cyclic amp-dependent protein kinase catalytic subunits ...the divergent pka2 catalytic subunit was...

EUKARYOTIC CELL, Feb. 2004, p. 14�26 Vol. 3, No. 11535-9778/04/$08.00�0 DOI: 10.1128/EC.3.1.14–26.2004Copyright © 2004, American Society for Microbiology. All Rights Reserved.

Cyclic AMP-Dependent Protein Kinase Catalytic Subunits HaveDivergent Roles in Virulence Factor Production in Two

Varieties of the Fungal PathogenCryptococcus neoformans

Julie K. Hicks,1† Cletus A. D’Souza,1†‡ Gary M. Cox,1,2 and Joseph Heitman1,2,3,4*Departments of Molecular Genetics and Microbiology,1 Medicine,2 and Pharmacology and Cancer Biology,3

and Howard Hughes Medical Institute,4 Duke University Medical Center,Durham, North Carolina 27710

Received 13 September 2003/Accepted 24 October 2003

Our earlier findings established that cyclic AMP-dependent protein kinase functions in a signaling cascadethat regulates mating and virulence of Cryptococcus neoformans var. grubii (serotype A). Mutants lacking theserotype A protein kinase A (PKA) catalytic subunit Pka1 are unable to mate, fail to produce melanin orcapsule, and are avirulent in animal models, whereas mutants lacking the PKA regulatory subunit Pkr1overproduce capsule and are hypervirulent. Because other mutations have been observed to confer differentphenotypes in two diverged varieties of C. neoformans (grubii variety [serotype A] and neoformans variety[serotype D]), we analyzed the functions of the PKA genes in the serotype D neoformans variety. Surprisingly,the Pka1 catalytic subunit was not required for mating, haploid fruiting, or melanin or capsule production ofserotype D strains. Here we identify a second PKA catalytic subunit gene, PKA2, that is present in both serotypeA and D strains of C. neoformans. The divergent Pka2 catalytic subunit was found to regulate mating, haploidfruiting, and virulence factor production in serotype D strains. In contrast, Pka2 has no role in mating,melanin production, or capsule formation in serotype A strains. Our studies illustrate how different compo-nents of signaling pathways can be co-opted and functionally specialized during the evolution of related butdistinct varieties or subspecies of a human fungal pathogen.

A highly conserved nutrient and G-protein-regulated cyclicAMP (cAMP) signaling cascade promotes mating and viru-lence in Cryptococcus neoformans (2, 22). The cAMP-depen-dent protein kinase (PKA) mediates most, if not all, of thephysiological effects of the second messenger cAMP in fungiand other multicellular eukaryotes (reviewed in reference 60).In the inactive state, PKA exists as a holoenzyme comprised oftwo catalytic subunits and two regulatory subunits. Upon acti-vation, the PKA catalytic subunits are released and phosphor-ylate target substrates that include metabolic enzymes andtranscription factors (21, 22).

The cAMP-dependent protein kinases have undergone geneduplication and functional divergence during genome evolu-tion. Multiple PKA isoforms are expressed in mammalian cellsand have tissue-specific roles indicative of functional diversity(60). In Saccharomyces cerevisiae, the three PKA catalytic sub-unit isoforms Tpk1, Tpk2, and Tpk3 play redundant roles inyeast vegetative growth but have specialized roles in pseudohy-phal filamentous growth. Tpk2 activates filamentation,whereas Tpk1 and Tpk3 repress filamentation (48, 51, 64). Theplant fungal pathogen Ustilago maydis expresses two PKA cat-alytic subunit isoforms, Adr1 and Uka1 (24). Adr1 is a key

regulator of filamentous growth, whereas Uka1 plays only aminor role. PKA also regulates morphogenesis and virulencein the plant pathogenic fungi Magnaporthe grisea and Cry-phonectria parasitica and in the human pathogen Candida al-bicans (27, 43, 58).

The basidiomycete C. neoformans is an opportunistic humanfungal pathogen that is the causative agent of cryptococcosis(reviewed in references 12 and 31). The prevalence of crypto-coccal meningitis has increased worldwide as a result of humanimmunodeficiency virus infections, solid organ transplants, cy-totoxic chemotherapy, and systemic corticosteroid use (42). C.neoformans serves as a model system for studies of fungalpathogenesis. Known virulence factors include an antiphago-cytic capsule, the antioxidant melanin, the enzymes urease andphospholipase, and the ability to grow at 37°C (17, 18, 28, 36).C. neoformans has a defined sexual cycle in which compatiblecells mate in response to peptide pheromones and nutritionalsignals to produce a dikaryotic mycelium, basidia, and basid-iospores (33, 44, 55). � strains also differentiate by haploidfruiting involving filamentation and sporulation in response tonitrogen limitation, desiccation, and MFa pheromone (70, 71).

Based on the antigenicity of the capsular polysaccharidecomponents, C. neoformans can be classified into four sero-types: A, B, C, and D (35). Serotype A strains are the mostcommon clinical isolates in the world, representing 99% ofisolates from AIDS patients (38). Although clinically less com-mon, serotype D and AD hybrid strains are still importantpathogens and can represent 10% or more of clinical isolates insome regions of the world such as Europe (6, 34).

* Corresponding author. Mailing address: Department of MolecularGenetics and Microbiology, 322 CARL Bldg., Duke University Med-ical Center, Research Dr., Durham, NC 27710. Phone: (919) 684-2824.Fax: (919) 684-5458. E-mail: [email protected].

† J.K.H. and C.A.D. contributed equally to this work.‡ Present address: Biotechnology Laboratory, University of British

Columbia, Vancouver, Canada.

14

on April 4, 2020 by guest

http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

Population genetic studies have shown that the A and Dserotypes diverged �18 million years ago (25, 72). Serotype Astrains are designated C. neoformans var. grubii, and serotypeD strains are designated C. neoformans var. neoformans (26).This distinction into two different varieties or sibling species isfurther supported by molecular genetic studies. The Ste12�transcription factor is required for virulence in serotype D butnot in serotype A (14, 15, 73). Conversely, the mitogen-acti-vated protein kinase component Ste20� contributes to fullvirulence in serotype A but is dispensable for virulence in thecongenic serotype D strains (69). Recent studies reveal thatalthough serotype A and D strains can mate, unusual serotypeAD diploid hybrids are produced that sporulate to producelargely inviable spores (39). Taken together, these findingsreveal significant differences between serotypes A and D andsuggest that these represent at least two varieties and possiblyeven distinct species.

It was previously demonstrated that cAMP-dependent pro-tein kinase controls mating and virulence of C. neoformansserotype A strains (22). Here, we report the identification of asecond PKA catalytic subunit (Pka2) and have taken a com-parative genetics approach to elucidate the functions of thePKA catalytic subunits Pka1, Pka2, and the regulatory subunitPkr1 in serotypes A and D. We demonstrate that Pka2 regu-lates mating and expression of virulence factors in serotype D,

whereas Pka1 plays this role in serotype A. Additionally, wefind that PKA is a key regulator of virulence in serotype A butnot in serotype D, despite the essential role of PKA in viru-lence factor production in serotype D. These findings revealthat functional specialization of cAMP-dependent protein ki-nase catalytic subunits has occurred during the evolution of C.neoformans into the serotype A and D varieties such that thefunctions of the Pka1 and Pka2 subunits have been inter-changed.

MATERIALS AND METHODS

C. neoformans strains and media. All strains used in this study are listed inTable 1. Uracil auxotrophic strains were isolated by selecting spontaneous mu-tants resistant to 5-fluoroorotic acid as described previously (37). C. neoformansstrains were grown on standard S. cerevisiae media (56). The selective mediumfor biolistic transformation (66), Niger seed medium (2), low-iron medium(LIM) (68), filament agar (FA) (71), V8 medium (2), and modified Eagle’smedium (MEM) (28) were prepared as previously described. For cAMP rescueexperiments, cAMP was added at a concentration of 10 mM to media. The pka1mutant strains CDC36 (MATa pka1::ADE2 ade2) and CDC40 (MAT�pka1::ADE2 ade2) were isolated as segregants from a cross of strains CDC25(MATa pka1::ADE2 ura5 ade2) and JEC50 (MAT� ade2). The pka2 serotype Dmutant strains CDC99 (MAT�pka2�::URA5 ura5) and CDC101 (MATapka2�::URA5 ura5) strains were isolated from a cross of strains CDC89 (MAT�pka2�::URA5 ura5 ade2) and JEC34 (MATa ura5). The pka1 pka2 double mutantserotype D strains CDC103 (MAT� pka1::ADE2 pka2�::URA5) and CDC106(MATa pka1::ADE2 pka2�::URA5) were constructed by crossing strains CDC89

TABLE 1. Strains

Strain Genotype Source orreference

Serotype D strainsJEC20 MATa 44JEC21 MAT� 44JEC34 MATa ura5 J. C. EdmanJEC43 MAT� ura5 J. C. EdmanJEC50 MAT� ade2 J. C. EdmanJEC155 MAT� ura5 ade2 J. C. EdmanJEC156 MATa ura5 ade2 J. C. EdmanJEC170 MAT� ade2 lys2 J. C. EdmanBAC21 MAT� gpa1::ADE2 A. AlspaughBAC37 MAT� gpa1::ADE2 A. AlspaughCDC25 MATa pka1::ADE2 ura5 ade2 This studyCDC36 MATa pka1::ADE2 ade2 This studyCDC40 MAT� pka1::ADE2 ade2 This studyCDC43 MAT� pka1::ADE2 ade2 This studyCDC68 MAT� pkr1�::URA5 ura5 This studyCDC85 MAT� pka2�::URA5 ura5 This studyCDC89 MAT� pka2�::URA5 ura5 ade2 This studyCDC99 MAT� pka2�::URA5 ura5 This studyCDC101 MATa pka2�::URA5 ura5 This studyCDC103 MAT� pka1::ADE2 pka2�::URA5 ura5 ade2 This studyCDC106 MATa pka1::ADE2 pka2�::URA5 ura5 ade2 This studyCDC127 MAT� pka2�::URA5 ura5 ura3 (FOAr) This studyJKH15 MAT� pka2�::URA5 ura5 ura3/MATa PKA2 ura5 URA3 (diploid) This studyJKH16 MAT� URA5 ade2 lys2/MATa ura5 ADE2 LYS2 (diploid) This studyJKH19 MATa pka2�::ura5 ura5 (FOAr) This studyJKH20 MAT� pka2�::URA5 ura5 ura3/MATa pka2�::ura5 ura5 URA3 (dikaryon) This studyJKH21 MATa pka2�::URA5 ura5 PKA2 This study

Serotype A strainsH99 MAT� 50F99 MAT� ura5 (FOAr) 69CDC1 MAT� pka1::ADE2 ade2 22JKH4 MAT� pka2�::URA5 ura5 This studyCHM3 MAT� lac1�::NAT This study

VOL. 3, 2004 PKA CATALYTIC SUBUNITS HAVE DIVERGENT FUNCTIONS 15


http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

and CDC25. The serotype D PKA2/pka2� diploid strain JKH15 was generated bymating strains CDC127 (MAT� pka2�::URA5 ura5 ura3) and JEC34 (MATaPKA2 ura5 URA3) and selecting for diploids on minimal YNB medium (67%yeast nitrogen base without amino acids [Difco], 2% glucose, 20 g of Bacto agar[Difco]). The diploid or heterokaryon strains JKH16 and JKH20 were generatedby mating JEC34 (MATa ura5 ADE2 LYS2) � JEC170 (MAT� URA5 ade2 lys2)and CDC127 (MAT� pka2�::URA5 ura5 ura3) � JKH19 (MATa pka2::ura5 ura5URA3), respectively. The resulting prototrophic strains were selected on minimalYNB medium. The diploid or heterokaryon status of the strains was confirmedby using PCR primers specific for the MATa and MAT� alleles of the STE20gene, as described previously (39). The serotype D PKA2 reconstituted strainJKH21 was generated by biolistic transformation of strain JKH19 with plasmidpJH26 containing the wild-type serotype D PKA2 gene. Transformants wereselected on synthetic medium lacking uracil and containing 1 M sorbitol. Geno-type was confirmed both by Southern hybridization and expression analysis toconfirm the presence of the wild-type transcript in the reconstituted strain.

Isolation of the C. neoformans PKA1, PKA2, and PKR1 genes. The PKA1 genefrom serotype D was isolated by PCR with JEC20 genomic DNA as the templateand the serotype A primers JOHE3104 (AAAGGATCCTAATGTTCCAAAAGGTGTCCG) and JOHE3098 (AAAGGATCCCTAAAACTCCACGAAGAAATG). The 1.9-kb PCR product was digested with BamHI and inserted intoplasmid pUC19 to yield plasmid pCD18. The PKA1 sequence has been depositedin the National Center for Biotechnology Information (NCBI) GenBank data-base. The serotype D PKA2 gene was isolated by using the U. maydis Uka1polypeptide sequence (24) to perform a BLASTp search of the Stanford C.neoformans genome sequence database (October 2000 release, 2.8� coverage;R. Hyman and R. Davis) (http://www-sequence.stanford.edu/group/C.neoformans/index.html). This yielded a 680-bp sequence from a second PKA catalytic sub-unit gene in the C. neoformans genome. Primer pairs based on this sequencewere used for inverse PCR to obtain flanking sequences. Primers JOHE6074(CTTCGATCCGCTGTCAAA) and JOHE6075 (GCCTAGAGAGGAAGGAAA) were used in an inverse PCR with SacI-digested and religated genomicDNA from the serotype D strain JEC21 as the template. Primers JOHE6077(GTGGATACCGCTGGAAAT) and JOHE6078 (ATCGCAATATGGAGGGTG) were used in an inverse PCR with NheI-digested and religated genomicDNA from the serotype D strain JEC21 as the template. PCR products werecloned and sequenced, revealing sequence similarity to known PKA catalyticsubunits. Based on the sequence generated by the inverse PCR strategy, the PCRprimers JOHE6195 (TCCTTCACATGTCGATCC) and JOHE6079 (CACCCTCCATATTGCGAT) were utilized in a PCR with genomic DNA from the sero-type D strain JEC21 as the template. The resulting 1.6-kb PCR product was usedto screen a size-selected library of �4-kb ClaI fragments of JEC21 genomic DNAcloned in pBluescript. One positive clone (pCD49) was identified and sequenced.The serotype D PKA2 gene was independently isolated from a JEC21 bacterialartificial chromosome (BAC) library. An �4-kb EcoRI fragment containing thePKA2 gene was isolated and cloned into plasmid pJAF7, a transformation vectorcontaining the URA5 selectable marker, to yield plasmid pJH26.

The serotype D PKR1 gene from JEC21 was isolated by using primersJOHE3245 (TTCCTCCACAGCTCTACTC) and JOHE3180 (CTGTTGGAGAAAATCAGGG). The 2.5-kb PCR product was cloned in the vector pCRIITopo(Invitrogen) to yield plasmid pCD28. The PKR1 sequence has been deposited inthe NCBI GenBank database.

The PKA2 gene from the serotype A strain H99 was isolated by PCR withgenomic DNA from the serotype A strain, H99, and the serotype D PKA2primers JOHE6905 (GCTGATCAAGGTACATCC) and JOHE6515 (GCCTTGTTGTTTGCGAAG). An �2.5-kb product was amplified and cloned into thepCR2.1 vector (Invitrogen) to yield plasmid pCD80. The PKA2 serotype Asequence has been deposited in the NCBI GenBank database.

Disruption of the PKA1, PKR1, and PKA2 genes. To disrupt the PKA1 gene inserotype D, a 2.5-kb BamHI fragment containing the ADE2 gene from plasmidpRCD28 (59) was inserted into a unique BglII site within the 1.9-kb fragmentbearing the PKA1 gene in plasmid pCD18 to yield plasmid pCD24. To increasethe efficiency of homologous recombination, the termini of a 5-kb BamHI frag-ment containing the pka1::ADE2 allele were treated with terminal transferaseand ddATP. This fragment was then used to transform the ade2 ura5 auxotrophicstrain JEC156 to adenine prototrophy by biolistic transformation with previouslydescribed methods (65). Stable transformants were selected on synthetic mediumlacking adenine and containing uracil and 1 M sorbitol. pka1::ADE2 mutantstrains were identified by PCR, confirmed by Southern hybridization, and ob-tained at a frequency of 6% (3 of 47).

The serotype D PKR1 and PKA2 genes were disrupted by PCR overlap (19).To construct the pkr1�::URA5 allele, in the first round, the 5� end of the PKR1gene was amplified with primers JOHE3245 and JOHE5951 (GGTCGAGCAA

CTTCGCTCCGTATTTTCGTCCTCTTC), the 3� end of the gene was amplifiedwith primers JOHE5952 (CCACCTCCTGGAGGCAAGCTCGGGCACTCCTGTATT) and JOHE3180, and the URA5 gene was amplified with primersJOHE5950 (GAAGAGGACGAAAATACGGAGCGAAGTTGCTCGACC) andJOHE5953 (AATACAGGAGTGCCCGAGCTTGCCTCCAGGAGGTGG).Primers JOHE3245 and JOHE3180 were then used in an overlap amplificationwith the first three products as templates to yield the 4.5-kb PCR product bearingthe pkr1�::URA5 allele. The pkr1�::URA5 allele deletes 11 bp of the PKR1 geneand the URA5 insertion truncates the Pkr1 protein, resulting in the loss of bothcAMP binding sites. The PCR product was enzymatically treated with terminaltransferase and ddATP to dideoxyadenylate the termini and then directly intro-duced into the MAT� ura5 serotype D strain JEC43 by biolistic transformation(65). Stable transformants were selected on synthetic medium lacking uracil andcontaining 1 M sorbitol. A mutation in the PKR1 gene was obtained at a fre-quency of �1% (1 of 96) and was confirmed by PCR and Southern hybridization.

To construct the serotype D pka2�::URA5 allele, in the first round, the 5� endof the PKA2 gene was amplified with primers JOHE6195 (ATGGAGGGTGTTGTCGAA) and JOHE6148 (GGTCGAGCAACTTCGCTCCAAGAACGATAGACGCGA), the 3� end of the gene was amplified with primers JOHE6145(CCACCTCCTGGAGGCAAGGCAAGACCGTACATTCAC) and JOHE6079(CACCCTCCATATTGCGAT), and the URA5 gene was amplified with primersJOHE6144 (TCGCGTCTATCGTTCTTGGAGCGAAGTTGCTCGACC) andJOHE6146 (GTGAATGTACGGTCTTGCCTTGCCTCCAGGAGGTGG). Prim-ers JOHE6195 and JOHE6079 were then used in an overlap amplification with thefirst three products as templates to yield the 3-kb pka2�::URA5 allele. The 114-bpdeletion of PKA2 in the pka2�::URA5 allele results in truncation of the Pka2 proteinwith subsequent loss of the kinase active site motif. The ddATP-blocked PCRproduct was then directly introduced into the ura5 serotype D strain JEC155 bybiolistic transformation. Stable transformants were selected on synthetic mediumlacking uracil and containing 1 M sorbitol. Mutants with a null PKA2 gene wereobtained at a frequency of 3% (3 of 102). This was confirmed by Southern hybrid-ization and PCR with primers JOHE6143 (TCGCGTCTATCGTTCTTG) andJOHE6147 (GTGAATGTACGGTCTTGC). The pka1 pka2 double mutants wereisolated from progeny of a cross between the pka1 mutant strain CDC25 and thepka2 mutant strain CDC89.

The serotype A PKA2 gene was identified by using the serotype D PKA2sequence in a BLASTn search of the Duke Bioinformatics website (December2001 release; http://cneo.genetics.duke.edu/menu.html). The pka2�::URA5 dis-ruption allele was engineered by using the PCR overlap technique. The 5� end wasamplified by using primers JOHE7873 (CACCATCAATCATCAAGC) andJOHE7875 (GGTCGAGCAACTTCGCTCCAAGGACGATAGACGCGA). The3� end of the gene was amplified using primers JOHE6145 (see above) andJOHE7876 (TGGTATTGAGGAGAGGTG). The URA5 gene was amplified withprimers JOHE6146 (see above) and JOHE7874 (TCGCGTCTATCGTCCTTGGAGCGAAGTTGCTCGACC). Primers JOHE7873 and JOHE7876 were then usedin an overlap amplification with the first three products to yield a 3-kb pka2�::URA5allele. The ddATP-blocked PCR product was then directly introduced into the ura5serotype A strain H99 FOAr (F99) by biolistic transformation. Transformants wereselected on synthetic medium lacking uracil and containing 1 M sorbitol. Mutantswith a null PKA2 gene were obtained at a frequency of �8% (2 or 24). Deletionmutants were confirmed by Southern hybridization and PCR. Primers used fordistinguishing wild-type and pka2�::URA5 serotype A alleles were JOHE8192 (GGACCAGATCAATGTCTATA) and JOHE8193 (TTGTGAGAGAACGATCTCTG).

Identification of mutant alleles. Genomic DNA was isolated as describedpreviously (50). The DNA was digested with restriction enzymes, subjected toagarose gel electrophoresis, and analyzed by Southern blotting (54). The radio-labeled probe was synthesized with the Prime-It II DNA labeling kit (Stratagene)and [32P]dCTP (Amersham). The template for the serotype D PKA1 probe wasa 1.6-kb BamHI cDNA fragment of the PKA1 gene, and the template for theserotype D PKR1 gene was a 1.6-kb BamHI cDNA fragment. Low-stringencyhybridization to determine the presence of additional PKA catalytic subunitgenes was carried out by hybridization at 50°C overnight in Church buffer (U.S.Biochemicals) followed by three washes in 2� SSC (1� SSC is 0.15 M NaCl plus0.015 M sodium citrate)–0.2% sodium dodecyl sulfate at room temperature for15 min. The template for the serotype A PKA2 probe used in identifyingpka2�::URA5 deletion strains was an �2.5-kb EcoRI fragment from plasmidpCD80 (described above). DNA was probed by using a nonradioactive alkalinephosphatase kit (Amersham) according to the manufacturer’s instructions.

Expression analysis. Each strain was inoculated into 5 ml of yeast extract-peptone-dextrose medium and grown overnight at 30°C. Five milliliters of yeastextract-peptone-dextrose medium in 125-ml flasks was inoculated with 200 �l ofthe overnight cultures and grown at 250 rpm and 30°C for 5 h prior to harvesting.

16 HICKS ET AL. EUKARYOT. CELL


http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

RNA was isolated from the harvested cultures with Trizol (Gibco-BRL) byfollowing the manufacturer’s instructions. Fifteen micrograms (as indicated byspectrophotometric measurement) of RNA was separated on a denaturing geland transferred to a nylon membrane. The resulting blots were probed with a548-bp PKA2-specific PCR product and a 647-bp PKA1-specific PCR fragment.We note that a small transcript was observed in the pka1 mutants of bothserotype A and serotype D.

Fusion assays. Cell fusion efficiency was measured by mixing 2 � 106 cells ofstrains JEC34 and JEC170, JKH19 and JKH170, and JKH19 and CDC127. A5-�l drop of the cell suspension was inoculated onto V8 medium (pH 7.0) andallowed to incubate for 24 h at 25°C in the dark. Four plates were prepared foreach of the three cell mixes. The cells were then resuspended in 1 ml of double-distilled H2O, and 200 �l of the suspension (�105 total cells) was plated ontoYNB medium. The number of colonies on each plate was determined after 3 days(1).

Microscopy. All images of fusion, mating, haploid fruiting, and confrontationassays were captured with a Nikon Eclipse E400 microscope equipped with aNikon CoolPix digital camera. Differential interference microscopy and fluores-cent images were taken with a Zeiss Axioskop 2 Plus fluorescent microscopeequipped with an AxioCam MRM digital camera.

To visualize nuclei, diploid cells were grown in 10 ml of YNB liquid mediumovernight at 37°C. The cells were then pelleted, washed twice with phosphate-buffered saline (PBS), and incubated for 15 min in cold 70% ethyl alcohol to fixthe cells. The cells were again washed twice with PBS and permeabilized in PBScontaining 1% Triton X-100. Following two more PBS washes, the cells wereresuspended in 10 �l of PBS. One microliter of this suspension was mixed with1 �l of 4�,6�-diamidino-2-phenylindole (DAPI) mix (2 �g of DAPI/ml, 1 mg ofantifade/ml, and 45% glycerol). The cells were then observed with a 40� objec-tive. The percentage of dikaryotic cells in each diploid was determined by ran-domly examining 300 DAPI-stained cells from each diploid and then countingthe number of those cells that were dikaryotic.

Virulence assays. Groups of 10 DBA mice were intranasally infected with 5 �104 yeast cells of the isogenic serotype D wild-type strain (JEC21) and the pka1(CDC40), pka2 (CDC99), pka1 pka2 (CDC103), and pkr1 (CDC68) mutantstrains. Mice were anesthetized by intraperitoneal administration of phenobar-bital and suspended by the incisors on a silk thread, and 50 �l of inocula wasslowly dispensed directly into the nares. The mice were fed ad libitum andmonitored with twice-daily inspections. Mice that appeared moribund or in painwere sacrificed by CO2 inhalation. Statistical analysis was performed by theKruskal-Wallis test for significance (P values of �0.05 were considered signifi-cant).

Capsule assays. To document capsule and cell size, each strain, except for theserotype D diploid strains, was inoculated into 5 ml of MEM or LIM containingan iron chelator and grown at 30°C for 5 days. Twenty cells from each culturewere chosen at random, and the cell and capsule sizes were measured by usingAxiovision 3.0 software (Carl Zeiss Microscopy). The serotype D diploid strains(JKH15, JKH16, and JKH20), the serotype D PKA2 reconstituted strains(JKH21, JKH22, and JKH23), and the appropriate control strains were inocu-lated onto plates of MEM and grown at 30°C for 5 days. Aliquots of all cultureswere stained with India ink and viewed with a 100� objective.

Laccase assays. To quantitate melanin production in serotype A, 108 cells ofthe wild-type strain (H99) and the pka1 (CDC1), pka2 (JKH4), and lac1 (CHM3)mutant strains were grown in 25 ml of L-DOPA medium (0.1% L-asparagine,0.1% glucose, 0.3% KH2PO4, 0.01% L-DOPA, 0.025% MgSO4 � 7H2O [pH 5.6])at 30°C and 250 rpm for 16 h. The cultures were then incubated at 25°C and 250rpm for 24 h, then 1 ml of culture was centrifuged, and the supernatant wasspectrophotometrically read for optical density at a wavelength of 475 nm.

Nucleotide sequence accession number. The serotype D PKA1, serotype DPKR1, serotype D PKA2, and serotype A PKA2 sequences have been depositedin the NCBI GenBank database and given accession no. AF481770, AF481771,AF481772, and AY144605, respectively.

RESULTS

Identification of a second cAMP-dependent protein kinasecatalytic subunit in C. neoformans. S. cerevisiae and U. maydisboth express multiple PKA catalytic subunits that have over-lapping, redundant, or opposing functions. We sought to es-tablish whether C. neoformans also expresses multiple PKAsubunits and, if so, what function each subunit performs. It waspreviously demonstrated that the Pka1 catalytic subunit con-

trols virulence factor production and is necessary for patho-genesis (22). Here we identified a second PKA catalytic sub-unit, Pka2, and analyzed the functions of Pka1 and Pka2 inboth serotypes A and D.

Sequences of both the PKA1 gene and a gene encoding asecond PKA catalytic subunit homolog (PKA2) were identifiedin the Stanford C. neoformans genome sequence database(Hyman and Davis; http://www-sequence.stanford.edu/group/C.neoformans/index.html). The sequences for the PKA2 genewere found by a BLAST search of the database by using thecAMP-dependent protein kinase Uka1 protein sequence fromU. maydis. The serotype A and D Pka1 and Pka2 proteinsequences were aligned and compared to the Uka1 catalyticsubunit (Fig. 1A). A phylogenetic tree (Fig. 1B) revealed thatthe Pka1 subunit was similar to the S. cerevisiae Tpk2 as well asthe catalytic subunits from C. albicans, Aspergillus nidulans,and U. maydis, whereas the C. neoformans Pka2 subunit wasmuch more distantly related to these other catalytic subunits.Inter- and intravarietal comparisons of the PKA catalytic sub-units revealed that while the serotype A Pka1 and Pka2 pro-teins share 93% identity with their respective counterparts inserotype D, there is only 33 to 35% identity shared between thePka1 and Pka2 subunits, either within or between varieties.

cAMP-dependent protein kinase is not essential in C. neo-formans. To determine the biological functions of PKA in se-rotype D, the PKA1 gene was disrupted by inserting the ADE2gene into the kinase domain at the same site at which theserotype A PKA1 gene was previously disrupted. A large re-gion of the kinase domain of the PKA2 gene was deleted andreplaced with a URA5 cassette by PCR overlap (Fig. 2A).These mutations truncate the Pka1 and Pka2 proteins to re-move most of the catalytic region to result in a complete loss ofenzymatic activity. The pka1::ADE2 and pka2�::URA5 alleleswere introduced into recipient strains MATa ura5 ade2(JEC156) and MAT� ura5 ade2 (JEC155), respectively, by bi-olistic transformation and selection of ADE� or URA� iso-lates. pka1 and pka2 mutants in which gene replacement with-out ectopic integration had occurred were identified by PCRand Southern blotting (data not shown). pka1 pka2 doublemutants were readily isolated from a cross between compatiblepka1 and pka2 mutant strains, and the pka1 and pka2 singlemutants and the pka1 pka2 double mutant strains exhibited noovert growth defects. Northern blot analysis confirmed thatboth the PKA1 and PKA2 genes are expressed and that thepka1 and pka2 mutations lead to a loss in expression of thewild-type messages (Fig. 2B, lanes 1 to 4). In the pka1 disrup-tion strains, a small aberrant transcript was observed (data notshown). To determine whether the transcript affected the ob-served pka1 mutant phenotypes, the PKA1 open reading framewas completely deleted in both serotypes A and D. Phenotypicstudies revealed that the pka1 deletion strains were indistin-guishable from the original disruption strains with regard tomelanin and capsule production and the ability to mate. Thesedata show that the melanin, capsule, and mating phenotypesobserved in the original pka1 insertional mutants were notcaused by a dominant-negative effect resulting from the smalltranscript.

Extensive searches in the Stanford (13� coverage) and TheInstitute for Genomic Research (9.5� coverage; http://www.TIGR.org/tdb/e2k1/cna1/) serotype D databases and the



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

FIG. 1. Two C. neoformans genes encode PKA catalytic subunits. (A) PKA catalytic subunits identified in C. neoformans were aligned with theUka1 protein from the related basidiomycete fungal pathogen U. maydis with the CLUSTALW feature of MacVector (Oxford Molecular).Conserved residues are boxed, with identities shown in bold letters and darkly shaded boxes, and similarities are indicated by lightly shaded boxes.(B) The evolutionary relationships between the PKA catalytic subunits of C. neoformans, U. maydis, S. cerevisiae, A. nidulans, and C. albicans werecompared by using the neighbor-joining function of the ClustalX program (62). A phylogenetic tree was then constructed with the TreeViewprogram (46). The proteins are positioned on the tree according to sequence similarity. Evolutionary dissimilarity is proportional to the horizontaldistance joining any two proteins on the tree.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

Whitehead Institute and Duke University serotype A database(11� coverage) did not reveal any additional PKA catalyticsubunits. In summary, these findings demonstrate that thecAMP-dependent protein kinase is dispensable for viability inC. neoformans.

Pka2 is required for virulence factor production in serotypeD. C. neoformans elaborates two specialized virulence factors,capsule and melanin, which both contribute to virulence. Be-cause signaling by cAMP and Pka1 promotes melanin andcapsule production in C. neoformans serotype A strains (22),we tested whether PKA plays a similar role in serotype D. Inresponse to carbon source limitation, C. neoformans producesthe enzyme laccase, which oxidizes diphenolic substrates toproduce melanin as a protective antioxidant shell in the cellwall (45, 52, 74). To assay melanin production, colonies werecultured on Niger seed agar. As shown in Fig. 3A, the serotypeD pka2 mutant exhibited a severe defect in melanin produc-tion, whereas the pka1 mutant produced melanin to the same

extent as the wild type. A pka1 pka2 double mutant displayeda melanin defect similar to the pka2 single mutant strain.cAMP restored melanin production in a gpa1 mutant (as pre-viously observed) but not in the pka2 mutant (Fig. 3A), con-sistent with the model that Gpa1, a heterotrimeric G proteinG� subunit, regulates cAMP production, whereas Pka2 isdownstream of the site of cAMP activation. These findingsdemonstrate that the Pka2 catalytic subunit, and not the Pka1subunit, promotes melanin production in serotype D strains.

The Pka2 catalytic subunit was also found to regulate cap-sule production. Capsule is induced in vivo by host conditionsthat can be mimicked in vitro by growing cells in the presenceof 5% CO2–HCO3

� or iron-limiting conditions (28, 67). Thecapsule protects fungal cells from phagocytosis and can bedetected by microscopic examination of cells suspended inIndia ink. Yeast cells were grown in liquid LIM containing aniron chelator for 5 days at 30°C to induce capsule formation. Incontrast to the wild-type and pka1 mutant strains, which pro-

FIG. 2. PKA1 and PKA2 are not expressed in pka1 and pka2 mutant strains. (A) Identical regions of the PKA1 and PKA2 genes were disrupted(PKA1) or deleted (PKA2). In the PKA1 gene, the ADE2 gene was inserted at the region of DNA representing amino acid 325 in both serotypesA and D. In the PKA2 gene, the URA5 gene was used to replace the region representing amino acids 332 to 376 in both serotypes A and D. (B) pka1and pka2 mutants lack their respective wild-type transcripts. PCR fragments generated from regions of the PKA1 and PKA2 genes downstream ofthe inserted ADE2 or URA5 genes, respectively, were used to probe RNA blots. RNA was generated from serotype D wild-type (WT) (JEC21),pka1 (CDC43), pka2 (CDC85), and pka1 pka2 (CDC106) strains and from serotype A wild-type (H99), pka1 (CDC1), pka1 plus PKA1 (CDC13),and pka2 (JKH4) mutant strains. Ethidium bromide-stained rRNA served as a loading control.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

duce capsule, the gpa1, pka2, and pka1 pka2 mutants werehypocapsular (Fig. 3B). cAMP restored capsule production inthe gpa1 mutant but not in the pka2 mutant. In summary, thePka2 catalytic subunit regulates capsule and melanin produc-tion, and the Pka1 subunit does not.

To verify that Pka2 is responsible for the observed pheno-types in serotype D, we isolated a reconstituted pka2 plusPKA2 strain. The genotype of the strain was verified by South-ern hybridization, and the presence of wild-type transcripts andtranscript levels were determined by RNA analysis (data notshown). Melanin and capsule production in the reconstitutedstrain were comparable to those of the wild-type strain (Fig.4A). The mating defect observed in the bilateral pka2 � pka2mating (see below) was also complemented in the reconsti-tuted strain (Fig. 4B).

PKA is not required for virulence of C. neoformans serotypeD strains. We next tested whether Pka2 is required for viru-lence of C. neoformans in a murine model of cryptococcosis.Mice were infected by inhalation with 5 � 104 cells of the wildtype, the pka1 and pka2 single mutants, the pka1 pka2 doublemutant, and the pkr1 mutant strain, and animal survival wasmonitored over a period of 100 days (Fig. 3C). In this model,the cryptococcus infection initiates in the lungs and hematog-enously disseminates to the brain and other tissues (spleen andkidney), and the majority of animals develop severe hydro-cephalus and die from central nervous system infection. Theaverage survival of mice infected with the wild-type strain was61.6 days compared to 68 days for the pka1 mutant (P 0.1),65.6 days for the pka2 mutant (P 0.21), and 62.4 days for thepka1 pka2 mutant (P 0.88) (Fig. 3C). Thus, no significantdifferences were found in survival rates of mice infected withthe mutants in comparison to those of mice infected with thewild-type strain. Therefore, we conclude that neither Pka1 norPka2 is essential for virulence in the murine intranasal modelof cryptococcosis in the serotype D strains of C. neoformans,even though Pka2 contributes to capsule and melanin produc-tion. Additionally, virulence of the pkr1 mutant was not signif-icantly different from that of a wild-type strain (P 0.08) (Fig.3C), indicating that Pkr1 also does not regulate virulence inserotype D strains under the conditions tested.

Pka2 is not necessary for virulence factor production inserotype A. Previous work has shown that, in contrast to sero-type D, the Pka1 catalytic subunit is essential for virulencefactor production in serotype A (22). Therefore, we deter-mined what role, if any, the serotype A Pka2 catalytic subunitplays in these processes. To address this, we deleted the sameregion of the serotype A PKA2 gene as was deleted in serotypeD (Fig. 2A). Northern blots confirmed that the serotype Apka1 and pka2 mutants lacked the wild-type PKA1 and PKA2transcripts, respectively (Fig. 2B). We examined capsule andmelanin formation in the resulting pka2�::URA5 mutantstrain. Melanin production was examined by growth on Nigerseed agar at 30°C (serotype D) or 37°C (serotype A) (Fig. 5A).Quantitation of melanin production in serotype A strains con-firmed that the pka2 strain produced melanin at levels similarto those of the wild type, whereas the pka1 mutant producedlittle or no melanin, similar to a lac1 mutant that is deficient inlaccase production (Fig. 5B).

Capsule production by the serotype A and D wild-typestrains and the pka1 and pka2 mutant strains was examinedfollowing incubation in MEM (28). Cell and capsule size ofserotype A and serotype D pka1 and pka2 mutants were quan-titated by measuring both cell diameter and capsule radius. Incontrast to the severe hypocapsular phenotype observed inserotype D pka2 mutants, serotype A pka2 mutants producedwild-type levels of capsule and the serotype A pka1 mutantsexhibited a clear defect, as previously reported (22). Thus, thefunctions of the Pka1 and Pka2 catalytic subunits have beenreversed during the divergence of serotypes A and D.

Pka2 is required for mating and haploid fruiting in serotypeD. The role of the cAMP-dependent protein kinase in matingin serotype D strains was investigated. pka1 and pka2 singlemutants and pka1 pka2 double mutant cells were capable ofmating with a wild-type strain of the opposite mating type(data not shown). Next, bilateral mutant crosses were per-

FIG. 3. The Pka2 catalytic subunit controls virulence factor pro-duction, but not virulence, in serotype D. (A) Melanin production wasassessed by growing cells of serotype D wild-type (WT) (JEC21), gpa1(BAC21), pka1 (CDC40), pka2 (CDC99), pka1 pka2 (CDC103), andpkr1 (CDC68) mutant strains on Niger seed (Guizotia abyssinica) andincubating them for 3 days at 30°C. (B) Capsule production was as-sessed in wild-type (JEC21), gpa1 (BAC37), pka1 (CDC40), pka2(CDC99), and pka1 pka2 (CDC106) mutant strains. Cells were grownin LIM in the presence or absence of cAMP at 30°C for 5 days, andcapsules were detected with India ink. Magnification, �1,000.(C) Groups of 10 DBA mice were infected with 5 � 104 yeast cells ofwild-type, pka1, pka1 pka2, and pkr1 mutant strains by inhalation. Thepercent survival was plotted over 100 days. The average survival ofanimals infected with the wild-type strain was 61.6 days, compared to68 days for the pka1 mutant (P 0.1), 65.6 days for the pka2 mutant(P 0.21), and 62.4 days for the pka1 pka2 mutant (P 0.88).



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

formed in which MAT� pka1 mutant cells were cocultured withMATa pka1 cells and MAT� pka2 mutant cells were incubatedwith MATa pka2 cells on V8 mating medium. In the pka2 �pka2 bilateral mutant mating, following coincubation for 2days, we observed the production of short mating filamentswith long chains of basidiospores (Fig. 6A). After several moredays, the basidial heads and basidiospores were overgrown byvegetative growth. This is in contrast to wild-type and pka1 �pka1 matings that formed abundant filaments and basidio-spores (Fig. 6A). pka1 pka2 � pka1 pka2 bilateral matingswere morphologically similar to the pka2 � pka2 mutantcrosses (Fig. 6A). These observations indicate that Pka2 isrequired in at least one partner for wild-type mating in C.neoformans serotype D, whereas Pka1 is not.

We next tested whether the pka2 mutation results in a cellfusion defect. To address this, cell fusion assays were per-formed in which equal numbers of cells from two marked

strains (wild type � wild type, wild type � pka2, or pka2 �pka2) were cocultured on V8 medium (pH 7.0) for 24 h priorto resuspension in water and plating on YNB medium. After 3days of incubation at 25°C, the number of colonies on eachplate were counted. The relative fusion efficiency betweenPKA2 � PKA2, PKA2 � pka2, and pka2 � pka2 crosses wasdetermined and calculated by normalization to the PKA2 �PKA2 fusion. The absolute fusion efficiency of the wild-typemating, based on the 4 replica plates, was �1.2% (499 colo-nies/4 � 105 total cells plate). This is similar to previouslyreported fusion efficiencies for C. neoformans wild-type strains(20). Figure 7B shows that in contrast to the colonies producedby the PKA2 � PKA2 or the PKA2 � pka2 fusion products, thecolonies resulting from the pka2 � pka2 cross were morpho-logically defective, producing basidia and basidiospores fromvery short filaments, and the fusion efficiency was reduced by�5-fold (Fig. 7B).

To determine whether the pka2 mutation results in a nuclearfusion defect that is responsible for the observed aberrantmating in a bilateral cross, we utilized strains with selectablemarkers to create a homozygous wild-type diploid (PKA2/PKA2), a presumptive homozygous pka2 mutant diploid (pka2/pka2), and a heterozygous PKA2/pka2 diploid. In addition toobserving thermal dimorphism (filamentation at 25°C andyeast cells at 37°C) that is characteristic of diploid strains (57),we used MATa- and MAT�-specific PCR primers to confirmthat both mating types were present in the resulting strains(data not shown). DAPI staining of the diploid cells revealedthat while only a small number of the homozygous PKA2/PKA2diploid cells and the heterozygous PKA2/pka2 diploid cellswere dikaryotic (7 and 3%, respectively), 18% of the cells fromthe homozygous pka2 � pka2 mutant fusion product weredikaryotic (Fig. 7A, compare panels 2 and 3 with panel 4).Examination of the parental pka2� cells revealed that thesecells were uninucleate. These data suggest that PKA2 promotesboth nuclear fusion and filamentation.

Next, we addressed whether a loss of pheromone productionor an inability to respond to pheromones contributed to themating defect observed in a pka2 bilateral mutant cross. Cellsof pka1 and pka2 single mutants and pka1 pka2 double mutantswere confronted with either wild-type cells or the correspond-ing mutant strain of the opposite mating type on FA mediumin the dark at room temperature. Figure 6B shows the resultsof the confrontation assays with the pka2 mutant strains. FAmedium provides nitrogen starvation and desiccation signalsthat support production and response to pheromones. In thenormal response to pheromones, MAT� strains produce abun-dant conjugation tubes after 2 days (data not shown) andinitiate haploid fruiting upon longer incubation (Fig. 6C).MATa cells produce fewer conjugation tubes and dramaticallyenlarge in response to pheromones. We found that pka1 mu-tants responded normally (data not shown), whereas pka2 mu-tants exhibited greatly reduced filamentation in a bilateral con-frontation. An attenuation in pheromone response was alsoobserved in wild-type cells confronted with pka2 mutants.These data suggest that Pka2 is essential for pheromone re-sponse and is partially required for efficient pheromone pro-duction, whereas Pka1 is not. MAT� cells produce hyphal fil-aments and basidiospores in response to nitrogen starvationand desiccation through a process known as monokaryotic or

FIG. 4. A serotype D wild-type (WT) PKA2 gene restores matingand capsule and melanin production in a pka2 mutant. (A) Serotype Dwild-type (JEC21) and pka2 (JKH19) mutant strains and the pka2 plusPKA2 reconstituted strain (JKH21) were grown on Niger seed agarmedium and incubated for 3 days at 30°C (upper panel). The samestrains were also grown on a MEM plate for 3 days at 30°C, andcapsule production was detected with India ink (lower panel). Magni-fication, �1,000. (B) Serotype D wild-type (JEC34), pka2 (JKH19),and the pka2 plus PKA2 reconstituted (JKH21) strains were coculturedwith either a wild-type (JEC170) or pka2 (CDC127) strain on V8 agarmedium for 3 days at room temperature, and edges of the matingmixture were photographed. Magnification, �100. In the pka2 � pka2mating, an arrow points to a basidium produced on a short filament.The indicated basidium is shown at higher magnification (�200) in theinsert.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

haploid fruiting. As shown in Fig. 6C, wild-type and isogenicpka1 mutants undergo haploid fruiting, whereas pka2 mutantsdo not.

DISCUSSION

A highly conserved G� protein-cAMP-PKA signaling path-way controls differentiation and virulence of serotype A patho-genic isolates of C. neoformans (2, 3, 21). In this report weshow that components of the PKA pathway have differentfunctions in serotype A and D strains of C. neoformans. Inserotype A, the Pka1 catalytic subunit is required for melaninand capsule production and virulence while the Pka2 subunit isdispensable for these processes (Fig. 5). In contrast, serotypeD strains do not require Pka1 to regulate capsule and melaninproduction, but instead, Pka2 has assumed this regulatory role.The addition of exogenous cAMP did not restore melaninproduction and capsule formation in pka2 mutant strains, in-dicating that Pka2 is the target of cAMP signaling (Fig. 3A andB). Thus, although different PKA catalytic subunits play dis-tinct roles in serotype A and D strains, the function of the PKApathway is conserved with respect to virulence factor produc-tion. Surprisingly, PKA is not required for virulence in sero-type D, whereas PKA plays a prominent role in virulence inserotype A (Fig. 3C) (22).

PKA promotes melanin and capsule production in serotypeD, but the pka2 mutant strains still produce some melanin andcapsule and are thus hypocapsular and not acapsular (Fig. 5).Melanin contributes to but is not essential for virulence, andeven albino strains lacking laccase are capable of lethal infec-tion (53). The difference in virulence between serotype Astrain H99, in which PKA controls virulence, and the serotypeD strain JEC21, in which PKA does not control virulence, aredramatic. Nintety-five percent of all C. neoformans infections arecaused by serotype A strains (12). Furthermore, it has been re-ported that frequently those infections that have been attributedto serotype D strains are likely the result of serotype AD hybrids(8). Differences in capsule and melanin content and productionmay contribute to the observed differences in virulence.

Pka2 is required for both mating and haploid fruiting inserotype D strains, whereas Pka1 is not required for eitherprocess in serotype D but is essential for mating in serotype A(Fig. 6A and C) (22). The Pka2 requirement in mating includesmorphological changes involved in pheromone sensing, celland nuclear fusion, and also possibly a role in promoting pher-omone production and hyphal maturation (4, 69). The PKApathway may, for example, promote expression of elements ofthe pheromone response pathway during nutritional limitation.Thus, it appears that the cAMP signaling pathway regulates

FIG. 5. Functions of Pka1 and Pka2 in capsule and melanin biosynthesis are reversed in serotype A and D. (A) Serotype D wild-type (WT)(JEC21), pka1 (CDC43), and pka2 (CDC85) mutant strains and serotype A wild-type (H99), pka1 (CDC1), and pka2 (JKH4) mutant strains weregrown on Niger seed agar and incubated for 3 days at 30°C (sero D) or 37°C (sero A). Melanin production was not observed in the serotype Dpka2 strain or the serotype A pka1 strain but was produced in the wild type and produced at slightly elevated levels in the serotype D pka1 andserotype A pka2 strains. (B) Melanin production in serotype A strains was quantitated by growing 108 cells of either wild-type (H99), pka1 (CDC1),pka2 (JKH4), or lac1 (CHM3) mutant strains at 250 rpm and 30°C for 16 h in L-DOPA medium prior to transferring cultures to 250 rpm and 25°Cfor 24 h. Spectrophotometric readings of the culture supernatants were taken as the optical density at 475 nm: 1 U of laccase optical densityat 475 nm/0.001 (74). Each bar represents an average of the results from four independent replicates with the standard errors of the meansindicated by error bars. (C) Serotype D wild-type (JEC21), pka1 (CDC43), and pka2 (CDC85) mutant strains and serotype A wild-type (H99), pka1(CDC1), and pka2 (JKH4) mutant strains were inoculated in 5 ml of MEM (28), incubated at 30°C for 5 days, and examined with India ink.Magnification, �1,000. (D) Twenty cells from the cultures used for panel B were measured for cell (diameter) and capsule (radius) size. Each barrepresents the average of 20 measurements.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

filamentation in this fungus in parallel with the mitogen-acti-vated protein kinase cascade, reminiscent of the regulation ofpseudohyphal growth in S. cerevisiae (32, 40, 41).

Our findings are compatible with at least two models toexplain the roles of the divergent PKA subunits. In one model,the two PKA subunits act differentially on the same targets,whereas in the second model, they act on different targets.PKA targets in other systems include Hgl1 and Prf1 in U.maydis (23, 29), Flo8 and Sfl1 in S. cerevisiae (16, 48, 49, 51),and Rst2 in Schizosaccharomyces pombe (30). Studies are in

progress to identify PKA targets in C. neoformans to test thesemodels.

Divergence of PKA function is also observed in other fungi.In the yeast S. cerevisiae, the regulatory and catalytic subunitsof PKA are encoded by the BCY1 and TPK1, TPK2, and TPK3genes, respectively (63, 64). PKA regulates growth, metabo-lism, stress responses, entry into stationary phase, andpseudohyphal differentiation (reviewed in references 47 and61). The three PKA isoforms have overlapping functions ingrowth, stress, and stationary phase responses. In pseudohy-

FIG. 6. The Pka2 catalytic subunit promotes mating, pheromone production, and haploid fruiting. (A) To assess mating, MAT� wild-type(JEC21), pka1 (CDC43), pka2 (CDC85), and pka1 pka2 (CDC103) mutant strains were cocultured with a corresponding MATa partner on V8 agarmedium for 4 days at room temperature, and the edges of the mating mixture were photographed. Magnification, �100. An arrow points to chainsof basidiospores produced on short hyphae and a basidium observed in the bilateral mating between pka2 mutant strains. The indicated basidiumis shown at a higher magnification (�200) in the inserts. (B) MATa cells of serotype D wild-type and pka2 mutant strains were confronted witheither a MAT� wild-type or a MAT� pka2 mutant partner on FA at room temperature in the absence of light for 2 days. Morphological responsesto pheromone were photographed. Magnification, �400. (C) To assess haploid fruiting, the MAT� wild-type, pka1, pka2, and pka1 pka2 mutantstrains were grown on FA in the dark for 16 days at room temperature, and edges of the growth patch were photographed. Magnification, �400.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

phal differentiation, the three isoforms have opposing rolesand Tpk2 activates while Tpk1 and Tpk3 repress pseudohyphalgrowth (48, 51). Tpk2 activates pseudohyphal growth by induc-ing expression of the cell wall flocculin gene FLO11. Tpk2activates the Flo8 transcription factor, which positively regu-lates expression of FLO11, and represses Sfl1, a negative reg-ulator of FLO11 (48, 51). Furthermore, genome-wide tran-scriptional profiling revealed that Tpk2 negatively controls iron

uptake genes and positively controls genes involved in treha-lose degradation and water homeostasis (51). Functional spe-cialization of PKA catalytic subunits in pseudohyphal differ-entiation is attributable to subtle differences in the conservedcatalytic regions and not to differences in gene regulation orstructural differences in the more divergent amino-terminalregions of these proteins (48). Additionally, functional diver-sity of PKA isoforms may be due to differences in subcellular

FIG. 7. Pka2 is involved in both nuclear and cell fusion. (A) Nuclear fusion was assessed by engineering homozygous PKA2/PKA2 (panel 2),heterozygous PKA2/pka2� (panel 3), or homozygous pka2�/pka2� diploid or dikaryotic strains (panel 4). Cells were grown in YNB mediumovernight at 37°C prior to fixation and staining with DAPI (top panel). White arrows in panel 4 indicate examples of dikaryotic cells. Cells froma wild-type haploid strain (JEC34) are shown in panel 1. Differential interference microscopy images of the same cells are shown in the bottompanel. Cells were photographed at a magnification of �400. The percentages indicate the number of dikaryotic cells that were found in a total of300 cells that were examined per diploid strain. (B) Cell fusion efficiency was examined between wild-type (JEC34) and wild-type (JEC170) strains,wild-type (JEC170) and pka2� (JKH19) strains, or pka2� (JKH19) and pka2� (CDC127) strains. Equal numbers of cells were cocultured on V8medium (pH 7.0) overnight. Cells were then transferred to YNB medium, grown at 25°C for 3 days on YNB agar medium, and photographed.Magnification, �100. The number of fusion products on each plate were also counted. Four replica plates were generated for each fusion. Thearrow points to a basidium produced by the pka2 � pka2 fusion product (insert). Magnification, �200. The percentage of cell fusion wasnormalized to the wild-type � wild-type colony number and is listed as a percentage below the fusion product images.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

localization. Inactivation of all three isoforms of the PKAcatalytic subunit in S. cerevisiae results in loss of cell viability,indicating that the three PKA isoforms are functionally redun-dant for growth (64). In contrast, C. neoformans pka1 pka2mutants are viable, indicating that PKA is not essential forgrowth in C. neoformans. In accord with this model, adenylylcyclase mutants are also viable in C. neoformans but inviable inS. cerevisiae (3, 9, 13).

In the human fungal pathogen C. albicans, PKA regulatesthe serum-induced yeast-hypha dimorphic transition (5, 10,11). Ability to undergo the dimorphic transition has beenlinked to virulence and is regulated by the two PKA catalyticsubunit isoforms Tpk1 and Tpk2 (7, 58). The C. albicans Tpk1and Tpk2 proteins have redundant functions in growth andstress responses but exhibit functional differences in morphogen-esis depending on environmental conditions (7). Tpk1 is requiredfor filamentation on solid inducing media, whereas Tpk2 is re-quired for filamentation in liquid inducing media. The functionalspecialization of PKA isoforms in the filamentation response ismediated by the catalytic domains, whereas agar invasion by yeastcells is mediated by the unique N-terminal domain of Tpk2. It isnot yet known what structural features contribute to the func-tional specialization of the PKA isoforms in serotype A and Dstrains of C. neoformans, but domain exchange experiments canbe performed to address this question.

ACKNOWLEDGMENTS

We thank Cristl Arndt for technical assistance; Andy Alspaugh,Toshiaki Harashima, and Xuewen Pan for scientific discussions; JillBlankenship for assistance with genetic crosses and basidiospore dis-section; Chris Martin and Marisol de Jesus-Berrios for the laccasemutant strain; James Fraser for the pJAF7 transformation vector; andConnie Nichols, John Perfect, and Andy Alspaugh for comments onthe manuscript.

This work was supported by NIAID R01 grants AI39115 andAI42159 (J.H.) and P01 award AI44975 from the NIAID to the DukeUniversity Mycology Research Unit. G.C. is a Burroughs WellcomeNew Investigator in Molecular Pathogenic Mycology. J.H. is a Bur-roughs Wellcome Scholar in Molecular Pathogenic Mycology and anAssociate Investigator of the Howard Hughes Medical Institute.

REFERENCES

1. Alspaugh, J. A., L. M. Cavallo, J. R. Perfect, and J. Heitman. 2000. RAS1regulates filamentation, mating and growth at high temperature of Crypto-coccus neoformans. Mol. Microbiol. 36:352–365.

2. Alspaugh, J. A., J. R. Perfect, and J. Heitman. 1997. Cryptococcus neofor-mans mating and virulence are regulated by the G-protein � subunit GPA1and cAMP. Genes Dev. 11:3206–3217.

3. Alspaugh, J. A., R. Pukkila-Worley, T. Harashima, L. M. Cavallo, D. Fun-nell, G. M. Cox, J. R. Perfect, J. W. Kronstad, and J. Heitman. 2002.Adenylyl cyclase functions downstream of the G� protein GPA1 and con-trols mating and pathogenicity of Cryptococcus neoformans. Eukaryot. Cell1:75–84.

4. Ayad-Durieux, Y., P. Knechtle, S. Goff, F. Dietrich, and P. Philippsen. 2000.A PAK-like protein kinase is required for maturation of young hyphae andseptation in the filamentous ascomycete Ashbya gossypii. J. Cell Sci. 113:4563–4575.

5. Bahn, Y. S., and P. Sundstrom. 2001. CAP1, an adenylate cyclase-associatedprotein gene, regulates bud-hypha transitions, filamentous growth, and cyclicAMP levels and is required for virulence of Candida albicans. J. Bacteriol.183:3211–3233.

6. Bennett, J. E., K. J. Kwon-Chung, and D. H. Howard. 1977. Epidemiologicdifferences among serotypes of Cryptococcus neoformans. Am. J. Epidemiol.105:582–586.

7. Bockmuhl, D. P., and J. F. Ernst. 2001. A potential phosphorylation site foran A-type kinase in the Efg1 regulator protein contributes to hyphal mor-phogenesis of Candida albicans. Genetics 157:1523–1530.

8. Boekhout, T., B. Theelen, M. Diaz, J. W. Fell, W. C. J. Hop, E. C. A. Abeln,F. Dromer, and W. Meyer. 2001. Hybrid genotypes in the pathogenic yeastCryptococcus neoformans. Microbiology 147:891–907.

9. Boutelet, F., A. Petitjean, and F. Hilger. 1985. Yeast cdc35 mutants aredefective in adenylate cyclase and are allelic with cyr1 mutants while CAS1,a new gene, is involved in the regulation of adenylate cyclase. EMBO J.4:2635–2641.

10. Brown, A. J., and N. A. Gow. 1999. Regulatory networks controlling Candidaalbicans morphogenesis. Trends Microbiol. 7:333–338.

11. Calderone, R. A., and W. A. Fonzi. 2001. Virulence factors of Candidaalbicans. Trends Microbiol. 9:327–335.

12. Casadevall, A., and J. R. Perfect. 1998. Cryptococcus neoformans. ASMPress, Washington, D.C.

13. Casperson, G. F., N. Walker, and H. R. Bourne. 1985. Isolation of the geneencoding adenylate cyclase in Saccharomyces cerevisiae. Proc. Natl. Acad.Sci. USA 82:5060–5063.

14. Chang, Y. C., L. A. Penoyer, and K. J. Kwon-Chung. 2001. The secondSTE12 homologue of Cryptococcus neoformans is MATa-specific and playsan important role in virulence. Proc. Natl. Acad. Sci. USA 98:3258–3263.

15. Chang, Y. C., B. L. Wickes, G. F. Miller, L. A. Penoyer, and K. J. Kwon-Chung. 2000. Cryptococcus neoformans STE12� regulates virulence but is notessential for mating. J. Exp. Med. 191:871–882.

16. Conlan, R. S., and D. Tzamarias. 2001. Sfl1 functions via the co-repressorSsn6-Tup1 and the cAMP-dependent protein kinase Tpk2. J. Mol. Biol.309:1007–1015.

17. Cox, G. M., H. C. McDade, S. C. Chen, S. C. Tucker, M. Gottfredsson, L. C.Wright, T. C. Sorrell, S. D. Leidich, A. Casadevall, M. A. Ghannoum, andJ. R. Perfect. 2001. Extracellular phospholipase activity is a virulence factorfor Cryptococcus neoformans. Mol. Microbiol. 39:166–175.

18. Cox, G. M., J. Mukherjee, G. T. Cole, A. Casadevall, and J. R. Perfect. 2000.Urease as a virulence factor in experimental cryptococcosis. Infect. Immun.68:443–448.

19. Davidson, R. C., J. R. Blankenship, P. R. Kraus, M. d. Berrios, C. M. Hull,C. D’Souza, P. Wang, and J. Heitman. 2002. A PCR-based strategy togenerate integrative targeting alleles with large regions of homology. Micro-biology 148:2607–2615.

20. Dong, H., and W. Courchesne. 1998. A novel quantitative mating assay forthe fungal pathogen Cryptococcus neoformans provides insight into signallingpathways responding to nutrients and temperature. Microbiology 144:1691–1697.

21. D’Souza, C., and J. Heitman. 2001. Conserved cAMP signaling cascades regu-late fungal development and virulence. FEMS Microbiol. Rev. 25:349–364.

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

23. Durrenberger, F., R. D. Laidlaw, and J. W. Kronstad. 2001. The hgl1 gene isrequired for dimorphism and teliospore formation in the fungal pathogenUstilago maydis. Mol. Microbiol. 41:337–348.

24. Durrenberger, F., K. Wong, and J. W. 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.

25. Fan, M., B. P. Currie, R. R. Gutell, M. A. Ragan, and A. Casadevall. 1994.The 16S-like, 5.8S and 23S-like rRNAs of the two varieties of Cryptococcusneoformans: sequence, secondary structure, phylogenetic analysis and re-striction fragment polymorphisms. J. Med. Vet. Mycol. 32:163–180.

26. Franzot, S. P., I. F. Salkin, and A. Casadevall. 1999. Cryptococcus neofor-mans var. grubii: separate varietal status for Cryptococcus neoformans sero-type A isolates. J. Clin. Microbiol. 37:838–840.

27. Gao, S., and D. L. Nuss. 1996. Distinct roles for two G protein � subunits infungal virulence, morphology, and reproduction revealed by targeted genedisruption. Proc. Natl. Acad. Sci. USA 93:14122–14127.

28. Granger, D. L., J. R. Perfect, and D. T. Durack. 1985. Virulence of Crypto-coccus neoformans: regulation of capsule synthesis by carbon dioxide. J. Clin.Investig. 76:508–516.

29. Hartmann, H. A., R. Kahmann, and M. Bolker. 1996. The pheromoneresponse factor coordinates filamentous growth and pathogenicity in Ustilagomaydis. EMBO J. 15:1632–1641.

30. Higuchi, T., Y. Watanabe, and M. Yamamoto. 2002. Protein kinase A regu-lates sexual development and gluconeogenesis through phosphorylation ofthe Zn finger transcriptional activator Rst2p in fission yeast. Mol. Cell. Biol.22:1–11.

31. Hull, C. M., and J. Heitman. 2002. Genetics of Cryptococcus neoformans.Annu. Rev. Genet. 36:557–615.

32. Kubler, E., H. U. Mosch, S. Rupp, and M. P. Lisanti. 1997. Gpa2p, aG-protein alpha-subunit, regulates growth and pseudohyphal developmentin Saccharomyces cerevisiae via a cAMP-dependent mechanism. J. Biol.Chem. 272:20321–20323.

33. Kwon-Chung, K. J. 1976. Morphogenesis of Filobasidiella neoformans, thesexual state of Cryptococcus neoformans. Mycologia 68:821–833.

34. Kwon-Chung, K. J., and J. E. Bennett. 1984. Epidemiologic differencesbetween the two varieties of Cryptococcus neoformans. Am. J. Epidemiol.120:123–130.

35. Kwon-Chung, K. J., J. E. Bennett, and J. C. Rhodes. 1982. Taxonomic



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

studies on Filobasidiella species and their anamorphs. Antonie Leeuwenhoek48:25–38.

36. Kwon-Chung, K. J., and J. C. Rhodes. 1986. Encapsulation and melaninformation as indicators of virulence in Cryptococcus neoformans. Infect.Immun. 51:218–223.

37. Kwon-Chung, K. J., A. Varma, J. C. Edman, and J. E. Bennett. 1992.Selection of ura5 and ura3 mutants from the two varieties of Cryptococcusneoformans on 5-fluoroorotic acid medium. J. Med. Vet. Mycol. 30:61–69.

38. Kwon-Chung, K. J., A. Varma, and D. H. Howard. 1990. Ecology of Cryp-tococcus neoformans and prevalence of its two varieties in AIDS and non-AIDS associated Cryptococcosis, p. 103–113. In H. Vanden Bossche,D. W. R. Mackenzie, G. Cauwenbergh, J. V. Cutsem, E. Drouhet, and B.Dupont (ed.), Mycoses in AIDS patients. Plenum Press, New York, N.Y.

39. Lengeler, K. B., G. M. Cox, and J. Heitman. 2001. Serotype AD strains ofCryptococcus neoformans are diploid or aneuploid and are heterozygous atthe mating-type locus. Infect. Immun. 69:115–122.

40. Liu, H., C. A. Styles, and G. R. Fink. 1993. Elements of the yeast pheromoneresponse pathway required for filamentous growth of diploids. Science 262:1741–1744.

41. Lorenz, M. C., and J. Heitman. 1997. Yeast pseudohyphal growth is regu-lated by GPA2, a G protein � homolog. EMBO J. 16:7008–7018.

42. Mitchell, T. G., and J. R. Perfect. 1995. Cryptococcosis in the era of AIDS—100 years after the discovery of Cryptococcus neoformans. Clin. Microbiol.Rev. 8:515–548.

43. 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.

44. Moore, T. D. E., and J. C. Edman. 1993. The �-mating type locus of Cryp-tococcus neoformans contains a peptide pheromone gene. Mol. Cell. Biol.13:1962–1970.

45. Nurudeen, T. A., and D. G. Ahearn. 1979. Regulation of melanin productionby Cryptococcus neoformans. J. Clin. Microbiol. 10:724–729.

46. Page, R. D. 1996. TreeView: an application to display phylogenetic trees onpersonal computers. Comput. Appl. Biosci. 12:357–358.

47. Pan, X., T. Harashima, and J. Heitman. 2000. Signal transduction cascadesregulating pseudohyphal differentiation of Saccharomyces cerevisiae. Curr.Opin. Microbiol. 3:567–572.

48. Pan, X., and J. Heitman. 1999. Cyclic AMP-dependent protein kinase reg-ulates pseudohyphal differentiation in Saccharomyces cerevisiae. Mol. Cell.Biol. 19:4874–4887.

49. Pan, X., and J. Heitman. 2002. Protein kinase A operates a molecular switchthat governs yeast pseudohyphal growth. Mol. Cell. Biol. 22:3981–3993.

50. Perfect, J. R., N. Ketabchi, G. M. Cox, C. W. Ingram, and C. L. Beiser. 1993.Karyotyping of Cryptococcus neoformans as an epidemiological tool. J. Clin.Microbiol. 31:3305–3309.

51. 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.

52. Rosas, A. L., J. D. Nosanchuk, M. Feldmesser, G. M. Cox, H. C. McDade,and A. Casadevall. 2000. Synthesis of polymerized melanin by Cryptococcusneoformans in infected rodents. Infect. Immun. 68:2845–2853.

53. Salas, S. D., J. E. Bennett, K. J. Kwon-Chung, J. R. Perfect, and P. R.Williamson. 1996. Effect of the laccase gene, CNLAC1, on virulence ofCryptococcus neoformans. J. Exp. Med. 184:377–386.

54. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: alaboratory manual. 2nd ed. Cold Spring Harbor Laboratory Press, ColdSpring Harbor, N.Y.

55. Shen, W.-C., R. C. Davidson, G. M. Cox, and J. Heitman. 2002. Pheromonesstimulate mating and differentiation via paracrine and autocrine signaling inCryptococcus neoformans. Eukaryot. Cell 1:366–377.

56. Sherman, F. 1991. Getting started with yeast, p. 3–21. In C. Guthrie andG. R. Fink (ed.), Methods in enzymology, vol. 194. Academic Press, Inc., SanDiego, Calif.

57. Sia, R. A., K. B. Lengeler, and J. Heitman. 2000. Diploid strains of thepathogenic basidiomycete Cryptococcus neoformans are thermally dimorphic.Fungal Genet. Biol. 29:153–163.

58. Sonneborn, A., D. P. Bockmuhl, M. Gerads, K. Kurpanek, D. Sanglard, andJ. F. Ernst. 2000. Protein kinase A encoded by TPK2 regulates dimorphismof Candida albicans. Mol. Microbiol. 35:386–396.

59. Sudarshan, S., R. C. Davidson, J. Heitman, and J. A. Alspaugh. 1999. Molecularanalysis of the Cryptococcus neoformans ADE2 gene, a selectable marker fortransformation and gene disruption. Fungal Genet. Biol. 27:36–48.

60. 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.

61. Thevelein, J. M., L. Cauwenberg, S. Colombo, J. H. D. Winde, M. Donation,F. Dumortier, L. Kraakman, K. Lemaire, P. Ma, D. Nauwelaers, F. Rolland,A. Teunissen, P. V. Dijck, M. Versele, S. Wera, and J. Winderickx. 2000.Nutrient-induced signal transduction through the protein kinase A pathwayand its role in the control of metabolism, stress resistance, and growth inyeast. Enzyme Microb. Technol. 26:819–825.

62. Thompson, J. D., F. Gibson, F. Plewniak, F. Jeanmougin, and D. G. Higgins.1997. The CLUSTAL_X windows interface: flexible strategies for multiplesequence alignment aided by quality analysis tools. Nucleic Acids Res. 25:4876–4882.

63. Toda, T., S. Cameron, P. Sass, M. Zoller, J. D. Scott, B. McMullen, M.Hurwitz, E. G. Krebs, and M. Wigler. 1987. Cloning and characterization ofBCY1, a locus encoding a regulatory subunit of the cyclic AMP-dependentprotein kinase in Saccharomyces cerevisiae. Mol. Cell. Biol. 7:1371–1377.

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

65. Toffaletti, D. L., and J. R. Perfect. 1994. Biolistic DNA delivery for Crypto-coccus neoformans transformation, p. 303–308. In B. Maresca and G. S.Kobayashi (ed.), Molecular biology of pathogenic fungal: a laboratory man-ual. Telos Press, New York, N.Y.

66. Toffaletti, D. L., T. H. Rude, S. A. Johnston, D. T. Durack, and J. R. Perfect.1993. Gene transfer in Cryptococcus neoformans by use of biolistic delivery ofDNA. J. Bacteriol. 175:1405–1411.

67. Vartivarian, S. E., E. J. Anaissie, R. E. Cowart, H. A. Sprigg, M. J. Tingler,and E. S. Jacobson. 1993. Regulation of cryptococcal capsular polysaccha-ride by iron. J. Infect. Dis. 167:186–190.

68. Vartivarian, S. E., R. E. Cowart, E. J. Anaissie, T. Tashiro, and H. A. Sprigg.1995. Iron acquisition by Cryptococcus neoformans. J. Med. Vet. Mycol.33:151–156.

69. Wang, P., C. B. Nichols, K. B. Lengeler, M. E. Cardenas, G. M. Cox, J. R.Perfect, and J. Heitman. 2002. Mating-type-specific and nonspecific PAKkinases play shared and divergent roles in Cryptococcus neoformans. Eu-karyot. Cell 1:257–272.

70. Wang, P., J. R. Perfect, and J. Heitman. 2000. The G-protein subunitGPB1 is required for mating and haploid fruiting in Cryptococcus neofor-mans. Mol. Cell. Biol. 20:352–362.

71. Wickes, B. L., M. E. Mayorga, U. Edman, and J. C. Edman. 1996. Dimor-phism and haploid fruiting in Cryptococcus neoformans: association with the�-mating type. Proc. Natl. Acad. Sci. USA 93:7327–7331.

72. Xu, J., R. J. Vilgalys, and T. G. Mitchell. 2000. Multiple gene genealogiesreveal dispersion and hybridization in the human pathogenic fungus Cryp-tococcus neoformans. Mol. Ecol. 9:1471–1481.

73. Yue, C., L. M. Cavallo, J. A. Alspaugh, P. Wang, G. M. Cox, J. R. Perfect, andJ. Heitman. 1999. The STE12� homolog is required for haploid filamenta-tion but largely dispensable for mating and virulence in Cryptococcus neo-formans. Genetics 153:1601–1615.

74. Zhu, X., J. Gibbons, J. Garcia-Rivera, J. Casadevall, and P. R. Williamson.2001. Laccase of Cryptococcus neoformans is a cell wall-associated virulencefactor. Infect. Immun. 69:5589–5596.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

cyclic amp-dependent protein kinase catalytic subunits ...the divergent pka2 catalytic subunit was...

Documents