mutation of the regulator of g protein signaling crg1 ...health sciences center, and research...

EUKARYOTIC CELL, Aug. 2004, p. 1028–1035 Vol. 3, No. 41535-9778/04/$08.00�0 DOI: 10.1128/EC.3.4.1028–1035.2004Copyright © 2004, American Society for Microbiology. All Rights Reserved.

Mutation of the Regulator of G Protein Signaling Crg1 IncreasesVirulence in Cryptococcus neoformans

Ping Wang,1,2,3* Jim Cutler,1,2,3 Jill King,3 and Daniel Palmer3

Departments of Pediatrics1 and Microbiology, Immunology, and Parasitology,2 Louisiana State UniversityHealth Sciences Center, and Research Institute for Children,3 New Orleans, Louisiana

Received 14 April 2004/Accepted 8 June 2004

The regulator of G protein signaling homolog Crg1 was found to be a key regulator of pheromone-responsivemating in the opportunistic human fungal pathogen Cryptococcus neoformans. A mutation in the CRG1 gene hasgreatly increased virulence in the prevalently distributed MAT� strains of the fungus. Mouse survival time wasshortened by 40%, and the lethal dosage was 100-fold less than that of wild-type strains. In addition, the in-creased virulence of crg1 mutant strains was dependent on the transcription factor homolog Ste12� but not onthe mitogen-activated protein kinase homolog Cpk1. The enhanced mating due to CRG1 mutation, however,was still dependent on Cpk1. Interestingly, crg1 mutants of MAT� cells produced dark melanin pigment undernormally inhibitory conditions, which may relate to the mechanism for increased virulence.

Cryptococcus neoformans is a ubiquitous soil-borne basidio-mycetous fungus that is the leading cause of fungal meningitisin high-risk individuals such as those with human immunode-ficiency virus infection, high-dose chemotherapy, or solid or-gan and bone marrow transplants (27, 31). Haploid strains ofthis pathogen are either of the MAT� or MATa mating typesand, of the MAT� strains, the var. grubii accounts for �95% ofclinical and environmental isolates, with only two MATa iso-lates being described thus far (33, 36). The MAT� mating-type-associated virulence has been attributed to components of theMAT� mating-type locus, such as STE20�, whose mutationresulted in attenuation of virulence (38). Mating, haploid dif-ferentiation, and the virulence traits of polysaccharide capsuleand melanin production are regulated via two major G proteinsignal transduction pathways (1, 39). Crg1, a regulator of Gprotein signaling (RGS) protein homolog, was found to have akey role in the pheromone responsive mating of var. grubiiMAT� and var. gattii strains (17, 33). Thus, the identification ofCrg1 has provided a new insight into the role of G proteinsignaling in the physiology of the fungus.

The first RGS protein, initially identified from the baker’syeast Saccharomyces cerevisiae, acts as the GTPase-activatingprotein (GAP) for the G� subunit Gpa1 by terminating itsactivation (3, 5, 6, 13, 22). RGS proteins can also directlyinterfere with the activation of effector molecules (20, 24) orprevent spontaneous receptor-independent G protein signal-ing (34). RGS and RGS-like proteins are widely distributedthroughout eukaryotic organisms and have an impressive arrayof regulatory functions affecting cellular growth and develop-ment (4, 11, 12, 19). The present study reveals a novel role foran RGS protein in a pathogenic microorganism linking en-hanced pheromone response to increased virulence. The re-sulting hypervirulent C. neoformans strain will be of significantimportance in studying the mechanism for fungal virulence.

MATERIALS AND METHODS

Strains and media. C. neoformans var. grubii MAT� strain H99 and MATastrain KN99 and C. neoformans var. neoformans MAT� strain JEC21 and MATastrain JEC20 were previously described (32, 33, 35). KN99 is the tenth backcrossprogeny of H99 and KN5, a progeny of a cross between a var. grubii subset MAT�strain, 8-1, and the MATa 125.91 strain (33). F99 and F99a are Ura� derivativesof H99 and KN99, respectively.

YPD, YNB, 10% V8 juice agar, 5-fluoroorotic acid, and Nigerseed agar wereprepared as described previously (37, 39).

Characterization of the CRG1 gene and disruption of the CRG1 gene in aMATa strain. A 2.9-kb fragment containing the complete CRG1 coding domainwas amplified by PCR from the prototype strain H99 and sequenced. Intronswere determined by reverse transcription-PCR (RT-PCR), and the completecoding domain was determined by 5� and 3� rapid amplification of cDNA ends byusing a commercial kit (Invitrogen).

A 1.3-kb fragment containing the partial CRG1 gene was initially used togenerate a crg1::URA5 gene disruption allele that was, in turn, used to transformthe F99 strain by biolistic transformation (33). The same crg1::URA5 allele wasalso used to create a crg1 mutant strain from the MATa strain F99a. To com-plement crg1 mutant strains of both mating types, the 2.9-kb fragment containingthe wild-type CRG1 gene was inserted in the plasmid pGMC200 containing anourseothricin resistance marker gene (NATR), and the construct was used totransform the crg1::URA5 mutant strains. Complemented strains were verified byPCR amplification and Southern hybridization analysis. Other strains with theNATR marker were constructed by using a similar approach.

To obtain double-mutant strains, the PKA1 gene encoding Pka1 (AF288613),the PKR1 gene encoding Pkr1 (AF288614), the CPK1 gene encoding Cpk1(AF414186), and the STE12� gene encoding Ste12� (AF542529) were all dis-rupted in the crg1::ura5 mutant derivative to obtain crg1 pka1, crg1 pkr1, crg1cpk1, and crg1 ste12� strains. The crg1::ura5 strains were obtained by plating thecrg1::URA5 strain on 5-fluoroorotic acid medium. The derived � crg1::ura5 strainwas again transformed with the pka1::URA5, pkr1::URA5, cpk1::URA5, andste12�::URA5 gene disruption alleles, respectively. Mutant strains were screenedby PCR with respective gene-specific primers and CRG1 gene-specific primers.

The STE12a gene was isolated from the strain 125.91 (29) by using primers5�-TGCGATCAGGCTACAACCTCT and 5�-TAATCACAGTATGCATTGGAT, and the amplified fragment (4.2 kb) was inserted in pGMC200 and used totransform a crg1 ste12� strain. Expression of the STE12a gene was examined byRT-PCR.

Phenotype characterization of crg1 mutant strains. Pheromone response wasrepresented by formation of conjugation tubes which was induced at 22°C bystreaking cells of opposite mating type on filament agar (39). Mating was per-formed by coculturing a MAT� strain with a MATa strain of the same variety(var. grubii, KN99) or a cross variety (var. neoformans, JEC20) on V8 agarcontaining 0.1 or 1.5% glucose. Plates were incubated at 22 and 37°C.

Melanin was induced by growing cells in the diphenolic compound-rich Ni-gerseed agar medium. Melanin-producing strains exhibit dark brown pigment,

* Corresponding author. Mailing address: Department of Pediatrics,Research Institute for Children, Children’s Hospital, 200 Henry ClayAve., New Orleans, LA 70118. Phone: (504) 896 2739. Fax: (504) 894-5379. E-mail: [email protected].

1028

on February 13, 2021 by guest

http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

whereas melanin-defective strains appear as white colonies. The whole-cell lac-case activity was measured to provide a quantitative assessment of melanin with2% azinobis(3-ethylbenzthiazoline-6-sulfonate) (ABTS; Southern BiotechnologyAssociates, Inc.) as the substrate by a previously described method (14). Activ-ities were obtained for cells incubated in minimal asparagine medium withoutglucose for 4 h at 30 and 37°C. One unit of laccase activity was defined as a 0.01absorbance at 10 min (107 cells).

Capsule was induced by growing cells in Dulbecco modified Eagle mediumfor 3 days (18) or overnight in liquid Sabouraud medium (45). Cells grown inSabouraud medium were washed and grown again for 2 days at 30°C in a 10�dilution of the same medium that was buffered with HEPES (pH 7.3) (45). Cellswere stained with India ink and observed with an Olympus microscope (BX51)equipped with a digital camera. Images of capsules were printed, and the thick-nesses of capsule from 100 cells were measured. A Student t test was performedto compare capsules between the wild-type and crg1 mutant strains.

Murine models of cryptococcosis. Cryptococcal yeast cells were washed inphosphate-buffered saline, and 50 �l of suspension each was used to infectgroups of 4- to 6-week-old female mice. For A/JCR mice, groups of 5 and 10mice were used for each strain. For C57BL/6 mice, groups of five mice were usedper strain. For tests with serially diluted fungal cells, groups of four (A/JCR) orthree (C57BL/6) mice per strain were used. Mice were first anesthetized byxylazine-ketamine peritoneal injection, and yeast cells were then applied directlyinto the mouse’s nostrils with a pipette (7, 8). Infected mice were monitoredtwice daily, and those that appeared to be moribund or in pain were sacrificed byCO2 inhalation. Mouse survival was analyzed by the Kaplan-Meier method forsurvival by using Prism 4 (GraphPad Software, Inc., San Diego, Calif.). A P valueof �0.05 was considered statistically significant.

Brain, liver, spleen, and kidney tissues of infected mice were collected,weighed, homogenized with a Dounce glass homogenizer, and plated on YPDmedium after serial dilutions from 1:10 to 1:1,000 were made to determine CFU.Recovered fungal cells were also genotyped for strain verification by PCR. Forhistopathology study, fresh tissues were collected from moribund mice, im-

mersed in buffered neutral formalin, embedded in wax, and stained with theperiodic acid-Schiff reagent. Images were obtained with an Olympus microscope(BX51) equipped with a digital camera.

RESULTS

Crg1 is a key regulator of pheromone response in bothMAT� and MATa strains of var. grubii. The CRG1 gene existsas a single-copy gene in both MAT� and MATa strains ofC. neoformans var. grubii, and it consists of 1,923 nucleotidesencoding a polypeptide of 640 amino acids (GenBank acces-sion no. AY341339). The Crg1 protein shares, respectively, 60,38, and 24% amino acid identity with Schizophyllum communeThn1 (AAF78951), Aspergillus nidulans FlbA (AAA35104),and S. cerevisiae Sst2 (S60771) within the RGS domain (Fig. 1).

A crg1 mutant strain was generated from the MATa strainsF99a (a Ura� derivative of KN99 that is congenic to H99) viabiolistic transformation with the same crg1::URA5 allele thatwas previously used to generate the � crg1 strain (33). In therelated var. neoformans for which congenic strains are readilyavailable, MAT� cells (strain JEC21) responded to MATa(strain JEC20) pheromones by producing conjugation tubes, apheromone response, whereas JEC20 produced swollen cellsin response to JEC21 pheromones. Whereas the H99 strain didnot respond to JEC20 pheromone nor JEC20 to H99, theMAT� crg1 strain produced an abundance of conjugation tubeswhen confronted with JEC20 that also responded by produc-

FIG. 1. C. neoformans Crg1 shares a highly conserved RGS domain with other fungal RGS proteins. The amino acid identity within the RGSdomain between Crg1 (AY341339) and those of S. commune Thn1 (AAF78951), A. nidulans FlbA (AAA35104), and S. cerevisiae Sst2 (S60771)are 60, 38, and 24%, respectively. The S. cerevisiae Sst2 sequence was truncated, as indicated by “#” and “❋,” for alignment purposes.

VOL. 3, 2004 MUTATION OF Crg1 INCREASES VIRULENCE 1029


http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

ing swollen cells (Fig. 2A) (33). JEC21 (var. neoformans) pro-duced conjugation tubes when confronted with the MATa crg1mutant (var. grubii) but not in response to the parental KN99strain (Fig. 2A). When either a MAT� or a MATa crg1 mutantstrain was confronted with a compatible wild-type strain of thesame var. grubii, conjugation tubes rarely formed (Fig. 2A);however, when both MAT� and MATa crg1 mutant strainswere confronted, a profusion of conjugation tubes were ob-served in both mating-type cells (Fig. 2A), demonstrating thatCrg1 negatively regulates pheromone response in both � and acells of var. grubii.

In a qualitative mating assay in which observations weremade at an early stage (3 days) after coculture of compatiblestrains, crosses between crg1 mutant strains (var. grubii) andcompatible wild-type strains of either var. grubii or var. neofor-mans (unilateral) exhibited increased heterokaryotic filamen-tation, leading to basidia and basidiospore formation (data notshown), and the cross between � crg1 and a crg1 mutant strains(bilateral) exhibited the most robust mating phenotype (Fig.2B). In addition, both the unilateral and bilateral crosses thatinvolved at least one crg1 mutant strain(s) exhibited the mating

phenotype at 37°C on V8 agar supplemented with 1 to 1.5%glucose, conditions normally unfavorable for mating (data notshown). Supplementing the � crg1 mutant strain with the wild-type CRG1 gene reduced the mutant’s mating behavior similarto that of the wild-type strain (Fig. 2B).

Mutation of CRG1 increases virulence in MAT� cells. Thewild-type H99, crg1 mutant, and crg1�CRG1 complementedstrains were serially diluted and plated on YPD and YNBmedia for growth at 30 and 37°C. No apparent difference incolony size was observed among these three strains, suggest-ing CRG1 mutation did not result in growth defect (data notshown). When grown in diluted Sabouraud, a medium thatinduces the virulence factor antiphagocytic polysaccharide cap-sule (25, 26), the average size of 100 crg1 mutant cells was 10.70�m (1.54), whereas the average size of 100 wild-type cellswas 10.92 �m (1.32) (P 0.17). However, the average cap-sule for the crg1 mutant cell was 2.75 �m (0.62) thick, whichwas thicker than that of the wild-type cell (2.25 �m [0.55])(P � 0.0001).

Also interestingly, a difference in the production of the an-tioxidant virulence factor melanin (21, 40, 41) was detected in

FIG. 2. C. neoformans Crg1 is a key regulator of pheromone response in MAT� and a strains (A and B) and in melanin formation (C and D).Confrontation tests were performed on filament agar for 3 days at 25°C (A; magnification, �35). Mating was carried out on V8 agar for 3 days at 25°C(B; magnification, �14). Cells were grown on Nigerseed agar to induce melanin for 2 days (30°C) (0.1% glucose, panel C, top row) or 3 days (37°C)(1.5% glucose, panel C, bottom row). �WT and aWT are the wild-type strains of var. grubii H99 and KN99, respectively; JEC21 and JEC20 arethe wild-type strains of the var. neoformans � and a mating types, respectively. Arrows in panel A indicate swollen cells. The lac1 control strainis defective in melanin biosynthesis in panel C. The data for the laccase activity assay was from one experiment in panel D.

1030 WANG ET AL. EUKARYOT. CELL


http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

the � crg1 mutant strain in which melanin production wasshown to occur under conditions unfavorable for productionin the wild-type strain (Fig. 2C). This result was confirmed byquantification of the laccase production that is required formelanin production. The laccase activity was measured in mostof the strains when induced in the minimal asparagine mediumat 30°C but was much more prominent in the crg1 mutantstrain when it was induced at 37°C (Fig. 2D). Two independentcrg1 CRG1 complemented strains exhibited similar intermedi-ate levels of melanin pigment and laccase activity (Fig. 2C andD).

The melanin production by MAT� crg1 mutants under thenormally inhibitory conditions prompted us to test virulence byusing a murine model of cryptococcosis. This model mimickedthe natural infection route since fungal cells were placed di-rectly into nasal cavities, promoting inhalation of the organisminto the lungs and eventual dissemination to the central ner-vous system. The wild-type H99 strain produced 100% mortal-ity in A/JCR mice by day 37 when introduced at 5 � 104 yeastcells per mouse (Fig. 3A). In contrast, 100% mortality oc-curred by day 22 in mice infected with the � crg1 mutant strain(5 � 104 yeast cells per mouse, Fig. 3A and B [P 0.0001]).

Mice infected with the � crg1 mutant complemented with theCRG1 gene exhibited a virulence profile similar to that of H99(45 days, Fig. 3A [P 0.299]), indicating that the shortenedsurvival time of the crg1-infected mice was due to the CRG1mutation.

To ensure that our findings were not mouse strain specific,the virulence test was repeated in C57BL/6 mice. Again, theincreased virulence in MAT� cells due to the CRG1 mutationwas observed (Fig. 3C and data not shown). To further assessthe magnitude of the virulence increase, we inoculated micewith serially diluted H99 and � crg1 mutant cells at concentra-tions of 5 � 104, 5 � 103, 5 � 102, 50, and 5 cells per mouse(Fig. 3C). In general, survival time correlated with the amountof cells inoculated for each strain. As anticipated by the resultsin the A/JCR mice, the � crg1 mutant strain was more virulentthan the wild-type strain H99. At 5 � 102 yeast cells per mouse,100% mortality was achieved at day 30 in crg1-infected miceand at day 46 in the wild-type-infected mice (Fig. 3C, P 0.0295). Survival of mice infected with the crg1 strain at 5 � 102

cells per mouse was the same as that of the mice infected with5 � 104 wild-type cells (Fig. 3C, P 0.0753).

We examined the fungal burden in brains, spleens, livers,

FIG. 3. Mutation of the CRG1 gene dramatically increased virulence in MAT� cells of C. neoformans. Ten female A/JCR mice per strain (Aand D) were infected by nasal inhalation with 50 �l of yeast cells (5 � 104) with the wild-type, crg1, crg1 CRG1, crg1 pka1, crg1 pkr1, and crg1 ste12�strains, and five mice were used in a separate test involving the pkr1 mutant (B). (C) In the test with serial diluted strains, three C57BL/6 micewere used for each strain. Five- to six-week-old mice were used in panels A to C, and 7-week-old mice were used in panel D. Each panel alsorepresents an independent virulence test. Survival was monitored twice daily, and the number of surviving mice was plotted against time. P valueswere calculated by the Kaplan-Meier survival analysis with Prism 4 software.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

and kidneys of moribund mice that were infected, at 5 � 104

cells per mouse, with the wild-type H99 and � crg1 mutantstrains. Due to variances of CFU within the same group, pos-sibly caused by a viability difference in yeast cells from freshlysacrificed mice and those that had already deceased, directcomparison was not possible. Nevertheless, fresh kidneys col-lected from crg1 mutant-infected mice at the lower dose of 50cells per mouse, which had undergone a longer disease period,did reveal large cryptococcoma lesions packed with yeastthroughout the cortical region compared to the kidneys ob-tained from animals that were infected with the same amountof the wild-type strain that showed only occasional and smalllesions in the cortical region with a few cryptococcal yeast cells(Fig. 4). This difference is dramatic; however, whether it re-lates to hypervirulence remains to be determined since bothgroups of mice were moribund when the kidneys were col-lected.

CRG1 mutation-increased virulence is dependent on thetranscription factor homolog Ste12� but not on the mitogen-activated protein (MAP) kinase homolog Cpk1. Previous stud-ies have shown that at least two G protein signaling pathwaysfunction in C. neoformans to regulate mating, filamentation,and virulence (1, 28, 39). The enhanced pheromone responseof the crg1 mutant strains implied that Crg1 functions in themating pathway. Alternatively, the effect of CRG1 mutation onmelanin formation could also suggest that Crg1 functions inthe Gpa1/cyclic AMP (cAMP)-dependent signaling pathwaysince this pathway plays a prominent role in controlling mela-nin and capsule biosynthesis. To test the genetic interactionbetween Crg1 and components of these two pathways, crg1pka1, crg1 pkr1, crg1 cpk1, and crg1 ste12� double-mutantstrains were created for epistasis tests.

The crg1 pka1 strain was defective in melanin productionand was avirulent, suggesting that cAMP signaling is requiredfor the virulence of crg1 strains (Fig. 3A, wild type versus crg1pka1, P � 0.0001). Mutation of PKR1 activated the cAMP

signaling pathway, resulting in formation of cells with largercapsules that also enhanced virulence (Fig. 3B, wild type versuspkr1, P 0.0173) (10). Interestingly, mutation of both CRG1and PKR1 genes did not result in any synergistic effect invirulence because the crg1 pkr1 strain exhibited the virulenceprofile similar to that of the pkr1 strain (Fig. 3A and B).

The crg1 cpk1 mutant strain was defective in conjugationtube formation and mating (Fig. 5A), suggesting that Crg1functions upstream of Cpk1 in the MAP kinase mating path-way. Melanin and laccase at 37°C, however, were not affectedby CPK1 mutation (Fig. 5), and the crg1 cpk1 strain was alsohypervirulent (Fig. 3D, crg1 cpk1 versus wild type, P 0.0002;crg1 cpk1 versus crg1, P 0.372). In contrast, mutation of thetranscription factor STE12� gene in the crg1 strain did notaffect conjugation tube formation and only marginally reducedmating to KN99 (Fig. 5A and B). However, melanin and lac-case activity at 37°C were affected by mutation of STE12� (Fig.5C and D), and the crg1 ste12� strain was not hypervirulent,exhibiting a virulence profile similar to that of the wild-typestrain (Fig. 3A, wild type versus crg1 ste12�, P 0.846). Pre-viously, mutation of the var. grubii STE12� gene had resultedin only a moderate reduction in mating (with the serotype DMATa tester strain JEC20) but had no impact on virulence(44). Finally, the expression of the allelic STE12a gene in thecrg1 ste12� strain resulted in a restoration of robust mating(data not shown), but the crg1 ste12� STE12a reconstitutedstrain remained nearly unchanged in the formation of melaninand laccase activity (Fig. 5C and D).

DISCUSSION

G-protein-mediated signal transduction has evolved to playan important role that allows C. neoformans to sense not onlyexternal cues, such as pheromones, but also host-associatedchanges, such as changes in temperature, alterations in pH andCO2 concentrations, and the hosts’ immune defense, and to

FIG. 4. Histology study of C. neoformans-infected mouse renal tissue. Three sets of kidneys were collected from moribund mice that wereinfected, respectively, with 50 cells of wild-type H99 and MAT� crg1 mutants. Tissues were fixed in neutral buffered formalin, embedded, sectioned,and stained with the periodic acid-Schiff reagent. Arrows indicate cryptococcal yeast cells.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

respond by altering gene expression regulating growth andproliferation. Thus, mutation of CRG1 resulted in enhancedpheromone response and increased virulence in MAT� cellsare likely caused by appropriate alterations of G-protein-me-diated signal transduction events.

Sequence conservation between Crg1 and other fungal RGSfamily proteins, such as S. cerevisiae Sst2, within the RGSdomain suggest Crg1 could have a function in pheromoneresponsive mating. Indeed, mutation of the CRG1 gene re-sulted in enhanced pheromone response in both MAT� andMATa strains of var. grubii (33) (the present study) and in thedistant var. gattii (17). The function of two additional fungalRGS proteins, the S. commune Thn1 and A. nidulans FlbA, inmating is not known, but the corresponding mutant strainsboth exhibited severe defects in vegetative growth (16, 43).Surprisingly, mutation of the CRG1 gene dramatically in-creased virulence in the prevalently distributed and clinicallyimportant MAT� cells. This is in sharp contrast to other sig-naling components of C. neoformans or other model fungi inwhich gene deletion often had no impact or only resulted inattenuation or loss of virulence. Nevertheless, “hypervirulent”strains due to mutation of “anti-virulence” genes (15) have

been reported recently in the bacterium Mycobacterium tuber-culosis and the protozoan parasite Leishmania major (9, 30). InL. major, mutation of the PTR1 gene encoding pteridine re-ductase I resulted in a �50-fold increase over the wild-typestrain in lesion formation in a murine model (9). In fungi,transformation of Colletotrichum coccodes with the NEP1 geneencoding a phytotoxic protein increased its ability to kill theweed Abutilon by a factor of nine (2). In a clinical isolate of thebudding yeast S. cerevisiae, a strain carrying a mutation in theSSD1 gene encoding a 160-kDa cytoplasmic protein thought toalter the cell surface structure exhibited increased virulence inDBA/2 mice due to its ability to overstimulate the proinflam-matory response (42). However, both the clinical and the ssd1mutant yeast strains were avirulent in other mouse strains suchas BALB/c (42). Finally, hypervirulence has also been demon-strated in the pathogenic fungus Candida glabrata, in which amutation of the S. cerevisiae transcription factor Ace2 homologresulted in host tissue penetration by the pathogen, pathogenescape from the blood vessels, and overstimulation of thehost’s proinflammatory response in a murine model of invasivecandidiasis (23). In C. neoformans, mutation of the PKR1 geneencoding the cAMP-dependent protein kinase A regulatory

FIG. 5. Crg1 regulates pheromone responsive mating and virulence via distinct pathways. (A and B) The increased pheromone response andmating is dependent on the MAP kinase homolog Cpk1, cAMP signaling, and partially on the transcriptional factor homolog Ste12�. Conditionsfor confrontation and mating were as described in Fig. 2. (C and D) Melanin formation and the laccase activity of the crg1 strain under the normallyinhibitive condition are dependent on the transcription factor Ste12� but not the MAP kinase Cpk1 or Ste12a. Conditions for inducing melaninand measurement of the laccase activity were the same as described in Fig. 2. (D) The crg1 pkr1 strain was not tested.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

subunit Pkr1 resulted in cells with large capsules and an in-crease in virulence (Fig. 3B) (14). Cryptococcal cells with sim-ilarly large capsules can also be achieved by mutation of theSCH9 gene encoding a S. cerevisiae Sch9 protein kinase ho-molog. However, the sch9 mutant strain was attenuated invirulence despite the presence of a large capsule (P. Wang etal., unpublished data). In contrast, mutation of CRG1 has notresulted in a dramatically larger capsule similar to that pro-duced by the pkr1 or sch9 strain. An increase in capsule thick-ness was detected in the crg1 cells under the capsule inducingconditions compared to the wild-type strain; however, the in-crease in capsule thickness was not to the extent found in thepkr1 or sch9 strains. Thus, the greater virulence increase due toCRG1 mutation is likely via distinct regulatory circuitry.

The enhanced pheromone response due to the mutation ofCRG1 suggests that Crg1 interacts with one of the unreportedG protein � subunits and that the deletion of CRG1 resulted inprolonged activation of the Gpb1-protein-mediated matingpathway. Efforts are under way to identify additional Crg1target proteins. Our study supports a model in which Crg1functions upstream of the Gpb1/MAP kinase cascade for reg-ulating mating and conjugation tube formation, as well as up-stream of the transcription factor Ste12� to promote melaninformation under normally prohibitive conditions and increasevirulence. Unlike S. cerevisiae, Ste12� of C. neoformans is lesslikely to serve as the direct downstream component of themating pathway and is likely involved in several pathways,including mating and filamentation (10). The key element inCRG1 mutation-associated hypervirulence could be the diver-gent function of Ste12�/a. Activation of mating due to CRG1mutation may cause the upregulation of the Ste12� pathwaythat subsequently promotes melanin formation under stressedconditions, such as at a higher glucose level or higher temper-ature. Increased melanin may help to promote stress tolerance,benefiting fungal survival, or alter the host immune response,thereby increasing virulence.

In summary, our study for the first time links enhancedmating of MAT� cells to increased virulence and created anovel hypervirulent C. neoformans strain that will allow us tofurther elucidate the mechanisms of fungal virulence.

ACKNOWLEDGMENTS

We thank A. Alspaugh and J. Heitman for providing strains anddiscussion; N. Cutler, M. Duverney, and A. Prat for technical assis-tance; and D. Fox for reviewing the manuscript.

This research was supported in part by a fund from the Children’sHospital in New Orleans, La., and grants from the National Institutesof Health to P.W. (AI054958) and to J.E.C. (AI24912).

REFERENCES

1. 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.

2. Amsellem, Z., B. A. Cohen, and J. Gressel. 2002. Engineering hypervirulencein a mycoherbicidal fungus for efficient weed control. Nat. Biotechnol. 20:1035–1039.

3. Apanovitch, D. M., K. C. Slep, P. B. Sigler, and H. G. Dohlman. 1998. Sst2is a GTPase-activating protein for Gpa1: purification and characterization ofa cognate RGS-G� protein pair in yeast. Biochemistry 37:4815–4822.

4. Berman, D. M., and A. G. Gilman. 1998. Mammalian RGS proteins: bar-barians at the gate. J. Biol. Chem. 273:1269–1272.

5. Chan, R. K., and C. A. Otte. 1982. Isolation and genetic analysis of Saccha-romyces cerevisiae mutants supersensitive to G1 arrest by a factor and � factorpheromones. Mol. Cell. Biol. 2:11–20.

6. Chan, R. K., and C. A. Otte. 1982. Physiological characterization of Saccha-

romyces cerevisiae mutants supersensitive to G1 arrest by a factor and � factorpheromones. Mol. Cell. Biol. 2:21–29.

7. Cox, G. M., H. C. McDade, S. C. A. Chen, S. C. Tucker, M. Gottredsson,L. C. Wright, T. C. Sorrell, S. D. Leidich, A. Casadevall, M. A. Ghannoum,and J. R. Perfect. 2001. Extracellular phospholipase activity is a virulencefactor for Cryptococcus neoformans. Mol. Microbiol. 39:166–175.

8. 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.

9. Cunningham, M. L., R. G. Titus, S. J. Turco, and S. M. Beverley. 2001.Regulation of differentiation to the infective stage of the protozoan parasiteLeishmania major by tetrahydrobiopterin. Science 292:285–287.

10. Davidson, R. C., C. B. Nichols, G. M. Cox, J. Perfect, and J. Heitman. 2003.A MAP kinase cascade composed of cell type specific and nonspecific ele-ments controls mating and differentiation of the fungal pathogen Cryptococ-cus neoformans. Mol. Microbiol. 49:469–485.

11. De Vries, L., and M. G. Farquhar. 1999. RGS proteins: more than just GAPsfor heterotrimeric G proteins. Trends Cell Biol. 9:138–144.

12. De Vries, L., B. Zheng, T. Fischer, E. Elenko, and M. G. Farquhar. 2000. Theregulator of G protein signaling family. Annu. Rev. Pharmacol. Toxicol. 40:235–271.

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

14. 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.

15. Foreman-Wykert, A. K., and J. F. Miller. 2003. Hypervirulence and pathogenfitness. Trends Microbiol. 11:105–108.

16. Fowler, T. J., and M. F. Mitton. 2000. Scooter, a new active transposon inSchizophyllum commune, has disrupted two genes regulating signal transduc-tion. Genetics 156:1585–1594.

17. Fraser, J. A., R. L. Subaran, C. B. Nichols, and J. Heitman. 2003. Recapit-ulation of the sexual cycle of the primary fungal pathogen Cryptococcusneoformans var. gattii: implications for an outbreak on Vancouver Island,Canada. Eukaryot. Cell 2:1036–1045.

18. 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.

19. Hepler, J. R. 1999. Emerging roles for RGS proteins in cell signalling. TrendsPharmacol. Sci. 20:376–382.

20. Hepler, J. R., M. D. Berman, A. G. Gilman, and T. Kozasa. 1996. RGS4 andGAIP are GTPase-activating proteins for Gq� and block activation of phos-pholipase C� by �-thio-GTP-Gq�. Proc. Natl. Acad. Sci. USA 94:428–432.

21. Jacobson, E. S., and S. B. Tinnell. 1993. Antioxidant function of fungalmelanin. J. Bacteriol. 175:7102–7104.

22. Kallal, L., and R. Fishel. 2000. The GTP hydrolysis defect of the Saccharo-myces cerevisiae mutant G-protein Gpa1(G50V). Yeast 16:387–400.

23. Kamran, M., A.-M. Calcagno, H. Findon, E. Bignell, M. D. Jones, P. Warn,P. Hopkins, D. W. Denning, G. Butler, T. Rogers, F. A. Muhlschlegel, and K.Haynes. 2004. Inactivation of transcription factor gene ACE2 in the fungalpathogen Candida glabrata results in hypervirulence. Eukaryot. Cell 3:546–552.

24. Koelle, M. R. 1997. A new family of G-protein regulators: the RGS proteins.Curr. Opin. Cell Biol. 9:143–147.

25. Kozel, T. R., and E. C. Gotschlich. 1982. The capsule of Cryptococcus neo-formans passively inhibits phagocytosis of the yeast by macrophages. J. Im-munol. 129:1675–1680.

26. Kozel, T. R., G. S. Pfrommer, A. S. Guerlain, B. A. Highison, and G. J.Highison. 1988. Role of the capsule in phagocytosis of Cryptococcus neofor-mans. Rev. Infect. Dis. 10(Suppl. 2):S436–S439.

27. Kwon-Chung, K. J., and J. E. Bennett. 1992. Cryptococcosis, p. 397–446. InMedical mycology. Lea & Febiger, Malvern, Pa.

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

29. Lengeler, K. B., P. Wang, G. M. Cox, J. R. Perfect, and J. Heitman. 2000.Identification of the MATa mating type locus of Cryptococcus neoformansreveals a serotype A MATa strain thought to have been extinct. Proc. Natl.Acad. Sci. USA 97:14455–14460.

30. Manca, C., L. Tsenova, A. Bergtold, S. Feeman, M. Tovey, J. M. Musser,C. E. Barry, V. H. Freedman, and G. Kaplan. 2001. Virulence of a Myco-bacterium tuberculosis clinical isolate in mice is determined by failure toinduce Th1 type immunity and is associated with induction of IFN-�/�. Proc.Natl. Acad. Sci. USA 98:5752–5757.

31. 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.

32. Moore, T. D. E., and J. C. Edman. 1993. The �-mating type locus of Cryp-



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

tococcus neoformans contains a peptide pheromone gene. Mol. Cell. Biol.13:1962–1970.

33. Nielsen, K., G. M. Cox, P. Wang, D. L. Toffaletti, J. Perfect, and J. Heitman.2003. Sexual cycle of Cryptococcus neoformans var. grubii and virulence ofcongenic a and � isolates. Infect. Immun. 71:4831–4841.

34. Siekhaus, D. E., and D. G. Drubin. 2003. Spontaneous receptor-independentheterotrimeric G-protein signalling in an RGS mutant. Nat. Cell Biol. 5:231–235.

35. 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.

36. Viviani, M. A., M. C. Esposto, M. Cogliate, M. T. Montagna, and B. L.Wickes. 2001. Isolation of a Cryptococcus neoformans serotype A MATastrain from the Italian environment. Med. Mycol. 39:383–386.

37. Wang, P., M. E. Cardenas, G. M. Cox, J. Perfect, and J. Heitman. 2001. Twocyclophilin A homologs with shared and distinct functions important forgrowth and virulence of Cryptococcus neoformans. EMBO Rep. 2:511–518.

38. 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.

39. 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.

40. Wang, Y., P. Aisen, and A. Casadevall. 1995. Cryptococcus neoformans mel-anin and virulence: mechanism of action. Infect. Immun. 63:3131–3136.

41. Wang, Y., and A. Casadevall. 1994. Susceptibility of melanized and nonmela-nized Cryptococcus neoformans to nitrogen- and oxygen-derived oxidants.Infect. Immun. 62:3004–3007.

42. Wheeler, R. T., M. Kupiec, P. Magnelli, C. Abeijon, and G. R. Fink. 2003. ASaccharomyces cerevisiae mutant with increased virulence. Proc. Natl. Acad.Sci. USA 100:2766–2770.

43. 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.

44. 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.

45. Zaragoza, O., and A. Casadevall. 2004. Experimental modulation of capsulesize in Cryptococcus neoformans. Biol. Proced. Online 6:10–15.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

mutation of the regulator of g protein signaling crg1 ...health sciences center, and research...

Documents