For quality control purposes, the following Candida type strains were used in the identification procedures of Candida isolates: C. albicans CECT 1472, C. dubliniensis, strain provided by Biognostica from United Kingdom National External Quality Assessment Service, C. glabrata ATCC 2001, C. guilliermondii ATCC 6260, Candida kefyr ATCC 204093, C. krusei ATCC 6258, C. parapsilosis ATCC 22019, and C. tropicalis ATCC 750. These strains were maintained on Sabouraud dextrose agar.

Samples were collected from a total of 97 asymptomatic individuals (77 females and 20 males) attending a dental clinic in Braga, Portugal, over a 12-month period (May 2005–2006). Each of the potential subjects was informed of the aims and methods of the study, and anticipated benefits and potential risks, according to the World Medical Association Declaration of Helsinki. Data on patient age, oral hygiene habits, health status, medications, and prosthesis wearing were collected.

Samples were collected by passing a sterile swab (UNI-TER, MEUS, Padua, Italy) across the oral mucosa, tongue, hard palate, and gums, and replacing it in its sterile container tube. Samples were kept at 4°C and analysed within 24h.

CHROMagar Candida medium (CHROMagar, Paris, France) was prepared according to the manufacturer's instructions. The swab was inoculated into CHROMagar Candida medium rotating the swab head on the surface of the medium. The plates were incubated at 37°C for 48h. Colony morphology and colour description were assigned in a standard manner by a single investigator. Presumptive species identification was performed according to Odds and Bernaerts.20 At least one colony exhibiting each colour was streaked into a new CHROMagar Candida plate and then cryopreserved.

Yeast DNA was extracted using the QIAamp® DNA Mini Kit (QIAGEN, Lisbon, Portugal) according to the manufacturer's instructions. Genomic DNA content was determined by spectrophotometry readings at 260nm. Aliquots of 10μl were analysed by electrophoresis in a 0.8% agarose (Bio-Rad, Lisbon, Portugal) gel in 1×TBE buffer (Bio-Rad), and visualized with a UV transilluminator, after ethidium bromide (Bio-Rad) staining (0.5mg/ml).

To assess Candida speciation, a polymerase chain reaction method (PCR) previously described14 was followed. This method uses primer mixtures of the Candida DNA topoisomerase II genes (Table 1). In a multiplex PCR strategy it allows identification of C. albicans, C. guilliermondii, and C. parapsilosis using primer set A (Table 1) and C. dubliniensis, C. krusei, C. kefyr, and C. glabrata using primer set B (Table 1). C. tropicalis is identified in a third PCR reaction using a single pair of primers (Table 1).

PCR amplification was performed in 25μl volume consisting of 1× PCR buffer (160mM (NH4)2SO4, 670mM Tris–HCl pH 8.8, 0.1% Tween 20, and 25mM MgCl2; Bioron, Porto, Portugal); dNTP mixture (200μM each; Bioron); primer mixture (300nM each); 1.25U Taq DNA polymerase (Bioron); and 10–100ng genomic DNA template; the remaining volume consisted of sterilized ultrapure water. PCR was carried out in a MyCycler thermal cycler (Bio-Rad) under the following cycling conditions: 35 cycles of 30s at 94°C, 15s at 57°C, and 45s at 65°C, after a 10-min initial period of DNA denaturation and enzyme activation at 94°C. One blank reaction was performed simultaneously per every 10 tests run by replacing the template DNA by sterilized ultrapure water. DNA from type strains was also included in each reaction as positive and negative controls.

An aliquot of 15μl of each PCR product was analysed by electrophoresis in a 1.2% agarose gel in 1× TBE buffer. Fragments were visualized by ethidium bromide staining (0.5mg/ml) with a UV transilluminator. The size of the amplified DNA fragments was determined by comparison with a 100-bp DNA marker (Bioron).

All isolates whose presumptive identification in CHROMagar Candida did not correspond to the molecular identification were re-tested by two independent researchers in a blind assay.

CHROMagar Candida medium sensitivity was calculated as: [No. of true positives×100/(No. of true positives+No. of false negatives)], and specificity as: [No. of true negatives×100/(No. of true negatives+No. of false positives)]. Statistical analysis was performed using GraphPad Prism, version 5.00 software for Windows. Data were analysed using two-tailed chi-square test or Fischer test to measure association between Candida species distribution within groups. A statistical confidence interval of 95% was established.

Presumptive identification of yeasts clinical isolates was based on their colour on CHROMagar Candida medium. Samples were processed in parallel with C. albicans, C. krusei, and C. tropicalis type strains that presented the expected colours on CHROMagar Candida medium: green, pink, and blue, respectively.

For each culture, representative isolates were identified by PCR using Candida specific primers pairs for the genomic sequences of DNA topoisomerase II gene (Table 1).14 Reference DNA strain was included in each assay as control. DNA of C. albicans (Fig. 1A, lane 1), C. guilliermondii (Fig. 1A, lane 3), and C. parapsilosis (Fig. 1A, lane 4) were amplified using primer set A (Table 1). Primer set B (Table 1) allowed identification of C. glabrata (Fig. 1B, lane 2), C. dubliniensis, C. krusei, and C. kefyr reference strains (data not shown). Finally, the set of primers C (Table 1) allowed identification of C. tropicalis reference strain (Fig. 1C, lane 5). For all Candida species the amplicon size obtained (Fig. 1) was as expected (Table 1). Clinical isolate identification was based on the comparison of size of the amplified DNA products (assessed by the DNA ladder) with the respective type strain PCR product. Fig. 1 shows an example of the identification of an isolate of C. albicans (lane 6), C. glabrata (lane 7), C. guilliermondii (lane 8), C. parapsilosis (lane 9), and C. tropicalis (lane 10).

Fig. 1.

Candida species-specific amplification of DNA topoisomerase II fragments. Genomic DNA was amplified using different sets of species-specific primers (from A to C), in accordance with Table 1. Lanes: M, 100-bp DNA marker and their molecular size in bp is indicated in the left margin; 1. C. albicans CECT 1472; 2. C. glabrata ATCC 2001; 3. C. guilliermondii ATCC 6260; 4. C. parapsilosis ATCC 22019; 5. C. tropicalis ATCC 750; 6–10. Example of clinical isolates of each species; 11. Blank. C. dubliniensis, C. krusei ATCC6258, and C. kefyr ATCC 204093 identification was omitted for simplicity. Arrows on the right indicate molecular weight of the amplified products.

(0.19MB).

CHROMagar Candida phenotypic characteristics of the Candida species identified by PCR are specified in Table 2. As expected, C. glabrata, and C. parapsilosis did not present a distinguishable colour in this medium. Using PCR as a standard method, and considering the CHROMagar Candida identifying colours claimed by the manufacturer, CHROMagar Candida sensitivity and specificity for C. albicans were found to be 97.9% and 83.3%, respectively. For C. tropicalis CHROMagar Candida sensitivity was 66.7%, and specificity 100%.

From the 97 patients evaluated, 53 were identified as oral Candida carriers: 81.1% were females (n=43), and 18.9% males (n=10), with ages ranging from 28 to 91 years old (mean=61 and median=62 years old). Denture wearers accounted for 84.9% of the individuals (n=45).

C. albicans was identified in 79% of the samples, being the predominant Candida species. Additionally, C. parapsilosis comprised 6.5% of the isolates, followed by C. glabrata (4.8%), C. tropicalis (3.2%), and C. guilliermondii (1.6%). Five percent of the CHROMagar Candida positive samples (Table 2) were not identified. The distribution of Candida isolates within gender, and age groups is presented in Table 3. There was no association between C. albicans or NCAC species carriage within (i) gender (P=0.7), (ii) the age groups defined (P=0.83), or (iii) denture wearing (P=1).

Seventeen percent of the individuals (n=9) presented more than one Candida species per sample (Table 4). These individuals (seven females, two males) had a mean age of 58 years old (range=38–88 years old). The percentage of patients wearing denture in this sub-population (88.9%) was similar to that observed in the population studied.