ABSTRACT
Four methods for the accurate delineation of epidemiologically related and unrelated strains of Candida lusitaniae were compared. Three pulsed-field electrophoretic methods, including two contour-clamped homogeneous field gel electrophoresis methods (EKP-1 and EKP-2) yielding electrophoretic karyotype patterns of intact chromosomal DNA and a method in which the chromosomal DNA was macrodigested with the endonuclease SfiI prior to pulsed-field electrophoresis (MDP), and a random amplified polymorphic DNA (RAPD) assay were evaluated. A selected panel of 21 well-characterized isolated representing 13 strains of C. lusitaniae, including 7 epidemiologically related isolates of one strain (group I-A), 3 epidemiologically related isolates of another strain (group I-B), and 11 epidemiologically unrelated isolates (group II), were tested. All isolates were coded and tested in a blinded manner. All seven group I-A isolates were confirmed to be a single strain by the EKP-1 and MDP methods, and the three group I-B isolates were shown to be a single strain by the EKP-1, EKP-2, MDP, and RAPD methods. Subtle differences were noted with two of the group I-A isolates by the EKP-2 method, whereas three of these isolates were different by the RAPD method. Each group II isolate had distinct patterns by all four methods. These data support the fact that the three pulsed-field electrophoretic methods and the RAPD method can be used to delineate strains of C. lusitaniae. The EKP-1, EKP-2, and MDP gave results that correlated with the epidemiologic characteristics of the isolates tested in the study, whereas the RAPD method was perhaps too sensitive in detecting DNA changes for epidemiologic studies.
