Comparative genomics of emerging pathogens in the Candida glabrata clade

Abstract

Background: Candida glabrata follows C. albicans as the second or third most prevalent cause of candidemia worldwide. These two pathogenic yeasts are distantly related, C. glabrata being part of the Nakaseomyces, a group more closely related to Saccharomyces cerevisiae. Although C. glabrata was thought to be the only pathogenic Nakaseomyces, two new pathogens have recently been described within this group: C. nivariensis and C. bracarensis. To gain insight into the genomic changes underlying the emergence of virulence, we sequenced the genomes of these two, and three other non-pathogenic Nakaseomyces, and compared them to other sequenced yeasts.

Results: Our results indicate that the two new pathogens are more closely related to the non-pathogenic N. delphensis than to C. glabrata. We uncover duplications and accelerated evolution that specifically affected genes in the lineage preceding the group containing N. delphensis and the three pathogens, which may provide clues to the higher propensity of this group to infect humans. Finally, the number of Epa-like adhesins is specifically enriched in the pathogens, particularly in C. glabrata.

Conclusions: Remarkably, some features thought to be the result of adaptation of C. glabrata to a pathogenic lifestyle, are present throughout the Nakaseomyces, indicating these are rather ancient adaptations to other environments. Phylogeny suggests that human pathogenesis evolved several times, independently within the clade. The expansion of the EPA gene family in pathogens establishes an evolutionary link between adhesion and virulence phenotypes. Our analyses thus shed light onto the relationships between virulence and the recent genomic changes that occurred within the Nakaseomyces.

Sequence accession numbers: Nakaseomyces delphensis: CAPT01000001 to CAPT01000179Candida bracarensis: CAPU01000001 to CAPU01000251Candida nivariensis: CAPV01000001 to CAPV01000123Candida castellii: CAPW01000001 to CAPW01000101Nakaseomyces bacillisporus: CAPX01000001 to CAPX01000186.

Figure 1

Maximum likelihood species tree of 22 Saccharomycotina species. The tree was reconstructed based on the analysis of a concatenated alignment of one-to-one orthologs of 603 widespread genes. Species names in red and with an asterisk indicate human fungal pathogens. Underlined species names correspond to the newly sequenced Nakaseomyces species. Important evolutionary events such as the Whole Genome Duplication (WGD) or the genetic code transition in the Candida clade (CTG) are marked on the tree. All aLRT-based supports were maximal and a single node with a bootstrap support below 100% is indicated. This topology is also the most parsimonious in terms of inferred duplications in 4,965 individual gene phylogenies, as assessed by a Gene Tree Parsimony approach implemented in duptree [20].

Figure 2

MAT-like cassettes in the sequenced genomes.HMR-, HML- and MAT-like are represented as rectangles, with red ones containing a-type information and blue ones containing alpha-type information. MAT cassettes of each type are shown for N. bacillisporus, which is diploid, and bicolored MAT cassettes in haploid species indicate possible switching in culture. Approximate coordinates on corresponding scaffold are shown below each cassette.

Figure 3

Summary of the main findings. The phylogenetic tree represents the evolution of the Nakaseomyces species, using S. cerevisiae as outgroup. Blue numbers indicate the number of genes that are predicted to have been gained at each lineage during the evolution of the Nakaseomyces. Red numbers indicate the yeast genes that have been lost. Species names coloured in red indicate the human pathogens, lighter colouring indicates recently-reported emerging pathogens. Coloured branches can be matched to the corresponding coloured boxes below, which list important events occurring at that lineage in the evolutionary history of Nakaseomyces. (Abbreviations: Ploidy: H: Haplobiontic, D: Diplobiontic; Sexual reproduction: H: homothallic, N.O.: not observed).

Figure 4

Phylogenetic profiles of specific gene families and pathways. The phylogenetic tree represents the evolution of the Nakaseomyces species, with the pathogens colored. White boxes indicate absence of a particular family or pathway in a given species, while numbers in colored boxes indicate the number of paraloguous copies of that gene family or the number of components of a given pathway. Intensity of the colors is proportional to the number of paralogs present.

Figure 5

Trees of the three groups of adhesins containing more than one member. Phylogenetic trees were reconstructed using the same approach as for the trees present in the phylome, and are visualized using ETE [50], A) proteins grouped with the known C. glabrata EPA genes, B) proteins similar to the yeast Flo protein, C) group of unknown adhesins, and D) sequences related to AWP adhesins. Labels indicate protein names and are colored depending on the species they belong to: C. glabrata in dark blue, a second strain of C. glabrata (BG2) in light blue, C. bracarensis in red, C. nivariensis in green, N. delphensis in purple, N. bacillisporus in yellow and C. castellii in orange. Logos at the lower right part of the image represent the conservation in each of the adhesin groups of sites that are considered important in the structure of EPA1[51].

Figure 6

Tandem repeats of eight genes identified in C. glabrata and their homologs in the ‘glabrata group’. A: MNT3 homologs in C. glabrata are present in a tandem repeat of eight genes and in a tandem of two genes, with additional isolated copies. Closest species exhibit a tandem repeat of two genes with mixed synteny at the borders. B: YPS homologs in C. glabrata are present in a tandem repeat of eight genes, with additional isolated copies. Closest species exhibit a tandem repeat of two genes with conserved synteny at the borders.