Differentiation of Candida albicans Species Complex by Tobacco Agar Obtained from Different Cigarette Brands Available in Colombia

Aim To evaluate the efficacy of different brands of cigarettes in the preparation of tobacco agar, for the differentiation of these related yeasts. Methodology. Tobacco agar was prepared using six brands and four varieties of cigarettes, and 125 clinical isolates previously identified by PCR and Maldi-Tof were used. To determine whether the results of the microbiological tests were associated with similarities in the chemical components of cigarettes, thin-layer chromatography was performed. Results Candida dubliniensis colonies presented hue differences according to the incubation temperature and the brand or variety of cigarette used, except in the tobacco agar produced with Marlboro Xpress cigarette, where its differentiation was not possible. The chromatograms showed few differences among apolar and medium polarity extract components. Conclusions Tobacco agar is a low-cost tool used for the differentiation of Candida dubliniensis; however, incubation temperature and cigarette brand affect the performance of the media. No relationship was found between the microbiological results and the chemical similarity of the extracts of the cigarettes by chromatography.


Introduction
Candida albicans is the main human fungal pathogen; however, Candida dubliniensis and Candida africana (a biovariety of C. albicans) are emerging yeasts gaining clinical and epidemiological relevance that share morphological and physiological characteristics, thus forming the Candida albicans complex [1,2].
Due to the phenotypic similarities, these isolates can be identified as C. albicans, causing an overestimation in their epidemiology [3]. Although several phenotypic identification systems are designed to differentiate C. dubliniensis from C. albicans, they do not allow identifying Candida africana [4]. e discrimination between these species is relevant for the understanding of their clinical importance and epidemiological role in human infections [5][6][7]. e tobacco agar allows, in most cases, the differentiation between C. dubliniensis and C. albicans; but it is not used in Colombia as a routine medium. In addition, there are no reports of their use for C. africana isolates. e aim of this work is to evaluate the effectiveness of cigarettes available in Colombia for the preparation of tobacco agar on the differentiation of C. albicans, C. dubliniensis, and C. africana. e mixes were boiled for 30 minutes in a water bath and then filtered with cheesecloth to obtain the extracts, which were then diluted to 250 mL with distilled water. 5 g of agar was added to the extracts and sterilized at 121°C for 15 minutes. Finally, each medium was cooled to 55°C and poured into 90 mm petri dishes.

Culture.
Fresh cultures on the Sabouraud agar (peptone 1%, glucose 2%, and agar 2%) were used and then streaked on tobacco agar plates of each cigarette. Plates were incubated at 37°C and 28°C for 7 days, separately. Variations of macromorphology (colony color and margin) and micromorphology (presence of chlamydoconidia at 100x and 400x magnification when the microorganism was stained with lactophenol cotton blue) were registered daily. e experiment was carried out in triplicate. C. albicans ATCC 28367, C. albicans ATCC 90028, and C. dubliniensis ATCC 3949 and ATCC 11473 were used as control strains.

in-Layer Chromatography.
is technique was performed to determine commonalities between the microbiological results and the TLC banding patterns of the extracts. e cigarettes were randomly numbered as follows: (1) Fortune, (2) Rhotmans, (3) L & M, (4) Marlboro Classic, (5) Marlboro fusion, (6) Marlboro gold, (7) Marlboro Xpress, (8) Lucky Strike, (9) Chesterfield, and (10) Philip Morris (of Argentine origin). 3 g of each variety of cigarette was subjected to consecutive extraction with petroleum ether, dichloromethane, and ethanol by cold maceration in an orbital shaker at 114 rpm for 120 hours [8]. e extracts were concentrated by using a rotavapor (Buchi R-111) at a maximum temperature of 40°C. Later, the residual solvent was brought to dryness in a vacuum oven at 30°C. e most appropriate mobile phase was selected by performing preliminary thin-layer chromatography in which the dried extracts were redissolved and then eluted on Silica Gel 60 G with fluorescence marker F254 as the stationary phase. e petroleum ether, dichloromethane, and ethanol extracts were eluted with petroleum ether : acetone (8 : 2), toluene : acetone (8 : 2), and three proportions of dichloromethane : ethanol (8 : 2, 7 : 3 and 6 : 4), respectively. Long-wave ultraviolet (UV) light (365 nm), short-wave UV light (254 nm), and 5% (w/v) vanillin-sulfuric acid were used as visualization agents [9].

Results
In all media and at both incubation temperatures, C. albicans and C. albicans var. africana produced smooth-edged, whiteto-cream color colonies, without chlamydoconidia and without or with poorly developed pseudomycelium. No variations were observed when extending the incubation up to 10 days.
Colonies of C. dubliniensis displayed differences in hue, depending on the incubation temperature and the brand of cigarette used; at 37°C, yellowish brown colonies (darker color) were observed on agar media prepared with Fortuna and Marlboro Classic cigarettes (Figure 1 At 25°C, C. dubliniensis produced yellowish brown colonies (darker color) on tobacco agar prepared from Marlboro Classic, Marlboro Gold, Fortuner, and Marlboro Fussion cigarettes and colonies with pale color (weak positive) in tobacco agar prepared from Rhotmans, Lucky Strike, L & M, and Chesterfield cigarettes. No differences were observed between the colony color of C. dubliniensis, C. albicans and C. africana on the agar prepared with Marlboro Xpress cigarettes. On agar media prepared from Phillip Morris cigarettes (external control), C. dubliniensis produced intense yellowish-brown colonies. Remarkably, in all media, C. dubliniensis produced scalloped margin colonies (Table 1).
Microscopic examination of C. dubliniensis growing on Marlboro Classic, Marlboro Gold, and Marlboro Fussion revealed abundant production of large and spherical chlamydoconidia (Figure 1(b)), whereas on tobacco agar prepared from Fortuna cigarettes, only few small and subspherical chlamydoconidia were detected. e chromatogram obtained with petroleum ether : acetone (8 : 2) and revealed with vanillin had differences in the spots of the Lucky Strike cigarette samples (8) (A, with Rf � 0.31) and 10 (B, with Rf � 0.28 and C, with Rf � 0.65), as well as a higher concentration in some bands (Figure 2(a)). In the other samples, similar migrations of the bands were observed. With short wave UV light, the same migration bands were observed in all the samples, while with long UV light, the same bands were observed, but with a higher concentration in samples 8 and 10 (Lucky Strike and Phillip Morris).
Concentrated spots were observed in samples 2, 6, 7, 8, and 10 on TLC plates eluted with toluene : acetone (8 : 2) and revealed with vanillin. In sample 2 (Rhotmans), two 2 International Journal of Microbiology   additional spots of lower polarity were observed, and the sample 10 (Phillip Morris) presented 3 spots of different polarities, which were not found in the other samples. e bands of less migration were similar in all the samples (Figure 2(b)). With short-wave UV light, two spots were observed in sample 2 and one spot in sample 10, which did not appear after the revelation with vanillin. In addition, some bands had slight differences between the samples. In the visualization with long UV light, the spots of samples 5,6,7,8,9, and 10 showed a higher concentration. Furthermore, some of the bands of less migration were similar in all the samples.
Additional TLC of the ethanol extract samples 3 and 5 (which showed differences with the rest) was performed. e plates were eluted with three proportions of dichloromethane : ethanol (8 : 2, 7 : 3, and 6 : 4). Finally, a predominance of yellow and red bands was observed with vanillin reagent, an indicative of flavonoid and terpene compounds (Figure 2(c)).

Discussion
In this study, all C. dubliniensis produced brown or yellowish colonies with chlamydoconidia in most media, and all C. albicans produced smooth white-to-cream colonies without chlamydoconidia.
ese results are similar to those described by Pineda et al. in Argentina, who used Exceter ® pipe tobacco and 100% C. dubliniensis isolates produced brownish and rough colonies, whereas 98.5% of C. albicans isolates produced white and smooth colonies. Incubation was mantained for 48-72 hours at 30°C (n � 212 isolates) [10]. albicans produced scarce and solitary chlamydoconidia and all the C. dubliniensis isolates produced abundant chlamydoconidia in clusters. Moreover, 8/10 C. dubliniensis produced rough and yellowish colonies, and all (183) C. albicans isolates produced smooth and white-to-creamcolored colonies [11].
In Brazil, Silveira-Gomes et al. used tobacco from Fumo Extra Forte ® , and the culture plates were incubated at 28°C and observed daily up to 96 h. Morphological aspects suggestive of C. dubliniensis on tobacco agar (rough/brownish-yellow colonies and hyphae with chlamydoconidia) were observed in nine isolates [12]; Monteiro included 200 isolates, presumptively identified as C. albicans or C. dubliniensis, that were inoculated on tobacco agar and incubated at 28°C for 4 days. However, all the isolates were further identified as C. albicans by PCR [13].
Recently, Oliveira-Liveiro (Brazil, 2017) reported a sensitivity of 81.5%, when comparing the phenotypic results of C. albicans and C. dubliniensis, with the results of molecular techniques. In the study, six strains produced brownish and rough colonies and developed chlamydoconidia, but only two were identified as C. dubliniensis by PCR. Furthermore, one of the isolates, identified as C. dubliniensis by PCR, did not produce chlamydoconidia on tobacco agar [7].
In our study, C. dubliniensis macroscopic differences were subtle or absent in the incubation at 37°C, so that we suggest the incubation at 28°C on agar media prepared from Marlboro Classic, Marlboro Gold, or Marlboro Fussion cigarettes. Additionally, it was not possible to discriminate C. dubliniensis on tobacco agar produced from Marlboro Xpress cigarettes, and no agar allowed the differentiation between C. albicans and C. albicans var. africana.
After an exhaustive bibliographic search for the literature in English and Spanish, using the electronic database Medline and LILACS and the terms "cigarette," "tobacco agar," and "Candida," it was not possible to find a study comparing different brands of cigarettes for the preparation of tobacco agar, nor with the use of thin-layer chromatography. However, when the search was performed for the gray literature in Google and Google scholar, only one document was found comparing the discrimination of C. dubliniensis from C. albicans using three brands of local cigarettes in Perú; no differences were found according the cigarette brand [14]. e extracts of cigarettes analyzed by TLC were similar to each other regarding apolar and medium polarity components (simple phenols, monoterpenes, sesquiterpenes, and diterpenes), according to the color of the blue spots revealed with long UV light and with vanillin. In extracts of medium polarity, the spots of all the samples probably represent some alkaloids and triterpene molecules. e presence of a small compound, similar to formic acid, was also detected. e toluene : acetone fraction corresponds to medium polarity, which means that these compounds show variation in all 10 cigarette samples.
In the polar (ethanolic) extract, blue fluorescent bands were present under long-wave UV light. ese bands are characteristic of complex phenols like flavonoids, anthocyanins, and tannins and some derivatives of quinones and anthraquinones.

Conclusions
In conclusion, the tobacco agar described by Tendolkar et al. for the differentiation of Cryptococcus neoformans, [15] and later by Khan et al. for the differentiation of C. dubliniensis from C. albicans, [16] varies in capacity of discrimination of C. dubliniensis according to the brand of cigarette used and the incubation temperature. Interestingly, no relationship was found between the microbiological results and the TLC band patterns of the extracts; therefore, an exhaustive analysis of the chemical composition and its relationship with the performance of tobacco agar can be achieved with gas chromatography and mass spectrometry.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request. 4 International Journal of Microbiology