Genotyping and antifungal susceptibility testing of Cryptococcus neoformans isolates from Cameroonian HIV-positive adult patients

Authors


Corresponding author: M. Mallié, Laboratoire de Parasitologie et Mycologie Médicale, UFR de Pharmacie, 15 Avenue Charles Flahault, 34093 Montpellier Cedex 05, France
E-mail: michele.mallie@univ-montp1.fr

Abstract

Cryptococcus neoformans is the most common cause of meningitis amongst adult Africans with HIV/AIDS. The widespread use of fluconazole may lead to the emergence of isolates with reduced susceptibility. We studied C. neoformans isolates from HIV-infected patients with cryptococcal meningitis. Genotyping and antifungal testing were performed to assess the genetic diversity, occurrence of mixed infections and in vitro activity of antifungal agents. Isolates were recovered from cerebrospinal fluid prior to systemic antifungal treatment. Six isolates were studied for each sample (a total of 114 isolates from 19 patients). Serotyping was performed via LAC 1 and CAP 64 gene amplification and genotyping was performed using phage M13 core, (GACA)4 and (GTG)5 primers and restriction polymorphism analysis of the URA5 gene. Susceptibilities for amphotericin B, flucytosine, fluconazole, voriconazole and posaconazole were tested by the Sensititre YeastOne® method. All strains were identified as C. neoformans var. grubii serotype A. We identified nine major genotypes. Up to two genotypes were identified in the same sample. None of the isolates were resistant to the studied drugs. However, 13 of 114 strains exhibited a reduced susceptibility to fluconazole and 13 of 114 strains exhibited a reduced susceptibility to flucytosine. No correlation was found between the genotype and susceptibility. This study confirms the prevalence of C. neoformans serotype A in Cameroon. Two genotypes may be responsible for a single episode of cryptococcosis. The possibility of mixed infection and diminished susceptibility to fluconazole or flucytosine must be considered for the management of cryptococcosis.

Introduction

Cryptococcal meningitis is one of the most common life-threatening opportunistic fungal infections in immunocompromised patients, particularly among those with AIDS in sub-Saharan Africa [1]. Cryptococcus neoformans represents the main microorganism that causes meningitis in African adults with HIV/AIDS infection. [2]

Two distinct varieties of C. neoformans have been described and are associated with three serotypes: C. neoformans var grubii (serotype A), C. neoformans var neoformans (serotype D) and AD hybrids [3]. C. gattii, another species of the genus Cryptococcus, consists of serotypes B and C [3] and is also capable of forming hybrids with C. neoformans [4]. Molecular typing has resulted in the further subdivision of these two species into eight major molecular types: VNI and VNII (serotype A; var grubii), VNIII (serotype AD; var neoformans), VNIV (serotype D; var neoformans), VGI, VGII, VGIII and VGIV (serotypes B and C; var gattii) [5]. Within C. neoformans var grubii, the VNI molecular type is predominant worldwide [6].

Several molecular typing methods have been used in the study of the epidemiology of C. neoformans. The most commonly used approaches involve amplified fragment length polymorphism (AFLP) analysis, PCR fingerprinting and/or PCR-restriction fragment length polymorphism (RFLP) analysis, as well as mating and/or serotype-specific PCR [5,7,8]. Microsatellites are popular molecular typing targets because they provide cost-effective genotyping with rapid turnaround times and a high discriminatory power. Similar to multilocus sequence typing (MLST) data, microsatellite typing data are transportable and exchangeable between laboratories [9].

Differences in biology, epidemiology, pathogenicity, clinical features and drug susceptibility have been associated with species, varieties and molecular types in the genus Cryptococcus [10,11]. Moreover, mixed infection by different molecular types within a given patient has recently been reported [12].

Different treatment strategies are used in patients with cryptococcal meningitis. Amphotericin B with or without flucytosine remains the ‘reference standard’ antifungal for induction therapy. A high oral dose of fluconazole with flucytosine is not as effective as amphotericin B associated with flucytosine [13]. In Cameroon, fluconazole is the most commonly administered drug for the treatment of cryptococcosis.

The objective of the present study was to determine the existence and the occurrence of mixed Cryptococcus infections in HIV-positive patients from Cameroon. To achieve this objective, we investigated the genetic diversity of six different isolates in a single sample for each patient using several molecular typing methods. In addition, the in vitro antifungal susceptibility of these different isolates was evaluated for the improved management of cryptococcal disease and investigation of potential antifungal drug resistance.

Materials and Methods

Yeast isolates

This study was a prospective study of 23 patients who exhibited neuromeningeal symptoms with a suspicion of cryptococcosis. The patients included in the study were HIV positive and hospitalized in the Hôpital Central, Yaoundé. None of them received a systemic antifungal treatment prior to the study. Cryptococcosis was diagnosed and confirmed at the Laboratory of Microbiology and Hôpital de Jour, Hôpital Central, Yaounde, by positive India ink stains, cryptococcal antigen latex agglutination slide tests (Crypto La-test® Fumouze Diagnostics, Levallois-Perret, France) in the cerebrospinal fluid (CSF) and/or cultures associated with urea-indole test. Only 19 of 23 samples exhibited a positive culture for Cryptococcus for the CSFs collected from patients. The average age was 38.1 years, with a range between 25 and 56 years. Cultures were encoded with the first letters of patient’s last and first names. The identification of each isolate after culturing was performed twice using a commercial identification kit (ID32C, Biomérieux). Each isolate was stored on cryobeads as recommended by the manufacturer. Six isolates were studied for each sample (five isolated colonies from a culture plus the entire culture) for a total of 114 isolates from the 19 samples that yielded positive cultures. This procedure was especially designed for the present study. All isolates were culture-purified and grown for 3 days on yeast extract potato dextrose agar prior to use.

Every patient included in this study was informed of the risks and gave his or her consent. The present study received the approval of the Cameroonian Human Investigations Committee and was supported by SIDACTION.

Susceptibility testing

The Sensititre antifungal susceptibility method was performed according to the manufacturer’s instructions. A final microorganism concentration of 1.5–8 × 103 CFU/mL was recommended. The final inoculum in the broth was inoculated onto YeastOne® plates within 15 min. Twenty microlitres of the yeast suspension were transferred into 11 mL of YeastOne® inoculum broth. Plates containing serial two-fold dilutions of the antifungal agents across 11 to 12 dilutions were inoculated using the prepared inoculum and incubated (35°C, 48 h or 72 h). The minimal inhibitory concentration (MIC) was recorded as the lowest concentration of antifungal agents that prevented the development of a red colour. Each test consisted of a microtitre plate, which contained dried serial dilutions of the eight antifungal agents, posaconazole (0.008–8 mg/L), amphotericin B (0.008–16 mg/L), fluconazole (0.125–256 mg/L), itraconazole (0.008–16 mg/L), ketoconazole (0.008–16 mg/L), flucytosine (0.03–64 mg/L), voriconazole (0.008–16 mg/L) and caspofungin (0.008–16 mg/L) in individual wells. The wells also contained a colorimetric indicator, which improves the end point readability according to a colour change from blue to purple. MICs for the Sensititre YeastOne® assay were read after 48 h of incubation. The Clinical and Laboratory Standards Institute M27-A3 protocol (CLSI M27-A3) was used for the MIC interpretative guidelines [14–16]. The interpretive MIC criteria for fluconazole were as follows: susceptible (S) ≤8 mg/L, susceptible-dose-dependent (SDD) 16–32 mg/L and resistant (R) ≥ 64 mg/L. The interpretive MIC criteria for flucytosine were as follows: S ≤ 4 mg/L, intermediate (I) 8–16 mg/L and R ≥ 32 mg/L. The interpretive MIC criterion for voriconazole (S ≤ 1 mg/L) was based on the work of Pfaller et al. [17]. For Cryptococcus, interpretative criteria have not yet been defined for posaconazole, amphotericin B, itraconazole and ketoconazole. Caspofungin is not active against Cryptococcus and thus was not considered.

The test organisms included two American Type Culture Collection (ATCC) strains that have been established as quality control strains (Candida parapsilosis ATCC 22019 and Candida krusei ATCC 6258) by the CLSI [14]. These isolates were tested between 5 and 10 times, with ≥98% of the MICs being within the reference ranges.

Geometric means MICs (G-MICs) were determined for each drug that was tested (except for caspofungin, which is not active against C. neoformans).

PCR fingerprinting

Primers for the minisatellite-specific core sequence of the wild-type phage M13 (5′-GAGGGTGGCGGTTCT-3′) and the microsatellite-specific sequences (GTG)5 and (GACA)4 were used as single primers in the PCR procedure, as described by Meyer et al. [10] Briefly, the amplifications were performed in a final volume of 50 μL containing 40.5 μL of sterile deionised water, 50 ng of DNA, 5 μL of 1× PCR buffer (10 mM Tris-HCl, pH 8.3, 50 mM KCl, 1.5 mM MgCl2 and 0.01% gelatine) (Sigma-Aldrich, Saint-Quentin-Fallavier, France), 0.2 mM of each dNTP, (Sigma-Aldrich, Fr), 20 μM primer and 2.5 U Taq DNA Polymerase from Thermus aquaticus (Sigma-Aldrich). Each PCR was performed in a Primus 96 plus (MWG AG Biotech, Ebersberg, Germany) thermal cycler, using the following conditions.

  • M13 and (GTG)5: initial denaturation (94°C, 3 min), 40 cycles of denaturation (94°C, 30 s), annealing (50°C, 1 min) and extension (72°C, 1.5 min) and a final extension step at 72°C for 6 min.

  • -(GACA)4: the microsatellite-specific primer (GACA)4 (5′ GACAGACAGACAGACA 3′) was used as the only primer in the PCR reactions [18]. The amplification conditions were as follows: initial denaturation (94°C, 3 min), 40 cycles of denaturation (94°C, 30 s), annealing (40°C, 1 min) and extension (72°C, 1.5 min) and a final extension step at 72°C for 7 min.

    • Amplification products were separated by electrophoresis in a 6% Poly(NAT) Wide Mini S-2×25 gel (Elchrom Scientific, Ebersberg, Germany) in 1× Tris-acetate EDTA buffer at 10 V/cm for 105 min. The bands were visualized under UV light. A 100 bp molecular-size marker (Promega, Ebersberg, Germany) was loaded in each gel as a control.

  • URA5 RFLP analysis: PCRs were performed with the primers URA5 (5′-ATGTCCTCCCAAGCCCTCGACTCCG-3′) and SJ01 (5′-TAAGACCTCTGAACACCGTACTC-3′) [7]. The reaction conditions were as follows: initial denaturation (94°C, 2 min), 35 cycles of denaturation (94°C, 45 s), annealing (61°C, 1 min), extension (72°C, 2 min) and a final extension cycle at 72°C for 10 min.

    • Fifteen microlitres of PCR products were double digested with Sau96I (10 U/ml) and HhaI (20 U/ml) for 5 h at 37°C. The digested fragments were separated on 1.5% agarose gels stained with ethidium bromide.

Serotyping by multiplex PCR

Four primers designed for cloning LAC1 and a pair of primers for CAP64 were used (Table 1). The primers for CAP64 were based on DNA sequences reported by Ito-Kuwa et al. [19]. The primers for LAC1 and CAP64 were prepared at a concentration of 10 μM. For PCRs with the simultaneous use of the six primers for LAC1 and CAP64, 1.65 μL of each LAC1 primer and 0.85 μL of each CAP64 primer were added to a 50 μL reaction mixture. The PCR amplification was initiated at 94°C for 3 min, followed by 35–40 cycles of denaturation (94°C, 30 s), annealing (47°C, 60 s) and polymerization (72°C, 1.5 min), and a final extended polymerization step (72°C, 7 min). The amplified products were run at 100 V for 20 min on 1.5% agarose gels stained with ethidium bromide.

Table 1. Primers used for LAC1 and CAP64 amplification
 Primer sequences
LAC11 5′-GGAACAGCAACCACACTACTG-3′
2 5′-CATATTGGGTGGCATCTTACTGAGGGA-3′
3 5′-CCAGGGAACATGTTGTTGAC-3′
4 5′-GTTGTGGAAGGCAAAGAAAC-3′
CAP641 5′-GCCAAGGGAGTCTTATATGG- 3′
2 5′-GCAAAGGGTTCACCAAATCG- 3′

The molecular profiles obtained by PCR fingerprinting were analysed based on the presence or absence of readily apparent and well-defined bands in the digitised gel images. The molecular types were assigned by comparison with the reference strains (ATCC MYA-4564, ATCC MYA-4565, ATCC MYA-4566, ATCC MYA-4567) for VNI to VNIV. A dendrogram based on genotypic data was generated using the UPGMA algorithm.

Results

Table 2 summarizes the in vitro susceptibility data of 114 isolates of Cryptococcus to posaconazole, amphotericin B, fluconazole, itraconazole, ketoconazole, flucytosine and voriconazole as determined using the Sensititre YeastOne® assay. Caspofungin results are not presented in this table because this drug is inactive against C. neoformans.

Table 2. Minimum inhibitory concentration of the 114 C. neoformans strains tested for seven antifungal drugs determined by the Sensititre YeastOne microdilution method
 PosaconazoleaAmphotericin BaFluconazoleaItraconazoleaKetoconazoleaFlucytosineaVoriconazolea
  1. aPatients for whom between-strain differences in antifungal susceptibility were of three dilutions or larger figure in bold.

  2. bOriginal sample +5 colonies.

  3. cGeometric minimal inhibitory concentration.

SamplebRangec (mg/L)
ELOGA0.03–0.50.25–0.51–160.015–0.120.015–0.251–160.015–0.12
BAY 0.060.50.12–0.25 2160.03–0.12 0.0150.122–80.03–0.12
BEL 0.030.250.12–0.5 2–8 <0.008–0.06 <0.008–0.12 28 <0.008–0.06
MYA0.060.25 0.06–1 2–80.015–0.120.015–0.121–8 0.0150.12
BEL INN0.06–0.250.12–0.25 0.580.015–0.060.015–0.061–8<0.008–0.06
IRO0.06–0.25 0.060.51–8 0.0150.06 0.0150.061–16<0.008–0.06
LON 0.0080.50.06–0.25 116 0.0150.12 0.0150.122–4 0.0150.12
KEH 0.060.50.12–0.25 2160.015–0.12 <0.0080.122–8 <0.0080.12
BAI0.06–0.50.06–0.252–320.03–0.120.015–0.1240.03–0.12
ADJ0.060.52–40.03–0.060.015–0.030.5–20.03
OVA0.060.52–40.03–0.060.0152–80.015
NYA0.03–0.060.520.015–0.03<0.008–0.0152–4<0.008–0.015
NDO0.03–0.060.25–0.51–20.015–0.03<0.008–0.0081–2<0.008–0.015
BAH0.008–0.0150.25–0.50.5–1<0.008<0.0082–4<0.008
NGO0.03–0.060.51–20.015–0.03<0.008–0.0082–4<0.008–0.008
SER0.03–0.060.520.030.008–0.0150.5–2<0.008–0.015
KEN0.015–0.060.25–0.520.015–0.03<0.008–0.0151–40.015
NJO0.03–0.060.25–0.520.0150.01520.015
MFO0.06–0.120.25–0.52–40.015–0.060.008–0.0151–20.015–0.03
G MICc0.100.313.220.050.082.760.06

All isolates exhibited low MIC to posaconazole, amphotericin B, itraconazole, ketoconazole and voriconazole. As expected, all isolates exhibited high MIC to caspofungin. The lowest MIC values were observed for voriconazole, with MIC50 and MIC90 values of 0.015 and 0.12 mg/L, respectively.

High MIC values were determined for fluconazole (MIC50 and MIC90 values of 2 and 16 mg/L, respectively) and flucytosine (MIC50 and MIC90 values of 2 and 8 mg/L, respectively).

The G-MICs ranged from 0.06 mg/L for voriconazole to 3.22 mg/L for fluconazole.

When considering each isolate, 13/114 (11%) were categorized as SDD for fluconazole and I for flucytosine. Five of 19 patients studied exhibited isolates with different MIC categories for fluconazole in the same sample. Seven of 19 patients exhibited isolates with different MIC categories for flucytosine in the same sample. Briefly, patients ELOGA, BAY, BAI, LONG and KEH exhibited isolates that were both S and SDD for fluconazole in the same sample and patients ELOGA, BEL, OVA, BAY, MYA, BEL INN and KEH exhibited isolates that were both S and I for flucytosine in the same sample.

Genotyping

All isolates were identified as Cryptococcus neoformans var. grubii (serotype A) (Fig. 1). Despite the high discriminatory power, the analysis of the PCR patterns did not allow for the detection of the variability between both strains based on URA5 fragment length polymorphism and phage M13 core sequence primer amplification. These methods generated VNI profiles for all the studied isolates. (GACA)4 associated with (GTG)5 generated 20 different genotypes grouped into nine major molecular types (i.e. genotypes with >95% of similarity) for the 114 studied isolates. Eight patients exhibited C. neoformans with two different major molecular types in the same sample.

Figure 1.

 Relationship between the obtained genotypes from the 19 patients studied. Genotype data were generated by URA5 gene RFLP and micro- and minisatellite polymorphism. The dendrogram obtained is based on UPGMA clustering. The scale bar indicates the percentage of similarity. Each major group was determined when percentage was >95% of similarity. Braces determine major molecular types.

Discussion

The in vitro activity results for the main antifungal drugs are comparable with previously published values [12]. Nevertheless, the MIC90 values for fluconazole and flucytosine were greater in the present work than in previous studies [20], although one prior study identified a strain resistant to flucytosine [21]. As previously suggested, the extensive use of fluconazole could explain the elevated MICs observed in this study [22]. However, previous reports have already demonstrated the low activity of this drug against C. neoformans even though fluconazole has been proven to be active in vitro [23].

The present work demonstrated for the first time that at least two isolates with different antifungal susceptibilities (i.e. with a 3- to 4-fold difference in the MICs) can be found in a single sample despite the lack of antifungal treatment administered to the patients prior to the study. Although resistance was not observed in the present collection of isolates, these results may imply the presence of potentially resistant isolates in a given sample.

As demonstrated in previous studies, where C. neoformans var. grubii (serotype A) was the most common causative agent isolated from patients with cryptococcosis in sub-Saharan Africa, all of the patients studied in the Hôpital Central of Yaounde, Cameroon, exhibited C. neoformans with serotype A [24]. Moreover, Dromer et al. [25] demonstrated that cryptococcal infections with serotype A strains exhibited a higher mortality rate than other serotypes even after treatment. Surprisingly the URA5 RFLP and M13 primers used here did not exhibit the expected high extent of polymorphism between isolates [7,8,11,12]. The molecular types obtained were comparable with those described for VNI [10,11]. This result is in agreement with serotyping results because all C. neoformans var. grubii belong to the VNI molecular type [12,26]. Microsatellite polymorphism revealed nine major molecular types and eight patients exhibited at least two different molecular types. This has already been shown for aspergillosis and candidiasis [27,28]. Evidence of mixed infections was previously reported by Illnait Zaragozi et al. in a recent study of serial Cryptococcus isolates from Cuban patients [12]. Moreover, a study by Desnos-Ollivier et al. [29] recently reported a high frequency of mixed infections in 20% of patients with cryptococcal disease. The results of the present study are in agreement with this paper because 42% of our patients exhibited two different molecular type. Nevertheless, although a correlation between high MIC values and strain genotype has recently been demonstrated [30], such a relationship was not found in the present study. As suggested by Illnait-Zaragozi et al. [12], these genotypic differences occur due to a microevolution of the original strain. However, the present study demonstrated that the within-host Cryptococcus are heterogeneous (i) genotypically and (ii) with regard to their antifungal susceptibility. This has never been demonstrated in Cameroon and the present study provides new data for a better understanding of the biology and the epidemiology of C. neoformans.

Conclusion

Cryptococcosis remains one of the most life threatening opportunistic infections in HIV-positive patients in Africa. The present study analysed the ‘within-host’ diversity of Cryptococcus. The results demonstrated the high genetic variability among clinical isolates and the presence of isolates with different MICs in a given sample. A larger study with sequential isolates should be conducted to confirm these results and the clinical implications of these findings require further investigation.

Acknowledgements

We thank D. Castel for technical help.

Fundings

This study was supported by SIDACTION AI 19-01-396.

Transparency Declarations

None to declare.

Ancillary