Detecting Introgression Between Members of the Fusarium fujikuroi and F. oxysporum Species Complexes by Comparative Mitogenomics

The Fusarium fujikuroi species complex (FFSC) and F. oxysporum species complex (FOSC) are two related groups of plant pathogens causing a wide diversity of diseases in agricultural crops world wide. The aims of this study are (1) to clarify the phylogeny of the FFSC, (2) to identify potential deviation from tree-like evolution, (3) to explore the value of using mitogenomes for these kinds of analyses, and (4) to better understand mitogenome evolution. In total, we have sequenced 24 species from the FFSC and a representative set of recently analyzed FOSC strains was chosen, while F. redolens was used as outgroup for the two species complexes. A species tree was constructed based on the concatenated alignment of seven nuclear genes and the mitogenome, which was contrasted to individual gene trees to identify potential conflicts. These comparisons indicated conflicts especially within the previously described African clade of the FFSC. Furthermore, the analysis of the mitogenomes revealed the presence of a variant of the large variable (LV) region in FFSC which was previously only reported for FOSC. The distribution of this variant and the results of sequence comparisons indicate horizontal genetic transfer between members of the two species complexes, most probably through introgression. In addition, a duplication of atp9 was found inside an intron of cob, which suggests that even highly conserved mitochondrial genes can have paralogs. Paralogization in turn may lead to inaccurate single gene phylogenies. In conclusion, mitochondrial genomes provide a robust basis for phylogeny. Comparative phylogenetic analysis indicated that gene flow among and between members of FFSC and FOSC has played an important role in the evolutionary history of these two groups. Since mitogenomes show greater levels of conservation and synteny than nuclear regions, they are more likely to be compatible for recombination than nuclear regions. Therefore, mitogenomes can be used as indicators t