Fluconazole and amphotericin-B resistance are associated with increased catalase and superoxide dismutase activity in Candida albicans and Candida dubliniensis

Introduction: Candida dubliniensis, a new species of Candida that has been recovered from several sites in healthy people, has been associated with recurrent episodes of oral candidiasis in AIDS and HIV-positive patients. This species is closely related to C. albicans. The enzymatic activity of C. dubliniensis in response to oxidative stress is of interest for the development of drugs to combat C. dubliniensis. Methods: Fluconazoleand amphotericin B-resistant strains were generated as described by FeketeForgács et al. (2000). Superoxide dismutase (SOD) and catalase assays were performed as described by McCord and Fridovich (1969) and Aebi (1984), respectively. Results: We demonstrated that superoxide dismutase (SOD) and cata lase activities were signifi cantly higher (p<0.05) in the fl uconazoleand amphotericin B-resistant strains of C. dubliniensis and C. albicans than in the sensitive strains. The catalase and SOD activities were also signifi cantly (p<0.01) higher in the sensitive and resistant C. albicans strains than in the respective C. dubliniensis strains. Conclusions: These data suggest that C. albicans is better protected from oxidative stress than C. dubliniensis and that fl uconazole, like amphotericin B, can induce oxidative stress in Candida; oxidative stress induces an adaptive response that results in a coordinated increase in catalase and SOD activities.

of unusual carbohydrate assimilation profi les, which can be identifi ed using commercially available systems such as the API ® ID 32C and API ® 20C AUX kits, were one of the earliest indications that C. dubliniensis represented a new species within the Candida genus 4 .Studies have since documented a high prevalence of this organism in patients suffering from diabetes (3.6 to 18.2%), cystic fi brosis (11.1%), cancer (2 to 4.6%), vulvovaginal candidiasis (0.17 to 2.4%), and candidemia (0.5 to 7.9%) 6 .
Candida dubliniensis is more susceptible to polyenes and azoles than C. albicans; however, a remarkable feature of C. dubliniensis is the rapid development of secondary resistance to fl uconazole 7 .Azoles, such as fl uconazole, act on the ergosterol biosynthesis pathway by inhibiting the enzyme C14-α-demethylase, a cytochrome P450-dependent enzyme 8 .Amphotericin B, a polyene antifungal agent, binds to sterols such as ergosterol and disrupts the osmotic integrity of the fungal membrane, resulting in the leakage of intracellular ions and metabolites 9 .Despite intense study of C. dubliniensis, the enzymatic activities of superoxide dismutase (SOD) and catalase in fl uconazole-resistant and amphotericin B-resistant C. dubliniensis are unknown.In Candida, SOD plays an important protective role against attack by macrophages and neutrophils, which generate superoxide radicals 10 .Superoxide dismutase converts damaging superoxide radicals (O 2 .-), one type of reactive oxygen species (ROS), to the less damaging Linares CEB et al -Candida albicans and Candida dubliniensis METHODS hydrogen peroxide (H 2 O 2 ), which can be converted into water by catalase 11,12 .Catalase has been implicated in amphotericin B resistance by inhibiting the oxidative damage produced by this polyene 13 .Ergosterol, a sterol present in the fungal plasma membrane, is the target of amphotericin B. The binding of amphotericin B to ergosterol produces channels that leak potassium ions, resulting in the disruption of the proton gradient 14,15 .The antioxidant enzymes SOD and catalase may act synergistically to protect cells from oxidative damage 16 .Some authors have suggested that the coordinated inactivation of ROS by the sequential action of superoxide-and hydrogen peroxide-inactivating enzymes is essential for the effective protection of cells against ROS toxicity 17 .
To further explore the antioxidant enzymatic activities and the coordinated inactivation of ROS by SOD and catalase in C. dubliniensis and C. albicans, we investigated the SOD and catalase activities in fl uconazole-and amphotericin B-sensitive and resistant strains of C. dubliniensis and C. albicans.

Strains
The Candida albicans and Candida dubliniensis strains used in this study [Ca 10, Ca 81, Ca 119, Ca 170, Ca ATCC 44373, Cd 1, Cd 2, Cd 3, Cd 4, Cd 5, Cd 7, Cd 9, Cd 11, and Cd 13] were isolated from patients with oral candidiasis at the University Hospital of Universidade Federal de Santa Maria (Brazil).Species identifi cation was based on phenotypic and genotypic characteristics.The C. dubliniensis genotype was confi rmed by random amplifi cation of polymorphic deoxyribonucleic acid (DNA) using the primers CDU (5′ GCG ATC CCC A 3′) and B-14 (5′ GAT CAA GTC C 3′) 18,4 .The C. albicans strain ATCC 44373 from the American Type Culture Collection was used as a control strain.

Generation of fl uconazole-resistant strains
The minimal inhibitory concentration (MIC) of fl uconazole was first determined for C. albicans and C. dubliniensis according to the Clinical and Laboratory Standards Institute (CLSI) document M27-A3 19 .After the MIC determinations, the cultures were grown overnight in Sabouraud glucose broth (SDB), and the cells were added to fl asks containing 20ml of media to achieve a fi nal absorbance of 0.1 at 640nm.The cultures were incubated at 30°C for 10h, and fl uconazole was then added to a fi nal concentration of 8μg ml −1 .After 14h of further incubation, the cells of the fl uconazole-containing culture were consecutively subcultured 3 times in fresh SDB medium containing 8μg ml −1 fl uconazole; in each case, the cultures were incubated at 30°C with shaking for 24h.After the third incubation, the cells were added to fl asks containing 20ml of SDB medium containing 8μg ml −1 fl uconazole, and the turbidity of the cultures was adjusted to achieve a fi nal absorbance of 0.1 at 640nm.After a 10h incubation, fl uconazole was added at a fi nal concentration of 16μg ml −1 , and after 14h, the cells from this culture were subcultured 3 times in fresh SDB medium containing 16μg ml −1 fl uconazole; each subculture was incubated at 30°C with shaking for 24h.The fl uconazole concentration was always duplicated under this procedure, and the fi nal concentration was 64μg ml −1 .Cells from this culture were plated on Sabouraud glucose agar (SDA), and the colonies were designated C. albicans fl uconazole resistant and C. dubliniensis fl uconazole resistant 20 .

Generation of amphotericin B-resistant strains
Induction of amphotericin B resistance was performed as described for the generation of the fl uconazole-resistant strains, with some modifi cations.The time needed for cell growth between exposures to different amphotericin B concentrations was 48h.Resistant cells obtained at 2µg ml -1 were plated on SDA, and the colonies were designated C. albicans amphotericin B-resistant and C. dubliniensis amphotericin B-resistant.

Susceptibility tests
Antifungal agents: standard fluconazole (Pfizer Central Research) and amphotericin B (Bristol-Myers, Squibb Australia Pty Ltd) powders were obtained from their respective manufacturers.Stock solutions were prepared in water (fl uconazole) and dimethyl sulfoxide (amphotericin B).Serial two-fold dilutions were prepared exactly as outlined in the CLSI document M27-A3 19 .Final dilutions were made in Roswell Park Memorial Institute (RPMI) 1640 medium buffered to pH 7.0 with 0.165mmol l -1 morpholinopropanesulfonic acid (MOPS).Aliquots (100µl) of each antifungal agent at a 2X fi nal concentration were dispensed into the wells of plastic microdilution trays, which were sealed and frozen at -70ºC until use.The fl uconazole dilutions ranged from 0.12 to 128μg ml -1 , and the amphotericin B dilutions ranged from 0.0313 to 16μg ml -1 .
Inoculum standardization -inoculum suspensions of the yeast strains were prepared in sterile saline (0.85% NaCl) from 24h cultures grown on SDA at 35°C.The turbidity of the suspensions was adjusted to achieve a fi nal absorbance of 0.1 at 530nm 8,19 .
CLSI broth microdilution method (document M27-A3): microplates containing 100μl of two-fold serial dilutions of the antifungal drugs in standard RPMI 1640 medium (0.2% glucose) were inoculated with 100μl of inoculum containing 1 × 10 3 to 5 × 10 3 colony-forming units per milliliter (CFU/ml).Following inoculation, the reference microdilution plates were incubated at 35°C, and the Minimum Inhibitory Concentration MICs were determined after 24 and 48h.The fl uconazole and amphotericin B reference MICs corresponded to the lowest drug dilutions that resulted in prominent growth inhibition (≥ 80%) or the absence of growth, respectively 19 .

Enzyme assays
Candida albicans and Candida dubliniensis strains: strains were cultured for 48h in SDA containing 64μg ml −1 fl uconazole or 2μg ml −1 amphotericin B. The yeast cells were added to sterile saline (0.85%) to achieve a fi nal absorbance of 0.1 at a wavelength of 625nm.Aliquots (1ml) of this suspension were added to 50ml of SDB containing 4% glucose and 1% mycological peptone, and these cultures were incubated at 30ºC with shaking for 24h for the fl uconazole experiments and for 48h for the amphotericin B experiments.After centrifugation, the cells were washed 3 times with sterile water, and a crude extract was prepared.Preparation of cell-free extracts: the crude extracts were prepared using the glass bead lysis method.Cells that had been cultured in SDB supplemented with fluconazole or amphotericin B were washed 3 times with MilliQ water and resuspended in lysis buffer (50mmol l -1 potassium phosphate pH 7.0) containing 0.5g of 500-µm-diameter glass beads.The mixture was homogenized in a Tefl on-glass homogenizer for 6 cycles of alternating homogenization for 20s and cooling on ice.The mixture was then centrifuged for 30 min in a refrigerated centrifuge to remove the cell debris and glass beads.The supernatant was used for enzyme assays.

Superoxide dismutase activity was assayed according to
McCord and Fridovich 21 .The assay was performed in a total volume of 1ml containing 50mmol l -1 glycine buffer (pH 10), 60mmol l -1 epinephrine, and the enzyme.Epinephrine was added, and adrenochrome formation was recorded at 480nm with a ultraviolet-visible (UV-Vis) spectrophotometer for 4 min.One unit of SOD activity was equivalent to the amount of enzyme required to inhibit epinephrine oxidation by 50% under the experimental conditions.The assays were performed in triplicate.

Catalase activity
Catalase activity was determined in the cell-free extracts using the method of Aebi et al. 22 .H 2 O 2 solution (10mM), enzyme extract, and 50mM phosphate buffer (pH 7) were pipetted into a cuvette.The reduction of H 2 O 2 was followed at a wavelength of 240nm for 4 min against a blank containing 50mM phosphate buffer and enzyme extract.Catalase activity was expressed in ΔE min -1 (mg protein) -1 .The assays were performed in triplicate.

Protein analysis
Protein concentrations were measured using the Bradford method and Coomassie Blue; serum albumin was used as the standard 23 .

Statistics
Paired t tests were used to compare the sensitive and resistant cells.Comparisons among the C. albicans and C. dubliniensis groups were made using unpaired t tests.The calculations were performed using the GraphPad InStat � statistical program.

Ethical considerations
This protocol was approved by the Bioethics Committee of the Universidade Regional Integrada do Alto Uruguai e das Missões from Frederico Westphalen -RS, Brazil, under registration n° 061-2/PIH/04.Fluconazole-and amphotericin B-resistant C. dubliniensis and C. albicans strains were obtained by exposing sensitive isolates to increasing concentrations of these antifungal agents.For the sensitive C. dubliniensis cells, the MICs ranged from 0.06 to 0.5μg ml -1 for fl uconazole and from 0.0312 to 0.125μg ml -1 for amphotericin B; for the sensitive C. albicans cells, the MICs ranged from 0.5 to 4.0μg ml -1 for fl uconazole and from 0.0625 to 0.25μg ml -1 for amphotericin B. After the induction of resistance, tests of the susceptibility to fl uconazole and amphotericin B demonstrated that resistance developed in all strains assayed.
Table 1 shows the SOD and catalase activities in the cell-free extracts of the fl uconazole-sensitive and fl uconazole-resistant C. dubliniensis strains.In general, the mean SOD activity was 1.97-fold greater for fl uconazole-resistant C. dubliniensis than for the sensitive cells.The difference between the 2 groups was signifi cant (p < 0.01).The catalase activity of the fl uconazoleresistant cells was 1.58-fold greater than that of the fl uconazolesensitive cells; this difference was signifi cant (p < 0.01).
Table 2 shows the SOD and catalase activities of the C. dubliniensis strains resistant to 2µg ml -1 amphotericin B. The mean SOD activity of the C. dubliniensis strain resistant to 2µg ml -1 amphotericin B was 1.97-fold greater than that of the sensitive cells; this difference was statistically signifi cant (p < 0.01).The mean catalase activity of the group resistant to 2µg ml -1 amphotericin B was 2-fold greater than that of the sensitive group.Statistical analyses showed signifi cant differences (p < 0.01) between the C. dubliniensis catalase activities of the cells resistant to 2µg ml -1 amphotericin B and the sensitive cells.
The SOD and catalase activities of the fl uconazole-sensitive and fluconazole-resistant C. albicans cells are shown in Table 3.The mean SOD activity of the fl uconazole-resistant C. albicans was 1.3-fold greater than that of the sensitive cells; this difference was statistically signifi cant (p < 0.01).The mean catalase activity of the fl uconazole-resistant group was 1.57fold greater than that of the fl uconazole-sensitive group; this difference was statistically signifi cant (p < 0.01).
Table 4 shows the SOD and catalase activities of the C. albicans strains that were sensitive or resistant to 2µg ml -1

DISCUSSION
Since the recognition of C. dubliniensis as a new species in 1995, many studies have investigated the differences between this species and C. albicans.Studies of the responses of these two species under identical conditions have uncovered phenotypic as well as genotypic differences between these two species.Enzymes involved in virulence, such as aspartyl proteinases, phospholipases, chondroitin sulfatase, and hyaluronidase, have been thoroughly studied.De Bernardis et al. 24 and Linares et al. 25 demonstrated that the level of aspartyl proteinase was signifi cantly higher in C. albicans than in C. dubliniensis, while no differences in phospholipase, chondroitin sulfatase, and hyaluronidase activities were observed [24][25][26] .In this study, we investigated the effect of fl uconazole and amphotericin B resistance on the SOD and catalase activities of C. albicans and C. dubliniensis.These enzymes are not related to virulence factors but play important roles in the oxidative stress response in both species 27 .SOD activity is involved in the detoxifi cation of free superoxide radicals, such as the radical produced by phagocyte nicotinamide adenine dinucleotide phosphate (NADPH) oxidase 29 .Phagocytic cells have 2 antimicrobial systems that are responsible for the generation of reactive oxygen species (ROS) and reactive nitrogen species (RNS).To counter these antimicrobial systems, microorganisms such as Candida have multiple defense responses that can be adjusted according to the nature and amount of ROS produced by the host or environment 30 .Our results revealed that SOD activity was signifi cantly increased in C. dubliniensis and C. albicans strains that were resistant to amphotericin B and fl uconazole compared to sensitive strains.The increase in SOD activity in response to amphotericin B may be due to the production of oxidative damage in Candida through lipid peroxidation 15 .In our study, yeast cultures were exposed to increasing concentrations of amphotericin B, up to a maximum of 2 µg ml -1 .In general, at this concentration, amphotericin B kills Candida by inducing oxidative damage through lipid peroxidation; however, we determined that Candida SOD activity was activated in response to amphotericin B exposure.The mean SOD activity of the C. dubliniensis and C. albicans strains resistant to 2µg ml -1 amphotericin B was 1.97-fold greater than that of the sensitive strains.This SOD activation likely occurred due to the induction of an adaptive response upon exposure of Candida to sublethal concentrations of peroxide or superoxide.This adaptive response results in a several-fold increase in catalase, SOD, and glucose 6-phosphate dehydrogenase (G6PDH) activity 29,31,32 .In addition, the elevation of SOD activity in the yeast strains resistant to amphotericin B may contribute to an increased resistance to phagocytic cell attack, making the yeast cells more resistant to host defenses.We also observed that the SOD activities of the C. albicans strains that were sensitive to fl uconazole and amphotericin B were signifi cantly (p < 0.01) higher than the SOD activities of the sensitive C. dubliniensis strains.When SOD activity was compared among the resistant groups, C. albicans also expressed signifi cantly (p < 0.01) higher levels of SOD than C. dubliniensis.
As previously demonstrated for other enzymes (e.g., proteinases), SOD activity was higher in C. albicans than in C. dubliniensis.This study is the fi rst to evaluate the effect of fl uconazole resistance on SOD activity in C. albicans and C. dubliniensis.Our results suggest that fl uconazole has an oxidant effect on Candida, resulting in the activation of its antioxidant system, including SOD.The mean SOD activities of the fl uconazole-resistant C. albicans and C. dubliniensis strains were 1.93-and 1.97-fold higher, respectively, than those of the sensitive strains.If fl uconazole induces SOD activity, like amphotericin B, then this reaction to fl uconazole would contribute to the increased resistance of Candida to phagocytic cell attack.
Similar to SOD activity, catalase activity was elevated in the fl uconazole-and amphotericin B-resistant C. albicans and C. dubliniensis cells compared to the sensitive groups.The enzyme catalase converts hydrogen peroxide (H 2 O 2 ) to oxygen (O 2 ) and water (H 2 O) 16 .In Candida, catalase has been suggested to be involved in the resistance to amphotericin B and is involved in hypha formation 15,28 .Catalase activity has not been previously studied in fl uconazole-resistant C. albicans and C. dubliniensis strains.Here, we demonstrated that the catalase activities of the amphotericin B-resistant C. dubliniensis and C. albicans strains were 2-and 1.77-fold higher, respectively, than those of the sensitive groups.The catalase activities of the fl uconazoleresistant C. albicans and C. dubliniensis groups were 1.58-and 1.57-fold higher, respectively, than those of the sensitive groups.These fi ndings demonstrate that similar changes in catalase and SOD activities occur in response to exposure to fl uconazole and amphotericin B under the same conditions and corroborate the hypothesis that an adaptive response occurs when Candida is exposed to oxidative stress 32 .Based on the literature, we expected that these enzymes would have similar responses because the product of SOD is substrate for catalase 11 .Here, we confi rmed this enzymatic profi le in C. albicans and C. dubliniensis.Similar to SOD, catalase activity was signifi cantly (p < 0.01) higher in the sensitive and resistant C. albicans strains compared to the respective C. dubliniensis groups.In combination with the increased susceptibility of C. dubliniensis to hydrogen peroxide and macrophage killing, this result suggests that C. albicans is better protected from oxidative stress than C. dubliniensis, in accordance with the higher virulence of C. albicans 33,34,35 .
The results obtained in this study also demonstrate a synergism between SOD and catalase in the C. dubliniensis and C. albicans strains that were resistant to fl uconazole and amphotericin B, in which protection by SOD generates hydrogen peroxide but not other ROS, such as lipid hydroperoxides or the reactive nitric intermediate (RNI) peroxynitrite, which would be more detrimental to the cells due to the absence of defenses against these types of compounds.Because the hydrogen peroxide generated by SOD activity is still toxic to the cell, it must be eliminated by catalase, which converts this toxic compound into water 14 .
In conclusion, this study is the fi rst to determine the effect of fl uconazole and amphotericin B resistance on catalase and SOD activities in C. albicans and C. dubliniensis.The stress-inducing effect of fl uconazole on Candida has not been previously described, and we have demonstrated that catalase and SOD activities are increased in fl uconazole-resistant cells.We also confi rmed that C. albicans has more active antioxidant pathways than C. dubliniensis.

TABLE 2 -
Superoxide dismutase and catalase activities of the amphotericin B-sensitive and amphotericin B-resistant Candida dubliniensis strains.The data are the mean ± SD of triplicate experiments; U: unit of SOD; ΔE: absorbance change at 240nm over 240s.a Amphotericin B-sensitive C. dubliniensis; b C. dubliniensis resistant to 2µg/ml -1 amphotericin B; c Statistical analyses indicated signifi cant differences between the SOD activities of the C. dubliniensis cells resistant to 2µg/ml -1 amphotericin B and the sensitive C. dubliniensis cells (p < 0.01).d Statistical analyses indicated signifi cant differences between the catalase activities of the C. dubliniensis cells resistant to 2µg/ml -1 amphotericin B and the sensitive C. dubliniensis cells (p < 0.01).SOD: superoxide dismutase; SD: standard deviation; C.: Candida.

TABLE 3 -
Superoxide dismutase and catalase activities of the fl uconazole-sensitive and fl uconazole-resistant Candida albicans strains.
The data are the mean ± SD of triplicate experiments.U: unit of SOD; ΔE: absorbance change at 240nm over 240s.a Fluconazole-sensitive C. albicans; b Fluconazole-resistant C. albicans (64µg/ml -1 ); c Statistical analyses revealed signifi cant differences between the SOD activities of the fl uconazole-resistant and fl uconazole-sensitive C. albicans groups (p < 0.01); d Statistical analyses revealed signifi cant differences between the catalase activities of the fl uconazole-resistant and fl uconazole-sensitive C. albicans groups (p < 0.01).SOD: superoxide dismutase; SD: standard deviation; C.: Candida.
Linares CEB et al -Candida albicans and Candida dubliniensis

TABLE 4 -
Superoxide dismutase and catalase activities of the amphotericin B-sensitive and resistant Candida albicans strains.The data are the mean ± SD of triplicate experiments.U: unit of SOD.ΔE: absorbance change at 240nm over 240s.aAmphotericin B-sensitive C. albicans; b C. albicans resistant to 2 µg ml -1 amphotericin B; c Statistical analyses revealed signifi cant differences between the SOD activities of the C. albicans cells resistant to 2µg/ml -1 amphotericin B and the sensitive C. albicans cells (p < 0.01); d Statistical analyses revealed differences between the catalase activities of the C. albicans cells resistant to 2µg/ml -1 amphotericin B and the sensitive C. albicans (p < 0.05).SOD: superoxide dismutase; SD: standard deviation; C.: Candida.amphotericinB. The mean SOD activity of the C. albicans strain resistant to 2µg ml -1 amphotericin B was 1.97-fold greater than that of the sensitive group; this difference was statistically signifi cant (p < 0.01).The mean catalase activity of the cells resistant to 2µg ml -1 amphotericin B was 1.77-fold greater than that of the sensitive group; this difference was statistically signifi cant (p < 0.05).