Introduction

Heterotrimeric () guanine-nucleotide binding proteins (G-proteins) are activated by the seven transmembrane spanning family of receptors (Malbon, 2005). Binding of agonists, such as chemo-attractants, neurotransmitters, hormones, odorants and taste ligands, to such receptors promotes exchange of GDP to GTP on G and leads to the dissociation of G from the G heterodimer (Dohlman and Thorner, 2001). Either G or G, or both are then free to activate downstream effectors (Dohlman and Thorner, 2001). Signaling persists until GTP is hydrolyzed owing to the intrinsic GTPase activity of G and consequently the GDP-G and G reassociate. Hence, duration of G-protein signaling is controlled by the guanine nucleotide state of G (Dohlman and Thorner, 2001). Members of the RGS protein family share a conserved domain of 120 amino acids and function as key negative regulators of G-protein signaling pathways (Dohlman et al, 1996; Koelle and Horvitz, 1996; Siderovski et al, 1996). RGS proteins function primarily as GTPase accelerating proteins (GAPs) and increase the hydrolysis rate of GTP bound to the G subunits (Siderovski and Willard, 2005). Structures of the G_i1–GDP–AlF₄^-–RGS4 and G_t–GDP–AlF₄^-–RGS9 complexes revealed that RGS proteins exert their GAP activity primarily by stabilizing the transition state for GTP hydrolysis (Tesmer et al, 1997a; Slep et al, 2001). Based on sequence similarity and biological properties, 16 distinct mammalian G subunits have been classified into four families: G_s, G_i, G_q and G_12/13 (Fukuhara et al, 2001). To date, RGS proteins have been definitively shown to regulate G_i, G_q and G_12/13, but it is controversial as to whether a bona fide GAP exists for a G_s subunit (Zheng et al, 2001; Castellone et al, 2005).

G-protein-mediated signaling is one of the most important mechanisms by which eukaryotic cells sense extracellular signals and integrate them into intrinsic signal transduction pathways. In fungi, G-proteins are involved in the regulation of a variety of cellular functions in vegetative growth and/or pathogenic development, such as conidiation, infection structure differentiation and pathogenicity (Bolker, 1998; Lengeler et al, 2000; Yu, 2006).

The filamentous ascomycete Magnaporthe grisea causes the devastating rice-blast disease (Ou, 1985). The appressorium, a unicellular infection structure employed by M. grisea to breach the plant surface, is elaborated by a conidium at the tip of a germ tube in response to specific environmental and physicochemical surface signals (Lee and Dean, 1993; Gilbert et al, 1996). Hydrophobic, but not hydrophilic, membranes promote appressoria formation in vitro (Lee and Dean, 1994) and hence surface hydrophobicity has been implicated as an important trigger for this process (Talbot et al, 1993, 1996; Beckerman and Ebbole, 1996). However, it is unclear whether hydrophobicity is the only signal necessary for appressorium differentiation or whether other surface characteristics are also important.

Several studies have implicated cyclic AMP as a critical mediator of appressorium development in Magnaporthe (Mitchell and Dean, 1995; Choi and Dean, 1997). Loss of adenylyl cyclase Mac1 or cAMP-dependent protein kinase A activity leads to a failure in appressorium development (Mitchell and Dean, 1995; Choi and Dean, 1997). A MAP kinase cascade has also been identified as an essential signaling pathway involved in appressorium formation during pathogenic growth (Xu and Hamer, 1996; Xu, 2000; Bruno et al, 2004).

M. grisea contains three distinct G subunits (MagA, MagB and MagC; Liu and Dean, 1997), two G (Mgb1 and Mgb2; Nishimura et al, 2003) and one G subunit (MG10193.4; Dean et al, 2005). Our sequence comparisons revealed that MagA, MagB and MagC belong to the G_s, G_i and fungal-specific G_II subfamilies, respectively. Previous characterization of G-deletion strains (Liu and Dean, 1997), the magB^G42R mutant (Fang and Dean, 2000) and the mgb1 mutant (Nishimura et al, 2003) has indicated functions for G-protein signaling in vegetative growth, sexual reproduction and pathogenicity in M. grisea. The mgb1 mutant fails to produce appressoria, whereas elevated Mgb1p levels promote precocious appressoria formation (Nishimura et al, 2003).

Here, we describe TMT1398, a novel Magnaporthe mutant that uncouples thigmo-morphogenesis (response to mechanical sensation) from hydrophobicity signaling and pathogenesis, and leads to appressoria formation on non-inductive surfaces. TMT1398 harbors a loss-of-function allele of a gene we designate as RGS1, as it encodes an RGS domain-containing protein. Based on the characterization of an rgs1 mutant, we report that surface hardness-dependent signaling is critical for appressorium development. Genetics and biochemical analyses support the notion that Rgs1 is a cross-subfamily regulator of all G subunits (MagA, MagB and MagC) and modulates conidiogenesis, surface signaling and pathogenesis in Magnaporthe.

Results

Rgs1 is important for proper appressorium differentiation

An insertion mutant, TMT1398, with pleiotropic defects in vegetative growth and appressorium development, was isolated in an Agrobacterium Transferred-DNA (T-DNA)-mediated forward screen for nonpathogenic mutants in M. grisea. Wild-type conidia could form appressoria only on inductive surface, whereas TMT1398 elaborated appressoria efficiently on the inductive as well as the non-inductive surfaces (Figure 1A). TMT1398 strain was found to contain a single-copy T-DNA insertion that disrupted a candidate gene with extensive sequence similarity to FLBA (Yu et al, 1996) and SST2 (Dohlman et al, 1996), both of which encode RGS-domain-containing proteins; hence, we designated the TMT1398 locus as RGS1 in M. grisea. Figure 1B schematically shows the T-DNA disruption of the RGS1 locus in TMT1398 and depicts the RGS1 annotation on contig 2.276 and 2.277. The full-length cDNA of RGS1 contains an open reading frame (ORF) encoding a 714-amino acid protein (GenBank accession number DQ335135). The SMART database (http://smart.embl-heidelberg.de) predicts two DEP (Disheveled, Egl-10 and Pleckstrin) domains at the N-terminus and an RGS domain at the C-terminus of the Rgs1 protein (Figure 1B). The primary sequence and predicted secondary structure of the RGS domain from Rgs1 compares favorably to canonical RGS domains using multiple sequence alignment tools (Supplementary Figure S1) suggesting that Rgs1 most likely is a functional GAP.

To ascertain that the defects in TMT1398 were a consequence of the loss of RGS1 function, we created an rgs1 strain by replacing the entire ORF of RGS1 with the hygromycin resistance cassette (HPH1; Figure 1B and C). The resultant rgs1 strain recapitulated the defects shown by TMT1398 and was able to form appressoria on non-inductive membrane surfaces (Figure 1A). Furthermore, genetic complementation using a 7.1-kb fragment carrying the entire RGS1 locus completely suppressed the appressorial defects of rgs1 (Figure 1A, complement) and TMT1398 (data not shown). The complemented rgs1 strain behaved like the wild type, elaborating appressoria only on inductive, but not non-inductive, surfaces (Figure 1A). Based on comparative immunoblotting of Rgs1p levels (Figure 1D and E), the TMT1398 was considered a loss-of-function allele of RGS1, as Rgs1 protein was undetectable in the TMT1398 and rgs1 mutant, but was clearly detected in the wild-type and the complemented strain (Figure 1D). We conclude that the remarkable ability of the TMT1398 and rgs1 strains to form appressoria on non-inductive surfaces was solely due to the loss of RGS1, thus implicating Rgs1 as an important negative regulator of appressorium development and likely an inhibitor of precocious appressorium formation on undesirable surfaces.

Surface hardness stimulus is essential for appressorium differentiation in Magnaporthe

Surface hydrophobicity is considered to be an inductive stimulus for appressorium formation in Magnaporthe (Lee and Dean, 1994; Talbot et al, 1996). To identify additional signals involved in the regulation of appressorium differentiation, we tested rgs1 and wild-type strains on artificial surfaces with different physicochemical properties. Neither wax nor petroleum jelly (considered hydrophobic and soft), nor freshly polymerized agar blocks (deemed hydrophilic and soft) could trigger appressorium differentiation in the rgs1 or wild-type strains (Figure 2A). Germ tube emergence and growth remained unperturbed on both these surfaces. Wild-type M. grisea behaved similarly on the non-inductive hydrophilic surface and the soft hydrophobic surface and failed to produce appressoria (Figure 2A), suggesting that surface hydrophobicity alone is not sufficient to induce appressorial development. We concluded that the signaling for appressorium differentiation was not active equivalently on all the surfaces in the rgs1 strain.

Compared with membrane surfaces used earlier, both petroleum jelly and fresh agar surfaces are softer. We therefore hypothesized that contact with a hard surface might be required for appressorium differentiation. Additional appressorial assays were thus performed on agar surfaces hardened by drying (Figure 2B). Controlled evaporation of water from the agar block did not alter the hydrophilic nature of the surface, as judged by measuring the contact angles made by water droplets on the fresh agar and dried agar surfaces (data not shown; see Supplementary Figure S2). In addition, nano-indentation analyses (Supplementary Figure S2 for methodology) revealed that increasing the drying time increased the hardness of the resultant agar surfaces. Rather strikingly, the ability to induce appressorium development was directly related to surface hardness (Figure 2B). As shown in Figure 2C, appressoria were rarely elaborated by the wild-type or rgs1 conidia germinated on fresh agar block, whereas the agar blocks dried for 12 and 24 h could induce appressorium differentiation in 213.6 and 622.2%, respectively, of the rgs1 conidia. Under these conditions, appressoria were induced in 18.24.3 and 55.71.1%, respectively, of the wild-type conidia (Figure 2C). Nano-indentation experiments revealed that the average hardness (mean s.d.; 90 indents over three replicates) of the fresh agar block was 734 kPa, whereas after drying for 24 h, the resultant hardness of the dried agar block averaged 15111 kPa. The mean hardness value of a 2-week-old barley leaf was found to be 19421 kPa. Thus, RGS1 likely uncouples hardness signal sensing for appressorium development from other inductive stimuli, and that the rgs1 mutant bypasses the requirement of surface hydrophobicity but not that of surface hardness during appressorium differentiation. Taken together, we conclude that a thigmotropic cue acts as an important trigger for infection-related morphogenesis in Magnaporthe.

Timing of the thigmotropic signal sensing

Magnaporthe conidia begin to germinate immediately upon hydration and elaborate short germ tubes within 1–3 h. At about 6 h post inoculation, the tip of the germ tube undergoes swelling and hooks back to initiate appressorium differentiation. To better understand the relationship between surface hardness sensing and appressorium development, we studied the timing of hardness perception. An initial contact for 2 h with a hard surface (GelBond membrane) was sufficient to induce appressorium formation in the rgs1 (Figure 3B) and the wild-type strain (data not shown). In contrast, a hanging-drop setup that prevented rgs1 conidia from contacting the hard surface led to a total failure in appressorium formation (Figure 3C). We conclude that contact with a hard surface is necessary for efficient appressorium differentiation, and further suggest that such thigmo-morphogenetic signal is sensed within 2 h of conidial germination.

Rgs1 regulates MagA, the G_s subunit, during surface signaling

RGS proteins serve as negative regulators of G-protein signaling by virtue of their ability to accelerate the intrinsic GTPase activity of target G subunits. It is thus probable that Rgs1 also functions as a GAP for G subunit(s) in M. grisea. Therefore, we created individual G-deletion mutants (magA, magB and magC) in Magnaporthe and analyzed them for defects in appressorium differentiation. A G302S mutation that renders the G subunit RGS-insensitive has been characterized in yeast (DiBello et al, 1998) and also in mammalian G_o, G_i and G_q (DiBello et al, 1998; Lan et al, 1998). We introduced the corresponding RGS-insensitive mutation (Figure 4A) individually in the three G subunits, resulting in magA^G187S, magB^G183S and magC^G184S strains. Like the wild-type, the magA strain formed appressoria only on inductive surface (Figure 4B), whereas the magA^G187S mutant elaborated appressoria efficiently on both the inductive and the non-inductive surfaces, a phenotype reminiscent of rgs1.

Next, we created and tested a magA^Q208L mutant for appressoria formation. The magA^Q208L corresponds to the Q204L mutation that abolishes the GTPase activity and RGS interaction of the G_s subunit (Graziano and Gilman, 1989; Berman et al, 1996a) and leads to the constitutive activation of G_s owing to higher GTP occupancy. The magA^Q208L mutant was highly active and elaborated precocious appressoria on inductive and non-inductive surfaces alike (Figure 4B). Compared with the magA^G187S mutant, the magA^Q208L showed two major differences on the inductive surface: the germ tubes were extremely short and were barely visible in these magA^Q208L assays, and secondly, a significant proportion of conidia (23.72.2%; P=0.005) elaborated multiple appressoria from their terminal cells (Figure 4B, inset). These results indicate that Rgs1 genetically interacts with MagA during appressorium development in Magnaporthe.

As RGS proteins act by directly binding to their partner G subunits (Siderovski and Willard, 2005), we tested whether Rgs1 interacts physically with MagA, by using Escherichia coli-purified GST-RGS and MBP-MagA fusion proteins. As shown in Figure 4C, GST-RGS and MBP-MagA proteins indeed interacted in the presence of GDP and AlF₄^-, suggesting that the transition state for nucleotide hydrolysis (Berman et al, 1996b) in MagA likely mediates its binding to Rgs1. MBP-MagA could also physically associate with native Rgs1 from the wild-type M. grisea strain (Figure 4D). We conclude that Rgs1 physically interacts with, and regulates, MagA during appressorium initiation in M. grisea. Additionally, we construe that the GTP-bound MagA serves as a dominant signaling moiety (likely independent of G) during surface signaling and pathogenic differentiation in M. grisea.

As the magB strain failed to conidiate, it was impossible to perform appressorium formation assays with this strain. However, examination of the magB^G183S (RGS-insensitive) strain indicated that MagB also plays an important role in appressoria formation. On inductive surfaces, the frequency of appressorium formation was similar (88%) in the magB^G183S and wild-type strains. However, on non-inductive surface, the magB^G183S showed highly elongated germ tubes and a significant increase in appressoria formation (64 versus 12%) when compared with the wild-type (Supplementary Figure S3). To further confirm these findings and to identify the active state of magB during early surface signaling, we tested a GTPase-deficient magB^Q204L mutant on the above-mentioned surfaces. Compared with the magB^G183S, appressoria formation in the magB^Q204L mutant was delayed and occurred at a significantly reduced frequency on the inductive as well as the non-inductive surfaces, resulting in highly elongated germ tubes and immature appressoria on both types of surfaces (Supplementary Figure S3). Additionally, we created and analyzed a magB^G42R mutant (predicted to be constitutively active) that has been shown to form appressoria on non-inductive surfaces (Fang and Dean, 2000). We found that appressoria formation in the magB^G42R mutant was similar to that seen in the magB^G183S, with germ tubes showing extended growth predominantly on non-inductive surfaces before developing precocious appressoria. Taken together, these data suggest that the G_i subunit MagB regulates germ tube elongation during the pathogenic phase and, likely through its GTP-bound state, restricts appressorium formation on undesirable surfaces. No defects in appressorium development were evident in magC or magC^G184S mutants (data not shown).

Rgs1-dependent regulation of cyclic AMP levels

The wild-type strain could not form appressoria on non-inductive surfaces, but addition of the cell-permeable cAMP analog 8-Br-cAMP or the phosphodiesterase inhibitor IBMX to the germinating conidia induced appressorium formation (Figure 5A). These findings were reminiscent of the phenotype of the rgs1 and magA^G187S mutants on non-inductive surfaces. G_s subunits (such as MagA) function to elevate cAMP levels by directly activating adenylyl cyclase (Tesmer et al, 1997b). Therefore, we quantified steady-state cAMP levels in the rgs1, magA^G187S, magA^Q208L and wild-type strains. Compared with the wild-type, the levels of cAMP were 4–5-fold higher in the rgs1, magA^G187S and magA^Q208L mutants (Figure 5B). The cAMP levels in the magB^Q204L mutant were found to be about 10–15% lower than that in the wild type (data not shown). These observations suggest that G_s (MagA)-based signaling directly regulates cAMP levels in Magnaporthe, and that its activation, owing to the loss/reduction of Rgs1-based negative regulation, likely results in elevated intracellular cAMP levels during the initiation step of pathogenic development.

Relationship between thigmotropism and cAMP levels

Next, we tested whether increasing cAMP levels during conidial germination could help bypass the surface dependency for appressoria formation. Exogenous addition of 8-Br-cAMP was unable to induce appressoria in wild-type conidia inoculated on non-inductive soft surfaces such as agar blocks (Figure 6A). Furthermore, cAMP concentrations as high as 30 mM failed to induce proper appressorium formation in the wild-type conidia seeded on soft agar surfaces. Conidia from wild-type and rgs1 strains germinated normally on agar blocks containing 10 mM 8-Br-cAMP, but the resultant germ tubes failed to establish the hooking stage necessary to initiate appressoria formation (Figure 6A). Wild-type and rgs1 conidia germinated efficiently in the presence of 30 mM exogenous 8-Br-cAMP, but the germ tubes showed highly melanized and irregular structures at their tips (Figure 6A). The aberrant apical structures were separated by septa from the germ tube proper (data not shown), but failed to develop into normal appressoria (Figure 6A). Compared with the wild-type, the rgs1 strain showed an overall higher frequency (2–3-fold) of such aberrant structures upon addition of the equivalent amount of 8-Br-cAMP (30 mM; Figure 6A).

To investigate the relationship between surface hardness stimulus and cAMP, we quantified and compared the intrinsic cAMP levels within the germ tubes produced on hard membrane surfaces to those formed on soft agar surfaces. Steady-state cAMP levels were significantly higher in the germ tubes harvested from the hard hydrophilic surfaces (Figure 6B), suggesting that cAMP accumulation is indeed coupled with proper perception and integration of the initial thigmotropic cue. We thus conclude that during pathogenic development, contact of the germ tubes with a hard surface directly regulates cAMP accumulation, likely through activation of G-protein signaling, and leads finally to formation of functional appressoria.

Rgs1 regulates MagB, the G_i subunit, during conidiogenesis

Besides its role in appressorium development, Rgs1 was also found to be involved in asexual development (or 'conidiation'). Microscopic observations indicated that rgs1 colonies produced more conidia than the wild type, and that excess Rgs1 (RGS1 overexpression) inhibited conidiation (Figure 7A). This implicated Rgs1 as a negative regulator of conidiation in Magnaporthe. We examined conidiation in G-deletion mutants (magA, magB and magC), RGS-insensitive mutants (magA^G187S, magB^G183S and magC^G184S) and GTPase-deficient alleles of magA and magB. Loss of MAGB function completely abolished conidiation, whereas the magB^G183S, magB^Q204L and magB^G42R mutants each showed hyper-conidiation defects (Figure 7A). Furthermore, the increased ability to conidiate in the rgs1 strain was completely abolished upon deletion of the MAGB gene, and the resultant rgs1 magB double mutant did not conidiate at all (Figure 7A). No apparent conidiation defects were evident in the magA, magA^G187S, magC or magC^G184S mutants. A surprising finding was the significant reduction (87%) in conidiation in the magA^Q208L mutant. This decrease in conidiation was most likely a consequence of the huge reduction in the number of aerial hyphae and conidiophores in the magA^Q208L mutant (Figure 7A). The magA^Q208L strain also showed precocious pigmentation at the tips of the aerial hyphae and the conidiophores (Figure 7A). The bar chart in Figure 7B depicts the quantification of the total number of conidia produced by the respective mutants mentioned above. We conclude that Rgs1 and MagB act in concert to regulate conidia formation in Magnaporthe and infer that the GTP-bound MagA likely functions as a negative regulator of aerial hyphae and conidiophore development.

E. coli-purified GST-RGS and MBP-MagB fusion proteins were found to physically interact in a GDPAlF₄^--dependent manner (Figure 7C). Moreover, full-length Rgs1 protein was found to associate with MBP-MagB in pull-down assays using total lysates from wild-type but not from rgs1 (Figure 7D). We conclude that Rgs1 physiologically interacts with, and regulates, the G_i MagB during conidiation in M. grisea.

Discussion

Physicochemical properties of the plant surface, which can be mimicked in vitro to some extent, are believed to trigger the formation of appressoria in M. grisea. Surface hydrophobicity is considered to be important, as wild-type conidia form appressoria on hydrophobic, but not hydrophilic, surfaces, thus leading to a general classification of these two surfaces as inductive and non-inductive, respectively. In this study, we characterized a novel Magnaporthe mutant, TMT1398, which abolished surface dependency during appressoria formation and responded efficiently to inductive, as well as non-inductive, surfaces. Through genetic complementation analysis, we ascertained that the defects in TMT1398 (and an rgs1 mutant) were due solely to the loss of Rgs1 protein, thus implicating Rgs1 to be a negative regulator of appressoria formation that couples surface dependency with infection-related morphogenesis.

A rather surprising finding was the inability of wild type and rgs1 to elaborate appressoria on softer surfaces irrespective of their hydrophobic or hydrophilic nature. This suggested that surface hydrophobicity alone is insufficient to induce appressorium development, and that soft surfaces such as agar blocks or wax are non-inductive. Surface hardness, thus, represents a critical signal for appressorium development, as observed in appressorium formation assays on dried agar surfaces (derived from non-inductive fresh agar blocks), where rgs1 and wild-type were now fully capable of forming functional appressoria. We estimated the critical surface hardness required for initiating appressoria, and also demonstrated that the accumulation of cAMP, an important second messenger (Lee and Dean, 1993; Choi and Dean, 1997), depends on contact with a requisite hard surface. Furthermore, exogenous addition of 8-Br-cAMP to germinating wild-type or rgs1 conidia failed to induce appressoria on soft surfaces. This suggests that although cAMP-mediated signaling is essential, it is not sufficient to initiate appressoria on inappropriate surfaces and that contact with a hard surface is indeed necessary to trigger pathogenic behavior. Initiation of appressoria on leaf surfaces or on hydrophobic hard surfaces was comparatively more efficient than it was on dried agar surfaces, thus implying that surface hardness, as well as hydrophobicity, is important for efficient development of appressoria in M. grisea.

As Rgs1 showed a high degree of similarity to G-protein regulators FlbA (Yu et al, 1996), Sst2 (Dohlman et al, 1996) and CpRGS1 (Segers et al, 2004), we investigated the function(s) of G subunits MagA, MagB and MagC as potential substrates of Rgs1 GAP activity. Interestingly, the RGS-insensitive G_S magA^G187S and the GTPase-deficient magA^Q208L were each capable of forming appressoria on both the inductive and non-inductive surfaces similar to the rgs1 strain. Furthermore, Rgs1 physically interacted with an activated form of MagA, thus suggesting that Rgs1 directly regulates MagA during appressoria initiation. Like the rgs1, the magA^G187S and magA^Q208L strains accumulated high levels of cAMP, suggesting that a role for Rgs1 in appressorium formation is to regulate the nucleotide state of MagA and thereby control GTP-MagA-dependent adenylyl cyclase activation and the resultant levels of intracellular cAMP. Comparative analysis of the magA^Q208L and magA^G183S mutants showed that GTP-MagA is indeed a dominant signaling moiety that likely activates downstream effector(s) independent of the G dimer. These data help explain the lack of appressorial defects in the magA strain and are consistent with earlier results related to the G subunit in M. grisea (Nishimura et al, 2003).

Analysis of the magB^G183S, magB^G42R and magB^Q204L strains revealed that a probable function of GTP-bound MagB is to suppress appressoria formation on undesirable surfaces, indirectly resulting in extensive elongation of germ tubes. The duration and strength of G-protein-mediated signaling is regulated by the rate at which GTP is hydrolyzed by G (Dohlman and Thorner, 2001). It is probable that GTP hydrolysis is weak or insufficient in the magB^G183S mutant (and residual in the magB^G42R mutant), thus delaying the reassociation of the G dimer with the mutant magB alleles. We believe that delayed appressorium formation in the magB mutants partly reflects the counteracting function(s) of the dissociated G and the compromised GTPase function that increases the GTP occupancy variably on these mutant MagB variants. A preliminary finding that cAMP levels are subnormal in the magB^Q204L mutant supports a possibility that GTP-MagB, true to its G_i subfamily classification, inhibits adenylyl cyclase activity in M. grisea. In future experiments, we will focus on discerning the exact role of MagB in regulating the accumulation of appressorial cAMP.

FlbA and CpRGS1 are both known to positively regulate asexual development (Yu et al, 1996; Segers et al, 2004). In contrast, we found that Rgs1 negatively regulates conidiation in M. grisea. Elevated Rgs1 levels inhibited conidiation, whereas loss of Rgs1 led to profuse conidiation. An unexpected finding was the inhibitory role of GTP-bound MagA (based on magA^Q208L) during the differentiation of aerial hyphae and conidiophores, indirectly leading to reduced conidiation. It remains to be seen if such reduction in aerial growth (and conidiation) is due to the excessive cAMP in the magA^Q208L strain. However, Rgs1 function during conidia formation per se was mediated primarily through regulation of MagB. MagB was found to be essential for conidiation, and deletion of MAGB in the rgs1 background abolished the production of conidia. Furthermore, the magB^G183S, magB^G42R and magB^Q204L mutants showed excessive conidiation reminiscent of the rgs1 strain. These data suggest that GTP-bound MagB is likely a major signaling moiety during conidia formation. The genetic interactions discussed above were ably supported by biochemical studies that showed Rgs1 physically interacts with MagB, in a manner suggesting a possible GAP activity for Rgs1.

Interestingly, we also found direct interactions between activated MagC (G_II subunit) and Rgs1 in M. grisea (Supplementary Figure S5). Although its biological significance is presently unclear, preliminary results implicate MagC in the maintenance of mycelial hydrophobicity during vegetative growth (data not shown). Overall, our findings suggest that Rgs1 is a unique regulator of G-protein signaling and negatively controls three distinct G subunits during different, albeit very important, biological functions such as asexual development, vegetative growth and infection-related morphogenesis in the model phytopathogen M. grisea.

Future experiments will be aimed at identifying the molecular basis of thigmotropism in Magnaporthe, and its functional relevance (if any) to heterotrimeric G-protein signaling. Additionally, we hope to identify the specific downstream effector(s) of the individual G subunits during the developmental stages mentioned above. We also intend to analyze whether G proteins serve specific function(s) during the host penetration and in planta developmental phase in Magnaporthe.

Materials and methods

Fungal strains and culture conditions

The M. grisea wild-type strain B157 was obtained from the Directorate of Rice Research (Hyderabad, India). For culture maintenance and conidiation, wild-type and mutant strains were grown on prune agar medium (PA) as described (Soundararajan et al, 2004). Mycelia used for genomic DNA, total RNA and total protein extraction were harvested from cultures grown in liquid complete medium (CM) for 2 days.

Appressorium formation assay

Conidia harvested from 10-day-old cultures were filtered through Miracloth (Calbiochem, CA, USA) and resuspended to 1 10⁵ conidia per milliliter in sterile water. Droplets (20–50 l) of conidial suspension were placed on membranes or other tested surfaces and incubated under humid conditions at room temperature. Microscopic observations were made at requisite intervals. Treatment with IBMX (2.5 mM) or 8-Br-cAMP (at 10 or 30 mM) was carried out for 16 h. Photomicrographs were taken using a Nikon Eclipse 80i compound microscope with differential interference contrast optics.

Isolation of TMT1398 and molecular cloning of RGS1

The M. grisea mutant TMT1398 was isolated in an Agrobacterium T-DNA-mediated insertional mutagenesis screen for pathogenesis-related defects (H Liu and NI Naqvi, unpublished data). Genomic DNA extracted from TMT1398 was digested with restriction enzyme BglII and used for Southern analysis to determine the number of T-DNA insertion sites. DNA sequences flanking the single T-DNA insertion in TMT1398 were amplified using the TAIL PCR method (Liu and Whittier, 1995) and sequenced. Full-length cDNA encoded by the candidate gene, RGS1, disrupted in TMT1398, was obtained with primers NIN749 and NIN670 using a one-step RT–PCR kit (Qiagen Corporation, USA). The Magnaporthe Genome Database (http://www.broad.mit.edu/annotation
/fungi/magnaporthe; Dean et al, 2005) was used to anchor the RGS1 locus to the genome contig(s) and the requisite overlapping clones obtained from a BAC library (Clemson University Genomics Institute, USA).

RGS1 deletion, complementation analysis and RGS1 overexpression

The RGS1-deletion mutant was generated using the standard one-step gene replacement strategy. Briefly, about 1 kb of 5' UTR and 3' UTR regions was PCR amplified (see Supplementary Figure S4 for primer sequences and restriction enzyme sites) and ligated sequentially to flank the hygromycin resistance cassette in pFGL59. The gene replacement construct was introduced into M. grisea B157 via Agrobacterium-mediated transformation. CM containing 250 g/ml hygromycin was used for selection. Correct gene replacement event (rgs1HPH1) was confirmed by PCR and Southern analysis. A 7.1-kb BglII–EcoRI fragment from BAC clone 02N7, which contained the entire RGS1 locus, was cloned into the BamHI–EcoRI sites in pFGL97 (bialaphos resistance) and used to complement the rgs1 strain. Resistance to ammonium glufosinate (Cluzeau Labo, France) was used to select the requisite transformants, and single-copy integrants ascertained by DNA gel blots.

G-deletion mutants and site-directed mutagenesis

The same strategy as used for the RGS1 deletion was followed to individually delete the G genes, using their respective 5' UTR and 3' UTR fragments. Two fragments amplified with primers NIN577/NIN578 and NIN579/NIN580, respectively, were cloned subsequently and used for MAGA deletion. The primer pairs NIN714/NIN715 and NIN716/NIIN717 were similarly used to amplify fragments for MAGB replacement, and primer pairs NIN573/NIN574 and NIN575/NIN576 for MAGC deletion (Supplementary Figure S4).

For creating RGS-insensitive mutations (G302S equivalents) in the G subunits, we introduced the desired mutation on the PCR fragments used for making the constructs, which were then used for homology-dependent replacement of the WT locus with the mutant allele, such that in each instance the mutant allele was the sole copy of that particular G and was placed under its native regulation. For MAGA, the fragment amplified with primers NIN865/866 was cloned into the PstI and HindIII sites in pFGL347 to obtain pFGL369. Fragments amplified with primers NIN861/NIN862 (XhoI/SpeI) and NIN863/NIN864 (SpeI/EcoRI) were cloned into the XhoI–EcoRI sites of pFGL369 (hygromycin resistance). The resultant pFGL370 was introduced into M. grisea B157 to finally obtain the magA^G187S mutant strains. A similar strategy was used to create the magA^Q208L, magB^G183S, magB^G42R, magB^Q204L and magC^G184S strains. The mutant alleles were confirmed by requisite PCR amplification and sequencing analyses. The primers used in each instance are listed in Supplementary Figure S4.

Rgs1 antiserum and protein analysis

A fusion protein containing the GST tag and the proximal 1–531 aa of Rgs1 was expressed in E. coli BL21(DE3) and purified using glutathione–Sepharose 4B (Amersham Biosciences, USA) and eluted with 10 mM GSH. A total of 500 g (1 mg/ml) GST-RGS protein was emulsified with an equal volume of adjuvant and injected intramuscularly into 2-month-old New Zealand White female rabbits. Rabbits were boosted at 2-week intervals. Antiserum specificity was ascertained by Western blot analysis of total protein lysates from wild-type and rgs1 mycelia. Rgs1 antiserum was affinity purified using standard protocols.

Protein-related methods

Total protein extracts were obtained by grinding 2-day-old mycelial cultures in liquid nitrogen and resuspended in 300 l of extraction buffer (10 mM Na₂HPO₄ pH 7.0, 0.5% SDS, 1 mM DTT and 1 mM EDTA). Lysates were cleared by centrifugation at 12 000 g for 20 min at 4°C. Protein concentrations in the supernatant were determined by the Bradford assay (Bio-Rad, USA). Protein sample (100 g) from each extract was fractionated by SDS–PAGE, transferred onto a PVDF membrane (Millipore Corporation, USA) and immunoblotted with Rgs1 antiserum (1:1000 dilution). Secondary antibodies conjugated to horseradish peroxidase were used at 1:10 000 dilution. The SuperSignal kit (Pierce, USA) was used to detect the chemiluminescent signals as instructed.

To examine the in vitro interaction between RGS domain and G proteins, 40 g of GST or GST-RGS (amino acids 531–714) bound to glutathione-conjugated beads was incubated for 30 min at room temperature in binding buffer (50 mM Tris–HCl, pH 7.5, 50 mM NaCl and 2 mM DTT) supplemented with 50 M GDP, 50 M GTPS and 5 mM MgCl₂ or 50 M GDP, 5 mM MgCl₂, 10 mM NaF and 30 M AlCl₃. Then, 80 g MBP-MagA or MBP-MagB or MBP-MagC protein was added and incubated at 4°C for 2 h. The beads were washed five times with the binding buffer containing the respective nucleotides and/or AlCl₃ and NaF. The agarose beads and associated proteins were boiled for 6 min, fractionated on SDS–PAGE and immunoblotted with anti-MBP or anti-GST antisera (Sigma-Aldrich, USA).

To obtain total lysates, mycelia were ground into a fine powder and extracted with non-denaturing NP-40 buffer (6 mM Na₂HPO₄, 4 mM NaH₂PO₄, 1% Triton X-100, 200 mM NaCl and 2 mM EDTA). The crude lysate was mixed with MBP or MBP-MagA or MBP-MagB or MBP-MagC proteins, conjugated to amylose beads for 2 h at 4°C in the presence of 50 M GDP, 5 mM MgCl₂, 10 mM NaF and 30 M AlCl₃. The beads were washed five times with the binding buffer. Affinity-purified proteins were immunoblotted with anti-Rgs1 and later reprobed with anti-MBP.

Quantification of intracellular cAMP

Two-day-old liquid mycelial cultures, or conidia germinated for 3 h on 1% soft agar surface or hydrophilic GelBond membranes (Cambrex BioScience, USA), were harvested, frozen in liquid nitrogen and lyophilized for 16 h. The dried samples were individually ground to a fine powder in liquid nitrogen and resuspended in 200 l ice-cold 6% TCA and incubated on ice for 10 min. After centrifugation at 4000 r.p.m. for 15 min at 4°C, the supernatant was collected and washed four times with five volumes of water-saturated diethyl ether. The remaining aqueous extract was lyophilized and dissolved in the assay buffer. The cAMP levels were quantified according to the cAMP Biotrak Immuno-assay System (Amersham Biosciences, NJ, USA).

Supplementary data

Supplementary data are available at The EMBO Journal Online (http://www.embojournal.org).

Acknowledgements

We thank members of the Fungal Patho-Biology Group for helpful suggestions and discussions. We thank S Naqvi, G Jedd and S Oliferenko for comments on the manuscript. Work in the Siderovski laboratory was funded by NIH grant R01 GM074268. This work was supported by intramural research funds (to NIN) from Temasek Life Sciences Laboratory, Singapore.

References

Beckerman JL, Ebbole DJ (1996) MPG1, a gene encoding a fungal hydrophobin of Magnaporthe grisea, is involved in surface recognition. Mol Plant Microbe Interact 9: 450–456 | PubMed | ChemPort |

Berman DM, Kozasa T, Gilman AG (1996b) The GTPase-activating protein RGS4 stabilizes the transition state for nucleotide hydrolysis. J Biol Chem 271: 27209–27212 | Article | PubMed | ISI | ChemPort |

Berman DM, Wilkie TM, Gilman AG (1996a) GAIP and RGS4 are GTPAse-activating proteins for the G_i subfamily of G protein subunits. Cell 86: 445–452 | Article | PubMed | ISI | ChemPort |

Bolker M (1998) Sex and crime: heterotrimeric G proteins in fungal mating and pathogenesis. Fungal Genet Biol 25: 143–156 | Article | PubMed | ChemPort |

Bruno KS, Tenjo F, Li L, Hamer JE, Xu JR (2004) Cellular localization and role of kinase activity of PMK1 in Magnaporthe grisea. Eukaryot Cell 3: 1525–1532 | Article | PubMed | ISI | ChemPort |

Castellone MD, Teramoto H, Williams BO, Druey KM, Gutkind JS (2005) Prostaglandin E2 promotes colon cancer cell growth through a Gs–axin-beta-catenin signaling axis. Science 310: 1504–1510 | Article | PubMed | ISI | ChemPort |

Choi W, Dean RA (1997) The adenylate cyclase gene MAC1 of Magnaporthe grisea controls appressorium formation and other aspects of growth and development. Plant Cell 9: 1973–1983 | Article | PubMed | ChemPort |

Dean RA, Talbot NJ, Ebbole DJ, Farman ML, Mitchell TK, Orbach MJ, Thon M, Kulkarni R, Xu JR, Pan H, Read ND, Lee YH, Carbone I, Brown D, Oh YY, Donofrio N, Jeong JS, Soanes DM, Djonovic S, Kolomiets E, Rehmeyer C, Li W, Harding M, Kim S, Lebrun MH, Bohnert H, Coughlan S, Butler J, Calvo S, Ma LJ, Nicol R, Purcell S, Nusbaum C, Galagan JE, Birren BW (2005) The genome sequence of the rice blast fungus Magnaporthe grisea. Nature 434: 980–986 | Article | PubMed | ISI | ChemPort |

DiBello PR, Garrison TR, Apanovitch DM, Hoffman G, Shuey DJ, Mason K, Cockett MI, Dohlman HG (1998) Selective uncoupling of RGS action by a single point mutation in the G protein alpha-subunit. J Biol Chem 273: 5780–5784 | Article | PubMed | ISI | ChemPort |

Dohlman HG, Song J, Ma D, Courchesne WE, Thorner J (1996) Sst2, a negative regulator of pheromone signaling in the yeast Saccharomyces cerevisiae: expression, localization, and genetic interaction and physical association with Gpa1 (the G-protein alpha subunit). Mol Cell Biol 16: 5194–5209 | PubMed | ISI | ChemPort |

Dohlman HG, Thorner JW (2001) Regulation of G protein-initiated signal transduction in yeast: paradigms and principles. Annu Rev Biochem 70: 703–754 | Article | PubMed | ISI | ChemPort |

Fang EG, Dean RA (2000) Site-directed mutagenesis of the magB gene affects growth and development in Magnaporthe grisea. Mol Plant Microbe Interact 13: 1214–1227 | PubMed | ChemPort |

Fukuhara S, Chikumi H, Gutkind JS (2001) RGS-containing RhoGEFs: the missing link between transforming G proteins and Rho? Oncogene 20: 1661–1668 | Article | PubMed | ISI | ChemPort |

Gilbert RD, Johnson AM, Dean RA (1996) Chemical signals responsible for appressorium formation in the rice blast fungus Magnaporthe grisea. Physiol Mol Plant Pathol 48: 335–346 | Article | ISI | ChemPort |

Graziano MP, Gilman AG (1989) Synthesis in Escherichia coli of GTPase-deficient mutants of Gs. J Biol Chem 264: 15475–15482 | PubMed | ISI | ChemPort |

Koelle MR, Horvitz HR (1996) EGL-10 regulates G protein signaling in the C. elegans nervous system and shares a conserved domain with many mammalian proteins. Cell 84: 115–125 | Article | PubMed | ISI | ChemPort |

Lan KL, Sarvazyan NA, Taussig R, Mackenzie RG, DiBello PR, Dohlman HG, Neubig RR (1998) A point mutation in Galphao and Galphai1 blocks interaction with regulator of G protein signaling proteins. J Biol Chem 273: 12794–12797 | Article | PubMed | ISI | ChemPort |

Lee YH, Dean RA (1993) cAMP regulates infection structure formation in the plant pathogenic fungus Magnaporthe grisea. Plant Cell 5: 693–700 | Article | PubMed | ChemPort |

Lee YH, Dean RA (1994) Hydrophobicity of the contact surface induces appressorium formation in Magnaporthe grisea. FEMS Microbiol Lett 115: 71–76 | Article |

Lengeler KB, Davidson RC, D'Souza C, Harashima T, Shen WC, Wang P, Pan X, Waugh M, Heitman J (2000) Signal transduction cascades regulating fungal development and virulence. Microbiol Mol Biol Rev 64: 746–785 | Article | PubMed | ISI | ChemPort |

Liu S, Dean RA (1997) G protein alpha subunit genes control growth, development, and pathogenicity of Magnaporthe grisea. Mol Plant Microbe Interact 10: 1075–1086 | PubMed | ISI | ChemPort |

Liu YG, Whittier RF (1995) Thermal asymmetric interlaced PCR: automatable amplification and sequencing of insert end fragments from P1 and YAC clones for chromosome walking. Genomics 25: 674–681 | Article |