Australia in the global picture of the molecular epidemiology of Cryptococcus gattii molecular type VGII

Cryptococcosis, a life-threatening disease of the lung and central nervous system of humans and a broad range of other animals, is caused by the basidiomycetous yeasts Cryptococcus neoformans and C. gattii¹. Although most cases of infection in the world are caused by C. neoformans, there is an important prevalence of C. gattii among clinical and veterinary samples in Australia^2–4 and the natural habitat of the yeast is strongly associated with native Eucalyptus species^4,5, which together position Australia as an endemic area for the less common cryptococcal species. From the numerous C. gattii infections that have been reported in Australia, the molecular type VGII, amongst the four recognised molecular types (VGI-VGIV), has been associated with a high occurrence and a number of small cryptococcosis outbreaks, with most of the isolates belonging to the clonal subtype VGIIb², which was initially described in 1999 causing part of the ongoing cryptococcosis outbreak on Vancouver Island, British Columbia, Canada⁶. These findings indicate that Australia is an important stepping-stone in the global dispersion of this outbreak-related subtype and highlight the need for continuous surveillance.

Since it was first reported in Vancouver Island and because of an increased number of human and veterinary C. gattii infections, previously uncommon in North America^6,7, C. gattii has been the focus of several studies conducted to elucidate the underlying evolutionary mechanisms involved in its emergence and dispersal in temperate climates and to determine the potential geographical origin of this fungus. At first, two distinct subtypes among the molecular type VGII isolates responsible for the Canadian outbreak were characterised by multilocus sequence typing (MLST), the clinical and environmental common highly virulent VGIIa subtype, and the rarer, less virulent VGIIb subtype⁸. The early finding that both subtypes have only mating type alpha isolates that were fertile, led to the suggestion that same-sex mating between two alpha cells was the driving force for the emergence of the outbreak⁸. Within a decade, an additional novel VGII subtype, VGIIc, also with increased virulence, but not associated with the Vancouver Island outbreak, was identified in the Pacific Northwest, which emphasised the continuous emergence of new genotypes among C. gattii⁷.

MLST analysis⁹ of a larger number of globally collected clinical, veterinary and environmental isolates showed that the subtypes VGIIa and VGIIb are not only present in North America, but also in South America and Europe, with the subtype VGIIb especially being also found in Asia and Australia^2,10,11. Supported by the early association reported between C. gattii and Eucalyptus species^4,5 and the extensive introduction of these trees to different places in the world, such as California in the USA, the identification of the subtype VGIIb in Australia, suggested an Australian origin of the outbreak isolates⁸. However, the presence of VGIIa and VGIIb in South America and the close association of C. gattii with a number of tropical trees that has been reported in this region, indicated strongly that both genotypes originated from South America and have since then been dispersed and introduced on several occasions to other parts of the world, including North America¹⁰.

To shed light on the speculations that have arisen on the origins of the outbreak strains, the Molecular Mycology Research Laboratory at the University of Sydney is leading an ongoing research collaboration among different institutions in Australia, South and North America, Asia, Africa and Europe to carry out global epidemiological studies based on MLST typing⁹. When looking worldwide at the geographical distribution of the genetic diversity, shown by this approach, the highest number of sequence types has been detected so far in South America, while Australia harbors only few sequence types (Figure 1). High levels of genetic diversity are a strong indication of ancestral origins, hence, this extended genetic diversity seems likely to corroborate the idea that the Vancouver Island outbreak subtypes have originated from South America^2,10. In contrast, the low genetic diversity combined with the over-representation of some sequence types in Asia and Australia suggest that a clonal expansion has occurred and that the colonisation of such clones may lead to the development of local outbreaks, such as the one observed in a group of sheep in Western Australia⁴.

Figure 1. Minimum spanning tree of global Cryptococcus gattii VGII isolates representing the continent where the different sequence types (ST) have been identified. The tree was calculated by goeBURST analysis implement in Phyloviz software (http://www.phyloviz.net/wiki/) and which used the same priority rules for linking STs as eBURST but with a global optimisation. A total of 78 STs representing 529 isolates was included. Size of the circles corresponds to the number of isolates within each ST. Number of nucleotide polymorphisms are shown on each branch. Solid, grey and dashed branches represent at least one, two to three and more than 4 differences, respectively. The three subtypes VGIIa (ST20), VGIIb (ST7), and VGIIc (ST6) are highlighted in the picture while the remaining circles represents the other VGII lineages.

Taking advantage of the more comprehensive view obtained by investigating the whole genome, whole genome sequencing (WGS) was recently performed on 118 C. gattii VGII isolates from five continents, representing most of the MLST haplotypes previously identified^2,10–12. Even though the subtypes from the Pacific Northwest of the USA were completely clonal (Figure 2), as shown already by MLST (Figure 1), but with the much greater resolution obtained by WGS, they showed various genetic differences with the other VGII lineages, including mutations, deletions, transpositions, recombination events, and gene content differences which are potentially related to habitat adaptation, virulence, and pathology¹². Presence or absence of genes among the different isolates is currently being characterised to understand any functional role, by gene knockout studies carried out in our laboratory. WGS analysis showed again, that i) the highest genetic diversity within the VGII population is present in isolates from South America, ii) the major outbreak genotypes VGIIa and VGIIb are both present in South America, and iii) the Pacific Northwest genotype VGIIc is closely related to a South American isolate, which all support the evolutionary origin and dispersal of C. gattii from this part of the world (Figure 2)¹². These findings were independently confirmed in a parallel study investigating 53 different VGII isolates¹³.

Figure 2. Single nucleotide polymorphism (SNP) phylogeny of Cryptococcus gattii molecular type VGII isolates with geographic annotation. Maximum parsimony phylogenetic analysis was performed on SNPs from 118 C. gattii VGII genomes including Pacific Northwest and global diversity. 310,969 SNPs were identified (221,248 parsimony informative). CI = 0.401.

Our current data indicate that Australia is not the origin of the North American outbreaks but a major stepping-stone in the global spread of outbreak-related C. gattii genotypes. As seen with the outbreaks in North America, C. gattii will continue to expand its ecological niche, because there is still a constant and dynamic process driving its evolution. Thus, further characterisation of isolates from regions like Australia, Brazil and Colombia, where a relatively high incidence of cryptococcosis due to C. gattii occurs in some native animals and indigenous human populations^3,4,14,15, need to be undertaken in order to better understand the key processes for the emergence of C. gattii epidemics around the globe, especially in regions where the yeast is not thought to be endemic.