Previous Article | Next Article ![]()
Applied and Environmental Microbiology, July 2002, p. 3622-3627, Vol. 68, No. 7
0099-2240/02/$04.00+0 DOI: 10.1128/AEM.68.7.3622-3627.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
R. G. Price,2* G. Midgley,3 W. Khamri,1 and A. C. Richardson1
PPR Diagnostics Ltd., London E1W 1AT,,1 Division of Life Sciences, King's College London, London SE1 9NN,2 Department of Medical Mycology, St. John's Institute of Dermatology, St. Thomas' Hospital, London SE1 7EH, United Kingdom3
Received 21 December 2001/ Accepted 11 April 2002
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
Currently, two chromogenic agars are widely used in clinical mycology laboratories for presumptive detection and identification of Candida spp., particularly C. albicans. These are Candida ID agar, a product of bioMerieux (3, 15), and CHROMagar Candida, produced by CHROMagar Company Ltd. (3, 16, 18, 24). Candida ID agar, which superseded Albicans ID2 medium (3, 7, 14, 20), is based on a chromogenic indolyl glucosaminide substrate which is hydrolyzed by C. albicans to give a turquoise or blue insoluble product. C. tropicalis, Candida lusitaniae, and Candida guilliermondii appear pink on this agar, and other species of Candida are white. CHROMagar Candida also uses a chromogenic ß-glucosaminidase substrate, which is metabolized to give green colonies of C. albicans, steel blue colonies of C. tropicalis, and fuzzy rose-colored colonies of C. krusei.
In the present study, a number of novel chromogenic glucosaminide substrates (1) were evaluated for their usefulness in differentiating Candida spp. in agar media. This led to development of a new agar medium containing the substrate ammonium 4-{2-[4-(2-acetamido-2-deoxy-ß-D-glucopyranosyloxy)-3-methoxyphenyl]-vinyl}-1-(propan-3-yl-oate)-quinolium bromide (VLPA-GlcNAc) (Fig. 1). This medium was optimized for sensitivity to C. albicans, C. tropicalis, and Candida kefyr and was tested with a wide range of yeasts and some molds. The efficacy of the new medium was compared to the efficacies of the two commercially available chromogenic agars described above.
|
| MATERIALS AND METHODS |
|---|
|
|
|---|
Chromogenic substrates.
The chromogenic substrates usedwere closely related to the phenols described by Aamlid et al. (1) and were obtained from PPR Diagnostics Ltd., London, United Kingdom. They were 2-{2-[4-(2-acetamido-2-deoxy-ß-D-glucopyranosyloxy)-3-methoxyphenyl]-vinyl}-3-methyl-benzothiazolium iodide (VBzTM-GlcNAc), 4-{2-[4-(2-acetamido-2-deoxy-ß-D-glucopyranosyloxy)-3-methoxyphenyl]-vinyl}-1-ethyl-quinolinium iodide (VLE-GlcNAc), 4-{2-[4-(2-acetamido-2-deoxy-ß-D-glucopyranosyloxy)-3-methoxyphenyl]-vinyl}-1-methyl-quinolinium iodide (VLM-GlcNAc), VLPA-GlcNAc (Fig. 1), 2-{2-[4-(2-acetamido-2-deoxy-ß-D-glucopyranosyloxy)-3-methoxyphenyl]-vinyl}-1-ethyl-quinolinium iodide (VQE-GlcNAc), 2-{2-[4-(2-acetamido-2-deoxy-ß-D-glucopyranosyloxy)-3-methoxyphenyl]-vinyl}-1-methyl-quinolinium iodide, and 5-[4-(2-acetamido-2-deoxy-ß-D-glucopyranosyloxy)-3-methoxyphenylmethylene]-2-thioxothiazolidin-4-one-3-acetic acid (VRA-GlcNAc). In water, most of the substrates had comparatively low solubilities (<2.0 mM); the only exception was VLPA-GlcNAc, whose solubility was >20 mM.
Chromogenic media.
Sabouraud dextrose agar (SDA) (Oxoid) was the preferred basal medium to which novel glucosaminide substrates were added; different batches of this agar and substrates were used in a 2-year period. The SDA (CM41; Oxoid) contained (per liter) 10.0 g of mycological peptone, 40.0 g of glucose, and 15.0 g of agar (Bacteriological No. 1). SDA plus a chromogenic glucosaminide substrate (0.32 g liter-1) was quickly heated to the boiling point in glass bottles over a Bunsen burner and tripod and then removed from the heat. After it cooled to about 55°C, the medium was poured into 90-mm petri dishes. After the medium set, the plates were surface dried for 20 min at 37°C and used immediately or stored at 4°C for 6 weeks. The pH of agar plates was 5.6 ± 0.2, as determined with a flat pH electrode (Gelplas combination electrode; Merck Ltd., Poole, United Kingdom).
In the initial experiments, the new chromogenic substrates (at concentrations of 0.35 and 0.7 mM) were also tested in a variety of basal media (obtained from Oxoid, except where indicated otherwise). These media included (i) media formulated for the growth of fungi, including SDA, Sabouraud maltose agar, malt extract agar, potato dextrose agar, corn meal agar, and rice extract agar (with and without Tween 80; Becton Dickinson and Co., Cowley, Oxfordshire, United Kingdom); and (ii) media usually used for bacterial growth, including nutrient agar (Oxoid; Merck, Darmstadt, Germany), Lab-lemco agar, cystine-lactose-electrolyte-deficient medium (CLED), modified CLED (with and without lactose), standard plate count agar, and yeast extract agar. The substrates were added to the basal agar in each of the following ways: as a powder before boiling as described above; as a powder before autoclaving; and as a filter-sterilized solution after the agar was autoclaved and cooled to 55°C.
In addition, VLPA-GlcNAc hydrolysis was compared by using SDA obtained from a number of suppliers, including MAST Group Ltd. (DM200D; Bootle, Merseyside, United Kingdom), Lab M (Lab009; Bury, United Kingdom), Difco (SDA 210950 and modified SDA; purchased from Becton Dickinson and Co.), BBL (SDA 4311584 and SDA Emmons 4311589; purchased from Becton Dickinson and Co.), Oxoid (CM41), bioMerieux, and Merck (Sabouraud agar containing 2% glucose and Sabouraud agar containing 4% glucose). VLPA-GlcNAc was boiled with each test agar (as described above) before plates were poured.
Optimization of the VLPA-GlcNAc medium.
During development of the medium, the effects of chromogenic substrate concentration, incubation temperature, and pH on the development of colony color were investigated. In addition, the effects of including the following groups of compounds in the medium were determined: inducers of yeast morphogenesis, including N-acetyl-D-glucosamine (0.25 to 5 g liter-1) (5, 22) and glucose (0.5 to 100 g liter-1 (11, 19); an inducer of germ tube production, hemin (10 to 50 mg liter-1) (4); an inducer of chlamydospore production, Tween 80 (2 to 10 g liter-1); sugars assimilated by C. albicans, including trehalose (2 to 20 g liter-1), raffinose (1 to 30 g liter-1), and sucrose (2 to 50 g liter-1); and cell wall-permeabilizing agents, including n-octylglucoside (0.003 to 0.09%, wt/vol), sodium dodecyl sulfate (0.005 to 0.08%, wt/vol), dithiothreitol (0.005 to 0.1%, wt/vol), and 2-mercaptoethanol (0.05 to 1%, vol/vol) (13).
Inoculation of media.
Test yeasts and fungi were streaked onto chromogenic media by using sterile plastic inoculating loops. Inocula were grown on SDA (Oxoid) plates for 24 to 48 h at 37°C or, as indicated below, for 48 or 72 h at 30°C.
Comparison of the new Candida agar with commercially available media.
A total of 125 test strains (98 Candida strains and 27 non-Candida strains) were used. The media evaluated were (i) Candida diagnostic agar (CDA) developed in this study and containing VLPA-GlcNAc, (ii) Candida ID agar ready-poured plates (code 4354093; bioMerieux), and (iii) CHROMagar Candida ready-poured plates (code 43591; Becton Dickinson and Co.). Candida ID agar and CHROMagar Candida plates were stored at 4°C in the dark for a maximum of 7 days. CDA plates were similarly stored but for up to 6 weeks. Test strains were grown on SDA (Oxoid) and streaked onto plates of CDA, Candida ID agar, and CHROMagar Candida, as described above. Plates were incubated at 37°C and observed to determine colony coloration and morphology at 24, 48, and 72 h. In accordance with the manufacturer's instructions, Candida ID agar plates were incubated in the dark. Experiments were replicated in those cases in which strains gave atypical results. In all cases, the results of repeated tests were the same as the results of the original tests.
| RESULTS AND DISCUSSION |
|---|
|
|
|---|
The most intense coloration of C. albicans colonies was observed on corn meal agar (substrate concentration, 0.7 mM) at an incubation temperature of 37°C; the colonies appeared pink-orange (VBzTM-GlcNAc), pink (VLE-GlcNAc), brown (VLM-GlcNAc and VLPA-GlcNAc), orange-brown (VQE-GlcNAc), or orange (VRA-GlcNAc) after 48 h. However, although the colony colors were intense, the colored chromophores also diffused into the agar surrounding the colonies. Thus, the most effective substrate-medium combination was VLPA-GlcNAc in SDA, in which, unusually, colonies of C. albicans appeared white with deep-red spots (Fig. 2) while the background agar remained yellow.
|
SDA preparations from several different companies were used in combination with VLPA-GlcNAc to determine whether the use of different medium formulations affected colony coloration. Deep-red multispotted colonies of C. albicans were observed with SDA formulations from Oxoid and MAST, but the latter medium resulted in slightly smaller colonies. The SDA formulations from Lab M, Difco, and BBL gave C. albicans colonies which were orange with red spots, and there was also some diffusion of the orange color into the surrounding agar; in addition, red spots were also evident in colonies of C. tropicalis and C. kefyr. When the SDA formulation from bioMerieux was used, C. albicans and C. tropicalis gave unspotted yellow colonies, although C. kefyr colonies were white with deep-red spots. The Oxoid SDA formulation was therefore chosen for use in the new medium as clearly defined spotted colonies were produced by C. albicans and uniformly pink colonies were produced by C. tropicalis and C. kefyr.
Optimization of the chromogenic medium for detection of Candida spp.
A series of experiments were undertaken to optimize the medium by maximizing C. albicans, C. tropicalis, and C. kefyr color production.
(i) Substrate concentration.
Increasing the concentration of VLPA-GlcNAc (bromide form; 0.1 to 1.0 mM) increased the number of spots in C. albicans colonies and the pink color of C. tropicalis and C. kefyr colonies. Maximal coloration for all three species was achieved with concentrations of
0.5 mM.
(ii) Incubation time and temperature.
Candida species are grown routinely at 37°C or occasionally 30°C. When test plates were incubated at 30°C, it was noted that the spots in colonies of C. albicans tended to merge together to give colonies a ringed appearance at 48 h. At 25°C, growth of Candida spp. was too slow (3 to 4 days) for a routine clinical agar, although spots were formed. At 42°C, growth of the three strains of C. dubliniensis was inhibited, but spot development in C. albicans was greatly reduced. Thus, the most suitable incubation temperature appeared to be 37°C, with plates observed at 24 and 48 h.
(iii) Effect of heat on the substrate.
The substrate decomposed when it was autoclaved in SDA, giving a brown agar. However, no color change was observed when the substrate was boiled in SDA (see Materials and Methods).
(iv) pH.
Maximal spot formation in C. albicans colonies was observed when SDA with a pH between 5.0 and 7.5 was used; the pH of unmodified SDA medium (pH 5.6) is within this range.
(v) Additions to the medium.
Addition to SDA of cell wall-permeabilizing agents, metabolizable sugars, and inducers of yeast morphogenesis, germ tube production, and chlamydospore production (see Materials and Methods) failed to enhance the color reactions of C. albicans or C. tropicalis colonies.
The results suggested that the most appropriate formulation for a differential Candida medium was simply SDA (Oxoid) plus VLPA-GlcNAc (0.55 mM or 0.3223 g liter-1). This agar was designated Candida diagnostic agar (CDA), and plates of this medium were prepared by briefly boiling the constituents prior to pouring.
Colony coloration on CDA.
Ninety-eight Candida strains and 28 non-Candida strains were streak plated onto CDA and incubated at 37°C. The plates were observed for coloration for 48 h or for 72 h if there was any indication of a color change at 48 h. Among the Candida spp., deep-red-spotted colonies of C. albicans were observed (Table 1 and Fig. 2) after 24 h of incubation, and the spots increased in size and number when plates were incubated for 48 h. The pattern of spots varied from many small (diameter, <1 mm) pinprick spots per colony to a smaller number of large spider-like spots (diameter, 2 to 4 mm). C. dubliniensis strains also gave spotted colonies; however, strains of this species could be differentiated by their much poorer growth if plates were incubated at 42°C. Colonies of most C. tropicalis strains were pale pink at 24 h, and maximum coloration was observed at 48 h; however, maximum coloration of strains EL804, 258, El8047, and EM5579 required 72 h of incubation. The C. kefyr strain gave pink colonies at 24 h. Colonies of other Candida species were white. Except for Trichosporon spp., other yeasts (S. cerevisiae, M. pachydermatis, B. capitatus, H. anomala, and Cryptococcus spp.) also gave all-white colonies. Colonies of 7 the 10 Trichosporon strains, representing five species (see Materials and Methods), had some spots. However, colonies of these strains were readily differentiated from those of C. albicans and C. dubliniensis by their folded-lace appearance.
|
In addition to differentiation of C. albicans plus C. dubliniensis strains, the chromogenic agars tested also differentiated additional groups of strains. These groups were C. tropicalis plus C. kefyr strains for CDA and CHROMagar Candida and C. tropicalis plus C. kefyr, C. lusitaniae, and C. guilliermondii strains for Candida ID agar. The sensitivity (Table 2) and specificity for differentiation of these secondary groups of strains were, however, much greater (both 100%) for CDA than for Candida ID agar (94.7 and 93.8%, respectively) and CHROMagar Candida (72.7 and 98.1%, respectively). CHROMagar Candida has also been reported to differentiate C. krusei; the present study included only four test strains of this species, and three gave the predicted (pink) color reaction. C. albicans was clearly distinguishable from other known pathogenic microorganisms in mixed cultures
|
CHROMagar Candida is reported to give green colonies of C. albicans and steel blue colonies of C. tropicalis. In this study, most C. albicans strains gave green colonies after 48 h of incubation; the exception was strain M207, which gave blue-turquoise colonies. However, two additional strains (JP43 and EM628) gave blue-turquoise colonies at 24 h. Colonies of the three C. dubliniensis strains were green. Only 7 of 10 C. tropicalis strains gave the predicted steel blue colony color. CHROMagar Candida is also reported to detect C. krusei colonies by their fuzzy rose color. Three of the four test strains gave pink colonies after 48 h of incubation, but after only 24 h of incubation the colony colors of two of these strains varied from pink to purple. The remaining C. krusei strain gave purple or white colonies even after 48 h of incubation. In addition, the usefulness of colony color in identification of C. krusei appears to be limited as several other yeasts gave pink colonies on CHROMagar Candida, including C. lusitaniae UK NEQAS4620, C. parapsilosis EK6021 and EL7825, and B. capitatus EQ161 and PMC2558. Many other strains gave purple colonies (Table 1), which might also be confused with C. krusei colonies.
In addition to the strains listed in Table 1, nine Trichosporon strains (representing five species) were also streaked on plates of the three test agars. Except for both strains of T. capitatum and one of the two test strains of T. mucoides, the colony colors were similar to those of C. albicans on all three media. However, as noted above, Trichosporon strains are readily differentiated from C. albicans by their colony morphology (2). Thus, the similarity of colony colors does not appear to affect the usefulness of the chromogenic agars for Candida identification.
All of the chromogenic media tested (CHROMagar Candida, Candida ID agar, and CDA) appeared to be useful in presumptive identification of Candida spp. from clinical specimens, although there was variation in the range of species differentiated and in the sensitivity and specificity for target groups. All of the test media had a high sensitivity for C. albicans detection but failed to distinguish strains of this species from C. dubliniensis. This was not unexpected as the newly recognized species C. dubliniensis is known to be closely related to C. albicans and was formerly referred to as atypical C. albicans. It is most commonly isolated from immunosuppressed patients and intravenous drug users who are not infected with human immunodeficiency virus, but it represents only a small proportion of the total Candida isolations in clinical laboratories. C. albicans may be differentiated from C. dubliniensis by PCR methods (12) or by its poorer growth at high incubation temperatures (9). The manufacturers claim that on CHROMagar Candida, C. dubliniensis appears as dark green colonies whereas colonies of C. albicans are light green. Tintelnot et al. (23) found that 56% of C. dubliniensis test strains gave dark green colonies, but it is generally agreed that this feature alone is not sufficient for differentiation of C. dubliniensis (21).
The 100% sensitivity and 100% specificity of CDA for detection of the C.albicans-C. dubliniensis and C. tropicalis-C. kefyr groups provide a significant advantage over Candida ID agar and CHROMagar Candida. A further advantage is that for strains of other important Candida spp. and non-Candida yeast strains, colony color was consistent (white). In contrast, for example, colonies of C. glabrata and C. parapsilosis strains were white or pink on Candida ID agar and purple, white and purple, pink, or gray on CHROMagar Candida (Table 1). In addition, a general difficulty in the use of CHROMagar Candida was that colonies of all of the test strains were various shades of blue, turquoise, purple, or pink. In contrast, on CDA colonies were strikingly red spotted, uniformly pink, or white. Colony colors on Candida ID agar (mostly turquoise, pink, or white) were also relatively easy to differentiate; however, a disadvantage of this agar is that plates must be incubated in the dark.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
| REFERENCES |
|---|
|
|
|---|
-methyl-D-glucoside as determined by the API 20C AUX and Vitek YBC systems. J. Clin. Microbiol. 37:3804-3808.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| J. Bacteriol. | Microbiol. Mol. Biol. Rev. | Eukaryot. Cell | All ASM Journals |
|---|