Towards understanding host specificity in the smut fungus Sporisorium reilianum - probable role of a diversity cluster

The two host-adapted varieties of the smut fungus Sporisorium reilianum (S. reilianum f. sp. zeae, SRZ, and S. reilianum f. sp. reilianum, SRS) produce spores either on maize or on sorghum. For plant infection, mating compatible haploid sporidia need to fuse and form infectious dikaryotic filaments that infect at seedling stage and form spore in inflorescences of the plant. Despite both formae speciales being very closely related on the genomic level why they differ in terms of host selection remains elusive. To know the host determining factors we generated sexual spores by crossing the compatible SRZ strain SRZ1_5-2 (mating type a1b1) with the SRS strain SRS2_H2-7 (mating type a2b6). Meiotic progeny (SRSZ) with the mating type a1b1 were selected and tested for virulence on sorghum after mating with SRS_H2-7. Virulence assays showed that the progeny were either non-virulent (107 offspring) or showed various degrees of disease phenotype.We selected 107 non-virulent, 40 full-virulent and 41 offspring with intermediate virulence for genotype analysis. 190 strains (188 offspring and 2 parental strains) were sequenced using Illumina technology. Mapping of the reads against the two assembled parental genomes showed that parental origin of chromosomal regions could be unambiguously assigned for all SRSZ strains. Interestingly, many strains contained partially duplicated genomic regions, i.e. carrying the same chromosomal fragments from both parents. Correlation of phenotype with parental origin of genomic loci revealed a region of 51 genes in the left arm of chromosome 7 potentially associated with the virulence phenotype on sorghum. Analysis of this region revealed a cluster of 9 genes that were more highly expressed in planta, showed the least amount of sequence conservation and coded for proteins carrying predicted secretion signal peptides. Deletion of the gene cluster in SRS strains is under way to show whether this gene cluster is necessary for full virulence on sorghum.