Use of Bulk Segregant Analysis for Determining the Genetic Basis of Azole Resistance in the Opportunistic Pathogen Aspergillus fumigatus

Abstract

A sexual cycle was described in 2009 for the opportunistic fungal pathogen Aspergillus fumigatus, opening up for the first time the possibility of using techniques reliant on sexual crossing for genetic analysis. The present study was undertaken to evaluate whether the technique 'bulk segregant analysis' (BSA), which involves detection of differences between pools of progeny varying in a particular trait, could be applied in conjunction with next-generation sequencing to investigate the underlying basis of monogenic traits in A. fumigatus. Resistance to the azole antifungal itraconazole was chosen as a model, with a dedicated bioinformatic pipeline developed to allow identification of SNPs that differed between the resistant progeny pool and resistant parent compared to the sensitive progeny pool and parent. A clinical isolate exhibiting monogenic resistance to itraconazole of unknown basis was crossed to a sensitive parent and F1 progeny used in BSA. In addition, the use of backcrossing and increasing the number in progeny pools was evaluated as ways to enhance the efficiency of BSA. Use of F1 pools of 40 progeny led to the identification of 123 candidate genes with SNPs distributed over several contigs when aligned to an A1163 reference genome. Successive rounds of backcrossing enhanced the ability to identify specific genes and a genomic region, with BSA of progeny (using 40 per pool) from a third backcross identifying 46 genes with SNPs, and BSA of progeny from a sixth backcross identifying 20 genes with SNPs in a single 292 kb region of the genome. The use of an increased number of 80 progeny per pool also increased the resolution of BSA, with 29 genes demonstrating SNPs between the different sensitive and resistant groupings detected using progeny from just the second backcross with the majority of variants located on the same 292 kb region. Further bioinformatic analysis of the 292 kb region identified the presence of a cyp51A gene variant resulting in a methionine to lysine (M220K) change in the CYP51A protein, which was concluded to be the causal basis of the observed resistance to itraconazole. The future use of BSA in genetic analysis of A. fumigatus is discussed.

Keywords: CYP51; antifungal; filamentous fungi; genomics; itraconazole; next-generation sequencing.

Figure 1

Itraconazole ETEST sensitivity testing of Aspergillus fumigatus strains 47-308 (A) and 47-51 (B) and progeny from back crossing. The ETEST has a concentration of 32 mg L^-1 itraconazole at the top of the strip which decreases down the strip towards the bottom. (A) Isolate 47-308 shows full growth over the range of itraconzole concentrations, indicating a resistance MIC of >32 mg L^-1. (B) Isolate 47-51 shows inhibition of growth at itraconazole concentrations above 0.75 mg L^-1 indicating a sensitive phenotype. (C) Representative ETEST results of progeny isolated from the third backcross between a resistant progeny isolate and the itraconazole sensitive parental isolate 47-51. Results demonstrate a 1:1 segregation of the resistant (top row) and sensitive (bottom row) phenotype.

Figure 2

Methodology of bulk segregant analysis used in the present study. An azole resistant isolate (47-308) carrying an unknown mutation was sexually crossed with a sensitive isolate (47-51). F1 progeny were collected and up to six rounds of backcrossing with the sensitive parent undertaken with arising resistant progeny. At certain stages the collected progeny were separated into azole resistant and sensitive pools, which were genome sequenced in order to identify genomic regions within the resistant progeny containing a candidate gene(s) of interest.

Figure 3

Distribution plots of SNP variants identified from bioinformatic analysis mapped onto A. fumigatus JGI reference A1163 genome contigs (not to scale) from bulk segregant analysis (BSA) applied following six rounds of backcrossing (BC6) using 40 progeny per BSA pool. Each red bar represents one or more (if in close proximity) variant sites exhibiting a consistent difference between the sensitive and resistant groupings. Data includes sites present in gene coding and non-coding regions. The red asterisk above contig DS_499598 indicates the position of the M220K causal variant in cyp51A.

Figure 4

Distribution plots of SNP variants identified from bioinformatic analysis mapped onto A. fumigatus JGI reference A1163 genome contigs (not to scale) from bulk segregant analysis (BSA) applied following two rounds of backcrossing (BC2) using 80 progeny per BSA pool. Each red bar represents one or more (if in close proximity) variant sites exhibiting a consistent difference between the sensitive and resistant groupings. Data includes sites present in gene coding and non-coding regions. The red asterisk above contig DS_499598 indicates the position of the M220K causal variant in cyp51A.

Figure 5

Variant ratio plots to show extent of heterozygosity or homozygosity at SNPs identified on contig DS_499598 from bioinformatic analysis. (A) Data analysis of resistant parent and progeny pool (n=40) derived after six back crosses (BC6). (B) Data analysis of resistant parent and progeny pool (n=80) derived after two back crosses (BC2). Values were calculated as the ratio of the average coverage supporting that variant compared to the average total coverage at that given position across the three replicates in the progeny pools. The red asterisk indicates the position of the M220K causal variant in cyp51A.