Eight RGS and RGS-like Proteins Orchestrate Growth, Differentiation, and Pathogenicity of Magnaporthe oryzae

A previous study identified MoRgs1 as an RGS protein that negative regulates G-protein signaling to control developmental processes such as conidiation and appressorium formation in Magnaporthe oryzae. Here, we characterized additional seven RGS and RGS-like proteins (MoRgs2 through MoRgs8). We found that MoRgs1 and MoRgs4 positively regulate surface hydrophobicity, conidiation, and mating. Indifference to MoRgs1, MoRgs4 has a role in regulating laccase and peroxidase activities. MoRgs1, MoRgs2, MoRgs3, MoRgs4, MoRgs6, and MoRgs7 are important for germ tube growth and appressorium formation. Interestingly, MoRgs7 and MoRgs8 exhibit a unique domain structure in which the RGS domain is linked to a seven-transmembrane motif, a hallmark of G-protein coupled receptors (GPCRs). We have also shown that MoRgs1 regulates mating through negative regulation of Gα MoMagB and is involved in the maintenance of cell wall integrity. While all proteins appear to be involved in the control of intracellular cAMP levels, only MoRgs1, MoRgs3, MoRgs4, and MoRgs7 are required for full virulence. Taking together, in addition to MoRgs1 functions as a prominent RGS protein in M. oryzae, MoRgs4 and other RGS and RGS-like proteins are also involved in a complex process governing asexual/sexual development, appressorium formation, and pathogenicity.


Introduction
Signal transduction cascades are the primary means by which external cues are communicated to the nuclei of eukaryotic organisms including fungi. Heterotrimeric guanine-nucleotide binding protein (G-protein) signaling is one of the most important mechanisms by which eukaryotic cells sense extracellular signals and integrate them into intrinsic signal transduction pathways, such as the cyclic AMP (cAMP)-dependent signaling pathway. Heterotrimeric G-proteins are activated by the seven-transmembrane-spanning family of receptors [1]. Binding of signal ligands to such receptors promotes an exchange of GDP to GTP on the Ga subunit, which then triggers a reciprocal conformational change and dissociation from the Gbc heterodimer [2]. Either Ga or Gbc, or both, are then free to activate downstream target effectors such as phosphodiesterase, protein kinases, adenylyl cyclases, phospholipases, and ion channels [3][4][5][6]. The activated G-proteins are later desensitized by the intrinsic GTPase activity of the Ga subunit, followed by re-association with the Gbc complex. Therefore, the guanine nucleotide state of the Ga subunit plays a critical role in controlling G-protein signaling [2]. 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 [7][8][9].
Regulators of G-protein signaling (RGS) proteins primarily function as GTPase-accelerating proteins (GAPs) that promote GTP hydrolysis by the Ga subunits, thereby inactivating the Gprotein and rapidly switching off G protein-coupled signaling pathways [10,11]. All RGS proteins contain a conserved domain of ,120 amino acids that are required for activity and function as key negative regulators of G-protein signaling pathways [12][13][14]. The budding yeast Saccharomyces cerevisiae contains four RGS and RGS-like proteins: Sst2, Rgs2, Rax1, and Mdm1. The archetypical RGS protein Sst2 possesses two N-terminal DEP (Disheveled, EGL-10, Pleckstrin) homology domains and a C-terminal RGS domain, Rgs2 has an N-terminal RGS domain, Rax1 has an Nterminal RGS domain and three C-terminal trans-membrane motifs, and Mdm1 contains an N-terminal PXA and a C-terminal PX domain in addition to an RGS domain [15].
The ascomycete Magnaporthe oryzae is pathogenic to important crops such as rice, barley, wheat, and millet. Rice blast, caused by this heterothallic haploid fungus, is one of the most severe fungal diseases of rice throughout the world [16]. Genetic studies of this important pathogen have advanced dramatically in the past decade, and thus it is an excellent model system for investigating plantpathogen interactions. M. oryzae infects rice plants in a manner typical of many other foliar pathogens. Germ tubes produced from conidia attached to leaf surfaces differentiate into specialized infection structures called appressoria. The enormous turgor pressure generated in appressoria by the accumulation of high concentrations of glycerol is used to penetrate the underlying plant surface [17]. Mutants blocked at appressorium formation or appressorial turgor generation fail to infect healthy rice plants [18]. After penetration, infection hyphae grow in and between plant cells, and eventually result in lesion formation on the plant. Thousands of conidia are produced on the lesions and then released to initiate a new disease cycle on new plant tissues within 3-5 days. Initiation of appressorium formation in M. oryzae was shown to require G-protein and cAMP signaling, because loss of Ga MoMagB and adenylyl cyclase MoMac1 leads to failure in appressorium formation [19,20]. A MAP kinase cascade has also been identified as an essential signaling pathway involved in appressorium formation during pathogenic development [21][22][23]. M. oryzae contains three distinct Ga proteins (MoMagA, MoMagB, and MoMagC), two Gb subunits (MoMgb1 and MoMgb2), and one Gc subunit [19,24,25]. Previous studies revealed that a constitutively active allele, MoMagB G42R , and MoMgb1 have affected Gprotein signaling in vegetative growth, sexual reproduction, and pathogenicity in M. oryzae [24,26]. The DMomgb1 mutant also has a defect in appressoria formation, whereas increased MoMgb1 levels promote precocious appressoria formation [24]. Moreover, expression of a dominant active allele of MoMAGB caused appressoria to form on non-inductive surfaces, while exogenous cAMP can activate appressorium formation in a DMomagB mutant [19,26,27], indicating that MoMagB may sense surface cues and stimulate cAMP synthesis. The regulator of G protein signaling Rgs1, which interacts with all three Ga subunit, was shown to negatively regulate G-protein signaling. Deletion of MoRGS1 leads to a significant increase in intracellular cAMP levels and conidiation, and DMorgs1 mutants also form appressoria on non-inductive hydrophilic surfaces [28]. These observations suggest that G-protein signaling and its regulators play important roles in activating the downstream cAMP pathway and regulating vegetative growth and pathogenic development. Further characterization of G-protein regulators will be helpful in better understanding the role of G-protein-mediated signaling in the regulation of early events during plant infection by the rice blast fungus.
Here, we systematically characterized all eight RGS proteins (MoRgs1-8) in M. oryzae. We found MoRgs1 has a role in regulating cell wall integrity and surface hydrophobicity, in addition to a role in mycelia growth, conidiation, sexual reproduction, and pathogenicity as previously reported [28]. All RGS proteins were involved to certain degree in the regulation of intracellular cAMP levels. Other RGS proteins, MoRgs4 in particular, also exhibit various degree of roles in conidiation, vegetative growth, asexual and sexual development, appressorium formation, and pathogenicity.

M. oryzae contains eight genes encoding RGS and RGSlike proteins
MoRgs1 was first identified as a negative regulator of the Gprotein signaling pathway during the important developmental events such as conidial and appressorium formation in M. oryzae [28]. To obtain a comprehensive understanding of RGS protein functions, seven additional genes encoding RGS and RGS-like proteins were identified and their biological functions characterized.

Author Summary
Regulator of G-protein signaling (RGS) proteins play pivotal roles in modulating heterotrimeric G-protein signal transduction cascades that enable eukaryotic cells to perceive external cues and elicit appropriate physiological and biochemical responses. A previous study for one of the RGS proteins, MoRgs1, has demonstrated that these important roles also exist in the rice blast pathogen Magnaporthe oryzae. To further study functions of RGS proteins in this pathogen, we have characterized additional seven RGS and RGS-like proteins. This has been the largest number of RGS and RGS-like proteins ever discovered from a single fungal organism. We provided evidence to demonstrate that, in addition to MoRgs1 functioning as a prominent RGS protein, other RGS and RGS-like proteins are also involved in a complex process to control asexual/sexual development, appressorium differentiation and penetration, and pathogenicity of M. oryzae. significant (MoRGS1 . 12-fold; MoRGS3 . 10-fold; MoRGS3 . 17-fold), in comparison to MoRGS2 (. three-fold), MoRGS4 (. 1.9-fold), MoRGS5 (. two-fold), MoRGS6 (. four-fold), and MoRGS7 (. two-fold) ( Table 1). These results suggest that RGS proteins in M. oryzae likely play various roles in conidiogenesis, appressorium formation, and infection of the host plant.
Opposed roles of MoRgs1 and MoRgs4, and MoRgs2 and MoRgs3 in asexual reproduction Mutant strains specific to each RGS gene were generated and verified ( Figure S1A-D). Since G-protein and the cAMP pathway are important in conidium formation and loss of MoRgs1 and phosphodiesterase MoPdeH led to enhanced conidiation [28,29], the role of the additional RGS proteins in conidiogenesis was examined. Under normal conditions, only DMorgs4 and DMorgs6 mutant strains exhibited darkened colony with less dense hyphal mat and only DMorgs1 exhibited progressive autolysis ( Figure 2A). Microscopic observations indicated that DMorgs1 and DMorgs4 produced fewer conidiophores and conidia than the wild type strain. In contrast, DMorgs2 and DMorgs3 showed enhanced conidiophore and conidium formation, and DMorgs5, DMorgs6, DMorgs7 and DMorgs8 exhibited no observable changes in the production of conidiophores and conidia ( Figure 2B). To validate these findings, the numbers of conidia produced from each plate were quantified. Consistently, the conidia number of DMorgs1 and DMorgs4 was reduced to 0.64-fold and 0.038-fold that of the wild type, while that of DMorgs2 and DMorgs3 was increased 1.59-fold and 1.55-fold, and DMorgs5, DMorgs6, DMorgs7 and DMorgs8 strains produced nearly the same number of conidia, 1.02-fold and 1.08-fold, as the wild type, respectively ( Table 2). Our finding suggests differentiated roles in conidiophore development and conidia formation by various RGS proteins in M. oryzae: MoRgs1 and MoRgs4 have positive roles, while MoRgs2 and MoRgs3 have negative effect on the regulation of these processes.
To further evaluate the role of MoRgs1 and MoRgs4 in conidiation, a DMorgs1 DMorgs4 double mutant strain was generated and characterized ( Figure S2). The mutant showed even more pronounced defect in vegetative growth with almost no conidia or conidiophores found ( Figure 2C and 2D), indicating that MoRgs1 and MoRgs4 function on different targets in conidiogenesis.

Effects of MoRgs2-8 on appressorium formation
In M. oryzae, physical cues of an inductive surface, such as hardness and hydrophobicity, are required for appressorium formation [28]. However, appressorium can be induced on noninductive surfaces in the presence of exogenous cAMP or inhibitors of cAMP phosphodiesterase [30]. Since MoRgs1 regulates cAMP levels and the DMorgs1 mutant formed normal appressoria on noninductive surfaces [28], we studied the functions of other RGS and RGS-like proteins in appressorium formation. No appressoria formation was observed in mutant strains of DMorgs2, DMorgs3, DMorgs4, DMorgs5, DMorgs6, DMorgs7, and DMorgs8 on noninductive surfaces ( Figure 3). On inductive surfaces, all of the mutant strains, except DMorgs5 and DMorgs8, often produced two appressoria, either on branched germ tubes or on two germ tubes that emerged from one conidial cell after extended incubation of over 24 hours (Figure 3). The percentage of conidia forming two appressoria on germ tubes emerging from a single conidial cell was as high as 15% in these mutants. The appressoria formed on the   secondary branching germ tubes were usually smaller than those formed on the primary germ tubes ( Figure 3). These results indicate that, apart from MoRgs5 and MoRgs8, MoRgs2, MoRgs3, MoRgs4, MoRgs6, and MoRgs7 may also be involved in germ tube growth and appressoria formation.

MoRgs1 and MoRgs4 are indispensable for sexual reproduction
A previous study revealed that G-protein signaling is involved in sexual development in M. oryzae [26]. To determine whether additional proteins play any roles in mating, the DMorgs mutant and the wild-type strains (Guy11, MAT1-2) were crossed to a standard tester strain, TH3 (MAT1-1). After 3 weeks, numerous perithecia were observed at the junctions of the cross between the wild-type strain and TH3, complement transformants (DMorgs1/ MoRGS1 and DMorgs4/MoRGS4)/TH3, but no or only a few perithecia were found for crosses between DMorgs1 and TH3, and

MoRgs1 is essential for the maintenance of cell wall integrity
In our most recent study, deletion of the MoPDEH gene encoding phosphodiesterase resulted in a cell wall integrity defect,  and the DMopdeH mutant also underwent progressive autolysis of mycelia after incubation on CM agar plates for over 14 days [31]. This is similar to autolysis occurred in mutant strains of DMomps1 and DMomck1 mutants [32,33]. MoMps1 and MoMck1 are homologs of S. cerevisiae Slt2 and Bck1 proteins that are involved in cell wall integrity. Because RGS proteins are generally negative regulators of G-protein signaling, deletion of RGS genes may activate downstream cAMP signals, as occurred in the DMopdeH mutant [31]. We tested all DMorgs mutants on CM agar plates for autolysis. Only the mycelia of the DMorgs1 mutant underwent progressive autolysis after incubation for 14 days, similar to the DMopdeH mutant, and none of the mutant strains showed any autolysis ( Figure 2A). Moreover, the autolysis phenotype of the DMorgs1 mutant was suppressed by addition of 1 M sorbitol to the culture medium (Figure 2A), as also found with the DMopdeH and DMomps1 mutants [32,33]. These results suggest that MoRgs1 plays an important role in the maintenance of cell wall integrity.

MoRgs1 and MoRgs4 are required for surface hydrophobicity
Disruption of several hydrophobin genes of M. oryzae, including MoMPG1 and MoMHP1, and MoPDEH resulted in a water-or detergent-soaked, easily wettable phenotype [34,35,36,37,38,39,40]. To determine whether any RGS proteins are involved in surface hydrophobicity, all DMorgs mutant strains were tested with water and detergent solutions. Compared with wild type and complement transformants (DMorgs1/MoRGS1 and DMorgs4/ MoRGS4), none showed an easily wettable phenotype when incubated with water droplets (10 ml) after several hours. However, aerial hyphae of DMorgs1 and DMorgs4 mutants that were grown on CM agar were more readily wettable with a solution containing both 0.02% SDS and 5 mM EDTA within 5 min ( Figure 5). This is similar to the DMomhp1 and DMopdeH mutants [31]. Because the expression levels of MoMPG1 and MoMHP1 were altered in the DMopdeH mutant [31], we speculate that the surface hydrophobicity defect may also be related to MoMpg1 and MoMhp1. To test this hypothesis, we examined the expression levels of MoMPG1 and MoMHP1. Like the DMopdeH mutant, the MoMPG1 expression level showed a significant decrease (. 1000-fold) in the DMorgs1 and DMorgs4 mutants, and a relatively small decrease in the DMorgs2 mutant (. 30-fold). In contrast, MoMHP1 expression increased to different extents in all of the mutants: the fold increase for DMorgs1, DMorgs2, DMorgs3, DMorgs4, DMorgs5, DMorgs6, DMorgs7, and DMorgs8 was two, 10, 30, three, three, six, two, and three-fold, respectively ( Figure 5). These results indicate that MoRgs1 and MoRgs4 play a role in regulating surface hydrophobicity, likely through regulation of the MoMPG1 expression levels.

RGS proteins regulate MoPTH11 expression
The pathogenicity factor MoPth11 was reported to be involved in the cAMP pathway, as its transcription is regulated by exogenous cAMP [41,42]. Disruption of MoPDEH also affected the expression of MoPTH11 during plant infection [31]. The expression of MoPTH11 was thus assessed and found to be downregulated in all of the mutants. The transcription of MoPTH11 decreased more than 30-, 17-, 20-, and 20-fold, in DMorgs1, DMorgs2, DMorgs4, and DMorgs8 mutants, respectively. However, in the mutants DMorgs3, DMorgs5, DMorgs6, and DMorgs7, the expression of MoPth11 was not obviously changed ( Figure 6). These results indicated that the RGS proteins have differentiated roles in the transcription of PTH11 in M. oryzae.

MoRgs4 affects extracellular laccase and peroxidase activities
To determine whether any RGS proteins are involved in the regulation of laccase activity, a pathogenicity factor of certain fungi [43][44][45][46][47], we tested the mutant strains on CM agar and liquid medium supplemented with 0.2 mM 2, 29-azino-di-3-ethylbenzthiazoline-6-sulfonate (ABTS). In each case, decreases in laccase activity were only seen in the DMorgs4 mutant, with a lessoxidized dark purple stain around colonies of the mutant and a lower level of laccase activity in the culture filtrate compared with the wild-type strain ( Figure 7A and 7B). Consistent with these observations, the expression levels of two extracellular laccase genes, MGG11608.6 and MGG13464.6, were also significantly down in DMorgs4 mutants ( Figure 7C).
The Congo red degradation reaction is catalyzed by peroxidase, which requires H 2 O 2 as a limiting substrate [48,49,50]. Discolored halos were observed beyond the wild-type colony margins when cultured on CM agar plates with Congo red, but there was no color change with the DMorgs4 mutant, implying that MoRgs4 is involved in peroxidase activity ( Figure 8A). Enzyme activity assays using ABTS as substrate revealed that the DMorgs4 mutant almost lost its peroxidase activity in the extracellular culture filtrate ( Figure 8B). We further examined the transcriptional level of five peroxidase-encoding genes that possess a signal peptide. The expression levels of MGG08200.6, MGG07790.6, MGG_ 01924.6, and MGG_13291.6 were dramatically downregulated in the DMorgs4 mutant and, in contrast, only MGG_11856.6 was upregulated in this mutant. A lesser degree of downregulation was observed for the other DMorgs mutants ( Figure 8C). These data suggested that RGS proteins might be all involved in the regulation of extracellular peroxidases with MoRgs4 playing a more prominent role.
MoRgs1, MoRgs3, MoRgs4, and MoRgs7 are required for full virulence According to the expression profiles (Table 1), the expression of RGS genes, especially MoRGS1, MoRGS3 and MoRGS7, was significant altered during infectious growth in planta, suggesting their potential roles in pathogenicity. To further test virulence involvement of these RGS proteins, susceptible rice seedlings of CO-39 were sprayed with conidia of various DMorgs mutants. Very few lesions were found up to 7 days post-inoculation with DMorgs1, DMorgs3, DMorgs4, and DMorgs7 mutants. In contrast, rice seedlings sprayed with DMorgs2, DMorgs5, DMorgs6, and DMorgs8 mutants under the same conditions developed numerous typical rice blast lesions similar to the wild type strain ( Figure 9A). Since the DMorgs1, DMorgs3, DMorgs4, and DMorgs7 mutants exhibited normal appressoria formation, we examined the ability of the appressoria in penetration of the onion epidermal cells. Interestingly, only DMorgs3 and DMorgs7 showed decreased penetration efficiency (61% and 21%) compared to the wild type (85%), while no change was found for the other DMorgs mutants (Table 2). However, the majority of the appressoria in the DMorgs2, DMorgs5, DMorgs6, and DMorgs8 mutants formed invasive hyphae, while only a few limited infectious hyphae developed from appressoria of DMorgs1, DMorgs3, DMorgs4, and DMorgs7 mutants (data not shown). To further validate this observation, we repeated the test with rice sheath cells and found that the appressoria of the DMorgs2, DMorgs5, DMorgs6, and DMorgs8 mutants as well as the wild type strain, produced branching infectious hyphae, but the appressoria of the DMorgs1, DMorgs3, DMorgs4, and DMorgs7 mutants produced only limited infectious hyphae ( Figure 9B). These results indicate that MoRgs1,  MoRgs3, MoRgs4, and MoRgs7 have significantly reduced virulence resulting from the defect in infectious hyphal growth.
Moreover, we examined the collective effect of MoRgs1 and MoRgs4 on pathogenicity on detached barley leaves. Consistently, DMorgs1 and DMorgs4 were less virulent than the wild type strain Guy11 and the DMorgs1DMorgs4 double mutant strain lost all pathogenicity ( Figure 9C). Further observation indicated that the DMorgs1DMorgs4 was unable to form appressorium on induction surfaces ( Figure 9D).

Functions of M. oryzae RGS proteins in cAMP and G protein signaling
MoRgs1 plays an important role in regulation of the intracellular cAMP level in M. oryzae [28]. To determine whether other RGS and RGS-like proteins are also involved in this process, we measured intracellular cAMP levels of the mutants in the hyphal stage and compared with that of the wild-type and the DMomac1 mutant strains. The results indicated that all DMorgs mutant strains accumulate somewhat higher levels of cAMP than the wild-type strain. Compared with wild-type, DMorgs1 showed a ,3.7-fold higher intracellular cAMP level, which is consistent with the earlier study [28]. Surprisingly, an increase of five-fold was found in the DMorgs2 mutant while the DMorgs3-8 mutants accumulated 2.8-, 2.8-, 2.2-, 2.1-, 3-, and 3-fold higher levels of cAMP, respectively ( Figure 10). While the mechanism remains unclear, particularly for DMorgs4 and DMorgs5 mutant strains, these results suggest that RGS proteins have important roles in regulating intracellular cAMP levels.
The RGS domain of RGS proteins has a high affinity to Ga and the binding specificity between RGS and Ga proteins often determine signal specificity and amplitude. In S. cerevisiae, Sst2 functions as a negative regulator of pheromones and mating by interacting with Ga Gpa1 [12] and Rgs2 down-regulates glucose activation of the cAMP pathway through direct inhibition of Gpa2 [51]. In a study by Chasse et al., Sst2, Rgs2, Rax1, and Mdm1 were all found to bind Gpa1 and affect Gpa1 signaling, although Sst2 still remained the most regulatory role in Gpa1 signaling and mating [15]. M. oryzae MoRgs1 was found to interact with MoMagA for pathogenicity and MoMagB for conidiation [28]. To find out whether other RGS proteins also function similarly by binding to all or specific Ga proteins, a yeast two-hybrid (Y2H) assay was conducted. In this assay, MoRgs2, MoRgs5, MoRgs7 and MoRgs8 were found to interact with MoMagB, while MoRgs7 only interacted with MoMagA. Surprisingly, all RGS and RGS-like proteins interacted with MoMagC ( Figure 11A and 11B). MoRgs1 failed to interact with MoMagA in contrast to previously reported may due to the different experimental conditions utilized [28].
To further investigate the interaction between MoMagB and MoRgs2, a DMomagBDMorgs2 double mutant strain was generated and analyzed. In addition, two RGS-insensitive mutant alleles, MoMAGB G183S and MoMAGB Q204L allele were also introduced respectively into the DMomagB mutant. Consistent with a role for MoRgs2 in conidiation that is mediated through MoMagB, no conidia was found in the DMomagBDMorgs2 mutant, whereas a significant increase in conidiation was found for DMorgs2 expressing MoMAB G183S and MoMAGB Q204L alleles ( Figure 12).
These results indicated that MoRgs2 functions upstream of MoMagB in conidiation. MoRgs7 regulates pathogenicity may also be mediated through MoMagB, similar to MoRgs1. The

Discussion
Heterotrimeric G-proteins play crucial roles in the regulation of fungal developmental processes and pathogenicity [7]. RGS proteins function as negative regulators to enhance the intrinsic GTPase activity of Ga subunits, thereby inactivating the Gproteins and rapidly switching off the cellular response. A large number of mammalian RGS proteins have been shown to play important roles in various signaling pathways [52]. Many fungal RGS protein homologs have also been well characterized, such as Sst2 of S. cerevisiae [12], FlbA and RgsA of Aspergillus nidulans [53][54][55], CPRGS-1 of Cryphonectria parasitica [56], and Cag8 of Metarhizium anisopliae [57], as well as MoRgs1 of M. oryzae [28]. Here, we identified and characterized seven RGS proteins in addition to MoRgs1 of M. oryzae. Our findings revealed that there are as many as eight RGS and RGS-like proteins in M. oryzae, the most ever found in fungi, which play various roles in the modulation of vegetative growth, asexual/sexual development, cell wall integrity, surface hydrophobicity, appressorium formation and penetration, and pathogenicity in this pathogenic fungus (Table S2).
The M. oryzae MoRgs1 is highly homologous to A. nidulans FlbA and C. parasitica CPRGS-1. Consistent with studies of FlbA, and CPRGS-1 that positively regulate asexual development [54,56], our data indicated that MoRgs1 positively regulates asexual sporulation. This, however, contradicted with the previous report that MoRgs1 has a negatively role in asexual development reported. Regardless, our findings for the role of MoRgs1 are mostly consistent with the previous study [28]. In M. oryzae, deletion of the gene encoding MoMagA or MoMagC has no effect on vegetative growth, appressorium formation, or pathogenicity, except that deletion of MoMAGC encoding MoMagC exhibited reduced conidiation [19]. The DMomagB mutant, however, exhibited significantly reduced vegetative growth, conidiation, and appressorium formation, as well as virulence [19]. A reduction in perithecium formation was observed in DMomagA, but not DMomagC, and no perithecium was found in DMomagB ( Figure S3). The strains with the putative dominant active MoMAGB G42R allele formed appressoria on both hydrophobic and hydrophilic surfaces [26]. This phenotype was also observed in the transformant carrying multiple copies of MoMGB1 [24] and in a phosphodiesterase gene mutant DMopdeH [31], similar to results observed in the DMorgs1 mutant, suggesting that expression of MoMAGB G42R and multiple copies of MoMGB1, inactivation of MoPDEH, and deletion of MoRGS1 all enhance cAMP signaling. Because of this, it is feasible to hypothesize that MoRgs1 has a negative effect on Ga MoMagB signaling. The MoMAGB G42R mutation also reduced conidiation and virulence, failed to form asci, and caused autolysis of aged colonies [26], and the same phenotypes appeared in the DMorgs1 mutant, suggesting that MoRgs1 negatively regulates MoMagB. Surprisingly, MoRgs1 and MoMagB failed to interact with each other in our hand.
Instead, interactions between MoRgs7 and MoMagA, MoRgs2, MoRgs5, MoRgs7, MoRgs8 and MoMagB, and MoRgs1-8 and   MoRgs5, the DMorgs2 mutant did show increased conidiation, suggesting that MoRgs2 may negatively regulate MoMagB in conidiogenesis and pathogenesis. Since MoRgs4 and MoRgs5 are structurally more similar to S. cerevisiae Mdm1 whose role in G protein was not established, their roles in M. oryzae may be established independent of G protein signaling.
Interestingly, with the exception of DMorgs5 and DMorgs8, most DMorgs mutants often formed multiple appressoria on hydrophobic surfaces, which were also observed in the site-directed mutation transformant MoMGB1 D41N , as the aspartic acid residue at 41 is known to be involved in the interaction between MoMgb1 and MgSte20 [24]. This may indicate that RGS proteins could  play a role in activating the D41 of MoMgb1 or have cross-talk between MoSte20 (MgSte20) signaling for germ tube growth and correct regulation of appressorium formation.
The fungal cell wall plays important roles during cell division, growth, and morphogenesis, and in mediating all exchanges between the cell and its environment [58,59]. In pathogenic fungi, the ability to maintain cell wall integrity is critical to the establishment of disease in the host [33]. Several cell wall integrity-associated genes such as MoMps1 and MoMck1 have been characterized in M. oryzae. MoMps1 and MoMck1 have been described as essential for cell wall integrity and pathogenicity [32,33]. In our latest studies, a constitutive activating cAMP pathway mutant DMopdeH also showed an autolysis phenotype like that observed in the DMorgs1 mutant [31]. Thus, the G-protein/ cAMP signaling pathway may have cross-talk with the MAPK pathway in regulating cell wall integrity. Alternatively, like the DMopdeH mutant, the cell wall integrity defect in DMorgs1 could be due to the high intracellular cAMP level in the mutant. However, based on our results, there should be more regulators involved in regulation of cell wall integrity, because other DMorgs mutants besides DMorgs1 also have high intracellular cAMP levels but do not exhibit the autolysis phenotype.
Most hydrophobins confer surface hydrophobicity on fungi forming a spore rodlet layer. Deletion of several hydrophobin genes, including MoMPG1, resulted in a water-or detergentsoaked, easily wettable phenotype and these genes played important roles in multiple infection-related processes [34][35][36][37][38][39]. In previous studies, G-protein and cAMP signaling pathways have been reported to be involved in hydrophobin synthesis and surface hydrophobicity [31,57]. In the insect pathogenic fungus M. anisopliae, loss of the MoRGS1 homolog gene Mocag8 reduced the transcription of a hydrophobin-encoding gene [57]. In the chestnut blight fungus C. parasitica, the RGS protein CPRGS-1 is also known to regulate hydrophobin synthesis [56]. In M. oryzae, a constitutive activate cAMP signaling mutant DMopdeH showed a defect in surface hydrophobicity and a low level of MoMPG1 expression [31]. These results well support the surface hydrophobicity defects of the DMorgs1 and DMorgs4 mutants. Furthermore, they can also be taken as evidence explaining the cell wall integrity defect in DMorgs1.
The high expression level of RGS genes at late infection stages in infected rice leaves indicates their potential role in infectious growth and virulence. Mutation of MoRGS1, MoRGS3, MoRGS4, and MoRGS7 significantly reduced virulence and produced fewer lesions than the wild-type strain, implying that appressoria formed by these mutants are probably defective in penetration. It is likely that these four genes regulate processes involved in the early stages of appressorium penetration, such as development of the penetration peg or differentiation of infectious hyphae. The reduction in pathogenicity may be due to a reduction in development at the pre-penetration stages or a defect in infectious growth of DMorgs mutants in host cells. According to the penetration data, MoRGS1, MoRGS3 and MoRGS7 are indeed involved in penetration and infectious growth and well support the hypothesis. However, unlike DMorgs1, DMorgs3 and MoRGS7, the reduced virulence of DMorgs4 mainly resulted from the defect in infectious hyphal growth, indicating the unique functions of each RGS protein in M. oryzae. RGS proteins may act as different regulators in the regulation of different targets (such as Ga subunits in pathogenesis).
In the present study, DMorgs4 totally lost laccase and peroxidase activity. Laccases are copper-containing oxidases found in many plants, fungi, and microorganisms. Laccase activity has been reported to be involved in virulence in some fungi [43]. In M. oryzae, several virulence attenuation mutants also show loss or reduction of laccase and peroxidase activities [44,45,60]. Therefore, the loss of laccase and extracellular peroxidase activity might be one aspect of the reduced virulence in the DMorgs4 mutant. However, DMorgs1 and DMorgs3 also had reduced pathogenicity but with normal laccase and extracellular peroxidase activity, indicating regulators other than laccases and extracellular peroxidases must exist in the G-protein signaling pathway to control infection-related processes in M. oryzae. The disparity may also indicate that the function of MoRgs4 may be more in line with that of S. cerevisiae Mdm1 and could be independent of G protein signaling.
Overall, our results indicate that different RGS proteins control unique signal transduction pathways in M. oryzae, which are involved in asexual/sexual development, appressorium differentiation, penetration, and infectious growth. It will be important and interesting to distinguish specific functions associated with each RGS proteins and link G-protein signaling to the pathogenicity of the fungus.

Strains and culture conditions
The M. oryzae Guy11 and 70-15 strains were used as wild type for transformation in this study. All strains were cultured on complete medium (CM) agar plates [31]. Liquid CM medium was used to prepare the mycelia for DNA and RNA extraction. For conidiation, strain blocks were maintained on straw decoction and corn (SDC) agar media [31] at 28uC for 7 days in the dark followed by 3 days of continuous illumination under fluorescent light.

Targeted gene deletion and complementation
Standard DNA and RNA manipulations were performed as described previously [61]. The gene-deletion mutants were generated using the standard one-step gene replacement strategy. First, two 1.0 kb of sequences flanking of targeted gene were PCR amplified with primer pairs (Table S1), then a ,2-kb fragment containing the two flanking sequences was amplified by overlap PCR. All amplified sequences and fragments were sequenced and then ligated to flank the hygromycin resistance cassette, which was amplified with primers FL1111 & FL1112 (Table S1), into the pMD19-T vector (Takara Co. Dalian, China). The ,3.4-kb fragments, which contain the flanking sequences and hygromycin cassette, were amplified and transformed into protoplasts of wild type Guy11. The complement fragments, which contain the entire RGS genes and their native promoter regions, were amplified by PCR with primers (Table S1) and inserted into pCB1532 (sulphonylurea resistance) or pYF11 (bleomycin resistance) to complement the mutant strains, respectively.

Vegetative growth
Small agar blocks were cut from the edge of 4-day-old cultures and placed onto CM and SDC media for culturing in the dark at 28uC. The size and morphology of the colonies were examined each day for 10 days and then photographed. The experiment was performed in triplicate.

Hyphal growth and surface hydrophobicity assay
For hyphal growth, small agar blocks were cut from the edge of 4-day-old cultures and placed onto the CM and CM adding 1 M sorbitol and cultured in the dark at 28uC for two weeks. The size and morphology of the colonies were examined every day and photographed on day 14 after incubation. For surface hydropho-bicity assay, the strains were plated onto CM agar plates and incubated at 28uC for 14-day. Sterile distilled water (10 ml) was placed on the surface of cultures. In addition, wettability of aerial hyphae to solution containing both 0.02% SDS and 5 mM EDTA was also assessed as previously described [62].

Appressorium formation, cuticle penetration, and infection assays
Conidia were harvested from 10-day-old cultures, filtered through three layers of lens paper, and resuspended to a concentration of 5610 4 spores per milliliter in sterile water. For appressorium formation and cuticle penetration assays, droplets (30 ml) of conidial suspension were placed on plastic cover slips (hydrophobic), Gelbond films (hydrophilic) and onion epidermal cells and incubated under humid conditions at room temperature as described previously [63]. Appressorium formation and development of invasive hyphae were examined after incubation for 24 hours. For plant infection assays, conidia were resuspended to a concentration of 5610 4 spores per milliliter in a 0.2% (w/v) gelatin solution. Two-week-old seedlings of rice (Oryza sativa cv CO39) were sprayed with 5 ml of conidial suspension of each treatment. Inoculated plants were kept in a growth chamber at 25uC with 90% humidity and in the dark for the first 24 hours, followed by a 12/12 hours light/dark cycle [64]. Lesion formation was observed daily and photographed 7 days after inoculation.
For microscopic observation of penetration and infectious hyphae expansion in rice tissue, rice cultivar CO-39 were prepared as previously described [44] and inoculated with 100 ml of conidial suspension (1610 4 spores per milliliter) on the inner leaf sheath cuticle cells. After 48 hours incubation under humid conditions at room temperature, the leaf sheaths were observed under a microscope. Appressorium turgor was measured by incipient cytorrhysis (cell collapse) assay using a 1-5 molar concentration of glycerol solution as described previously [65].

Mating
Plugs of DMorgs mutants and control strain Guy11 (MAT1-2) and the mating partner strain TH3 (MAT1-1) were pointinoculated 3 cm apart on oatmeal agar medium and incubated at 20uC under constant fluorescent light for 3 to 4 weeks. Mature perithecia were crushed to examine the asci and ascospores approximately 20 to 25 days post-inoculation.

Intracellular cAMP, laccase and peroxidase activities assays
Two-day-old liquid mycelial cultures were harvested, frozen in liquid nitrogen and lyophilized for 16 hours. Intracellular cAMP extraction was followed as previously described [28]. The cAMP levels were quantified according to the cAMP Biotrak Immunoassay System (BD Bioscience, NJ, USA).
Laccase and peroxidase activities were measured from 2-day-old CM liquid cultures. Mycelia were removed completely by filtration and centrifugation (5,000 g at 4uC) and processed using a colorimetric determination as described previously [66].

Yeast two-hybrid assay
The bait constructs were generated by cloning MoMAGA, MoMAGB and MoMAGC full-length cDNAs into pGBKT7, respectively. The RGS cDNAs (MoRGS1, MoRGS4, MoRGS5, MoRGS6, MoRGS7 and MoRGS8: RGS domain only; MoRGS2 and MoRGS3: full-length) were cloned into pGADT7 as the prey constructs (see primers in Table S1). The resulting prey and bait constructs were confirmed by sequencing analysis and transformed in pairs into yeast strain AH109 as the description of BD library construction & screening kit (Clontech, USA). The Trp+ and Leu+ transformants were isolated and assayed for growth on SD-Trp-Leu-His-Ade medium and the expression of LacZ reporter gene following the instructions provided by Clontech. Yeast stains for positive and negative controls were from the Kit.

Co-immunoprecipitation (co-IP) and western blot analysis
The MoMAGB-3xFLAG and MoRGS2-GFP constructs were generated with the yeast gap repair approach [23,67] and confirmed by sequencing analysis. The resulting fusion constructs were co-transformed into protoplasts of 70-15. Transformants expressing the MoMAGB-3xFLAG and MoRGS2-GFP constructs were identified by PCR and confirmed by western blot analysis with an anti-FLAG antibody (Sigma-Aldrich, USA). For co-IP assays, total proteins were isolated from vegetative hyphae as described [23] and incubated with anti-FLAG M2 beads (Sigma-Aldrich). Western blots of proteins eluted from the M2 beads were detected with the anti-GFP [23], anti-FLAG and anti-actin (Sigma-Aldrich) antibodies with the ECL Supersignal System (Pierce, USA).