EUKARYOTIC CELL, Aug. 2005, p. 1465–1476 Vol. 4, No. 81535-9778/05/$08.00�0 doi:10.1128/EC.4.8.1465–1476.2005Copyright © 2005, American Society for Microbiology. All Rights Reserved.

The pkaB Gene Encoding the Secondary Protein Kinase A CatalyticSubunit Has a Synthetic Lethal Interaction with pkaA and Plays

Overlapping and Opposite Roles in Aspergillus nidulansMin Ni,1,3 Sara Rierson,1† Jeong-Ah Seo,1 and Jae-Hyuk Yu1,2*

Department of Food Microbiology and Toxicology & Food Research Institute,1 Molecular and EnvironmentalToxicology Center,2 and Department of Genetics,3 University of Wisconsin—Madison,

1925 Willow Drive, Madison, Wisconsin 53706

Received 12 April 2005/Accepted 8 June 2005

Filamentous fungal genomes contain two distantly related cyclic AMP-dependent protein kinase A catalyticsubunits (PKAs), but only one PKA is found to play a principal role. In Aspergillus nidulans, PkaA is theprimary PKA that positively functions in vegetative growth and spore germination but negatively controlsasexual sporulation and production of the mycotoxin sterigmatocystin. In this report, we present the identi-fication and characterization of pkaB, encoding the secondary PKA in A. nidulans. Although deletion of pkaBalone does not cause any apparent phenotypic changes, the absence of both pkaB and pkaA is lethal, indicatingthat PkaB and PkaA are essential for viability of A. nidulans. Overexpression of pkaB enhances hyphalproliferation and rescues the growth defects caused by �pkaA, indicating that PkaB plays a role in vegetativegrowth signaling. However, unlike �pkaA, deletion of pkaB does not suppress the fluffy-autolytic phenotyperesulting from �flbA. While upregulation of pkaB rescues the defects of spore germination resulting from�pkaA in the presence of glucose, overexpression of pkaB delays spore germination. Furthermore, upregulationof pkaB completely abolishes spore germination on medium lacking a carbon source. In addition, upregulationof pkaB enhances the level of submerged sporulation caused by �pkaA and reduces hyphal tolerance tooxidative stress. In conclusion, PkaB is the secondary PKA that has a synthetic lethal interaction with PkaA,and it plays an overlapping role in vegetative growth and spore germination in the presence of glucose but anopposite role in regulating asexual sporulation, germination in the absence of a carbon source, and oxidativestress responses in A. nidulans.

In eukaryotic cells, protein kinases constitute one of thelargest gene families and play a pivotal role in transducingexternal and/or internal signals to elicit appropriate cellularresponses. Among these kinases, cyclic AMP (cAMP)-depen-dent protein kinase A (PKA) serves as a prototypical example.In the absence of cAMP, the PKA holoenzyme exists as aninactive equimolar hetero-tetramer composed of a homo-dimeric regulatory subunit and two associated catalytic sub-units (reviewed in references 36 and 38). The regulatory sub-unit also functions in preventing the inactive PKA holoenzymefrom entering the nucleus (6, 11). The cooperative binding oftwo cAMP molecules to each regulatory subunit of the enzymeresults in dissociation of the active PKAs from the regulatorysubunits. These active PKAs can phosphorylate downstreamtarget proteins at serine or threonine residues (reviewed inreference 38). PKA signaling plays a central role in vegetativegrowth, development, mating, stress response, and pathogenic-ity in various fungi (reviewed in references 8, 21, and 22).

In the budding yeast Saccharomyces cerevisiae, three TPKgenes encode the closely related PKA isoforms that shareredundant and distinct functions in viability (39) and inpseudohyphal morphogenesis, respectively (26, 28). Similarly,

two PKA isoforms play redundant and distinct roles in Candidaalbicans, i.e., PKAs are essential for (normal) growth and stressresponse while playing distinct roles in hyphal morphogenesisand invasive growth (3). In the fission yeast Schizosaccharomy-ces pombe, one PKA, Pka1, has been characterized and shownto function in spore germination, growth, and sexual develop-ment (15, 24).

Unlike yeasts, two distantly related PKAs exist in most (ifnot all) filamentous or dimorphic fungi, but only one PKA hasbeen found to play a predominant role. For instance, in theplant pathogenic fungus Ustilago maydis, while Adr1 (class IPKA) was found to be crucial for development and pathoge-nicity, deletion of Uka1 (class II PKA) resulted in no detect-able changes (9) (see Fig. 1D, below). Likewise, one PKA hasbeen shown to play a primary role in normal growth, develop-ment, and pathogenicity in Magnaporthe grisea (1, 25, 43), Cryp-tococcus neoformans (7, 17), and Aspergillus fumigatus (23).

In the model filamentous fungus Aspergillus nidulans, PkaAwas shown to play a central role in various biological processes(10, 34). Deletion of pkaA resulted in delayed germination ofconidiospores (asexual spores), highly restricted growth, andelevated asexual sporulation (conidiation) and production ofthe mycotoxin sterigmatocystin (ST). Additionally, genetic ep-istatic analysis showed that PkaA functions downstream ofFadA, the G� subunit for a heterotrimeric G protein thatprimarily mediates vegetative growth signaling (34, 44). ThisFadA-mediated signaling is negatively controlled by FlbA, anRGS protein, which is necessary for the commencement of

* Corresponding author. Mailing address: Department of Food Mi-crobiology and Toxicology, 1925 Willow Drive, Madison, WI 53706-1187. Phone: (608) 262-4696. Fax: (608) 263-1114. E-mail: jyu1@wisc.edu.

† Present address: USDA-ARS, U.S. Dairy Forage Research Cen-ter, 1925 Linden Drive West, Madison, WI 53706.

1465

on March 18, 2021 by guest

http://ec.asm.org/

Dow

nloaded from

development and ST production (16, 20, 44). Both the loss-of-function flbA and the constitutively active fadA [e.g.,fadA(G42R) or fadA(Q204L)] mutations result in the absenceof development and ST production as well as hyphal disinte-gration (autolysis) due to uncontrolled activation of vegetativegrowth signaling (16, 20, 44, 45). Deletion of pkaA partiallyrestores conidiation in these proliferative mutants while retain-ing restricted growth (34), indicating that vegetative growth isprimarily mediated via FadA and PkaA. However, deletion ofpkaA does not completely eliminate PKA activity in A. nidu-lans (34), suggesting that multiple PKA catalytic subunits mayexist in A. nidulans. In addition to PkaA, a homolog of the S.cerevisiae Sch9 protein named SchA has been studied in A.nidulans (10). Deletion of schA alone caused no apparentphenotypic changes, but the combination of null mutations inboth pkaA and schA delayed conidiospore germination evenfurther (10). A recent study showed that TPKs and Sch9 func-tion in separate signaling cascades in S. cerevisiae (31), sug-gesting SchA and PkaA may function in distinct pathways in A.nidulans.

In this study, we have identified and characterized the pkaBgene encoding the secondary PKA in A. nidulans. PkaB isdistantly related to PkaA and is most similar to other PKAspreviously characterized as class II (see Fig. 1D, below). Al-though deletion of pkaB alone causes no apparent phenotypicchanges in growth, development, spore germination, ST pro-duction, or stress responses, the absence of both PkaB andPkaA is lethal, indicating that PkaB and PkaA constitute thesole PKA catalytic subunits that are essential for the viability ofA. nidulans. We further demonstrate that while PkaB andPkaA have overlapping functions in stimulating vegetativegrowth and spore germination in the presence of glucose, thetwo PKAs play opposite roles in regulating spore germinationin the absence of an external carbon source, oxidative stressresponses, and asexual sporulation in submerged culture.

MATERIALS AND METHODS

Aspergillus nidulans strains, growth conditions, and genetics. Strains used inthis study are listed in Table 1. Standard culture and genetics techniques were

used (27). Fungal strains were grown on minimal solid or liquid medium withsupplements (simplified as MM) as described previously (18). To examine thelevels of colony growth, all strains were inoculated on solid MM with variouscarbon sources and incubated at 37°C for 3 days (MMG, with 1% glucose) or 6days (MMT, with 100 mM threonine as a sole carbon source to induce overex-pression of pkaB). To check the development of conidiophores in liquid sub-merged culture, liquid MMG was inoculated with the conidia of individualstrains (108 spores/100 ml) and incubated at 37°C, 250 rpm. To examine the effectof overexpression or upregulation of pkaB by an ectopic copy of pkaB under thealcA promoter, relevant strains were grown in liquid MMG at 37°C, 250 rpm for14 h. Subsequently, mycelia were collected, rinsed with liquid medium without acarbon source, divided into two equal parts, and transferred into liquid MMG orliquid MMT. Then, the cultures were shaken under the same conditions de-scribed above, and development of wild-type and mutant strains was examinedunder a microscope at 0, 6, and 12 h posttransfer.

To examine the pkaB mRNA levels in the wild type, samples from liquidsubmerged and postdevelopmental induction cultures were collected as de-scribed elsewhere (12). Briefly, 5 � 107 conidia of FGSC4 were inoculated in 100ml of liquid MMG with 0.1% yeast extract in 250-ml flasks and incubated at 37°C,250 rpm. For the vegetative growth phase, samples were collected from liquidsubmerged cultures at designated time points, squeeze dried, and stored at�80°C until subjected to total RNA isolation. For sexual and asexual develop-mental induction, 18-h vegetatively grown mycelia were transferred to solidMMG and the plates were either air exposed for asexual developmental induc-tion or tightly sealed and shielded from light to induce sexual development.

Identification and structure of pkaB. The pkaB gene was identified throughtblastn search of the A. nidulans genome (The Broad Institute [http://www.genome.wi.mit.edu/annotation/fungi/aspergillus/index.html]) using Tpk1, Tpk2, and Tpk3of S. cerevisiae. To examine the positions of introns, reverse transcription-PCR,followed by sequencing analyses of pkaB, was carried out. The Smart RACE kit(Clontech) was used to determine the 5� and 3� ends of the pkaB transcript.

Construction of pkaB deletion and pkaB overexpression strains. The double-joint PCR method was used to generate the pkaB deletion mutant (46). Adeletion construct containing argB with approximately 1 kb each of the 5� and 3�flanking regions of pkaB (amplified with primer pairs of OKH56 with -57 andOKH58 with -59 (Table 2) was introduced into the recipient strain, PW1 (biA1argB2 methG1 veA1). The resulting transformants were randomly screened fordeletion of pkaB as described previously (46). The pkaB overexpression constructwas created by fusing the alcA promoter with the pkaB open reading frame(ORF). The alcA promoter was amplified by OJA106 and OJA107. The pkaBORF with an alcA(p) tail was amplified by OSR16 and OSR18. The two DNAamplicons were fused as described elsewhere (46), and the fusion product wasamplified by OJA108 and OSR17. The resulting alcA(p)::pkaB amplicon was cutwith BamHI and ligated into the BamHI-cut pSH96 plasmid (42). The sequence-verified alcA(p)::pkaB plasmid was introduced into FGSC237, and the single-copy integration of plasmid into the trpC locus was confirmed by a series of PCRanalyses. Three alcA(p)::pkaB transformants were isolated and sexually crossed

TABLE 1. A. nidulans strains used in this study

Strain Genotypea Source or reference

FGSC4 veA� FGSCb

FGSC26 biA1 FGSCFGSC237 pabaA1 yA2 trpC801 FGSCPW1 biA1 argB2 methG1 P. WeglenskiTKIS18.11 pabaA1 yA2 �pkaA::argB� �argB::trpC� trpC801 34TKIS20.1 pabaA1 yA2 alcA(p)::pkaA::trpC� 34TSR1 biA1 argB2 �pkaB::argB� methG1 This studyTSR53 biA1 argB2 �pkaB::argB� methG1 This studyRSA53.126 pabaA1 yA2 �pkaA::argB� �argB::trpC� trpC801 This studyRSA53.130 biA1 �pkaA::argB� �argB::trpC� trpC801 This studyRSA53.111 pabaA1 biA1 argB2 �pkaB::argB� This studyRSA53.143 biA1 argB2 �pkaB::argB� This studyRSA37.1C.12 biA1 �pkaA::argB� �argB::trpC� �pkaB::argB� alcA(p)::pkaA::trpC� This studyRNI5.1, -2, -3c biA1 alcA(p)::pkaB::trpC� This studyRNI6 pabaA1 yA2 �pkaA::argB� alcA(p)::pkaB::trpC� This studyRNI7 pabaA1 yA2 �pkaA::argB� �pkaB::argB� alcA(p)::pkaB::trpC� This studyRNI8 biA1 alcA(p)::pkaA::trpC� This study

a All strains carry the veA1 mutation except FGSC4.b FGSC, Fungal Genetics Stock Center.c Multiple isogenic alcA(p)::pkaB strains. RNI5.1, RNI5.2, and RNI5.3 behaved identically.

1466 NI ET AL. EUKARYOT. CELL

on March 18, 2021 by guest

http://ec.asm.org/

Dow

nloaded from

with �pkaA (RSA53.130) or �pkaA �pkaB alcA(p)::pkaA (RSA37.1C.12) strains.From these crosses, alcA(p)::pkaB (RNI5.3), �pkaA alcA(p)::pkaB (RNI6), and�pkaA �pkaB alcA(p)::pkaB (RNI7) strains were isolated and used in subse-quent overexpression studies.

Nucleic acid isolation and manipulation. Genomic DNA and total RNAisolation and Northern blot analyses were carried out as previously described (12,32). The DNA probes used to examine the mRNA levels of pkaB and brlA wereprepared by PCR amplification of the coding regions of individual genes usinggenomic DNA of FGSC4 as the template with appropriate oligonucleotide pairs(Table 2).

The pkaB mRNA levels in wild-type (FGSC26), �pkaB (RSA53.143), �pkaA(RSA53.130), alcA(p)::pkaB (RNI5.3), �pkaA alcA(p)::pkaB (RNI6), and �pkaA�pkaB alcA(p)::pkaB (RNI7) strains were examined as follows: (i) for conidia,total RNA was isolated from the freshly collected conidia of individual strainsgrown on solid MMG at 37°C for 5 days. (ii) In liquid medium without a carbonsource, approximately 109 spores of each strain were inoculated in 100 ml ofliquid medium without a carbon source and further incubated at 37°C, 250 rpmfor 24 h. Each sample was then collected by using a filter sterilization unit (0.45�m), transferred into a microcentrifuge tube, and subjected to total RNA iso-lation. (iii) In liquid MMG or MMT, approximately 5 � 107 or 108 spores of eachstrain were inoculated in 100 ml of liquid MMG or MMT, respectively, andfurther incubated at 37°C, 250 rpm for 24 h. Samples were then collected,transferred into microcentrifuge tubes, and subjected to total RNA isolation.Approximately 2 �g (from conidia) or 6 �g of total RNA was loaded per lane,separated by agarose electrophoresis, transferred to nylon membrane (Mag-naprobe; Osmonics), and hybridized with the pkaB probe amplified by OSR14and OSR15 (Table 2).

Germination of conidia. To examine germination levels, the conidia (�106

spores/plate) of wild-type (FGSC26), �pkaB (RSA53.143), �pkaA (RSA53.130),alcA(p)::pkaB (RNI5.3), �pkaA alcA(p)::pkaB (RNI6), and �pkaA �pkaBalcA(p)::pkaB (RNI7) strains were spread on solid MM containing glucose orthreonine, or on the medium lacking a carbon source, and incubated at 37°C. Thelevels of germination (isotropic and germ tube formation) were examined every2 h after 4 h of inoculation on solid MMG or MMT or after 8 h of inoculationon the medium lacking a carbon source.

Stress response. To examine the sensitivity of hyphae to H2O2, the conidia ofwild-type (FGSC26), �pkaB (RSA53.143), �pkaA (RSA53.130), alcA(p)::pkaB(RNI5.3), and alcA(p)::pkaA (RNI8) strains were incubated on solid MMG(�250 spores/plate) in triplicate, and the plates were incubated at 37°C for 26 h.This 26-h incubation time was chosen because conidia of all the strains testedgerminated and formed microscopically visible colonies without producing con-idiophores. In the wild type, the first conidiophore from a single conidium isproduced at 28 h of incubation at 37°C (44). The hyphae were overlaid with 10ml of 0, 10, 50, 150, or 300 mM H2O2 solution and incubated at room temper-

ature for 10 min. The H2O2 solution was then decanted, and the plates werewashed four times with 15 ml of sterile double-distilled H2O. Plates were thenincubated at 37°C for an additional 24 h. The number of surviving colonies wascounted and calculated as a percentage of the control (untreated). To examinethe sensitivity of conidia to H2O2, 1-ml aliquots of conidial suspensions contain-ing �105 spores with various H2O2 concentrations (0, 0.25, 0.50, 0.75, and 1.0 M)were incubated at room temperature for 30 min. Spores were then collected bycentrifugation at 10,000 � g for 1 min, washed twice with 1.5 ml of steriledouble-distilled H2O, resuspended, diluted with sterile distilled water, and inoc-ulated onto solid MMG. After incubation at 37°C for 24 h, the number ofcolonies was counted and calculated as a percentage of control. To examine thethermal tolerance of hyphae, the conidia of wild-type (FGSC26), �pkaB(RSA53.143), and alcA(p)::pkaB (RNI5.3) conidia (�250/plate) were inoculatedon solid MMG in triplicate. Plates were incubated at 37°C for 10 h and thenshifted to 50°C. After 2, 3, or 4 h of incubation at 50°C, the plates were shiftedback to 37°C for an additional 24 h, and the number of colonies was counted.

Microscopy. The colony photographs were taken using a Sony DSC-F707digital camera. Microscopic photographs were taken using an Olympus BH2compound microscope installed with an Olympus DP-70 digital imaging system.

RESULTS

Identification of pkaB encoding the secondary PKA in A.nidulans. Search of the A. nidulans genome employing Tpk1,Tpk2, and Tpk3 of S. cerevisiae resulted in the identification ofthe pkaB gene encoding an additional PKA (AN4717.2, chro-mosome III). The pkaB gene is composed of a 1,371-bp ORFwith three (62-bp, 64-bp, and 54-bp) introns and is predicted toencode a 396-amino-acid polypeptide which shows 40 to �53%amino acid identity to yeast PKAs (Fig. 1A and C). The pre-dicted PkaB protein contains a typical Ser/Thr protein kinasedomain and an extension to the Ser/Thr-type protein kinasedomain at the C terminus (14, 37). These domains are con-served in numerous PKA catalytic subunits. To ensure thatpkaB was expressed, the steady-state mRNA levels were ex-amined by Northern blot analysis. As shown in Fig. 1B, pkaBencodes a �2.4-kb transcript that accumulates at high levelsduring the vegetative growth phase and the early to middlephases of asexual/sexual development of A. nidulans.

TABLE 2. Oligonucleotides used in this study

Oligo Sequence Position or purpose

OKH60 GAC TCT ATA CCA CCG TAC GCC GAT AT Forward primer for argB�

OKH61 CAC CGG GTG CGA TTT GCC CCA TTT CC Reverse primer for argB�

OKH56 AAC GAA CCC AGC GTC GAT CTA T pkaB deletion 5� forwardOKH57 AGT CAA ATG AGG CCT CTA AAC TGG TCA ACT GTG CGC TTC

TCC CTG TCA TpkaB deletion 5� reverse with complementary

5� argB� taila

OKH58 AGC CAA GGT AGA TCC AGG CCT AAC ACA CA TTG CGG AATGGT ATC ACA CGT C

pkaB deletion 3� forward with complementary3� argB� taila

OKH59 TGG GCT TTG GTC CTC ATT GAG T pkaB deletion 3� reverseOKH74 CGT CCA CGC ACT TCC TCA pkaB deletion construct 5� nestOKH75 GGG ATA TCT GCC CGG CTA pkaB deletion construct 3� nestOJA107 ACA ATC AAT GTT CAA TGT AC 5� end of alcA(p)OJA106 TTT GAG GCG AGG TGA TAG GAT TGG A 3� end of alcA(p)OJA108 CG GGA TCC AGT GGT TCG GTA ATC alcA(p) 5� nested with BamHI taila

OSR16 TCC AAT CCT ATC ACC TCG CCT CAA A CA TTC AAA TGGCTT CAG GAA

pkaB overexpression 5�-alcA(p) taila

OSR17 TTC TTC GGC TAG GGA TCC pkaB overexpression construct 3� nestedOSR18 TTG CTT TAT ACT CTG GCT TT pkaB overexpression construct 3�OSR14 ACT AGT ACT CCG TAC ATG C pkaB forward primer for probeOSR15 TTG CTA GCC TTG GTA ATA pkaB reverse primer for probeOJA142 CTG GCA GGT GAA CAA GTC brlA forward primer for probeOJA143 AGA AGT TAA CAC CGT AGA brlA reverse primer for probe

a Tail sequence is shown in italics.

VOL. 4, 2005 A. NIDULANS SYNTHETIC LETHAL PKAs 1467

on March 18, 2021 by guest

http://ec.asm.org/

Dow

nloaded from

1468 NI ET AL. EUKARYOT. CELL

on March 18, 2021 by guest

http://ec.asm.org/

Dow

nloaded from

PkaB shows high amino acid identity with other fungalPKAs: 82% with PkaC2 of A. fumigatus (GI 40641908 [23]),53% with Tpk1p of S. cerevisiae (GI 6322297 [39]), and 49%with Uka1 of U. maydis (GI 2565330 [9]) (for alignment, seeFig. 1C). Phylogenetic analysis of Tpk1p, Tpk2p, Tpk3p, andSch9p of S. cerevisiae (39, 40), PkaA and PkaB of A. nidulans(34), PkaC1 and PkaC2 of A. fumigatus (23), two putativePKAs, XP_324862.1 (GI 32408763) and XP_326295.1 (GI32411629) of Neurospora crassa, two PKAs (CpkA and Cpk2)of Magnaporthe grisea (25) (GI 27802543 for Cpk2), andUka1p, Adr1p, and UM00484.1 of U. maydis has been carriedout (9) (Fig. 1D). As it is quite apparent, most filamentousfungi contain two distantly related PKAs (previously classifiedas I and II) and one Sch9-like kinase (10). Although PkaB wasoriginally identified using yeast PKAs, it is distantly related toTpk1p, Tpk2p, Tpk3p, and PkaA. Previous studies in U. maydis(9), M. grisea (25), C. neoformans serotype A (7), A. fumigatus(23), and A. nidulans (34) demonstrated that deletion of classI PKA resulted in profound effects, suggesting that class I PKArepresents the predominant (primary) unit in these fungi (foran exception, see Discussion).

Deletion of both pkaB and pkaA is lethal. The pkaB deletionmutant was isolated employing the double-joint PCR methodby replacing the pkaB ORF with argB� (46). Phenotypic char-acteristics of the �pkaB mutant were thoroughly examined forvegetative growth at various temperatures, asexual/sexual de-velopment, spore germination, ST production, growth on dif-ferent external carbon sources, and tolerance to variousstresses. However, no apparent phenotypic changes werecaused by �pkaB. This is consistent with the previous reportsthat class II PKAs may have no clear function (9, 23). Based onthese results, we hypothesized that PkaB may function as asecondary unit and may play a certain role in the absence of theprimary PKA, PkaA. To test this hypothesis, we attempted toisolate the �pkaA �pkaB double mutant via a meiotic cross of�pkaA (RSA53.126) and �pkaB (TSR1 and TSR53) strains. Ofthe 80 randomly chosen progeny, 40 had the pkaA� �pkaB orthe �pkaA pkaB� genotype, and no single �pkaA �pkaB mu-tant was isolated. Due to the presence of the endogenous argBmutations (�argB or argB2) in both deletion mutants (Table 1),no pkaA� pkaB� (wild-type) recombinants could grow on themedium lacking arginine. Thus, if all recombinant progenywere viable, one-third of progeny would be the pkaA� �pkaB,�pkaA pkaB�, or �pkaA �pkaB mutant. If this were the case,the likelihood of not isolating the �pkaA �pkaB double mutantamong 80 progeny by chance would be (1 � 0.33)80, or 1.22 �10�14, i.e., almost 0% probability to miss the �pkaA �pkaB

mutant. This strongly suggests that the �pkaA �pkaB mutant isnonviable and that PkaA and PkaB constitute the sole PKAcatalytic subunits required for the viability of A. nidulans.

To further test the synthetic lethal interaction between PkaAand PkaB, a number of �pkaA and �pkaB strains carrying anectopic copy of alcA(p)::pkaB or alcA(p)::pkaA were con-structed and meiotically crossed with each other. Via randomscreening of progeny, we were able to isolate the �pkaA �pkaBmutants only if an ectopic copy of alcA(p)::pkaB oralcA(p)::pkaA was present (Table 1). These results furtherconfirm that deletion of both pkaA and pkaB is lethal.

Overexpression of pkaB enhances vegetative growth and res-cues the growth defects caused by �pkaA. The synthetic lethalphenotype caused by the absence of both pkaB and pkaAimplies that PkaB may function in hyphal growth and/or sporegermination. To examine its potential role in activating vege-tative growth, the effect of overexpression of pkaB on hyphalproliferation was examined. As shown in Fig. 2, overexpressionof pkaB resulted in enhanced hyphal proliferation while re-pressing conidiation, resulting in a fluffy colony with delayedconidiation on inducing medium (MMT with or without yeastextract [YE]). Furthermore, overexpression of pkaB could par-tially rescue the growth defects caused by �pkaA (Fig. 2, lowermiddle panel). These results indicate that PkaB may stimulatevegetative growth, which indirectly represses conidiation in theair (see reduced conidiation in Fig. 2).

To test for a potential role of PkaB in transducing FadA(G�)-mediated vegetative signaling, we generated the �flbA�pkaB mutant and found that deletion of pkaB was not suffi-cient to suppress fluffy-autolytic phenotypes caused by the ab-sence of FlbA function. This result suggests that either FadA-mediated vegetative signaling is primarily transmitted viaPkaA, with PkaB functioning as a back-up unit, or that PkaBfunctions in a separate signaling branch (see Discussion).

Upregulation of pkaB rescues delayed germination causedby �pkaA in the presence of glucose. Previously, it was shownthat deletion of pkaA resulted in delayed spore germination(10). While deletion of pkaB had no effects on spore germina-tion (data not shown), the synthetic lethality resulting from�pkaB �pkaA implied a potential role of PkaB in spore ger-mination. Two copies of pkaB, an ectopic copy of alcA(p)::pkaBand the endogenous pkaB gene, in alcA(p)::pkaB strains re-sulted in �10-fold-elevated pkaB mRNA accumulation evenwhen glucose (1%) was used as a carbon source (Fig. 3A).Such upregulation of pkaB allowed us to test the effects ofelevated mRNA levels of pkaB on spore germination in thepresence of glucose. Approximately 106 spores of wild-type

FIG. 1. Summary of pkaB, encoding the secondary PKA catalytic subunit. A. A partial restriction map of the pkaB gene region is shown. ThepkaB ORF (filled box), the mRNA coding region (arrow), and the three introns (marked by discontinuities on the arrow) were determined byreverse transcription-PCR and random amplification of cDNA ends followed by sequence analyses. B. Steady-state mRNA levels of pkaB in variousgrowth and developmental stages. Numbers indicate the times of incubation in liquid MMG (Veg) or solid MMG under the conditions inducingasexual development or sexual development. The pkaB mRNA levels were low at 24 and 96 h post-asexual and -sexual developmental induction.C. Amino acid sequence alignment of A. nidulans PkaB (An_PkaB), S. cerevisiae Tpk1P (Sc_Tpk1), A. fumigatus PkaC2 (Af_PkaC2), and U. maydisUka1 (Um_Uka1), showing identities (black boxes) and similarities (gray boxes). The proteins were aligned with ClustalW (5) and displayed byBOXSHADE (http://www.ch.embnet.org/software/BOX_form.html). D. Phylogenetic tree of PkaA, PkaB, and SchA of A. nidulans, PkaC1 andPkaC2 of A. fumigatus, CpkA and Cpk2 of M. grisea, XP324862.1 and XP326095.1 of N. crassa, Tpk1p, Tpk2p, and Tpk3p of S. cerevisiae, Adr1,and Uka1 and UM00484.1 of U. maydis. The ClustalW alignment (5) of the fungal proteins was presented by TREEVIEW. The length of the barrepresents an evolutionary distance of 0.1 amino acid substitutions per site.

VOL. 4, 2005 A. NIDULANS SYNTHETIC LETHAL PKAs 1469

on March 18, 2021 by guest

http://ec.asm.org/

Dow

nloaded from

and relevant mutant strains (Fig. 3B) were inoculated ontosolid MMG, and conidial swelling (isotropic growth) and germtube emergence were microscopically observed. More than90% of the wild-type conidia began to form short germ tubes(data not shown) by 6 h, and all the conidia formed long germtubes by 10 h (Fig. 3B). The conidia of the �pkaB andalcA(p)::pkaB mutants exhibited a germination pattern similarto that of the wild-type conidia (�pkaB not shown). As re-ported, deletion of pkaA severely delayed spore germination,i.e., only 40% of the �pkaA mutant conidia began to producegerm tubes at 10 h. As if PkaB could replace PkaA in sporegermination, upregulation of pkaB could rescue germinationdefects caused by �pkaA, i.e., the conidia of the �pkaAalcA(p)::pkaB mutant exhibited a germination level similar tothat of the wild-type conidia (Fig. 3B). In supporting dose-dependent effects of pkaB, the conidia of the �pkaA �pkaBalcA(p)::pkaB triple mutant did not germinate as effectively asthose of the �pkaA alcA(p)::pkaB mutant (data not shown, butfor pkaB accumulation see Fig. 3A). These results provide

evidence that PkaB plays an overlapping role in spore germi-nation in the presence of glucose.

Overexpression of pkaB delays spore germination. The ef-fect of overexpression on conidial germination was examinedby direct inoculation of the conidia of wild-type, �pkaB,alcA(p)::pkaB, �pkaA, �pkaA alcA(p)::pkaB, and �pkaA�pkaB alcA(p)::pkaB strains on MMT. Because it was notpossible to determine the pkaB mRNA levels under theseexperimental conditions, the pkaB mRNA levels were assessedby incubating the conidia of individual strains in liquid MMT at37°C, 250 rpm, for 24 h (see Materials and Methods). ThepkaB transcript levels in strains with alcA(p)::pkaB were about20-fold higher than those in wild-type and �pkaA strains (notshown). Somewhat surprisingly, we found that overexpressionof pkaB on solid MMT severely delayed conidial germination(Fig. 4A). Because threonine is not one of the most favorablecarbon sources for A. nidulans, germination of the conidia ofboth wild-type and �pkaB strains on solid MMT was delayedabout 4 h, i.e., �90% of the wild-type and �pkaB conidia

FIG. 2. Overexpression of pkaB enhances vegetative growth. Photographs of the colonies of wild-type (WT; FGSC26), �pkaB (RSA53.143),alcA(p)::pkaB (RNI5.3), �pkaA (RSA53.130), �pkaA alcA(p)::pkaB (RNI6), and �pkaA �pkaB alcA(p)::pkaB (RNI7) strains grown on MMG(noninducing) or MMT plus YE (threonine plus 5 g/liter YE) for 3 days or on MMT for 6 days at 37°C are shown. Yeast extract was added toachieve both alcA induction and enhanced growth. While �pkaB caused no apparent phenotypic changes, overexpression of pkaB resulted inelevated hyphal proliferation and reduced asexual sporulation on both MMT plus YE and MMT (upper panel). Furthermore, overexpression ofpkaB rescued the growth defects caused by �pkaA (lower panel).

1470 NI ET AL. EUKARYOT. CELL

on March 18, 2021 by guest

http://ec.asm.org/

Dow

nloaded from

formed germ tubes within 10 h of incubation (Fig. 4A; �pkaBnot shown). However, the majority of the conidia of thealcA(p)::pkaB, �pkaA, and �pkaA alcA(p)::pkaB mutants didnot germinate by 10 h (Fig. 4A), and 80% of the conidia of thethree mutants began to produce short germ tubes at 13 h (notshown). These results suggest that overexpression of PkaBnegatively affects spore germination.

Upregulation of pkaB blocks spore germination in the ab-sence of a carbon source. We then examined germination ofthe conidia of wild-type, alcA(p)::pkaB, �pkaA, and �pkaAalcA(p)::pkaB strains on the solid medium without a carbonsource. Under these conditions, a few wild-type conidia beganto germinate at 10 h of incubation, and most formed elongatedthin germ tubes at 31 h (Fig. 4B). However, while conidia ofthe �pkaA mutant exhibited isotropic growth (swelling) andgerminated at low levels (�20%), the alcA(p)::pkaB mutantconidia showed no sign of germination (e.g., swelling of spores)at 31 h (Fig. 4B). Moreover, no conidia of the alcA(p)::pkaBmutant were able to germinate even after 5 days of incubation(not shown). Similar results were obtained when the conidia ofthe �pkaA alcA(p)::pkaB (Fig. 4B) and �pkaA �pkaBalcA(p)::pkaB (not shown) mutants were tested. Collectively,

while both PkaB and PkaA positively function in spore germi-nation in the presence of glucose, PkaB negatively controlsconidial germination in the absence of external carbon sources.

Upregulation of pkaB enhances submerged sporulationcaused by �pkaA. The �pkaA mutant was shown to produceconidiophore-like structures in some liquid submerged cultureconditions, where wild-type strains do not sporulate (34). Totest the effect of �pkaB and upregulation of pkaB on sporula-tion in liquid submerged cultures, wild-type, �pkaB,alcA(p)::pkaB, �pkaA, �pkaA alcA(p)::pkaB, and �pkaA�pkaB alcA(p)::pkaB strains were cultured in liquid MMG at37°C, 250 rpm, and the status of development was microscop-ically observed. While the wild type and the �pkaB mutant didnot sporulate, the �pkaA mutant produced a few conidio-phore-like structures at 20 h (Fig. 5). Interestingly, the �pkaAalcA(p)::pkaB and �pkaA �pkaB alcA(p)::pkaB (not shown),but not alcA(p)::pkaB, mutants elaborated near-complete con-idiophore-like structures abundantly at 20 h. We then exam-ined the mRNA levels of brlA, encoding a key transcriptionalactivator for conidiophore development (2). As shown in Fig.5, while derepression of the brlA expression under submergedculture conditions was caused by both �pkaA and �pkaA

FIG. 3. Upregulation of pkaB rescues germination defects caused by deletion of pkaA. A. The conidia of wild-type (WT; FGSC26), �pkaB(RSA53.143), �pkaA (RSA53.130), alcA(p)::pkaB (RNI5.3), �pkaA alcA(p)::pkaB (RNI6), and �pkaA �pkaB alcA(p)::pkaB (RNI7) strains wereinoculated (108 spores/100 ml) into liquid MMG and incubated at 37°C, 250 rpm, for 24 h. Total RNA was isolated from each sample, 6 �g of totalRNA was loaded per lane, and equal loading was confirmed by ethidium bromide staining of rRNA. Note that the presence of the alcA(p)::pkaBconstruct resulted in elevated pkaB mRNA levels (�10-fold). While deletion of pkaA did not affect the pkaB mRNA levels in the absence ofalcA(p)::pkaB, somehow �pkaA and �pkaA �pkaB caused reduced accumulation of the pkaB mRNA derived from the alcA promoter. Due to thedifferences in transcriptional termination, the pkaB mRNA derived from the alcA promoter is about 0.5 kb bigger than the endogenous pkaBtranscript shown in wild-type (FGSC26) and �pkaA (RSA53.130) strains. Because wild-type and �pkaA strains showed extremely low pkaB mRNAlevels compared to the strains containing alcA(p)::pkaB, different exposure conditions were employed to make the pkaB transcripts clearly visiblein all the strains tested. It is important to note that the pkaB transcripts shown in alcA(p)::pkaB, �pkaA alcA(p)::pkaB, and �pkaA �pkaBalcA(p)::pkaB strains resulted from a 20-h exposure with one intensifying screen, whereas the pkaB transcripts shown in wild-type and �pkaAstrains were visualized by a 24-h exposure with two intensifying screens. Therefore, the actual relative levels of the pkaB mRNA in wild-type and�pkaA strains must be reduced by twofold. B. The conidia of wild-type (FGSC26), alcA(p)::pkaB (RNI5.3), �pkaA (RSA53.130), and �pkaAalcA(p)::pkaB (RNI6) strains were inoculated on solid MMG and incubated at 37°C for 10 h. Note that the presence of an ectopic copy ofalcA(p)::pkaB could rescue the germination defects caused by �pkaA.

VOL. 4, 2005 A. NIDULANS SYNTHETIC LETHAL PKAs 1471

on March 18, 2021 by guest

http://ec.asm.org/

Dow

nloaded from

alcA(p)::pkaB, the level of brlA mRNA was much higher in the�pkaA alcA(p)::pkaB mutant, indicating that upregulation ofpkaB results in elevated derepression of brlA in the absence ofpkaA. These results suggest that PkaB may play an oppositerole in controlling conidiophore development in liquid sub-merged culture.

Upregulation of pkaB or �pkaA reduces the hyphal toler-ance to oxidative stress. Finally, we examined a potential roleof PkaB in the stress response. When the conidia or the 10-h-cultured hyphae of wild-type, �pkaB, and alcA(p)::pkaB strains(see Materials and Methods) were tested for thermal toler-ance, mutant and wild-type strains exhibited no detectabledifferences. We then examined the sensitivities of the conidiaand the hyphae of wild-type, �pkaB, alcA(p)::pkaB, �pkaA,and alcA(p)::pkaA strains to oxidative stress. When treatedwith various concentrations of H2O2 for 30 min, no differencesin survival rates between the wild-type and mutant spores wereobserved. However, survival rates of the hyphae of �pkaA andalcA(p)::pkaB [but not alcA(p)::pkaA or �pkaB] strains (Fig.6A) on solid MMG decreased drastically when treated withH2O2. While almost 100% of the wild-type hyphae survived,only 60% of the alcA(p)::pkaB and 30% of the �pkaA hyphaeformed visible colonies after treatment with 10 mM H2O2 (Fig.6B). Furthermore, no alcA(p)::pkaB and �pkaA colonies sur-vived after treatment with 50 mM or higher concentrations ofH2O2 (Fig. 6A and B). This suggests that upregulation of pkaB

or the absence of pkaA reduces the tolerance of hyphae tooxidative stress and that PkaB and PkaA play opposite roles inthe hyphal tolerance to oxidative stress.

DISCUSSION

All cells have the capacity to sense and respond to theenvironment via signal transduction cascades. Membrane-bound heterotrimeric G proteins (G proteins) play a key rolein transduction of the extracellular cues into cells to elicit theappropriate physiological and biochemical responses. In vari-ous fungi, the G protein-cAMP-dependent protein kinase(PKA) signaling pathway is a primary mechanism for vegeta-tive growth, morphological switch, nutrient sensing, mating,stress response, secondary metabolism and pathogenicity (re-viewed in references 8, 21, and 22). In this study, we haveidentified and characterized the second PKA catalytic subunitin the model ascomycete, A. nidulans. One of the critical find-ings in our study is that the absence of both pkaA and pkaB islethal, which indicates that PkaA and PkaB constitute the solePKA catalytic subunits essential for growth of A. nidulans.Moreover, our intensive genetic studies unveiled (partially) thecomplicated functions of this secondary PKA in controllingdiverse physiological processes.

Previous studies showed that the cellular PKA activity islargely dependent on class I PKA in various filamentous fungi,

FIG. 4. Inhibitory roles of pkaB in germination. A. The conidia of wild-type (WT; FGSC26), alcA(p)::pkaB (RNI5.2), �pkaA (RSA53.130), and�pkaA alcA(p)::pkaB (RNI6) strains were inoculated on solid MMT (inducing medium) and incubated at 37°C for 10 h. Note that while most ofthe wild-type conidia germinated, only a few mutant conidia germinated. The arrows show the alcA(p)::pkaB mutant conidia forming a germ tube.Although culture conditions were not identical, the pkaB mRNA levels in alcA(p)::pkaB (RNI5.2) and �pkaA alcA(p)::pkaB (RNI6) strains grownin liquid MMT for 24 h (see Materials and Methods) were about 20-fold higher than those of the wild type (not shown). B. The conidia of wild-type(WT; FGSC26), alcA(p)::pkaB (RNI5.2), �pkaA (RSA53.130), and �pkaA alcA(p)::pkaB (RNI6) strains were inoculated on the solid mediumlacking a carbon source and incubated at 37°C for 31 h. An electrophoresis-grade agarose was used as a solidifying agent to avoid the introductionof unwanted energy sources. At 31 h, most of the wild-type conidia were able to germinate and about 20% of the �pkaA mutant conidia could formgerm tubes (arrows). However, upregulation of pkaB caused by alcA(p)::pkaB completely blocked germination of spores regardless of the presenceor absence of pkaA. Although culture conditions were different, the pkaB mRNA levels in alcA(p)::pkaB (RNI5.2) and �pkaA alcA(p)::pkaB(RNI6) strains incubated in the liquid medium lacking a carbon source at 37°C for 24 h (see Materials and Methods) were about fivefold higherthan those of the wild type (data not shown).

1472 NI ET AL. EUKARYOT. CELL

on March 18, 2021 by guest

http://ec.asm.org/

Dow

nloaded from

such as A. nidulans (34), C. neoformans serotype A (7), M.grisea (1), and A. fumigatus (23). As it plays a predominant rolein these fungi, mutational inactivation of class I PKA results inprofound effects. However, we found that, compared to thoseof ascomycete fungi, the two PKAs in basidiomycetes appear tobe more closely related (see the distance between Adr1 andUka1 of U. maydis in Fig. 1D). Our additional phylogeneticanalysis employing Pka1 and Pka2 of the A and D serotypes ofC. neoformans (17) further confirmed that the two basidiomy-cete PKAs are somewhat closely related (data not shown, butsimilar to Adr1 and Uka1 of U. maydis in Fig. 1D) and sug-gested that the predominant PKA unit may not be clearlydistinguished in some basidiomycete fungi. In fact, a recentstudy showed that the predominant PKA is different dependingon the serotype of C. neoformans (17). While loss of Pka1(similar to Um_Adr1, class I [Fig. 1D]) in serotype A resultedin failure in mating, melanin, and capsule production as well asvirulence establishment (7), mutational inactivation of Pka2(similar to Um_Uka1, class II) in serotype D affected mating,haploid fruiting, and virulence factor (melanin) production(17).

Double deletion analyses have been carried out in two ba-sidiomycete fungi, C. neoformans and U. maydis. Deletion ofboth pka1 and pka2 in serotype D did not block fungal growth,indicating that PKA is not essential for the viability of C.neoformans (17). Similarly, in U. maydis, while the �uka1�adr1 double mutant formed noticeably less mycelial mass andlower levels of tumor than the adr1 single mutant, the viabilityremained unaffected, indicating that PKAs are not absolutely

required for growth of this fungus (9). On the contrary, ourstudies evidently demonstrate that mutational inactivation ofboth PkaB and PkaA is lethal and that PKAs are essential forthe viability of A. nidulans. A similar result was obtained in M.grisea, where the double mutant of cpkA and cpk2 is lethal(J. R. Xu, personal communication). This is somewhat discor-dant to the previous findings that the A. nidulans adenylatecyclase, CyaA, is not essential for fungal growth (10). As apossible explanation, while PKA catalytic subunits are gener-ally activated by cAMP binding to the PKA regulatory subunit(PkaR; C. d’Enfert, unpublished data), PKA can be activatedby other mechanisms in addition to CyaA-cAMP signaling,and/or PKAs may maintain a basal level activity even in theabsence of cAMP.

The fact that deletion of both pkaB and pkaA is lethal im-plies that PkaB plays a role in fungal growth and/or sporegermination. Similar to overexpression of pkaA (34), overex-pression of pkaB enhanced hyphal proliferation and reducedconidiation in air-exposed solid MMT, indicating that PkaBand PkaA have an overlapping role in vegetative growth anddevelopmental regulation. Moreover, overexpression or up-regulation of pkaB could partially rescue the growth and ger-mination defects caused by �pkaA, suggesting that PkaB can atleast partially replace PkaA functions in certain physiologicalprocesses. Vegetative growth signaling mediated by FadA(G�)-FlbA (RGS) is in part transduced by PkaA (34). Deletionof pkaB could not suppress fluffy-autolytic phenotypes causedby the absence of FlbA. This finding suggests that either PkaBfunctions in the FadA-mediated signaling pathway but its role

FIG. 5. Upregulation of pkaB enhances submerged sporulation caused by �pkaA. The conidia (108/100 ml) of wild-type (WT; FGSC26),alcA(p)::pkaB (RNI5.2), �pkaA (RSA53.130), and �pkaA alcA(p)::pkaB (RNI6) strains were inoculated in liquid MMG and incubated at 37°C,250 rpm, for 20 h. The arrows indicate conidiophore-like structures formed in �pkaA and �pkaA alcA(p)::pkaB strains. The steady-state mRNAlevels of brlA were examined at 26 h of incubation. Transcripts of brlA were detectable in the �pkaA and �pkaA alcA(p)::pkaB mutants. Note thatbrlA is highly accumulated by upregulation of pkaB in the absence of pkaA. For the pkaB mRNA levels in individual strains, see Fig. 3A.

VOL. 4, 2005 A. NIDULANS SYNTHETIC LETHAL PKAs 1473

on March 18, 2021 by guest

http://ec.asm.org/

Dow

nloaded from

in vegetative growth signaling is minute, or that PkaB primarilyfunctions in a separate signaling branch (Fig. 7; see below).

Spore germination is the first crucial step for fungal growthand colony formation. At least two signaling pathways (G pro-tein-cAMP and Ras) are associated with controlling spore ger-mination in A. nidulans (10). Fillinger et al. first reported thatcAMP signaling controls the early events of conidial germina-tion in response to external carbon sources (10). Disruption ofcyaA and pkaA caused delayed spore germination, and thedouble deletion mutant exhibited more pronounced delay thaneach single mutant. Because no cAMP was detectable in the�cyaA mutant, their results support the idea that PkaA (andprobably PkaB) in A. nidulans may retain some activity in theabsence of cAMP and that additional signaling pathways, in-cluding RasA, are involved in the spore germination process(10). Our findings indicate that two PKAs play overlappingroles in spore germination in the presence of glucose. How-ever, in contrast, overexpression of pkaB or �pkaA delayedgermination on solid MMT [inducing conditions foralcA(p)::pkaB]. Furthermore, upregulation of pkaB completelyblocked spore germination on the medium without an externalcarbon source (Fig. 4B). These results suggest that, whereas

PkaB can activate germination when glucose (a favorable car-bon source) is present, one of its important regulatory rolesmight be to negatively control spore germination under unfa-vorable conditions, such as a lack of external energy sources.Balanced activities of two PKAs may play an important role inregulating proper spore germination.

Besides its functions in controlling fungal growth and nutri-ent sensing, cAMP-PKA has been found to be involved instress responses in fungi. In S. cerevisiae, the G-protein-cou-pled receptor Gpr1 and the G� subunit Gpa2 stimulate cAMPproduction by adenylyl cyclase in response to glucose and otherfermentable carbon sources, thereby activating TPKs (29).This Gpr1-Gpa2-cAMP-PKA signaling is negatively regulatedby Rgs2 (an RGS protein [41]). Activities of TPKs antagonizethe induction of the general stress response in S. cerevisiae,which is mediated by Msn2p and Msn4p, positive transcrip-tional factors for the stress response pathway (35). Deletion ofRGS2 causes prolonged activation of Gpa2-PKA and therebyincreases the sensitivity of cells to thermal stress, while over-expression of RGS2 causes significant elevation in heat resis-tance (41). Recent studies showed that GanB (a G� similar toGpa2 [4]) functions in sensing external carbon sources and

FIG. 6. Upregulation of pkaB or �pkaA reduces hyphal tolerance to oxidative stress. The hyphae of wild-type (WT; FGSC26), �pkaB(RSA53.111), alcA(p)::pkaB (RNI5.2), �pkaA (RSA53.126), and alcA(p)::pkaA (RNI8) strains grown on solid MMG at 37°C for 26 h were treatedwith 0, 10, 50, 150, and 300 mM H2O2 for 10 min in triplicate. The plates were then incubated at 37°C for an additional 24 h. A. The colonies thatsurvived through 0 mM (control) or 50 mM H2O2 treatment are shown. Note that no alcA(p)::pkaB or �pkaA mutants survived after treatment,whereas the alcA(p)::pkaA and �pkaB mutants did not exhibit differences in survival rates compared to the wild type. B. The bar graph shows therelative survival rates of the hyphae of wild-type, alcA(p)::pkaB, and �pkaA strains.

1474 NI ET AL. EUKARYOT. CELL

on March 18, 2021 by guest

http://ec.asm.org/

Dow

nloaded from

controlling spore germination via a cAMP/PKA pathway in A.nidulans, where SfaD-GpgA (G) is predicted to function inactivating GanB (19, 30, 33). Furthermore, we demonstratedthat GanB signaling activates thermal and oxidative stress re-sponses, which are negatively controlled by RgsA (an RGSprotein similar to Rgs2 in yeast [13]). Contrary to the yeaststress response, deletion of rgsA results in the elevated thermaland oxidative stress tolerances of both hyphae and spores (13).In this study we demonstrated that, while no pkaB mutations(deletion and/or overexpression) affected thermal tolerance ofhyphae and spores, or tolerance of spores against oxidativestress, �pkaA or alcA(p)::pkaB resulted in significantly reducedtolerance of hyphae to oxidative stress. Our findings indicatethat, while RgsA-GanB activities may regulate the generalstress response in A. nidulans, the PKA-mediated stress re-sponse might be limited to oxidative stress and that PkaA andPkaB play opposite roles in conferring hyphal tolerance tooxidative stress.

Deletion or dominant-negative mutation of ganB as well asdeletion of pkaA result in formation of conidiophore-likestructures in liquid submerged culture (4, 34). We found that,while overexpression of pkaB alone did not cause sporulationin liquid submerged culture by itself, it enhanced the sub-merged sporulation phenotype caused by �pkaA. Collectively,based on the overlapping and opposite roles of the two PKAsin various physiological processes and the role of GanB-cAMP-PKA signaling in controlling spore germination, stressresponse, and asexual development, it can be speculated that

PkaA and PkaB function downstream of GanB. If this were thecase, GanB signaling might stimulate PkaA activity while in-hibiting PkaB function. Currently available information is in-tegrated and summarized in the model presented in Fig. 7. Inthis model, it is proposed that both FadA and GanB activatePkaA. PkaA functions as the predominant unit for vegetativegrowth, germination, and regulation of asexual developmentand ST production. PkaB is the secondary PKA and is likelyinhibited by GanB signaling, whereas overexpression and/orupregulation of pkaB may overcome the GanB-mediated re-pressive effects. Understanding the exact functions and thegenetic position of PkaB will require further studies.

ACKNOWLEDGMENTS

We are grateful to the Broad Institute for the genome database,Kap-Hoon Han for helpful discussion, Nancy Keller for providingpkaA mutant strains, and Ann Larson and Ellin Doyle for criticallyreviewing the manuscript.

This work was supported by a Hatch grant and National ScienceFoundation grant MCB-0421863 to J.H.Y.

REFERENCES

1. Adachi, K., and J. E. Hamer. 1998. Divergent cAMP signaling pathwaysregulate growth and pathogenesis in the rice blast fungus Magnaporthe grisea.Plant Cell 10:1361–1373.

2. Adams, T. H., M. T. Boylan, and W. E. Timberlake. 1988. brlA is necessaryand sufficient to direct conidiophore development in Aspergillus nidulans.Cell 54:353–362.

3. Bockmuhl, D. P., S. Krishnamurthy, M. Geards, A. Sonneborn, and J. F.Ernst. 2001. Distinct and redundant roles of the two protein kinase Aisoforms Tpk1P and Tpk2P in morphogenesis and growth of Candida albi-cans. Mol. Microbiol. 42:1243–1257.

FIG. 7. Proposed model for PKA-mediated signaling in A. nidulans. We propose that the two PKAs are differently regulated by G proteinsignaling and that overlapping and opposite roles of PkaA (primary PKA) and PkaB (secondary PKA) control various biological processes.FlbA-FadA (G�)-controlled vegetative growth signaling is in part transduced by PkaA, which in turn represses conidiation and ST production (34).The demonstration that GanB and SfaD::GpgA, as well as PkaA, are required for proper germination of spores (10, 19, 30, 33) indicates that PkaAmay be activated by both GanB and FadA. RgsA-controlled GanB signaling is also associated with activation of the stress response (13). The factthat �pkaA delayed conidial germination and reduced hyphal tolerance to oxidative stress further supports the idea that PkaA may be activatedby GanB. While PkaB can stimulate vegetative growth and activate conidial germination in the presence of glucose, upregulation of pkaB blocksgermination of spores in the absence of a carbon source and reduces hyphal tolerance to oxidative stress. This suggests that the resulting cellularresponses for PkaB and GanB-PkaA signaling may be opposite. Thus, if PkaB functions downstream of GanB, the activity of PkaB may benegatively regulated by GanB. GanB-mediated signaling may activate other stress responses via a different (possibly a mitogen-activated proteinkinase) downstream signaling cascade.

VOL. 4, 2005 A. NIDULANS SYNTHETIC LETHAL PKAs 1475

on March 18, 2021 by guest

http://ec.asm.org/

Dow

nloaded from

4. Chang, M.-H., K.-S. Chae, D.-M. Han, and K.-Y. Jahng. 2004. The GanBG�-protein negatively regulates asexual sporulation and plays a positive rolein conidial germination in Aspergillus nidulans. Genetics 167:1305–1315.

5. Chenna, R., H. Sugawara, T. Koike, R. Lopez, T. J. Gibson, D. G. Higgins,and J. D. Thompson. 2003. Multiple sequence alignment with the Clustalseries of programs. Nucleic Acids Res. 31:3497–3500.

6. Colledge, M., and J. D. Scott. 1999. AKAPs: from structure to function.Trends Cell Biol. 9:216–221.

7. D’Souza, C. A., J. A. Alspaugh, C. Yue, T. Harashima, G. M. Cox, J. R.Perfect, and J. Heitman. 2001. cAMP-dependent protein kinase controlsmating and virulence of the fungal pathogen Cryptococcus neoformans. Mol.Cell. Biol. 21:3179–3191.

8. D’Souza, C. A., and J. Heitman. 2001. Conserved cAMP signaling cascadesregulate fungal development and virulence. FEMS Microbiol. Rev. 25:349–364.

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

10. Fillinger, S., M. K. Chaveroche, K. Shimizu, N. P. Keller, and C. d’Enfert.2002. cAMP and ras signalling independently control spore germination inthe filamentous fungus Aspergillus nidulans. Mol. Microbiol. 44:1001–1016.

11. Griffioen, G., and J. M. Thevelein. 2002. Molecular mechanisms controllinglocalization of protein kinase A. Curr. Genet. 41:199–207.

12. Han, K.-H., J.-A. Seo, and J.-H. Yu. 2004. A putative G protein-coupledreceptor negatively controls sexual development in Aspergillus nidulans. Mol.Microbiol. 51:1333–1345.

13. Han, K.-H., J.-A. Seo, and J.-H. Yu. 2004. Regulators of G-protein signalingin Aspergillus nidulans: RgsA downregulates stress response and stimulateasexual sporulation through attenuation of GanB (G�) signaling. Mol. Mi-crobiol. 53:529–540.

14. Hanks, S. K., and T. Hunter. 1995. Protein kinases 6. The eukaryotic proteinkinase superfamily: kinase (catalytic) domain structure and classification.FASEB J. 9:576–596.

15. Hatanaka, M., and C. Shimoda. 2001. The cyclic AMP/PKA signal pathwayis required for initiation of spore germination in Schizosaccharomyces pombe.Yeast 18:207–217.

16. Hicks, J. K., J.-H. Yu, N. P. Keller, and T. H. Adams. 1997. Aspergillussporulation and mycotoxin production both require inactivation of the FadAG� protein-dependent signaling pathway. EMBO J. 16:4916–4923.

17. Hicks, J. K., C. A. D’Souza, G. M. Cox, and J. Heitman. 2004. CyclicAMP-dependent protein kinase catalytic subunits have divergent roles invirulence factor production in two varieties of the fungal pathogen Crypto-coccus neoformans. Eukaryot. Cell 3:14–26.

18. Kafer, E. 1977. Meiotic and mitotic recombination in Aspergillus and itschromosomal aberrations. Adv. Genet. 19:33–131.

19. Lafon, A., J.-A. Seo, K.-H. Han, J.-H. Yu, and C. d’Enfert. The heterotri-meric G-protein GanB(�)-SfaD()-GpgA() is a carbon source sensor in-volved in early cAMP-dependent germination in Aspergillus nidulans. Ge-netics, in press.

20. Lee, B. N., and T. H. Adams. 1994. Overexpression of flbA, an early regulatorof Aspergillus asexual sporulation, leads to activation of brlA and prematureinitiation of development. Mol. Microbiol. 14:323–334.

21. Lee, N., C. A. D’Souza, and J. W. Kronstad. 2003. Of smuts, blasts, mildewsand blights: cAMP signaling in phytopathogenic fungi. Annu. Rev. Phyto-pathol. 41:399–427.

22. Lengeler, K. B., R. C. Davidson, C. A. D’Souza, T. Harashima, W.-C. Shen,P. Wang, X.-W. Pan, M. Waugh, and J. Heitman. 2000. Signal transductioncascades regulating fungal development and virulence. Microbiol. Mol. Biol.Rev. 64:746–785.

23. Liebmann, B., M. Muller, A. Braun, and A. A. Brakhage. 2004. The cyclicAMP-dependent protein kinase a network regulates development and viru-lence in Aspergillus fumigatus. Infect. Immun. 72:5193–5203.

24. Maeda, T., Y. Watanabe, H. Kunitomo, and M. Yamamoto. 1994. Cloning ofthe pka1 gene encoding the catalytic subunit of the cAMP-dependent proteinkinase in Schizosaccharomyces pombe. J. Biol. Chem. 269:9632–9637.

25. Mitchell, T. K., and R. A. Dean. 1995. The cAMP-dependent protein kinasecatalytic subunit is required for appressorium formation and pathogenesis bythe rice blast pathogen Magnaporthe grisea. Plant Cell 7:1869–1878.

26. Pan, X., and J. Heitman. 1999. Cyclic AMP-dependent protein kinase reg-ulates pseudohyphal differentiation in Saccharomyces cerevisiae. Mol. Cell.Biol. 19:5352–5356.

27. Pontecorvo, G., J. A. Roper, L. M. Hemmons, K. D. Macdonald, and A. W. J.Bufton. 1953. The genetics of Aspergillus nidulans. Adv. Genet. 5:141–238.

28. Robertson, L. S., and G. R. Fink. 1998. The three yeast A kinases havespecific signaling functions in pseudohyphal growth. Proc. Natl. Acad. Sci.USA 95:13783–13787.

29. Rolland, F., J. H. De Winde, K. Lemaire, E. Boles, J. M. Thevelein, and J.Winderickx. 2000. Glucose-induced cAMP signalling in yeast requires both aG-protein coupled receptor system for extracellular glucose detection and aseparable hexose kinase-dependent sensing process. Mol. Microbiol. 38:348–358.

30. Rosen, S., J.-H. Yu, and T. H. Adams. 1999. The Aspergillus nidulans sfaDgene encodes a G protein subunit that is required for normal growth andrepression of sporulation. EMBO J. 18:5592–5600.

31. Roosen, J., K. Engelen, K. Marchal, J. Mathys, G. Griffioen, E. Cameroni,J. M. Thevelein, C. De Virgilio, B. De Moor, and J. Winderickx. 2005. PKAand Sch9 control a molecular switch important for the proper adaptation tonutrient availability. Mol. Microbiol. 55:862–880.

32. Seo, J.-A., Y. Guan, and J.-H. Yu. 2003. Suppressor mutations bypass therequirement of fluG for asexual sporulation and sterigmatocystin productionin Aspergillus nidulans. Genetics 165:1083–1093.

33. Seo, J.-A., K.-H. Han, and J.-H. Yu. Multiple roles of a heterotrimeric Gprotein subunit in governing growth and development of Aspergillus nidu-lans. Genetics, in press.

34. Shimizu, K., and N. P. Keller. 2001. Genetic involvement of a cAMP-dependent protein kinase in a G protein signaling pathway regulating mor-phological and chemical transitions in Aspergillus nidulans. Genetics 157:591–600.

35. Smith, A., P. W. Mary., and S. Garrett. 1998. Yeast PKA represses Msn2p/Msn4p-dependent gene expression to regulate growth, stress response andglycogen accumulation. EMBO J. 17:3556–3564.

36. Taylor, S. S., J. A. Buechler, and W. Yonemoto. 1990. cAMP-dependentprotein kinase: framework for a diverse family of regulatory enzymes. Annu.Rev. Biochem. 59:971–1005.

37. Taylor, S. S., and E. Radzio-Andzelm. 1994. Three protein kinase structuresdefine a common motif. Structure 2:345–355.

38. Taylor, S. S., J. Yang, J. Wu, N. M. Haste, E. Radzio-Andzelm, and G.Anand. 2004. PKA: a portrait of protein kinase dynamics. Biochimica. Bio-physica. Acta 1697:259–269.

39. Toda, T., S. Cameron, P. S. Sass, M. P. Zoller, and M. Wigler. 1987. Threedifferent genes in S. cerevisiae encode the catalytic subunits of the cAMP-dependent protein kinase. Cell 50:277–287.

40. Toda, T., S. Cameron, P. Sass, and M. Wigler. 1988. SCH9, a gene ofSaccharomyces cerevisiae that encodes a protein distinct from, but function-ally and structurally related to cAMP-dependent protein kinase catalyticsubunits. Genes Dev. 2:517–527.

41. Versele, M., J. H. de Winde, and J. M. Thevelein. 1999. A novel regulator ofG protein signalling in yeast, Rgs2, downregulates glucose-activation of thecAMP pathway through direct inhibition of Gpa2. EMBO J. 18:5577–5591.

42. Wieser, J., and T. H. Adams. 1995. flbD encodes a Myb-like DNA-bindingprotein that coordinates initiation of Aspergillus conidiophore development.Genes Dev. 9:491–502.

43. Xu, J. R., M. Urban, J. A. Sweigard, and J. E. Hamer. 1997. The CPKA geneof Magnaporthe grisea is essential for appressorial penetration. Mol. PlantMicrobe. Interact. 10:187–194.

44. Yu, J.-H., J. Wieser, and T. H. Adams. 1996. The Aspergillus FlbA RGSdomain protein antagonizes G-protein signaling to block proliferation andallow development. EMBO J. 15:5184–5190.

45. Yu, J.-H., S. Rosen, and T. H. Adams. 1999. Extragenic suppressors ofloss-of-function mutations in the Aspergillus FlbA regulator of G-proteinsignaling domain protein. Genetics 151:97–105.

46. Yu, J.-H., Z. Hamari, K.-H. Han, J.-A. Seo, Y. Reyes-Domınguez, and C.Scazzochio. 2004. Double-joint PCR: a PCR-based molecular tool for genemanipulation in filamentous fungi. Fungal Gen. Biol. 41:973–981.

1476 NI ET AL. EUKARYOT. CELL

on March 18, 2021 by guest

http://ec.asm.org/

Dow

nloaded from