What is the value of ITS sequence data in Colletotrichum systematics and species diagnosis? A case study using the falcate-spored graminicolous Colletotrichum group.

Because the genus Colletotrichum is among the most important groups of plant pathogenic fungi worldwide, the ability to accurately diagnose species is vital for the implementation of effective disease control and quarantine measures. Although the long-standing, unresolved taxonomic issues in the genus have recently begun to be addressed through multi-locus phylogenetic research, the tools most commonly used for Colletotrichum species identification are either insufficiently variable (e.g. morphology), or homoplasic (e.g. morphology and host range criteria). In this study, using the systematically well-defined falcate-spored, grass-associated group (FG) of Colletotrichum as a model, we test the utility of ITS sequence data to diagnose species affiliations through similarity-based searches of the NCBI GenBank database or by means of gene trees constructed using phylogenetic methods. 43% of all Colletotrichum sequences accessioned by GenBank are from the ITS region, making it the single most common sequence curated by the community; however, 34% of the ITS accessions existed only as sequence data in the database, with no associated publication. Using Colletotrichum ITS sequence data from 53 FG defined isolates and 16 falcate-spored, non-graminicolous isolates to perform GenBank BLASTN searches, we found that erroneous identifications occurred for 86% of the 14 species tested. In contrast, the phylogenetic tree generated by the ITS sequence data, although poorly supported by bootstrap values, correctly grouped most of the species, but 10% of the individual isolates were incorrectly placed. From this study, we conclude that the currently available infrastructure of Colletotrichum ITS sequence data may yield unreliable species diagnoses, particularly if sequence similarity alone is the only criterion applied.