Identification of Candida species from human immunodeficiency virus-infected patients in Ethiopia by combination of CHROMagar, Tobacco agar and PCR of amplified internally transcribed rRNA spacer region.
Abstract: The purpose of this study was to determine phenotypic and genotypic identification of Candida species from HIV infected patients in Ethiopia. Oral rinses from 13 human immunodeficiency virus (HIV) positive patients were inoculated on CHROMagar. Colonies were then sub cultured on Tobacco agar. DNA was extracted from 71 colonies and amplified by PCR targeting the conserved regions of 18S, 5.8S, and 28S rRNAs. The amplified DNA samples were analyzed their sequence and a part of them were examined by restriction fragment length polymorphism (RFLP). Oral Candida colonization was detected in all of patients with or without oral candidiasis. The number of Candida species on the CHROMagar was more than 1000 CFU/ml in 12 of 13 patients. We showed that phenotypic and genotypic identification process was important to determine the species. These results indicated that HIV-positive patients have high risk to C. albicans and the other Candida speceies infection in Ethipia.

KEY WORDS: Candida, human immunodeficiency virus infection, Ethiopia, rRNA spacer region, CHROMagar, Tobacco agar
Article Type: Report
Subject: Candida (Identification and classification)
Candida (Research)
HIV patients (Diseases)
HIV patients (Health aspects)
Polymerase chain reaction (Usage)
Authors: Isogai, Hiroshi
Mulu, Andargachew
Diro, Erimias
Tekleselassie, Henok
Kassu, Afework
Kimura, Koichi
Nishikawa, Takeshi
Isogai, Takeshi
Pub Date: 03/01/2010
Publication: Name: Journal of Applied Research Publisher: Therapeutic Solutions LLC Audience: Academic Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2010 Therapeutic Solutions LLC ISSN: 1537-064X
Issue: Date: March, 2010 Source Volume: 10 Source Issue: 1
Topic: Event Code: 310 Science & research
Geographic: Geographic Scope: United States Geographic Code: 1USA United States
Accession Number: 227012907

Infections due to Candida spp. and other fungi have increased dramatically in recent years and are of particular importance because of the rising number of immuno-compromised patients (3). Candida albicans accounts for approximately 90% of Candida spp. isolated from yeast infected patients. However, during the last 20 years a marked shift in the spectrum of Candida species such as C. tropicalis, C. glabrata and C. krusei has been noted among different immunocompromised patients, with CD4 lymphocyte counts (less than 200 cells/ [mm.sup.3]). These Candida species; Candida tropicalis, Candida parapsilosis, Candida glabrata, and Candida krusei are of increasing significance as they tend to be more resistant to antifungal agents (2-4). In Ethiopia, there have been no reports about Candida spp. from oral origin. A rapid screening approach is needed for the diagnosis.

While a number of useful tests based on phenotypic characteristics, such as discrimination based on colony color on CHROMagar Candida medium, growth at 42 to 45[degrees]C, and assimilation profiles, have been developed (5-8), they are not completely reliable (9). In this study, we determined in Candida species by combination of CHROMagar, Tobacco agar and polymerase chain reaction (PCR) of the amplified internally transcribed rRNA spacer region. We showed that phenotypic and genotypic identification process was important to determine the species. This work is a first step to determine the clinical features in HIV-associated candidiasis in Ethiopia.


Thirteen HIV positive patients with different clinical stages were randomly selected at the Black Lion Specialized Hospital, Addis Ababa, Ethiopia. After informed consent was obtained, patients were instructed to provide a rinse of oral cavity. Oral rinses were obtained from patients by asking them to rinse their mouths with 10 ml of sterile distilled water for 30 seconds and to expectorate the rinse in to a sterile container. Then the fluid was poured to a CHROMagar (Paris, France) plate and after 30 seconds the plates was drained. In preliminary experiment, we determined 283.6 [+ or -] 23.6 [micro]l of fluid remained on the plate. The plate was then incubated at 37[degrees]C for 48 h. A presumptive identification was made based on the color of the colonies.

The identities of C. dubliniensis and C. albicans isolates were further confirmed by Tobacco agar, which is a new medium for differentiating C. dubliensis from C. albicans. In addition, seventy-one clinical isolates each of Candida tropicalis, Candida glabrata, Candida parapsilosis, and Candida krusei were also tested for colony characteristics on tobacco agar. The method used for the preparation of tobacco agar was the same as that described by Tendolkar et al. (10), except that we used cigarette tobacco instead of tobacco leaves according to the method of Khan et al (11). Briefly, 50 g of tobacco from commercially available cigarette brands (Marlboro; tar, 8 mg; nicotine, 0.6 mg; Philip Morris Products SA, Richmond, Va.) was mixed with 1 liter of distilled water. The mixture was boiled for 30 min and then filtered through several layers of gauze. To this filtrate, 20 g of agar was added, and the volume was made up to 1 liter. It was autoclaved at 121[degrees]C for 15 min. All the test isolates were freshly sub-cultured on Sabouraud dextrose agar (Difco, Detroit, MI), and tobacco agar plates were streaked with a small amount of inoculum from the isolated colonies. The culture plates were incubated at 30[degrees]C and observed daily up to 96 h for colony characteristics, such as surface topography (rough or smooth), formation of hyphal fringes at the periphery, and color. In addition, tobacco agar prepared from Ethiopian cigarettes (Nyala, Ethiopia) was also used for comparison.

Seventy-one local clinical isolates and six reference strains were studied in this study. C. albicans IFM4009, C. glabrata IFM54350, C. parapsilosis IFM5774 and IFM5804, C. tropicalis IFM46821, and C. dubliniensis IFM54605 were kindly provided from Chiba University Research Center for Pathogenic Fungi and for Microbial Toxicoses.

The species identification of the test isolates was done by combination of PCR amplification of their rRNA genes corresponding to unique sequences within the internally transcribed spacer (ITS). Primers for PCR were designed for two separate areas of the DNA encoding the rRNA. The pair of primers, ITS1 (5'-TCC GTA GGT GAA CCT GCG G-3') and ITS4 (5'-TCC TCC GCT TAT TGA TAT GC-3'), targeting the conserved regions of 18S, 5.8S, and 28S rDNAs, have been reported in previous Studies (12-14). Furthermore, oligonucleotide primers were derived from rRNA genes of fungi and can be used for universal fungi PCR15. Forward primer ITS3 (5'-GCA TCG ATG AAG AAC GCA GC-3') corresponds to the 5.8S rRNA gene, and reverse primer ITS4 (5'-TCC TCC GCT TAT TGA TAT GC-3') corresponds to the 28S rRNA gene of fungi.

PCR (ITS1/ITS4 and ITS3/ITS4 primer set, respectively) was performed in a total volume of 50 [micro]l of mixture including 1 [micro]l of extracted specimens according to the manufacture protocol (Takara Premix gamma Taq Takara, Japan). Samples were placed on Astec Program temp control system PC-800 DNA thermal cycler. After an initial step of 5 min at 94[degrees]C, 35 cycles were performed for 30 sec at 94[degrees]C, 30 sec at 55[degrees]C, and 2 min at 72[degrees]C. Finally, an additional extension was achieved for 2 min at 72[degrees]C, and samples were cooled to 4[degrees]C. These products were kept at -20[degrees]C until further processing.

PCR products were digested individually with 10 U of restriction enzyme Hae III by overnight incubation at 37[degrees]C. The resulting restriction fragments were analysed by agarose gel electrophoresis.

Gel electrophoresis with 3% agarose gel was conducted with Tris-acetate-EDTA buffer (20 mM Tris-Acetate, 0.5 mM EDTA, pH 8.3). A Novagen PCR Marker, 50-2000bp (Merk & Co, NJ, USA) was run concurrently with amplicons for sizing of the bands. Gel was stained with 0.01% (v/v) ethidium bromide solution for 20 min and then photographed.

PCR products were puriied by LaboPass[TM] Products DNA Purification kit (COSMO Genetech Co., Korea) and dissolved in distilled water to a final concentration of 20 ng/[micro]l. The PCR products with sequencing primer ITS1 or ITS4 were automatically sequenced with PRISM 3100 genetic DNA analyzer (Applied Biosystems, Weiterstadt, Germany) using the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Weiterstadt, Germany).

This study was conducted after obtained institutional ethical clearance and informed consent from the study subjects. All work was conducted in accordance with the Declaration of Helsinki (1964).


Phenotypic examination showed that oral Candida colonization was detected in all of patients. Eight patients had oral candidiasis such as pseudomembraneus, candidal leukoplakia and erythematous angular cheilitis. The number of Candida species on the CHROMagar (CFU/ml in rinse) was more than 1000 in 12 of 13 patients. There were no significant differences of number of candidal colonization between symptomatic and asymptomotic groups. The species isolated most frequently from the patients was C. albicans (12 of 13 patients), followed by C. tropicalis (11/13), C. krusei (9/13), C. parapsilosis (4/13), C. guilliermondi (2/13), C. glabrata (2/13), and C. dubliniensis (2/13) in phenotypic examinations. Seven patients had only C. albicans and 1 patient had only C. tropicalis. Two or more Candida species were isolated from the other 5 patients.

The colony color and morphology of 71 isolates on CHROMagar were examined and Candida species were suspected. C. albicans and C. dubliniensis showed mint green colony. C. albicans showed slightly light color but it is dificult to differentiate C. dubliniensis from C. albicans. Forty-three isolates were considered to be C. albicans or C .dubliniensis. Purple to pink or purple 11 colonies were determined as C. tropicalis. Nine large rough colonies with whitish pink color were determined as C. krusei. Two isolates showed small smooth pink or red-purple colonies. These were determined as C. glabrata. Six noncolored isolates were suspected to be C. parapsilosis or the other species.

Fig. 1 shows flowchart for detection of yeast isolates from HIV-infected patients in Ethiopia. On Marlboro tobacco agar, 69 of 71 isolates showed white-to-cream colored colonies (Fig. 1). Two isolates produced rough, yellowish-brown colonies with peripheral hyphal fringes after incubation for 72 hours at 30[degrees]C and determined to C. dubliniensis. Thus, it is easy to differentiate C. dubliniensis from C. albicans. On this medium, major isolates (C. albicans) showed smooth, white-to-cream-colored colonies without hyphal fringes after extended incubation for up to 10 days. Like

C. albicans, none of the isolates of C. tropicalis and others formed yellowish-brown colonies on the Tobacco agar and thus were indistinguishable from each other. Similar observations were made when Nyala Tobacco agar prepared from Ethiopian cigarettes were used.


After determination of colony color, PCR was done by using primer sets as shown in Fig.1. All isolates were amplified using universal fungal primer ITS3-4. Most of isolates yielding a products as C. albicans, C. dubliniensis, C. tropicalis, C. krusei is approximately from 330 to 340 bp. Size of PCR products was 330 and 420 bp in six non-colored isolates. Two isolates of them were suspected as C. parapsilosis or the other species (Their species were identified as C. tropicalis after analyzing of DNA sequence). C. glabrata can be differentiated by the approximately 420 bp size of the PCR product. However, PCR products of former non-colored isolates also showed similar size. Size of PCR products was from 330 to 340 bp in reference strain of C. albicans IFM4009, C. parapsilosis IFM5774, C. tropicalis IFM46821, and C. dubliniensis IFM54605. The size was approximately 420 bp in C. glabrata IFM54350.


The intergenic spacer region (primer ITS1-4) was successfully amplified from all tested isolates (Fig. 1, Fig.2A). A distinct product size was obtained before and after HaeIII enzyme treatment (Fig. 1, Fig.2B). A product of approximately 500-530 bp was obtained from C. albicans, C. dubliniensis, C. tropicalis including 4 non-colored isolates on CHROMagar and C. krusei. A product of approximately 870 bp was obtained from C. glabrata. A product of approximately 720 bp was obtained from 2 isolates. These 2 isolates were determined as C. parapsilosis on CHROMagar and size of PCR products using primer ITS 1 and ITS4 was near to C. glabrata. The product of 720 bp could not be obtained from C. parapsilosis IFM5774 and C. glabrata IFM54350. These isolates were determined as C. kefyr after sequencing.

The isolates were studied further by RFLP analysis following digestion of the PCR products by the HaeIII. After HaeIII digestion, C. albicans, C.dubliniensis and C. tropicalis yielded 80-90 and 440 bp. C. krusei yielded 40, 90 and 370 bp. C. glabrata. yielded 200 and 670 bp. C. kefyr yielded 20, 70 and 630 bp.

Size of PCR products (ITS1-4) was from 520 to 540 bp in reference strain of C. albicans IFM4009, C. parapsilosis IFM5774, C. tropicalis IFM46821, and C. dubliniensis IFM54605. The size was approximately 870 bp in C. glabrata IFM54350. After HaeIII digestion, C. albicans IFM4009, C. tropicalis IFM46821, and C. dubliniensis IFM54605 yielded 90 and 430-440 bp. C. glabrata IFM54350 yielded 200 and 670 bp after the digestion. A product of approximately 340 bp (ITS3-4) or 540 bp (ITS1-4) was obtained from C. parapsilosis IFM 5774. After HaeIII digestion, the fragment size was 40, 110 and 390 bp, respectively, and such RFLP pattern was not observed in clinical isolates.

After sequence, the homology score between 18S ribosomal RNA gene from C. albicans (GenBank/EMBL/DDBJ accession No.AY342214), C. tropicalis (EF196807), C. krusei (EF198013), C. glabrata (AY939794), C. kefyr (AF543841) was 99% in isolates as C. albicans, C. tropicalis and C. kefyr and 96% in isolates as C. krusei and C. glabrata, respectively.


The results presented here indicate that major species of Candida from clinical isolates in Ethiopia should be assigned to the species C. albicans. PCR can be useful for the differentiation of isolates to the species level by using RFLP. The method that detects the presence and the size of the intron in the 25S rDNA is particularly easily adapted for use in reference laboratories for the rapid identification of large numbers of isolates. The results obtained in this study demonstrated that by using the restriction enzyme HaeIII the differentiation of C. albicans and non-albicans strains could be easy. Our isolates as C. albicans showed light mint color in CHROMagar. All of them showed two fragments of 90 and 430 bps after the enzyme treatment.

In this study, CHROMagar allowed the presumptive identification of C. albicans, C. tropicalis, C. krusei, C. parapsilosis. Some authors have observed that C. albicans and C. dubliniensis may show colonies with different shade of green (16, 17). The color intensity, however, is not restrictive for each isolates including C. albicans and C. dubliniensis. It is also dificult to determine them by using PCR and RFLP. Khan et al. reported that Tobacco agar provides a simple tool for presumptive differentiation of C. dubliniensis from C. albicans (11). Interestingly, we could conirm the usefulness of Tobacco agar for the aforementioned purpose.

There have been various molecular approaches for detection of fungi from clinical isolates. Targets of PCR amplification included 18S rDNA, 5.8S rDNA, intergenic spacer regions and 28S rDNA regions. The size of the 18S, 5.8S and 26S rDNA genes are essentially identical in all Candida species, while the lengths of the ITS regions depend on the species. The ITS region is located between the 18S and 26S rDNA genes and is subdivided into the ITS1 and ITS2 regions. The differences of sequences in these regions have been used to detect and identify fungi. We used the differences in the length of both ITS1 and ITS2 regions to determine Candida species. C. glabrata and C. kefyr could be easily identiied from C. albicans, C. tropicalis and C. krusei, because the PCR products from these 2 species were signiicantly larger than those of C. albicans and other species. However, molecular weights of PCR products were near among C. albicans, C. tropicalis, C. dubliniensis, parapsilosis and C. krusei. There was the risk for miss identification when species were determined by only a PCR amplification method.

Phenotypic characteristics on the selective agar medium such as CHROMagar and Tobacco agar are useful to determine Candida species with support of molecular diagnosis. The identification of Candida species with the amplification and sequencing of intergenic spacer regions and RFLP analysis is a practical and reliable method. It is useful for identification a part of clinically isolated Candida species, such as C. guiliermondii and C. parapsilosis identiied by CHROMagar in this study. Tobacco agar is useful tool for identifying C. dubliniensis. The composition of this medium is very simple, brands of cigarette is not concerned and easy to make. The combination of these methods is useful for diagnosis oral candidasis and risk judgment of the disease in patients with HIV.


Phenotypic and genotypic identification process was important to determine Candida species. C. albicans, C. tropicalis and C. krusei were major species. In Ethiopia, HIV-positive patients have high risk to C. albicans and the other Candida species infection.


We thank Chiba University Research Center for Pathogenic Fungi and for Microbial Toxicoses for providing the Candida strains.


(1.) Fraser VJ, Jones M, Dunkel J, et al. Candidemia in a tertiary care hospital: epidemiology, risk factors, and predictors of mortality. Clin Infect Dis. 1992; 15:415-421.

(2.) Fridkin SK, Jarvis WR. Epidemiology of nosocomial fungal infections. Clin Microbiol Rev. 1996; 9: 499-511.

(3.) Hazen KC. New and emerging yeast pathogens. Clin Microbiol Rev. 1995; 8: 462-478.

(4.) Pfaller M, Wenzel R. Impact of the changing epidemiology of fungal infections in the 1990's. Eur J ClinMicriobiol Infect Dis. 1992; 11:287-291.

(5.) Beighton D, Ludford R, Clark DT, et al. Use of CHROMagar Candida medium for isolation of yeasts from dental samples. J Clin Microbiol. 1995; 33: 3025-3037.

(6.) Gales AC, Pfaller MA, Houston AK, et al. Identification of Candida dubliniensis based on temperature and utilization of xylose and alpha-methyl-D-glucoside as determined with the API 20C AUX and Vitek YBC systems. J Clin Microbiol. 1999; 37: 3804-3808.

(7.) Sancak B, Rex JH, Paetznick V et al. 2003. Evaluation of a method for identification of Candida dubliniensis bloodstream isolates. J Clin Microbiol. 2003; 41:489-491.

(8.) Schoofs A., Odds FC, Colebunders R et al. 1997. Use of specialised isolation media for recognition and identification of Candida dubliniensis isolates from HIV-infected patients. Eur J Clin Microbiol Infect Dis. 1997; 16:296-300.

(9.) Gutierrez J, Morales P, Gonzalez MA et al. Candida dubliniensis, a new fungal pathogen. J Basic Microbiol. 2002; 42: 207-227.

(10.) Tendolkar U, Taniwala S, Jog S et al. Use of a new medium--tobacco agar, for pigment production of Cryptococcus neoformans. Indian J Med Microbiol. 2003; 21:277-279.

(11.) Khan ZU, Ahmad S, Mokaddas E et al. Tobacco Agar, a new medium for differentiating Candida dubliniensis from Candida albicans. J Clin Micro biol. 2004; 42:4796-4798.

(12.) Cirak MY, Kalkanci A., Kustimur S. Use of molecular method in identification of Candida species and evaluation of fluconazole resistance. Mem Inst Cruz, Rio de Janeiro, 2003; 98: 1027-1032.

(13.) Nho S, Anderson MJ, Moore CB et al. Species differentiation by internally transcribed spacer PCR and HhaI digestion of fluconazole-resistant Candida krusei, Candida inconspicua, and Candida norvegensis strains. J Clin Microbiol 1997; 35:1036-1039.

(14.) Williams DW, Wilson MJ, Lewis MA et al. Identification of Candida species by PCR and restriction fragment length polymorphism analysis of intergenic spacer regions of ribosomal DNA. J Clin Microbiol. 1995; 33:2476-2479.

(15.) Wahyuningsih R, Freisleben HJ, Sonntag HG et al. Simple and rapid detection of Candida albicans DNA in serum by PCR for diagnosis of invasive candidiasis. J. Clin. Microbiol. 2000; 38:3016-3021.

(16.) Kirkpatrick WR, Revankar SG, McAtee RK et al. Detection of Candida dubliniensis in oropharyngeal samples from human immunodeficiency virus-infected patients in North America by primary CHROMagar Candida screening and susceptibility testing of isolates. J Clin Microbiol. 1998; 36: 902-904.

(17.) Sullivan, D., and Coleman, D. 1998. Candida dubliniensis: characteristics and identification. J. Clin. Microbiol. 36: 329-334.

Hiroshi Isogai, DVD [1] Andargachew Mulu, MD [2] Erimias Diro, MD [3] Henok Tekleselassie, MD [4] Afework Kassu, MD [2] Koichi Kimura, MD [5] Takeshi Nishikawa, MD [6] Emiko Isogai, DVD* [7]

[1] Animal Research Center, Sapporo Medical University, Sapporo, Hokkaido 060-8556, Japan,

[2] Department of Microbiology and Parasitology, College of Medicine and Health Sciences, University of Gondar, Gondar, Ethiopia,

[3] Department of Internal Medicine, Facility of Medicine, Addis Ababa University, Addis Ababa, Ethiopia,

[4] Department of Surgery, Facility of Medicine, Addis Ababa University, Addis Ababa, Ethiopia,

[5] Faculty of Human Science, Hokkaido Bunkyo University, Eniwa, Hokkaido 061-1449, Japan,

[6] Hokkaido University of Education, Ainosato-Sapporo, 002-8502

[7] Department of Disease Control and Molecular Epidemiology, Health Sciences University of Hokkaido, Ishikari-Tobetsu, Hokkaido 061-0293, Japan

* Address correspondence to Dr. Emiko Isogai, Department of Disease Control and Molecular Epidemiology, Health Sciences University of Hokkaido, Ishikari-Tobetsu 1757, Japan, E-mail:
Gale Copyright: Copyright 2010 Gale, Cengage Learning. All rights reserved.