DNA Sequence-Based Identification of Poorly Identifiable Yeast Species

Abstract

Identification of Candida species has become more important because of an increase in infections caused by species other than Candida albicans, including species innately resistant to azole antifungal drugs. This study was done to compare the phenotypic identification of clinically important yeast species (using API 20 C AUX), with emphasis on poorly identifiable species, with rDNA sequence based identification. Clinical specimens from different sources including infected wounds, stool, urine, vaginal swabs and oral swabs were cultured directly. Suspected Yeast were identified by gram stain, germ tube test and API 20CAUX. The Internal Transcribed Spacers of the rDNA gene complex were amplified with fungal universal primers and partial sequences were obtained. Sequence data were used to identify the clinical isolates up to the species level using the CBS yeast database (http://www.cbs.knaw.nl/yeast/BioloMICSID.aspx). In addition to the DNA identification, the sugar assimilation results of the API system were used to obtain additional phenotypic identification using the polyphasic identification system of CBS yeast database. Twenty eight poorly identified isolates were sequenced using primer ITS4, and sequences data were then used to search for sequence homology in the CBS. Using API 20 C AUX system, 92.6% of all strains were well identified (i.e. with high level of identification) up-to-the species level and 7.4% of the isolates were poorly identified. DNA sequencing resulted in excellent identification (≥ 98% homology) with most of the tested isolated. Three yeast isolates had very low sequence homology (< 20%), which might represent novel new yeast species.‫‪The study concluded that rDNA sequencing for yeast identification appeared to be the useful tool for‬‬ ‫‪clinical microbiological laboratories for the identification of several yeast species which cannot be‬‬ ‫.‪identified by common phenotypic methods such as API 20C AUX‬‬