C-222. Comparison of the Aspergillus DiversiLab rep-PCR Assay to a Sequenced-Based Method Using Two Databases for the Identification of Aspergillus species

K. Eskey1, L. Hall2, T. Ross1, J. Gorthy1, W. G. Merz1;
1Johns Hopkins Med. Inst., Baltimore, MD, 2Mayo Clin., Rochester, MN.

Rapid, accurate identification of Aspergillus species is important for antifungal decisions and for epidemiologic investigations. The study was to compare the identification of Aspergillus species by two molecular methods, rep-PCR [DiversiLab(DVL)] to a sequence-based method. The gold standard was the identification of 46 type strains, ATCC strains or strains kindly provided by Marion Klich. For the rep-PCR, DNA was extracted using the ZR Fungal/Bacterial DNA Kit (Zymo Research), PCR was performed with the Aspergillus Kit, electrophoretic separation was performed using the Agilent 2100 Bioanalyzer (Agilent Technologies) and data analysis was conducted with the DiversiLab software. For the sequencing protocol, 1 ul loopful of fungus was lysed using PrepMan Ultra and sequenced with MicroSeq D2 large subunit rDNA fungal sequencing kit on a 3100 Genetic Analyzer (Applied Biosystems). Sequences were compared to Mayo Clinic Custom library (MCCL) and the Applied Biosystems D2 fungal library (ABD2L). For rep-PCR, 12/46 isolates were correctly identified including 3/3 A. fumigatus, 8/8 A. terreus, and 1/1 A. niger. Thirty-four isolates were identified incorrectly. The correct identifications were with species in the library whereas 32/34 incorrect identifications did not have species in the library, and 2/2 A. flavus were incorrectly identified despite being in the library. With sequencing, 41/46 were correct including 8/8 A. terreus, 2/2 A. flavus, 8/9 A. glaucus, 8/8 A. versicolor, 3/3 A. fumigatus, 1/1 A. niger, 6/6 A. nidulans, 1/1 A. pseudofischerii, 1/1 A. ochraceous, 1/1 A. fischerii and 2 were called A spp. using data from MCCL and ABD2L . The 5 incorrect were 1 A. glaucus, 1 A. carneus, 2 A. clavatus and 1 A. uniguis. With the rep-PCR, correct identification was dependant on the species representation in the library. Addition of more species to the library is required before the potential for this method can be assessed. The sequence-based method with the current databases accurately identified 89% of the Aspergillus spp. tested in this study, and the assay can be incorporated into algorithm(s) for the identification of medically-important fungi.