Secondary metabolism of the forest pathogen Dothistroma septosporum : a thesis presented in the partial fulfilment of the requirements for the degree of Doctor of Philosophy (PhD) in Genetics at Massey University, Manawatu, New Zealand

Abstract

Dothistroma septosporum is a fungus causing the disease Dothistroma needle blight (DNB) on more than 80 pine species in 76 countries, and causes serious economic losses. A secondary metabolite (SM) dothistromin, produced by D. septosporum, is a virulence factor required for full disease expression but is not needed for the initial formation of disease lesions. Unlike the majority of fungal SMs whose biosynthetic enzyme genes are arranged in a gene cluster, dothistromin genes are dispersed in a fragmented arrangement. Therefore, it was of interest whether D. septosporum has other SMs that are required in the disease process, as well as having SM genes that are clustered as in other fungi. Genome sequencing of D. septosporum revealed that D. septosporum has 11 SM core genes, which is fewer than in closely related species. In this project, gene cluster analyses around the SM core genes were done to assess if there are intact or other fragmented gene clusters. In addition, one of the core SM genes, DsNps3, that was highly expressed at an early stage of plant infection, was knocked out and the phenotype of this mutant was analysed. Then, evolutionary selection pressures on the SM core genes were analysed using the SM core gene sequences across 19 D. septosporum strains from around the world. Finally, phylogenetic analyses on some of the SM core genes were done to find out if these genes have functionally characterised orthologs. Analysis of the ten D. septosporum SM core genes studied in this project showed that two of them were pseudogenes, and five others had very low expression levels in planta. Three of the SM core genes showed high expression levels in planta. These three genes, DsPks1, DsPks2 and DsNps3, were key genes of interest in this project. But despite the different expression levels, evolutionary selection pressure analyses showed that all of the SM core genes apart from the pseudogenes are under negative selection, suggesting that D. septosporum might actively use most of its SMs under certain conditions. In silico predictions based on the amino acid sequences of the proteins encoded by SM core genes and gene cluster analyses showed that four of the SM core genes are predicted to produce known metabolites. These are melanin (DsPks1), cyclosporin (DsNps1), ferricrocin (DsNps2) and cyclopiazonic acid (DsHps1). Gene cluster analyses revealed that at least three of the D. septosporum SMs might be produced by fragmented gene clusters (DsPks1, DsNps1, DsNps2). This suggested that dothistromin might not be the only fragmented SM gene cluster in D. septosporum. According to phylogenetic analyses, some of the D. septosporum SM core genes have no orthologs among its class (Dothideomycetes), suggesting some of the D. septosporum SMs may be unique. One such example is the metabolite produced by DsNps3. Comparison of wild type and ΔDsNps3 D. septosporum strains showed that the ΔDsNps3 strain produces fewer spores, less hyphal surface network at an early stage of plant infection, and lower levels of fungal biomass in disease lesions compared to wild type, suggesting that the DsNps3 SM may be a virulence factor. Attempts to identify a metabolite associated with DsNps3, and to knockout another gene of key interest, DsPks2, for functional characterization were unsuccessful. Further work is required to confirm the gene clusters, characterise the SMs and their roles. However, the findings so far suggest that dothistromin is unlikely to be the only D. septosporum SM that is a virulence factor in since the DsNps3 SM also appears to be involved in virulence. Likewise the fragmented dothistromin cluster may not be the only one in the genome and there may be at least three more fragmented SM gene clusters.