Introduction
Oropharyngeal candidiasis (OPC) is the most common opportunistic infection during AIDS. It is associated with pain, limits food ingestion and can lead to cachexia. Fluconazole (FCZ) is commonly used for the treatment of OPC in AIDS patients because of its low toxicity, its pharmacokinetic properties that assure good saliva levels and a broad activity against yeasts. However, in vitro resistance to FCZ of Candida albicans isolates recovered from the oral cavities of HIV-infected patients who were clinically resistant to the drug has been frequently reported since the early 1990s [1-8]. FCZ resistance has been described as a rare phenomenon considering the large number of patients receiving FCZ prophylaxis or treatment world-wide.
The incidence of FCZ resistance has been reported to be approximately 10% among HIV-infected patients followed at the Pasteur Institute Hospital (B. Dupont, personal communication, 1994) and in other centres [5]. Resistance has been attributed to low-dose regimens (50 mg per day), and to prolonged and repeated courses of therapy in severely immunocompromised patients [2,5,6].
Decreased susceptibility to azoles has been attributed to diminished drug uptake, increased efflux, or a modification of the target enzyme, 14-α-demethylase [9]. Unlike bacteria, yeasts do not seem to inherit resistance by transfer of plasmids and, so far, researchers have failed to identify a genetic marker of resistance. They nevertheless demonstrated that resistant isolates were often genetically identical to pretreatment isolates [10-12] and that patients with AIDS could be infected with several isolates exhibiting various genotypes [12-14] and phenotypes [10,15].
The existence of innate resistance to FCZ has rarely been described for C. albicans [16,17] and has never been clearly demonstrated. This study was started when it was discovered that an HIV-positive patient suffering from his first episode of OPC failed to respond to FCZ, and his oral C. albicans isolate tested resistant to the drug. His long-time partner was an AIDS patient who had been repeatedly treated with FCZ and became FCZ-resistant over time. The question was asked whether, in addition to environmental pressure following FCZ intake, FCZ resistance could be spread by transmission of isolates between sexual partners. This occurrence would thus have explained the existence of FCZ-resistant isolates in patients who had never been treated with azoles.
Patients and methods
Patients
Twenty HIV-infected patients comprising 10 couples were studied. Amongst the 1200 HIV-infected patients followed at the Pasteur Institute Hospital, five male homosexual and five heterosexual couples were selected chronologically according to the following criteria: both partners were HIV-infected and both usually came to the consultation, and both were followed at our centre and agreed to participate in the study. The stability of the relationship was arbitrarily defined: a stable couple was defined as one that had remained faithful for at least 1 year. All participants were outpatients at the time of sampling and gave informed consent prior to entry into the study. Patients were evaluated for age, sex, degree of immunosuppression, number of previous episodes of OPC, previous treatments with azoles (cumulative dose), prior treatment failure with FCZ, arbitrarily defined by the physicians as the persistence of lesions (< 50% improvement) after more than 7 days of treatment with at least 100 mg of FCZ daily.
Both partners were sampled within 8 weeks of each other and usually on the same day. Subsequent samples were obtained from some patients during the following year, and from one patient an isolate that had been recovered 5 months earlier was also studied. A sterile cotton swab was drawn over the lesions or the oral cavity (cheeks and palate). Swabs were then rubbed onto Sabouraud dextrose-agar plates and the plates were incubated for 48 h at 28°C.
Isolates
From each sample, five colonies were selected from the primary culture plate and their purity was assessed by further subcloning. C. albicans was identified by chlamydospore formation on Rice-Tween agar. Species identification of the chlamydospore-negative colonies was based on their sugar assimilation pattern using ID32C commercial strips (BioMérieux, Marcy l'Etoile, France). Each colony (clone) was assigned a code number including the number corresponding to the couple (1-10), a letter for the individual (A and B), a number for the sample (1-4), and a letter for the clone (a-e). The clones and the initial sample were stored frozen at -20°C in phosphate-buffered saline containing 10% glycerol, a procedure that ensured viability of the isolates for several months (data not shown).
Antifungal susceptibility testing
Susceptibilities to FCZ and itraconazole (ITZ) was determined using a microbroth dilution method. Antifungal agents obtained as powders from the manufacturers (for FCZ, Pfizer Central Research, Sandwich, Kent, UK; for ITZ, Janssen Pharmaceuticals, Beerse, Belgium) were dissolved (640 μg/ml) in sterile distilled water for FCZ or dimethyl sulphoxide for ITZ, and then twofold dilutions were made in distilled water. Twenty microlitres of each drug dilution (columns 2-11) or distilled water (column 1 for the negative control and column 12 for the growth control) were then added to wells of a round-bottomed microculture plate. At this stage, plates were frozen for several weeks without modifying the results of the antifungal susceptibility tests (data not shown).
Yeasts were precultured for 24 h at 28°C on Sabouraud agar Petri dishes. Yeast suspensions were adjusted by turbidimetry (Hach Ratio XRE turbidimeter; Hach Co., Loveland, Colorado, USA) to 5 × 103/ml in RPMI-1640 medium (Sigma Chemical Co., St Louis, Missouri, USA), buffered with 0.165 mol/l morpholino-propanesulphonic acid (Sigma), as recommended by the National Committee for Clinical Laboratory Standards [18]. Medium alone (column 1) and yeast suspension (columns 2-12) were added in 180 μl aliquots giving final concentrations from 64 to 0.12 μg/ml for FCZ and from 16 to 0.03 μg/ml for ITZ. After 48 h of incubation at 35°C, minimum inhibitory concentrations (MIC) were determined spectrophotometrically at 492 nm as the concentration of the antifungal agent preventing 80% of the growth compared with the drug-free well. Resistance to FCZ and ITZ was defined as MIC ≥ 16 and ≥ 4 μg/ml, respectively. All the colonies recovered from each couple were studied simultaneously. MIC were also determined for the mixture of the five colonies for each sample. Each run included a quality control Candida kefyr isolate (no. 706) with known susceptibilities to FCZ and ITZ (FCZ MIC, 0.5-2 μg/ml; ITZ MIC, 0.06-0.25 μg/ml). MIC were determined by an observer who was unaware of the source of the isolates, and duplicate assays were performed for most of the isolates.
Restriction fragment length polymorphisms
The yeasts were cultured in Sabouraud broth overnight in a shaking incubator at 28°C. DNA extraction, digestion with the restriction enzymes EcoRI and HinfI, and electrophoresis were carried out as described previously [17]. The 1 kb DNA ladder (Gibco BRL, Life Technologies, Inc., Gaithersburg, Maryland, USA) was included in each gel as a molecular size standard. For each couple, the 20 digested DNA samples were analysed on the same gel. For the subsequent samples, at least one clone from the first sample of each patient was included for comparison (more in the case of heterogeneous flora).
The photographs of the ethidium bromide-stained gels were then examined to detect similarities and differences in banding patterns both visually and by using the Taxotron software package (P.A. Grimont, Institut Pasteur, Paris, France). The position and the number of the major bands were used for strain delineation. In a previous study comparing three genotyping methods [restriction fragment length polymorphism (RFLP) determination with EcoRI and HinfI digestion, arbitrarily primed polymerase chain reaction (PCR) and karyotyping], a complete agreement was shown between arbitrarily primed PCR and HinfI RFLP results [17]. Genetic identity between clones was defined as the presence of an equal number of bands of similar sizes after HinfI digestion, and confirmed by EcoRI patterns. Samples were considered to be identical when at least one clone was shared by both partners. Data were analysed blindly. Reproducibility of the RFLP patterns was checked twice with the first DNA extraction from all clones and for some clones that had been subjected to a second DNA extraction.
Statistical analysis
Clinical and experimental data were recorded using EpiInfo software (Centers for Disease Control, Atlanta, Georgia, USA; World Health Organization, Geneva, Switzerland). Comparisons were made using the Mann-Whitney U test or Fisher's exact test, where required.
Results
Characteristics of the patients studied
At entry, 11 patients had AIDS. The median CD4+ T-lymphocyte count was 56 × 106/l (range, 8-800 × 106/l). Twelve patients were asymptomatic carriers of yeasts (no mucosal lesions and no complaints) and eight had OPC, six of whom were being treated at the time of sampling (three with FCZ, one with ketoconazole, one with ITZ, one with miconazole). The clinical history of each patient is summarized in Fig. 1 and Table 1.
Antifungal susceptibility test results for the first sample from each patient
MIC varied widely from 1 to ≥ 64 μg/ml for FCZ (median, 32 μg/ml) and 0.015 to ≥ 16 μg/ml for ITZ (median, 0.25 μg/ml). Thirteen patients (53 clones) were colonized or infected with clones exhibiting high FCZ MIC. Five samples contained clones cross-resistant to ITZ (total, 12 clones). Samples from some patients (4A, 7A, 7B and 9A), contained FCZ-susceptible and -resistant clones, with a majority of FCZ-susceptible clones. When the median MIC for the clones and the MIC determined on the pooled clones were low, the sample was considered to be FCZ-susceptible.
At the end of this analysis, nine patients were considered to be infected (seven patients had lesions) or colonized (two asymptomatic carriers) by FCZ-resistant clones. These patients suffered from more severe immunodepression than the others (median CD4+ cells counts, 33 versus 211 × 106/l in patients with susceptible flora; P = 0.002, Mann-Whitney U test). For six patients, FCZ therapy failed, although it did not fail for patients with susceptible flora.
DNA typing results for the first sample from each patient
HinfI digestion generated a few major bands (one to three bands in the range of 4-6 kb) whereas EcoRI digestion produced a continuum of bands from 20 to 0.5 kb with a few major bands (Fig. 2).
RFLP analysis showed genetic diversity among the C. albicans clones recovered from the 20 HIV-infected patients with 14 and 17 different patterns observed with EcoRI and HinfI, respectively (Table 1). HinfI digestion revealed more than one genotype in samples from six patients (patients 2B, 4A, 4B, 7B, 8A and 8B; Fig. 2). Thus, genetic diversity was not restricted to patients who had clones exhibiting variable MIC. The patterns obtained with both enzymes showed that one genotype was characteristic of a given patient or a given couple, except for pattern E14/H17, which was found for clones recovered from patients 9B, 10A and 10B. No further information as to a possible relationship between patient 9B and patients 10A/10B was available. For six of the 10 pairs of samples studied (couples 1, 4, 5, 7, 8 and 10), the partners shared at least one identical clone (Table 1).
Evolution of the oral flora over time
Except for couple 2, all patients were sampled at least once more, up to a total of three times, either systematically during routine follow-up or after specific complaints from one of the partners.
In addition to the clones recovered from both partners and exhibiting genetic identity in the first sample (couples 1, 4, 5, 8 and 10), patients 6A and 6B shared the same clones within 1 month of the first sampling (Fig. 1). Thus, at the end of the study, the clones recovered from six couples showed genetic identity over a long period of time (three heterosexual couples: 1, 4 and 8; and three homosexual couples: 5, 6 and 10) and transiently for one heterosexual couple (couple 7), and concerned 11 of the 21 pairs of isolates studied (Fig. 1). Among the six couples with a prolonged sharing of isolates, four comprised patients with CD4+ cell counts less than 50 × 106/l.
Origin of in vitro FCZ resistance
As noted above, after the first sampling nine patients were considered to be colonized or infected with FCZ-resistant clones. Of these, only four patients (1A, 3B, 5A and 6B) had previously been repeatedly treated with azoles, whereas the remaining five patients (1B, 5B, 6A, 10A and 10B) had not (Table 1). However, the cumulative dose of azole intake for some of their partners (1A, 5A or 6B) was 10 g or greater. In these three cases, genotyping showed that the pairs of samples contained identical clones that were resistant to FCZ in vitro. The patient who seemed to have acquired the resistant isolates from his partner had clinical lesions at the time of most samplings, and was still colonized or infected with the acquired clones months after the death of the partner (Fig. 1).
Discussion
With an increasing number of immunocompromised patients, such as HIV-infected patients, or living together in the same environment in hospital wards, interhuman transmission of pathogenic fungi is likely to occur frequently. However, it is a rare event and has only recently been demonstrated by molecular typing methods for nosocomial candida infections in patients at risk for candidaemia [17,19,20]. These studies uncovered sources of infections (infusions, health-care workers) and led to the improvement of prophylactic programmes. Studies on the spread of superficial mycoses focused on sexual partners, and the transmission of isolates during superficial candidiasis was recently demonstrated in patients with recurrent vaginitis. Infecting and male partner strains were compared, and eight of the 10 paired strains were found to be identical or far more similar to one another than 'extra-couple' strains [21], thereby providing an explanation for the recurrence of the infection.
Transmission of isolates during OPC did not represent a threat until azole resistance developed. A few groups have reported C. albicans strain identity between sexual partners after sampling the oral cavity [14,22]. Different molecular typing methods showed that the close relatedness between the isolates shared by both partners could only be the result of transmission because of the known genetic diversity among oral isolates [13,15,23]. Our study confirmed these results using a different typing method that has proven effective to delineate C. albicans clones [17]. These results were somewhat contradictory to those reported by Schmid et al. [24], who reported that strains from AIDS patients were genetically less diverse than control strains. They discussed the possible selection of strains more adapted to the oral cavity and replacing over time the commensal flora in immunodepressed patients. The clustering of isolates in their study could also have been due to the inclusion of unidentified sexual partners.
Transmission cannot be demonstrated unless different isolates are found at time zero, with subsequent acquisition of an isolate shared by both partners and absence of recovery of this isolate from an environmental source such as food or water (a very unlikely source of contamination for OPC). It was not possible to exclude the fact that both the genetic identity of the clones and a similar drug susceptibility pattern could not have occurred by chance, erroneously suggesting that one of the clones was transmitted from one partner to the other. However, knowing that a patient was usually infected by his/her own isolate, that a given isolate was specific to a given individual, and that resistance to FCZ has rarely been observed in patients never treated with azoles, the most likely explanation for our finding of resistant isolates shared by both partners (one of whom had never been treated with azoles) was transmission, and our results are therefore discussed accordingly.
The frequency of genetic identity of oral isolates of C. albicans between HIV-positive partners was higher in our study (seven out of 10 couples) than in a previous study (five out of 19 couples) [14]. This increase could reflect differences in the populations examined (high proportion of patients with advanced AIDS in our study) or the fact that five clones per patient were studied in order to take into account the known diversity of the buccal flora. This approach probably increased the chance of finding a minor clone shared by both partners. The possibility of a lower discriminatory power of our genotyping method was ruled out by our previous study [17]. In any case, the frequency and timing of oral flora replacement by that of the sexual partner (HIV-infected or not) could only be evaluated in a prospective study.
The major question was whether transmitted isolates could be responsible for infections or were just colonizing flora. In their study, Boerlin and collaborators were unable to find an association between transmission and previous or present OPC [14]. The virulence of the acquired isolates in our patients was suggested by the fact that they were recovered from lesions in some patients (1B, 5B, 6A and 10A). Isolate replacement by that of the partner was thus often associated with infection in our study. Why these isolates seemed more virulent in these patients than their own flora remains to be determined. Selection of more virulent isolates may have occurred after treatment with azoles or during the course of HIV infection, as recently reported on a subset of isolates [25], and this possibility was repudiated by other authors [24,26]. Further studies are needed to better identify the virulence factors of oral isolates.
Another issue was the virulence of C. albicans isolates resistant to FCZ. The combination of clinical history and typing results (MIC and RFLP) clearly showed that FCZ-resistant isolates acquired from the partner could be identified as the infection-causing organism. These findings represent the first evidence that resistance to FCZ is not always associated with decreased virulence, as one might have extrapolated from experimental models of C. albicans [27] and Cryptococcus neoformans [28] infections in mice, which demonstrated the opposite. If resistant isolates could be transmitted and could cause infection, any attempt to prevent the emergence of FCZ resistance during the treatment of OPC in HIV-infected patients should have benefited not only the ailing patients but also their partners. The acquisition of an FCZ-resistant isolate could be even more damaging for the patient when cross-resistance to other azoles limits the choice of treatment. Although this is not yet a big issue, two of the nine patients with homogeneously FCZ-resistant flora, including one patient who seemed to have acquired the resistant clone from her partner (patient 1A), had clones resistant to ITZ.
In conclusion, our study suggests that C. albicans oral isolates could be transmitted between sexual partners and that replacement of the oral flora, especially a resistant clone, could cause OPC in HIV-infected patients. Our results also indicate that, without carefully reviewing the clinical and personal history of the patient, FCZ resistance could not be reported for patients never treated with azoles. In addition to FCZ resistance following drug intake, another possible explanation was interhuman transmission of resistant isolates. This possibility underlines the urgent need for antifungal susceptibility tests able to predict the emergence of resistance to be performed before treatment efficacy is affected or the clones has spread to other patients.
References
1. Cameron ML, Schell WA, Bruch S, Bartlett JA, Waskin HA, Perfect JR: Correlation of in vitro fluconazole resistance of Candida isolates in relation to therapy and symptoms of individuals seropositive for human immunodeficiency virus type I. Antimicrob Agents Chemother 1993, 37:2449-2453.
2. Chavanet P, Lopez J, Grappin M, et al.: Cross-sectional study of the susceptibility of Candida isolates to antifungal drugs and in vitro-in vivo correlation in HIV-infected patients. AIDS 1994, 8:945-950.
3. Coleman DC, Bennett DE, Sullivan DJ, et al.: Oral Candida in HIV infection and AIDS: new perspectives/new approaches. Crit Rev Microbiol 1993, 19:61-82.
4. Garcia-Hermoso D, Dromer F, Improvisi L, Provost F, Dupont B: Fluconazole concentrations in saliva from AIDS patients with oropharyngeal candidosis refractory to treatment with FCZ. Antimicrob Agents Chemother 1995, 39:656-660.
5. Newman SL, Flanigan TP, Fisher A, Rinaldi MG, Stein M, Vigilante K: Clinically significant mucosal candidiasis resistant to fluconazole in patients with AIDS. Clin Infect Dis 1994, 19:684-686.
6. Ruhnke M, Eigler A, Geiseler B, Engelmann E, Trautmann M: Emergence of fluconazole-resistant strains of Candida albicans in patients with recurrent oropharyngeal candidosis and human immunodeficiency virus infection. J Clin Microbiol 1994, 32:2092-2098.
7. Baily GG, Perry FM, Denning DW, Mandal BK: Fluconazole-resistant candidosis in an HIV cohort. AIDS 1994, 8:787-792.
8. Heinic GS, Stevens DA, Greenspan D, et al.: Fluconazole-resistant Candida in AIDS patients. Oral Surg Oral Med Oral Pathol 1993, 76:711-715.
9. White A, Goetz MB: Azole-resistant Candida albicans: report of two cases of resistance to fluconazole and review. Clin Infect Dis 1994, 19:687-692.
10. Bart-Delabesse E, Boiron P, Carlotti A, Dupont B: Candida albicans genotyping in studies with patients with AIDS developing resistance to fluconazole. J Clin Microbiol 1993, 31:2933-2937.
11. Millon L, Manteaux A, Reboux G, et al.: Fluconazole-resistant recurrent candidiasis in human immunodeficiency virus-positive patients: persistence of Candida albicans strains with the same genotype. J Clin Microbiol 1994, 32:1115-1118.
12. Redding S, Smith J, Farinacci G, et al.: Resistance of Candida albicans to fluconazole during treatment of oropahryngeal candidiasis in a patient with AIDS: documentation by in vitro susceptibility testing and DNA subtype analysis. Clin Infect Dis 1994, 18:240-242.
13. Barchiesi F, Hollis RJ, McGough DA, Scalise G, Rinaldi MG, Pfaller MA: DNA subtypes and fluconazole susceptibilities of Candida albicans isolates from the oral cavities of patients with AIDS. Clin Infect Dis 1995, 20:634-640.
14. Boerlin P, Addo M, Durussel C, Pagani JL, Chave JP, Bille J: Transmission of oral Candida albicans strains between HIV-positive patients [letter]. Lancet 1995, 345:1052-1053.
15. Dellion S, Improvisi L, Dupont B, Dromer F: Diversity among yeasts isolated from patients with AIDS and oral candidosis, especially regarding fluconazole susceptibility. J Mycol Méd 1995, 5:98-101.
16. Goff DA, Koletar SL, Buesching WJ, Barnishan J, Fass RJ: Isolation of fluconazole-resistant Candida albicans from human immunodeficiency virus-negative patients never treated with azoles. Clin Infect Dis 1995, 20:77-83.
17. Bart-Delabesse E, van Deventer H, Goessens W, et al.: Contribution of molecular typing methods and antifungal susceptibility testing to the study of a candidemia cluster in a burn care unit. J Clin Microbiol 1995, 33:3278-3283.
18. National Committee for Clinical Laboratory Standards: Reference Method for Broth Dilution Antifungal Susceptibility Testing of Yeasts. Proposed Standard. [Document M27-P.] Villanova: National Committee for Clinical Laboratory Standards; 1992.
19. Voss A, Pfaller MA, Hollis RJ, Rhine-Chalberg J, Doebbeling BN: Investigation of Candida albicans transmission in a surgical intensive care unit cluster by using genomic DNA typing methods. J Clin Microbiol 1995, 33:576-580.
20. Romano F, Ribera G, Giuliano M: A study of a hospital cluster of systemic candidosis using DNA typing methods. Epidemiol Infect 1994, 112:393-398.
21. Schmid J, Rotman M, Reed B, Pierson CL, Soll DR: Genetic similarity of Candida albicans strains from vaginitis patients and their partner. J Clin Microbiol 1993, 31:39-46.
22. Sangeorzan JA, Bradley SF, He X, et al.: Epidemiology of oral candidiasis in HIV-infected patients: colonization, infection, treatment, and emergence of fluconazole resistance. Am J Med 1994, 97:339-346.
23. Pfaller MA, Rhine-Chalberg J, Redding SW, et al.: Variations in fluconazole susceptibility and eletrophoretic karyotype among oral isolates of Candida albicans from patients with AIDS and oral candidiasis. J Clin Microbiol 1994, 32:59-64.
24. Schmid J, Odds FC, Wiselka MJ, Nicholson KG, Soll DR: Genetic similarity and maintenance of Candida albicans strains from a group of AIDS patients, demonstrated by DNA finger-printing. J Clin Microbiol 1992, 30:935-941.
25. McCullough M, Ross B, Reade P: Characterization of genetically distinct subgroup of Candida albicans strains isolated from oral cavities of patients infected with human immunodeficiency virus. J Clin Microbiol 1995, 33:695-700.
26. Hellstein J, Vawter-Hugart H, Fotos P, Schmid J, Soll DR: Genetic similarity and phenotypic diversity of commensal and pathogenic strains of Candida albicans isolated from the oral cavity. J Clin Microbiol 1993, 31:3190-3199.
27. Dupont B, Improvisi L, Eliaszewicz M, Pialoux G, GEMO: Resistance of Candida albicans to fluconazole (FCZ) in AIDS patients. 32nd International Conference on Antimicrobial Agents and Chemotherapy. Anaheim, October 1992 [abstract 1203].
28. Velez JD, Allendoerfer R, Luther M, Rinaldi MG, Graybill JR: Correlation of in vitro azole susceptibility with in vivoresponse in a murine model of cryptococcal meningitis. J Infect Dis 1993, 168:508-510.
© Lippincott-Raven Publishers.