Global Molecular Epidemiology of Cryptococcus neoformans and Cryptococcus gattii: An Atlas of the Molecular Types

Cryptococcosis is a fungal disease affecting more than one million people per year worldwide. The main etiological agents of cryptococcosis are the two sibling species Cryptococcus neoformans and Cryptococcus gattii that present numerous differences in geographical distribution, ecological niches, epidemiology, pathobiology, clinical presentation and molecular characters. Genotyping of the two Cryptococcus species at subspecies level supplies relevant information to understand how this fungus has spread worldwide, the nature of its population structure, and how it evolved to be a deadly pathogen. At present, nine major molecular types have been recognized: VNI, VNII, VNB, VNIII, and VNIV among C. neoformans isolates, and VGI, VGII, VGIII, and VGIV among C. gattii isolates. In this paper all the information available in the literature concerning the isolation of the two Cryptococcus species has been collected and analyzed on the basis of their geographical origin, source of isolation, level of identification, species, and molecular type. A detailed analysis of the geographical distribution of the major molecular types in each continent has been described and represented on thematic maps. This study represents a useful tool to start new epidemiological surveys on the basis of the present knowledge.


Introduction
According to the last report (December 2010) from the Joint United Nations Program on HIV/AIDS and the World Health Organization (http://www.unaids.org/), 34 million people worldwide suffer from HIV infection/AIDS, 2.7 million people are newly infected every year from this disease, and 1.8 million people die from AIDS-related causes. Neuropathological conditions are present in approximately 70% to 90% of AIDS patients [1]. Cryptococcal meningitis is considered an AIDS-de�ning condition [2][3][4], and it is the most common fungal infection of the central nervous system and the third most frequent neurological complication in AIDS patients [1]. e main etiological agents of cryptococcosis are the basidiomycetous yeasts Cryptococcus neoformans and Cryptococcus gattii which can also infect, although with a signi�cantly lower incidence, people with decreased immunity such as individuals with sarcoidosis, lymphoproliferative disorders, those undergoing immunosuppressive therapies [4][5][6], and more rarely immunocompetent people [7,8]. Worldwide, C. neoformans and C. gattii infections cause an estimated one million cases of cryptococcal meningitis per year among people with HIV/AIDS, resulting in nearly 625,000 deaths (Centers for Disease Control and Prevention, CDC, Atlanta, USA, http://www.cdc.gov/). e greatest burden of disease occurs in sub-Saharan Africa, where mortality is estimated to be 50% to 70% [9,10]. In the United States and other developed countries, cryptococcosis is decreasing among persons with HIV/AIDS due to the availability of high active antiretroviral therapy, and, at present, the mortality is around 12% (CDC, Atlanta, USA).
Although in the past the etiological agent of cryptococcosis was considered a homogeneous anamorphic species (C. neoformans), now the two species C. neoformans and C. gattii have been separated on the basis of numerous differences such as geographical distribution, ecological niches, epidemiology, pathobiology, clinical presentation, and molecular characters. C. neoformans species is further classi�ed in two varieties, C. neoformans var. grubii (serotype A) and C. neoformans var. neoformans (serotype D), which are also  able to recombine and to produce diploid or aneuploid intervarietal AD hybrids [78,131,132]. C. neoformans has been widely associated to avian excreta [44,97,98,[133][134][135][136] although it has been isolated also from other sources [11,28,60,99,100,137].
C. gattii species is classi�ed in two different serotypes, B and C, which have not yet elevated to the variety status. is species was thought to be restricted to tropical and subtropical regions but a recent outbreak due to C. gattii infection, which occurred in Vancouver Island and North �est Paci�c Coast of America [121,138], has expanded the geographical area of this pathogen also to temperate regions. In addition, interspecies hybrids between C. neoformans and C. gattii have been found in Colombia, Brazil, India, and e Netherlands [29,122,139,140].
Besides a prevalent asexual life cycle, both species have also a bipolar sexual cycle with two mating types, MATa and MAT , the latter being the most prevalently isolated from both patients and environment. Filobasidiella neoformans and Filobasidiella bacillispora are the sexual states of C. neoformans and C. gattii, respectively [141,142]. During sexual recombination, either �laments with clamp connections and basidiospores are produced [143]. Recombinant basidiospores are also produced via same-sex mating [144] and are thought to be the propagules responsible for the infection of the host [143].
e availability of whole genome sequences from C. neoformans and C. gattii strains and the recent progresses in the molecular biology have greatly advanced our understanding of this pathogenic yeast [145,146]. e disease aspects of cryptococcal infection are becoming better de�ned, while the life cycle of this fungus in the environment remains less well established. How this fungus has spread worldwide, the nature of its population structure, and how it evolved to be a deadly pathogen are ongoing research subjects that are key to our understanding of this environmental pathogen.
Due to the importance of the C. neoformans/C. gattii species complex as human fungal pathogens, several research groups are currently focusing on the molecular determination of the number of genetically diverged subgroups within each species. Several molecular typing techniques have been applied: multilocus enzyme electrophoresis (MLEE) [147]; DNA �ngerprinting [148]; random ampli�cation of polymorphic DNA (RAPD) [12,149]; P�R �ngerprinting [150]; ampli�ed fragment length polymorphism (AFLP) [11]; restriction fragment length polymorphism (RFLP) of PLB1 [13], GEF1 [37], or URA5 genes [64]; sequencing of ITS1-5.8S-ITS2 rDNA region [38] or intergenic spacer region (IGS) [151] and, more recently, multilocus sequence typing (MLST) [14, 45,53]; multilocus microsatellite typing (MLMT) [89,140]; matrix-assisted laser desorption ionization-time of �ight mass spectrometry-based method (MALDI-T�F) analysis [79,152,153]. e multitude of data obtained with different typing methods has raised the problem to compare the results and the need to standardize genotypes nomenclature among comparable methods. e �rst steps towards this direction have been recently achieved by the comparison of the results obtained using the most common Cryptococcus typing methods [15] and by the identi�cation of eight major molecular types among the two Cryptococcus species. A second step was to standardize a gold standard typing method able to produce unambiguous and comparable results [154]. Finally, a global database was implemented in order to collect the different genotypic pro�le and to make the data available for the research community [154]. is task is the main aim of the activity of the ISHAM Cryptococcus working Group for "Genotyping of Cryptococcus neoformans and Cryptococcus gattii" which promotes the genotyping of the two Cryptococcus species to elucidate the global epidemiology of this life-threatening pathogen.

Standardization of Cryptococcus Molecular Typing Methods
Although numerous molecular techniques have been applied to subtype C. neoformans and C. gattii strains, only three methods were proved to produce comparable results: PCR �ngerprinting, AF�P, and M��T. PCR �ngerprinting is based on the ampli�cation of DNA sequences �an�ed by simple DNA repeats which are used as single primers in the PCR. e ampli�cation produces a banding pro�le that discriminates the strains at subspecies level. e primers employed in PCR �ngerprinting include the minisatellite-speci�c core sequence of wild-type phage M13 (5 ′ -GAGGGTGGCGGTTCT-3 ′ ) and the microsatellite-speci�c primer (GACA) 4 . e technique was �rst applied to Cryptococcus typing in 1993 [150] to study a set of cryptococcal strains. A high polymorphism was detected among the investigated strains which could be separated in two groups corresponding to serotype A and serotype D and a third one including both serotypes B and C. In a different study, a variety of genotyping clusters was identi�ed during the investigation of some Italian C. neoformans isolates by (GACA) 4 PCR �ngerprinting [155]. e results showed a strong correlation between genotypes and serotypes. e more prevalent genotype was named VN1 and corresponded to C. neoformans var. neoformans, serotype D, a second  e AFLP typing method is based on digestion of DNA samples with a frequent and a rare cutting endonuclease enzyme combined with ampli�cation using an adaptor that creates speci�city at the restriction sites. Subsequent rounds of PCR are able to select a unique pro�le depending on the number of nucleotides added to the primers. Fluorescently labeled fragments are separated by an automated capillary sequencer and visualized as a virtual banding pro�le [165]. Application of this technique to Cryptococcus typing requires a digestion with MseI and EcoRI restriction enzymes and an ampli�cation with the two selective primers MseI-G and EcoRI-AC [11]. e analysis of 207 global C. neoformans and C. gattii isolates led to identify three AFLP genotypes (AFLP1, APLP2, and AFLP 3) among C. neoformans strains and three (AFLP4, AFLP5, and AFLP6) among C. gattii. In addition, two further subtypes of the genotype AFLP1 were identi�ed as AFLP1A and AFLB1B [11]. Since AFLP is a technique with a high discriminatory power, being able to assign a unique pro�le to each strain, it contributed to elucidate the cause of the outbreak of C. gattii in Vancouver Island [157]. In the study, two AFLP6 subtypes were clearly identi�ed as the cause of the outbreak: AFLP6A and AFLP6B. Since AFLP6A was isolated in 75% of the cases in Vancouver Island environment and AFLP6B isolates were less frequent, it was hypothesized that the former genotype was more virulent than the latter. e higher virulence of AFLP6A than AFLP6B F 6: Percentage of Cryptococcus neoformans and Cryptococcus gattii isolates ( ,647) identi�ed at species comple�, species, variety/serotype, or molecular type level (a). Prevalence of the di�erent VN and VG molecular types among the isolates identi�ed at molecular type level ( ) (b). Geographic distribution of the molecular types identi�ed in Africa (c). Molecular typing data have been combined from the following references: Senegal [14], Republic Democratic of Congo [11,53], Uganda [53,57], Rwanda [11], Tanzania [53], Malawi [53,58,59], Zimbabwe [11], Botswana [53,58,60,61], and South Africa [11,15,38,60,62,63].
was actually shown later in murine model [14]. Finally, AFLP application to Cryptococcus typing contributed to discover new interspecies hybrids between C. neoformans and C. gattii. Both AFLP1/AFLP4 (AFLP9) and AFLP3/AFLP4 (AFLP8) hybrid isolates were identi�ed during some studies carried out in e Netherlands [122,139,140].
MLST is a typing technique based on sequence analysis of a set of polymorphic loci. e combination of the different allele types of the selected loci determines the MLST genotype [166]. One hundred and two global C. neoformans var. grubii isolates were analyzed by MLST in a �rst study which employed the following set of 12 loci: CAP10, CAP59, GPD1, LAC1, MPD1, MP88, SOD1, TEF1 , TOP1, URE1, and IGS1 [53]. e results showed two major clades among the studied isolates, corresponding to PCR �ngerprinting molecular types VNI and VNII and a third new clade, VNB, including only isolates from Botswana.
A second MLST study investigated 202 global C. gattii isolates in order to elucidate the origin of Vancouver Island outbreak isolates [14]. MLST analysis, using seven loci (CAP10, GPD1, IGS1, LAC1, MPD1, PLB1, and TEF1 ) and the two mating-type speci�c loci SXI and SXIa, was able to di�erentiate all the four PCR �ngerprinting molecular types, VGI-VGIV, as well as both Vancouver Island outbreak subtypes, VGIIa and VGIIb.
neoformans var. grubii. MLST pro�les of 1�7 Ugandan C. neoformans var. grubii clinical isolates were shown to be associated with the host immunological response providing a new tool to predict virulence [57]. Finally, four serotype-C VGIV C. gattii clinical isolates were identi�ed in India, and their MLST pro�les were found to be strictly correlated to those from South African VGIV isolates [34]. In order to include all C. neoformans var. grubii sequences and genotypes obtained using the standard MLST scheme, a preliminary MLST database was constructed at the Imperial College of London (London, UK, http://www.mlst.net/). Unfortunately, this database has the limit to require a long time for sequence check before the sequences could be included and assigned with the right sequence type code. To overcome these limits, a new MLST database has been established at the Molecular Mycology Research Laboratory (University of Sydney, Sydney, Australia, http://www.mycologylab.org/). e system assigns the sequence code automatically when a sequence is compared with the database and a progressive sequence code to new sequences. Subsequently, the users are required to send all the data necessary for quality control as well as the clinical data to complete the database. At present, the C.  neoformans var. grubii database contains 355 strains with 110 sequence types, and the C. gattii database contains 400 strains with 160 sequence types [169].

Combined Epidemiological Analysis
A total of 68,811 C. neoformans and C. gattii isolates, reported by hundreds of global research studies, were analyzed. Data search was performed in PubMed database (http://www.ncbi.nih.gov/) using the keyword "cryptococcus" combined with a country name, that is, "cryptococcus italy. " Each reference from the resulting list of references was selected if it reported data concerning the isolation of one or more Cryptococcus species complex isolates. Isolates reported without an identi�cation code or without a citation were considered new isolates and included in the analysis, whereas isolates reported from more than one paper were considered only once. en, all the isolates were analyzed on Identi�cation at species level was performed for 62% of the isolates, variety or serotype was identi�ed in 7%, and molecular type in 22% ( ). Only a small percentage of the isolates (9%) was identi�ed as Cryptococcus species complex (Figure 2). A total of 1,328 C. gattii and 900 C. neoformans isolates were reported. Among the isolates identi�ed at molecular type level, VGI represented the more frequently isolated molecular type (39%) followed by VNI (27%) and VGII (22%), the other molecular types were less frequent. No VGIV isolates have been yet isolated from Oceania. Figure 2 shows the molecular type geographical distribution in the different countries of Oceania. Although C. gattii, with molecular types VGI, VGII, and VGIII, is prevalent in Australia and in Papua New Guinea, only two VGIII isolates were reported from New Zealand where, on the contrary, C. neoformans (VNI, VNII, and VNIV) is prevalently isolated.

3.2.
Asia. e combined analysis including all the Asian countries showed that a total of 19,651 C. neoformans and C. gattii isolates were reported. China, India, and ailand, together, mainly contributed to the study reporting the 80% of the Asian isolates. Six percent of the isolates were recovered from the environment or from animals in Turkey, Israel, Iran, India, Nepal, China, ailand, Malaysia, Taiwan, Republic of Korea, and Japan ( Figure 3). In most of the environmental surveys, C. gattii was isolated from tree samples, namely, from Syzygium cumini, Mimusops [32,33,177], while the only ten isolates from an animal source were recovered from koalas living in two different zoos in Japan [178,179]. On the contrary, C. neoformans was prevalently isolated from pigeon and other birds excreta [180] and less frequently from trees such as Eucalyptus tree, Tamarindus arjuna, Tamarindus indica, Cassia �stola, Syzygium cumini, and Ficus religiosa [33,177,181,182], as well as from some vegetables and fruit (tomato, carrot, banana, eggplant, papaya, apple, and guava) [183,184]. Among animals, few C. neoformans isolates were isolated from cat and dog and one from a bandicoot [185,186]. e majority of the Asian isolates (74%) were identi�ed just at species complex level, 7% at species level and 11% at variety/serotype level, while the molecular type was determined only in 8% ( ,708) ( Figure 4). C. neoformans was the species prevalently isolated in Asia ( ,192), being C. gattii about tenfold less frequently ( ) isolated. Eighty-one percent of the isolates belong to VNI and 13.2% to VGI molecular type. VGII molecular type is also represented, although in low percentages, in all the Asian countries included in the analysis, except for Israel and Taiwan. VNIII and VNIV molecular types are present in China and in India, as well as one VNIII isolate was reported from ailand, whereas they are absent in the other Asian , recovered only 1% out the total reported isolates ( Figure 5). C. neoformans was not only isolated from pigeon and other birds excreta but also from soil and house dust [60,187,188], as well as from trees such as Eucalyptus camaldulensis, mopane tree, and baobab [60,189]. C. gattii was isolated from soil, Eucalyptus camaldulensis, and almond tree [60,189]. Two veterinary isolates were also reported from two cases of cryptococcosis affecting South African cheetahs [38,190]. e majority of the studies reported only the species of the isolates (68%), 19% were reported as Cryptococcus species complex, and 11% as variety or serotype ( Figure 6). Molecular typing techniques were applied to identify 2% of the isolates ( ). Of these, 68% were molecular type VNI. VNII and VNIII represent 11% and 1% of the isolates, respectively, and have been reported only from Uganda and South Africa. irteen percent of the African isolates belongs to VNB molecular type, which was initially considered endemic of Botswana [53] but that is present also in South Africa, Rwanda, and Republic Democratic of Congo. In addition, a consistent population of the rare VGIV molecular type was isolated in Botswana and Malawi [58]. Only one isolate belonging to VGII and four belonging to VGI molecular type were reported from Senegal [14] and Republic Democratic of Congo [11], respectively, whereas VNIV was totally absent among the African isolates included in the present study ( Figure 6). Figure 7 shows the European countries reporting the isolation of at least one Cryptococcus species complex strain. Data were lacking from some Balkan and Eastern European countries. e majority of the isolates were reported from France, Spain, Italy, and United Kingdom representing 82% out of the total ( ,736). Nine percent of the isolates were detected from environmental and veterinary sources (Figure 7). C. neoformans not only isolated from pigeon and other birds excreta, but also from bat guano and red fox faeces [65,191,192]. Veterinary isolates include strains recovered from cat, dog, magpie, and some isolates from striped grass mouse and degu living in a zoo [193][194][195][196]. Few C. neoformans strains were isolated from trees, namely, from Eucalyptus camaldulensis and oak tree [74,197]. Most of the C. gattii natural isolates were from Eucalyptus camaldulensis, Douglas tree, carob tree, and stone pine [66,74], whereas C. gattii animal infections were reported in a ferret and in some goats [65,67].

Europe. e map in
Variety or serotype was determined in 34% of the European isolates, species in 25%, molecular type in 15%, while the 26% was reported as Cryptococcus species complex (Figure 8). A total of 6,371 isolates were identi�ed as C. neoformans and 94 as C. gattii.
European molecular typing data are shown in Figure 8. e majority of the isolates belong to VNI molecular type (59%), although VNIII and VNIV molecular types were also reported in most of the countries representing 18.5% and 18.3%, respectively. C. gattii molecular types distribution in Europe is not yet well de�ned. VGI is the prevalent molecular type, being 43 isolates reported from Portugal, Spain, Italy, and e Netherlands. Few VGII isolates were reported from Greece, Switzerland, e Netherlands, and Denmark, and only one VGIII isolate was found in Greece. A relevant observation is the absence of typing results from the United Kingdom, despite the fact that 12% of the European isolates were reported from this country (Figure 8).
Seventy-seven percent of the isolates were identi�ed at least at species level (32% as variety or serotype, 23% as species, and 22% as molecular type), and 23% were reported as Cryptococcus species complex ( Figure 10). C. neoformans was recognized in 6,665 and C. gattii in 1,464 isolates. e combined analysis of the molecular typing data reported from Brazil (1,439 isolates) showed that all the molecular types, except for VGIV, are represented in this country. e majority of the isolates in Brazil belong to VNI ( ) molecular type followed by VGII ( ), while VNII, VNIV, VGI, and VGIII occurred in a lower but similar percentage. Two isolates of VNIII as well as one VNI/VGII hybrid were also reported. In Colombia (542 isolates), the prevalence of molecular types was similar to that observed in Brazil, except for VGIII, which occurred in a higher percentage than VNII, VGI, and VGIV, as well as VNIV, that was recognized only in one isolate. VNIII AD hybrids seem to be absent in Colombia. Data from Argentina (94 isolates) showed that the VGI molecular type is the prevalent genotype among C. gattii isolates, in contrast to that observed in Brazil and Colombia, where it is the VGII. In Cuba, 198 VNI isolates were detected, whereas only one isolate was VGI. On the contrary, in the near Puerto Rico, only C. gattii isolates (16 VGII and one VGIV) were reported. Finally, all the four C. neoformans molecular types were reported from Chile, but no C. gattii isolates were found ( Figure 10).

North America.
A total of 7,922 C. neoformans and C. gattii isolates were reported from the USA (79%), Canada (15%), and Mexico (6%). Eighty percent of the isolates were from clinical sources, whereas 20% were recovered from the environment and animals ( Figure 11). Pigeon droppings were the main source for C. neoformans isolation [210,211], although, in Mexico, it was also isolated from fruit and vegetables [212]. ree C. neoformans isolates (two in Canada and one in the United States) caused infection in ferrets [170,213]. Isolation of C. gattii from the environment and from animals was widely reported from Canada during the monitoring of the Vancouver Island C. gattii outbreak. Soil, trees, and animals living in Vancouver Island (dogs, cats, horses, ferrets, and birds) resulted colonized or infected with this pathogen [121]. Outside Canada, VGIIa isolates were found in the environment and animals (air, water, soil, tree, cats, dogs, alpacas, and parrots) in Oregon and Washington State [123,124], and one VGI strain was isolated from Eucalyptus camaldulensis in Mexico [214].
Almost half of the isolates (49%) reported from North America were identi�ed only as Cryptococcus species complex, 10% were identi�ed at species level, while variety or serotype was reported for 21%. Molecular type was determined in 20% of the isolates ( ,707) ( Figure 12). Despite the fact that C. neoformans was more frequently isolated in North America than C. gattii (3,148 versus 885 isolates, resp.), 39% of the isolates identi�ed by molecular techniques belong to VGIIa molecular type. is is due to the extensive effort produced to discover the cause of Vancouver Island outbreak which, at present, includes 473 VGIIa, 57 VGIIb, and 70 VGI isolates [121]. In addition, a recent study has reported the infection of a Canadian patient with an interspecies VNI/VGI AB hybrid strain [122]. VNI was the prevalent molecular type in both the United States and Mexico, where VNII, VNIII, VNIV, and VGI are also present in lower percentages. VGII and VGIIa C. gattii molecular types were reported from the Northwest Paci�c Coast of United States, while VGIII was reported more frequently from Mexico and Southern California [64,120,125]. Five VGIV isolates were reported from Mexico [64,120], although this molecular type is absent in Canada and in the United States.

Concluding Remarks
e present combined analysis shows that about 68,811 C. neoformans/C. gattii isolates were reported in the world till now. e majority of the isolates were reported from Asia and Africa (19,651 and 19,647 198), and Brazil ( ,709). On the contrary, data are completely lacking from many countries of Africa, Asia, and Eastern Europe. United States is the country where the environment was more extensively surveyed (1089 isolates), followed by Brazil ( ), Australia ( ), Colombia ( ), and India ( ). Although 723 environmental isolates were also reported from Canada, they are not representative of the whole country since they were recovered during the monitoring of a restricted territory such as Vancouver Island area. In 38.5% of the isolates ( ,473) reported in the literature, the species was not determined, whereas among the isolates identi�ed at least at species level (61.5%, ,338), C. neoformans was about eightfold more frequently isolated than C. gattii (88.6% versus 11.4%). e C. neoformans/C. gattii ratio is variable for each continent being 68 : 1 in Europe, 33 : 1 in Africa, 7.6 : 1 in Asia, 4.5 : 1 in Central and South America, 3.5 : 1 in North America, and 1 : 1.5 in Oceania, where C. gattii is the prevalent species isolated.
Molecular type was determined for 8,077 isolates (12%) representing only a part of the world countries. Molecular data are absent from large parts of Africa, Asia, Eastern Europe, as well as from United Kingdom, Ireland, Norway, and Finland. VNI is the prevalent molecular type worldwide except in Australia and Papua New Guinea, where it is VGI. is latter molecular type was also found in 13.2% of the Asian, in 7% of the North American, in 4% of the Central and South American, and in 3.4% of the European isolates, while only four VGI strains have been reported from Africa. VNII is a rare molecular type which is reported from all the continents, except from Europe, in low percentages. However, a recent MLST study carried out in Italy has showed the presence of one VNB and three VNII strains also among a group of Italian clinical isolates, suggesting that these populations are underestimated in the European continent [215]. In addition, two VNB isolates from Brazil and Colombia, previously reported as VNII, were recognized by MLST analysis, con�rming that VNB molecular type is not endemic of Southern Africa [22]. e distribution and prevalence of the VGII molecular type is relevant to elucidate the origin of the Vancouver Island and Northwest Paci�c Coast C. gattii outbreak. e present analysis has identi�ed four main reservoirs of VGII molecular type: Brazil (266 isolates), Colombia ( ), Australia ( ), and Puerto Rico ( ). ese data con�rm the hypotheses suggested by other authors that the Vancouver outbreak could be originated from Australia [14] or from South America [22]. e VGIII molecular type has been prevalently detected in Latin American countries, including Mexico and Sothern California (134 isolates). In the other continents, VGIII is very rare, counting one isolate in Republic of Korea, one in Greece, and six in Oceania. e abundance of VNIII AD hybrids seems to be strictly related to the presence of VNIV molecular type. In Europe and in the USA, where the frequency of isolation of VNIV strains is higher than in other geographical areas (18% and 6%, resp.), a similar percentage of VNIII isolates has been observed, suggesting that in these regions hybridization between VNI and VNIV populations is occurring. VGIV molecular type was reported from Southern Africa ( ), India ( ), Colombia ( ), and Mexico ( ), but a recent MLST study, comparing these isolates, has revealed that Indian and Southern African isolates are strictly correlated and different from those from South America [34]. Finally, the interspecies C. neoformans/C. gattii hybrids have been rarely reported from different geographical areas, namely, one from India, one from Colombia, one from Brazil, one from Canada, and three from e Netherlands. However, due to the difficulty to identify these hybrids, it is likely that their prevalence is underestimated.
In conclusion, the present study describes the state of the art of C. neoformans and C. gattii genotyping by a detailed representation of the geographical distribution of the major molecular types, which could be a useful tool to start new epidemiological surveys on the basis of the present knowledge.
��n��c� �f �n�eres�s e author has no con�ict of interests.