The role of repellents and hydrophobins in Ustilago maydis

Abstract

Ustilago maydis is an important model organism to study fungal pathogenicity. U. maydis can grow yeast-like and filamentous. In the latter form this fungus infects maize. In my Thesis the expression and function of hydrophobins and repellents of U. maydis were studied. Hydrophobins are produced by fungi where they fulfill a wide spectrum of functions including attachment of the fungus to a hydrophobic surface and the formation of aerial hyphae. Repellents 1-10 and Rep1-c were isolated as peptides that are involved in the formation of aerial hyphae in U. maydis. This raised the question whether repellents have (partially) replaced hydrophobins in U. maydis or function indirectly by anchoring hydrophobins to the cell wall. Using a new in situ hybridization protocol based on PNA probes it was shown that the repellent gene was expressed in filaments but not in the yeast form. In the filaments, the mRNA of this gene locates at the tip. This is also the site where the protein is expected to be secreted in the cell wall. The genome of U. maydis contains the hydrophobin genes hum2 and hum3. Deletion of hum2 only slightly reduced formation of aerial hyphae and surface hydrophobicity, while inactivation of hum3 in the wild-type or in the ?hum2 strain had no effect. In contrast, inactivation of rep1 dramatically affected surface hydrophobicity and aerial hyphae formation. Deletion of the hydrophobin genes had no effect on attachment of hyphae to a hydrophobic substrate but attachment was reduced by 50% in a cross of rep1 deletion strains. Additional deletion of either or both the hydrophobin genes did not further reduce attachment. From these data it was concluded that hydrophobins have been functionally replaced in U. maydis by repellents. Thus, repellents do not function by anchoring hydrophobins to the cell wall. Hydrophobins self-assemble into amphipathic amyloid fibrils at hydrophobic-hydrophilic interfaces. Amyloids represent an ordered fibrillar structure of proteins in a stacked b-sheet conformation. I have shown that repellents form a similar structure in the cell wall of filaments of U. maydis. This explains why repellents have hydrophobin-like properties and functions. A strain in which the repellent gene was inactivated has a slightly different expression profile when compared to the wild-type. Only 31 genes had a 2-fold change in expression. From these results it is concluded that aerial hyphae of U. maydis do not result from a differentiation process. Rather, they may represent vegetative hyphae that happen to grow in the air. Of the 31 genes with a changed gene expression in the strain in which the repellent gene was inactivated, 22 genes were up-regulated. Of these, 11 belong to the class of small secreted proteins (SSP’s). Prediction programs indicate that these proteins have the tendency to form amyloid fibrils. It may thus be that the cell wall of U. maydis contains a variety of proteins with this structure. In the case of U. maydis this is a functional fold, whereas in humans amyloids have been associated with diseases like Huntington’s, Alzheimer’s and Creutzfeldt-Jacob’s.