Notes
Colony morphology of Candida spp. as a guide to species identification

https://doi.org/10.1016/S0732-8893(99)00051-6Get rights and content

Abstract

The presence of colony projections, often referred to as “feet,” have typically been considered a characteristic of Candida albicans. In the current study that examined the colony morphology of numerous different species of Candida, several clinical isolates of Candida tropicalis and Candida krusei were also noted to produce “feet.” The medium and growth conditions under which colony projections were produced by these species were characterized.

Introduction

Colonies of yeast are often found on routine primary isolation media and can be of clinical importance when recovered from clinically significant sites. Rapid tests are often performed to determine if an isolate is Candida albicans or another species of Candida. The procedure that is most often used by clinical laboratories is the germ tube test (GTT). Whereas the GTT is a relatively rapid test, taking 2 to 3 h, an even more rapid test for identification would be of value. Recently, two publications have suggested that the presence of colony protruding extensions termed “feet” could be used to directly identify C. albicans Nagaishi and Baron 1997, Calvin et al 1998.

Although footed colonies are highly suggestive of C. albicans, we were not convinced that previous studies had tested sufficient numbers of non-albicans species to prove that colony extensions alone can be used as a single criterion for the identification of C. albicans. In an attempt to clarify this issue, we performed a study in which four other species of Candida, in addition to C. albicans, were examined for the presence of colony formations known as “feet.”

We examined 125 isolates from five different species of yeast were examined. The organisms were clinical isolates from patients at the University of Iowa Hospitals and Clinics (Iowa City, IA, USA) collected over a four year period from March 1995 to March 1998. The isolates selected included 25 strains each of the following Candida species: C. albicans, C. krusei, C. tropicalis, and C. glabrata. These organisms were identified by using standard microbiologic techniques that included: the GTT, the Vitek YBC identification card (bioMerieux Vitek, Hazelwood, MO, USA), the API Yeast ID card, and colony and microscopic morphology. Isolates had been recovered from blood (n = 24), normally sterile body fluid specimens (n = 18), urine (n = 8), and from tissue biopsies or respiratory tract specimens (n = 75). All isolates were subcultured twice on potato dextrose agar before testing, and all organism identifications were additionally confirmed based on typical colony morphology on CHROMAgar (adds and Odds and Bernaerts 1994, Pfaller et al 1996). Organisms were tested in pure culture.

Each isolate was suspended in saline to a density equivalent to a 0.5 McFarland standard and then further diluted 1:100 in saline; 10 μL of this suspension was inoculated onto the surface of six agar plates, three plates each of 5% sheep blood agar and chocolate agar (Remel Laboratories, Lenexa, KS, USA). The inoculum was distributed evenly across the entire agar surface of the plates. Two of the three plates were incubated at 30 and 35°C in ambient air, whereas the third plate was incubated at 35°C in 5% CO2. Using this method, approximately 100 well-isolated colonies were typically present on each plate. After 48 h of incubation, colonies of each isolate were examined for the presence of “feet.”

Of the 25 isolates of C. albicans examined, colony extensions or “feet” were produced by all but one strain (96%) (Table 1; Figures 1a, b). The species identification of the single strain of C. albicans that did not produce “feet” under any condition was confirmed by the Vitek YBC system (biotype 551455411 with a 99% probability for C. albicans). This isolate was germ tube positive. “Feet” production with C. albicans occurred equally on blood agar and chocolate agar plates, but appeared to be enhanced by CO2 incubation. All of the “feet”-producing strains of C. albicans produced “feet” in CO2 whereas only 63% demonstrated this characteristic in ambient air (Table 2).

The concept we validated in our study was that two other species of Candida, i.e., C. tropicalis and C. krusei, also produced “feet” (Figures 2a, b). Seven of the 25 isolates of C. krusei (28%) examined produced “feet” within 48 h. Six of the 25 isolates of C. tropicalis (24%) also produced “feet” when grown under the appropriate conditions. We did not observe “feet” on the colonies of C. parapsilosis or C. glabrata under any test conditions. We noted that for those strains of C. krusei and C. tropicalis that produced “feet,” the colonies did so more frequently on blood agar plates (100% for both organisms) compared to chocolate agar plates (0% and 17%, respectively). These organisms were also more prone to produce “feet” following CO2 incubation (86% and 100%) than in ambient air (50% and 14%). These results are summarized in Table 2.

Nagaishi and Baron commented that C. krusei has a dry, fuzzy colony appearance that differs from most other yeasts and, as a result, this species would not be confused with C. albicans (1997). We found that with one strain of C. krusei, “feet” were evident at 48 h of incubation on blood agar in CO2, but the colony projections were obscured by the spreading colony morphology and were not apparent when the plate was reincubated for an additional 24 h period. Because isolates of C. krusei tend to grow as flat spreading colonies more readily on chocolate agar during the first 24 to 48 h, it may explain why feet were not apparent on this media.

We conclude that although most C. albicans isolates (96%) produce colony extensions, not all yeast that form “feet” are C. albicans. In particular, C. krusei and C. tropicalis can also produce colony extensions. These observations indicate that the presence of colony extensions cannot be used as a single criterion for the definitive identification of C. albicans. Isolates that produce “feet” can, however, be presumed to be Candida species other than C. parapsilosis and C. glabrata. We invite further discussion of these findings.

References (4)

There are more references available in the full text version of this article.

Cited by (9)

View all citing articles on Scopus
View full text