% identity and 89.3% similarity to S. cerevisiae Hog1 and 83.9% identity and 93.4% similarity to the Sty1 MAPK from Schizosaccharomyces pombe (Figure 2). The gene was named OSM1.

View larger version (105K): [in this window] [in a new window]   Figure 2. Predicted Amino Acid Sequence of the M. grisea OSM1 Gene. Amino acid sequence alignment of the predicted M. grisea OSM1 MAPK gene product with S. pombe Sty1, S. cerevisiae Hog1, and Candida albicans CaHog1. Sequences were aligned with the CLUSTALW program (Thompson et al. 1994 ). Identical amino acids are highlighted on a black background, conserved amino acids are shown on a dark gray background, and similar amino acids are shown on a light gray background. Dashes indicate gaps in the alignments. OSM1 contains the conserved TGY activation loop at positions 174 to 176 found in the stress-activated protein kinase subgroup of MAPKs (Robinson and Cobb 1997 ). OSM1 has GenBank accession number AF184980.

To test whether OSM1 could function in the yeast HOG pathway, we subcloned a full-length cDNA corresponding to OSM1 as a 1.17-kb XbaI fragment into the yeast expression vector pYES1 for expression under the S. cerevisiae GAL1 promoter in the hog1 mutant JBY10 (Brewster et al. 1993 ). Osmotic sensitivity of the resulting transformants was determined by streaking yeast cells onto galactose-containing agar medium supplemented with 0.4 M NaCl. Transformants carrying the pYES1:OSM1 construct complemented the osmotic sensitivity of hog1 (Figure 3A), although we found that the osmotolerance of these transformants under more severe stress was not as high as the (nonisogenic) wild-type S. cerevisiae strain tested (Figure 3B). Deletion of HOG1 in S. cerevisiae is known to result in morphological aberrations when cells are grown under osmotic stress (Brewster and Gustin 1994 ). Introduction of OSM1 complemented this phenotype, and cells shaped like the wild type were observed after exposure to 0.4 M NaCl (Figure 3C). We conclude that OSM1 is a homolog of HOG1 and encodes a functional MAPK.

View larger version (68K): [in this window] [in a new window]   Figure 3. The M. grisea OSM1 Gene Is a Functional Homolog of the S. cerevisiae Osmosensory MAPK-Encoding Gene HOG1. (A) Growth of S. cerevisiae strains on a galactose-supplemented minimal medium agar plate in the presence of 0.4 M NaCl. JBY10 is an osmotically sensitive hog1 strain. KPD400 is a transformant of JBY10 expressing the M. grisea OSM1 gene under the control of the S. cerevisiae GAL1 promoter. INVSc2 is a nonisogenic wild-type S. cerevisiae strain. (B) Growth of the same S. cerevisiae strains in galactose-supplemented minimal media in the presence of a range of NaCl concentrations. Growth was determined spectrophotometrically by recording the OD600nm. Square data points are INVSc2, circles are KPD400, and diamonds are JBY10. Expression of M. grisea OSM1 restored the ability of S. cerevisiae hog1 mutants to grow in conditions of hyperosmotic stress, although not to levels observed by a wild-type (nonisogenic) strain. Each data point represents the mean of three replicate samples. Error bars representing the standard deviation did not exceed the size of the symbols used to plot the data. (C) Restoration of wild-type morphology of the S. cerevisiae hog1 mutant JBY10 by expression of M. grisea OSM1. JBY10 cells (top) after exposure to 0.2 M NaCl were typically polarized with misshapen budding. Below, the wild-type morphology of KPD400 cells is shown under the same conditions. Bar = 10 µm.

Inactivation of OSM1 Leads to Osmotic Sensitivity
A osm1 deletion mutant was generated using a one-step gene replacement, as shown in Figure 4A to 4C. A gene replacement vector (pOSM1) was constructed to delete 1.1 kb of the OSM1 coding region and to replace it with the hygromycin B phosphotransferase (HPH) gene cassette (Carroll et al. 1994 ). The region deleted includes the threonine and tyrosine residues (Thr-174 and Tyr-176) within the activation loop of the conserved kinase domain that is known to mediate MAPK activity (Johnson et al. 1996 ; Robinson and Cobb 1997 ). Due to the likelihood that the gene replacement would lead to osmotic sensitivity, the M. grisea transformation protocol, which depends on regeneration of protoplasts on osmotically stabilized media, was altered. Protoplasts were regenerated in osmotically stabilized liquid medium for 6 hr before being plated directly to nonstabilized growth medium (see Methods). Transformants were selected on hygromycin B–containing medium. The presence of the osm1 deletion allele was verified by DNA gel blotting. Hybridization analysis with probe pBSSX8 showed that the isogenic wild-type strain Guy11 and one of the putative transformants, JH69, contained a 2.2-kb SacI fragment containing the 3' end of OSM1. Four of the transformants, JH73, JH33, JH28, and JH22, contained a larger hybridizing 4.1-kb Sac I fragment, which is indicative of a gene replacement (Figure 4D). Consistent with this, the same fragment hybridized with an HPH probe. One of the transformants selected showed an ~10-kb hybridizing fragment, indicating ectopic insertion of the gene replacement vector.

View larger version (38K): [in this window] [in a new window]   Figure 4. Targeted Replacement of the M. grisea OSM1 MAPK Gene. (A) Restriction map of the OSM1 locus. The arrow shows the orientation of the OSM1 open reading frame. (B) Gene replacement vector pOSM1. (C) Restriction map of the osm1 deletion allele. The arrow indicates the orientation of the HPH open reading frame. (D) DNA gel blot analysis of pOSM1 transformants. Genomic DNA was prepared from the wild-type strain Guy11 (lane 1), osm1 deletion mutants JH73, JH33, JH28, and JH22 (lanes 2 and 4 to 6), an ectopic integration transformant JH8 (lane 7), and nontransformed strain JH69 (lane 3). Genomic DNA was digested with SacI and separated in a 1% agarose gel. The blot was probed first with pBSSX8, which contains the 3' end of OSM1, and then with the HPH gene cassette (Carroll et al. 1994 ). The Xs between (A) and (B) indicate a crossover event. B, BamHI; C, ClaI; D, HindIII; E, EcoRI; R, EcoRV; S, SalI; Sc, SacI; X, XhoI.

To ensure that any phenotypes attributed to the osm1^- mutation were due to the gene replacement event, we crossed JH73 to an M. grisea strain of opposite mating type, TH3. The resulting random ascospore progeny were subjected to DNA gel blot analysis to determine whether OSM1 was present. All hygromycin-resistant progeny were found also to carry the osm1 null mutation (data not shown).

Inactivation of OSM1 did not cause any measurable defect in the growth and development of osm1 mutants in standard growth medium, although a reduction in conidiogenesis was consistently observed in osm1 mutants. Conidial numbers were ~10-fold lower in plate cultures of osm1 mutants (mean = 4.1 x 10⁵ mL^-1) than those of the isogenic strain Guy11 (3.8 x 10⁶ mL^-1). The osm1 deletion mutant showed dramatically reduced growth compared with the Guy11 strain after exposure to hyperosmotic stress, as shown in Figure 5A. The response to chronic hyperosmotic stress was determined based on growth in the presence of 0.4 M NaCl. The reduction in growth was found to be proportional to the external concentration of solute when analyzed by determination of dry weight (Figure 5B). Similar reductions in growth were observed when the osm1 mutant was subject to chronic hyperosmotic stress by exposure to concentrations of sorbitol or KCl generating identical osmotic potentials (data not shown). A pronounced morphological difference was also observed when the osm1 mutants were exposed to hyperosmotic stress. Hyphae became wider and took on a pseudohyphal, budding appearance (Figure 5C). Hyphae were also conspicuously less vacuolated than were those of the Guy11 strain under these conditions (Figure 5C).

View larger version (77K): [in this window] [in a new window]   Figure 5. Osmotic Sensitivity and Aberrant Hyphal Morphology of M. grisea osm1 Deletion Mutants. (A) M. grisea strain Guy11 and isogenic osm1 deletion mutant JH73 on CM agar and CM agar supplemented with 0.4 M NaCl. Plates were inoculated with a 5-mm2 plug of mycelium and allowed to grow for 12 days at 24°C. Strains on each plate are given in the panel on the left. (B) Growth of the wild-type M. grisea strain Guy11 and isogenic osm1 deletion mutant JH73 in CM broth supplemented with varying concentrations of NaCl. Liquid cultures were prepared in 20 mL of CM with a uniform inoculum of blended mycelia and allowed to grow for 4 days with aeration. The mycelia were removed by filtration, and dry weights were determined. The data points represent the means of three replicates. Square data points are measurements of the growth of JH73, and diamonds are that of Guy11. Error values represent the standard deviation. The experiment was repeated with very similar results. (C) Hyphal morphology of osm1 deletion mutant JH73 (top panel) and the iosgenic wild-type strain Guy11 after exposure to 0.4 M NaCl. Hyphae appeared bulbous and budded. The hyphae were stained with neutral red to reveal vacuoles and viewed under Hoffman modulation contrast. Guy11 showed increased vacuolation under hyperosmotic stress. This response was not observed in the osm1 deletion mutant JH73. Bar = 10 µm.

Accumulation of the Compatible Solute Arabitol Is Controlled by OSM1
To determine the effect of the osm1 mutation on the cellular response to hyperosmotic stress, we made mycelial extracts from JH73 cultures under isoosmotic and hyperosmotic conditions. Under normal conditions, the osm1 mutant accumulated trehalose normally, although amounts of the other major storage carbohydrate mannitol were depleted (Figure 6A). During hyperosmotic stress, arabitol did not accumulate to the high levels observed in the Guy11 strain (Figure 1, Figure 6A, and Figure 6B). In contrast, smaller amounts of glycerol accumulated in both osm1 and Guy11 mycelium in response to hyperosmotic stress. These results indicate that accumulation of the compatible solute arabitol is regulated by OSM1 and that the osm1 mutation may have further pleiotropic effects on polyol metabolism in M. grisea.

View larger version (19K): [in this window] [in a new window]   Figure 6. Compatible Solute Production by the osm1 Mutant JH73 during Hyperosmotic Stress. Mycelium was grown for 48 hr in CM broth before being transferred to CM or CM plus 0.4 M NaCl broth for a further 24 hr. Carbohydrates were extracted and quantified by gas–liquid chromatography. Error bars indicate standard deviation from three repeats of the experiment. (A) Major storage carbohydrates accumulating in M. grisea osm1 mycelium during growth in isoosmotic conditions (CM). 1, mannitol; 2, trehalose; 3, glycerol; 4, arabitol. (B) Major storage carbohydrates accumulating in M. grisea osm1 mycelium after acute hyperosmotic stress (CM broth plus 0.4 M NaCl). The osm1 mutation affects the ability to accumulate arabitol in response to hyperosmotic stress. Levels of mannitol stored during normal growth and development are also affected. 1, mannitol; 2, trehalose; 3, glycerol; 4, arabitol.

M. grisea osm1 Mutants Produce Functional Appressoria
The role of OSM1 in appressorium turgor generation was determined by estimation of internal turgor using cytorrhysis. The wild-type M. grisea strain Guy11 and the isogenic osm1 mutant JH73 both required >3 M glycerol to collapse 50% of the developed appressoria, suggesting a mean internal glycerol concentration of between 2 and 4 M, as shown in Figure 1C and Figure 7A. To confirm this result, we made appressorium extracts and assayed glycerol quantitatively by using an enzymatic assay. Glycerol was found in nearly identical quantities in both Guy11 and JH73 (mean = 3.13 ± 1.1 M after 48 hr). Appressorium-mediated penetration was determined by using an onion epidermis penetration assay (Chida and Sisler 1987 ) and by checking the pathogenicity of M. grisea strains on a susceptible rice cultivar, CO39. Appressorium-mediated penetration and the ability to cause rice blast disease symptoms were unaffected in the osm1 gene replacement mutant, as shown in Figure 7B. We conclude that OSM1 is not required for appressorium turgor generation by M. grisea.

View larger version (18K): [in this window] [in a new window]   Figure 7. M. grisea osm1 Deletion Mutants Produce Appressoria with High Cellular Turgor. (A) Appressorium turgor was determined using a cytorrhysis assay (Howard et al. 1991 ; de Jong et al. 1997 ). Appressoria were allowed to form on hydrophobic plastic cover slips for 50 hr and then incubated in a range of glycerol concentrations (conc.). The number of appressoria from the osm1 mutant JH73 collapsing after this treatment was recorded. In this experiment, 3 M glycerol caused ~35% of JH73 appressoria to collapse, suggesting a mean internal glycerol concentration of 2 to 4 M. Turgor was not significantly different from the wild-type M. grisea strain Guy11. Error bars indicate standard deviation from three repeats of the experiment. (B) Rice blast symptoms caused by inoculation of the susceptible rice cultivar CO39 with Guy11 and the osm1 mutant JH73. Lesion density values were not significantly different.

The fact that deletion of OSM1 did not affect appressorium function suggested that the M. grisea HOG pathway does not regulate turgor generation within infection cells. One possibility, however, is that the pathway does not operate within differentiated appressoria but only in hyphae exposed to hyperosmotic stress. Therefore, we decided to investigate the response of appressoria to hyperosmotic stress to determine whether the OSM1-mediated hyperosmotic signal transduction pathway is capable of functioning at the same time as appressorium turgor generation. Appressoria were allowed to form in water drops on the hydrophobic side of Gelbond for 24 hr and then subjected to acute osmotic stress by addition of 0.4 M NaCl. The appressorial contents were then extracted, and internal polyol concentrations were analyzed by gas–liquid chromatography. In unstressed wild-type appressoria, glycerol accumulated as expected (de Jong et al. 1997 ), but in appressoria exposed to acute stress with 0.4 M NaCl, trace amounts of arabitol were also present (Figure 8B). In contrast, the mutant did not accumulate arabitol in its appressoria after acute osmotic stress (Figure 8A). We conclude that the OSM1-mediated MAPK signaling pathway acts independently of the signal transduction pathway leading to appressorium turgor.

View larger version (8K): [in this window] [in a new window]   Figure 8. Compatible Solute Production in M. grisea Appressoria after Exposure to Acute Hyperosmotic Stress. Appressoria were allowed to form on the hydrophobic surface of Gelbond for 24 hr and then subjected to acute hyperosmotic stress by the addition of 0.4 M NaCl. Appressorial extracts were prepared and analyzed by gas–liquid chromatography. An enlarged section of the gas–liquid chromatography trace from this experiment is shown with retention times in minutes. Arabitol had a retention time of 10.04 min in this experiment (arrow). The large peak shown at 9.4 min represents the internal standard, ribose. (A) Gas–liquid chromatography trace from appressorial extract of the osm1 mutant JH73 exposed to hyperosmotic stress. (B) Gas–liquid chromatography trace from appressorial extract of Guy11 exposed to hyperosmotic stress.

Inactivation of OSM1 Has Pleiotropic Effects on Appressorium Morphogenesis
Although OSM1 is not required for appressorium function, deletion of OSM1 does have pronounced effects on appressorium development under conditions of chronic osmotic stress (Figure 9A to 9C). Conidia were germinated in the presence of 0.2 or 0.4 M NaCl and allowed to develop appressoria on the hydrophobic surface of Gelbond. Appressoria formed normally in the Guy11 strain. However, in osm1 mutants, multiple appressoria were produced from the ends of germ tubes. This response was proportional to the external osmotic potential, and as many as 24 ± 11% of the conidia produced compound appressoria after developing in the presence of 0.4 M NaCl. The presence of enhanced appressorium development during hyperosmotic stress in the absence of OSM1 may indicate that one function of the MAPK encoded by OSM1 is to negatively regulate appressorium development during severe osmotic stress. This suggests that the OSM1-mediated hyperosmotic signaling pathway may communicate with the known signal transduction cascade(s) required for appressorium morphogenesis in M. grisea.

View larger version (39K): [in this window] [in a new window]   Figure 9. Development of Multiple Appressoria by the M. grisea osm1 Mutant JH73 during Chronic Hyperosmotic Stress. M. grisea conidia of Guy11 and the osm1 mutant JH73 were allowed to germinate on hydrophobic Gelbond membranes in the presence of increasing concentrations of NaCl. (A) The number of conidia elaborating more than one appressorium was recorded. Each data point represents the mean percentage from observing 200 conidia. Error values are the standard deviation. (B) Normal infection structure in M. grisea. A conidium of Guy11 has elaborated a single appressorium from a short germ tube. (C) Multiple appressorium formation by the osm1 mutant JH73 in the presence of 0.4 M NaCl. Bar = 10 µm.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION METHODS REFERENCES

This study sought to characterize the control of cellular turgor in M. grisea. We found that two distinct polyols form the major compatible solutes during turgor generation either in response to hyperosmotic stress or during plant infection. The enormous turgor reported in appressoria of M. grisea apparently requires the highly soluble polyol glycerol, which can accumulate to the very high concentrations required without precipitation or toxicity (de Jong et al. 1997 ). In contrast, the five-carbon polyol arabitol accumulates in hyphae of M. grisea in addition to glycerol when they are exposed to chronic or acute hyperosmotic stress. Arabitol biosynthesis has not been studied extensively, although an NAD-dependent arabitol dehydrogenase–encoding gene has been isolated from the yeast Candida albicans (Wong et al. 1993 ), suggesting one possible route for its accumulation in M. grisea. Hyperosmotic stress tolerance is likely to be an important process during the life cycle of M. grisea, and high levels of both glycerol and arabitol have previously been reported in blast-infected rice leaves (Hwang et al. 1989 ).

To determine whether celluar turgor is controlled in M. grisea via the conserved eukaryotic signaling pathway, we isolated and characterized the OSM1 gene, which is functionally related to S. cerevisiae HOG1. Recent evidence has shown that fungal pathogens have adapted several conserved signaling pathways for regulating pathogenicity-related functions such as appressorium development or plant tissue colonization, including MAPK elements of the pheromone response pathway of S. cerevisiae, components of the Ras-adenylate cyclase pathway, and heterotrimeric G proteins (Gold et al. 1994 ; Mitchell and Dean 1995 ; Banuett and Herskowitz 1996 ; Gao and Nuss 1996 ; Xu and Hamer 1996 ; Alspaugh et al. 1997 ; Beckerman et al. 1997 ; Choi and Dean 1997 ; Liu and Dean 1997 ; Xu et al. 1997 ). Identification of OSM1, and its subsequent targeted deletion, revealed that M. grisea possesses a MAPK-mediated hyperosmotic stress pathway.

The OSM1 MAPK shows very high similarity to the stress-activated MAPKs Hog1 from S. cerevisiae and Sty1 from S. pombe. The similarity extends throughout the catalytic domains of the proteins, and the conservation of an activation loop within the kinase domain (the TGY motif at residues 174 to 176) indicates that OSM1 is activated via threonine and tyrosine phosphorylation by a MAPKK homologous to Pbs2 from S. cerevisiae or Wis1 from S. pombe (Herskowitz 1995 ; Robinson and Cobb 1997 ). Consistent with the high degree of conservation, OSM1 was able to complement the osmotic sensitivity and morphological aberrations of a hog1 mutant of S. cerevisiae. The morphological defects of S. cerevisiae hog1 mutants during osmotic stress are due to the role of Hog1 in regulating reorganization of the cytoskeleton during exposure to high solute concentrations (Brewster and Gustin 1994 ). The reduced polarization and budding morphology of osm1 null mutants after exposure to hyperosmotic stress indicate that OSM1 probably performs this function in M. grisea hyphae as well. Increased vacuolation of M. grisea hyphae during hyperosmotic stress was also observed and found to be OSM1 dependent. Formation of vacuoles is likely to be a consequence of compatible solute generation and adjustment of cellular turgor.

The HOG pathway in budding yeast appears to act specifically to regulate the cellular response to hyperosmotic stress, whereas in fission yeast, the same pathway, which is mediated by the Sty1 MAPK, controls the response to a wider selection of environmental stresses, including oxidative stress, ultraviolet light exposure, and heat shock. To determine whether the OSM1 pathway in M. grisea resembles either of these distinct situations, we exposed conidia from M. grisea osm1 mutants to UV light, oxidative stress (exposure to 10 mM H₂O₂), or heat shock. Conidial viability of osm1 mutants was reduced compared with the isogenic wild type after UV exposure but was not affected by the other stresses (data not shown). This indicates that the OSM1 signaling cascade may respond to stresses other than osmotic shock, resembling, at least in part, the Sty1/Wis1 pathway in fission yeast (Millar et al. 1995 ; Gustin et al. 1998 ).

Regulation of Appressorium-Mediated Infection in M. grisea
We tested whether OSM1 was required for appressorium function and found (to our surprise) that M. grisea appressoria can operate effectively in the absence of this MAPK. This indicates that the massive increase in cellular turgor that occurs in M. grisea appressoria during plant infection is not controlled using the conserved mechanism for modulation of cellular turgor. We were aware, however, that hyperosmotic stress experiments were all done by analyzing the response of fungal hyphae; therefore, there was a possibility that the OSM1 pathway cannot operate effectively in appressoria. We therefore exposed preformed appressoria to hyperosmotic stress and found that arabitol began to accumulate in an OSM1-dependent manner. This shows that the OSM1 pathway is still able to operate within appressoria and must therefore be independent of the pathway mediating turgor generation. Although we cannot preclude the possibility that there are shared components in both signaling cascades, the fact that osmotically stressed appressoria generate arabitol and glycerol points to both pathways being fully active at the same time.

OSM1 is the third MAPK-encoding gene identified in M. grisea, and all three MAPKs play a role in infection-related development. PMK1 is a functional homolog of the S. cerevisiae MAPK-encoding gene FUS3 from the pheromone response pathway (Xu and Hamer 1996 ). M. grisea PMK1 regulates the formation of appressoria, probably acting downstream of a cAMP-dependent pathway involving a heterotrimeric G-protein encoded by MAGB (Lee and Dean 1993 ; Liu and Dean 1997 ) and adenylate cyclase encoded by MAC1 (Choi and Dean 1997 ; Adachi and Hamer 1998 ). pmk1 mutants are unable to elaborate appressoria and are also blocked in all subsequent stages of pathogenic development because mutants fail to colonize plant tissue even when conidia are injected directly into leaves (Xu and Hamer 1996 ). The MPS1 MAPK is required for appressorium function in M. grisea. Here its function appears to be in regulating cell wall remodeling and polarity establishment for penetration peg emergence (Xu et al. 1998 ). MPS1 is a functional homolog of the MAPK-encoding gene SLT2, which is required for cell wall integrity under low osmotic conditions in budding yeast. mps1 mutants fail to penetrate plant cuticles, although they still generate turgor pressure and activate plant defense responses (Xu et al. 1998 ).

OSM1 does not regulate appressorium turgor, but it does appear to play a role in appressorium morphogenesis. Chronic osmotic stress induced a multiple appressorium phenotype in osm1 mutants, which indicates that one function for OSM1 might be to prevent cross-talk between the hyperosmotic stress pathway and the PMK1 appressorium morphogenesis pathway. In S. cerevisiae, for example, Hog1p prevents miscommunication between the HOG pathway and pheromone response pathway to ensure that mating does not occur during times of hyperosmotic stress (O'Rourke and Herskowitz 1998 ). Mutations in the HOG1 gene therefore allow osmolarity-induced activation of the pheromone response pathway in S. cerevisiae (Hall et al. 1996 ). In M. grisea, the PMK1 pathway, which controls appressorium development, is functionally related to the pheromone response pathway from S. cerevisiae. It may be worth speculating, therefore, that OSM1 normally acts to prevent inappropriate activation of this pathway during hyperosmotic stress (appressorium development is severely impeded in the presence of >0.5 M NaCl). In the absence of OSM1, the PMK1 pathway may be overstimulated in the presence of osmotic stress, thereby resulting in multiple rounds of appressorium formation. A summary of the MAPK signaling pathways of M. grisea is given in Figure 10.

View larger version (19K): [in this window] [in a new window]   Figure 10. Model Showing the MAPK Signaling Pathways in the Rice Blast Fungus M. grisea. Three genes encoding MAPKs have been identified and characterized in M. grisea. PMK1 regulates appressorium development and subsequent proliferation of the fungus in plant tissue. PMK1 may operate downstream of (or parallel to) the cAMP signaling pathway, which regulates appressorium formation (Xu and Hamer 1996 ). MPS1 regulates appressorium-mediated infection, potentially by controlling formation of the penetration peg. By analogy to S. cerevisiae, MPS1 may act downstream of a protein kinase C (PKC) pathway (Xu et al. 1998 ). OSM1 regulates hyperosmotic stress adaptation and clearly acts in response to that stress. OSM1 may also (?) prevent the PMK1 signaling pathway from operating during stressful environmental conditions.

Identification of OSM1 has shown that the action of M. grisea appressoria is likely to be regulated in a novel manner, distinct from the mechanism that underlies turgor generation in most eukaryotic cells. It is clear from many recent studies that fungal phytopathogens have coopted and adapted several regulatory pathways that condition mating, nutrition, or the response to environmental stress in budding yeast and used them to regulate pathogenicity functions, including infection cell development and plant tissue invasion (Banuett 1998 ). This study shows that in addition to these regulatory pathways, novel signaling networks may well have evolved specifically to regulate plant infection.

	METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION METHODS REFERENCES

Fungal Isolates
Strains of Magnaporthe grisea from this study are stored in the laboratory of N.J. Talbot. Strain TH3 was kindly donated by Dr. J.L. Notteghem (Centre de Coopération Internationale en Recherche Agronomique pour le Développement, Montpellier, France). Standard growth procedures were used for M. grisea (Crawford et al. 1986 ), with oatmeal and complete medium (Talbot et al. 1993 ). Long-term storage of M. grisea was performed by growing the fungus through sterile filter paper discs, desiccating these for 48 hr, and then storing them under desiccation at -20°C.

Genetic Crosses
Genetic crosses were performed as described by Valent et al. 1991 . Briefly, strains of opposite mating type were paired on oatmeal agar and grown until the mycelia joined. Approximately 14 days after inoculation, flask-shaped perithecia were visible at the mycelial margins. These were transferred to a 4% distilled water agar plate with the use of fine forceps. The perithecia were rubbed gently to remove adhering mycelium and conidia before being broken open to reveal asci. Mature asci were removed with a sterile fine needle, and ascospores were dissected from them. These ascospores were transferred individually to a 24-cell well plate (Sarstedt, Nümbrecht, Germany) containing complete medium and incubated for 4 to 5 days (Talbot et al. 1993 ). At this time, monoconidial isolations were made from each well and grown in individual plates for storage.

Rice Infections
Fourteen-day-old rice seedlings were infected with suspensions of M. grisea conidia prepared in 0.2% gelatin at a concentration of 10⁴ conidia mL^-1. Two-week-old seedlings of the susceptible rice cultivar CO39 were sprayed using an artist's airbrush (Badger Co., Franklin Park, IL). Plants were incubated in plastic bags for 48 hr to maintain high humidity and then transferred to controlled environment chambers at 24°C, 84% relative humidity, with 900 µE m^-2 sec^-1 tungsten illumination and a 14-hr day length. Plants were incubated for 96 to 144 hr for full disease symptoms to become apparent. The first disease symptoms were observed 96 hr after seedling inoculation. Lesion densities were routinely scored from 40 randomly chosen 5-cm-long leaf tips, and means and standard deviations were determined.

Assays for Infection-Related Morphogenesis
Appressorium development by M. grisea was observed on plastic cover slips (Fisher Scientific, Loughborough, UK) or the hydrophobic side of Gelbond membranes (FMC, Chicago, IL), as described by Hamer et al. 1988 . A 100-µL drop of a conidial suspension at a concentration of 10⁵ mL^-1 was placed on the surface of a cover slip and left in a humid environment at 24°C. The frequency and morphology of appressorium formation were determined by counting the number of appressoria that had developed from 300 conidia after 14 hr (Talbot et al. 1993 ). Hyphal staining of M. grisea to determine the extent of vacuolation was performed by staining with Neutral Red solution containing 0.15% (w/v) neutral red (BDH, Poole, Dorset, UK) and 3.75% (w/v) PVP (Sigma) in 50 mM Tris-maleate buffer, pH 7.5. Hyphae and appressoria were routinely viewed with an Optiphot-2 microscope (Nikon, Kingston, UK) under Hoffman modulation contrast.

DNA Isolation and Analysis
Genomic DNA was extracted from fungal mycelium by using a CTAB (hexadecyltrimethylammonium bromide) procedure described by Talbot et al. 1993 . Gel electrophoresis, restriction enzyme digestion, and DNA gel blot hybridizations were all performed using standard procedures (Sambrook et al. 1989 ). DNA hybridization probes were labeled by the random primer method (Feinberg and Vogelstein 1983 ), using the Stratagene Prime-It kit, and high-stringency washes were performed as previously described (Talbot et al. 1993 ). DNA sequencing was performed using an ABI 377 automated sequencer (Perkin-Elmer) and dye terminator cycle sequencing Ready Reaction kit (Amersham Pharmacia Biotechnology, Amersham, UK) with Ampli Taq DNA Polymerase FS (Amersham Pharmacia Biotechnology), according to the manufacturer's instructions. DNA/protein sequence databases were searched using the BLAST algorithm (Altschul et al. 1990 ) at the National Center for Biotechnology Information (Bethesda, MD), which can be accessed via the World Wide Web.

Identification of the OSM1 Mitogen-Activated Protein Kinase–Encoding Gene
OSM1 was cloned using polymerase chain reaction (PCR) with degenerate primers designed to conserved regions of mitogen-activated protein kinases (MAPKs). The gene was cloned by two independently working coauthors (K.P.D. and J.-R.X.). OSM1 was obtained as a 400-bp sequence amplified by PCR with the degenerate primers KPD1 (5'-CTTGGATCCMGNGCNCCNGARATHATGYTN-3') and KPD2 (5'-CTTGGATCCRTANGGRTGNGCNARNGC-3'), where R is purine; Y is pyrimidine; M is alanine or cytosine; H is thymine, cytosine, or alanine; and N is any nucleotide. Optimal conditions for the PCR reactions were as follows: 1 min at 94°C, 1 min at 45°C, and 1 min at 72°C for 35 cycles with a 10-min extension at 72°C. All PCR reactions were performed in a Perkin-Elmer GeneAmp PCR System 2400 machine using 1 unit of Tth DNA polymerase (Promega) per reaction. The resulting PCR amplicons were gel purified with standard methods using a Qia-ex kit (Qiagen, Chatsworth, CA), according to the manufacturer's instructions, and cloned into Promega pGEM-T, and the plasmids were digested with BamHI to liberate the recombinant insert. OSM1 was also isolated using primers MAK2, MAK4, and MEK3, described by Xu and Hamer 1996 and used previously to identify PMK1 and MPS1 from M. grisea (Xu and Hamer 1996 ; Xu et al. 1998 ). PCR amplicons showing similarity to S. cerevisiae HOG1 were used to screen a ZAPII (Stratagene) cDNA library from nitrogen-starved Guy11 mycelium (Xu and Hamer 1996 ) and a Guy11 GEM-11 genomic library (Talbot et al. 1993 ).

Complementation of Saccharomyces cerevisiae hog1
An S. cerevisiae hog1 strain (JBY10:MATa ura3 leu2 his3 ade2 lys2 try1 hog1-::TRP1) was kindly donated by M. Gustin (Rice University, Houston, TX). Plasmid pGAL1(p):OSM1, placing OSM1 under control of the yeast GAL1 promoter, was constructed by amplifying a 1168-bp fragment from a full-length cDNA clone (pH g110-3) containing the entire OSM1 open reading frame, using the primers KD5HOG (5'-GCTCTAGATGAAACCTGCTTCCCGACGGT-3') and KD3HOG (5'-GGTCTAGAGAATGAAATCATGAACGGAAA-3'), which also added XbaI sites to each end of the fragment. The amplicon was digested with XbaI and cloned into pYES2 (Invitrogen, Carlsbad, CA). Yeast cells of strain JBY10 (hog1) were transformed using the lithium acetate method described by Gietz and Schiestl 1995 . The resulting (Ura⁺) transformants were confirmed by colony PCR and dot blot hybridization with pH g110-3 (data not shown). S. cerevisiae Ura⁺ transformants were tested for osmosensitivity by growth on galactose-supplemented minimal media without uracil in the presence of 0.2 M NaCl (Brewster et al. 1993 ). For analysis of S. cerevisiae growth in liquid medium, precultures of the wild-type strain INVSc2, hog1 strain JBY10, and GAL1p:OSM1 strain KPD400 were prepared by overnight incubation at 30°C in yeast extract–peptone–dextrose medium (Brewster et al. 1993 ). The titer of cells was equalized, and 100 µL of preculture was used to inoculate 10-mL shaking cultures in supplemented minimal media in the presence of galactose and a range of NaCl concentrations from 0.1 to 0.4 M. Cultures were incubated overnight at 30°C while being shaken at 220 rpm. The OD_600nm of the resulting cell suspensions was determined in a Pharmacia Ultrospec 2000 spectrophotometer.

Construction of the Gene Replacement Vector pOSM1
Plasmid pOSM1 was constructed by first cloning a 1.9-kb EcoRI fragment (5' flanking region of OSM1) into pBCKS (Stratagene) to create pBHE12. This clone was digested with SacI and religated to create pBHS12, which carries a 1-kb SacI-EcoRI fragment (5' flank of OSM1). A 1.8-kb XhoI fragment (the 3' end of OSM1 and downstream flanking sequence) was cloned into pBHS12, and the orientation of the insert was checked by SalI digestion. The correct clone was named pNX2. The hygromycin phosphotransferase gene cassette (HPH) (Carroll et al. 1994 ) was inserted into the unique EcoRI site to create pOSM1.

Fungal Transformation
Protoplast preparation and transformation were performed as described previously (Talbot et al. 1993 ). The transformation protocol for generation of the osm1 gene replacement was adapted by incubating polyethylene glycol–treated transformed Guy11 protoplasts for 28 hr in liquid TB3 medium (3 g L^-1 yeast extract, 3 g L^-1 casamino acids, and 10 g L^-1 glucose) osmotically stabilized with 0.8 M sucrose. The regenerating protoplasts were collected by centrifugation at 1000g for 6 min and resuspended in unmodified TB3. The protoplasts were then immediately plated onto nonosmotically stabilized complete medium (CM; Talbot et al. 1993 ) agar containing 150 µg mL^-1 hygromycin B (Calbiochem), and an overlay of CM agar containing 250 µg mL^-1 hygromycin B was added after 18 hr. Transformants were selected at 24°C after 6 to 10 days.

Identification of Compatible Solutes in M. grisea
M. grisea liquid cultures were prepared, and resulting mycelia were harvested via filtration through sterile Miracloth (Calbiochem). The mycelium was washed in isoosmotic concentrations of CaCl₂, frozen in liquid nitrogen, and freeze dried under vacuum for 3 to 5 days. The dry weight of these samples was then recorded, and carbohydrates were extracted using methylene dichloride. Briefly, lyophilized samples were ground using a pestle with a mortar containing 4 mL of MMW (methanol–methylene dichloride–double-distilled H₂O [v/v] 12:5:3). A further 6 mL of MMW was added gradually to the ground mycelium, and the suspension was decanted into 50-mL sterile Oakridge tubes (Nalgene, Rochester, NY). The suspension was incubated at 40°C for 30 min, with vortexing every 10 min. Samples were then processed by centrifugation for 30 min at 3000g at 4°C. The supernatant was decanted into sterile Oakridge tubes and stored on ice. The pellet was reextracted with 5 mL MMW and vortexed for 1 min followed by centrifugation at 3000g. The resulting supernatants were pooled, and 7.5 mL of sterile double-distilled H₂O and 7.5 mL of methylene dichloride were added. This solution was vortexed for ~1 min, and the resulting emulsion was centrifuged at 3000g for 30 min at 4°C. The aqueous phase was then carefully decanted into a sterile universal and frozen at -80°C overnight. Extracts were then lyophilized for 3 to 5 days. The lyophilized material was resuspended in 0.5 to 1 mL of sterile double-distilled H₂O and stored at -20°C.

For gas–liquid chromatography, 500-µL aliquots of extracted carbohydrate sample were dried in a gentle stream of air in Reacti-vials (Pierce, Rockford, IL). In each case, a 50-µL aliquot of the internal standard (100 mM ribose) was first added. Samples were oximated with 500 µL of 0.3 M O-methyl hydroxyl amine (Sigma) in pyridine (HPLC grade; Sigma) by incubation at 70°C for 30 min and then silylated by the addition of 100 µL of N-trimethylsilylimidazole (Sigma) and incubation at room temperature for 60 min. The samples were extracted using 500 µL of hexane (Sigma) and analyzed using a capillary gas–liquid chromatograph (model GC14-A; Shimadzu, Columbia, MD) with a BP5 25-m column with an internal diameter of 0.25 mm. Quantitative estimates of carbohydrate concentration were made by integrating peaks, normalizing values to the internal standard, and comparing them with peaks from replicate derivatized (100 mM) standards of fructose, glucose, mannitol, arabitol, glycerol, and trehalose.

Appressorial glycerol was extracted by scraping cells from the surface of plastic cover slips (Fisher Scientific, Loughborough, UK) in 80% ethanol. The suspension was then sonicated on ice for 1 min. This was repeated five times. The appressorial debris was then removed by centrifugation at 3000g, and the supernatant was evaporated to dryness in a rotary evaporator (Jencons Scientific, Bridgeville, PA). The extract was resuspended in double-distilled H ₂O, and glycerol content was quantified using a spectrophotometric glycerol enzymatic assay (Roche Molecular Biochemicals, Lewes, UK).

	ACKNOWLEDGMENTS

We gratefully acknowledge Dr. Howard Stebbings (University of Exeter, UK) and Dr. John Hamer (Purdue University, West Lafayette, IN) for critically reading the manuscript. This work was supported by grants to N.J.T. from the Biotechnology and Biological Sciences Research Council (BBSRC; Grant No. 9/P08629) and the Royal Society. K.P.D. is the recipient of a BBSRC Ph.D. studentship. N.J.T. is a Nuffield Foundation Science Research Fellow.

	REFERENCES

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION METHODS REFERENCES

Adachi, K., and Hamer, J.E. (1998) Divergent cAMP signaling pathways regulate growth and pathogenesis in the rice blast fungus Magnaporthe grisea. Plant Cell 10:1361-1373[Abstract/Free Full Text].

Albertyn, J., Hohmann, S., Thevelein, J.M., and Prior, B.A. (1994) GPD1, which encodes glycerol-3-phosphate dehydrogenase, is essential for growth under osmotic stress in Saccharomyces cerevisiae and its expression is regulated by the high-osmolarity glycerol response pathway. Mol. Cell. Biol. 14:4135-4144[Abstract/Free Full Text].

Alspaugh, J.A., Perfect, J.R., and Heitman, J. (1997) Cryptococcus neoformans mating and virulence are regulated by the G-protein sub-unit GPA1 and cAMP. Genes Dev. 11:3206-3217[Abstract/Free Full Text].

Altschul, S.F., Gish, W., Miller, W., Myers, C.W., and Lipman, D.L. (1990) Basic local alignment search tool. J. Mol. Biol. 215:403-410[CrossRef][ISI][Medline].

Banuett, F. (1998) Signaling in the yeasts: An informational cascade with links to the filamentous fungi. Microbiol. Mol. Biol. Rev. 62:249-275[Abstract/Free Full Text].

Banuett, F., and Herskowitz, I. (1996) Discrete developmental stages during teliospore formation in the maize smut fungus Ustilago maydis. Development 122:2965-2976[Abstract].

Beckerman, J.L., Naider, F., and Ebbole, D.J. (1997) Inhibition of pathogenicity of the rice blast fungus by Saccharomyces cerevisiae -factor. Science 276:1116-1119[Abstract/Free Full Text].

Blomberg, A., and Adler, L. (1992) Physiology of osmotolerance in fungi. Adv. Microb. Physiol. 33:145-212[Medline].

Bourett, T.M., and Howard, R.J. (1990) In vitro development of penetration structures in the rice blast fungus Magnaporthe grisea. Can. J. Bot. 68:329-342.

Bourett, T.M., and Howard, R.J. (1992) Actin in penetration pegs of the rice blast fungus Magnaporthe grisea. Protoplasma 168:20-26[CrossRef][ISI].

Brewster, J.L., and Gustin, M.C. (1994) Positioning of cell growth and division after osmotic stress requires a MAP kinase pathway. Yeast 10:425-439[CrossRef][ISI][Medline].

Brewster, J.L., de Valoir, T., Dwyer, N.D., Winter, E., and Gustin, M.C. (1993) An osmosensing signal transduction pathway in yeast. Science 259:1760-1763[Abstract/Free Full Text].

Carroll, A.M., Sweigard, J.A., and Valent, B. (1994) Improved vectors for selecting resistance to hygromycin. Fungal Genet. Newsl. 41:22.

Chida, T., and Sisler, H.D. (1987) Restoration of appressorial penetration ability by melanin precursors in Pyricularia oryzae treated with antipenetrants and in melanin-deficient mutants. J. Pestic. Sci. 12:49-55.

Choi, W., and Dean, R.A. (1997) The adenylate cyclase gene MAC1 of Magnaporthe grisea control appressorium formation and other aspects of growth and development. Plant Cell 9:1973-1983[Abstract].

Chumley, F.G., and Valent, B. (1990) Genetic analysis of melanin deficient, nonpathogenic mutants of Magnaporthe grisea. Mol. Plant-Microbe Interact. 3:135-143.

Crawford, M.S., Chumley, F.G., Weaver, C.G., and Valent, B. (1986) Characterization of the heterokaryotic and vegetative diploid phases of Magnaporthe grisea. Genetics 114:1111-1129[Abstract/Free Full Text].

de Jong, J.C., McCormack, B.J., Smirnoff, N., and Talbot, N.J. (1997) Glycerol generates turgor in rice blast. Nature 389:244-245[CrossRef].

Feinberg, A.P., and Vogelstein, B. (1983) A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132:6-13[CrossRef][ISI][Medline].

Gao, S.J., and Nuss, D.L. (1996) Distinct roles for two G protein sub-units in fungal virulence, morphology, and reproduction revealed by targeted gene disruption. Proc. Natl. Acad. Sci. USA 93:14122-14127[Abstract/Free Full Text].

Gietz, R.D., and Schiestl, R.H. (1995) Transforming yeast with DNA. Methods Mol. Cell. Biol. 5:255-269.

Gold, S., Duncan, G., Barrett, K., and Kronstad, J. (1994) cAMP regulates morphogenesis in the fungal pathogen Ustilago maydis. Genes Dev. 8:2805-2816[Abstract/Free Full Text].

Gustin, M.C., Albertyn, J., Alexander, M., and Davenport, K. (1998) MAP kinase pathways in the yeast Saccharomyces cerevisiae. Microbiol. Mol. Biol. Rev. 62:1264-1300[Abstract/Free Full Text].

Hall, J.P., Cherkasova, V., Elion, E., Gustin, M.C., and Winter, E. (1996) The osmoregulatory pathway represses mating pathway activity in Saccharomyces cerevisiae—Isolation of a fus3 mutant that is insensitive to the repression mechanism. Mol. Cell Biol. 16:6715-6723[Abstract].

Hamer, J.E., Howard, R.J., Chumley, F.G., and Valent, B. (1988) A mechanism for surface attachment in spores of a plant pathogenic fungus. Science 239:288-290[Abstract/Free Full Text].

Han, J., Lee, J.D., Bibbs, L., and Ulevitch, R.J. (1994) A MAP kinase targeted by endotoxin and hyperosmolarity in mammalian cells. Science 265:303-311[Free Full Text].

Han, S.J., Choi, K.Y., Brey, P.T., and Lee, W.J. (1998) Molecular cloning and characterization of a Drosophila p38 mitogen-activated protein kinase. J. Biol. Chem. 273:369-374[Abstract/Free Full Text].

Herskowitz, I. (1995) MAP kinase pathways in yeast: For mating and more. Cell 80:187-197[CrossRef][ISI][Medline].

Hirayama, T., Maeda, T., Saito, H., and Shonozaki, K. (1995) Cloning and characterization of seven cDNAs for hyperosmolarity-responsiveness (HOR) genes of Saccharomyces cerevisiae. Mol. Gen. Genet. 249:127-138[CrossRef][Medline].

Howard, R.J., and Ferrari, M.A. (1989) Role of melanin in appressorium formation. Exp. Mycol. 13:403-418[CrossRef].

Howard, R.J., and Valent, B. (1996) Breaking and entering—Host penetration by the fungal rice blast pathogen Magnaporthe grisea. Annu. Rev. Microbiol. 50</