Introduction
Oral candidiasis (OC) has been closely linked to human immunodeficiency virus (HIV) infection since the pandemic began 1,2. It has become an important diagnostic and prognosis parameter, as well as a marker of the success of highly active antiretroviral therapy 3,4.
Extensive research has been done on OC, and prophylactic treatments developed and implemented for treatment eradication, but it continues to be the most frequent opportunistic oral infection in HIV+ patients, regardless of treatment with highly active antiretroviral therapy (4,5.
Candida species, mainly C. albicans, are the causal agents of OC. Normally commensal dimorphic fungi, under specific local and systemic circumstances they can become pathogens (6,7. The Candida species that colonize the oral cavity of HIV+ subjects exhibit greater antimycotic treatment resistance than strains from HIV-seronegative subjects. Resistance to antimycotic drugs is considered an important risk factor contributing to antimycotic resistance and consequent development of deep mycoses and candidemia 8.
Although C. albicans has a clonal mode of reproduction, phenotypic and genotypic traits can distinguish between strains 9. Strain-typing techniques have been developed to generate epidemiological data on C. albicans. These describe the behavior of different strains in distinct patient groups, thus identifying associations between genotypes and phenotypical traits, mainly antibiotic resistance and mechanisms of pathogenicity 9-13.
Genotype and phenotype variation in Candida species is a possible cause for increased antifungal resistance with consequent treatment failure and persistent infection. This suggests that oral candidiasis in HIV+ patients may be caused by different clones of the same species (12,14. Little data is available on the presence of C. albicans clones with different resistance profiles colonizing the oral cavity of HIV+ subjects. In addition, it is still under debate if certain C. albicans strains exhibit variation in their virulence capabilities.
The objective of this study was to use resistotyping to identify possible phenotypic differences between C. albicans strains isolated from the oral cavity of HIV+ and HIV-seronegative patients.
Methods
Ethics statement
The research protocol was approved by the bioethics committee of the Faculty of Dentistry of the Autonomous University of Yucatan (Universidad Autónoma de Yucatán [UADY]) (code FOCAI 04/02). It was classified as minimal risk in accordance with the Health Research Regulation, Article 17, General Health Law, Health Department, Government of Mexico.
Strain isolation and recovery
A total of 92 consecutive clinical isolates of oral C. albicans strains were obtained from oral swabs subsequently frozen at -80°C in heart infusion broth (Becton Dickinson, Mexico) containing 10% glycerol (Sigma-Aldrich, USA). Isolates were collected from 37 healthy subjects (≥ 21 years old) lacking clinical lesions indicating any clinical form of oral candidiasis, and from 55 HIV+ patients (>18-<50 years old) in the O’Horan General Hospital, Merida, Yucatan, Mexico. Inclusion criteria for the HIV+ patients were that they must have been asymptomatic for any clinical form of oral candidiasis at the time of sampling, and not have received any antifungal or antibacterial treatment at least six months prior to oral examination.
Strains were sown on CHROMagar Candida
(Becton Dickinson Microbiology Systems, USA) and incubated at 37ºC for 48 h. Isolates were identified by standard morphology.
Molecular identification
DNA EXTRACTION
Strains were grown on Sabouraud dextrose agar (SDA) (Becton Dickinson and Company, USA) for 48 hours at 37°C, suspended in 500 μL sterile distilled water, heated for 15 min at 95°C and then immediately frozen at -70°C for 15 min. After thawing to room temperature samples were centrifuged for 5 min at 16,000 x g, the supernatant placed in a sterile Eppendorf tube and stored at -20°C until analysis (6,15).
PCR
Two pairs of oligonucleotides (Invitrogen, USA) were simultaneously used in each PCR test: the first pair, specific to C. albicans, amplifies a 175 bp fragment of the 25S rRNA gene: CAL5 (5'-TGTTGCTCTCTCGGGGGCGGCCG-3') and NL4CAL (5'-AAGATCATTATGCCAACATCCTAGGTA/TAA-3'). The second pair, present in all Candida species, amplifies a 610 bp fragment of the 25S rRNA gene: RNAF (5'-GCATATCAATAAGCGGAGGAAAAG-3') and RNAR (5'-GGTCCGTGTTTCAAGACG-3'). PCR was done following Yang et al. (16) as described in Hernández-Solís et al.11.
Resistotyping
Resistogram biotyping of C. albicans strains was done following the McCreight and Warnock method as modified by Nakamura et al. 17. Briefly, stock solutions of boric acid, cetrimide, sodium periodate and sodium selenite (Sigma-Aldrich, USA) were prepared at a 20 mg/mL concentration. A silver nitrate (Sigma-Aldrich, USA) stock solution was prepared at a 2 mg/mL concentration.
The resistance profile was assessed based on different concentration series of each stock solution added to SDA: boric acid 1.15, 1.3, 1.45 and 1.6 mg/mL; cetrimide 0.06, 0.08, 0.1 and 0.12 mg/mL; silver nitrate 0.0075, 0.01, 0.0125 and 0.015 mg/mL; sodium periodate 0.01, 0.02,
0.03 and 0.04 mg/mL; and sodium selenite 0.1, 0.2, 0.3 and 0.4 mg/mL. Candida albicans (5 μL CFU), previously incubated for 24 h at 37°C and with an optical density ranging from 0.45 to 540 nm, were inoculated onto the agar plates containing one of the chemical reagents and incubated for 40h at 37°C.
Resistance was tested using a concentration for each stock solution that had exhibited clear differentiation between the studied strains: 0.3 mg/mL for sodium selenite; 1.6 for boric acid;
0.6 for cetrimide; 0.03 for sodium periodate; and 0.015 for silver nitrate (Table 1). Growth patterns were described following the criteria suggested by Khan et al. (18): confluent growth (resistance) with capital letter identifying reagent; non-confluent growth (sensitivity) with lowercase letter; and absence of growth indicated by a hyphen (-).
Results
Control group. Thirty-seven C. albicans isolates were collected from the HIV-seronegative patients. From these, 74 clones were obtained (2 clones per strain) and seven different resistotypes identified (Table 2). The most frequent (n=48 clones; 64.8%) resistotype profile was --CDE, followed by -bCDE (n= 8; 10.8%) and ---DE (n= 6 cases; 8.1%). One strain in this study group had two clones with different resistotype profiles.
HIV+ group. Three clones were obtained from each of the 55 C. albicans isolates from this group. Sixteen clones were lost during processing, leaving 149 clones to determine the resistotypes. Eleven different resistotype profiles were identified in this study group. The most frequent (n=70; 47%) was ABCDE, followed by A-CDE (n=37; 24.9%) and --CDE (n= 14; 9.4%). Twenty oral C. albicans strains were identified with two different resistotype profiles. The remaining strains had clones with the same resistotype profile (Table 3).
The frequency of the serotype --CDE was significantly higher in the HIV-seronegative group (64.8% versus 9.4%, p<0.05).
Discussion
Candida strain pathogenicity has been correlated to certain resistotypes 19. Variations in C. albicans clone biotype are known to exist in recurrent oral candidiasis in HIV+ patients (17,20. Antifungal therapy can lead to replacement of an initial biotype by a new one, even one resistant to antifungal drugs. Though infrequent, shifts in the resistogram biotypes of oral C. albicans isolates may also occur to a certain extent in normal subjects 17.
Based on this phenomenon it is possible that differences may be linked to selection of a more virulent switched phenotype, or that a switched phenotype of the same strain can exhibit variations in virulence in response to changing environmental conditions (18,21.
This being the case, C. albicans populations are mainly clonal in origin with gene content being apparently constant and stable between strains (9,22. Phenotypic differences between C. albicans strains are therefore probably due mainly to changes in expression levels of associated genes, and/or minimal modifications in their sequences, which may influence variations in the function of their encoded proteins (9). Determining if changes in genetic expression are a possible cause of the presence of different C. albicans strains in the oral cavity of HIV+ patients is beyond the scope of the present study and provides ample opportunity for future research. This will need to focus on identifying what factors are associated with changes in C. albicans phenotypes and biotypes, concentrating on these strains’ intergenic regions and protein coding sequences 9.
Resistotype --C-- is reported to be the most common of the highly pathogenic C. albicans isolates 23. This does not coincide with the present results in which only two strains exhibited this resistotype; the rarity of --C-- resistotypes has been reported elsewhere 23. In a study of sixteen different resistograms from 198 oral isolates from 22 normal subjects 24, it was found that a particular strain tends to persist in the oral cavity of normal subjects although some changes can occur in these C. albicans biotypes. Differences in oral colonization by Candida species may be strongly influenced by ethnic, geographical and sociodemographic factors (19,25,26.
A broad range of fungi-typing techniques currently exist with variations in throughput, cost, processing time and discriminatory power. What method is chosen will depend largely on the epidemiological data needed and available laboratory infrastructure (10, 12,27. Several technologies are currently available for biotyping and genotyping of C. albicans strains, molecular methods being the most precise methods for yeast biotyping (9,28-30. However, molecular biology methods are more expensive than phenotypic methods and are often only available in research or epidemiological reference laboratories (19,30). Phenotyping methods such as auxonotyping, enzymotyping and resistotyping may be easy to run but do not have the discriminatory power of molecular methods (19,29,31. Recurrent C. albicans isolates have been shown to exhibit a high frequency of morphological variation that is not related to genomic DNA fingerprinting 32.
The resource limitations of routine laboratories can make genotyping unfeasible, and morphotyping tests more promising (18, 19). However, discrimination between strains and results reproducibility require molecular typing methods to compare and confirm results. Most genotypic methods are expensive and involve highly specific infrastructure, as well as highly trained laboratory staff. This is a vital concern in developing and undeveloped countries, which have the highest HIV+/AIDS prevalence, since it prevents use of these highly specific techniques for routine strain identification. The resistotyping method used in the present study is an easy-torun, comparatively inexpensive method applicable to all C. albicans strains. It is a near ideal system in developing countries for phenotypic and genotypic identification of Candida strains since it allows for testing a large number of isolates, provides rapid results, involves relatively inexpensive procedures, and can be automated.
The usefulness of resistotyping for phenotypic discrimination of C. albicans strains is well known (18, 23, 31), as are the problems of using phenotypic tests, such as lengthy time requirements, difficulty in automating and inaccurate strain identification 33. The resistotype technique has been successful in identifying different C. albicans strains from areas other than the oral cavity. It has also been used to identify four different C. albicans resistotypes (-B- -F- was the most common) from the respiratory tract of TB-positive and TB-negative patients 34, and to identify C. albicans resistotypes from feces and vaginal mucus (A-C-F-, ABC-F- and A- - -F- were the most frequent) 23. One promising approach for improving results accuracy is to use different analyses in conjunction (e.g. resistograms in addition to an API system, or Odds and Abbott’s method) to increase discriminatory capacity beyond what each technique can offer individually (18, 31).
Conclusions
In this study, we have demonstrated the differences between strains isolated from HIV+ and seronegative patients. Resistotyping is easy to perform in laboratories with lower economic resources, and useful for epidemiological purposes.
Footnote
Funding/support
This study was supported in part by PADECCA 2015 and PRODEP.
Conflict of interest
The authors declare no conflict of interest.