INTRODUCTION
Candidiasis is the most frequent opportunistic mycosis in humans and is one of the major causes of mortality in immunocompromised patients1,2. The incidence of systemic fungal infections has increased steadily over recent years, mainly due to the growing number of organ and bone marrow transplants, cancer treatments, and AIDS cases, all of which are detrimental to the body’s natural defense system3. In view of the compromised immune system in such cases, the availability of effective antifungal agents is particularly critical to treatment. However, treatment can be hindered by the presence of Candida species displaying multiple drug resistance (MDR), which is becoming more frequently reported4,5.
Treatment for Candida infections normally involves triazole antifungals, of which fluconazole is the most commonly prescribed6,7. The azolic antifungals act by inactivating lanosterol 14α-demethylase, thereby inhibiting biosynthesis of ergosterol, a compound essential for cell integrity and membrane function8–9. This results in a decrease in the level of ergosterol available for membrane function and a concomitant increase in the number of intermediate metabolites10. Resistance to azole antifungals is commonly associated with continuous drug use, but can also because by intrinsic factors11,12, such as mutations or alterations in the expression of the CDR1, CDR2, PDR513,14, ERG315, MDR1, FLU116, and ERG1117 genes. The ERG11 and ERG3 genes encode proteins involved in the biosynthesis of ergosterol, and mutations at specific points in these two genes can critically alter the effectiveness of azolic antifungal drugs18.
Thus, this study aims to investigate polymorphisms in the ERG11 gene, and correlate this genetic data to the fluconazole susceptibility profile of different Candida species isolated from diabetic and kidney transplant patients.
METHODS
Biological material
Samples of saliva were collected from 90 outpatients, including 40 diabetic patients (12 male and 28 female), 19 kidney transplant recipients (9 male and 10 female), and 31 immunocompetent patients (3 male and 28 female; control group). The 40 diabetic patients were between 40 and 70 years old and had been diagnosed with type II diabetes over five years prior. All had hypertension but were not using insulin, and 40 were diagnosed with hyperglycemia. All transplant patients were between 30 and 60 years old, and had received a kidney transplant over one year before the study. In addition, 19 of the transplant patients were under treatment with prednisone, azathioprine, and cyclosporine. The control group was composed of individuals between the ages of 18 and 30, who had not been diagnosed with any disease, and were not using drugs with antimicrobial or anti-inflammatory activities. The patients within the kidney transplant, diabetic and control groups were not paired by gender or age. A total of 94 Candida isolates were obtained, and the species were identified by Benedetti et al.19. Among the 70 isolates used in this study, 58 were classified as Candida albicans, 6 as Candida glabrata and 8 as Candida tropicalis.
Ethical standards
This study was performed in accordance with the ethical standards of the 1964 Helsinki declaration, and was approved by the Ethics Committee (Conselho Nacional de Ética em Pesquisa – CONEP) under registration no. 19885/2011, Universidade Paranaense (Parana-Brazil). Informed consent was obtained from all individual participants involved in the study.
Fluconazole minimum inhibitory concentration
The fluconazole minimum inhibitory concentration (MIC) of Candida isolates was determined using an E-Test Kit (Biomerieux, France) in Mueller-Hinton media (Difco, USA), supplemented with 2% glucose and 0.5 μg/mL methylene blue, according to the protocol of the Clinical and Laboratory Standards Institute20. The isolates were classified as susceptible to fluconazole if their MIC was ≤8 μg/mL; dose-dependent MIC values were from 8 to 32 μg/mL; and resistant isolates had MIC ≤ 32 μg/mL21. As controls, three reference strains were used: C. albicans ATCC – 44858 (MIC – 1 μg/mL), C. glabrata ATCC – 2001(MIC – 4 μg/mL), and C. tropicalis ATCC – 28707 (MIC – 2 μg/mL). Analysis of variance (ANOVA), was used to compare MIC values of individual isolates, as well as isolate groups (diabetic, kidney transplant, and immunocompetent control) using GraphPad Prism software, version 5 (GraphPad, USA) with a significance level of 0.05. When the MIC value was higher than that reported for each species, the differences between reference strains and the isolates were confirmed with a t-test.
DNA isolation, amplification, and partial sequencing of the ERG11 gene
Total genomic DNA was purified from 2 day-old single colony cultures growing on Mueller-Hinton media (Difco, USA) using the Ultraclean Microbial DNA Isolation Kit(MoBio, USA) according to the manufacturer’s instructions. Partial sequences of the ERG11 gene were amplified using the following primer pairs: C. albicans9, Sec1A (forward); 5’-TTAGTGTTTTATTGGATTCCTTGGTT-3’, Sec1B (reverse); 5’-TCTCATTTCATCACCAAATAAAGATC-3’,C. tropicalis22, CtERGr1 – F (forward); 5’-TCTTTTGTCAACACAGTAATGGC-3’, CtERGr1 – R (reverse); 5’-AACACCTTTACCAAAAACAGGAG-3’, C. glabrata23, CgERGr1 – F (forward); 5’-ACTACAATAACATGTCCACTGA-3’, CgERGr1 – R (reverse); 5’-GGGTGGTCAAGTGGGAGTAA-3’. The amplification was performed as described by Xu et al.9 in a final volume of 12.5 μL, containing 1 X Tris Base buffer solution, dNTPs (0.2 µM, Invitrogen, USA), MgCl2 (1.6 mM), primers (15 pmol each), Taq DNA polymerase (0.5 U) (Invitrogen, USA), and template DNA (20 ng). The PCR products were purified using 7.5 M ammonium acetate (15 µL) and absolute ethanol (74 µL). Samples were incubated on ice for one hour, followed by centrifugation for 45 min at 23,100 g. The pellet was re-suspended in 12 µL of Milli-Q water. Sequencing of the PCR products was performed using a Big Dye Kit (Applied Biosystems, USA), followed by purification using Sephadex G-50 fine DNA grade resin (GE Healthcare, UK) in a MultiScreen Column Loader (Merck Millipore, USA), and analysis by electrophoresis in an ABI3500 Automatic Sequencer (Applied Biosystems, USA). The obtained sequences were visually inspected using the BioEdit program version 7.2.524, aligned using ClustalW25, and manually adjusted using MEGA software, version 626. Additional sequences were obtained from GenBank (Table 1) and alignments were generated in ClustalW. Amino acid prediction was performed using MEGA software, version 626.
TABLE 1: Polymorphic sites in sequences of the ERG11 gene from isolated species of C. albicans, C. tropicalis and C. glabrata, and associated amino acid prediction, fluconazole MIC values and GenBank codes.
Species | Isolate | Source | Mutation type* | Amino acid prediction** | MIC (μg/mL) | GenBank codes |
---|---|---|---|---|---|---|
C. albicans | ATCC- 44858 | ATCC | Synonymous | F105L / D116E / S137L | 1.0 | MF411488 |
CA_112_PC | Control | Missense | K128T | 0.2 | MF411470 | |
CA_2_PC | Control | Synonymous / Missense | F105L / K119N / K128T | 0.5 | MF411410 | |
CA_13_PC | Control | Synonymous | F105L / K119N | 0.5 | MF411414 | |
CA_136_PC | Control | Synonymous / Missense | F105L / K119N / K128T | 0.5 | MF411482 | |
CA_141_PC | Control | Synonymous | F105L | 0.5 | MF411487 | |
CA_115_PC | Control | Synonymous / Missense | F105L / S137L/ K128T | 0.75 | MF411471 | |
CA_124_PC | Control | Synonymous | F105L / D116E | 0.75 | MF411478 | |
CA_12_PC | Control | Synonymous / Missense | F105L / K128T | 1.0 | MF411413 | |
CA_128_PC | Control | Synonymous | F105L / D116E / K119N /S137L | 1.0 | MF411481 | |
CA_137_PC | Control | Synonymous | F105L / S137L | 1.0 | MF411484 | |
CA_9_PC | Control | Synonymous / Missense | F105L / K119N / K128T | 1.5 | MF411412 | |
CA_118_PC | Control | Synonymous / Missense | F105L / K119N / K128T | 1.5 | MF411473 | |
CA_122_PC | Control | Synonymous | F105L | 1.5 | MF411477 | |
CA_121_PC | Control | Synonymous / Missense | F105L / K128T | 2.0 | MF411476 | |
Arithmetic mean | 0.94 | |||||
Geometric mean | 0.69 | |||||
CA_30_PD | Diabetic | Synonymous / Missense | F105L / K128T | 0.5 | MF411422 | |
CA_35_PD | Diabetic | Synonymous / Missense | F105L / K119N / S137L / K128T | 0.75 | MF411426 | |
CA_79_PD | Diabetic | Synonymous / Missense | F105L / S137L/ K128T | 0.75 | MF411450 | |
CA_21_PD | Diabetic | Synonymous / Missense | F105L / K128T | 1.0 | MF411416 | |
CA_25_PD | Diabetic | Synonymous | F105L | 1.0 | MF411418 | |
CA_29_PD | Diabetic | Synonymous | F105L/ D116E/ K119N / S137L | 1.0 | MF411421 | |
CA_38_PD | Diabetic | Synonymous / Missense | F105L / K119N / S137L /K128T | 1.0 | MF411427 | |
CA_91_PD | Diabetic | Synonymous | F105L | 1.0 | MF411457 | |
CA_100_PD | Diabetic | Synonymous / Missense | F105L / K119N /K128T | 1.0 | MF411464 | |
CA_135_PD | Diabetic | Synonymous | F105L | 1.0 | MF411483 | |
CA_22_PD | Diabetic | Synonymous / Missense | F105L/ K119N / K128T | 1.5 | MF411417 | |
CA_28_PD | Diabetic | Synonymous | F105L | 1.5 | MF411420 | |
CA_31_PD | Diabetic | Synonymous | F105L | 1.5 | MF411423 | |
CA_33_PD | Diabetic | Synonymous / Missense | F105L / D116E / S137L / K128T | 1.5 | MF411425 | |
CA_39_PD | Diabetic | Synonymous / Missense | F105L / K128T | 1.5 | MF411428 | |
CA_76_PD | Diabetic | Synonymous | F105L / D116E / K119N / S137L | 1.5 | MF411449 | |
CA_81_PD | Diabetic | Synonymous / Missense | F105L / S137L/ K128T | 1.5 | MF411451 | |
CA_86_PD | Diabetic | Synonymous | F105L / D116E / K119N | 1.5 | MF411454 | |
CA_90_PD | Diabetic | Synonymous | F105L | 1.5 | MF411456 | |
CA_104_PD | Diabetic | Synonymous / Missense | F105L / K119N / K128T | 1.5 | MF411466 | |
CA_19_PD | Diabetic | Synonymous | F105L | 2.0 | MF411415 | |
CA_32_PD | Diabetic | Synonymous | F105L / D116E | 2.0 | MF411424 | |
CA_43_PD | Diabetic | Synonymous | F105L / D116E | 2.0 | MF411429 | |
CA_69_PD | Diabetic | Synonymous | F105L / K119N / S137L | 2.0 | MF411444 | |
CA_70_PD | Diabetic | Synonymous | F105L / D116E / K119N | 2.0 | MF411445 | |
CA_72_PD | Diabetic | Synonymous | F105L | 2.0 | MF411447 | |
CA_73_PD | Diabetic | Synonymous / Missense | F105L / K128T | 2.0 | MF411448 | |
CA_84_PD | Diabetic | Synonymous | F105L /D116E /K119N /S137L | 2.0 | MF411452 | |
CA_92_PD | Diabetic | Synonymous / Missense | F105L /D116E /K119N/S137L/ K128T | 2.0 | MF411458 | |
CA_93_PD | Diabetic | Synonymous / Missense | F105L / K128T | 2.0 | MF411459 | |
CA_98_PD | Diabetic | Synonymous / Missense | K119N / K128T | 2.0 | MF411462 | |
CA_99_PD | Diabetic | Synonymous | F105L | 2.0 | MF411463 | |
CA_85_PD | Diabetic | Synonymous | F105L /D116E / K119N /S137L | 3.0 | MF411453 | |
CA_96_PD | Diabetic | Synonymous | F105L | 3.0 | MF411461 | |
CA_102_PD | Diabetic | Synonymous | F105L | 3.0 | MF411465 | |
Arithmetic mean | 1.63 | |||||
Geometric mean | 1.49 | |||||
CA_57_PT | Transplanted | Synonymous | F105L | 0.75 | MF411436 | |
CA_46_PT | Transplanted | Synonymous / Missense | F105L / K119N / K128T | 1.0 | MF411430 | |
CA_51_PT | Transplanted | Synonymous | F105L / K119N | 1.5 | MF411433 | |
CA_52_PT | Transplanted | Synonymous / Missense | F105L / K128T | 1.5 | MF411434 | |
CA_49_PT | Transplanted | Synonymous | F105L | 2.0 | MF411432 | |
CA_58_PT | Transplanted | Synonymous | F105L / K119N | 2.0 | MF411437 | |
CA_63_PT | Transplanted | Synonymous | F105L / K119N / S137L | 2.0 | MF411441 | |
CA_62_PT | Transplanted | Synonymous | F105L | 3.0 | MF411440 | |
CA_67_PT | Transplanted | Synonymous / Missense | F105L / K119N / K128T | 3.0 | MF411443 | |
Arithmetic mean | 1.86 | |||||
Geometric mean | 1.70 | |||||
C. tropicalis | ATCC- 28707 | ATCC | Synonymous | T224C / G263A | 2.0 | MF414164 |
CT_40_PD | Diabetic | Synonymous | G263A | 1.5 | MF414156 | |
CT_42_PD | Diabetic | Synonymous | G263A | 1.5 | MF414157 | |
CT_55_PT | Transplanted | Synonymous | T224C / G263A | 1.5 | MF414158 | |
CT_129_PC | Control | Synonymous | T224C / G263A | 1.5 | MF414163 | |
CT_77_PD | Diabetic | Synonymous | T224C / G263A | 2.0 | MF414160 | |
CT_105_PD | Diabetic | Synonymous | T224C / G263A | 2.0 | MF414161 | |
CT_106_PD | Diabetic | Synonymous | T224C / G263A | 2.0 | MF414162 | |
Arithmetic mean | 1.71 | |||||
Geometric mean | 1.73 | |||||
C. glabrata | ATCC -2001 | ATCC | Not found | __ | 4.0 | MF414155 |
CG_113_PC | Control | Not found | __ | 1.5 | MF414152 | |
CG_123_PC | Control | Not found | __ | 1.5 | MF414153 | |
CG_45_PT | Transplanted | Not found | __ | 4.0 | MF414149 | |
CG_48_PT | Transplanted | Not found | __ | 4.0 | MF414150 | |
CG_78_PT | Diabetic | Not found | __ | 4.0 | MF414151 | |
Arithmetic mean | 3.0 | |||||
Geometric mean | 2.88 |
*Mutations F105L, D116E, K119N, S137L, T224C and G263A were classified as synonymous and K128T as missense. **Mutations previously described.
RESULTS
Fluconazole minimum inhibitory concentration
All isolates with MIC values ≤ 8 μg/mL21were classified as susceptible to fluconazole (Table 1). The mean MIC values was higher for the species C. glabrata (p=0.02), the mean MIC were 1.43 μg/mL for C. albicans, 1.71 μg/mL for C. tropicalis, and 3.0 μg/mL for C. glabrata. Interestingly, although all C. albicansisolates were characterized as susceptible to fluconazole (MIC ≤ 8 μg/mL)21, strains isolated from transplanted (MIC = 1.86 μg/mL) and diabetic (MIC =1.63 μg/mL) patients had higher fluconazole MIC values than the control group (MIC =0.94 μg/mL) (p=0.001) (Table 1).
Analysis of polymorphic sites in the ERG11 gene of Candida isolates
The GenBank codes for the partial sequences of the ERG11 gene from Candida isolates are listed in Table 1. The alignment of the ERG11 sequence of the C. albicans isolates was 482 bp-long, with 134 polymorphic sites. The 407 bp-long alignment of the C. tropicalis isolates to the ERG11 genus contained 12 polymorphic sites. No polymorphisms were found in the partial sequence alignment of the ERG11 gene of the six C. glabrata isolates.
On comparing the obtained ERG11 sequences to the published ERG11 sequences of fluconazole-sensitive Candida species, we observed that 32.9% (n=31) of isolates showed no nucleotide alteration in their partial ERG11 sequences. The remaining 67.1% (n=63) of isolates showed at least one mutation of these, 19.1% (n=18) had a unique polymorphic site, 21.3% (n=20) showed two substitutions, 17.1% (n=16) had three polymorphic sites, and 9.6% (n=9) had more than three variable sites in the partial sequence of the ERG11 gene (Table 1).
Analysis of the predicted amino acid sequence of ERG11 revealed that 82.9% (121 bp) of variable sites resulted in synonymous mutations, with no change in the predicted amino acid sequence in C. albicans or C. tropicalis isolates, while 17.1% (25 bp) resulted in missense mutations, which were observed only in C. albicans isolates. In addition, 41.8% (56 bp) of mutations in C. albicans isolates occurred in the amino acid lysine (Lys) position F105L, 18.7% (25) in arginine (Arg) position K119N, 11.1% (15) in phenylalanine (Phe) position S137L, and 9.7% (13) in leucine (Leu) position D116E. In C. tropicalis, the most frequent mutation (58.3%) was in the amino acid asparagine (Asn) position G263A, followed by threonine (Thr) position T224C (41.7%). The missense nucleotide mutation observed in the ERG11 gene of C. albicansisolates was a substitution of adenine with cytosine at codon 530, which resulted in the amino acid cysteine instead of phenylalanine (Table 1). In addition, of the 25 missense mutations observed, 92% (n=23) were heterozygous, and 8% were homozygous.
DISCUSSION
Candida infections are the most frequent cause of opportunistic diseases in immunocompromised patients27, and fluconazole is the first treatment option for such cases. However, several studies have documented the ability of Candida to develop high-level resistance to azole compound28, either through efflux pumps or alterations in the sterol 14-α-demethylase caused by mutations in the ERG11 gene29,30. For effective treatment of Candida infections, and selection of the most efficient prophylactic measures, it is necessary to identify the Candida species, evaluate its susceptibility profile to different antifungals, and determine the mechanisms involved in any observed resistance31.
The Candida isolates used in this study had been classified in a previous study21as the three different species; C. albicans, C. glabrata, and C. tropicalis. All isolates (70) were classified as susceptible to fluconazole, with MIC ≤ 8 μg/mL20. However, there were significant differences in the susceptibility ranges, with isolates from the control group displaying higher susceptibility to fluconazole than isolates from diabetic and kidney transplant patients (Table 1). This higher susceptibility of isolates from the control group maybe related to the absence of continuous fluconazole use, which has been suggested as one of the principal causes of fluconazole resistance in Candida12,32,33. In contrast, the higher fluconazole MIC values of strains isolated from transplant patients may be related to frequent contact with hospital patients, which facilitates oral colonization by less susceptible strains. Another contributor to this lower susceptibility may be the high cellular stress level of this group, caused by exposure to other drugs such as chemotherapeutics, corticosteroids, and antibiotics, which have been shown to activate resistance mechanisms in yeast34,35.
In addition, we observed a difference in the mean fluconazole MIC values between different Candidaspecies (Table 1). The lowest fluconazole MIC values were observed in C. albicans, followed by C. tropicalis, then C. glabrata. These differences may be related to intrinsic factors specific to each species36,37(Table 1). In the case of C. glabrata isolates, the relatively high tolerance to fluconazole is supported by the absence of any mutations in the ERG11 gene (Table 1), suggesting that the higher tolerance may be a common characteristic of this species. Consistent with this notion, previous studies have demonstrated that after exposure to fluconazole, C. glabrata isolates showed lower susceptibility profiles compared to other Candida species36,37.
Analysis of the ERG11 gene sequence showed that, despite the high number of polymorphic sites observed in C. albicans and C. tropicalis, none occurred in the region coding for the binding site of the antifungal, the enzyme sterol 14-α-demethylase38,39. Although specific ERG11 amino acid substitutions are known to be responsible for resistance to azolic compounds40,41, the mutations characterized as F105L, D116E, K119N, K128T and S137L in C. albicans isolates, and T224C and G263A in C. tropicalis (Table 1) were not associated with antifungal resistance9,23,42,43,44, in agreement with the MIC profiles. Although these mutations are synonymous, it is possible that they could influence the efficiency of translation, resulting in alterations in protein production, as has been suggested previously45. For example, the F105L mutation was previously correlated with fluconazole resistance40, where as another study suggested that this mutation alone does not result in an alteration in the active site of 14-α-demethylase46.
Several studies have recently investigated the fluconazole susceptibility of Candida isolates from HIV, cancer, and immunocompromised patients in Brazil47,48. However, no other studies have investigated the fluconazole susceptibility of Candida species associated with diabetic and kidney transplant patients. Although Chaves et al.49evaluated the diversity of Candida isolates from kidney transplant recipients in Brazil, the authors did not investigate the fluconazole susceptibility profiles
We therefore provide the first comparison of fluconazole susceptibility profiles of Candida species isolated from diabetic and kidney transplant patients in Brazil. Although the therapeutic and prophylactic use of fluconazole is widespread, fluconazole resistant Candida isolates were not observed in this study. However, fluconazole MIC values were higher for isolates from diabetic and kidney transplant patients compared with the control group. Further studies should therefore be performed to verify whether continued exposure to fluconazole may result in the identification of resistant Candida strains in these patients.