Context: The incidence of infections due to diverse Candida species is increasing, with correspondingly different antifungal susceptibility patterns. Routine yeast identification methods cause significant delays in appropriate patient management.
Objective: A DNA pyrosequencing strategy was evaluated for identification of pathogenic Candida species associated with human infections.
Design: Clinical (n = 51) and commercial (n = 9) Candida isolates were identified in a blinded, parallel study consisting of routine fungal cultures and biochemical analyses in comparison with DNA pyrosequencing.
Results: DNA pyrosequencing yielded species-level identification of all 60 Candida isolates, and sequencing interpretations agreed in all cases with results of biochemical and morphologic testing. Different Candida species were identified, such as C. albicans, C. dubliniensis, C. glabrata, C. guilliermondii, C. krusei, C. lusitaniae, C. parapsilosis, and C. tropicalis. Automated and manual approaches to DNA sequence interpretation, each coupled with the Identifire identification software, demonstrated 100% agreement with respect to Candida species identification. Twenty-one isolates yielded intraspecies DNA sequence differences (90%-98% nucleic acid sequence identity) by automated interpretation. Sequence differences resulted from single-nucleotide polymorphisms or single-base additions/deletions, in addition to interpretative challenges in homopolymeric tracts.
Conclusion: DNA pyrosequencing coupled with automated DNA sequence alignment provides a practical approach for accurate and timely identification of Candida pathogens. Relatively rapid and facile genotypic studies by DNA pyrosequencing matched the effectiveness of extensive biochemical/morphologic studies for yeast identification.