Rapid identification of Candida albicans and other human pathogenic yeasts by using short oligonucleotides in a PCR

Abstract

A PCR system that can quickly and accurately identify 14 species of human pathogenic yeasts was developed. The procedure distinguished between nine species of a closely related clade, Lodderomyces elongisporus, Candida parapsilosis, a new Candida sp., C. sojae, C. tropicalis, C. maltosa, C. viswanathii, C. albicans, and C. dubliniensis and between another five more divergent species, Pichia guilliermondii, C. glabrata, C. zeylanoides, C. haemulonii, and C. haemulonii type II. A rapid DNA extraction procedure that yields purified DNA in about 1 h is also described. The system uses uniform conditions with four primers for each reaction, two 40- to 50-mer universal primers that serve as a positive control and two 23- to 30-mer species-specific primers. Species-specific primers were derived from a 600-nucleotide variable region (D1/D2) at the 5' end of the large-subunit (26S) ribosomal DNA gene and were generally designed to use mismatches at the 3' end. Universal primers were developed from conserved nucleotide sequences in the small-subunit (18S) rRNA gene. In this system, a control 1,200- to 1,300-base DNA fragment was produced in all reactions and a smaller 114- to 336-base DNA fragment was produced if the chromosomal DNA from the target species was present. The PCR procedure is rapid and easy to interpret and may be used with mixed cultures.

FIG. 1

Phylogenetic tree of the Candida species studied and their nearest neighbors as represented by one of two most parsimonious trees derived from maximum-parsimony analysis. The phylogram was calculated from the divergence in large-subunit region D1/D2. Branch lengths are proportional to nucleotide differences as indicated on the marker bar. Numbers given on branches are the frequencies (expressed as percentages) with which a branch appeared in 100 bootstrap replicates. Frequencies under 50% are not shown. Tree length = 705; consistency index = 0.613; retention index = 0.749; rescaled consistency index = 0.459; homoplasy index = 0.387. Schizosaccharomyces pombe served as the outgroup species in the analysis.

FIG. 2

Agarose gel electrophoresis of PCR products from 14 species of yeasts by using the four-primer system. All reaction mixtures for panels A and B contain universal primers NS395F and NS1654R and C. albicans species-specific primers CAL5 and NL4CAL as well as chromosomal DNAs from the following yeast strains: L. elongisporus YB-4239 (lane 2), C. parapsilosis Y-12969 (lane 3), Candida sp. Y-17456 (lane 4), C. sojae Y-17909 (lane 5), C. tropicalis Y-12968 (lane 6), C. albicans Y-12983 (lane 7), C. dubliniensis Y-17841 (lane 8), C. maltosa Y-17677 (lane 10), C. viswanathii Y-6660 (lane 11), P. guilliermondii Y-2075 (lane 12), C. zeylanoides Y-1774 (lane 13), C. haemulonii Y-6693 (lane 14), C. haemulonii type II Y-17801 (lane 15), C. glabrata Y-65 (lane 16). PCR products were visualized on 1.0% agarose gels. Lanes 1 and 9 contain ΦΧ174 replicative-form DNA cut with HaeIII. Molecular weight markers (in base pairs) are indicated.

FIG. 3

Agarose gel electrophoresis showing PCR products from six strains of C. albicans and six strains of C. dubliniensis. Reactions and conditions in panels A and B are similar to those described in the legend for Fig. 2 with the exception that chromosomal DNAs from the following strains were used: C. albicans Y-12983 (lane 2), C. albicans Y-79 (lane 3), C. albicans YB-3898 (lane 4), C. albicans Y-17967 (lane 5), C. albicans Y-17968 (lane 6), C. albicans Y-17974 (lane 7), C. albicans Y-12983 (lane 9), C. dubliniensis Y-17841 (lane 10), C. dubliniensis Y-17512 (lane 11), C. dubliniensis Y-17969 (lane 12), C. dubliniensis Y-17971 (lane 13), C. dubliniensis Y-17972 (lane 14), C. dubliniensis Y-17973 (lane 15). Lanes 1 and 8 contain ΦΧ174 replicative-form DNA cut with HaeIII. Molecular weight markers (in base pairs) are indicated.