Fungal Genetics and Biology 49 (2012) 511–520

Contents lists available at SciVerse ScienceDirect

Fungal Genetics and Biology

journal homepage: www.elsevier .com/locate /yfgbi

Functional analyses of regulators of G protein signaling in Gibberella zeae

Ae Ran Park a, Ah-Ram Cho a, Jeong-Ah Seo b, Kyunghun Min a, Hokyoung Son a, Jungkwan Lee c,Gyung Ja Choi d, Jin-Cheol Kim d, Yin-Won Lee a,⇑a Department of Agricultural Biotechnology and Center for Fungal Pathogenesis, Seoul National University, Seoul 151-921, Republic of Koreab Science and Technology Division, Ministry for Food, Agriculture, Forestry and Fisheries, Gyeonggi-Do 427-712, Republic of Koreac Department of Applied Biology, Dong-A University, Busan 604-714, Republic of Koread Eco-friendly New Materials Research Group, Research Center for Biobased Chemistry, Division of Convergence Chemistry, Korea Research Institute of Chemical Technology,Daejeon 305-343, Republic of Korea

a r t i c l e i n f o

Article history:Received 19 September 2011Accepted 9 May 2012Available online 24 May 2012

Keywords:Gibberella zeaeFusarium graminearumRGSG protein signaling

1087-1845/$ - see front matter � 2012 Elsevier Inc. Ahttp://dx.doi.org/10.1016/j.fgb.2012.05.006

⇑ Corresponding author. Fax: +82 2 873 2317.E-mail address: lee2443@snu.ac.kr (Y.-W. Lee).

a b s t r a c t

Regulators of G protein signaling (RGS) proteins make up a highly diverse and multifunctional proteinfamily that plays a critical role in controlling heterotrimeric G protein signaling. In this study, sevenRGS genes (FgFlbA, FgFlbB, FgRgsA, FgRgsB, FgRgsB2, FgRgsC, and FgGprK) were functionally characterizedin the plant pathogenic fungus, Gibberella zeae. Mutant phenotypes were observed for deletion mutantsof FgRgsA and FgRgsB in vegetative growth, FgFlbB and FgRgsB in conidia morphology, FgFlbA in conidiaproduction, FgFlbA, FgRgsB, and FgRgsC in sexual development, FgFlbA and FgRgsA in spore germinationand mycotoxin production, and FgFlbA, FgRgsA, and FgRgsB in virulence. Furthermore, FgFlbA, FgRgsA,and FgRgsB acted pleiotropically, while FgFlbB and FgRgsC deletion mutants exhibited a specific defectin conidia morphology and sexual development, respectively. Amino acid substitutions in Ga subunitsand overexpression of the FgFlbA gene revealed that deletion of FgFlbA and dominant active GzGPA2mutant, gzgpa2Q207L, had similar phenotypes in cell wall integrity, perithecia formation, mycotoxin pro-duction, and virulence, suggesting that FgFlbA may regulate asexual/sexual development, mycotoxin bio-synthesis, and virulence through GzGPA2-dependent signaling in G. zeae.

� 2012 Elsevier Inc. All rights reserved.

1. Introduction

The ascomycete fungus, Gibberella zeae (anamorph Fusariumgraminearum), is the most common causal agent of Fusarium headblight (FHB), a devastating disease for cereal crops worldwide(Leslie and Summerell, 2006). FHB causes significant crop lossesby reducing grain yield and quality, as well as contaminating cere-als with mycotoxins. The main mycotoxins produced by G. zeae aretrichothecene deoxynivalenol (DON) and estrogenic toxin zearale-none (ZEA), both of which are hazardous to humans and livestock(Desjardins, 2006). As the primary and secondary inocula for epi-demics of FHB on cereal crops, G. zeae produces ascospores (sexualspores) and conidia (asexual spores), respectively. Ascospores arehypothesized to be of greater importance in treating FHB epidem-ics, since FHB inoculum requires aerial dispersal to the cereal headsthrough the forcible discharge of ascospores into the air from peri-thecia (Sutton, 1982; Trail, 2007).

Production of spores and secondary metabolites are frequentlylinked in filamentous fungi, and require morphological and physi-ological transitions that are tightly regulated by both environmen-tal signals and intracellular signaling pathways (Brodhagen and

ll rights reserved.

Keller, 2006). In Aspergillus nidulans, asexual development is usu-ally accompanied by sterigmatocystin (ST) production, and theseprocesses are regulated by G protein signaling pathways (Hickset al., 1997). G protein signaling has been well studied in eukary-otes, and has important roles in regulating various cellular func-tions, including mating, development, secondary metabolism, andpathogenicity in fungi (reviewed in Bölker, 1998). HeterotrimericG proteins consist of Ga, Gb, and Gc subunits, and mediate signaltransduction between transmembrane G protein coupled receptors(GPCRs) and intracellular effectors such as mitogen-activated pro-tein kinases, adenylyl cyclase-cAMP dependent protein kinases, ionchannels, and phospholipases (Neer, 1995; Neves et al., 2002; Si-mon et al., 1991). For example, upon receiving extracellular stim-uli, GPCRs interact with G proteins to induce a substitution ofGDP for GTP in the Ga subunit, which leads to dissociation of theactivated Ga subunit from the Gbc subunits. The slow, intrinsicGTPase activity of the activated Ga subunit then hydrolyzes GTP,thereby facilitating the re-association of the GDP-bound Ga sub-unit with Gbc and resetting the G protein cycle. The rate of GTPhydrolysis of the Ga subunit determines the intensity of the signal-ing produced, and plays a key role in controlling the rapid, yet pre-cise signaling responses in the cell (McCudden et al., 2005).

Regulator of G protein signaling (RGS) proteins possess a con-served RGS amino acid domain that interacts with the Ga subunit.

512 A.R. Park et al. / Fungal Genetics and Biology 49 (2012) 511–520

This interaction greatly facilitates the intrinsic GTPase activity ofGa subunit, and rapidly turns off activated G protein signals, there-by negatively regulating G protein signaling (Chidiac and Roy,2003). In addition to directly suppressing G protein signalingthrough GTPase-accelerating protein (GAP) activity, RGS proteinscan also enhance the activation of G-protein pathways by actingas effectors, effector antagonists, and scaffolding proteins forreceptors, G proteins, or other regulatory molecules (Ballon et al.,2006; Chen and Hamm, 2006; Zhong and Neubig, 2001). Subse-quent studies of amino acid substitutions in Ga subunits and dis-ruption of RGS genes have provided additional insight into themechanisms of G protein signaling and novel functions of RGS pro-teins in fungi (Hicks et al., 1997; Miyajima et al., 1989; Yu et al.,1996; Zhang et al., 2011). For example, in Saccharomyces cerevisiae,Sst2p has been found to negatively regulate pheromone signalingvia inhibition of the Ga subunit (Dietzel and Kurjan, 1987), whileother S. cerevisiae RGS proteins, including Rgs2p, Rax1p, andMdm1p, have been found to be required for stress responses, bud-ding polarity, and mating efficiency (Ballon et al., 2006; Chasseet al., 2006). In A. nidulans, FlbA is able to negatively regulate myce-lia proliferation and activates sterigmatocystin (ST) production andconidiation (Yu, 2006). RgsA of A. nidulans down-regulates thestress response and conidia germination, but stimulates ST produc-tion and conidiation (Han et al., 2004). Moreover, in the rice blastfungus, Magnaporthe oryzae, Rgs1 has been found to regulate viru-lence, asexual development, and thigmotropism (Liu et al., 2007),and in F. verticillioides, both FlbA2 and RgsB are required to controlvegetative growth and fumonisin B1 production (Mukherjee et al.,2010).

In our previous study of the phytopathogenic fungus G. zeae,three putative Ga protein coding genes, GzGPA1, GzGPA2, andGzGPA3, were individually deleted (Yu et al., 2008). When GzGPA1was deleted, female sterility was compromised and mycotoxin pro-duction was enhanced. When GzGPA2 was deleted, virulence wasreduced and chitin accumulation increased in the cell wall. How-ever, GzGPA3 null mutants did not exhibit any phenotypic changes.Although the key function of RGS proteins as the negative controllerof intracellular signaling through interactions with Ga subunitswas demonstrated in other fungi, the functions of RGS proteins inG. zeae remain largely unknown. Therefore, in this study, sevenRGS genes from G. zeae were characterized, and mechanistic rolesof FgFlbA in G protein signaling pathways were investigated bysite-directed mutagenesis of Ga proteins and overexpression stud-ies of the FgFlbA gene. The results demonstrate that FgFlbA is in-volved in various cellular processes including conidiation, sexualdevelopment, cell wall integrity, mycotoxin biosynthesis, and viru-lence via GzGPA2-dependent signaling pathways in G. zeae.

2. Materials and methods

2.1. Fungal strains and media

The wild-type strain Z-3639 of G. zeae (Bowden and Leslie,1999) and mutants derived from this strain were maintained inmedia according to the Fusarium laboratory manual (Leslie andSummerell, 2006). Conidia were induced in carboxyl methyl cellu-lose (CMC) medium (Cappellini and Peterson, 1965) and on yeastmalt agar (YMA) (Harris, 2005). All strains were stored as conidiasuspensions in 20% glycerol at �70 �C.

2.2. Nucleic acid manipulations, PCR primers, and PCR conditions

Genomic DNA was extracted using a cetyltrimethylammoniumbromide procedure (Leslie and Summerell, 2006). Restrictionendonuclease digestion, gel electrophoresis, gel blotting, and South-

ern hybridizations with 32P-labeled probes were performed accord-ing to standard procedures (Sambrook and Russell, 2001). PCRprimers used for this study (Table S1) were synthesized by the Bion-ics oligonucleotide synthetic facility (Seoul, Korea). Total RNA wasextracted using the Easy-Spin Total Extraction Kit (Intron Biotech,Seongnam, Korea), and first strand cDNA was synthesized withSuperScriptIII reverse transcriptase (Invitrogen, Carlsbad, CA, USA).

2.3. Targeted gene deletion and complementation

For targeted gene deletion, a double joint (DJ)-PCR strategy wasused (Yu et al., 2004). Briefly, 50 and 30 flanking regions of each tar-get gene were amplified by PCR using the primer pairs, RGS-5for/RGS-5rev and RGS-3for/RGS-3rev, from the wild-type strain,respectively. A geneticin-resistance cassette (gen) was amplifiedfrom pII99 (Namiki et al., 2001). These three amplicons (e.g., 50 re-gion, 30 region, and gen) were then mixed at a 1:1:2 M ratio andfused in a second round of PCR. Fused constructs were amplifiedusing the nested primers, RGS-nf and RGS-nr. To complement dele-tion mutants, RGS genes, including native promoter and termina-tor, were amplified from genomic DNA of the wild-type strainusing the RGS-nf and RGS-nr primer pairs. Intact copies of RGSgenes were then directly transformed into fungal protoplasts ofcorresponding deletion mutants, along with the vector pUCH1 car-rying the hygromycin-resistance gene (hyg) as a selective marker(Turgeon et al., 1987). Fungal transformation was performed aspreviously described (Han et al., 2007).

2.4. Germination test

Germination tests were performed as previously described (Leeet al., 2009b). Briefly, mycelia harvested from potato dextrosebroth (PDB) were spread on YMA and incubated for 2 d at 25 �C un-der near-UV light (wavelength: 365 nm, HKiv Import & Export Co.,Ltd., Xiamen, China) to induce conidiation. Conidia were then har-vested in distilled water, filtered through a layer of Miracloth(CalBiochem, San Diego, CA, USA), washed twice with distilledwater, and centrifuged (5000 rpm, 25 �C, 5 min). Ascospores wereharvested by placing carrot agar upside down to capture ascosp-ores discharged by mature perithecia on petri dish covers.

One ml of spore suspension (106 conidia/ml) was incubated in50 ml PDB. The total number of germinated spores at 0, 1, 2, 3, 4,5, and 6 h were then counted. One hundred conidia for each exper-iment were also observed by light microscopy, and these experi-ments were performed twice with three replicates.

2.5. Protoplast production assay

Protoplast production assays for G. zeae strains were performedaccording to Hou et al. (2002) with some modifications. Briefly,fungal conidia produced from CMC cultures were inoculated intoYPG liquid medium (3 g yeast extract, 10 g peptone, and 20 g glu-cose per liter) at 106 per ml and grown for 12 h with shaking at25 �C. Mycelia were harvested by filtration through sterile What-man No. 2 filter paper and incubated in 1 M NH4Cl containing 1%Driselase (Karlan Research Products, Santa Rosa, CA, USA) to gener-ate protoplasts for 90 min. Formation of protoplasts was monitoredusing a DE/Axio Image A1 microscope (Carl Zeiss, Oberkochen,Germany) with a CCD camera.

2.6. Sexual crosses and ascospore discharge

Mycelia grown on carrot agar for 5 d were mock fertilized forselfing as previously described (Leslie and Summerell, 2006). Foroutcrosses, mycelia of the female strain grown on carrot agarplates were fertilized with 1 ml male strain conidia suspensions

A.R. Park et al. / Fungal Genetics and Biology 49 (2012) 511–520 513

(1 � 106 conidia/ml). The heterothallic Dmat1 strain (Lee et al.,2003) carrying a MAT1-1 deletion was used as a tester strain foroutcrosses. Perithecia and ascospores were observed 9 d after fer-tilization. After sexual induction, all cultures were incubated undernear-UV light (HKiv Import & Export Co., Ltd.).

Ascospore discharge was observed in small acrylic chambers(1 � 2.5 � 5 cm) constructed to minimize free convection (Aylorand Anagnostakis, 1991). A semicircular agar block (11 mm indiameter) covered with mature perithecia was placed on a cover-slip in the chamber for 24 h (Min et al., 2010; Trail et al., 2005).The morphology of asci rosettes and ascospores present 7 and9 d after fertilization was then examined. Perithecia were dissectedon a glass slide in a drop of 15% glycerol and the asci were flattenedunder a coverslip (Son et al., 2011). Differential interferencecontrast (DIC) images were captured using a DE/Axio Image A1microscope (Carl Zeiss) with a CCD camera.

2.7. Mycotoxin analyses

G. zeae strains were inoculated for mycotoxin production onrice medium. Briefly, rice grains (30 g) were placed in 250 mlErlenmeyer flasks and incubated with G. zeae strains at 25 �C(Lee et al., 2002). Rice cultures were harvested 21 d later. ForDON analysis, rice cultures were harvested and extracted with anacetonitrile–water (84:16, v/v). Extracts (500 ll each) were thenconcentrated to dryness and dissolved in 50 ll Sylon BZT(BSA + TMCS + TMS1, 3:2:3, Supelco, Bellefonte, PA, USA), mixedgently, and heated at 60 �C for 5 min for derivatization. After thereaction, 200 ll n-hexane was added to the solution, then 200 lldistilled water, and this solution was allowed to stand until twolayers separated. The upper layer (1 ll) was analyzed using a Shi-madzu QP-5000 gas chromatograph–mass spectrometer (GC–MS,Shimadzu, Kyoto, Japan) and a selected ion monitoring mode pre-viously described (Seo et al., 1996).

For ZEA quantification, rice cultures were extracted with an ace-tonitrile–water (84:16, v/v). The extract was then diluted in meth-anol–water (65:35, v/v) and a Shimadzu LC-10 AD HPLC equippedwith an RF-10A fluorescence detector (Shimadzu) was used for theanalysis. A Symmetry C18 column (15 mm � 4.6 mm; particle size,5 lm; Waters Corp, Milford, MA, USA) was used and the detectionwavelength was 274–466 nm. The mobile phase was 65% aqueousmethanol and had a flow rate of 1 ml/min.

2.8. Virulence test

Fungal strain virulence was determined using the wheat culti-var, Eunpamil, as previously described (Lee et al., 2009a). Briefly,10 ll of conidia suspension (1 � 105 conidia/ml in 0.1% Tween20) was injected into the center spikelet of wheat heads at themid-anthesis stage. Ten wheat heads were injected for each treat-ment. After being incubated in a humidity chamber for 3 d at 25 �C,plants were placed in a greenhouse and spikelets exhibiting FHBsymptoms were counted 14 d after inoculation.

2.9. Site-directed mutagenesis of Ga proteins and RGS overexpression

To generate mutants carrying dominant active and interfering Gproteins, mutations were created from each Ga gene using site-di-rected mutagenesis according to previous reports (Fang and Dean,2000; Liu et al., 2007; Yu, 2006; Yu et al., 1996). The positions forthe site-directed mutagenesis in G. zeae were determined by se-quence alignment among Ga proteins of A. nidulans (Yu et al.,1996) and M. oryzae (Fang and Dean, 2000). Briefly, for the conver-sion of glycine 42 to arginine in GzGPA1 to generate a dominantactive GzGPA1 protein, two PCR fragments were amplified fromthe wild-type strain using primers, GzGPA1-5for/GzGPA1G42R-

3rev and GzGPA1G42R-5for/GzGPA1-3rev. For the conversion ofglycine 203 to arginine in GzGPA1 to generate a dominant interfer-ing GzGPA1 protein, two PCR fragment were amplified from thewild-type strain using primers, GzGPA1-5for/GzGPA1G203R-3revand GzGPA1G203R-5for/GzGPA1-3rev. These fragments were sub-sequently fused as previously described (Yu et al., 2004), and thefinal constructs were amplified using nested primers, GzGPA1-nfand GzGPA1-nr. Each construct with the gzgpa1G42R andgzgpa1G203R mutant alleles was then co-transformed into fungalprotoplasts of the GzGPA1 deletion mutant with the pUCH1 vector(Turgeon et al., 1987). The resulting transformants were initiallyscreened using hygromycin and geneticin, identified mutant alleleswith the absence or presence of the appropriate BsrFI and EaeIsites, and finally was confirmed by sequencing (Fig. S1). A similarstrategy was used to create mutants carrying dominant active orinterfering GzGPA2 and GzGPA3. Glutamine 207 and 204 in theGa subunits of GzGPA2 and GzGPA3 were converted with leucineto generate mutant with dominant active GzGPA2 (gzgpa2Q207L)and GzGPA3 (gzgpa3Q204L) proteins. Glycine 206 and 203 in theGa subunits of GzGPA2 and GzGPA3 were converted with arginineto generate mutants with dominant interfering GzGPA2(gzgpa2G206R) and GzGPA3 (gzgpa3G203R) proteins (Fig. S1).

Overexpression of the FgFlbA gene was achieved as previouslyreported (Lin et al., 2011). Briefly, the 50 and 30 flanking regionsof each target gene were amplified by PCR using the primer pairs,FgFlbA-5for/FgFlbA-OE 5rev and FgFlbA-OE 3for/FgFlbA-3rev,respectively from the wild-type strain. The elongation factor 1apromoter (PEF1a) of F. verticillioides was amplified from pSKGEN(Lee et al., 2011) with primers EF pro-5for and EF pro-3rev. Thesethree amplicons (e.g., 50 region, 30 region, and PEF1a) were thenmixed and fused using a second round of PCR. The resulting con-structs were amplified using the nested primers, FgFlbA-nf andFgFlbA-nr. Overexpression constructs of FgFlbA genes were then di-rectly transformed into fungal protoplasts of FgFlbA deletion mu-tant along with pUCH1, and overexpression of FgFlbA wasconfirmed by quantitative real-time PCR (qRT-PCR) with a SYBRGreen Supermix (Bio-Rad, Hercules, CA, USA) and a 7500 real-timePCR system (Applied Biosystems, Foster City, CA, USA). The endog-enous housekeeping gene, cyclophilin (Cyp1; Broad Institute ID:FGSG_07439.3), was used as a control for normalization (Kwonet al., 2009). PCR assays were performed in triplicate, with two rep-licates per run. For quantification of FgFlbA gene expression, datawere analyzed using SDS Software 1.3.1 (Applied Biosystems).

3. Results

3.1. Identification and deletion of RGS genes in G. zeae

Six RGS genes, including FgFlbA, FgFlbB, FgRgsA, FgRgsB, FgRgsC,and FgGprK, were previously identified in G. zeae (Lafon et al.,2006). In this study, an additional putative RGS gene (Broad Insti-tute ID: FGSG_08679.3) was identified using a library and genomeassignments server (http://supfam.cs.bris.ac.uk/SUPERFAMILY/)(Cuomo et al., 2007; Gough et al., 2001). This newly identified genewas named FgRgsB2.

The seven RGS proteins of G. zeae, as well as the respective pro-teins previously identified in other fungal species including A. nidu-lans, M. oryzae, S. cerevisiae, and F. verticillioides (Chasse et al., 2006;Lafon et al., 2006; Liu et al., 2007; Mukherjee et al., 2010), werethen classified into five subgroups using a neighbor-joining ap-proach available in the MEGA software package (Tamura et al.,2007) (Fig. S2). These RGS proteins contained at least one RGS do-main in combination with other domains such as: two Dishevelled,EGL-10, and pleckstrin (DEP) domains (FgFlbA and FgFlbB), a RGSdomain at the N-terminus (FgRgsA), transmembrane (TM) domains

514 A.R. Park et al. / Fungal Genetics and Biology 49 (2012) 511–520

at the C-terminus (FgRgsB and FgRgsB2), Phox (PX) homology andPX-associated (PXA) domains at the C- and N-termini, respectively(FgRgsC), and six TM domains at the N-terminus (FgGprK).

To determine the function of the seven RGS genes in G. zeae,each RGS open reading frame (ORF) of the wild-type strain was re-placed with the gen gene. In addition, each deletion mutant wascomplemented by introducing the wild-type allele with pUCH1.Both deletion and complementation of each gene were confirmedby Southern analysis (Fig. S3).

3.2. FgFlbA, FgFlbB, FgRgsA, and FgRgsB are involved in vegetativegrowth, conidia morphology, and/or conidia production

Deletion of FgRgsA and FgRgsB resulted in significantly reducedvegetative growth compared to wild-type and complementedstrains on potato dextrose agar (PDA) (Table 1). In addition, growthpatterns for each of the strains on minimal medium were similar tothose observed on PDA (data not shown). To investigate whetherRGS genes affect conidia morphology and production, the sizeand number of conidia produced by each strain were analyzedand recorded. A significant reduction (59%) in conidia productionrelative to wild-type and complemented strains was observed onlywith DfgflbA mutants, while DfgflbB mutants produced shorter andthinner conidia with fewer septa. In contrast, DfgrgsB mutants pro-duced longer and wider conidia than both the wild-type and com-plemented strains (Table 1).

3.3. Deletion of FgFlbA and FgRgsA alters the germination rate ofspores

Deletion of FgFlbA resulted in precocious germination of conidiacompared to wild-type and complemented strains, while deletionof FgRgsA led to a decrease in germination rate. In the wild-typestrain, 5% and 50% of conidia germinated 2 h and 3 h after incuba-tion, respectively. In DfgflbA mutants, approximately 50% ofconidia germinated 2 h after incubation, and 80% of conidia germi-nated by 3 h incubation in potato dextrose broth (PDB) (Fig. 1A). Incontrast, conidia of DfgrgsA mutants did not germinate by the 2 htime point, and only 25% of conidia germinated by the 3 h timepoint. Similar results were obtained in the ascospore germinationrates determined for the DfgrgsA mutants (Fig. 1B).

Table 1Vegetative growth, asexual reproduction, and virulence of G. zeae strains.

Strain Radial growth (mm)a Conidia morphologyb

Length (lm) Width (lm)

WT 84 Ae 45 A 4.7 ADfgflbA 85 A 45 A 4.6 ADfgflbB 85 A 36 B 4.3 BDfgrgsA 80 B 45 A 4.8 ADfgrgsB 51 C 51 C 5.1 CDfgrgsB2 84 A 45 A 4.8 ADfgrgsC 83 A 46 A 4.6 ADfggprK 83 A 44 A 4.8 AFgFlbAc 85 A 45 A 4.8 AFgFlbBc 85 A 46 A 4.7 AFgRgsAc 85 A 46 A 4.8 AFgRgsBc 83 A 46 A 4.7 AFgRgsB2c 84 A 45 A 4.7 AFgRgsCc 85 A 44 A 4.8 AFgGprKc 84 A 45 A 4.6 A

a Radial growth was measured 4 d after inoculation on potato dextrose agar (PDA).b One hundred conidia harvested from yeast malt agar (YMA) were observed for eachc Conidiation was measured by counting the number of conidia produced after 3-d-od Mean number of symptomatic spikelets was counted 14 d after inoculation.e Values within a column with different letters are significantly different according to T

obtained from three biological replicates.

3.4. FgFlbA, FgRgsB, and FgRgsC are involved in sexual development

G protein signaling is required for sexual development in G. zeae(Yu et al., 2008). To test whether RGS genes contribute to sexualdevelopment, we checked fertility of RGS deletion mutants. TheDfgflbA mutants did not develop perithecia by self-fertilization,and lost its capacity for female fertility (Fig. 2A). However, perithe-cia and viable ascospores were normally produced when conidia ofDfgflbA mutants were spermatized with the heterothallic strain,Dmat1, used as a female in outcrossing assays (data not shown).

Both DfgrgsB and DfgrgsC mutants produced normal peritheciaand the number of perithecia produced by both deletion mutantswas not different from that of the wild-type strain (Fig. 2A). How-ever, the number of forcibly discharged ascospores from theseperithecia was dramatically reduced (Fig. 2B). When ascosporeswere collected from the lid of culture plates 9 d after sexual induc-tion, the wild-type strain discharged �450 ascospores per perithe-cium while the DfgrgsB and DfgrgsC mutants discharged �1 and 3ascospores per perithecium, respectively. To examine peritheciamaturation of each mutant and the wild-type strains, we randomlyselected 50 perithecia, dissected, and then observed under micro-scope. Seven days after sexual induction, �90% of asci in thewild-type strain contained ascospores. In contrast, most of asci inDfgrgsB mutants did not contain ascospores 7 d after sexual induc-tion, and 9 d after sexual induction, the ascospore formation ofDfgrgsB was similar to that of wild-type and complemented strainsobserved on 7 d after sexual induction (Fig. 2C). For the DfgrgsCmutant, eight ascospores were detected in asci 7 d after sexualinduction, but the asci contained predominantly abnormal, two-celled, or irregularly curved or shortened ascospores compared tothe wild-type strain (Fig. 2C).

3.5. The effect of FgFlbA and FgRgsA deletions on mycotoxin production

Although the growth on the rice medium was not differentamong the G. zeae strains, the FgFlbA and FgRgsA deletion mutantsproduced higher levels of DON and ZEA than the wild-type strain.In contrast, the other five RGS gene deletions did not exhibit alteredtoxin levels (Fig. 3). Toxin production levels detected in the com-plemented strains was similar to that of the wild-type strain (datanot shown).

Conidia production (number/ml)c Disease indexd

No. of septum

4.3 A 1.7 � 106 A 7.0 A4.3 A 0.7 � 106 B 1.4 B3.6 B 1.5 � 106 AB 6.0 A4.0 A 2.1 � 106 A 1.1 B4.3 A 2.1 � 106 A 1.2 B4.0 A 1.6 � 106 A 7.1 A4.0 A 1.9 � 106 A 5.9 A4.0 A 1.7 � 106 A 6.7 A4.3 A 1.7 � 106 A 6.6 A4.3 A 1.7 � 106 A 7.4 A4.3 A 1.6 � 106 A 6.7 A4.4 A 1.6 � 106 A 7.4 A4.2 A 1.8 � 106 A 7.3 A4.3 A 1.8 � 106 A 7.2 A4.2 A 1.8 � 106 A 6.5 A

strain using light microscopy.ld cultures grown in carboxymethylcellulose (CMC) medium.

ukey’s test (p < 0.05) using SPSS 12.0 software (SPSS, Inc., Chicago, IL). All data were

Fig. 1. Influence of RGS deletion on germination of conidia (A) and ascospores (B). Spores were incubated in potato dextrose broth (PDB) on a rotary shaker (150 rpm) at 25 �Cand were analyzed 2, 3, 4, 5, and 6 h after incubation for germination. Asterisks indicate data that significantly differed (p < 0.05) based on Tukey’s test. WT, G. zeae wild-typestrain Z-3639; DfgflbA, FgFlbA deletion mutant; DfgrgsA, FgRgsA deletion mutant; FgFlbAc, DfgflbA-derived strain complemented with FgFlbA; FgRgsAc, DfgrgsA-derived straincomplemented with FgRgsA.

Fig. 2. Sexual development of G. zeae strains. (A) Each RGS deletion strain was inoculated on carrot agar for 5 d, then mock-fertilized to induce sexual reproduction. After 9 d,self-fertilization was examined using a dissection microscope. Black, ball-like structures are perithecia. Scale bar = 1 mm. (B) Forcible discharge of ascospores from peritheciaproduced on carrot agar. A semicircular agar block covered with perithecia was placed on a coverslip in each chamber to capture horizontally discharged ascospores. (C) Ascirosettes of each G. zeae strain were microscopically observed 7 and/or 9 d after sexual induction of each strain. Scale bar = 20 lm. WT, G. zeae wild-type strain Z-3639; DfgflbA,FgFlbA deletion mutant; DfgflbB, FgFlbB deletion mutant; DfgrgsA, FgRgsA deletion mutant; DfgrgsB, FgRgsB deletion mutant; DfgrgsB2, FgRgsB2 deletion mutant; DfgrgsC,FgRgsC deletion mutant; DfggprK, FgGprK deletion mutant; FgFlbAc, DfgflbA-derived strain complemented with FgFlbA; FgFlbBc, DfgflbB-derived strain complemented withFgFlbB; FgRgsAc, DfgrgsA-derived strain complemented with FgRgsA; FgRgsBc, DfgrgsB-derived strain complemented with FgRgsB; FgRgsB2c, DfgrgsB2-derived straincomplemented with FgRgsB2; FgRgsCc, DfgrgsC-derived strain complemented with FgRgsC; FgGprKc, DfggprK-derived strain complemented with FgGprK.

A.R. Park et al. / Fungal Genetics and Biology 49 (2012) 511–520 515

3.6. FgFlbA, FgRgsA, and FgRgsB are required for virulence

Virulence of each G. zeae strain on wheat heads was quantifiedby counting the number of wheat spikelets showing blight symp-tom. While wild-type and complemented strains were able to col-onize a fresh wheat spikelet and have FHB symptoms manifest onthe entire head within 14 d following inoculation, DfgflbA, DfgrgsA,and DfgrgsB mutants were only able to colonize inoculated spik-elets, yet not spread to other spikelets (Table 1 and Fig. 4). In con-trast, the other four RGS deletion mutants caused typical headblight symptoms similar to that of the wild-type strain.

3.7. Site-directed mutagenesis of Ga and RGS overexpression

To further investigate the correlations between Ga proteins(e.g., GzGPA1, GzGPA2, and GzGPA3) and RGS proteins, the

following mutants were created by site-directed mutagenesis:gzgpa1G42R, gzgpa1G203R, gzgpa2Q207L, gzgpa2G206R, gzgpa3Q204L, andgzgpa3G203R. The mutation in gzgpa1G42R is a conversion of glycine42 to arginine in the Ga subunit of GzGPA1, while the gzgpa2Q207L

and gzgpa3Q204L mutants include substitutions of glutamine 207and 204 to leucine in the Ga subunits of GzGPA2 and GzGPA3,respectively. These mutations were expected to disrupt the endog-enous GTPase activity of the Ga subunit, thereby resulting in dom-inant active G protein signaling. In contrast, gzgpa1G203R,gzgpa2G206R, and gzgpa3G203R mutants had glycine 203 or 206 re-placed with arginine in the Ga subunits as dominant interferingmutants. In addition, the oligomers designed for these site-directedmutagenesis reactions contained modified restriction sites to facil-itate rapid screening of desired mutants. Mutants containing thepresence or absence of the appropriate BsrFI, AccI, and EaeI siteswere then sequenced (Fig. S1A).

Fig. 3. Mycotoxin production by G. zeae strains. Deoxynivalenol (DON) andzearalenone (ZEA) were extracted from rice cultures 21 d after the inoculation ofeach strain, and levels were determined by GC–MS and HPLC, respectively. WT, G.zeae wild-type strain Z-3639; DfgflbA, FgFlbA deletion mutant; DfgflbB, FgFlbBdeletion mutant; DfgrgsA, FgRgsA deletion mutant; DfgrgsB, FgRgsB deletion mutant;DfgrgsB2, FgRgsB2 deletion mutant; DfgrgsC, FgRgsC deletion mutant; DfggprK,FgGprK deletion mutant.

516 A.R. Park et al. / Fungal Genetics and Biology 49 (2012) 511–520

To generate FgFlbA overexpression mutants, the FgFlbAOE strainwas modified to express FgFlbA under control of the promoter fromthe highly expressed EF1a (PEF1a) instead of its native promoter(Fig. S4A). The expression of FgFlbA in the FgFlbAOE strain wasfound to be up-regulated compared to the wild-type strain in com-plete medium (Fig. S4B).

Fig. 4. Virulence of G. zeae strains on wheat heads. The center spikelet of each wheat headDfgflbA, FgFlbA deletion mutant; DfgflbB, FgFlbB deletion mutant; DfgrgsA, FgRgsA deletDfgrgsC, FgRgsC deletion mutant; DfggprK, FgGprK deletion mutant; FgFlbAc, DfgflbA-derivwith FgFlbB; FgRgsAc, DfgrgsA-derived strain complemented with FgRgsA; FgRgsBc, Dfgcomplemented with FgRgsB2; FgRgsCc, DfgrgsC-derived strain complemented with FgRg

3.8. FgFlbA regulates various cellular processes via GzGPA2-mediatedsignaling pathways

Phenotypic analyses of Ga protein and the FgFlbA mutants wereperformed to determine the functional counterpart of FgFlbAamong the three Ga proteins. In these assays, both gzgpa1G42R

and gzgpa1G203R mutants of GzGPA1 did not produce any initialstructures for the perithecium, and the medium of the dominantinterfering GzGPA1 mutant (gzgpa1G203R) was highly melanizedfollowing sexual induction, unlike gzgpa1G42R mutant (Fig. 5). How-ever, perithecium development and pigmentation of the mutantscarrying dominant active (gzgpa2Q207L) and interfering(gzgpa2G206R) GzGPA2 proteins was similar to that exhibited bydeletion and overexpression mutants of FgFlbA, respectively(Fig. 5). Moreover, despite discrepancies of vegetative growth be-tween FgFlbA deletion strain and gzgpa1G42R and gzgpa2Q207L mu-tants, similar phenotypes were observed in sexual/asexualreproduction and virulence assays as well (Table 2, Figs. 5 and 6).The GzGPA3 mutants did not exhibit any phenotypic changes insexual/asexual reproduction or virulence (data not shown). Poten-tial defects in cell wall composition in G. zeae strains were also ana-lyzed based on the efficiency of protoplast production followingtreatment with a fungal carbohydrolase mixture. The GzGPA2 dele-tion mutant did not produce protoplasts up to 90 min after Drise-lase treatment as previously reported (Yu et al., 2008), yet both thedominant active GzGPA2 mutant (gzgpa2Q207L) and the FgFlbA dele-tion strain were found to be highly sensitive to Driselase (Fig. 7).The other mutants exhibited protoplast production similar to thatof the wild-type strain.

In combination, these data suggest that FgFlbA regulates sexual/asexual reproduction, virulence, and cell wall composition in aGzGPA2-dependent manner. However, the developmental pheno-types during mycelial growth and germination rates were ob-served to be inconsistent with other phenotypic patterns(Figs. S5 and S6).

3.9. FgFlbA is differentially involved in DON and ZEA biosynthesis viaregulation of GzGPA1 and GzGPA2 signaling pathways

The gzgpa1G42R and gzgpa2Q207L mutants produced significantlyhigher levels of DON than the wild-type strain. In particular, the

was injected with 10 ll of conidia suspension. WT, G. zeae wild-type strain Z-3639;ion mutant; DfgrgsB, FgRgsB deletion mutant; DfgrgsB2, FgRgsB2 deletion mutant;ed strain complemented with FgFlbA; FgFlbBc, DfgflbB-derived strain complementedrgsB-derived strain complemented with FgRgsB; FgRgsB2c, DfgrgsB2-derived strainsC; FgGprKc, DfggprK-derived strain complemented with FgGprK.

Fig. 5. Self-fertility of G. zeae strains. Each strain was inoculated on carrot agar for 5 d, then mock-fertilized to induce sexual reproduction. These images were obtained usinga dissection microscope after an additional 9 d of incubation. WT, G. zeae wild-type strain Z-3639; DfgflbA, FgFlbA deletion mutant; FgFlbAOE, DfgflbA-derived straintransformed with FgFlbA driven by PEF1a; gzgpa1G42R, GzGPA1 mutant with G42R mutation; gzgpa1G203R, putative dominant interfering GzGPA1 mutant; gzgpa2Q207L, putativeconstitutively active GzGPA2 mutant; gzgpa2G206R, putative dominant interfering GzGPA2 mutant.

Table 2Vegetative growth and conidiation of G. zeae strains.

Strain Radial growth (mm)a Conidia production (number/ml)b

WT 76 Ac 2.0 � 106 Agzgpa1G42R 55 B 2.5 � 103 Bgzgpa1G203R 44 B 1.7 � 106 Agzgpa2Q207L 48 B 2.0 � 104 Bgzgpa2G206R 72 A 2.9 � 106 CDfgflbA 78 A 4.5 � 105 BFgFlbAOE 80 C 1.9 � 106 A

a Radial growth was measured 4 d after inoculation on potato dextrose agar(PDA).

b Conidiation was measured by counting the number of conidia produced after 3-d-old cultures grown in carboxymethylcellulose (CMC) medium.

c Values within a column with different letters are significantly differentaccording to Tukey’s test (p < 0.05) using SPSS 12.0 software (SPSS, Inc.). All datawere obtained from three biological replicates.

A.R. Park et al. / Fungal Genetics and Biology 49 (2012) 511–520 517

dominant active GzGPA2 mutant, gzgpa2Q207L, accumulated �280-fold higher levels of DON than the wild-type strain as did theFgFlbA deletion mutant (Fig. 8). For ZEA production, the gzgpa1G42R

mutant produced a higher level of mycotoxin than the wild-typestrain as the FgFlbA deletion mutant. The G protein dominantinterfering mutants, gzgpa1G203R and gzgpa2G206R, and the FgFlbAoverexpression mutant, produced similar levels of mycotoxinscompared to wild-type (Fig. 8).

Fig. 6. Virulence of the G. zeae strains on wheat heads. The center spikelet of 10 wheat heaDfgflbA, FgFlbA deletion mutant; FgFlbAOE, DfgflbA-derived strain transformed with FgFputative dominant interfering GzGPA1 mutant; gzgpa2Q207L, putative constitutively active

4. Discussion

G protein signaling pathways play key roles in the regulation offungal development, secondary metabolism, and virulence (Bölker,1998; Liu and Dean, 1997). Correspondingly, G proteins of G. zeaehave been shown to contribute to the asexual and sexual develop-ment, toxin production, and virulence (Yu et al., 2008). Based onthe capacity for RGS proteins to mediate G protein signaling inyeast and filamentous fungi (Dietzel and Kurjan, 1987; Lee andAdams, 1994; Yu, 2006), it is hypothesized that RGS proteins alsohave regulatory roles in G protein signaling cascades in G. zeae.Therefore, in this study, seven putative RGS genes in the phyto-pathogenic fungus G. zeae were functionally characterized. Of theseven RGS genes studied, FgFlbA, FgFlbB, FgRgsA, FgRgsB, and FgRgsCwere found to have roles in vegetative growth, conidiation, sporemorphology and germination, sexual development, mycotoxin pro-duction, and/or virulence. Similarly, a recent study in M. oryzae re-vealed that RGS and RGS-like proteins are also involved in acomplex processes governing asexual/sexual development, appres-sorium formation, and virulence (Zhang et al., 2011).

Despite RGS proteins are highly conserved among ascomycetefungi, mutations of these genes frequently exhibit different pheno-typic changes. While deletion of FgFlbA, FgRgsA, or FgRgsB signifi-cantly reduced virulence in G. zeae, deletion of RGS genes wasnot found to affect virulence in F. verticillioides (Mukherjee et al.,

ds were injected with 10 ll conidia suspension. WT, G. zeae wild-type strain Z-3639;lbA driven by PEF1a; gzgpa1G42R, GzGPA1 mutant with G42R mutation; gzgpa1 G203R,

GzGPA2 mutant; gzgpa2G206R, putative dominant interfering GzGPA2 mutant.

Fig. 7. Protoplasts from mycelia of G. zeae strains. Microscopic observation was performed 90 min after a fungal carbohydrolase mixture treatment of each G. zeae strain. WT,G. zeae wild-type strain Z-3639; DfgflbA, FgFlbA deletion mutant; FgFlbAOE, DfgflbA-derived strain transformed with FgFlbA driven by PEF1a; Dgzgpa2, GzGPA2 deletion mutant;gzgpa2Q207L, putative constitutively active GzGPA2 mutant; gzgpa2G206R, putative dominant interfering GzGPA2 mutant. Scale bar = 50 lm.

Fig. 8. Production of deoxynivalenol (DON) and zearalenone (ZEA) by G. zeaestrains. G. zeae strains were grown on rice for 21 d before levels of DON and ZEAwere analyzed by GC–MS and HPLC, respectively. WT, G. zeae wild-type strain Z-3639; DfgflbA, FgFlbA deletion mutant; FgFlbAOE, DfgflbA-derived strain trans-formed with FgFlbA driven by PEF1a; gzgpa1G42R, GzGPA1 mutant with G42Rmutation; gzgpa1G203R, putative dominant interfering GzGPA1 mutant; gzgpa2Q207L,putative constitutive active GzGPA2 mutant; gzgpa2G206R, putative dominantinterfering GzGPA2 mutant.

518 A.R. Park et al. / Fungal Genetics and Biology 49 (2012) 511–520

2010). FgFlbA deletion of G. zeae manifested a similar conidiationphenotype as A. nidulans, yet a different phenotype in regard tomycotoxin production (Fig. 9). A. nidulans FlbA deletion mutantswere previously found to have a block in ST production (Adamsand Yu, 1998), whereas deletion of FgFlbA in G. zeae enhancedDON and ZEA production. Deletion of RGS genes in F. verticillioidesalso triggered overproduction of mycotoxins, where FlbA2 andRgsB1 were associated with the production of Fumonisin B1

(Mukherjee et al., 2010).Although RGS proteins are well known as negative regulators of

cell signaling via G protein cascades, RGS deletion mutants in G.zeae exhibited similar phenotypes as G protein deletion mutantsin regard to mycelial growth, sexual development, virulence, andmycotoxin production. This observation is consistent with studiesof Sst2 and Gpa1 deletion mutants in S. cerevisiae, where constitu-tively active pheromone responses and failure to resume growthwere observed. As a result, it has been hypothesized that the neg-ative action of Sst2p requires the presence of Gpa1 (Dohlman et al.,1996; Miyajima et al., 1987). However, another possibility may bethe contributions of functionally active Gbc heterocomplexes(Fang and Dean, 2000). Therefore, it is possible that Gbc remainsfunctionally active in both deletion and active Ga mutants. In addi-tion, the study of FlbA overexpression in A. nidulans suggests thatFlbA has Ga-independent functions, and these activities involveinteractions with Gbc subunits (Yu et al., 1996). Accordingly, it re-mains to be understood how GzGPB1 propagate signals requiredfor virulence and mycotoxin production in G. zeae (Yu et al., 2008).

In G. zeae, DfgflbA mutant did not exhibit a fluffy, autolytic phe-notype, yet affected cell wall integrity. In filamentous fungi, theability to maintain cell wall integrity is critical for establishing dis-ease in a host, as well as for fungal development, including asexualand sexual reproduction. For example, in F. oxysporum, G. zeae,Ustilago maydis, and Botrytis cinerea (Garcerá-Teruel et al., 2004;Kim et al., 2009; Martín-Udíroz et al., 2004; Soulié et al., 2003), dis-ruption of chitin synthase genes have been shown to reduce path-ogenicity and to result in abnormal hyphae formation. Theincreased sensitivity of the gzgpa2Q207L and DfgflbA mutants toDriselase observed in the present study strongly suggests thatFgFlbA is involved in cell wall integrity by negatively regulatingGzGPA2 activity (Fig. 9) and subsequently their gene mutationsaffects virulence of the phytopathogenic fungus, G. zeae.

FgFlbA is also closely related to DON biosynthesis via regulationof GzGPA2 signaling in G. zeae. Two Ga protein mutants,

Fig. 9. Comparative analysis of G protein signaling pathways in G. zeae, A. nidulans, and M. oryzae. FlbA primarily governs vegetative growth, development, and ST productionin A. nidulans. Rgs1 regulates MagA during pathogenesis in M. oryzae. FgFlbA in G. zeae may be involved in cellular asexual/sexual reproduction, cell wall integrity, mycotoxinproduction, and virulence based on its capacity to negatively regulate GzGPA2 activity, with more complex regulations than M. oryzae and A. nidulans. Dotted lines representless defined interactions between Ga proteins and RGS proteins.

A.R. Park et al. / Fungal Genetics and Biology 49 (2012) 511–520 519

gzgpa1G42R and gzgpa2Q207L, produced higher levels of DON whileonly the gzgpa1G42R mutant produced higher level of ZEA similarto FgFlbA deletion mutants (Fig. 8). Similarly, heterologous expres-sion of the A. nidulans FadAG42R dominant mutant allele was foundto enhance T-2 toxin production in F. sporotrichioides (Tag et al.,2000). However, the dominant active allele of FadA inhibits myco-toxin production, including ST and aflatoxin, in A. nidulans (Yuet al., 1996), A. parasiticus (Hicks et al., 1997), and A. flavus(McDonald et al., 2005). In combination, these results indicate thatG proteins and RGS proteins mediate different responses throughfine-tuned mechanisms in mycotoxin production according to fun-gi. It is possible that the functional diversity associated with G pro-tein signaling pathways in phytopathogenic fungi may be a resultof phenotypic variations, such as differences in host range, repro-ductive manner, and secondary metabolites. In addition, the vari-ous fungal cellular processes may be differentially regulated byactivation/suppression of downstream genes through G proteinsignaling pathways (Brown et al., 1996; Keller et al., 1997; Kimet al., 2005; Proctor et al., 2003).

Although many previous genetic studies of G protein in fungihave used the G42R and Q204L mutations of Ga proteins to gener-ate dominant active Ga (Fang and Dean, 2000; Hicks et al., 1997;Liu et al., 2007; Shimizu and Keller, 2001; Yu, 2006; Yu et al.,1996; Zhang et al., 2011), a recent study of human Gi1 protein,homolog of GzGPA1, demonstrated that Gi1 mutant carryingG42R mutation is unable to move into the fully-active form andtherefore gzgpa1G42R may not carry dominant active form ofGzGPA1 (Bosch et al., 2012). In G. zeae, several phenotypes of thegzgpa1G42R mutants were not contrary to those of dominant inter-fering mutants, gzgpa1G203R in sexual development and virulence.Therefore, further studies are needed on dominant active mutantgzgpa1Q204L to clarify the correlation between RGS and GzGPA1proteins in G protein signaling pathways.

In conclusion, this study has shown that RGS proteins exhibitdiverse functions in asexual/sexual development, mycotoxin pro-duction, and virulence in G. zeae. In addition, FgFlbA was foundto have roles at various developmental stages in mediatingGzGPA2-dependent signaling pathways, and on regulating produc-tion of mycotoxins via unique signal transduction pathways in G.zeae. Further studies are needed to better understand the

interaction mechanisms between RGS proteins and heterotrimericG protein subunits, and the link between G protein signaling anddownstream regulators in G. zeae.

Acknowledgment

This work was supported by a National Research Foundation ofKorea (NRF) Grant funded by the Korean Government (MEST)(2012-0000575).

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.fgb.2012.05.006.

References

Adams, T.H., Yu, J.-H., 1998. Coordinate control of secondary metabolite productionand asexual sporulation in Aspergillus nidulans. Curr. Opin. Microbiol. 1, 674–677.

Aylor, D.E., Anagnostakis, S.L., 1991. Active discharge distance of ascospores ofVenturia inaequalis. Phytopathology 81, 548–551.

Bölker, M., 1998. Sex and crime: heterotrimeric G proteins in fungal mating andpathogenesis. Fungal Genet. Biol. 25, 143–156.

Ballon, D.R. et al., 2006. DEP-domain-mediated regulation of GPCR signalingresponses. Cell 126, 1079–1093.

Bosch, D.E. et al., 2012. A P-loop mutation in Ga subunits prevents transition to theactive state: Implications for G-protein signaling in fungal pathogenesis. PLoSPathog. 8, e1002553.

Bowden, R.L., Leslie, J.F., 1999. Sexual recombination in Gibberella zeae.Phytopathology 89, 182–188.

Brodhagen, M., Keller, N.P., 2006. Signalling pathways connecting mycotoxinproduction and sporulation. Mol. Plant Pathol. 7, 285–301.

Brown, D.W. et al., 1996. Twenty-five coregulated transcripts define asterigmatocystin gene cluster in Aspergillus nidulans. Proc. Natl. Acad. Sci. USA93, 1418–1422.

Cappellini, R.A., Peterson, J.L., 1965. Macroconidium formation in submergedcultures by a non-sporulating strain of Gibberella zeae. Mycologia 57, 962–966.

Chasse, S.A. et al., 2006. Genome-scale analysis reveals Sst2 as the principalregulator of mating pheromone signaling in the yeast Saccharomyces cerevisiae.Eukaryot. Cell. 5, 330–346.

Chen, S., Hamm, H.E., 2006. DEP domains: more than just membrane anchors. Dev.Cell. 11, 436–438.

Chidiac, P., Roy, A.A., 2003. Activity, regulation, and intracellular localization of RGSproteins. Recept. Channels 9, 135–147.

Cuomo, C.A. et al., 2007. The Fusarium graminearum genome reveals a link betweenlocalized polymorphism and pathogen specialization. Science 317, 1400–1402.

520 A.R. Park et al. / Fungal Genetics and Biology 49 (2012) 511–520

Desjardins, A.E., 2006. Fusarium Mycotoxins: Chemistry, Genetics, and Biology. APSPress, St. Paul, MN.

Dietzel, C., Kurjan, J., 1987. Pheromonal regulation and sequence of theSaccharomyces cerevisiae SST2 gene: a model for desensitization topheromone. Mol. Cell. Biol. 7, 4169–4177.

Dohlman, H.G. et al., 1996. Sst2, a negative regulator of pheromone signaling in theyeast Saccharomyces cerevisiae: expression, localization, and genetic interactionand physical association with Gpa1 (the G-protein a subunit). Mol. Cell. Biol. 16,5194–5209.

Fang, E.G.C., Dean, R.A., 2000. Site-directed mutagenesis of the magB gene affectsgrowth and development in Magnaporthe grisea. Mol. Plant–Microbe Interact.13, 1214–1227.

Garcerá-Teruel, A. et al., 2004. Loss of virulence in Ustilago maydis by Umchs6 genedisruption. Res. Microbiol. 155, 87–97.

Gough, J. et al., 2001. Assignment of homology to genome sequences using a libraryof hidden Markov models that represent all proteins of known structure. J. Mol.Biol. 313, 903–919.

Han, K.-H. et al., 2004. Regulators of G protein signalling in Aspergillus nidulans:RgsA downregulates stress response and stimulates asexual sporulationthrough attenuation of GanB (Ga) signalling. Mol. Microbiol. 53, 529–540.

Han, Y.-K. et al., 2007. A novel F-box protein involved in sexual development andpathogenesis in Gibberella zeae. Mol. Microbiol. 63, 768–779.

Harris, S.D., 2005. Morphogenesis in germinating Fusarium graminearummacroconidia. Mycologia 97, 880–887.

Hicks, J.K. et al., 1997. Aspergillus sporulation and mycotoxin production bothrequire inactivation of the FadA Ga protein-dependent signaling pathway.EMBO J. 16, 4916–4923.

Hou, Z. et al., 2002. A mitogen-activated protein kinase gene (MGV1) in Fusariumgraminearum is required for female fertility, heterokaryon formation, and plantinfection. Mol. Plant–Microbe Interact. 15, 1119–1127.

Keller, S.E. et al., 1997. Factors affecting the growth of Fusarium proliferatum and theproduction of fumonisin B1: oxygen and pH. J. Ind. Microbiol. Biotechnol. 19,305–309.

Kim, J.-E. et al., 2009. Gibberella zeae chitin synthase genes, GzCHS5 and GzCHS7, arerequired for hyphal growth, perithecia formation, and pathogenicity. Curr.Genet. 55, 449–459.

Kim, Y.-T. et al., 2005. Two different polyketide synthase genes are required forsynthesis of zearalenone in Gibberella zeae. Mol. Microbiol. 58, 1102–1113.

Kwon, S.-J. et al., 2009. Proteomic analysis of fungal host factors differentiallyexpressed by Fusarium graminearum infected with Fusarium graminearum virus-DK21. Virus Res. 144, 96–106.

Lafon, A. et al., 2006. G-protein and cAMP-mediated signaling in aspergilli: agenomic perspective. Fungal Genet. Biol. 43, 490–502.

Lee, B.N., Adams, T.H., 1994. Overexpression of flbA, an early regulator of Aspergillusasexual sporulation, leads to activation of brIA and premature initiation ofdevelopment. Mol. Microbiol. 14, 323–334.

Lee, J. et al., 2003. Shifting fungal reproductive mode by manipulation of matingtype genes: obligatory heterothallism of Gibberella zeae. Mol. Microbiol. 50,145–152.

Lee, S.-H. et al., 2009a. Roles of the glyoxylate and methylcitrate cycles in sexualdevelopment and virulence in the cereal pathogen Gibberella zeae. Eukaryot.Cell. 8, 1155–1164.

Lee, S.-H. et al., 2009b. GzSNF1 is required for normal sexual and asexualdevelopment in the ascomycete Gibberella zeae. Eukaryot. Cell. 8, 116–127.

Lee, S. et al., 2011. Functional analyses of two acetyl coenzyme A synthetases in theascomycete Gibberella zeae. Eukaryot. Cell. 10, 1043–1052.

Lee, T. et al., 2002. Tri13 and Tri7 determine deoxynivalenol- and nivalenol-producingchemotypes of Gibberella zeae. Appl. Environ. Microbiol. 68, 2148–2154.

Leslie, J.F., Summerell, B.A., 2006. The Fusarium Laboratory Manual. Blackwell Pub,Ames, IA.

Lin, Y. et al., 2011. A putative transcription factor MYT1 is required for femalefertility in the ascomycete Gibberella zeae. PloS ONE 6, e25586.

Liu, H. et al., 2007. Rgs1 regulates multiple Ga subunits in Magnaporthepathogenesis, asexual growth and thigmotropism. EMBO J. 26, 690–700.

Liu, S., Dean, R.A., 1997. G protein a subunit genes control growth, development,and pathogenicity of Magnaporthe grisea. Mol. Plant–Microbe Interact. 10,1075–1086.

Martín-Udíroz, M. et al., 2004. Role of chitin synthase genes in Fusarium oxysporum.Microbiology 150, 3175–3187.

McCudden, C.R. et al., 2005. G-protein signaling: back to the future. Cell. Mol. LifeSci. 62, 551–577.

McDonald, T. et al., 2005. RNA silencing of mycotoxin production in Aspergillus andFusarium species. Mol. Plant–Microbe Interact. 18, 539–545.

Min, K. et al., 2010. A novel gene, ROA, is required for normal morphogenesis anddischarge of ascospores in Gibberella zeae. Eukaryot. Cell. 9, 1495–1503.

Miyajima, I. et al., 1989. GPA1Val-50 mutation in the mating-factor signaling pathwayin Saccharomyces cerevisiae. Mol. Cell. Biol. 9, 2289–2297.

Miyajima, I. et al., 1987. GPA1, a haploid-specific essential gene, encodes a yeasthomolog of mammalian G protein which may be involved in mating factorsignal transduction. Cell 50, 1011–1019.

Mukherjee, M. et al., 2010. Regulators of G protein signalling in Fusariumverticillioides mediate differential host–pathogen responses on nonviableversus viable maize kernels. Mol. Plant Pathol. 12, 479–491.

Namiki, F. et al., 2001. Mutation of an arginine biosynthesis gene causes reducedpathogenicity in Fusarium oxysporum f. sp. melonis. Mol. Plant–Microbe Interact.14, 580–584.

Neer, E.J., 1995. Heterotrimeric G proteins: organizers of transmembrane signals.Cell 80, 249–257.

Neves, S.R. et al., 2002. G protein pathways. Science 296, 1636–1639.Proctor, R.H. et al., 2003. Co-expression of 15 contiguous genes delineates a

fumonisin biosynthetic gene cluster in Gibberella moniliformis. Fungal Genet.Biol. 38, 237–249.

Sambrook, J., Russell, D.W., 2001. Molecular Cloning: a Laboratory Manual. ColdSpring Harbor Laboratory Press, New York.

Seo, J.-A. et al., 1996. Variation in 8-ketotrichothecenes and zearalenone productionby Fusarium graminearum isolates from corn and barley in Korea.Mycopathologia 134, 31–37.

Shimizu, K., Keller, N.P., 2001. Genetic involvement of a cAMP-dependent proteinkinase in a G protein signaling pathway regulating morphological and chemicaltransitions in Aspergillus nidulans. Genetics 157, 591–600.

Simon, M.I. et al., 1991. Diversity of G proteins in signal transduction. Science 252,802–808.

Son, H. et al., 2011. Meiotic silencing in the homothallic fungus Gibberella zeae.Fungal Biol. 115, 1290–1302.

Soulié, M.C. et al., 2003. Disruption of Botrytis cinerea class I chitin synthase geneBcchs1 results in cell wall weakening and reduced virulence. Fungal Genet. Biol.40, 38–46.

Sutton, J., 1982. Epidemiology of wheat head blight and maize ear rot caused byFusarium graminearum. Can. J. Plant Pathol. 4, 195–209.

Tag, A. et al., 2000. G protein signalling mediates differential production of toxicsecondary metabolites. Mol. Microbiol. 38, 658–665.

Tamura, K. et al., 2007. MEGA4: molecular evolutionary genetics analysis (MEGA)software version 4.0. Mol. Biol. Evol. 24, 1596–1599.

Trail, F., 2007. Fungal cannons: explosive spore discharge in the Ascomycota. FEMSMicrobiol. Lett. 276, 12–18.

Trail, F. et al., 2005. Ejection mechanics and trajectory of the ascospores of Gibberellazeae (anamorph Fuarium graminearum). Fungal Genet. Biol. 42, 528–533.

Turgeon, B.G. et al., 1987. Development of a fungal transformation system based onselection of sequences with promoter activity. Mol. Cell. Biol. 7, 3297–3305.

Yu, H.-Y. et al., 2008. Functional analyses of heterotrimeric G protein Ga and Gbsubunits in Gibberella zeae. Microbiology 154, 392–401.

Yu, J.-H., 2006. Heterotrimeric G protein signaling and RGSs in Aspergillus niduians. J.Microbiol. 44, 145–154.

Yu, J.-H. et al., 2004. Double-joint PCR: a PCR-based molecular tool for genemanipulations in filamentous fungi. Fungal Genet. Biol. 41, 973–981.

Yu, J.-H. et al., 1996. The Aspergillus FlbA RGS domain protein antagonizes G proteinsignaling to block proliferation and allow development. EMBO J. 15,5184–5190.

Zhang, H. et al., 2011. Eight RGS and RGS-like proteins orchestrate growth,differentiation, and pathogenicity of Magnaporthe oryzae. PLoS Pathog. 7,e1002450.

Zhong, H., Neubig, R.R., 2001. Regulator of G protein signaling proteins: novelmultifunctional drug targets. J. Pharmacol. Exp. Ther. 297, 837–845.