Founder events influence structures of Aspergillus flavus populations

Abstract

In warm regions, agricultural fields are occupied by complex Aspergillus flavus communities composed of isolates in many vegetative compatibility groups (VCGs) with varying abilities to produce highly toxic, carcinogenic aflatoxins. Aflatoxin contamination is reduced with biocontrol products that enable atoxigenic isolates from atoxigenic VCGs to dominate the population. Shifts in VCG frequencies similar to those caused by the introduction of biocontrol isolates were detected in Sonora, Mexico, where biocontrol is not currently practiced. The shifts were attributed to founder events. Although VCGs reproduce clonally, significant diversity exists within VCGs. Simple sequence repeat (SSR) fingerprinting revealed that increased frequencies of VCG YV150 involved a single haplotype. This is consistent with a founder event. Additionally, great diversity was detected among 82 YV150 isolates collected over 20 years across Mexico and the United States. Thirty-six YV150 haplotypes were separated into two populations by Structure and SplitsTree analyses. Sixty-five percent of isolates had MAT1-1 and belonged to one population. The remaining had MAT1-2 and belonged to the second population. SSR alleles varied within populations, but recombination between populations was not detected despite co-occurrence at some locations. Results suggest that YV150 isolates with opposite mating-type have either strongly restrained or lost sexual reproduction among themselves.

Fig 1

Approximate collection locations of Aspergillus flavus VCG YV150 isolates used in the current study. Isolates were recovered from different substrates from 1987 to 2008. AZ, Arizona, USA; TX, Texas, USA; AK, Arkansas, USA; MS, Mississippi, USA; GA, Georgia, USA; SON, Sonora, Mexico; SIN, Sinaloa, Mexico; NAY, Nayarit, Mexico. No isolates with MAT1‐2 idiomorph were detected below the blue line. [Color figure can be viewed at wileyonlinelibrary.com]

Fig 2

The output of SplitsTree using the neighbour net and phi‐test approaches (Huson and Bryant, 2006). Out of 36 haplotypes, two a posteriori populations were determined: MAT1‐1 and MAT1‐2 populations. [Color figure can be viewed at wileyonlinelibrary.com]

Fig 3

The posterior probability [ln P(D)] averaged across 20 simulations for each K (data not shown) was used to calculate the optimal number of populations, delta K, using Structure Harvester (Earl and VonHoldt, 2012), following the Evanno method (Evanno et al., 2005).

Fig 4

Structure output based on 20 simulations for each K (K = 1–8, the eight a priori populations) for 36 clone‐corrected haplotypes of Aspergillus flavus VCG YV150 collected from 1987 to 2008 in the United States and Mexico. The graphic represents the output for one of the simulations for K = 2. Each vertical line represents a haplotype along the x‐axis. The proportion of membership (Q) in a genetic cluster is denoted by colour. Red = alleles associated with MAT1‐1, and green = alleles associated with MAT1‐2. Haplotypes for isolates containing MAT1‐1 and MAT1‐2 are distributed along the x‐axis. Haplotypes associated with one mating‐type were not detected in the other. [Color figure can be viewed at wileyonlinelibrary.com]

Fig 5

The output of SplitsTree comparing haplotypes of the two YV150 mating‐type idiomorphs with haplotypes of VCGs OD02, MR17, CRG136 (Grubisha and Cotty, 2010) and YV36 (Grubisha and Cotty, 2015) (including the haplotype of AF36, the active ingredient fungus of the aflatoxin biocontrol agent Aspergillus flavus AF36). [Color figure can be viewed at wileyonlinelibrary.com]