Ecoepidemiology of Cryptococcus gattii in Developing Countries

Cryptococcosis is a systemic infection caused by species of the encapsulated yeast Cryptococcus. The disease may occur in immunocompromised and immunocompetent hosts and is acquired by the inhalation of infectious propagules present in the environment. Cryptococcus is distributed in a plethora of ecological niches, such as soil, pigeon droppings, and tree hollows, and each year new reservoirs are discovered, which helps researchers to better understand the epidemiology of the disease. In this review, we describe the ecoepidemiology of the C. gattii species complex focusing on clinical cases and ecological reservoirs in developing countries from different continents. We also discuss some important aspects related to the antifungal susceptibility of different species within the C. gattii species complex and bring new insights on the revised Cryptococcus taxonomy.


Introduction
Cryptococcosis is a systemic fungal disease caused by yeasts belonging to the Cryptococcus neoformans/C. gattii species complexes [1], affecting both immunocompetent and immunocompromised hosts and causing devastating diseases [2]. Cryptococcal meningitis is the most common mycosis associated with acquired immune deficiency syndrome (AIDS) patients with significant morbidity and mortality especially in sub-Saharan Africa, Asia, and Latin America [3]. It is estimated that approximately 225,000 new cryptococcal meningitis cases occur globally each year, the majority of which (73%) occur in sub-Saharan Africa [3].
C. tetragattii (genotype AFLP7/VGIV) has been found in Puerto Rico from tree detritus [46] and the molecular type VGIV has also been found in México and Colombia, but AFLP or MLST genotyping has not been performed to differentiate C. tetragattii (genotype AFLP7/VGIV) and C. decagattii (genotype AFLP10/VGIV) from each other [33][34][35]. Some of these isolates have been recently investigated and shown to belong to C. decagattii rather than to C. tetragattii [1].

Africa
On the African continent, most of the literature data consists in descriptions of the C. neoformans species complex's ecological distribution and clinical involvement. But only few studies have found new C. gattii species complex members. In South Africa, HIV-positive and HIV-negative children and adults were reported to have cryptococcosis caused by C. gattii s.l., but no genotyping was performed to determine the species [51][52][53]. The same holds true for Botswana [49] and Rwanda, where cryptococcal meningitis cases with C. gattii s.l. were found [50]. Environmental niches of C. gattii s.l. were investigated in Zambia, where positive samples were found in Colophospermum mopane, Julbernadia globiflora, Eucalyptus spp., Brachystegia spp., fig tree, and feces from Hyrax midden, but no genotyping was performed [54].

Asia
There have been many studies performed in developing Asian countries reporting C. gattii s.l. causing human diseases and the ecological niche of this species complex. Unfortunately, most of the isolates were not genotyped. Chen and colleagues (2000) performed a study in Taiwan with clinical cases of cryptococcosis during the 1980s and 1990s [70]. Infections by C. gattii s.l. occurred in 35.6% of patients during the study period. The cryptococcosis cases included both immunocompetent and immunocompromised patients with a predominance of central nervous system (CNS) diseases [70]. In China, C. gattii s.l. was isolated from a surgical wound [61]. In addition, in India, it was isolated from HIV-positive and HIV-negative patients [62,63]. Environmental niches of C. gattii s.l. in India were recognized being tree hollows of Syzygium cumini, Ficus religiosa, Polyalthia longifolia, Azadirachta indica, Cassia fistula, Mimusops elengi, and Cassia marginata [64][65][66][67][68], and flowers, bark, and detritus of Eucalyptus camaldulensis and E. tereticornes [69].
C. deuterogattii (genotype AFLP6/VGII) has been isolated from chronic meningitis in Malaysia [80,81], HIV-negative patients from Korea and India [79,84,85], and immunocompetent patients in China [71,74,75]. In Thailand, this species was reported causing disease in HIV-positive and HIV-negative patients, as well as causing primary cutaneous cryptococcosis [83,86]. However, the environmental niche of this species has not been reported.
C. bacillisporus (genotype AFLP5/VGIII) was found causing diseases in patients from Korea [79,85], which is interesting because in Asia C. gattii s.s. and C. deuterogattii have a predominance among clinical samples. In India, the first environmental C. bacillisporus isolate was recovered from decaying wood of Manilkara hexandra [78].
C. tetragattii (genotype AFLP7/VGIV) was isolated in India from several clinical sources, including an HIV-positive patient with meningitis, cutaneous lesions, and granulomas in HIV-negative patients. All these isolates were genetically similar to C. tetragattii found in Botswana, Africa. However, only one patient had previously travelled to Egypt [87].
A hypothesis to explain the differences in geographic distribution of C. gattii/C. neoformans species complexes was put forward by Casadevall and colleagues (2017). These authors hypothesized that it may be attributed to the breakup of the supercontinent Pangea. The physical separation of Cryptococcus species complexes was an important point for its speciation [89]. In addition, it was suggested that environmental events, such as wind, ocean currents, and animals, would be involved, driving the more recent speciation of Cryptococcus species complexes [89]. Despite all epidemiological studies carried out, there are many countries where the presence of cryptococcal molecular genotypes has not yet been explored [90].
However, different antifungal susceptibility profiles are also present within species of the C. gattii species complex. C. deuterogattii (genotype AFLP6/VGII) has higher geometric mean MICs for flucytosine, fluconazole, voriconazole, itraconazole, posaconazole, and isavuconazole than C. gattii s.s. (genotype AFLP4/VGI) [95]. Lockhart and colleagues (2012) investigated the correlation of C. gattii species complex and its antifungal susceptibility [96]. C. deuterogattii (genotype AFLP6/VGII) had the highest geometric mean MIC for fluconazole, followed by C. bacillisporus (genotype AFLP5/VGIII), genotype VGIV (AFLP non-genotyped), while C. gattii s.s. (genotype AFLP4/VGI) had the lowest among species [96]. Trilles and colleagues (2012) also observed that C. deuterogattii isolates had higher MICs of azoles than C. gattii s.s. [93]. An Indian study showed that clinical and environmental C. gattii s.l. isolates had high fluconazole MICs [68]. However, despite these differences in antifungal susceptibility among cryptococcal species, the initial cryptococcosis therapy is the same; the clinical management changes are according to presentations and immune status, but does not consider the species involved in the disease [97].
Another important point is the phenomenon of heteroresistance, the ability of adaptation to a high concentrations of drugs, observed in C. gattii s.l. to itraconazole and fluconazole [98,99]. The development of heteroresistance is related to phenotypic changes, such as a decrease in cell and capsule size, low ergosterol content in the cell wall, less susceptibility to oxidative stress, and a great ability to proliferate inside macrophages [98,99]. This intrinsic mechanism present in members of the C. neoformans/C. gattii species complexes may contribute to a relapse of cryptococcosis during maintenance therapy [98,100]. However, the clinical importance of heteroresistance is not yet clear and requires further investigation [100,101].

The C. gattii Species Complex: Four Molecular Types, Five Genotypes or Five Species?
The taxonomy of the tremellomycetous yeasts has recently been revised [102,103]. Since the genus Cryptococcus was described, it has grown out as a highly polyphyletic one that contained more than 100 species within the orders Filobasidiales, Tremellales, and Trichosporonales [102,103]. The taxonomic revision of the genus Cryptococcus has been extensively discussed over the past two decades. At the 6th International Conference on Cryptococcus and Cryptococcosis (ICCC) debate, "How many species and varietal states are there?" [104], different hypotheses were discussed about the status of the C. neoformans/C. gattii species complex: Should the situation be kept in a "two-species division"? [105,106]. Should it be divided into six species? [107,108]. Eight? [109]. The hypotheses were supported based on different opinions about the definition of species. The first one was supported by the idea that phenetic, biological, and cladistic species concepts need to be used together to proper classify the agents of cryptococcosis, because genetic variation as shown by the molecular types does not always reflect their biological characteristics [105,106,110]. However, the second hypothesis was based on phylogenetic support that included analysis of mitochondrial, ribosomal, and nuclear genes to investigate the relationship among the various C. neoformans and C. gattii genotypes. The different genotypes clustered in six monophyletic lineages for all loci studied, suggesting that C. neoformans serotype A and D represent two different species and that C. gattii genotypes represent four individual taxa [107,108]. The third hypothesis goes a little further, considering that each genotype within C. neoformans and C. gattii has sufficient genotypic variation to be considered a different species [109].
Phenotypic diversity within the C. gattii species complex is also supporting the division of five species. Capsule and cell size showed to be variable within the complex, C. gattii s.s. (genotype AFLP4/VGI) had the largest capsules but smaller cells compared to the other species, while C. deuterogattii (genotype AFLP6/VGII) has the largest cells but smaller capsules [111]. All species in the C. gattii species complex have the ability to grow at 25, 30, and 35 • C, but with variable tolerance to 37 • C [1,111]. C. deuterogattii (genotype AFLP6/VGII) has the highest thermotolerance to 37 • C, while C. gattii s.s. (genotype AFLP4/VGI), C. bacillisporus (genotype AFLP/VGIII), and C. tetragattii (genotype AFLP7/VGIV) have less growth at 37 • C than 30 • C [1,111,112]. There is no significant difference in tolerance to oxidative or osmotic stresses among species [111,112].
The understanding of genetic diversity is an important step for the discovery of previously unrecognized phenetic differences [111]. The exact moment that individuals in an ancestral species are split into progeny species is not recognized for any method of species delimitation, because this process needs time until the changes in morphology, mating behavior, or gene sequences may be recognized in the progeny species [113]. Phylogenomic analyses calculated the time since divergence of the C. neoformans species complex and the C. gattii species complex to be~34 million years ago (mya) [114]; the divergence between C. deuterogattii (AFLP6/VGII) and the other species of the C. gattii species complex occurred~12 mya [114,115]; and the divergence of C. neoformans (AFLP1/VNI/AFLP1; VNII/AFLP1A/1B) and C. deneoformans (VNIV/AFLP2),~24 mya [115]. The divergence among the species within the C. neoformans/C. gattii species complexes occurred recently, and will most likely continue as an ongoing process. The occurrence of interspecies hybrids may also be attributed to the recent divergence event, because species currently hybridizing are most likely the youngest [116].
Although a revision of the cryptococcal taxonomy has been published, part of the cryptococcal research community is not fully in favor of using the 'seven species recognition'. Some investigators believe that it will lead to taxonomic instability due to the fact that there are most likely more species present. Many points have been discussed, including the number of isolates used, the use of phylogenetic approaches for species delineation, the accommodation of hybrids in the new taxonomy, and the fact that the new names may cause confusion between the published literature and clinical practice [110]. With these points of view, Kwon-Chung and colleagues (2017) suggested the use of the "C. neoformans species complex" and the "C. gattii species complex" as an intermediate step, instead of using the seven species nomenclature, until biological and clinical relevant differences become clear [110]. Although, according to Hagen and colleagues (2017), it is important to consider the presence of different species inside the complexes to avoid delay in the clinical progress [117].

Final Remarks
Clinical and environmental occurrence of the C. gattii species complex is related to geographic location, which may be attributed to the (micro)climate, or even a lack of diagnosis/environmental isolation. Cryptococcosis in most developing countries is underreported and the precise burden of cryptococcosis caused by the C. gattii species complex is uncertain. In addition, not all clinical laboratories differentiate the pathogenic Cryptococcus species. In the environment, many tree species have been described as a reservoir, proving that the C. gattii species complex has no tree species-specific relation, and is widely spread in the environment.
C. gattii species complex members differ in phenotypic traits, as capsule and cell size, thermotolerance and antifungal susceptibility. Many studies have demonstrated higher MICs of azoles for members of the C. gattii species complex compared to the C. neoformans species complex. Difference in antifungal susceptibility has also been observed within the C. gattii species complex, with C. deuterogattii (genotype AFLP6/VGII) being less susceptible to azoles than C. gattii s.s (genotype AFLP4/VGI). However, in vitro antifungal susceptibility does not correlate to in vivo susceptibility. Clinical manifestations in patients with C. gattii s.l. infections tend to be more severe than C. neoformans. In the former, cerebral involvement causes more hydrocephalus, focal CNS signs, as well as papilledema, ataxia, hearing loss, altered mentation, and neurological sequelae. Usually, meningo-encephalitis caused by C. gattii s.l. is followed by higher intracranial pressures, sometimes irresponsible to multiple LPs and/or CNF shunts. Simultaneous pulmonary involvement in >50% of patients is also observed, and mass lesions (cryptococcomas) are associated to a prolonged clinical course and respond slowly to therapy.
A new taxonomy of the polyphyletic genus Cryptococcus has been published, including the medical important species complexes C. neoformans and C. gattii. In addition, there are different opinions about the new classification. The presence of genetic differences within the C. gattii species complex needs to be considered in future studies to correlate genotypic and phenotypic traits of each species to diseases clinical presentation.