Construction and functional assignment of a manually annotated expressed sequence tag (EST) library from the pathogenic fungus Ophiostoma novo-ulmi




Pinchback, Michael

Journal Title

Journal ISSN

Volume Title



A genetic catalogue was generated from expressed sequence tags (ESTs) from the pathogenic filamentous fungus Ophiostoma novo-ulmi. Rather than full sequencing of the entire genome, fragments of each gene being actively expressed at a specific point in time were catalogued and annotated for identity and function. This catalogue represents a resource of considerable depth for the purposes of gene discovery, genetic regulation, protein expression, pathogenicity, and growth state studies. An online database was generated to serve as a powerful tool for downstream applications, facilitating and enhancing future research in all of these important areas of fungal biology. The ascomycetous fungus Ophiostoma novo-ulmi represents an excellent model organism for genetic experimentation. A diversity of physiological functions, including dimorphism, pathogenicity, melanin biosynthesis, and glycoprotein secretion at high levels mean that principles elucidated from this fungus are likely of broad application. Ophiostoma novo-ulmi has been identified as the causative agent of Dutch elm disease, which has become an economic and horticultural pandemic in North America. As a result, the mechanisms of host-pathogen interaction of this fungus are of particular interest. Initial attempts at disruption of pathogenicity, most commonly by disruption of single genes identified as potential pathogenicity factors, have met with little success. As our understanding of the complexity and co-ordination of proteins involved with host-pathogen interaction deepens, the discovery of a single dominant pathogenicity gene is becoming increasingly unlikely. As such, a broader genomics approach was employed to work towards identification of groups, or networks of genes that operate in a concerted manner, regulating pathogenicity or parasitic fitness. A low redundancy library was constructed from Ophiostoma nova-ulmi complementary DNA, producing a total of 4386 readable expressed sequence tags (ESTs) from 5760 clones. Of these, 2093 sequences matched with sequences found in public databases while 2293 represented orphan sequences. Of the sequences in the former group, 1761 sequences matched with known proteins while 332 sequences matched with hypothetical/predicted proteins. Sequences matching known proteins included 880 singletons, corresponding to 49.97 % of the ESTs in this category. Extrapolating this proportion to the sequences matching hypothetical proteins estimated the number of singletons in this category to be 166. Similarly, 1835 orphan sequences were estimated to contain 917 unique sequences. Singletons matching entries in public databases (n=880) were manually annotated into functional categories as established by the Munich Information Centre for Protein Sequences (MIPS). Metabolism (21%), Protein Synthesis (10%), Subcellular Localization (10%), Biogenesis of Cell Components (8%), and Transcription (8%) categories were the most highly represented.



Ophiostoma, Dutch elm disease