Molecular and genetic characterization of a serotype A MATa Cryptococcus neoformans isolate

S. M. Keller1, M. A. Viviani2, M. C. Esposto2, M. Cogliati2 and B. L. Wickes1

1 Department of Microbiology, Mail Code 7758, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78229-3900, USA
2 Laboratorio di Micologia Medica, Instituto di Igiene e Medicina Preventiva, Universita degli Studi di Milano, IRCCS Ospedale Maggiore, Milano, Italy

Correspondence
B. L. Wickes
wickes{at}uthscsa.edu


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Cryptococcus neoformans comprises two varieties (neoformans and gattii) and four serotypes (A, B, C and D). Fertile isolates of both mating types have been identified in serotypes B, C and D; however, a fertile serotype A MATa strain has not been confirmed, although serotype A MAT{alpha} strains will mate with serotype D MATa strains. Preliminary analysis of a recent Italian environmental isolate (IUM 96-2828) suggested that this strain was haploid, serotype A and MATa. In this study, IUM 96-2828 has been characterized in detail. A mating reaction between IUM 96-2828 and H99 (serotype A MAT{alpha}) produced abundant spores with an equal distribution of MATa and MAT{alpha} progeny, all of which were serotype A. Karyotypic analysis of F1 spores revealed evidence of recombination, confirming that IUM 96-2828 was fertile. The MATa pheromone gene from IUM 96-2828 was sequenced and found to be most closely related to the serotype D MATa pheromone gene. Phylogenetic comparisons of other genes not linked to mating type also suggested IUM 96-2828 was most closely related to serotype A strains. Biochemical analysis showed that the carbon assimilation profiles of H99 and IUM 96-2828 were identical for 97 % (30/31) of the substrates while isozyme analysis showed 89 % (17/19) identity. Assays of major virulence factors found no difference between H99 and IUM 96-2828. Virulence studies using the mouse model demonstrated that IUM 96-2828 was virulent for mice, although it was less virulent than H99. These data strongly suggest that IUM 96-2828 is a true haploid serotype A MATa isolate that is fertile.

The GenBank accession number for the MFa1A pheromone sequence reported in this paper is AY129299.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Cryptococcus neoformans is a heterothallic basidiomycetous fungus with two mating types, MATa and MAT{alpha} (Kwon-Chung, 1975, 1976). It is primarily an opportunistic pathogen that causes meningoencephalitis in immunocompromised patients, but can also cause illness in healthy individuals (Casadevall & Perfect, 1998). During the 1980s and early 1990s the incidence of cryptococcal infection rose steadily due to factors such as AIDS, cancer therapy, organ transplant and other immunosuppressive conditions (Casadevall & Perfect, 1998; Mitchell & Perfect, 1995). In recent years the incidence of cryptococcosis has been decreasing due to better management of AIDS, which is the greatest risk factor for infection (Hamilton & Goodley, 1996). Unfortunately, the decrease in new infections does not end the seriousness of cryptococcosis for AIDS patients. For those AIDS patients who were infected, cryptococcosis is often treated with a life-long antifungal regimen (Pinner et al., 1995). Further compounding this problem is the inability of current AIDS therapies to eradicate HIV from the body (Cohen, 1997), which has led, predictably, to an increasing failure rate of these therapies (Cohen, 1998). The most recent data for HIV infection have cited an increase in drug-resistant HIV and an increase in new infections in certain areas in the USA (Matsushita, 2000; Piot et al., 2001). Therefore, cryptococcosis still remains a serious disease for this patient population.

The life-cycle of C. neoformans has been well defined. Mating occurs when MATa and MAT{alpha} cells physically contact one another and fuse to form a diploid, which then differentiates into a dikaryotic hypha with fused clamp connections. Basidia eventually form at the tips of these hyphae and provide the site where meiosis and sporulation occur. Haploid spores of each mating type form in four random chains on the basidial surface by successive mitotic divisions. Once released from the spore chains, individual basidiospores eventually germinate to form haploid budding yeast cells, completing the life-cycle (Kwon-Chung, 1975, 1976).

After the life-cycle was described, it was used to help classify C. neoformans into two varieties, which were further subdivided into serotypes (Kwon-Chung et al., 1982a). C. neoformans can be separated into four serotypes (A, B, C and D), with a fifth serotype (AD) occasionally observed. These serotypes have been grouped into two varieties, neoformans (A, D) and gattii (B, C) (Evans, 1949, 1950; Evans & Kessel, 1951; Vogel, 1966), although a new variety designated C. neoformans var. grubii has recently been proposed for serotype A isolates (Franzot et al., 1999). The four serotypes are not equally distributed throughout the world. The gattii variety is found in tropical and subtropical areas while the neoformans variety is found globally (Kwon-Chung, 1987). Serotype A is the most frequent serotype recovered from clinical isolates (Kwon-Chung & Bennett, 1992). This frequency is reduced, however, in tropical and subtropical areas where the gattii variety is found, as well as in parts of Europe where serotype D is frequently found (Bennett et al., 1977; Viviani et al., 2000). In spite of some of these geographical variations, serotype A isolates cause the overwhelming majority of cryptococcal infections throughout the world (Kwon-Chung & Bennett, 1992).

Similar to the bias in serotype prevalence, a number of studies have also documented a bias in mating-type representation in clinical specimens. These studies have found that MAT{alpha} isolates comprise more than 95 % of all clinical isolates (Bennett et al., 1977; Hironaga et al., 1983; Jong et al., 1982; Madrenys et al., 1993; Yan et al., 2002). Although this bias suggests that the MAT{alpha} mating type is more virulent than the MATa mating type, this assumption is complicated. On the one hand, Kwon-Chung et al. (1992) showed that MAT{alpha} cells are more virulent than MATa cells by using congenic strains, which only differ at the mating-type locus, in a mouse survival model. On the other hand, the MAT{alpha} bias in clinical isolates also extends into environmental samples (Kwon-Chung & Bennett, 1978), which allows for the possibility that exposure frequency may play a role in the clinical bias. Therefore, although most evidence points to MAT{alpha} cells being innately more virulent than MATa cells, additional studies need to be performed to confirm this.

Presently, most genetic studies of C. neoformans are conducted in serotype D strains because of the existence of both MATa and MAT{alpha} strains as well as a congenic pair of strains, which serves as a standard laboratory pair. The inability to identify a serotype A MATa strain suggested that these strains were either extinct or existed in an ecological niche that remained to be found. In 2000, Lengeler reported the identification of a serotype A MATa clinical isolate from a Tanzanian AIDS patient (Lengeler et al., 2000). This report provided the first evidence that a serotype A MATa locus existed; however, the strain was sterile. Another study designed to establish the origin of serotype AD strains, which are diploid or aneuploid hybrids derived from a fusion of serotype A and D parents, showed that several of these unique strains harboured a serotype A MATa locus (Lengeler et al., 2001). The high frequency of the serotype A MATa mating-type locus in diploid strains suggested that these strains do, in fact, exist in nature and can be readily recovered (Cogliati et al., 2001). It is, therefore, curious as to why haploid serotype A MATa isolates still remain elusive.

Here we report the detailed characterization of a fertile serotype A MATa C. neoformans isolate. This isolate, designated IUM 96-2828, was initially recovered as an environmental isolate from Italy (Viviani et al., 2001). We show in this study that IUM 96-2828 mates in the laboratory at high frequency and produces recombinant progeny, which demonstrates that it is MATa in mating type. Serotypic analysis by agglutination and phylogenetic analysis of genes not linked to mating type revealed that the strain is most closely related to serotype A strains, confirming that it is a true serotype A isolate. Analysis of a variety of phenotypic traits revealed that this strain had no unusual characteristics with regard to phenotype, as all were wild-type. Taken together, these results demonstrate that IUM 96-2828 is a true, fertile serotype A MATa isolate.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Media, strains and plasmids.
YPD consisted of 1 % yeast extract, 2 % peptone and 2 % glucose, and was solidified with 2 % agar as needed. Minimal medium (Min) consisted of 2 % agar, 2 % glucose (Min-dex) and 0·67 % yeast nitrogen base (YNB) without amino acids (Difco). V8 agar was modified from the original description (Kwon-Chung et al., 1982a); it contained 10 % filtered V8 juice, 0·5 g KH2PO4 l-1 and 4 % agar, and was adjusted to pH 6. Egg yolk agar consisted of 2 % agar, 2 % glucose, 1 M NaCl, 5 mM CaCl2, 0·67 % YNB without amino acids and 10 % egg yolk (Price et al., 1982). Asparagine agar contained 1 g L-asparagine l-1, 1 g glucose l-1, 1 mM dopamine, 2 % agar and 0·67 % YNB without amino acids and without ammonium sulfate (Kwon-Chung et al., 1982b; Rhodes et al., 1982). Modified low iron medium (LIM) consisted of 2 % agar, 0·5 % glucose, 0·04 % KH2PO4, 0·5 % L-asparagine and 10 ml trace elements l-1 (Wills et al., 2001). Strains used in this study are listed in Table 1.


View this table:
[in this window]
[in a new window]
 
Table 1. Strains used in this study

 
DNA preparation and analysis.
Genomic DNA was prepared using the bead-beating method (Clarke et al., 2001). DNA (3–5 µg) for Southern blots was run on 0·6 % agarose gels at 65 V for 4 h and then transferred onto Hybond-N+ nylon membranes (Amersham Pharmacia). All hybridizations were done in Rapid-Hyb buffer (Amersham Pharmacia) according to manufacturer's instructions. Probes were randomly labelled using the High Prime kit (Roche) according to the manufacturer's instructions. After washing, membranes were exposed to BioMax MS X-Ray film (Kodak).

Serotyping.
Serotyping was performed using the Crypto Check slide agglutination kit (Iatron Labs) according to the manufacturer's instructions.

TA cloning, sequencing and oligonucleotide synthesis.
PCR products were TA cloned into pCR2.1 (Invitrogen) according to the manufacturer's instructions. Templates were prepared for sequencing using the Wizard miniprep kit (Promega) and sequenced at the Advanced Nucleic Acids Core Facility at the University of Texas Health Science Center at San Antonio. Data were analysed and aligned using MACVECTOR and ASSEMBLYLIGN (Oxford Molecular Group). Oligonucleotides were obtained from the Advanced Nucleic Acids Core Facility at the University of Texas Health Science Center at San Antonio.

PCR.
All reactions were performed in a GeneMate thermocycler (Techne) using a 50 µl reaction volume with 2·5 U Taq DNA polymerase (Invitrogen BRL) and 25–50 ng DNA as template. Primers are listed in Table 2. The serotype D MATa pheromone (MFa1) was amplified as an ~800 bp fragment from JEC-20 template DNA using primers All.MFa.5' and All.MFa.3'. Amplification conditions were 94 °C for 2 min, an annealing temperature of 62 °C for 30 s and an extension step of 72 °C for 60 s, which was then repeated 29 times except that the initial 94 °C denaturation step was reduced to 30 s. The final cycle included a 3 min extension step at 72 °C. The serotype A MATa pheromone (MFa1A) was amplified as an ~800 bp fragment from IUM 96-2828 DNA using the same primers and conditions as for the serotype D MFa1 gene, and then TA cloned and sequenced. The serotype A MAT{alpha} pheromone gene coding sequence was amplified from H99 using primers H99.MF.F-2 and H99.MF.R. Amplification conditions were 94 °C for 2 min, an annealing temperature of 56 °C for 15 s and an extension step of 72 °C for 30 s, which was then repeated 29 times except that the initial 94 °C denaturation step was reduced to 15 s. The final extension was extended for an additional 3 min at 72 °C. The ~100 bp H99 A MF{alpha}1 fragment was then TA cloned and sequenced.


View this table:
[in this window]
[in a new window]
 
Table 2. Primers used in this study

 
As an additional control for screening purposes and linkage analysis, primers were designed in a serotype-specific and mating-type-specific manner to amplify the STE20 gene (Table 2) using a strategy similar to that described by Lengeler et al. (2001). STE20 sequences from H99 (STE20{alpha}), JEC-21 (STE20{alpha}) and JEC-20 (STE20a) were aligned in order to design serotype and mating-type-specific primers. The primers for the IUM 96-2828 STE20a sequence were the same primers used in the study of STE20 genes from serotype AD hybrids (Lengeler et al., 2001). Amplification of all STE20 genes utilized a standard protocol with an initial step of 94 °C for 2 min, a primer-dependent annealing temperature (D STE20a, 62 °C; A STE20a, 58 °C; D STE20{alpha}, 58 °C; A STE20{alpha}, 55 °C) for 20 s, 72 °C for 1 min, followed by 24 additional cycles where the 94 °C step was reduced to 20 s, and a final extension of 3 min. D STE20a was amplified with the same protocol except 29 cycles were used instead of 24. Primers used and product sizes obtained included serotype D STE20a (STE20a.D.F/STE20a.D.R, ~250 bp), serotype D STE20{alpha} (STE20alpha.D.F/STE20alpha.D.R, ~200 bp), serotype A STE20a (STE20a.A1/STE20a.A2, ~850 bp) and serotype A STE20{alpha} (H99STE20.F/H99STE20.R, ~200 bp). All products were TA cloned and sequenced to confirm the accuracy of the primers.

Accession number.
The MFa1A pheromone sequence has been deposited in GenBank under accession number AY129299.

Electrophoretic karyotyping.
Seed cultures of 50 ml YPD were inoculated with approximately 1x106 cells and grown at 30 °C for 16 h with shaking at 300 r.p.m. before 100 ml pre-warmed YPD broth was added. Shaking and incubation continued for 4–6 h, and then cells were harvested. Plugs were prepared as described previously (Wickes et al., 1994), except that sphaeroplasting took place in solution prior to agarose addition using 950 mg lysing enzymes from Trichoderma harzianum (Sigma) per 1 ml buffer. Samples were run in a 0·7 % PFGE-grade agarose (Bio-Rad) gel with 0·5xTBE running buffer using the following protocol: 110 V, switch times of 50–130 s for 23 h followed by 170–400 s for 47 h, at a gel temperature of 10 °C. Southern blots were prepared as follows. After staining in ethidium bromide for 30 min, the gel was nicked in a Stratalinker (Stratagene) set at 60 mJ. Following a 15 min wash in 0·4 M NaOH/1·5 M NaCl, a standard capillary transfer was set up using a Hybond-N+ nylon membrane. Transfers continued until 2 l buffer had passed through the gel, or approximately 48 h. The membranes were neutralized after transfer by washing in 0·5 M Tris/HCl (pH 7) for 5 min and then rinsed briefly in 2xSSC. Hybridizations of karyotype blots followed Hybond-N+ membrane instructions, except that Rapid-Hyb buffer (Amersham Pharmacia) was used.

Phylogeny study.
Several genes were chosen for phylogenetic analysis based on the availability of sequences from all four serotypes for comparison. Accession numbers are listed in Table 3. C. neoformans CAP10, CNLAC1 and URA5 genes were amplified from IUM 96-2828 using primers to conserved regions identified from alignments of sequences from the other serotypes (Table 2). A 480 bp CAP10 fragment was amplified with primers CAP10.F and CAP10.R using an initial step of 94 °C for 2 min, 63 °C for 12 s, 72 °C for 30 s, followed by 29 additional cycles where the 94 °C cycle was reduced to 12 s and a final extension of 1 min at 72 °C. A 650 bp CNLAC1 fragment was amplified with primers LAC.F and LAC.R using an initial step of 94 °C for 2 min, 50 °C for 30 s, 72 °C for 1 min 30 s, followed by 29 additional cycles where the 94 °C step was reduced to 30 s and then a final extension of 3 min at 72 °C. A 780 bp URA5 fragment was amplified using primers URA5.F and URA5.R with an initial step of 94 °C for 2 min, 58 °C for 20 s, 72 °C for 45 s, followed by 29 additional cycles where the 94 °C step was reduced to 20 s and a final extension of 2 min at 72 °C. All PCR products were TA cloned and then sequenced.


View this table:
[in this window]
[in a new window]
 
Table 3. Accession numbers

 
Biochemistry.
Carbon assimilation was tested using the API ID32 kit (BioMérieux); isozyme analysis was performed using the API ZYM kit (BioMérieux). Assays were performed independently three separate times.

Virulence studies.
Virulence assays utilized 10 female BALB/c mice per isolate, which were injected with either H99 or IUM 96-2828 via the lateral tail vein. Inocula were prepared as previously described (Kwon-Chung et al., 1992) and then used to inject each mouse with 0·2 ml of a suspension containing 5x106 organisms ml-1. Mice were observed daily and scored for mortality. Data were analysed by Kaplan–Meier survival estimates using JMP software (SAS Institute). Egg yolk agar was used for testing phospholipase activity (Price et al., 1982; Vidotto et al., 1998). Strains were inoculated onto egg yolk agar as a suspension (1x108 cells ml-1) after growth on YPD for 24 h. Cultures were observed for the formation of a precipitate around the colony for up to 7 days incubation at 30 °C. Melanin production was tested on asparagine agar (Kwon-Chung et al., 1982b; Rhodes et al., 1982). A 24 h seed culture was grown on YPD at 30 °C. A suspension of approximately 1x108 cells ml-1 was prepared and 5–10 µl of this were dropped onto the agar. Positive reactions consisted of a dark-brown colony colour within 24–48 h incubation at 30 °C. Capsule production was assayed as described using a modified LIM (Wills et al., 2001). Plates were inoculated from a 24 h YPD culture and grown at 25 °C for 5–7 days. Cells were then mounted with India ink to visualize the capsule microscopically.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
IUM 96-2828 is serotype A
The serotype of IUM 96-2828 was determined by agglutination reactions and PCR amplification of the serotype-specific alleles. The Iatron slide agglutination reaction was performed four separate times on both young (24 and 48 h) and old (21 day) cultures. In all slide agglutination reactions, IUM 96-2828 was agglutinated by serum 1 (polyvalent) and serum 7 (serotype A-specific) in young cultures, as demonstrated in our previous study (Viviani et al., 2001), as well as in older cultures. To confirm these results, we performed crosses using H99 as the other parent to ensure that this strain would mate efficiently but, most importantly, would produce progeny that could be scored for serotype. Macroscopic and microscopic inspection of mating reactions revealed that IUM 96-2828 mated vigorously with H99 and produced abundant chains of basidiospores (Fig. 1). Forty-eight progeny from this cross as well as MAT{alpha} and MATa control progeny from crosses between JEC-20 and JEC-21, H99 and JEC-20, and IUM 96-2828 and JEC-21 were recovered and scored for mating type and serotype (Table 4). All progeny from the cross between IUM 96-2828 and H99 were serotype A and all progeny from the control cross between JEC-20 and JEC-21 were serotype D. Interserotype control crosses between serotype A and D strains (H99xJEC-20 and IUM 96-2828xJEC-21) yielded strains that were MATa, MAT{alpha} and MATa/{alpha} based on STE20 analysis. Our interserotype crosses only yielded serotype AD progeny, although due to the intentionally small sample size of these classes, these results were not unexpected since these types of crosses usually yield hybrid strains (AD) that are diploid or aneuploid (Lengeler et al., 2001). The interserotype crosses (IUM 96-2828xJEC-21 and H99xJEC-20) were included to demonstrate that serotype D antigenicity could be detected in serotype AD hybrids with different mating-type locus combinations (D MATa/A MAT{alpha} or D MAT{alpha}/A MATa) in case the diploid mating locus state influenced serotype expression. As predicted, all of the a/{alpha} progeny from the cross between H99 and JEC-20 contained both the serotype A STE20{alpha} and serotype D STE20a alleles, whereas all of the a/{alpha} progeny from the cross between IUM 96-2828 and JEC-21 contained the serotype A STE20a and serotype D STE20{alpha} alleles. In each mating-type background of these hybrids, serotype D antigenicity was detected. Serotype D has been shown to be dominant in interserotype crosses (Lengeler et al., 2001). If serotype D antigenicity were present in IUM 96-2828, it should have been detected in the repeated serotyping of the parental IUM 96-2828 strain, or uncovered during the analysis of the progeny derived from the IUM 96-2828 crosses. Since all 48 F1 progeny from the IUM 96-2828xH99 cross were serotype A, and IUM 96-2828 always serotyped as A regardless of culture conditions, we conclude that there is no evidence of serotype D antigenicity in IUM 96-2828.



View larger version (68K):
[in this window]
[in a new window]
 
Fig. 1. Mating reactions of IUM 96-2828. (a) Qualitative mating assays on V8 agar. 1, JEC-20 (D MATa)xJEC-21 (D MAT{alpha}); 2, IUM 96-2828 (A MATa)xH99 (A MAT{alpha}); 3, IUM 96-2828 (A MATa)xJEC-20 (D MATa); 4, IUM 96-2828 (A MATa)xJEC-21 (D MAT{alpha}). (b) Microscopic view of basidia and basidiospores from the cross of IUM 96-2828xH99. Numerous chains of basidiospores (solid arrows) are visible at the end of basidia.

 

View this table:
[in this window]
[in a new window]
 
Table 4. Progeny analysis from IUM 96-2828xH99

Serotypes were determined by agglutination using the Iatron kit. Mating types were determined by backcrossing to MAT{alpha} and MATa tester strains. The outcomes of these reactions were confirmed by PCR using the STE20 primer set, which consists of primer combinations specific for both serotype and mating type.

 
IUM 96-2828 is MATa in mating type
IUM 96-2828 yields a PCR product when amplified with primers specific for the serotype D MATa pheromone gene, but does not yield a product with primers specific for the serotype D MAT{alpha} gene. However, because C. neoformans pheromones are highly conserved regardless of mating type, additional analysis of the pheromone sequence was necessary. To obtain the sequence for the IUM 96-2828 pheromone, primers designed to amplify the serotype D MATa pheromone gene were used for the amplification and cloning of an ~800 bp fragment from IUM 96-2828, which was predicted to contain the pheromone coding sequence. Sequencing of the fragment revealed a 126 bp ORF, which was preceded by a consensus TATA box. Comparison to the serotype D MATa pheromone sequence revealed consensus 5' and 3' splice sites for a 69 bp intron immediately following the stop codon in a position similar to the positions of the introns in the serotype D MATa and MAT{alpha} pheromone genes (McClelland et al., 2002). A consensus poly A signal was also found in the 3' flanking region. The IUM 96-2828 pheromone gene sequence, designated MFa1A, is predicted to encode a protein of 41 amino acids in length with a molecular mass of 4·27 kDa. Analysis of the predicted protein revealed that MFa1Ap contained the three conserved regions characteristic of other C. neoformans pheromones (McClelland et al., 2002), including the carboxy terminal CVIA motif. This motif fits the CAAX consensus sequence observed in numerous other fungal pheromone genes (Caldwell et al., 1995). Comparison of the MFa1Ap amino acid sequence to both serotype D MATa and serotype A and D MAT{alpha} pheromone sequences revealed that MFa1Ap was most closely related to the serotype D MATa pheromone, displaying 88·1 % amino acid identity (Fig. 2). Similar identity was observed among the H99 and JEC-21 pheromone sequences, which are 86·8 % identical. Identities between pheromones of different mating types were found to be conserved, but reduced compared to same mating-type pheromones. The identities between H99 and JEC-20 were 52·4 %, 51·2 % between IUM 96-2828 and H99, and 52·4 % between IUM 96-2828 and JEC-21.



View larger version (34K):
[in this window]
[in a new window]
 
Fig. 2. Pheromone alignment. Pheromone sequences from JEC-20, JEC-21, IUM 96-2828 and H99 were aligned in various combinations. Regions of identity are shaded. All pheromones contained the three conserved regions characteristic of C. neoformans pheromones (underlined). These sites are predicted to be processing sites, which are recognized during pheromone maturation (McClelland et al., 2002).

 
A panel of strains representative of all four serotypes and both mating types were probed with the two serotype A pheromone genes (Fig. 3). Southern blots of this panel were probed with the MFa1A pheromone gene from IUM 96-2828, stripped and then re-probed with the MF{alpha}1A pheromone gene from H99. MFa1A only hybridized to JEC-20 and IUM 96-2828 DNA, both of which belong to variety neoformans. Conversely, MF{alpha}1A hybridized to MAT{alpha} DNA from all four serotypes. The hybridization pattern of the H99 MAT{alpha} pheromone gene to MAT{alpha} strains of all four serotypes was consistent with the pattern for the serotype D MAT{alpha} pheromone gene, which also hybridizes to MAT{alpha} DNA from all four serotypes (McClelland et al., 2002). The MFa1A hybridization pattern, however, differed from the MF{alpha}1A hybridization pattern. In this case, MFa1A only hybridized to MATa strains that were serotype A or D. This pattern matches the hybridization pattern of the serotype D MATa pheromone gene (McClelland et al., 2002). Therefore, it appears that C. neoformans var. neoformans MATa pheromones are conserved only within the variety whereas the MAT{alpha} pheromones are common to the species.



View larger version (29K):
[in this window]
[in a new window]
 
Fig. 3. Hybridization of MF{alpha}1A (top panel) or MFa1A (bottom panel) to genomic DNA from strains representing both mating types of all four serotypes. Lanes: 1, JEC-20; 2, JEC-21; 3, IUM 96-2828; 4, H99; 5, WSA-11; 6, WSA-10; 7, WSA-16; 8, WSA-15. Mating types of strains are indicated below each lane. Serotypes are shown above each pair of strains.

 
In addition to the pheromone gene, we also used the STE20 gene to determine cell type. Serotype and mating-type-specific PCRs with STE20 primers were used to trace linkage between mating type and the STE20 gene. Analysis of the progeny of the cross between IUM 96-2828 and H99 revealed that STE20a always co-segregated with the MATa mating type while STE20{alpha} always co-segregated with the MAT{alpha} mating type (Table 4). In crosses that yielded diploid progeny, the diploid strains carried two copies of STE20 in a pattern dependent on the input strains. In the cross between H99 and JEC-20, all MATa progeny had a serotype D-derived STE20a gene while all MAT{alpha} progeny had a serotype A-derived STE20{alpha} gene. Likewise, in the cross between IUM 96-2828 and JEC-21, all MATa progeny had a serotype A-derived STE20a gene and all MAT{alpha} progeny had a serotype D-derived STE20{alpha} gene. In each cross, regardless of whether the progeny were haploid or diploid, the STE20 allele amplified in patterns that were predicted from the parental serotypes and mating types. These results corroborate the serotyping results and provide additional molecular evidence that IUM 96-2828 is a true serotype A MATa strain.

Progeny of H99xIUM 96-2828 are recombinant
PFGE was used to compare the karyotypes of IUM 96-2828, H99, one F1 MATa and one F1 MAT{alpha} progeny (Fig. 4). The karyotypes of these four strains were highly conserved and were also conserved when compared to karyotype patterns previously reported for serotype A strains (Wickes et al., 1994). The chromosomes in these karyotypes ranged from 770 kb to 3·8 Mb in size, which agrees with data previously reported for serotype A karyotypes (Wickes et al., 1994). Both F1 progeny displayed non-parental karyotypes with at least one band being non-parental. The karyotype of the F1 MATa strain contained one band that was derived from the H99 (MAT{alpha}) parent, while the karyotype of the F1 MAT{alpha} strain contained one band derived from the IUM 96-2828 (MATa) parent, but lacked another band present in the H99 karyotype. The F1 progeny were themselves fertile when backcrossed to either compatible parent. In Southern blots of whole karyotypes, IUM 96-2828 and the F1 MATa strain from the cross between IUM 96-2828 and H99 only hybridized to the MFa1A probe from IUM 96-2828 (Fig. 4). Conversely, H99 and the F1 MAT{alpha} strain from the cross between IUM 96-2828 and H99 only hybridized to the MF{alpha}1A probe from H99. The probes hybridized to the band corresponding to the MAT chromosome (Clarke et al., 2001), which is approximately 1·8 Mb in size. The size and location of the MAT chromosome in the karyotypes of these strains is conserved in both serotypes.



View larger version (75K):
[in this window]
[in a new window]
 
Fig. 4. IUM 96-2828xH99 progeny are recombinant. Karyotype analysis was used to demonstrate recombinant progeny based on identification of non-parental patterns (arrows indicate recombinant chromosomes). Lanes: S.c., Saccharomyces cerevisiae size standard; 1, IUM 96-2828; 2, H99; 3, F1 MATa; 4, F1 MAT{alpha}. PFGE blots were probed with MFa1A (from IUM 96-2828) or MF{alpha}1A (from H99) to identify the MAT chromosome.

 
IUM 96-2828 is serotype A by phylogenetic analysis of genes not linked to mating type
The pheromone and STE20 genes from IUM 96-2828 are each linked to mating type. Additional sequences not linked to mating type were compared from all four serotypes to determine where the IUM 96-2828 sequences would cluster. Sequences from three other IUM 96-2828 genes (CAP10, CNLAC1 and URA5) were obtained in order to build phylogenetic trees for comparing this strain to representative strains of all four serotypes (Fig. 5). Sequences from serotypes A and D and serotypes B and C consistently clustered together, which was expected based on the classification of strains into the varieties neoformans (serotypes A and D) and gattii (serotypes B and C). All three IUM 96-2828 sequences clustered with the corresponding sequence from H99 (serotype A). Therefore, in addition to mating-type-specific sequences, sequences of IUM 96-2828 not linked to mating type also placed this strain closest to H99, a serotype A strain.



View larger version (11K):
[in this window]
[in a new window]
 
Fig. 5. Phylogenetic analysis of genes not linked to mating type. Genes were chosen based on the availability of sequences for all four serotypes. Sequences from IUM 96-2828 always clustered with sequences from H99 (serotype A MAT{alpha}).

 
Biochemical analyses
Carbon assimilation and isozyme profiles of IUM 96-2828 and H99 were compared (Table 5). Of the 31 carbon sources tested with the API ID32 kit, only aesculin assimilation differed, with IUM 96-2828 unable to use this compound as a carbon source. In spite of this difference, both strains were identified as C. neoformans by the kit. The isozyme activities of IUM 96-2828 and H99 were identical for 17 of the 19 enzymes assayed. The two differences included ß-glucosidase (H99 positive, IUM 96-2828 negative) and N-acetyl-ß-glucosaminidase (H99 negative, IUM 96-2828 positive). Aesculin assimilation requires ß-glucosidase activity (Miskin & Edberg, 1978), which was consistent with the aesculin assimilation and ß-glucosidase activity results for IUM 96-2828. Aesculin assimilation and isozyme patterns are variable among strains of C. neoformans (Edberg et al., 1980; Garcia-Martos et al., 2001; Schonheyder & Stenderup, 1982); therefore, IUM 96-2828 does not display any unusual phenotypic traits with regard to its carbon assimilation or isozyme patterns.


View this table:
[in this window]
[in a new window]
 
Table 5. Biochemical analyses of strains H99 and IUM 96-2828

 
Virulence for mice
Previous virulence studies have established that in a congenic pair of serotype D isolates, the MAT{alpha} mating type is more virulent than the MATa mating type in a mouse tail vein model (Kwon-Chung et al., 1992). Serotype A MAT{alpha} isolates are known to be virulent based on their prevalence in clinical isolates and have been shown to be virulent in multiple animal models and multiple inoculation sites (Cox et al., 2001; Cruz et al., 1999; Yue et al., 1999). We were interested in determining the level of virulence of IUM 96-2828 using the mouse tail vein injection model since this model is the most common in vivo model of C. neoformans virulence. IUM 96-2828 was shown to be virulent, but significantly less virulent than H99 (P<0·0001) (Fig. 6). In this experiment IUM 96-2828 took almost nine times longer than H99 to kill all of the mice. H99 displayed 100 % mortality by day 9 while IUM 96-2828 took until day 79 to kill all of the mice. Based on the low level of virulence demonstrated by IUM 96-2828 when compared to H99, several virulence factors were tested to ensure that IUM 96-2828 was not defective or reduced for one of the major virulence factors. IUM 96-2828 and H99 were both able to grow at 37 °C (Fig. 7a), with each displaying a doubling time of approximately 120 min. Both strains produced a dark pigment on asparagine agar, suggesting that their melanin production was similar (Fig. 7b). There were no clear differences in the capsule sizes (Fig. 7c) or phospholipase activities (Fig. 7d) of IUM 96-2828 and H99. Therefore, the difference in virulence between the two strains is not due to any of the major virulence factors.



View larger version (17K):
[in this window]
[in a new window]
 
Fig. 6. IUM 96-2828 versus H99 mouse survival. Ten BALB/c mice were inoculated for each strain and followed for 80 days. {lozenge}, IUM 96-2828; {blacklozenge}, H99.

 


View larger version (44K):
[in this window]
[in a new window]
 
Fig. 7. Comparison of virulence factors from IUM 96-2828 and H99. (a) Growth of serial dilutions (1x105 to 1x101 cells) at 37 °C on YPD. (b) Melanin production on asparagine agar. Strain MO92 is a melanin-negative control. (c) Capsule mounted with India ink after induction on LIM agar. Strain B4131 is a capsule-negative control. (d) Phospholipase B activity on egg yolk agar. Plb- is the negative control.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In 1999, C. neoformans was selected for a genome-sequencing project because it is a significant human pathogen for which current therapies are unsatisfactory (Heitman et al., 1999). It is also a good model for studying many aspects of fungal pathogenesis because it has well-defined genetic and molecular systems, and a variety of animal models that mimic the course of disease in humans. A primary consideration in the decision of which strain to use for the genome-sequencing project was mating type. JEC-21, a MAT{alpha} serotype D strain, was chosen for the genome project because it is the MAT{alpha} isolate of a congenic pair of strains that is useful for genetic analysis. JEC-21 is also a descendant of the strain (B-3501) that was used for most of the early genetic and molecular manipulations. While serotype A MAT{alpha} strains make up the majority of all clinical isolates, the lack of a serotype A MATa strain precluded traditional genetic crosses, which made the serotype D strain the more attractive choice for the sequencing project. Despite the usefulness of the congenic serotype D strains, however, the search for a serotype A MATa isolate has been ongoing. To date, more than 700 clinical and environmental isolates have been surveyed (K. J. Kwon-Chung, personal communication), with only two serotype A MATa strains being identified. The reasons behind the mating-type bias are still unclear. Although serotype D MATa strains have been found, the bias is quite similar to the serotype A bias. A possible reason for this bias is that the primary niche has not yet been discovered. The fact that C. neoformans var. neoformans appears to be ubiquitous may make a reservoir that yields large numbers of both mating types difficult to find, if it even exists. Alternatively, MATa strains may be heading towards extinction due to lack of a competitive advantage. However, this possibility must be viewed cautiously since clinical frequencies of certain strains actually represent select portions of environmental populations which have been filtered by host pressure.

The first confirmed serotype A MATa isolate was identified from a Tanzanian clinical isolate. Although the strain was shown to be virulent in mice, it did not produce mating pheromone, lacked one or more components of the signal transduction pathway required for mating and consequently, was sterile (Lengeler et al., 2000). IUM 96-2828, the second putative serotype A MATa strain to be identified, was initially recovered during a study of European isolates and identified by PCR (Viviani et al., 2001). The results of the present study provide strong evidence that IUM 96-2828 is a true serotype A MATa isolate, which is fertile. The purity of the progeny from the cross between IUM 96-2828 and H99 with regard to parental serotype suggested that there is no evidence of serotype D antigenicity in the IUM 96-2828 genomic background. Importantly, the clustering of IUM 96-2828 with other serotype A strains in the phylogenetic analysis of genes not linked to mating type also argues against any serotype D ancestry in this strain, especially since this analysis is molecular-based and independent of the agglutination assay. Therefore, by multiple techniques IUM 96-2828 always behaved unambiguously as a serotype A strain in our hands.

We also used multiple approaches to confirm that IUM 96-2828 was MATa in mating type. Southern blots of both mating types of all four serotypes revealed that the IUM 96-2828 MFa1A pheromone probe hybridized to genomic DNA from IUM 96-2828 and JEC-20, but did not hybridize to DNA from representative strains of the gattii variety (serotypes B and C). This pattern matches the hybridization pattern of the serotype D MATa pheromone (McClelland et al., 2002). In contrast, the MF{alpha}1A pheromone gene from H99 hybridized to all four serotypes. This hybridization pattern matched the serotype D MAT{alpha} pattern (McClelland et al., 2002). Xu et al. (2000) suggested that C. neoformans var. neoformans diverged from C. neoformans var. gattii approximately 37 million years ago and then separated into serotypes A and D approximately 18·5 million years ago. Although little is known about the evolution of mating types in C. neoformans, the MATa pheromones may have evolved after the divergence into the two varieties, and almost certainly after the MAT{alpha} pheromone. The hybridization pattern of MFa1A and the sequence conservation when compared to the serotype D MATa pheromone would be consistent with this hypothesis. Based on the sequence of the IUM 96-2828 pheromone, this strain would be predicted to mate as a MATa strain. In a standard mating reaction, IUM 96-2828 mated with H99 and produced abundant basidiospores. It also mated with a serotype D MAT{alpha} strain (JEC-21) but did not mate with a serotype D MATa strain (JEC-20). In a standard mating reaction with strains in which one or both backgrounds are unknown, it is possible to produce basidiospores that are not meiotic products of the two input parents by two ways. The first way is that the spores can be products of monokaryotic fruiting and therefore, are non-recombinant because they arise from a single strain in the absence of mating (Wickes et al., 1996). The second way is that the spores can be produced from a cross to a diploid or aneuploid self-fertile strain, which contains both a MATa and MAT{alpha} mating-type locus. In this case, although the spores could technically be recombinant, they would not be the product of a mating with the intended parent. These possibilities were excluded by the 1 : 1 mating-type segregation pattern of the progeny from the cross to H99, as well as the hybridization results with the pheromone probes and finally, the STE20 PCRs. The karyotype results in which PFGE was used to compare karyotypes from IUM 96-2828, H99 and the F1 progeny were also consistent with IUM 96-2828 being a fertile serotype A MATa. Both F1 progeny displayed non-parental karyotypes, providing evidence that the progeny were meiotic products. These results demonstrate the utility of IUM 96-2828 as a compatible tester strain to H99 and confirm that IUM 96-2828 is MATa in mating type.

Since IUM 96-2828 is serotype A and MATa, and has been shown to mate efficiently in a standard mating reaction, we have begun the construction of congenic strains in both the IUM 96-2828 and H99 backgrounds. Both strains mate efficiently and we are currently at generation F8. During mating analyses, we have observed that matings between IUM 96-2828 and H99 do not produce the intense hyphal fringe observed in crosses between the serotype D congenic strains JEC-20 and JEC-21 (see Fig. 1). However, the number of basidia produced in IUM 96-2828xH99 crosses is much higher and the spore chains produced appear to be substantially longer that those formed during JEC-20xJEC-21 crosses. Although mating efficiency can vary greatly depending on strain, our results suggest that IUM 96-2828 and H99 will be useful for most genetic analyses. To enhance their use for basic genetic analyses, we have also begun to insert various genetic markers into the strains and have succeeded in recovering an ura5 auxotroph, which can be complemented by transformation with the native gene. In the serotype D congenic pair, mating is markedly enhanced when two auxotrophs are crossed. Therefore, having marked strains will provide genetic flexibility for the study of serotype A strains and will also provide the appropriate hosts for transformation.

In addition to the utility of a congenic pair for future genetic studies, the two congenic pairs will provide an opportunity to test the role of mating type in the virulence of serotype A strains. In 1992, Kwon-Chung reported an association between mating type and virulence during a study of the serotype D congenic strains JEC-20 and JEC-21 (Kwon-Chung et al., 1992). The study concluded that the MAT{alpha} strain was significantly more virulent than the MATa strain. We were interested in determining the virulence of IUM 96-2828 in a mouse model in order to have background data for a future congenic virulence study. Our results demonstrated that IUM 96-2828 was virulent; however, it was significantly less virulent than H99. Interestingly, the Tanzanian isolate reported by Lengeler et al. (2000) was also significantly less virulent than H99. A number of factors are required for virulence in C. neoformans, which include, but are not restricted to, the production of capsule, melanin and phospholipase and growth at 37 °C (Buchanan & Murphy, 1998; Hamilton & Goodley, 1996; Kozel, 1995). IUM 96-2828 showed no difference from H99 in any of these phenotypic traits. Therefore, an explanation for the reduced virulence of IUM 96-2828 does not include a mutation in one of the above virulence factors. While it is tempting to conclude that reduced virulence is linked to the MATa mating-type locus of IUM 96-2828, in reality there could be many other reasons independent of mating type that explain the virulence difference. We identified slight differences in carbon assimilation and isozyme patterns; however, since C. neoformans strains are variable in these phenotypic traits, it is unclear what effect, if any, these differences would have on virulence. To address the association of mating type with virulence in serotype A strains, we plan to repeat the serotype D congenic strain study with the serotype A strains. The congenic strain strategy circumvents the problem of assembling a large, diverse collection of MATa and MAT{alpha} isolates because it allows the testing of each mating type in a normalized genetic background. By preparing two pairs of strains, one set in each parental background, the role of mating type in virulence can be more rigorously tested. Since H99 is a clinical isolate and IUM 96-2828 is an environmental isolate, if the MAT{alpha} strain in both backgrounds is significantly more virulent, this outcome in combination with the previous serotype D congenic virulence study should provide conclusive evidence for the role of mating type in the virulence of C. neoformans var. neoformans.


   ACKNOWLEDGEMENTS
 
We thank John R. Perfect (Department of Medicine, Duke University Medical Center, Durham, NC 27710, USA), Gary M. Cox (Departments of Medicine and Microbiology, Duke University Medical Center, Durham, NC 27710, USA) and June Kwon-Chung (Molecular Microbiology Section, Laboratory of Clinical Investigation, National Institute of Allergy and Infectious Diseases, National Institute of Health, Bethesda, MD 20892, USA) for strains, and Peter Williamson, June Kwon-Chung and Gary Cox for helpful comments. S. M. K. was supported in part by NIH training grant T32AI07271 to the Department of Microbiology, UTHSCSA. B. L. W. is a Burroughs–Wellcome New Investigator in Molecular Pathogenic Mycology and is supported by US Public Health Service Grant R29AI43522 from the National Institutes of Health. M. A. V. is supported by grant IRCCS Ospedale Maggiore di Milano RC 02/1999.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bennett, J. E., Kwon-Chung, K. J. & Howard, D. H. (1977). Epidemiologic differences among serotypes of Cryptococcus neoformans. Am J Epidemiol 105, 582–586.[Abstract]

Buchanan, K. L. & Murphy, J. W. (1998). What makes Cryptococcus neoformans a pathogen? Emerg Infect Dis 4, 71–83.[Medline]

Caldwell, G. A., Naider, F. & Becker, J. M. (1995). Fungal lipopeptide mating pheromones: a model system for the study of protein prenylation. Microbiol Rev 59, 406–422.[Abstract]

Casadevall, A. & Perfect, J. R. (1998). Cryptococcus neoformans, pp. 9, 418–438. Washington, DC: American Society for Microbiology.

Clarke, D. L., Woodlee, G. L., McClelland, C. M., Seymour, T. S. & Wickes, B. L. (2001). The Cryptococcus neoformans STE11{alpha} gene is similar to other fungal mitogen-activated protein kinase kinase kinase (MAPKKK) genes but is mating type specific. Mol Microbiol 40, 200–213.[CrossRef][Medline]

Cogliati, M., Esposto, M. C., Clarke, D. L., Wickes, B. L. & Viviani, M. A. (2001). Origin of Cryptococcus neoformans var. neoformans diploid strains. J Clin Microbiol 39, 3889–3894.[Abstract/Free Full Text]

Cohen, J. (1997). The daunting challenge of keeping HIV suppressed. Science 277, 32–33.[Free Full Text]

Cohen, J. (1998). AIDS therapy. Failure isn't what it used to be... but neither is success. Science 279, 1133–1134.[Free Full Text]

Cox, G. M., McDade, H. C., Chen, S. C. & 8 other authors (2001). Extracellular phospholipase activity is a virulence factor for Cryptococcus neoformans. Mol Microbiol 39, 166–175.[CrossRef][Medline]

Cruz, M. C., Cavallo, L. M., Gorlach, J. M., Cox, G., Perfect, J. R., Cardenas, M. E. & Heitman, J. (1999). Rapamycin antifungal action is mediated via conserved complexes with FKBP12 and TOR kinase homologs in Cryptococcus neoformans. Mol Cell Biol 19, 4101–4112.[Abstract/Free Full Text]

Edberg, S. C., Chaskes, S. J., Alture-Werber, E. & Singer, J. M. (1980). Esculin-based medium for isolation and identification of Cryptococcus neoformans. J Clin Microbiol 12, 332–335.[Medline]

Evans, E. E. (1949). An immunologic comparison of twelve strains of Cryptococcus neoformans (Torula histolytica). Proc Soc Exp Biol Med 71, 644–646.

Evans, E. E. (1950). The antigenic composition of Cryptococcus neoformans. I. A serologic classification by means of the capsular agglutinations. J Immunol 64, 423–430.

Evans, E. E. & Kessel, J. F. (1951). The antigenic composition of Cryptococcus neoformans. II. Serologic studies with capsular polysaccharide. J Immunol 67, 109–114.

Franzot, S. P., Salkin, I. F. & Casadevall, A. (1999). Cryptococcus neoformans var. grubii: separate varietal status for Cryptococcus neoformans serotype A isolates. J Clin Microbiol 37, 838–840.[Abstract/Free Full Text]

Garcia-Martos, P., Marin, P., Hernandez-Molina, J. M., Garcia-Agudo, L., Aoufi, S. & Mira, J. (2001). Extracellular enzymatic activity in 11 Cryptococcus species. Mycopathologia 150, 1–4.[CrossRef][Medline]

Hamilton, A. J. & Goodley, J. (1996). Virulence factors of Cryptococcus neoformans. Curr Top Med Mycol 7, 19–42.[Medline]

Heitman, J., Casadevall, A., Lodge, J. K. & Perfect, J. R. (1999). The Cryptococcus neoformans genome sequencing project. Mycopathologia 148, 1–7.[CrossRef][Medline]

Hironaga, M., Ikeda, R., Fukazawa, Y. & Watanabe, S. (1983). Mating types and serotypes of Cryptococcus neoformans isolated in Japan. Sabouraudia 21, 73–78.[Medline]

Jong, S. C., Bulmer, G. S. & Ruiz, A. (1982). Serologic grouping and sexual compatibility of airborne Cryptococcus neoformans. Mycopathologia 79, 185–188.[Medline]

Kozel, T. R. (1995). Virulence factors of Cryptococcus neoformans. Trends Microbiol 3, 295–299.[CrossRef][Medline]

Kwon-Chung, K. J. (1975). A new genus, Filobasidiella, the perfect state of Cryptococcus neoformans. Mycologia 67, 1197–1200.[Medline]

Kwon-Chung, K. J. (1976). Morphogenesis of Filobasidiella neoformans, the sexual state of Cryptococcus neoformans. Mycologia 68, 821–833.[Medline]

Kwon-Chung, K. J. (1987). Filobasidiaceae–a taxonomic survey. Stud Mycol 30, 75–85.

Kwon-Chung, K. J. & Bennett, J. E. (1978). Distribution of {alpha} and a mating types of Cryptococcus neoformans among natural and clinical isolates. Am J Epidemiol 108, 337–340.[Abstract]

Kwon-Chung, K. J. & Bennett, J. E. (1992). Medical Mycology. Philadelphia: Lea & Febiger.

Kwon-Chung, K. J., Bennett, J. E. & Rhodes, J. C. (1982a). Taxonomic studies on Filobasidiella species and their anamorphs. Antonie van Leeuwenhoek 48, 25–38.[Medline]

Kwon-Chung, K. J., Polacheck, I. & Popkin, T. J. (1982b). Melanin-lacking mutants of Cryptococcus neoformans and their virulence for mice. J Bacteriol 150, 1414–1421.[Medline]

Kwon-Chung, K. J., Edman, J. C. & Wickes, B. L. (1992). Genetic association of mating types and virulence in Cryptococcus neoformans. Infect Immun 60, 602–605.[Abstract]

Lengeler, K. B., Wang, P., Cox, G. M., Perfect, J. R. & Heitman, J. (2000). Identification of the MATa mating-type locus of Cryptococcus neoformans reveals a serotype A MATa strain thought to have been extinct. Proc Natl Acad Sci U S A 97, 14455–14460.[Abstract/Free Full Text]

Lengeler, K. B., Cox, G. M. & Heitman, J. (2001). Serotype AD strains of Cryptococcus neoformans are diploid or aneuploid and are heterozygous at the mating-type locus. Infect Immun 69, 115–122.[Abstract/Free Full Text]

Madrenys, N., De Vroey, C., Raes-Wuytack, C. & Torres-Rodriguez, J. M. (1993). Identification of the perfect state of Cryptococcus neoformans from 195 clinical isolates including 84 from AIDS patients. Mycopathologia 123, 65–68.[Medline]

Matsushita, S. (2000). Current status and future issues in the treatment of HIV-1 infection. Int J Hematol 72, 20–27.[Medline]

McClelland, C. M., Fu, J., Woodlee, G. L., Seymour, T. S. & Wickes, B. L. (2002). Isolation and characterization of the Cryptococcus neoformans MATa pheromone gene. Genetics 160, 935–947.[Abstract/Free Full Text]

Miskin, A. & Edberg, S. C. (1978). Esculin hydrolysis reaction by Escherichia coli. J Clin Microbiol 7, 251–254.[Medline]

Mitchell, T. G. & Perfect, J. R. (1995). Cryptococcosis in the era of AIDS–100 years after the discovery of Cryptococcus neoformans. Clin Microbiol Rev 8, 515–548.[Abstract]

Pinner, R. W., Hajjeh, R. A. & Powderly, W. G. (1995). Prospects for preventing cryptococcosis in persons infected with human immunodeficiency virus. Clin Infect Dis 21 Suppl. 1, S103–S107.[Medline]

Piot, P., Bartos, M., Ghys, P. D., Walker, N. & Schwartlander, B. (2001). The global impact of HIV/AIDS. Nature 410, 968–973.[CrossRef][Medline]

Price, M. F., Wilkinson, I. D. & Gentry, L. O. (1982). Plate method for detection of phospholipase activity in Candida albicans. Sabouraudia 20, 7–14.[Medline]

Rhodes, J. C., Polacheck, I. & Kwon-Chung, K. J. (1982). Phenoloxidase activity and virulence in isogenic strains of Cryptococcus neoformans. Infect Immun 36, 1175–1184.[Medline]

Schonheyder, H. & Stenderup, A. (1982). Isolation of Cryptococcus neoformans from pigeon manure on two media inducing pigment formation. Sabouraudia 20, 193–197.[Medline]

Vidotto, V., Leone, R., Sinicco, A., Ito-kuwa, S. & Criseo, G. (1998). Comparison of phospholipase production in Cryptococcus neoformans isolates from AIDS patients and bird droppings. Mycopathologia 142, 71–76.[CrossRef][Medline]

Viviani, M. A., Swinne, D., Kouzmanov, A. & 18 other authors (2000). Survey of cryptococcosis in Europe. The ECCM working group report. Rev Iberoam Micol 17, S115.

Viviani, M. A., Esposto, M. C., Cogliati, M., Montagna, M. T. & Wickes, B. L. (2001). Isolation of a Cryptococcus neoformans serotype A MATa strain from the Italian environment. Med Mycol 39, 383–386.[Medline]

Vogel, R. A. (1966). The indirect fluorescent antibody test for the detection of antibody in human cryptococcal disease. J Infect Dis 116, 573–580.[Medline]

Wickes, B. L., Moore, T. D. & Kwon-Chung, K. J. (1994). Comparison of the electrophoretic karyotypes and chromosomal location of ten genes in the two varieties of Cryptococcus neoformans. Microbiology 140, 543–550.[Abstract]

Wickes, B. L., Mayorga, M. E., Edman, U. & Edman, J. C. (1996). Dimorphism and haploid fruiting in Cryptococcus neoformans: association with the {alpha}-mating type. Proc Natl Acad Sci U S A 93, 7327–7331.[Abstract/Free Full Text]

Wills, E. A., Roberts, I. S., Del Poeta, M., Rivera, J., Casadevall, A., Cox, G. M. & Perfect, J. R. (2001). Identification and characterization of the Cryptococcus neoformans phosphomannose isomerase-encoding gene, MAN1, and its impact on pathogenicity. Mol Microbiol 40, 610–620.[CrossRef][Medline]

Xu, J., Vilgalys, R. & Mitchell, T. G. (2000). Multiple gene genealogies reveal recent dispersion and hybridization in the human pathogenic fungus Cryptococcus neoformans. Mol Ecol 9, 1471–1481.[CrossRef][Medline]

Yan, Z., Li, X. & Xu, J. (2002). Geographic distribution of mating type alleles of Cryptococcus neoformans in four areas of the United States. J Clin Microbiol 40, 965–972.[Abstract/Free Full Text]

Yue, C., Cavallo, L. M., Alspaugh, J. A., Wang, P., Cox, G. M., Perfect, J. R. & Heitman, J. (1999). The STE12alpha homolog is required for haploid filamentation but largely dispensable for mating and virulence in Cryptococcus neoformans. Genetics 153, 1601–1615.[Abstract/Free Full Text]

Received 31 July 2002; revised 20 September 2002; accepted 24 September 2002.