Population genomics analysis of US Fusarium graminearum isolates and identification of the genetic basis of fitness traits

Abstract

Fusarium graminearum is a filamentous ascomycete fungus that causes Fusarium head blight (FHB) disease. FHB is present throughout the world and is one of the major constraints to wheat and barley production. In North America, F. graminearum is the major Fusarium species causing FHB, although many other Fusarium species are known to cause the disease. In the United States, F. graminearum exists in populations correlated to trichothecene chemotypes such as DON, 3ADON, 15ADON, NX-2 and NIV. Even though F. graminearum is one of the most well-studied plant pathogens, large-scale population level studies in F. graminearum are limited. Population level studies help define the population structure and diversity of the pathogen in the field and detect the direction and amount of gene flow between the populations, which is crucial in making disease management decisions. Genomic patterns in isolates across populations also can be harnessed to study how the populations are evolving and the influence of natural selection in pathogen adaptation. Genotypic variation uncovered through population genomics studies in combination with phenotypic variation can be used to identify the genetic determinants of pathogen traits using quantitative genomics approaches. The broader goal of my research is to identify the genetic basis of pathogen fitness traits in F. graminearum. Growth rate, sexual fertility, aggressiveness, mycotoxin production, and fungicide sensitivity are traits of economic interest and the focus of my research projects. I attempted to identify the genetic basis of F. graminearum fitness through the indirect approach of scanning genomes of F. graminearum populations to detect the signs of recent positive natural selection. The assumption was that the regions under the influence of positive natural selection are likely to provide fitness advantage to the fungus. I identified 594 and 200 genomic regions in three F. graminearum populations under the influence of selection using two bioinformatics software packages. The regions include many genes with unknown function and some genes with roles in plant-microbe interaction, fungicide/drug resistance, mycotoxin production, sexual spore production, transportation and genes that code for components of membranes and cellular organelles. Other population genomic analyses conducted identified frequent recombination, a moderate level of gene flow and high evolutionary potential in F. graminearum populations from the U.S. I also used a more direct approach to detect the genetic basis of fitness traits in F. graminearum using genome wide association (GWAS) mapping. I identified a total of 13 significant quantitative trait nucleotides (QTNs) for growth rate, 30 for sexual fertility, 48 for propiconazole sensitivity, 37 for tebuconazole sensitivity, 8 for DON production in vitro, 3 for 15ADON production and 2 for aggressiveness in two wheat genotypes. Some of these QTNs fall on or are in close proximity to the genes with known roles for the respective trait. However, many of the QTNs are novel and provide new candidates for functional studies. Finally, I sequenced the genomes of four isolates of F. graminearum using Oxford Nanopore sequencing, assembled the genomes de novo and used the assembled genomes to identify structural rearrangements and infer their role in F. graminearum adaptation and fitness. I detected a total of 87 inversions, 159 translocations, 245 duplications, 58,489 insertions and 34,102 deletions. Regions with high recombination rates are associated with structural rearrangements, and a significant proportion of inversions, translocations, and duplications overlap with the repeat content of the genome. Structural rearrangements in F. graminearum play an important role in shaping pathogen-host interactions by introducing presence-absence polymorphisms in secondary metabolite clusters and predicted effector genes, shuffling genomic segments and changing protein products and gene regulation. Using GWAS and selection scans, I have identified candidate loci potentially determining fitness traits in F. graminearum. These loci provide new candidates for functional studies and, once verified, can be used as markers to monitor change in the genetic composition of field populations with respect to fitness-related traits such as fungicide sensitivity, aggressiveness and mycotoxin production.