An official website of the United States government.
<P> Mycotoxins, and especially the aflatoxins produced by Aspergillus flavus, are an enormous problem in agriculture, with aflatoxin B1 being the most carcinogenic known natural compound. Direct losses due to aflatoxins are estimated at $240 million annually in the US alone (Miller et al. 2014). Recent efforts to reduce aflatoxin concentrations in seeds and grains have focused on the use of nonaflatoxigenic A. flavus strains as biological control agents (Abbas et al. 2005; Miller et al. 2014). Although reported as an effective method for aflatoxin control, the extent to which these biological control strains recombine with native strains and the overall effect on microbial community spatiotemporal dynamics is unknown. Our overarching goal is to link the population processes of genetic divergence, adaptation and drift to specific biotic and abiotic components of agricultural ecosystems. Our objectives are (1) to characterize the genetic, phenotypic and genomic plasticity observed in experimentally evolved and natural populations of Aspergillus, to better understand fungal adaptation, (2) to identify significant genotype-by-environment and genotype-phenotype associations that are important in the production of mycotoxins and other secondary metabolites, and (3) to understand the genetic basis of fertility in Aspergillus species. </P>
<P>NON-TECHNICAL SUMMARY: Mycotoxins, and especially the aflatoxins produced by Aspergillus flavus, are an enormous problem in agriculture. Direct losses due to aflatoxins are estimated at $240 million annually in the United States alone. Recent efforts to reduce aflatoxin concentrations in seeds and grains have focused on the use of nonaflatoxigenic A. flavus strains as biological control agents. The goal of this project is to examine the underlying genetic and genomic processes that result in population shifts favoring nonaflatoxigenic fungi. Specifically, we seek to integrate molecular genetic and phenotypic data across experimental and natural population samples and characterize the interactions among biotic and abiotic components of agricultural ecosystems. This will lead to a better understanding of fungal adaptation, the genetic basis of fertility in Aspergillus, and the impact that genome plasticity has in response to changing environmental and ecological pressures. This knowledge will be important in developing more cost-effective and sustainable biocontrol strategies. </P>
<P>APPROACH: Microbial populations continually adjust in response to changing environmental and ecological pressures (Bock et al. 2004; Cotty & Jaime-Garcia 2007). The importance of clonal and sexual reproduction in adaptation and diversification in Aspergillus is unknown (Horn 2005; Horn et al. 2013b). The integration of experimental and natural population sampling will provide the resolution necessary to understand the maintenance of important functional traits such as nonaflatoxigencity and sterility in populations (Olarte et al. 2012). Female sterility occurs naturally and at a high frequency in populations of these fungi worldwide (Moore et al. 2013), but even rare sex can influence aflatoxigenicity (Horn et al. 2013a; Horn et al. 2009a; Horn et al. 2011; Horn et al. 2009b; Horn et al. 2009c). The proposed project will benefit from our existing culture collection of over 5,000 Aspergillus spp. isolates that were collected across five continents. We will use high throughput genotyping methods to provide dense marker coverage across all A. flavus chromosomes and populations. Soil composition and cropping practices are reported to influence A. flavus population shifts (Zablotowicz et al. 2007). Spectroscopic and microscopic techniques will be used to examine soil composition and chemistry. Molecular sequence variation, phenotypic and environmental data will be analyzed to establish correlations between abiotic and biotic components of soil and crop ecosystems, as a basis for modeling and making predictions on how changing specific components of ecosystems can influence aflatoxigenic fungi (Wu et al. 2011). Our sampling over consecutive years will allow us to estimate genotype turnover rates and model population shifts. We will also explore new statistical methods to establish correlations between abiotic and biotic components of soil and crop ecosystems. Our inferences of significant genotype-by-environment or genotype-by-genotype interactions that may be important in population shifts will be the basis for further experiments in the laboratory that are essential for model validation and refinement. Knowledge of the abiotic and biotic components of agricultural ecosystems will be beneficial in improving biocontrol strategies. These strategies currently utilize nontoxigenic A. flavus strains but it is possible that other species may be more suitable. Biocontrol artificially shifts mating-type distributions in populations in favor of MAT1-2; both approved biocontrol strains are MAT1-2 (Olarte et al. 2012). Previous studies of mating-type distributions reveal that A. flavus populations that favor asexual reproduction have a higher percentage of nontoxigenic strains (Moore et al. 2013; Moore et al. 2009). While mating-type distributions are a useful indicator of the potential for toxin production, our ability to predict future population shifts is limited, as changes in agronomic practices (Jaime-Garcia & Cotty 2006), insect damage (Dowd 1998; Lussenhop & Wicklow 1990; McMillian et al. 1985; Windham et al. 1999) and environmental parameters such as temperature and precipitation (Cotty & Jaime-Garcia 2007) will directly impact fungal populations. However, the results of this project will identify the biotic and abiotic components of agricultural ecosystems that are responsible for population shifts. The model we develop can be further tested by conducting laboratory as well as field tests where we can control soil composition and host and allow only abiotic conditions to impact populations. We expect drought associated plant stress to have the largest impact on fungal population density but our knowledge of population structure when conditions are favorable for crop production, is currently limited. Computational Methods: We will continue with the development, maintenance, testing and evaluation of our online SNAP toolkit (Aylor et al. 2006; Monacell & Carbone 2014; Price & Carbone 2005). The Mobyle SNAP Workbench web-portal allows researchers to seamlessly manage and execute complex command-line programs with multiple input files and parameters on a high-performance Linux cluster; these tools are parameter-rich or memory-intensive, often requiring several days or weeks to run (Monacell & Carbone 2014). Our optimization includes selecting appropriate number of compute nodes and machine architecture for MPI programs and when possible, parallelization of computational tasks to execute on multiple machines. We will apply best practices in our implementation of workflows that link together several programs sequentially. This automation will allow data from sequencing, taxonomy and systematic studies to be integrated in a way that brings out significant patterns and correlations that exist across different data sets. This will greatly facilitate exploratory population genetic and genomic analyses for the novice user but will also allow for efficient use of computational resources. </P>