Genome structure and pathogenicity of the fungal wheat pathogen Mycosphaerella graminicola

The phytopathogenic fungus Mycosphaerella graminicola (Fuckel) J. Schröt. in Cohn (asexual stage: Zymoseptoria tritici (Desm.) Quaedvlieg & Crous) causes septoria tritici leaf blotch (STB) in wheat and is one of the most important diseases of this crop worldwide. However, STB control, mainly based on the use of resistant cultivars and fungicides, is significantly hampered by the limited understanding of the genetic and biochemical bases of pathogenicity, and mechanisms of infection and resistance in the host. M. graminicola has a very active sexual cycle under field conditions, which is an important driver of STB epidemics. Moreover, it results in high genetic diversity of field populations that causes a major challenge for the development and sustainable management of resistant cultivars and the discovery of new antifungal compounds. Understanding the role of the sexual and asexual life cycles on genome composition of this versatile pathogen and its infection strategy is crucial in order to develop novel control methods. Chapter 1 is an introduction to the biology and pathogenicity of M. graminicola. In addition, it shortly describes the impact of improved and novel technologies on the speed, scope and scale of comparative genomics research. Chapter 2 provides detailed genetic analyses of two M. graminicola mapping populations, using mainly DArT markers, and the analysis of the meiotic transmission of unequal chromosome numbers. Polymorphisms in chromosome length and number were frequently observed in progeny isolates, of which 15–20% lacked one or more chromosomes despite their presence in one or both parents, but these had no apparent effect on sexual and pathogenic fitness. M. graminicola has up to eight so called dispensable chromosomes that can be easily lost - collectively called the dispensome - which is, so far, the highest number of dispensable chromosomes reported in filamentous fungi. They represent small-sized chromosomes and make up 38% of the chromosome complement of this pathogen. Much of the observed genome plasticity is generated during meiosis and could explain the high adaptability of M. graminicola in the field. The generated linkage map was crucial for finishing the M. graminicola genome sequence. Chapter 3 describes the M. graminicola genome sequence with highlights on genome structure and organization including the eight dispensable chromosomes. The genome comprises a core set of 13 chromosomes and a dispensome, consisting of eight chromosomes that are distinct from the core chromosomes in structure, gene and repeat content. The dispensome contains a higher frequency of transposons and the genes have a different codon use. Most of the genes present one the dispensome are also present on the core chromosomes but little synteny is observed neither between the M. graminicola dispensome and the core chromosomes nor with the chromosomes of other related Dothideomycetes. The dispensome likely originates from ancient horizontal transfer(s) from (an) unknown donor(s). Chapter 4 shows a global analysis of proteins secreted by M. graminicola in apoplastic fluids during infection. It focuses mainly on fungal proteins secreted in a compatible interaction. The study showed that many of the annotated secreted proteins have putative functions in fungal pathogenicity, such as cell wall degrading enzymes and proteases, but the function of a substantial number of the identified proteins is unknown. During compatible interactions proteins are primarily secreted during the later stages. However, many pathogenesis-related host proteins, such as PR-2, PR-3 and PR-9, accumulated earlier and at higher concentrations during incompatible interactions, indicating that fungal effectors are recognized by resistant plants and trigger resistant gene-mediated defence responses, though without a visible hypersensitive response. Chapter 5 further details the initial identification and characterization of necrosis-inducing proteins that are produced in culture filtrates (CFs) of M. graminicola. The necrosis-inducing activity of CFs is light dependent and inactivated by proteinase K and heat treatment (100C). This is reminiscent of the necrosis-inducing properties of host selective toxins of other Dothideomycete pathogens such as Stagonospora nodorum and Pyrenophora tritici-repentis. Subsequent purifications of CFs and mass spectrometry identified several candidate proteins with necrosis-inducing activity. Heterologous expression of the two most prominent proteins in Pichia pastoris produced sufficient quantities for infiltration assays in a panel of wheat cultivars that showed differential responses, suggesting specific recognition. Chapter 6 provides a general discussion of the thesis and puts the results obtained in a broader perspective with a focus on the genome structure of M. graminicola and its function. In addition, aspects of the hemi-biotrophic lifestyle, the relevance of secreted proteins for the wheat-M. graminicola pathosystem in relation to gene-for-gene models and the potential implications for resistance breeding strategies are discussed.