Real time RT-PCR quantification and Northern analysis of cerato-ulmin ( CU) gene transcription in different strains of the phytopathogens Ophiostoma ulmi and O. novo-ulmi.

Cerato-ulmin is a surface protein that belongs to the class of fungal proteins known as hydrophobins. This class II hydrophobin is produced throughout the life cycle and in all developmental stages of Ophiostoma novo-ulmi and O. ulmi; the aggressive and non-aggressive pathogens responsible for Dutch elm disease. Since yeast/mycelial transitions are often important to pathogenesis in dimorphic fungi such as Ophiostoma, we have examined the levels and abundance of cu mRNA in the yeast and mycelial stages of this fungus. The fungus contains one copy of the cu gene per haploid genome, located on chromosome IV. Our studies have been done using phosphoimager-based Northern analysis and real-time quantitative RT-PCR (qRT-PCR) to measure levels of cu mRNA. These measurements were made in both yeast-like and mycelial stages of the pathogen. Two wild-type, aggressive, strains of O. novo-ulmi (VA30 and H327) and one wild type non-aggressive strain of O. ulmi (H5) were analysed. As controls, we have utilized two types of mutants that we had previously generated, the null cu mutants THEK5-8 and THEK5-8-1, and a cu over-expression mutant, H5-tf16. Data generated by both Northern hybridization and real-time qRT-PCR analyses demonstrate that there is no cu mRNA transcription in the null mutants. The Northern analysis clearly showed that the over-expressing mutant H5-tf16 produces much more cu mRNA than the non-aggressive or aggressive strains. The quantitative data generated using qRT-PCR demonstrated that mycelium generally had 20-60% more cu mRNA than the yeast form. The non-aggressive strain of O. ulmi (H5) produces one-tenth as much cu mRNA as the aggressive strains (VA30 and H327). When transformed with additional copies of the cu gene, this same non-aggressive strain (H5-tf16) expressed about 20 times more cu mRNA than the wild type H5 strain. These data were consistently generated in multiple real-time qRT-PCR experiments with different RNA preparations, clearly demonstrating that the quantitative abundance values obtained were reproducible. Our study represents the first report on the use of real-time qRT-PCR to compare and quantify gene transcription in different growth phases of a fungal pathogen.