Multilocus sequence typing of the pathogenic fungus Aspergillus fumigatus

Abstract

A multilocus sequence typing (MLST) scheme was devised for Aspergillus fumigatus. The system involved sequencing seven gene fragments and was applied to a panel of 100 isolates of A. fumigatus from diverse sources. Thirty different sequence types were found among the 100 isolates, and 93% of the isolates differed from the other isolates by only one allele sequence, forming a single clonal cluster as indicated by the eBURST algorithm. The discriminatory power of the MLST method was only 0.93. These results strongly indicate that A. fumigatus is a species of a relatively recent origin, with low levels of sequence dissimilarity. Typing methods based on variable numbers of tandem repeats offer higher levels of strain discrimination. Mating type data for the 100 isolates showed that 71 isolates were type MAT1-2 and 29 isolates were MAT1-1.

FIG. 1.

The MAT locus in A. fumigatus. Both sexual idiomorphs are shown aligned together as described by Paoletti et al. (37). The upper sequence has only the 3′ portion of the MAT1-2 ORF and the complete MAT1-1 ORF. The lower sequence has the complete MAT1-2 ORF. Dashed lines indicate areas of sequence homology between both idiomorphs. Primers AFM1, AFM2, and AFM3, used for the multiplex PCR to distinguish the sexual idiomorph genotypes, are indicated. The black box indicates the location of MLST fragment MAT1-2.

FIG. 2.

Clonal cluster and five single STs generated by eBURST analysis of MLST data, representing all 30 STs found from 100 A. fumigatus isolates. The putative founder of the complex is ST 20. The diameter of each circle represents the number of members within each ST group. Connecting lines are of arbitrary lengths but indicate that STs linked by a single line are different only at one allele out of seven. The boxed letters indicate genotypes assigned in a previous study (39).

FIG. 3.

UPGMA dendrogram for 100 A. fumigatus isolates sequenced for MLST. The large cluster, which contained 94% of the isolates, was subdivided into two groups (indicated at the right) based on the high bootstrap values at the cluster nodes (numbers shown on the dendrogram).