INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Cryptococcus neoformans is an encapsulated, basidiomycetous yeast and is the causative agent of cryptococcosis, a rare but potentially serious disease of humans and animals (5). Two varieties of C. neoformans exist, and these differ biochemically, genetically, ecologically, and epidemiologically (13). C. neoformans var. neoformans has a worldwide distribution and has been associated with a variety of environmental sources, in particular, bird excreta (8) and decaying wood, forming hollows in a number of tree species (18). C. neoformans var. gattii has a more restricted global distribution, occurring in tropical and subtropical climates. Since 1989, C. neoformans var. gattii has been shown to have a specific ecological association with a number of eucalyptus species; Eucalyptus camaldulensis (river red gum), Eucalyptus tereticornis (forest red gum), Eucalyptus rudis (West Australian flooded gum), Eucalyptus gomphocephala (tuart) (4, 21, 22), Eucalyptus grandis (flooded gum), Eucalyptus blakelyi (Blakely's red gum), and Angophora costata (smooth-barked apple) (unpublished data). These trees are native to Australia, where a relatively high incidence of cryptococcosis due to C. neoformans var. gattii occurs in some native animals and indigenous human populations (6). They have also been exported to other tropical parts of the world, and C. neoformans var. gattii infections are also found in these regions (4). However, the role that the tree plays in the life cycle of this fungus and the nature of the infectious propagule are not well understood. Viable C. neoformans var. gattii cells have been found in the woody debris and detritus associated with the Eucalyptus species, and there appears to be a correlation between the flowering of the trees and the dispersal of the fungus (3). However, it is not known whether the fungus completes its life cycle on the tree to be shed as basidiospores or whether it propagates asexually and disperses as desiccated yeast cells. Basidiospores are favored as the infectious propagule, as they are small (<2 µm), can penetrate into the lung alveoli, and are more resistant to desiccation than yeast cells (13).

C. neoformans exists as two mating types, mating types  and a, with the mating type determined by a single locus with two idiomorphic alleles (10). In laboratory crosses, equal numbers of offspring of the a and  mating types are produced (10). However, surveys of C. neoformans var. neoformans isolates from clinical and environmental sources have shown that the  mating type outnumbers its a counterpart by ratios of 30:1 and 40:1, respectively (12, 19). In addition, population genetic studies of C. neoformans var. neoformans have indicated a clonal structure which may either influence or be influenced by this imbalance of mating types (7). The limited data available for C. neoformans var. gattii isolates found 84% of the clinical isolates to be of the  mating type, but no data are available for the mating type frequencies of C. neoformans var. gattii isolates in the environment (19).

C. neoformans var. neoformans strains of the  mating type have been associated with increased virulence in mice, which has prompted studies into the mating type locus (MAT) (15). Molecular studies have estimated MAT to be at least 75 kb in size (30). Within this locus, two -mating-type genes have been identified: MF and STE12. The MF gene contains a 114-bp open reading frame that encodes a pheromone precursor and that has homology to other fungal mating factors (20). STE12 is a homologue of the Saccharomyces cerevisiae STE12 gene and is predicted to encode an 855-amino-acid protein. Between amino acids 85 to 201 lies a homeodomain with a high degree of identity to the STE12 homeodomains of other fungal species (30). All molecular work on the mating-type locus of C. neoformans done to date has used C. neoformans var. neoformans. However, hybridization studies with C. neoformans var. gattii DNA have indicated the presence of similar or identical MF and STE12 genes in strains of the  mating type (30, 32).

The current study was undertaken to survey for the presence of the two mating types, mating types  and a, in clinical and environmental collections of C. neoformans var. gattii from different regions of Australia. As mating between isolates of C. neoformans var. gattii was difficult to induce, we used a molecular approach, using specific PCR primers to selectively amplify the MF and STE12 sequences from -mating-type strains.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Collection of C. neoformans var. gattii isolates. One hundred thirty-one environmental isolates were obtained from various trees in three Australian states: South Australia (SA), New South Wales (NSW), and Queensland (QLD). The isolates were collected either from a single tree or from a number of trees (in close proximity). In addition, a few of the environmental isolates came from dead branches used as perches within a koala enclosure in a wildlife park. The details about these isolates are summarized in Table 1.

View larger version (17K): [in this window] [in a new window]   FIG. 1.   Distribution of environmental and clinical isolates used in the study: 1, Balranald; 2, Hay; 3, Adelaide; 4, Gold Coast; 5, Renmark; 6, Sydney; 7, Coffs Harbour; 8, Pilliga; 9, Breza/Tamworth; 10, Port Macquarie; 11, West Wyalong; 12, Perth; 13, Dubbo; 14, Newcastle; 15, Alice Springs; 16, Darwin; 17, Townsville; 18, Melbourne; 19, Arnhem Land; 20, Brisbane; 21, Toowomba.

Mating type crosses. The reference strains B-3501 (mating type ) and B-3502 (mating type a) of Filobasidiella neoformans var. neoformans (10) and strains CBS 6956 (mating type ) and CBS 6955 (mating type a) of F. neoformans var. bacillispora (16) were used in all mating-type crosses. The media used for the crosses included V8 juice agar (14), sucrose biotin agar (11), eucalyptus seed agar (50 g of E. camaldulensis seed, 1 g of glucose, 1 g of KH₂PO₄, 1 g of creatinine, 15 g of Bacto Agar, 1,000 ml of distilled H₂O, 1 ml of penicillin G [20 U/ml], and 1 ml of gentamicin [80 mg/ml]), and moist autoclaved E. camaldulensis bark. A loopful of 2-day-old yeast cells of the strain being studied was mixed with a loopful of an  or a mating type tester strain in the center of the plate or bark (25), and the plate or bark was incubated at 25°C for 2 weeks. This was observed periodically for the development of the perfect state.

DNA isolation. The chromosomal DNA extraction procedure was based on the Novozyme 234, dodecyltrimethylammonium bromide, and hexadecyltrimethylammonium bromide method described by Wen et al. (29) with the following modifications: approximately 0.75 g of cells (wet weight) grown on Sabouraud dextrose agar were collected, the protoplasting solution was made with 10 mg of Novozyme 234 per ml of SCS buffer (20 mM sodium citrate, 1 M sorbitol), and all centrifugation steps were performed at 12,879 × g. The DNA pellet was resuspended in 100 µl of Tris-EDTA buffer (10 mM Tris-HCl, 1 mM EDTA [pH 8]) containing 10 µg of RNase A (Progen) per ml. DNA was diluted 1:10 for PCR amplification.

Primer design. The MF primers (primers MFU and MFL) were designed from within the open reading frame of the MF pheromone gene (20) and were expected to amplify a 109-bp fragment from -mating-type strains. The STE12 primers (primers STE12U and STE12L) were designed from within the homeodomain of the STE12 gene (30) and were expected to amplify a 150-bp fragment from MAT strains. Primers 660U and 660L were designed from an anonymous DNA fragment amplified by randomly amplified polymorphic DNA (RAPD) analysis-PCR from a C. neoformans var. gattii strain. These primers were used in coamplifications with the MF primers and were designed to amplify a 216-bp fragment from all C. neoformans var. gattii strains. Oligo 5.0 software was used to optimize the design of all primers. Primer sequences are listed in Table 4.

PCR amplification and sequencing. PCR amplifications were performed in 50-µl volumes containing 1× PCR buffer (0.1 M Tris-HCl [pH 8.3], 0.5 M KCl, 15 mM MgCl₂, 0.1% gelatin), 5% glycerol, 250 µM deoxynucleoside triphosphates, 2.5 U of Taq polymerase, 1 µl of diluted template DNA, and either 30 pmol of primers MFU and MFL plus 25 pmol of primers 660U and 660L or 25 pmol of primers STE12U and STE12L. The reaction mixtures were overlaid with sterile mineral oil (Sigma). Amplification conditions for PCR were 94°C for 5 min, followed by 30 cycles of 94°C for 1 min, 55 or 50°C for 1 min, and 72°C for 1 min, and a final extension at 72°C for 7 min. All amplifications were carried out in a Perkin-Elmer Cetus model 480 Thermal Cycler. A total of 10 µl of each amplification product was electrophoresed at 10 V/cm and 40 mA in 2% agarose gels containing 0.50 ng of ethidium bromide per ml. The gels were visualized by UV transillumination and were photographed.

Nucleotide sequence accession numbers. The GenBank accession numbers for the MF sequences are AF 155335, AF 155336, AF 155337, AF 155338, AF 155339, AF 155340, and AF 155341. The GenBank accession numbers for the STE12 sequences are AF 155342, AF 155343, AF 155344, AF 155345, AF 155346, AF 155347, AF 155348, and AF 155349.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Mating-type crosses. The  and a mating type test strains of F. neoformans var. neoformans were observed to mate only on the V8 juice agar, producing a dense white mycelial phase at the edge of the yeast colony. Basidium-producing chains of basidiospores and hyphae with clamp connections were observed under the microscope. These strains did not mate on any of the other media tested.

Coamplification with primers MF and 660. The MF primers successfully amplified a 109-bp fragment from all culture collection strains of both varieties known to be of the  mating type but did not produce a fragment from any of the strains of the a mating type. Coamplification with the MF and 660 primers was performed with DNAs from all of the environmental and clinical isolates listed in Tables 1, 2, and 3. The CBS strains characterized as being of the  and a mating types were included in each PCR run to act as positive controls. Representative profiles are shown in Fig. 2 and 3a, and a complete summary of the ratio of  mating types:a mating types is given in Table 5.

View larger version (46K): [in this window] [in a new window]   FIG. 2.   Representative gel of DNA from clinical and environmental C. neoformans var. gattii isolates coamplified with the MF and 660 primers. Lanes: 1 and 24, pGEM size marker (Promega); 2, GC5; 3, GC12; 4, GC17; 5, GC22; 6, GC27; 7, Ad5; 8, Ad12; 9, Ad17; 10, Ad22; 11, Ad26; 12, 1408; 13, 571 146; 14, 571 067; 15, 1409; 16, 571 178; 17, H28; 18, H23; 19, H13; 20, H8; 21, CBS 5757 (); 22, CBS 6998 (a); 23, negative control for PCR amplification. The upper band of 216 bpl of 109 bp was amplified by the 660 primers; the lower band was amplified by the MF primers. Primer dimer can be seen below many of the amplified fragments. Lanes 12, 15, and 18 show slight variation in size of 660 fragment.

View larger version (87K): [in this window] [in a new window]   FIG. 3.   Representative gel of DNA from Balranald isolates coamplified by the MF and 660 primers (a) and the STE12 primers (b). Lanes 1: pGEM size marker; 2, Bal 23; 3, Bal 24; 4, Bal 25; 5, Bal 26; 6, Bal 27; 7, Bal 28; 8, Bal 29; 9, Bal 30; 10, 401 Bal 2; 11, 402 Bal 6; 12, 402/1; 13, 403 Bal 8; 14, 405 Bal 2c; 15, 406 Bal 2d; 16, 407 Bal 2f; 17, 408 B1; 18, 409 B2; 19, 410 B5; 20, 404 H22b; 21, CBS 5757 (); 22, CBS 6998 (a). Primer dimers can be seen in many of the lanes.

neoformans MF

View larger version (29K): [in this window] [in a new window]   FIG. 4.   Sequence alignments of the 109-bp MF fragment with the MF gene (GenBank accession no. S56460), which covers nucleotide positions 10 to 118 (a), and the 149-bp STE12 fragment with the corresponding segment of the STE12 GenBank sequence (accession no. AF012924), which covers nucleotide positions 1194 to 1343 (b). Both GenBank sequences were from C. neoformans var. neoformans isolates. The primer sequences are shown in bold, and asterisks indicate positions which are identical in all isolates.

Amplification with STE12 primers. The STE12 primers were designed to confirm the results already obtained from the MF-660 coamplifications, as failure to produce the MF band could also be due to a polymorphism(s) in the primer binding sites. Representative amplification profiles are shown in Fig. 3b. Previously characterized CBS strains of the  and a mating types were again included as positive controls. All isolates that did not produce the MF band also failed to produce the STE12 band. All remaining isolates produced both bands. Direct sequencing of the STE12 PCR fragments from five C. neoformans var. gattii isolates and three C. neoformans var. neoformans isolates found that they had a high degree of homology to the corresponding segment of the published STE12 gene, with polymorphisms shared between isolates belonging to each variety (Fig. 4b) (30).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

For more than a decade, C. neoformans var. gattii has been known to have a specific ecological association with a number of eucalyptus species. Epidemiological studies have supported the association between human clinical disease and the natural reservoir of the fungus (24). The current study was undertaken to survey for the presence of the  and a mating types in C. neoformans var. gattii strains isolated from the environment and from humans and animals with cryptococcosis. The eventual aim of this work will be to determine whether the fungus completes its life cycle in association with the eucalyptus host, undergoing sexual recombination and producing potentially infectious basidiospores.

Mating analyses failed to determine the mating type of any of the isolates used in this study, despite the use of a variety of different isolates and a range of conditions. Kwon-Chung and colleagues (17) experienced similar difficulties with inducing mating in four strains of C. neoformans var. gattii isolated from E. camaldulensis trees. We therefore used a molecular approach to determine mating type.

To date, only the -mating-type locus has been isolated and sequenced, and we were able to assign the a mating type to an isolate only by failure to amplify the -mating-type-specific sequences. A positive control was therefore included in the MF amplifications to ensure that the absence of the MF fragment was not due to inhibition of the PCR. Two isolates, isolates H26 and Bal 3, did not produce a band with the positive control 660 primers, but amplification was successful with the MF primers. RAPD analysis-PCR has since shown Bal 3 to be C. neoformans var. neoformans (data not shown), but isolate H26 is definitely C. neoformans var. gattii as it is serotype B. It is therefore likely that this isolate has a polymorphism at one of the 660 primer binding sites. In addition, slight differences in the sizes of the product amplified by the 660 primers were seen between some of the clinical isolates (Fig. 2). These isolates have been found to be of the VGII type, a minor variant of C. neoformans var. gattii previously reported by Sorrell et al. (24).

A total of 100% of the human and 97.4% of the animal clinical isolates produced both the MF band and the STE12 band and can therefore be assumed to be of the  mating type. This imbalance is similar to that found in studies of clinical isolates of C. neoformans var. neoformans (12, 19, 26) and may indicate that, like C. neoformans var. neoformans, C. neoformans var. gattii  isolates are more virulent than a isolates and are therefore more likely to infect humans and animals (15). The only MATa clinical isolate, 571 093, came from a dog with respiratory failure from which C. neoformans var. gattii was cultured from the lower respiratory tract. No unusual or atypical symptoms were noted in this cryptococcal infection.

Of the 131 environmental isolates of C. neoformans that were collected to determine their mating types, 130 were C. neoformans var. gattii, and 91 (70%) of these were of the  mating type. This is substantially lower than the result of a similar study with C. neoformans var. neoformans, in which 97.5% of the isolates were of the  mating type (12); however, the current study may have been biased by the inclusion of many isolates from single trees. Of the 31 isolates taken from separate trees, 80.6% were of the  mating type. This is still less than the proportion for C. neoformans var. neoformans, although the  mating type remains predominant. In C. neoformans var. neoformans, the bias in mating type has been postulated to be due to haploid fruiting. In this process, haploid  cells form extensive hyphae in the absence of the opposite mating type, producing abundant blastospores and basidia bearing viable basidiospores that are all of the  mating type. This has also been observed in C. neoformans var. gattii strains, but the hyphae are less extensive, and while blastospores are produced, no basidiospores have been detected. Haploid fruiting apparently cannot occur in a-mating-type strains of either variety (31).

MATa isolates were found in collections from two areas, Renmark and Balranald, which lie less than 300 km apart and which share very similar geographies and floras. All the Balranald isolates came from two separate samplings from an individual tree. This tree was unique in that the C. neoformans var. gattii isolates obtained from it were overwhelmingly of the a mating type, especially those isolated in 1996 (27 of 29). These isolates came from a large amount of debris in a semihollow area located at the base of the tree. When isolates taken in 1989 and 1990 were examined, 7 of 10 isolates were of the a mating type, indicating that the predominance of a-mating-type isolates in 1996 may have been caused by sampling artifact. Alternatively, the a mating type may be becoming more predominant over time via clonal propagation. We are currently using RAPD analysis-PCR to assess the genetic diversity in this collection.

The results presented in this report indicate that, in contrast to C. neoformans var. neoformans, both mating types can be found in some populations of C. neoformans var. gattii. In particular, in the population from Renmark, the ratio of the two mating types (10 :6 a) approximated the 50:50 ratio expected to result from sexual outcrossing. The ability of pathogens to recombine sexually is important for their ability to survive the host immune response and to adapt to novel environmental challenges, including antimicrobial therapy. The clonal population structure reported for C. neoformans var. neoformans (7) is unusual compared to those reported for other medically important fungi, in which a history of sexual exchange has been seen (1, 2). Our results suggest that C. neoformans var. gattii may also reproduce sexually, but further studies analyzing the association of molecular markers are necessary to test this hypothesis (28). The mating type survey indicates that Renmark will be a suitable region as a target for future studies of genetic recombination in C. neoformans var. gattii.