ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Oct. 2004, p. 3845–3849 0066-4804/04/$08.00⫹0 DOI: 10.1128/AAC.48.10.3845–3849.2004 Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Vol. 48, No. 10
Caspofungin Uptake Is Mediated by a High-Affinity Transporter in Candida albicans Padmaja Paderu, Steven Park, and David S. Perlin* Public Health Research Institute at the International Center for Public Health, Newark, New Jersey Received 2 April 2004/Returned for modification 14 May 2004/Accepted 18 June 2004
The uptake of the echinocandin drug caspofungin acetate in Candida albicans was evaluated at drug levels at or near the MIC for the organism. Maximal uptake was achieved in 10 min and was energy independent. A saturable transport system, consistent with a facilitated-diffusion carrier, was observed with the unlabeled drug competing with the labeled drug for uptake and efflux. More than 90% of the transported drug was observed in a single kinetic compartment that was available for efflux, indicating that the drug was free in the cytoplasm following uptake. Efflux was also energy independent but was sensitive to the presence of a fully loaded carrier on both faces of the bilayer. Overall, the data presented are consistent with the presence of a high-affinity facilitated-diffusion transporter that mediates caspofungin uptake and could be a potential source of transport-related reduced susceptibility. Currently, there is a fundamental gap in our knowledge of echinocandin interactions with the fungal cell. A basic description of cellular transport of caspofungin would provide a better understanding of drug action. Equally important, an analysis of transport kinetics could help explain differential effects of echinocandins on diverse yeasts and molds. In this report, we present the basic characteristics of caspofungin transport into cells of C. albicans and provide evidence for a saturable facilitated-diffusion carrier mechanism to account for high-affinity transport.
The echinocandin caspofungin acetate (Cancidas) is the first representative of a new class of -(1,3)-D-glucan synthesis inhibitors approved for the treatment of invasive aspergillosis and esophageal candidiasis. Caspofungin is a water-soluble, semisynthetic polypeptide amine derivative of the natural product pneumocandin B0 with potent activity against Candida spp., Aspergillus spp., and Cryptococcus spp. It targets the 1,3-D-glucan synthase, which is a GTP-dependent enzyme composed of soluble GTP-binding and plasma membrane-bound components (10). The genes encoding the major subunits have been identified from several diverse fungi, including Saccharomyces cerevisiae (GSC1 [FKS1] and GSC2 [FKS2]) (5, 11), Candida albicans (GSC1, GSL1, and GSL2) (12), Aspergillus nidulans (FKSA) (7), Paracoccidioides brasiliensis (FKS1) (13), and Cryptococcus neoformans (FKS1) (17). In vitro drug resistance has been linked to mutations in FKS1 (6, 9), supporting it as a potential site of interaction, although little is known about the molecular nature of drug-target interactions that lead to inhibition and resistance. Clinical exposure to caspofungin is growing, and development of full resistance or reduced susceptibility to caspofungin appears to be a rare event (3). As a new drug class, crossresistance with existing polyene and azole drugs is not expected and has not been observed. Echinocandins appear to be poor substrates for multidrug efflux pumps (15), and other nonFKS1-mediated resistance mechanisms have not been described. Uptake of caspofungin into the cell is required for inhibition of glucan synthase activity, although little or nothing is known about drug uptake in yeasts. In many microorganisms, changes in net uptake either by alterations in an uptake transporter or by increasing efflux can induce drug resistance phenotypes. The presence of a high-affinity transporter active at or below the MIC of the drug could have significance for emergence of potential resistance.
MATERIALS AND METHODS Strains and growth conditions. Wild-type C. albicans strain ATCC 90028 (MIC, 0.25 g/ml) was used in this study and was grown to mid-log phase in YPD medium (1% yeast extract, 2% peptone, 2% dextrose, pH 5.7) at 37°C for 3 to 4 h following transfer from a confluent overnight culture in the same medium. Cells were harvested by centrifugation at 3,000 ⫻ g for 10 min at 4°C and washed twice by resuspension in ice-cold MES buffer (5 mM morpholineethanesulfonic acid, 150 mM NaCl, pH 7.0, sterilely filtered) and centrifugation as described above. The final cell pellet was resuspended in 5 ml of MES buffer. The cells were depleted of energy by incubation at 30°C in MES buffer without a carbon source for 1 h with constant agitation or overnight (14 h) at 4°C and assayed for medium acidification as previously described (16). The carbon-starved cells were then diluted to a final concentration of 1.2 ⫻ 108/ml. Transport kinetics. A 5-ml basic reaction medium consisted of MES buffer and Candida cells at 1.2 ⫻ 108/ml at 22°C. In some experiments, a portion of the cells was re-energized following a 20-min preincubation with 2% (wt/vol) glucose. The transport (uptake) reaction was initiated by the addition of [3H]caspofungin (1 to 11 g/ml; kindly provided by Merck Research Labs). Aliquots (250 l) were removed in triplicate and rapidly centrifuged at top speed (12,000 ⫻ g) in a Microfuge for 20 s to pellet the cells. The supernatant was rapidly removed, and the cell pellet was washed two times by resuspension with an excess (1 ml) of ice-cold MES buffer with or without glucose, depending on the treatment of the cells, followed by centrifugation as described above. The washed cell pellet was resuspended in 180 l of MES buffer and divided into 60-l aliquots (triplicate) that were added to scintillation vials containing 2 ml of scintillation fluid (Ready-Safe liquid scintillation cocktail) and counted in a Beckman LS6500 scintillation counter. Efflux experiments were performed by first preloading cells with [3H]caspofungin following a 20-min preincubation at 22°C. The cells were harvested by centrifugation and diluted into a large excess (50- to 100-fold) of MES buffer. Aliquots were removed for centrifugation in the Microfuge and evaluated for residual drug in the pellet. Some transport assays were also conducted by using a filtration-based procedure to separate cells from the transport medium in the place of centrifugation. Aliquots (50 l) from the transport
* Corresponding author. Mailing address: Public Health Research Institute at the International Center for Public Health, 225 Warren St., Newark, NJ 07103. Phone: (973) 854-3200. Fax: (973) 854-3101. Email:
[email protected]. 3845
3846
PADERU ET AL.
ANTIMICROB. AGENTS CHEMOTHER.
FIG. 1. Effect of cellular energy status on caspofungin uptake. The uptake of [3H]caspofungin was monitored in 3 ml of reaction medium containing 1.2 ⫻ 108 cells/ml. The transport reaction was initiated by the addition of caspofungin at a final concentration of 1 g/ml to either energized cells maintained in the presence of glucose (⽧) or de-energized cells starved of a carbon source and pretreated with 1 mM NaN3 (䊐). The assay was performed in triplicate as shown.
reaction mixture were added in quadruplicate with a multichannel pipette to a 96-well filtration plate containing Durapore hydrophilic low-protein-binding membranes with a pore size of 0.65 or 1.2 m in a Millipore Multiscreen filtration apparatus. The trapped cells were washed free of excess drug by thrice passing ice-cold MES buffer (0.5 ml) through the filter. The maximum concentration of drug in the cell at the steady state was calculated by assuming an average cell volume of 76 m3 (2). Reagent. The reagent used was caspofungin acetate (MK-0991, L-743,872; Merck Pharmaceuticals). It has a molecular weight of 1,213.42.
RESULTS Drug aggregation. The uptake of [3H]caspofungin was initially evaluated in a rapid filtration assay. Unfortunately, this assay could not adequately distinguish cell-mediated uptake of the labeled drug from that of the drug remaining in the reaction medium because of the presence of a high background level. The principal cause of the high background level was a significant population of large drug complexes formed in free solution that were trapped on 1.2-m-pore-size filters during filtration. These large aggregates were readily observed following filtration of the transport medium in the absence of cells. The aggregates appeared to be in equilibrium with a smaller monomer population, since a portion (5 to 10%) of the labeled drug passed through the filters. As would be expected, the extent of the aggregated drug pool was concentration dependent. The aggregation phenomenon could be minimized below 10 g/ml, but it was still significant, which prevented a filtration assay from being used. It was found that, to separate cells from the labeled drug in solution, rapid centrifugation at 12,000 ⫻ g enabled cells to be effectively separated from all forms (monomeric and multimeric) of the drug in free solution. Active versus passive drug uptake. The drug uptake was assessed for cells under energized and de-energized cellular conditions to determine if the transport process was energy dependent. Addition of [3H]caspofungin to cells preincubated with glucose resulted in rapid uptake of the drug, which
reached saturation within about 10 min (Fig. 1). To determine whether the uptake kinetics were energy dependent, cells were first de-energized by incubation of cells without a carbon source at 30°C for 1 h and then preincubated for 20 min with the metabolic inhibitor sodium azide (1 mM). The energy status of the cells was assessed by measuring H⫹-ATPasemediated proton efflux (medium acidification) from cells in response to the addition of glucose (14). Addition of glucose to cells resulted in strong medium acidification over 30 min, while in the absence of an added carbon source and in the presence of the metabolic poison azide, cells failed to show acid efflux, confirming their energy-replete status (data not shown). When energy-depleted cells were challenged with the drug, the initial rate and saturation kinetics for uptake were nearly identical to that of glucose-metabolizing cells, although the final uptake level in energized cells was consistently lower (Fig. 1). There was no impact of the drug on cell viability in the time frame examined, which could have influenced uptake. Overall, the nearly identical nature of the net uptake kinetics for energized and de-energized cells indicates that transport is independent of the energy state of the cell and is not mediated via active transport. Drug uptake appeared to be diffusion limited, since the calculated concentration of labeled drug in the cell (1.4 ⫾ 0.21 M) was not statistically significantly different from the starting drug concentration outside the cell (0.96 M). Evidence for carrier-mediated transport. A series of experiments were initiated to examine saturation kinetics for uptake as a function of concentration to determine whether drug uptake was mediated via a facilitated-diffusion carrier. Because of limited amounts of labeled drug at relatively low specific activity, it was not possible to adopt a classical approach to assess uptake with increasing amounts of labeled drug over a wide range of concentrations. Two alternative approaches were used to provide evidence for the presence of a saturable carrier. First, cells were preloaded with increasing amounts of
VOL. 48, 2004
FIG. 2. Net uptake of caspofungin into cells preloaded with unlabeled drug. Glucose-energized cells were preloaded with unlabeled drug at 1 (⽧), 6 (䊐), or 11 (Œ) g/ml for 30 min and then diluted into an excess volume containing 1.2 g of [3H]caspofungin per ml. Uptake of label was monitored in triplicate over 30 min.
unlabeled drug (0, 1, 6, and 11 g/ml) for 30 min and then rapidly diluted into an excess volume containing labeled drug at 1.2 g/ml; uptake was determined as a function of time. Figure 2 shows that the level of uptake was largely independent of the amount of unlabeled drug preloaded in the cells and is consistent with an exchange carrier moving unlabeled drug out of the cell and labeled drug into the cell. If the system were purely a dissipative channel or diffusion through the lipid phase, then the high concentration of unlabeled drug within the cell would prevent uptake of labeled drug. The reduced net uptake at 30 min with the highest level of unlabeled preloaded drug can be explained by proposing that as unlabeled drug leaked from the cells, it competed with labeled drug at the uptake face of the carrier. To further assess the role of a facilitated-diffusion carrier, uptake of labeled drug (1.2 g/ml) was monitored as a function of time in the presence of increasing amounts of external unlabeled drug (0, 1, 5, 10 g/ml). Figure 3 shows that the presence of unlabeled drug competed with labeled drug, effec-
FIG. 3. Effect of competition by external drug on net uptake. Cells were added to a medium containing [3H]caspofungin at 1.2 g/ml/and unlabeled drug at 0 (〫), 1 (䊐), 5 (⫻), and 10 (‚) g/ml. Uptake was measured in triplicate over 30 min.
CASPOFUNGIN UPTAKE IN CANDIDA ALBICANS
3847
FIG. 4. Efflux of caspofungin from preloaded cells. Cells were preloaded with [3H]caspofungin at 11.2 g/ml for 30 min. The cells were diluted into a 50-fold excess volume of buffer containing 5 g of unlabeled caspofungin per ml, and the amount of labeled drug associated with the cells was determined as a function of time. Efflux was evaluated in both glucose-energized (⽧) and de-energized (carbon starved and treated with 1 mM NaN3) (䊐) cells.
tively reducing the apparent net drug uptake. Both internal loading and external competition data are consistent with the presence of a saturable facilitated-diffusion carrier that is operational at low drug levels. Equilibrium exchange was used to show that the putative transporter had a Km of ⱕ0.96 M (data not shown). Efflux kinetics. Net drug uptake is the sum of unidirectional influx and efflux as a function of time. The efflux component for caspofungin was determined by preloading cells with saturating amounts of drug at 11.2 g/ml for 10 min and then diluting the cells into medium in the absence of the drug to determine the amount of drug left in the cells as a function of time. Efflux was extremely rapid with a simple declining exponential decay, suggesting the emptying of a single transport compartment (Fig. 4). The rate of efflux was independent of energy status, as glucose-energized and carbon-starved and azole-treated cells showed the same apparent rate of efflux (Fig. 4). Importantly, more than 90% of the drug was removed from the cells, indicating that most of the transportable drug was in free solution and was not bound significantly to cellular components. Effect of external drug on efflux kinetics. To determine whether drug efflux was occurring via reversal of an uptake carrier, energy-replete cells preloaded with labeled drug were diluted into medium containing increasing amounts of unlabeled drug ranging from 0 to 10 g/ml. Figure 5 shows the exponential efflux of drug plotted as a log function of time. In the absence of drug or at low levels outside, the initial rate of efflux was extremely fast, exceeding the zero-time point, which was actually between 10 and 15 s. As the level of drug was increased on the outside, the apparent efflux was slowed up to a saturation of 5 g/ml. The second, slower phase of efflux, especially in the absence of external unlabeled drug, reflects a basic property of most carriers in that the conformational changes of the loaded carrier are faster than those of the
3848
PADERU ET AL.
FIG. 5. Effect of external drug on efflux kinetics. Cells were preloaded with [3H]caspofungin at 11.2 g/ml for 30 min and then diluted into a 50-fold excess volume containing unlabeled drug at 0 (〫), 1 (䊐), 5 (‚), and 10 (⫻) g/ml. Labeled drug remaining in the cells was measured in triplicate over a 45-min period.
unloaded carrier. If the unloaded carrier returns slowly, this will be the rate-limiting step of the overall transport scheme and entry of labeled drug will be limited by the rate of return of the free carrier. DISCUSSION The transport of most drugs across the cell membrane represents a critical step in the interaction of a drug with its cellular target and hence can influence its overall efficacy. For many drugs, aqueous diffusion is an important part of the overall mechanism of drug transport, since it is this process that delivers drug molecules to and from the nonaqueous barrier. In this investigation, we have performed a preliminary characterization of the transport properties of the new echinocandin drug caspofungin across the cell membrane of C. albicans. Evidence was provided for a saturable, facilitateddiffusion carrier (Fig. 1, 2, and 5). Drug uptake by this system is energy independent (Fig. 1), as is measured efflux from preloaded cells (Fig. 4). In each case, drug transport showed concentration-dependent saturation and was diffusion limited since the calculated level of the drug approximated that in the external medium. Drug entering the cell was not bound and appeared in a single kinetic compartment presumed to be the cytoplasm, since more than 90% of the drug in preloaded cells was recovered as a single declining exponential decay (Fig. 4). The data support the presence of a facilitated-diffusion carrier that transports the drug at low levels (Km, ⬍1.2 g/ml) and operates freely in both directions. It cannot be ruled out that non-carrier-mediated diffusion occurs at higher levels or that separate transporters exist for influx and efflux. The presence of a defined facilitated-diffusion-type carrier system for caspofungin uptake at fungicidal levels is significant because it suggests an alternative mechanism that could promote reduced susceptibility. Provided that drug uptake is required for inhibition of glucan synthase, a decrease in the affinity of the drug carrier or a change in the drug translocation rate could potentially result in reduced susceptibility. The development of secondary resistance by Candida spp. following
ANTIMICROB. AGENTS CHEMOTHER.
clinical drug exposure is a rare event (3). Yet, limited resistance has been observed in in vitro studies and has been linked to mutations in the FKS1 subunit of the glucan synthase enzyme complex (4–6, 9). Non-FKS1-mediated resistance mechanisms have been proposed, including low-level resistance due to overexpression of Cdr2p (15) or Golgi protein Sbe2p (8). Furthermore, microarray-based scanning of the Saccharomyces cerevisiae genome has revealed a cascade of cell wall biosynthetic genes being overexpressed in response to a caspofungin challenge, as well as many other essential genes, including putative transporters, suggesting that numerous genes could influence the development of resistance (1). The presence of a defined high-affinity uptake transporter for caspofungin may provide another potential mechanism to develop resistance if cellular drug uptake is essential for its mechanism of action. Identifying the transporter will be an important first step toward exploring this possibility. Finally, a dual-uptake model is proposed to account for caspofungin transport. At low drug levels, at or below ⬃1 g/ml, a high-affinity facilitated-diffusion carrier is suggested to mediate drug uptake into the cell. At higher drug levels, nonspecific drug uptake could occur through normal diffusion pathways across the bilayer of the plasma membrane. A diffusion coefficient for caspofungin has not been determined for the low-affinity uptake system, although such measurements will be complicated by the physical aggregation properties of the drug observed in high-affinity transport experiments. ACKNOWLEDGMENT This work was supported by a Merck Medical School grant to D.S.P. REFERENCES 1. Agarwal, A. K., P. D. Rogers, S. R. Baerson, M. R. Jacob, K. S. Barker, J. D. Cleary, L. A. Walker, D. G. Nagle, and A. M. Clark. 2003. Genome-wide expression profiling of the response to polyene, pyrimidine, azole, and echinocandin antifungal agents in Saccharomyces cerevisiae. J. Biol. Chem. 278: 34998–35015. 2. Baev, D., X. S. Li, J. Dong, P. Keng, and M. Edgerton. 2002. Human salivary histatin 5 causes disordered volume regulation and cell cycle arrest in Candida albicans. Infect. Immun. 70:4777–4784. 3. Denning, D. 2003. Echinocandin antifungal drugs. Lancet 362:1142–1151. 4. Douglas, C. M. 2001. Fungal (1,3)-D-glucan synthesis. Med. Mycol. 39:55– 66. 5. Douglas, C. M., F. Foor, J. A. Marrinan, N. Morin, J. B. Nielsen, A. M. Dahl, P. Mazur, W. Baginsky, W. Li, M. el-Sherbeini, et al. 1994. The Saccharomyces cerevisiae FKS1 (ETG1) gene encodes an integral membrane protein which is a subunit of 1,3--D-glucan synthase. Proc. Natl. Acad. Sci. USA 91:12907–12911. 6. Douglas, C. M., J. A. Marrinan, W. Li, and M. B. Kurtz. 1994. A Saccharomyces cerevisiae mutant with echinocandin-resistant 1,3--D-glucan synthase. J. Bacteriol. 176:5686–5696. 7. Kelly, R., E. Register, M. J. Hsu, M. Kurtz, and J. Nielsen. 1996. Isolation of a gene involved in 1,3--glucan synthesis in Aspergillus nidulans and purification of the corresponding protein. J. Bacteriol. 178:4381–4391. 8. Kontoyiannis, D. P., R. E. Lewis, G. S. May, N. Osherov, and M. G. Rinaldi. 2002. Aspergillus nidulans is frequently resistant to amphotericin B. Mycoses 45:406–407. 9. Kurtz, M. B., G. Abruzzo, A. Flattery, K. Bartizal, J. A. Marrinan, W. Li, J. Milligan, K. Nollstadt, and C. M. Douglas. 1996. Characterization of echinocandin-resistant mutants of Candida albicans: genetic, biochemical, and virulence studies. Infect. Immun. 64:3244–3251. 10. Mazur, P., and W. Baginsky. 1996. In vitro activity of 1,3--D-glucan synthase requires the GTP-binding protein Rho1. J. Biol. Chem. 271:14604–14609. 11. Mazur, P., N. Morin, W. Baginsky, M. el-Sherbeini, J. A. Clemas, J. B. Nielsen, and F. Foor. 1995. Differential expression and function of two homologous subunits of yeast 1,3--D-glucan synthase. Mol. Cell, Biol. 15: 5671–5681. 12. Mio, T., M. Adachi-Shimizu, Y. Tachibana, H. Tabuchi, S. B. Inoue, T. Yabe, T. Yamada-Okabe, M. Arisawa, T. Watanabe, and H. Yamada-Okabe. 1997. Cloning of the Candida albicans homolog of Saccharomyces cerevisiae. GSC1/
VOL. 48, 2004 FKS1 and its involvement in -1,3-glucan synthesis. J. Bacteriol. 179:4096– 4105. 13. Pereira, M., M. S. Felipe, M. M. Brigido, C. M. Soares, and M. O. Azevedo. 2000. Molecular cloning and characterization of a glucan synthase gene from the human pathogenic fungus Paracoccidioides brasiliensis. Yeast 16:451–462. 14. Perlin, D. S., D. Seto-Young, and B. C. Monk. 1997. The plasma membrane H⫹-ATPase of fungi. A candidate drug target? Ann. N. Y. Acad. Sci. 834: 609–617. 15. Schuetzer-Muehlbauer, M., B. Willinger, G. Krapf, S. Enzinger, E. Presterl, and K. Kuchler. 2003. The Candida albicans Cdr2p ATP-binding cassette
CASPOFUNGIN UPTAKE IN CANDIDA ALBICANS
3849
(ABC) transporter confers resistance to caspofungin. Mol. Microbiol. 48: 225–235. 16. Soteropoulos, P., T. Vaz, R. Santangelo, P. Paderu, D. Y. Huang, M. J. Tamas, and D. S. Perlin. 2000. Molecular characterization of the plasma membrane H⫹-ATPase, an antifungal target in Cryptococcus neoformans. Antimicrob. Agents Chemother. 44:2349–2355. 17. Thompson, J. R., C. M. Douglas, W. Li, C. K. Jue, B. Pramanik, X. Yuan, T. H. Rude, D. L. Toffaletti, J. R. Perfect, and M. Kurtz. 1999. A glucan synthase FKS1 homolog in Cryptococcus neoformans is single copy and encodes an essential function. J. Bacteriol. 181:444–453.
