Alternative identification test relying upon sexual reproductive abilities of Candida lusitaniae strains isolated from hospitalized patients.

The in vitro mating ability of Candida lusitaniae (teleomorph Clavispora lusitaniae) clinical isolates has been investigated. Studying the effects of culture conditions, we showed that ammonium ion depletion in the medium is a major trigger of the sexual cycle. Moreover, a solid support is required for mating, suggesting a role for adhesion factors in addition to the mating type gene recognition function. Monitoring of mating and meiosis efficiency with auxotrophic strains showed great variations in ascospore yields, which appeared to be strain and temperature dependent, with an optimal range of 18 to 28 degrees C. The morphogenetic events taking place from mating to ascospore release were studied by scanning and electron microscopy, and the ultrastructure of the conjugation canal, through which intercellular nuclear exchanges occur, was revealed. Labeling experiments with a lectin-fluorochrome system revealed that the nuclear transfer was predominantly polarized, thus allowing a distinction between the nucleus donor and the nucleus acceptor strains. The direction of the transfer depended on the strain combination used, rather than on the genotypes of the strains, and did not appear to be controlled by the mating type genes. Finally, we demonstrated that all of the 76 clinical isolates used in this study were able to reproduce sexually when mated with an opposite mating type strain, and we identified a 1:1 MATa/MATalpha ratio in the collection. These results support the idea that there is no anamorph state in C. lusitaniae. Accordingly, the mating type test, which is easy to use and can usually be completed within 48 h, is a reliable alternative identification system for C. lusitaniae.

Comparative evaluation of a commercial system for identification of Candida lusitaniae.

The ability of the API Candida system (bioMérieux, France) to identify Candida lusitaniae isolates was evaluated in comparison to the Auxacolor and ID 32C systems using 52 clinical isolates previously identified on the basis of their morphology and their biochemical reactions in the Auxacolor and ID 32C systems. The API Candida system failed to definitively identify most of the strains tested within 24 h. No beta-maltosidase activity was detected in 28 strains, and supplementary tests were required to discriminate Candida lusitaniae, Candida famata and Candida guilliermondii. The API Candida system is not suitable for identification of Candida lusitaniae. In comparison, the Auxacolor system is easy to use and interpret, allowing rapid identification of this species; however, the ID 32C system is required for identification of atypical strains.

Evaluation of the E test for antifungal susceptibility testing of Candida glabrata.

The E test was compared to the reference NCCLS broth macrodilution method for susceptibility testing of Candida (Torulopsis) glabrata. The MICs of amphotericin B, flucytosine, fluconazole and itraconazole were determined using the appropriate culture media (RPMI 1640 agar with 2% glucose, Casitone agar or Antibiotic Medium 3 agar) according to the drug tested. Agreement between the two methods was within plus/minus two dilutions for 77-100% of test results, according to the drug/medium combination. The study revealed problems in determining the MICs of azoles using the E test, and confirmed the suitability of Casitone agar for susceptibility testing of fluconazole even if results were read within 24 h.

Amphotericin B susceptibility testing of Candida lusitaniae isolates by flow cytofluorometry: comparison with the Etest and the NCCLS broth macrodilution method.

A flow cytofluorometric susceptibility test (FCST) was used for rapid determination of the susceptibility of Candida lusitaniae isolates to amphotericin B. The test is based on the decrease in fluorescence intensity of cells stained with 3,3'-dipentyloxacarbocyanine iodide (DiOC5(3)), a membrane potential-sensitive cationic dye, after drug treatment. A total of 58 C. lusitaniae clinical isolates including strains known to be amphotericin B-resistant on the basis of in-vivo and/or in-vitro data were tested. MICs were determined concurrently by the NCCLS broth macrodilution method and the Etest, both with antibiotic medium 3. Regression analysis demonstrated that the data from the FCST and the Etest were better correlated (r = 0.93, n = 59, P < 0.001) than those from the FCST and the NCCLS method (r = 0.63, n = 59, P < 0.001). The FCST readily identified a series of putatively susceptible and resistant isolates. Our study points out the advantages of the flow cytometry approach in antifungal susceptibility testing of yeasts, since speed remains a major problem in conventional tests.

Evaluation of a flow cytofluorometric method for rapid determination of amphotericin B susceptibility of yeast isolates.

A rapid-flow cytofluorometric susceptibility test for in vitro amphotericin B testing of yeasts was evaluated and compared to the National Committee for Clinical Laboratory Standards (NCCLS) M27-T reference broth macrodilution method. The flow cytofluorometric method is based on the detection of decreased green fluorescence intensity of cells stained with DiOC5(3), a membrane potential-sensitive cationic dye, after drug treatment. Testing was performed on 134 clinical isolates (Candida spp. and Torulopsis glabrata). From the dose-response curve obtained for each isolate, three endpoints were calculated by computer analysis (the concentrations at which the fluorescence intensity was reduced by 50, 80, and 90%, i.e., 50% inhibitory concentration [IC50], IC80, and IC90, respectively). A regression analysis correlating these endpoints with the M27-T MICs showed that the best agreement was obtained with IC80. The flow cytofluorometric method showed good reproducibility with control strains. These initial results suggest that the flow cytofluorometric method is a valid alternative to the NCCLS reference method.

In-vitro susceptibility pattern of Candida lusitaniae and evaluation of the Etest method.

The antifungal susceptibility of 35 Candida lusitaniae isolates was determined in vitro by the National Committee for Clinical Laboratory Standards (NCCLS) M27-P macrodilution methodology. All the isolates were susceptible to ketoconazole, itraconazole and fluconazole. Of the 35 isolates, eight (23%) were resistant to flucytosine. For amphotericin B, M27-P yielded a narrow range of MICs (0.06-0.5 mg/L). We therefore investigated the activity of this drug by reading MICs at 72 h and by using alternative media, namely casitone complex medium (CCM) and antibiotic medium 3 (M-3). Reading at 72 h did not generate reproducible results. CCM and M-3 provided better discrimination between the isolates but did not change the rank order of the MICs. We thus concluded that all the isolates were susceptible to amphotericin B. We also conducted an evaluation with the Etest method according to the manufacturer's instructions with RPMI 1640 agar, CCM and the alternative semi-synthetic medium (SSM). For RPMI 1640, agreements +/-2 dilutions were 58% for amphotericin B, 92% for flucytosine, 57% for ketoconazole, 92% for fluconazole and 74% for itraconazole. CCM did not improve the agreement rates between the two methods but it led to better growth of all the isolates. The suitability of SSM was pointed out with itraconazole. The poor agreement rates for amphotericin B and ketoconazole call for further evaluation of the Etest method to assess several drug-organism combinations.