Fungal Genetics and Biology xxx (2009) xxx–xxx

ARTICLE IN PRESS

Contents lists available at ScienceDirect

Fungal Genetics and Biology

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

Protein kinase A regulates production of virulence determinantsby the entomopathogenic fungus, Metarhizium anisopliae

Weiguo Fang *, Monica Pava-ripoll, Sibao Wang, Raymond St. LegerDepartment of Entomology, University of Maryland, 4112 Plant Science Building, College Park, MD 20742, USA

a r t i c l e i n f o a b s t r a c t

Article history:Received 2 October 2008Accepted 5 December 2008Available online xxxx

Keywords:cAMP–PKAEntomopathogenic fungiMetarhizium anisopliae

1087-1845/$ - see front matter Published by Elsevierdoi:10.1016/j.fgb.2008.12.001

* Corresponding author.E-mail address: wfang1@umd.edu (W. Fang).

Please cite this article in press as: Fang, W(2009), doi:10.1016/j.fgb.2008.12.001

Metarhizium anisopliae is a model system for studying insect fungal pathogenesis. The role of cAMP signaltransduction in virulence was studied by disrupting the class I PKA catalytic subunit gene (MaPKA1). ThePKA mutant (DMaPKA1) showed reduced growth and greatly reduced virulence. PKA was dispensable fordifferentiation of infection structures (appressoria), but differentiation was delayed and the appressoriawere defective because of reduced turgor pressure. DMaPKA1 germinated at similar rates as the wild typein glucose and glycerol, but germination was delayed on alanine. Conidial adhesion and appressoriumformation by DMaPKA1 against a plastic surface was fully inhibited with glucose as sole nutrient source.Adhesion to plastic was not inhibited with glycerol or alanine, but appressorium formation was delayed.DMaPKA1 showed reduced tolerance to the oxidative agent diamide, but not to H2O2 and methyl-violo-gen. Comparative transcriptome analysis of DMaPKA1 and the wild type strain showed that PKA isresponsible for up-regulating approximately one-third of the genes induced by insect cuticle, includingsubsets of those responsible for differentiation of appressoria and penetration pegs, cuticle degradation,nutrient acquisition, pH regulation, lipid synthesis, cell cycle control and the cytoskeleton. PKA was nothowever required for expression of toxin-producing genes. We conclude therefore that MaPKA1 isrequired for sensing host-related stimuli and transduction of these signals to regulate many infectionprocesses.

Published by Elsevier Inc.

1. Introduction

Biological control agents, such as insect pathogenic fungi, offeran environmentally friendly alternative to chemical pesticides.However, their use has been limited by poor efficacy (St. Legeret al., 1996). Detailed knowledge of the mechanisms of fungalpathogenesis is needed for mycoinsecticide improvement. Theascomycete Metarhizium anisopliae has been used as model tostudy insect fungal pathogenesis. Its conidia adhere to the insectcuticle, germinate and the germ tubes differentiate into swolleninfection structures called appressoria. The appressoria producepenetration pegs which penetrate the insect cuticle via a combina-tion of mechanical pressure and cuticle degrading enzymes, princi-pally proteases and chitinases. Hyphae proliferate as a yeast likephase (blastospores) within the insect which is killed by a combi-nation of fungal growth and toxins. Hyphae then emerge from andconidiate on the cadaver. Several key genes involved in these pro-cesses have been identified including an adhesin (MAD1) andhydrophobins that are responsible for adherence to the cuticle(Wang and St. Leger, 2007a; St. Leger et al., 1992). The cuticledegrading enzymes and their genes have also been characterized

Inc.

. et al., Protein kinase A reg

(Bagga et al., 2004). A regulator of the G protein signaling pathwayis involved in conidiation and hydrophobin synthesis (Fang et al.,2007). An osmosensor signals to penetrant hyphae that they havereached the hemocoel (Wang et al., 2008) and a perilipin (the firstcharacterized in fungi) regulates the turgor pressure of infectionstructures (Wang and St. Leger, 2007b). The production of MCL1is required for evading insect immune responses. It contains a col-lagen domain, and is so far unique to Metarhizium (Wang and St.Leger, 2006).

We have not yet identified the signal transduction pathwaysthat are the master regulators of these virulence determinants.However, in previous studies we found that specific inhibitors ofprotein kinase A (PKA) delayed both appressorium formation andexpression of cuticle degrading enzymes by M. anisopliae (St. Legeret al., 1990). Sensing of environmental stimuli and transduction ofthe corresponding signal via the cAMP–PKA signal pathway playsan essential role in the virulence of a variety of human and plantpathogenic fungi. PKA is required for the production of functionalappressoria and pathogenicity in the rice blast pathogen Magna-porthe grisea (Mitchell and Dean, 1995). The disruption of PKA re-duced capsule size and attenuated virulence in the humanpathogen Cryptococcus neoformans (D’Souza et al., 2001). A PKAdeficient mutant of Ustilago maydis has a constitutively filamen-tous phenotype and is nonpathogenic (Gold et al., 1994; Larraya

ulates production of virulence determinants ..., Fungal Genet. Biol.

2 W. Fang et al. / Fungal Genetics and Biology xxx (2009) xxx–xxx

ARTICLE IN PRESS

et al., 2005). Using microarray analysis, SAGE (serial analysis ofgene expression) and proteomic approaches to analyze mutantsof the cAMP cascade has shown that cAMP–PKA signaling regulatesvarious genes involved in cell wall synthesis, translation, transportfunctions, the tricarboxylic acid cycle and glycolysis in C. neofor-mans (Hu et al., 2007), Candida albicans (Harcus et al., 2004) andAspergillus fumigatus (Grosse et al., 2008).

In this study, a PKA catalytic subunit gene Mapka1 (Metarhiziumanisopliae protein kinase A 1) was cloned and disrupted in M. ani-sopliae. The mutant had impaired appressorium development andwas almost avirulent. Transcriptomes of WT and the mutant wereprofiled by microarray analysis to identify genes regulated by PKA.We identified 244 down-regulated genes and one up-regulatedgene (Pr1D) in DMaPKA1, and found that links between PKA, com-ponents of the translation machinery, transport, stress responseand metabolic functions were conserved in M. anisopliae. However,we also found that down-regulated genes included those involvedin appressorium and penetration peg formation, cuticle degrada-tion and pH regulation (Fig. 5). M. anisopliae also differed fromother fungi in that sterol metabolism, rather than phospholipidsynthesis was controlled by PKA.

2. Materials and methods

2.1. Gene disruption

We constructed a master Ti vector, pFBARGFP, for gene disrup-tion in M. anisopliae. The Hygromycin resistance gene cassette inpPk2 (Covert et al., 2001) was replaced by the herbicide resistancebar gene cassette from pBARGPE1 (Pall and Bruhelli, 1993) to formpBAR. The bar cassette was inserted into EcoR I and Xba I sites. Theegfp cassette was excised from SK–GFP (Fang et al., 2006) by diges-tion with EcoRI and SpeI and inserted into the corresponding sitesin pBAR to form pFBARGFP. Using this vector, disruptants willshow herbicide resistance but no fluorescence.

To construct the vector for MaPKA1 disruption, the 50 end ofMaPKA1 was cloned by PCR with primers HMaPKA5-1 and HMaP-KA5-2 based on the previously obtained MaPKA1 sequence(AF116597) (Supplement 1). The 50 end of MaPKA1 was then in-serted into the Xba I site of pFBARGFP to form pFBARGFP-MaP-KA1-5. The 30 end of MaPKA1 was cloned with primersHMaPKA3-1 and HMaPKA3-2 (Supplement 1), and was then clonedinto pFBARGFP-MaPKA1-5 to form disruption vector pFBARGFP–MaPKA1. The introduction of the disruption vector into Agrobacte-rium tumefaciens AGL-1 and fungal transformation protocols wereas previously described (Fang et al., 2006). The strategy used forconfirmation of the disruption of MaPKA1 by PCR and Southernblotting was as previously described (Fang et al., 2008). The prim-ers for PCR confirmation are listed in Supplement 1.

To rescue DMaPKA1, MaPKA1 including its upstream and down-stream regulatory sequence was cloned by PCR with primers MaP-KA5 and MaPKA3 (Supplement 1). The resulting PCR product wassequenced to confirm no mutation, and subsequently inserted intoXba I and Sma I sites of pBENFGFP (Fang et al., 2006). The insertionof MaPKA1 into DMaPKA1 was confirmed by PCR and Southern blot-ting. The expression of this gene was confirmed by RT-PCR withprimers RT-PKA5 and RT-PKA3 (Supplement 1). RT-PCR was con-ducted using the one-step RT-PCR kit from Qiagen (Valencia, CA).

2.2. Bioassay

WT and mutant M. anisopliae were bioassayed using Galleriamellonella from Pet Solutions (Beavercreek, OH). Insects were inoc-ulated by immersion in conidial suspensions or by injection ofconidia into the hemoceol. Conidia were collected in 0.01% Triton

Please cite this article in press as: Fang, W. et al., Protein kinase A reg(2009), doi:10.1016/j.fgb.2008.12.001

X-100 solution, which was then filtered through glass wool toeliminate hyphae. For topical application, conidial suspensions atthe concentrations from 1 � 106 to 2 � 107 conidia/ml were used.Insects were inoculated by being immersed in a suspension of con-idia for 5 s, and then were transferred individually into a container.After death, a sterile wet cotton ball was placed into the containerto promote fungal emergence from cadavers. To bypass the cuticle,3 ll of a conidial suspension (104 conidia ml�1) was injected intoeach insect. Both experiments were repeated three times with 30insects per replicate. Mortality was recorded every day with topicalinoculation and every 12 h with injection.

2.3. Germination and appressoria formation against insect cuticles andplastic surface

Observations of appressorium formation against insect cuticlesand a hydrophobic plastic surface were conducted as describedpreviously (Wang and St. Leger, 2005a). Housefly wings, locusthind wings and G. mellonella larval cuticles were used. The cuticleof G. mellonella larvae was prepared as described previously (St. Le-ger et al., 1986). Cuticles were inoculated with 3 ll of conidial sus-pension (�107 ml�1) and incubated on 1.5% water agar. To testinduction against a hydrophobic surface, conidia were inoculatedinto 5.5-cm polystyrene Petri dishes containing 2 ml of 0.0125%YE (yeast extract). Appressorium formation against the insect cuti-cle and Petri dish were observed on three consecutive days at 12 hintervals.

The cAMP–PKA pathway is believed to be involved in the germi-nation of M. anisopliae conidia in response to different carbonsources (St. Leger et al., 1994). To compare germination behaviorof DMaPKA1 in different carbon sources with WT, 30 ll of freshconidial suspension (108 ml�1) was inoculated in 5.5-cm polysty-rene petri dishes containing 2 ml of basal medium (0.2% NaNO3,0.1% KH2PO4) supplemented with 1% glucose, 1% glycerol or 1% ala-nine. A nutrition rich medium SDB (Sabouraud dextrose broth) wasused as a control. Plates were incubated at 27 �C for 9 h, and germi-nation was checked at 3 h intervals. One hundred conidia fromeach of five replicates were scored microscopically to assess thegermination frequency and appresorium formation. All germina-tion, growth and appressorium observation in this study were re-peated three times with 3 replicates each time.

2.4. Measuring turgor pressure of appressorium

PEG8000 solutions with concentrations ranging from 10% to120% (w/v) were used to generate turgor pressure as describedby Wang et al. (2007b). The turgor pressure in appressoria was re-corded as the pressure which collapsed 50% of appressoria after10 min incubation.

2.5. Effect of cuticular melanin on germination and appressoriumformation

Tyrosinase on insect cuticle oxidizes L-DOPA (L-3,4-dihydroxy-phenylalanine) to generate melanin. The resulting products aretoxic to M. anisopliae (St. Leger et al., 1988). The toxicity of cuticularmelanin to WT and DMaPKA1 was tested in vitro. To obtain cutic-ular tyrosinase, cuticles of 20 G. mellonella larva were removed,comminuted under liquid nitrogen and transferred into 20 ml ofice-cold water. After shaking for 20 min, the insect cuticle andother debris was removed by filtration and 0.1 g L-DOPA wasadded to the enzyme solution rapidly producing a suspension ofblack melanin (St. Leger et al., 1988). Germination and appressoriaformation by the WT and mutant was checked every 6 h in themelanin suspension with 0.0125% YE as described above.

ulates production of virulence determinants ..., Fungal Genet. Biol.

W. Fang et al. / Fungal Genetics and Biology xxx (2009) xxx–xxx 3

ARTICLE IN PRESS

2.6. Lipid quantification and neutral lipid observation withfluorescence microscopy

Staining of neutral lipid droplet with Bodipy (Invitrogen) and li-pid quantification in conidia and hyphae were conducted as de-scribed in Wang et al. (2007b).

2.7. Oxidative stress, high osmotic stress and fungicide tolerance test

For the oxidative stress test, conidial germination was mea-sured in 2 ml of SDB supplemented with H2O2 (0.05%, 0.005% or0.0005%), diamide (10 lg ml�1, 3 lg ml�1 and 0.3 lg ml�1) ormethyl-viologen (45 lg ml�1, 22.5 lg ml�1 or 2.25 lg ml�1). Thirtymicroliter of a conidial suspension (108 ml�1) was inoculated intoeach Petri dish. Plates were incubated at 27 �C for 9 h, and germi-nation was checked every subsequent 3 h.

To test the effects of osmotic stress, 5 ll of a conidial suspension(107 ml�1, 106 ml�1, 105 ml�1 or 104 ml�1) was applied onto a PDA(potato dextrose agar) plate containing 0.7 M KCl or 1.5 M KCl.Diameter of the colony was measured daily.

For fungicide tolerance test, conidial suspensions were appliedto PDA plates containing Congo red (from10 lg ml�1 to1000 lg ml�1 or hygromycin B (from 100 lg ml�1 to 1000 lgml�1). Five microliter of a conidial suspension (from107 ml�1 to104 ml�1) were applied to the plates. Two days after inoculation,the diameter of each colony was measured.

2.8. Growth on PDA with different pH value

To test the relationship between MaPKA1 and pH regulation, theWT and DMaPKA1 were inoculated on PDA plates with pH valuesranging from 4.0 to 10.0. The diameters of colonies were recordeddaily. To test the ability of M. anisopliae to alter environmental pHvalue, 30 ll of a conidial suspension (107 ml�1) was inoculated into50 ml PDB (potato dextrose broth) with initial pH values rangingfrom 4.0 to 10.0. The pH value in each flask was monitored every12 h.

2.9. Microarray analysis

The impact of disrupting PKA on the transcriptome was investi-gated by microarray analysis. A standard mycelial inoculum (1.5 gwet weight) produced by inoculating conidia into SDB and cultur-ing at 27 �C for 30 h, was transferred into minimal medium con-taining 1.5% locust cuticle. After 24 h at 27 �C, the mycelium washarvested by filtration and RNA was extracted with the Plant miniRNA preparation kit (Qiagen, Valencia, CA). Locust cuticle was pre-pared according to St. Leger et al. (1986).

cDNA slides for microarray were prepared as previously de-scribed (Wang et al., 2005b). In total, 1730 amplified clones wereprinted in triplicates on the slides. Additional background controlwas provided by 30 randomly distributed spots of 3� SSC buffer.Printing, hybridization, and scanning of slides was as described be-fore (Freimoser et al., 2005). Microarray data analysis was con-ducted with a TIGR TM4 system. Fold differences greater than 1.8was considered as significant differences.

2.10. Real time RT-PCR

Real time RT-PCR was used to confirm gene expression differ-ences identified by microarray analysis. The genes analyzed andtheir primers for real time RT-PCR analysis are listed in Supple-ment 1. Gpd, tef and tyr genes were used as reference genes as pre-viously described (Fang and Bidochka, 2006). First-strand cDNAwas synthesised using a Quantitect Reverse Transcription Kit (Qia-gen, Valencia, CA). Real time PCR was performed with the Quanti-

Please cite this article in press as: Fang, W. et al., Protein kinase A reg(2009), doi:10.1016/j.fgb.2008.12.001

tect SYBR Green PCR Kit (Qiagen, Valencia, CA) using a Lightcycler480 (Roche, Indianapolis, IN). The PCR protocol included a 2 mininitial denaturation step at 94 �C, followed by 40 cycles of 15 s at95 �C, 30 s at 60 �C, and 30 s at 72 �C. Fluorescence was collectedat each polymerization step.

Primers were considered useful when PCR efficiency (standardcurve slope) was more than 1.9. DNA in each treatment was quan-tified from its respective standard curve. The relative expressionlevel of each gene was then normalized against the three referencegenes (gpd, tef and tyr), using geNORM (Vandesompele et al., 2002).

3. Results

3.1. Gene cloning and characterization

A full length PKA catalytic subunit gene MaPKA1 and its up-stream and downstream regulatory sequences were cloned. TheORF (open reading frame), which is interrupted by four introns,is 1569 bp long and predicts a protein with 522 amino acid resi-dues. MaPKA1 has a typical catalytic domain (S_TKc) of Serine/Threonine protein kinases. MaPKA1 showed 59.1%, 56.9%, 60%and 66% amino acid identity to the PkaA from Aspergillus nidulans(Fillinger et al., 2002), adr1 from Ustilago maydis (U23730) (Dür-renberger et al., 1998), TPK1 from Saccharomyces cereviae (Todaet al., 1987), and CPKA from Magnaporthe grisea (U12335) (Adachiand Hamer, 1998), respectively. These are all class I PKA catalyticsubunits. The peptide sequence of MaPKA1 is only 36.6% aminoacid identical to the class II catalytic subunit PkaB from A. nidulans(Ni et al., 2005).

Southern blotting using the PKA ORF as a probe showed thatMaPKA1 exists as a single copy in the genome of M. anisopliae (datanot shown). RT-PCR showed MaPKA1 was constitutively expressedin conidia and mycelia collected from PDA plates or harvested frominfected insect (data not shown).

3.2. Disrupting PKA greatly reduces the virulence of M. anisopliae tocaterpillars

MaPKA1 was disrupted using homologous recombination. Dis-ruption of MaPKA1 was confirmed by Southern blotting and PCR(data not shown) and the transcript could not be detected withRT-PCR in the disruptants. The phenotype of DMaPKA1 on PDAplates is similar to the WT strain, but with fewer aerial hyphaeso it appeared less ‘‘fluffy”. The dry weight biomass of the WTwas �twice that of DMaPKA1 after three days growth in SDB. Thecomplemented DMaPKA1 was similar to WT both in phenotypeand biomass.

The virulence of DMaPKA1 against G. mellonella was sharply re-duced. Two weeks after inoculation by immersing insects in conid-ial suspensions, both the WT strain and the complementedDMaPKA1 produced >90% mortality following topical infectionwith all four conidial suspensions (1 � 106, 5 � 106, 1 � 107 and2 � 107 conidia ml�1). DMaPKA1 produced 4% and 6% mortalitywith 1 � 107 and 2 � 107 conidia ml�1, respectively, and insectswere apparently unaffected by infection with 1 � 106 and5 � 106 conidia ml�1. Bypassing the cuticle by injection of 3 llWT conidial suspension (104 conidia ml�1) per caterpillar caused100% mortality within 7 days, and mycelium emerged from eachcadaver to conidiate. No mortality was produced by injecting in-sects with conidia of DMaPKA1.

3.3. MaPKA1 and germination behavior

Using glucose as sole carbon source, WT conidia adhered to thesurface of a plastic petri dish and germinated within 16 h. Within

ulates production of virulence determinants ..., Fungal Genet. Biol.

4 W. Fang et al. / Fungal Genetics and Biology xxx (2009) xxx–xxx

ARTICLE IN PRESS

48 h, >90% of hyphal tips had swollen to produce appressoria. Incontrast, conidia of DMaPKA1 germinated without adhesion or for-mation of appressoria, and grew as mycelial pellets suspended inthe medium. Similar growth is shown by the WT in SDB shake cul-tures that prevent adhesion of conidia to the surface of the flask(Fig. 1). In glycerol medium, both WT and DMaPKA1 showed>90% germination by 20 h, and adhered to the plastic. However,3 d after inoculation 85 ± 2.8% of WT hyphae, and only 10 ± 1.1%of DMaPKA1 hyphae had formed appressoria. Both WT and DMaP-KA1 grew more slowly in alanine than in glucose and glycerol. By48 h after inoculation, 80 ± 3.3% of WT conidia had germinatedand produced appressoria. In contrast, 55 ± 1.1% of DMaPKA1 con-idia had germinated and only 30 ± 1.9% of these germlings had pro-duced appressoria.

3.4. MaPKA1 and appressorium formation

Conidia were inoculated into Petri dishes containing 0.0125% YE(reproduces strain ARSEF2575’s growth patterns in vivo) to deter-mine germination and growth rates, and quantify appressoriumformation (St. Leger et al., 1991). Both WT and DMaPKA1 had100% germination rates 12 h after inoculation, and grew at thesame rates before the initiation of appressoria formation. Typically,WT germlings show a sinusoidal growth pattern with the germtube tips changing direction of growth 15 ± 1.9 times before differ-entiating appressoria. (Table 1). By 27 h after inoculation, 93 ± 3.2%of the hyphal tips had swollen (Fig. 2). Some subterminal appresso-ria were observed. Appresorium formation in DMaPKA1 was de-layed, and only 5 ± 1.3% of germlings had begun to differentiateat 27 h post-inoculation (Table 1) and (Fig. 2). The DMaPKA1 germ-lings showed more linear growth than the WT, with �5 changes indirection of growth before differentiation, though 95 ± 1.9% ofDMaPKA1 germlings had produced appressoria 3 days after inocu-lation (Table 1).

Fig. 1. WT and DMaPKA1 grown in medium using glucose as sole carbon resource in a Peof conidial suspension (107 ml�1) was inoculated into 2 ml of medium and grown for 5 dmycelial pellets which are similar to the growth in SDB shaking culture.

Table 1Appressorium formation on plastic surface and on insect cuticle.

Plastic surface

The number of changes in direction of growth precedingappressorium formation

Percent of germliappressorium

Time (h) WT DMaPKA1 WT

14 3 ± 1.2 0 020 15 ± 1.9 0 10% ± 1.527 16 ± 2.1 3 ± 1.3 93% ± 3.272 16 ± 2.8 5 ± 2.1 97% ± 2.9

Please cite this article in press as: Fang, W. et al., Protein kinase A reg(2009), doi:10.1016/j.fgb.2008.12.001

Adhesion and appressorium formation by the WT and DMaPKA1was also studied on cuticles from several host species (house flywings, locust wings and G. mellonella larval cuticles). The attach-ment of DMaPKA1 and the WT was tested on the cuticle of G. mello-nella larvae. Sixteen hours after inoculation, most WT conidia(92 ± 5.4%) had attached to the insect cuticle, as compared to only15 ± 4.9% of the DMaPKA1 conidia. Twenty four hours after inocu-lation, 100% of both WT and DMaPKA1 conidia that had attachedhad germinated, but the WT had produced three times moreappressoria (Table 1). By three days post-inoculation, the DMaP-KA1 mutant had caught up with the WT and rates of differentiationwere similar on housefly, locust and G. mellonella cuticles (Table 1),confirming that the cAMP–PKA pathway is dispensable for appres-sorium formation, but in its absence differentiation is delayed. Thecomplemented DMaPKA1 had similar rates of differentiation ofappressoria as the WT both on the plastic petri dish surface andon insect cuticles (data not shown).

To determine if the appressoria produced byDMaPKA1 are fullyfunctional, we tested their turgor pressure as turgor pressure is re-quired for mechanical penetration of the cuticle (Wang and St. Le-ger, 2007b). The turgor pressure of WT appressoria formed againstinsect cuticle is 10.1 ± 1.1 MPa, as compared to 6.5 ± 0.8 MPa inDMaPKA1. Turgor pressure in appressoria was previously corre-lated to total lipid amount and lipid droplet number (Wang andSt. Leger, 2007b). However, there was no significant difference inthe number of lipid droplets and total lipid between the WT andDMaPKA1 (data not shown). The complemented DMaPKA1 hadsimilar turgor pressure, lipid amount and lipid droplets as the WT.

3.5. MaPKA1 and stress tolerance

Oxidative stress was provided by three different oxidativeagents: H2O2, diamide and methyl-viologen. All three oxidativeagents partially inhibited the germination of WT and DMaPKA1

tri dish with hydrophobic surface. The nitrogen resource is NaNO3. Thirty microliterays. WT (A) grew evenly on the bottom of the Petri dish, while DMaPKA1 (B) formed

Insect cuticle

ng generating Locust wing (percent of germling differentiated intoappressorium)

DMaPKA1 Time (h) WT DMaPKA1

0 24 38% ± 2.1 10% ± 1.40 72 62% ± 3.2 60.7% ± 2.45% ± 1.395% ± 1.9

ulates production of virulence determinants ..., Fungal Genet. Biol.

Fig. 2. Appressorium formation in YE in Petri dishes with a hydrophobic surface. (a) WT (left) and DMaPKA1 (right) grown in YE for 16 h. WT shows a sinusoidal growthpattern and starts to form appressoria, while DMaPKA1 has linear growth. (b) WT (left) and DMaPKA1 (right) grown in YE for 27 h. 93% of WT germlings formed appressoria. Incontrast, only 5% of DMaPKA1 germlings produced appressoria. A: appressorium. C: conidium.

Fig. 3. The tolerance of WT and DMaPKA1 to oxdative stress generated by diamide.WT and DMaPKA1 grown in SDB supplemented with diamide (0.3 mg ml�1) for16 h. WT had around 50% of germination. Structures resembling bubble wereformed in both germlings and ungerminated conidia (indicated by arrow).DMaPKA1 had around 15% of germination without bubble-like structure ingermlings and ungerminated conidia.

W. Fang et al. / Fungal Genetics and Biology xxx (2009) xxx–xxx 5

ARTICLE IN PRESS

in SDB, but disruption of MaPKA1 made M. anisopliae significantlymore sensitive to diamide. Thus, 48.4 ± 2.3% of WT conidia andonly 17.2 ± 3.1% of DMaPKA1 conidia had germinated after 16 hin SDB containing 0.3 mg ml�1 diamide. In the WT hyphae, Struc-tures resembling bubbles were formed, but these were absent inhyphae of DMaPKA1 (Fig. 3). These bubble-like structure couldhave a protective role as the WT germinated more rapidly. The sen-sitivity of DMaPKA1 to methyl-viologen and H2O2 was similar toWT. The complemented DMaPKA1 had similar oxidative stress tol-erance to WT (data not shown).

The addition of KCl to PDA produced no differences in growthbetween WT, DMaPKA1 and the complemented DMaPKA1, suggest-ing that MaPKA1 is not required for osmotic stress tolerance. Like-wise, disruption of MaPKA1 did not increase sensitivity of M.anisopliae to Hygromycin B (up to 2 mg ml�1) or Congo Red (upto 1 mg ml�1).

3.6. MaPKA1. and resistance to cuticular melanin

The sensitivity of DMaPKA1 to melanin produced from cuticulartyrosinase was compared with WT. Conidia of both WT and DMaP-KA1 showed 100% germination rates in a suspension of melaninsupplemented with YE. By 24 h post inoculation, the length of

Please cite this article in press as: Fang, W. et al., Protein kinase A reg(2009), doi:10.1016/j.fgb.2008.12.001

WT hyphae was 42.4 ± 12.4 lm and 53 ± 1.8% of germlings hadformed appressoria. At this time, DMaPKA1 had similar hyphallengths, but only 15 ± 2.1% of germlings had formed appressoria.At 36 h after inoculation, 95 ± 4.1% of WT and 92 ± 3.2% of DMaP-KA1 germlings had produced appressoria. The complementedDMaPKA1 was similar to WT in germination and appressorium for-mation in the melanin solution.

3.7. Identification of putative PKA target genes by transcriptomeanalysis

Disruption of MaPKA1 renders M. anisopliae almost nonpatho-genic, suggesting that it regulates virulence factors. The transcrip-tome of DMaPKA1 was profiled against the WT to identify thesefactors. For this purpose, the WT and deletion mutant were pre-germinated in SDB and an equal biomass of mycelia was trans-ferred to cuticle-containing medium to induce expression of viru-lence determinants. This experimental set-up excluded possibleside effects from differences in germination and growth rates. Ofthe 1730 ESTs arrayed, 244 (14.1%) were down-regulated in DMaP-KA1 compared to the WT, and these were subdivided into 7 of thefunctional groups previously described (Freimoser et al. 2003)(Supplement 2). Of the 244 down-regulated genes, 5.7% were in-volved in stress response, 16.4% in cell structure and function,3.6% in RNA metabolism, 9.8% in protein metabolism, 18.4% in cellmetabolism and 2.0% in energy metabolism. Other significantlydown regulated genes were in the hypothetical and unknown func-tion group. Transposable elements did not show altered regulation.Only one EST corresponding to the subtilisin-like protease, Pr1Dwas up-regulated in DMaPKA1 compared to WT. RT-PCR analysisof 13 randomly selected ESTs was in good accordance with themicroarray data (Table 2).

3.8. Deletion of MaPKA1 down regulates many pathogenicity genes

The virulence genes regulated by MaPKA1 are involved in differ-ent stages of pathogenicity, including appressorium formation,cuticle degradation, and hyphal body formation.

The impact of disrupting MaPKA1 on appressorium formation islikely pleiotropic. Down regulated genes included three sterol syn-

ulates production of virulence determinants ..., Fungal Genet. Biol.

Table 2Fold-change comparison for selected genes between real time RT-PCR and microarray analysis.

Clone name Genbank accession number Fold change for

Microarray RT-PCR

Subtilisin-like protease Pr1I AJ273412 2.34 11.25Subtilisin-like protease Pr1K AJ274144 2.28 10.34Subtilisin-like protease Pr1F AJ251967 1.94 9.56Subtilisin-like protease Pr1E AJ251967 1.94 8.34Subtilisin-like protease PR1C AJ273540 1.80 4.12Subtilisin-like protease Pr1D AJ272861 �2.05 �6.34pH Signal transduction protein PalI AJ273283 2.06 8.96Tetraspanin Tsp3 CN809207 1.83 3.29Related to chitinase AJ274366 2.78 14.15Anaerobically expressed form of translation initiator AJ273702 2.09 10.01Related to oxysterol-binding protein CN808236 2.03 9.87C14 Sterol reductase AJ273501 2.34 10.14Sterol delta 5,6-desaturase ERG3 CN808158 2.15 9.22

Fig. 4. WT and DMaPKA1 grown in alkaline condition. As a representative, thisfigure shows the growth curves at pH 9.0.

6 W. Fang et al. / Fungal Genetics and Biology xxx (2009) xxx–xxx

ARTICLE IN PRESS

thesis genes (CN808158, AJ273149 and AJ273501) and a regulatorygene for sterol synthesis (CN808236). Sterols are essential compo-nents of the fungal cell membrane, influencing membrane perme-ability to solutes, including glycerol. Glycerol metabolismincreases turgor pressure in appressoria (Xu et al., 1996), andCN808700 with high similarity (2e�29) to a glycerol kinase in Rat-tus norvegicus was also down regulated in the deletion mutant.

Appressorial formation involves a dramatic remodeling of thecell skeleton (Xu et al., 1996), and MaPKA1 also regulates keycytoskeletal elements including CN808760 that is highly similar(6e�86) to tubulin-folding cofactor. Hydrolase genes involved inmodifying the cell wall are also regulated by PKA including: 1)AJ273420 similar (8.9e�13) to b-1, 3 exoglucanase precursor; 2)AJ274019 with high similarity (1e�79) to b-1, 3 glucosidase, and3) AJ273400 with high similarity (3e�55) to b -glucanase.CN809207 with high identity (1e�23) to tetraspanin-like proteinin M. grisea was down-regulated in DMaPKA1. The tetraspanin-like protein is involved in penetration peg formation in M. grisea(Clergeot et al., 2001).

M. anisopliae uses multiple hydrolases to degrade insect cuticle.A chitinase (AJ274366), a chymotrypsin and five subtilisin-like pro-teases (Pr1C, Pr1E, Pr1F, Pr1I and Pr1K), were down regulated inDMaPKA1. But the subtilisin-like protease Pr1D was up-regulated.Intracellular degradation of proteins may also be regulated byMaPKA1 as CN809052 with high similarity (2e�79) to a proteasomecomponent precursor from S. pombe is down regulated in DMaP-KA1. Other nutrient acquisition processes are up-regulated byMaPKA1. Several transporter proteins including two sugar trans-porters (AJ273923 and CN809107), an oligopeptide transporter(AJ273118) and a carboxylic acid transport protein (AJ273226),were down regulated in DMaPKA1. Taken together, it is clear thata primary function of the cAMP–PKA signaling pathway is a reorga-nization of uptake and assimilative functions to utilize the alteredspectrum of nutrients available in insect cuticle.

PKA is also required for up-regulation of several genes encodingproteins that confer resistance to antifungals, including CN808204that is highly similar (5e�38) to squalene epoxidase 1 andCN808298, with high identity (2e�74) to phenol 2-monooxygenase.These could detoxify cuticular components including phenolicssuch as melanin.

Several genes that are highly expressed by M. anisopliae inhemolymph (Wang et al., 2005b), were down-regulated in DMaP-KA1 including CN808348 encoding a catalase, which detoxifies per-oxide and AJ273139 with high similarity to an acid phosphataseinvolved in utilization of organic phosphate in hemolymph, andpossibly in weakening the host immune response (Xia et al.,2001). Genes for phosphate assimilation and catalase are also con-trolled by PKA in the basidiomycete C. neoformans (Hu et al., 2007).

Please cite this article in press as: Fang, W. et al., Protein kinase A reg(2009), doi:10.1016/j.fgb.2008.12.001

AJ273702 with high similarity (1e�51) to an anaerobically ex-pressed form of a translation initiator was down regulated inDMaPKA1. This potentially could interfere with the pathogen’s re-sponse to hypoxia in the hemolymph.

3.9. MaPKA1 and pH regulation

An EST (AJ273283) with similarity (5e�13) to the pH signaltransduction protein PalI (AFUB-041590) in Aspergillus fumigatuswas down regulated in DMaPKA1, suggesting that the cAMP–PKApathway is involved in responding to environmental pH. To con-firm this, the growth of DMaPKA1 was tested on PDA at pH valuesranging from pH 4.0 to pH 10.0. At pH values below 8.0, there wasno significant difference in growth between WT and DMaPKA1,with both the WT and DMaPKA1 being slightly inhibited at pH val-ues below 5.0. Growth of WT and DMaPKA1 was also inhibited atpH values above 8, but the extent of DMaPKA1 inhibition wasmuch greater than WT (Fig. 4). Therefore, MaPKA1 is involved inadapting M. anisopliae to alkaline environments.

3.10. MaPKA1 and growth

The microarray results are consistent with the reduced growthrate of DMaPKA1. Housekeeping genes down regulated in DMaP-KA1 included seven translation factors (AJ272931, CN809118,CN809110, CN808160, CN808711, AJ273876 and AJ273115), two

ulates production of virulence determinants ..., Fungal Genet. Biol.

Fig. 5. Proposed model for MaPKA1-mediated signaling in M. anisopliae.

W. Fang et al. / Fungal Genetics and Biology xxx (2009) xxx–xxx 7

ARTICLE IN PRESS

transcription factors (AJ274214 and AJ274125), five genes relatedto transportation of nutritions (AJ273093, AJ273923, AJ273223,AJ273849 and CN809107), two senescence-associated proteins(AJ273870 and AJ272904), a protein involved in autophagy(AJ273516), three proteins related to genome replication(CN809658, AJ273275 and AJ272826) and a spindle assemblycheckpoint protein (CN808886).

3.11. Crosstalk of cAMP–PKA pathway with other signal pathways

Adenylate cyclase synthesizes cAMP in response to signalstransferred from G proteins. The binding of cAMP to the PKA com-plex dissociates the regulator and catalytic subunits. Several up-stream components of MaPKA1 in the signal pathway were downregulated in DMaPKA1, including the adenylate cyclase gene(AJ251971), a GTPase-activator protein for Rho-like GTPase(AJ272954) and a GTPase activating protein homolog (AJ273268).This suggests that MaPKA1 or its targets feedback to control cAMPsynthesis. Likewise, crosstalk between the cAMP–PKA and MAPK(Mitogen-activated protein kinase) pathways is suggested by thedown regulation of MAPKK (AJ273356) in DMaPKA1.

4. Discussion

PKA is a central enzyme of cAMP signaling. Previously, weused exogenously applied cAMP and PKA inhibitors to implicatecAMP–PKA signaling in the differentiation of appressoria by ger-minating conidia (St. Leger et al., 1990). In this study, we furthercharacterized the role of the cAMP–PKA pathway by disruptionof a catalytic subunit of PKA. DMaPKA1 is almost avirulent andthis presumably results from the pleiotropic effect of the manygrowth functions controlled by PKA in its role as a master regu-lator. These include reduced growth rate and reduced produc-tion of cuticle degrading enzymes (Fig. 5). Similar to what isseen in the plant pathogen M. grisea (Xu et al., 1996), disruptionof PKA in M. anisopliae delayed appressorial differentiation andthe appressoria were dysfunctional. How the disruption of PKAimpairs appressorial function in M. grisea has not been deter-mined, but it reduces turgor pressure in the appressoria ofDMaPKA1 and this will impair the generation of mechanicalpressure (Wang et al., 2007b).

Please cite this article in press as: Fang, W. et al., Protein kinase A reg(2009), doi:10.1016/j.fgb.2008.12.001

The appressoria of DMaPKA1 may also be defective in penetra-tion peg formation, as tetraspanin is down regulated in DMaPKA1and tetraspanin is required for peg formation in M. grisea (Clergeotet al., 2001). The mode of regulation and signal pathways involvedin regulation of tetraspanin has not been determined in M. grisea.The MAPK and cAMP–PKA pathways of M. grisea coordinate to reg-ulate appressoria formation and pathogenic growth (Xu et al.,1996). An EST corresponding to MAPKK (MAPK kinase) wasdown-regulated when DMaPKA1 was inoculated into insect cuticlemedium. It is up-regulated in the WT (Wang et al., 2005b), indicat-ing that PKA cross talks with the MAPK pathway to regulate path-ogenesis in M. anisopliae as well as in M. grisea. This suggests thatthe signal transduction pathways regulating appressoria formationand pathogenic growth are conserved in plant and insectpathogens.

No one oxidant reproduces all kinds of oxidative stress, sothree different agents were used to cover the range of oxidativestresses M. anisopliae might be exposed to when challenged withthe insect immune system. Hydrogen peroxide (H2O2) is highlyreactive with a broad spectrum of targets. Methyl-viologen is aredox cycling agent that generates superoxide by reducingmolecular oxygen at the expense of NADPH in an aerobicallygrowing cell. Diamide produces oxidative stress indirectly byoxidizing glutathione and sulfhydryl groups on proteins. Onlydiamide delayed germination of DMaPKA1 compared to theWT. Therefore, MaPKA1 seems to be selectively involved inresisting the effects of oxidant(s) which target the thiol groupon proteins. In contrast to M. anisopliae, the disruption of com-ponents of the cAMP–PKA pathway in A. fumigatus increasedsusceptibility to multiple oxidants (Zhao et al., 2006). In M. ani-sopliae, as in Saccharomyces cerevisiae, cells may have distinctmechanisms to protect against different reactive oxygen species(Thorpe et al., 2004).

M. anisopliae is able to grow over a wide range of pH values.This is important for the ability of conidia and mycelia to storeglycerol and trehalose, and for the shelf life of M. anisopliae asa mycoinsecticide (Hallsworth and Magan, 1996). The ability ofM. anisopliae to alter environment pH allows it to regulate theexpression of some virulence genes to the pH at which theyare most active, e.g., the cuticle degrading subtilisin-like prote-ases are only produced at alkaline pH (St. Leger et al., 1999).

ulates production of virulence determinants ..., Fungal Genet. Biol.

8 W. Fang et al. / Fungal Genetics and Biology xxx (2009) xxx–xxx

ARTICLE IN PRESS

However, the mechanism by which M. anisopliae responds toenvironmental pH has not been identified. In this study, the dis-ruption of MaPKA1 reduced the ability of M. anisopliae to grow inan alkaline environment (pH > 8). The microarray data suggeststhat the cAMP–PKA pathway regulates the pH response of M.anisopliae by controlling the expression level of a gene(AJ273283) with similarity (4e�13) to the A. fumigatus pH sensorPalI.

PKA regulates production of phospholipids, but not sterolsynthesis in C. neoformans (Hu et al., 2007). However, of sixdown regulated genes in DMaPKA1 involved in lipid metabolism,three are involved in ergosterol synthesis and one in regulatingthe pathway. Ergosterol influences the permeability of glycerolunder hypoosmotic stress (Toh et al., 2001) and glycerol metab-olism is important for turgor pressure in appressorium in M. gri-sea (Xu et al., 1996) and M. anisopliae (Wang and St. Leger,2007b). Thus, dysfunctional ergosterol synthesis in DMaPKA1may contribute to the reduced turgor pressure in appressoria.

Our previous microarray analysis of gene expression by WTARSEF2575 on locust cuticle identified �700 differentially ex-pressed genes which are considered directly or indirectly to be in-volved in the pathogenesis of M. anisopliae (Freimoser et al., 2005;Wang et al., 2005b). Out of the 244 down regulated genes inDMaPKA1, 237 were up-regulated by the WT when grown in lo-cust medium i.e., PKA regulates �34% of the genes induced in re-sponse to cuticle. Only 85 genes are down regulated in C. albicansdisrupted in adenylate cyclase (Harcus et al., 2004). Many of thePKA-regulated genes in M. anisopliae are for secreted productssuggesting a major role for PKA in regulating expression of se-creted products in response to host signaling and nutrient sens-ing. In contrast, only nine genes for cell wall, cell surface andextracellular proteins were down regulated in PKA disputant mu-tants of C. neoformans (Hu et al., 2007). Of proteins secreted by M.anisopliae, only Pr1D was up-regulated by DMaPKA1 to a greaterextent than the WT on locust cuticle. Pr1D is one of 11 subtilasesexpressed by M. anisopliae and five were up-regulated by PKA.These proteases differ in their primary and secondary substratespecificities, adsorption properties to cuticle and alkaline stability,indicative of functional differences (Bagga et al., 2004). This studyshows that they also differ in their mode of regulation which pre-sumably decreases the pathogens dependence on a single regula-tory system, and permits the pathogen to respond to differentconditions.

Notably absent from the genes up-regulated by PKA in M. ani-sopliae were any involved in toxin production such as the polyke-tide synthases, although these are regulated by PKA in A.fumigatus (Grosse et al., 2008). Conversely, PKA up-regulated stressresponse genes such as heat shock proteins and senescence-associ-ated protein in M. anisopliae (supplemental 1). These were alsoidentified as being regulated by PKA in C. neoformans, but not inA. fumigatus. Similar to A. fumigatus (Grosse et al., 2008), and otherfungi, the PKA of M. anisopliae was required for up-regulation ofsome genes required for protein synthesis such as translation elon-gation factors, and for genes involved in cytoskeleton assembly andcell cycle regulation. These results suggest that PKA will have bothconserved and functionally diverse roles in different fungi com-mensurate with their different lifestyles. Further, a large percent-age of the genes up-regulated by PKA in M. anisopliae have noknown homologs, suggesting that the organism has undergonesubstantial evolutionary adaptations to the pathogenic lifestyleand PKA has been co-opted to regulate these ‘‘Metarhizium-spe-cific” processes.

Acknowledgment

This work was supported by NSF Grant MCB-0542904.

Please cite this article in press as: Fang, W. et al., Protein kinase A reg(2009), doi:10.1016/j.fgb.2008.12.001

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.fgb.2008.12.001.

References

Adachi, K., Hamer, J.E., 1998. Divergent cAMP signaling pathways regulate growthand pathogenesis in the rice blast fungus Magnaporthe grisea. The Plant Cell 10,1361–1373.

Bagga, S., Hu, G., Screen, S.E., St. Leger, R.J., 2004. Reconstructing the diversificationof subtilisins in the pathogenic fungus Metarhizium anisopliae. Gene 324, 159–169.

Clergeot, P.H., Gourgues, M., Cots, J., Laurans, F., Latorse, M.P., Pepin, R., Tharreau, D.,Notteghem, J.L., Lebrun, M.H., 2001. PLS1, a gene encoding a tetraspanin-likeprotein, is required for penetration of rice leaf by the fungal pathogenMagnaporthe grisea. Proc. Natl. Acad. Sci. USA 98, 6963–6968.

Covert, S.F., Kapoor, P., Lee, M.H., Briley, A., Nairn, C.J., 2001. Agrobacteriumtumefaciens-mediated transformation of Fusarium circinatum. Mycol. Res. 105,259–264.

D’Souza, C., Alspaugh, J.A., Yue, C., Harashima, T., Cox, G.M., Perfect, J.R., Heitman, J.,2001. Cyclic AMP-dependent protein kinase controls virulence of the fungalpathogen Cryptococcus neoformans. Mol. Cell Biol. 21, 3179–3191.

Dürrenberger, F., Wong, K., Kronstad, W.J., 1998. Identification of a cAMP-dependent protein kinase catalytic subunit required for virulence andmorphogenesis in Ustilago maydis. Proc. Natl. Acad. Sci. USA 95, 5684–5689.

Fang, W., Pei, Y., Bidochka, M.J., 2006. Transformation of Metarhizium anisopliaemediated by Agrobacterium tumefaciens. Can. J. Microbiol. 52, 623–626.

Fang, W., Bidochka, M.J., 2006. Expression of genes involved in germination,conidiogenesis and pathogenesis in Metarhizium anisopliae using quantitativereal-time RT-PCR. Mycol. Res. 110, 1165–1171.

Fang, W., Pei, Y., Bidochka, M.J., 2007. A regulator of a G protein signalling (RGS)gene, cag8, from the insect-pathogenic fungus Metarhizium anisopliae isinvolved in conidiation, virulence and hydrophobin synthesis. Microbiology153, 1017–1025.

Fang, W., Scully, L., Zhang, L., Pei, Y., Bidochka, M.J., 2008. Implication of a regulatorof G protein signalling (BbRGS1) in conidiation and conidial thermotolerance ofthe insect pathogenic fungus Beauveria bassiana. FEMS. Microbiol. Lett. 279,146–156.

Fillinger, S., Chaveroche, M.K., Shimizu, K., Keller, N.P., d’Enfert, C., 2002. cAMP andras signalling independently control spore germination in the filamentousfungus Aspergillus nidulans. Mol. Microbiol. 44, 1001–1016.

Freimoser, F.M., Hu, G., St. Leger, R.J., 2005. Variation in gene expression patterns asthe insect pathogen Metarhizium anisopliae adapts to different host cuticles ornutrient deprivation in vitro. Microbiology 151, 361–371.

Grosse, C., Heinekamp, T., Kniemeyer, O., Gehrke, A., Brakhage, A.A., 2008. Proteinkinase A regulates growth, sporulation, and pigment formation in Aspergillusfumigatus. Appl. Environ. Microbiol. 74, 4923–4933.

Hallsworth, J.E., Magan, N., 1996. Culture age, temperature, and pH affect the polyoland trehalose contents of fungal propagules. Appl. Environ. Microbiol. 62,2435–2442.

Harcus, D., Nantel, A., Marcil, A., Rigby, T., Whiteway, M., 2004. Transcriptionprofiling of cyclic AMP signaling in Candida albicans. Mol. Biol Cell 15, 4490–4499.

Gold, S., Duncan, G., Barrett, K., Kronstad, J., 1994. cAMP regulates morphogenesis inthe fungal pathogen Ustilago maydis. Genes Dev. 8, 2805–2816.

Hu, G., Steen, B.R., Lian, T., Sham, A.P., Tam, n., Tangen, K.L., Kronstad, J.W., 2007.Transcriptional Regulation by Protein Kinase A in Cryptococcus neoformans. PLoSPathogens 3, e42.

Larraya, L.M., Boyce, K.J., So, A., Steen, B.R., Jones, S., Marra, M., Kronstad, J.W., 2005.Serial analysis of gene expression reveals conserved links between proteinkinase a, ribosome biogenesis, and phosphate metabolism in Ustilago maydis.Eukaryot. Cell 4, 2029–2043.

Mitchell, T.K., Dean, R.A., 1995. The cAMP-dependent protein kinase catalyticsubunit is required for appressorium formation and pathogenesis by the riceblast pathogen Magnaporthe grisea. The Plant Cell 7, 1869–1878.

Ni, M., Rierson, S., Seo, J., Yu, J., 2005. The pkaB gene encoding the secondary proteinkinase a catalytic subunit has a synthetic lethal interaction with pkaA and playsoverlapping and opposite roles in Aspergillus nidulans. Eukaryot. Cell 4, 1465–1476.

Pall, M.L., Brunelli, J.P., 1993. A series of six compact fungal transformation vectorscontaining polylinkers with multiple unique restriction sites. Fungal Genet.Newslett. 40, 59–62.

St. Leger, R.J., Charnley, A.K., Cooper, R.M., 1986. Cuticle-degrading enzymes ofentomopathogenic fungi: cuticle degradation in vitro. J. Invertebr. Pathol. 47,167–177.

St. Leger, R.J., Cooper, R.M., Charnley, A.K., 1988. The effect of Melanization ofManduca Sexta cuticle on growth and infection by Metarhizium anisopliae. J.Invertebr. Pathol. 52, 459–470.

St. Leger, R.J., Butt, T.M., Staples, R.C., Roberts, D.W., 1990. Second messengerinvolvement in differentiation of the entomopathogenic fungus Metarhiziumanisopliae. J. General Microbiol. 136, 1779–1789.

ulates production of virulence determinants ..., Fungal Genet. Biol.

W. Fang et al. / Fungal Genetics and Biology xxx (2009) xxx–xxx 9

ARTICLE IN PRESS

St. Leger, R.J., Roberts, D.W., Staples, R.C., 1991. A model to explain differentiation ofappressoria by germlings of Metarhizium anisopliae. J. Invert. Pathol. 57, 299–310.

St. Leger, R.J., Staples, R.C., Roberts, D.W., 1992. Cloning and regulatoryanalysis of starvation–stress gene, ssgA, encoding a hydrophobin-likeprotein from the entomopathogenic fungus, Metarhizium anisopliae. Gene120, 119–124.

St. Leger, R.J., Bidochka, M.J., Roberts, D.W., 1994. Germination triggers ofMetarhizium anisopliae conidia are related to host species. Microbiology 140,1651–1660.

St. Leger, R.J., Joshi, L., Bidochka, M.J., Roberts, D.W., 1996. Construction of animproved mycoinsecticide overexpressing a toxic protease. Proc. Natl. Acad. Sci.USA 93, 6349–6354.

St. Leger, R.J., Nelson, J.O., Screen, S.E., 1999. The entomopathogenic fungusMetarhizium anisopliae alters ambient pH, allowing extracellular proteaseproduction and activity. Microbiology 145, 2691–2699.

Toda, T., Cameron, S., Sass, P.S., Zoller, M.P., Wigler, M., 1987. Three different genesin S. cerevisiae encode the catalytic subunits of the cAMP-dependent proteinkinase. Cell 50, 277–287.

Toh, T.H., Kayingo, G., van der Merwe, M.J., Kilian, S.G., Hallsworth, J.E., Hohmann, S.,Prior, B.A., 2001. Implications of FPS1 deletion and membrane ergosterolcontent for glycerol efflux from Saccharomyces cerevisiae. FEMS. Yeast Res. 1,205–211.

Thorpe, G.W., Fong, C.S., Alic, N., Higgins, V.J., Dawes, I.W., 2004. Cells havedistinct mechanisms to maintain protection against different reactiveoxygen species: oxidative-stress-response genes. Proc. Natl. Acad. Sci. USA101, 6564–6569.

Vandesompele, J., De Preter, K., Pattyn, F., Poppe, B., Van Roy, N., De Paepe, A.,Speleman, F., 2002. Accurate normalization of real-time quantitative RT-PCRdata by geometric averaging of multiple internal control genes. Genome Biology

Please cite this article in press as: Fang, W. et al., Protein kinase A reg(2009), doi:10.1016/j.fgb.2008.12.001

3. <http://genomebiology.com/2002/3/7/research/0034research0034.1-0034.11>.

Wang, C., St. Leger, R.J., 2005. Developmental and transcriptional responses to hostand nonhost cuticles by the specific locust pathogen Metarhizium anisopliae varacridum. Eukaryot. Cell 4, 937–947.

Wang, C., Hu, G., St. Leger, R.J., 2005. Differential gene expression by Metarhiziumanisopliae growing in root exudate and host (Manduca sexta) cuticle orhemolymph reveals mechanisms of physiological adaptation. Fungal. Genet.Biol. 42, 704–718.

Wang, C., St. Leger, R.J., 2006. A collagenous protective coat enables Metarhiziumanisopliae to evade insect immune responses. Proc. Natl. Acad. Sci. USA 103,6647–6652.

Wang, C., St. Leger, R.J., 2007a. The MAD1 adhesin of Metarhizium anisopliae linksadhesion with blastospore production and virulence to insects, and the MAD2adhesin enables attachment to plants. Eukaryot. Cell 6, 808–816.

Wang, C., St. Leger, R.J., 2007b. The Metarhizium anisopliae perilipin homolog MPL1regulates lipid metabolism, appressorial turgor pressure, and virulence. J. Biol.Chem. 282, 21110–21115.

Wang, C., Duan, Z., St. Leger, R.J., 2008. MOS1 osmosensor of Metarhizium anisopliaeis required for adaptation to insect host hemolymph. Eukaryot. Cell. 7, 302–309.

Xia, Y., Clarkson, J.M., Charnley, A.K., 2001. Acid phosphatases of Metarhiziumanisopliae during infection of the tobacco hornworm Manduca sexta. Arch.Microbiol. 176, 427–434.

Xu, J.R., Hamer, J.E., 1996. MAP kinase and cAMP signaling regulate infectionstructure formation and pathogenic growth in the rice blast fungusMagnaporthe grisea. Genes Dev. 10, 2696–2706.

Zhao, W., Panepinto, J.C., Fortwendel, J.R., Fox, L., Oliver, B.G., Askew, D.S., Rhodes,J.C., 2006. Deletion of the regulatory subunit of protein kinase A in Aspergillusfumigatus alters morphology, sensitivity to oxidative damage, and virulence.Infect. Immun. 74, 4865–4874.

ulates production of virulence determinants ..., Fungal Genet. Biol.

10 W. Fang et al. / Fungal Genetics and Biology xxx (2009) xxx–xxx

ARTICLE IN PRESS

Please cite this article in press as: Fang, W. et al., Protein kinase A regulates production of virulence determinants ..., Fungal Genet. Biol.(2009), doi:10.1016/j.fgb.2008.12.001