INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

The synthetic cyclopeptamine A172013 (Fig. 1) has some similarity in structure to the echinocandins and aculeacins. These compounds are noncompetitive inhibitors of -(1,3)-glucan synthesis (22, 23, 32) and have fungicidal activity against some fungi (3). Compound A172013 differs from the lipopeptide echinocandin B in the following four residues; (i) 4-aminoproline in the place of 3-hydroxy-4-methylproline, (ii) hydrogen substituted for the hydroxyls at C-3 and C-4 of homotyrosine, (iii) hydrogen substituted for the hydroxyls at C-4 and C-5 of ornithine, and (iv) a pentyloxyterphenyl group in place of the linoleoyl side chain. Compounds of this class are composed of a somewhat hydrophilic peptide core and a lipophilic side chain. Generally, their antifungal activity requires conservation of the homotyrosine, the two -hydroxy amino acids adjacent to the prolines, and a lipophilic side chain (3, 31).

View larger version (15K): [in this window] [in a new window]   FIG. 1.   Chemical structure of A172013.

Although these compounds inhibit -(1,3)-glucan production and thus affect the integrity of the cell wall, the mechanism of drug transport has not yet been elucidated. The lipophilic side chain most likely assists in the partitioning of these compounds into the amphiphilic fungal membrane, and it has been reported that the inhibition of glucan synthesis can be partially attributed to the membrane-disruptive effects of lipophilic agents (20). Cilofungin, a semisynthetic lipopeptide, has been used as a fluorescent probe to demonstrate the direct interaction of this drug with fungal microsomal membranes and phosphatidylcholine membrane vesicles (21). In order to better understand the events occurring at the cellular level, we need to know how this class of compound enters the fungal cells, where the drug localizes, and what residual activity can be detected at the critical cellular component (microsomes). In this report, we demonstrate the energy-independent passive diffusion of a radiolabeled cyclopeptamine into fungi, as well as the localization and residual anti--glucan synthase activity of the drug in cellular components of Candida albicans CCH442.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Chemicals. The cyclopeptamine A172013 was prepared at Abbott Laboratories, Abbott Park, Ill. (25). [terphenyl-³H]A172013 (910 mCi/mmol) was also prepared at Abbott Laboratories. UDP-[U-¹⁴C]glucose (0.6 mCi/mmol) and [5,6-³H]uridine (45 Ci/mmol) were purchased from Amersham Corp., Arlington Heights, Ill. Bovine serum albumin (BSA) was obtained from Calbiochem, San Diego, Calif. Growth media were purchased from Difco Laboratories, Detroit, Mich., and from Bio 101, Vista, Calif.; silicone oil was purchased from Accumetric Inc., Elizabethtown, Ky.; and other chemicals were purchased from Sigma Chemical Co., St. Louis, Mo.

Organisms and growth conditions. C. albicans CCH442, a standard culture at Abbott Laboratories, was obtained as a clinical isolate from Cook County Hospital, Chicago, Ill. C. albicans LP3-1 was produced by UV mutagenesis of strain CCH442 (14). C. albicans HOG301 was obtained from R. Poulter, Department of Biochemistry, University of Otago, Otago, New Zealand. Saccharomyces cerevisiae SEY2101 was obtained from V. Mrsa, Laboratory of Biochemistry, University of Zagreb, Zagreb, Croatia. Aspergillus niger ATCC 16404 was obtained from the American Type Culture Collection, Rockville, Md. C. albicans CCH442 (lipopeptide/cyclopeptamine sensitive) and LP3-1 (lipopeptide/cyclopeptamine resistant) were grown at 30°C in yeast nitrogen base (YNB) plus 0.5% (wt/vol) glucose. C. albicans HOG301 (filamentous growth) and S. cerevisiae SEY2101 were grown at 30°C in YNB plus 0.5% glucose and complete supplement mixture (CSM). All cultures were grown to a final A₄₂₀ of 1.0. A. niger ATCC 16404 spores were initially grown on 1% (wt/vol) yeast extract, 2% (wt/vol) peptone, and 0.5% (wt/vol) glucose (YPD)-agar plates at 35°C for 3 to 6 days to induce additional spore production. The harvested spores (5.5 × 10⁸) were then germinated in 50 ml of YNB plus 2% glucose at 35°C for 12 h.

MIC determinations. Organisms were grown at 30°C in YNB plus 0.5% (wt/vol) glucose or in YNB plus 2% glucose and CSM to approximately 10⁹ to 10¹⁰ CFU/ml. MICs were determined by the broth microdilution method previously described (10). The cells were diluted to provide a final inoculum of 2 × 10⁵ CFU/ml. An equal volume of the cell suspension was then added to a microtiter plate of twofold serial dilutions of test compounds. Growth was determined visually after 24 h at 30°C.

Cyclopeptamine uptake and displacement studies. Cell cultures were harvested by centrifugation (at 10,000 × g for 10 min) and washed once with transport buffer (0.01 M potassium phosphate [pH 6.8]-YNB [without amino acids]-5% glucose) (7, 28). Cells were resuspended in 1/10 volume transport buffer. Resuspended cells (A₄₂₀ = 10) (1.5 × 10⁸ CFU/ml) were mixed with an equal volume of transport buffer containing 0.05 to 10 µg of [³H]A172013/ml (0.08 to 1.7 µCi/ml) and were incubated at 23°C for 5 min or for the time periods indicated in figures. Cells were separated from the transport buffer by centrifugation (at 13,000 × g for 1 min) through silicone oil (specific gravity, 1.02 g/ml) as previously described (5, 6, 11). Dry weights of cells were determined by trapping cells on Whatman GF/F glass fiber filters and drying by microwave (2). Uptake was expressed as picomoles of drug per milligram of cells (dry weight).

RNA labeling with [³H]uridine. C. albicans CCH442 cells were grown in YNB plus CSM containing 10 µg of uracil/ml, 1% glucose, and 125 µCi of [³H]uridine (45 Ci/mmol). Cells were harvested by filtration at an A₄₂₀ of 8.0, washed, and resuspended in medium containing 100 µg of unlabeled uridine/ml. Cells were grown to an A₄₂₀ of 0.9, harvested by centrifugation, washed once with transport buffer, and then resuspended in transport buffer at an A₄₂₀ of 10. Radiolabeled cells were treated with 0.1 µg of unlabeled A172013/ml at 23°C, centrifuged through silicone oil at various time points after drug addition, and then processed as described above. The comparison of radiolabel (tritiated RNA) in the cell pellets of drug-treated versus untreated groups provided a measure for cell lysis.

Cyclopeptamine localization studies. A 200-ml C. albicans CCH442 culture (A₄₂₀ = 1.0) was incubated at 30°C for 5 min with 1.0 µg of [³H]A172013/ml (0.9 µCi/ml). Cells were harvested by centrifugation (at 1,000 × g for 5 min), and the cell pellet was processed for protoplast isolation (30). Briefly, the radiolabeled cell pellet was pretreated with 8 ml of 20 mM Tris-hydrochloride (Tris-HCl), pH 7.75, containing 0.5 mg of pronase/ml, 50 mM dithiothreitol (DTT), and 5 mM EDTA. Cells were shaken at 110 rpm for 30 min at 26°C. Cells were then washed twice (first salt wash) with 40 ml of 0.6 M KCl (4°C) and centrifuged at 1,000 × g for 10 min. The resulting pellet was resuspended in 4 ml of 0.6 M KCl containing 10 mg of Novozyme 234/ml and incubated at 26°C for 30 min with shaking (at 110 rpm). Protoplasts were harvested by centrifugation at 150 × g for 10 min and were washed twice (second salt wash) with 0.6 M KCl at 4°C. All supernatant fractions up to this point were considered part of the cell wall fraction. Microsomes were prepared from protoplasts by resuspending cells in 4 ml of lysis buffer (70 mM Tris-HCl [pH 8.0] containing 125 mM sucrose, 2 mM EDTA, 1.5 mM EGTA, 4 mM DTT, 10 mM -mercaptoethanol, and 25 µM GTP) and homogenizing in a Dounce homogenizer at 4°C. A sample was centrifuged at 650 × g for 10 min, and the supernatant was retained for ultracentrifugation while the pellet was rehomogenized in a Dounce homogenizer at 4°C. A sample from the Dounce homogenizer was centrifuged at 150 × g for 10 min, and supernatants were combined with those from the previous spin (the pellet contained nuclei). The supernatants were recentrifuged at 100,000 × g for 1 h to obtain a microsome membrane pellet. Microsome preparations were stored in 50 mM Tris-HCl (pH 7.5) containing 1 mM EDTA, 1 mM DTT, and 33% (vol/vol) glycerol at 80°C.

-(1,3)-Glucan synthase assay. The assays of microsome preparations were carried out in a final volume of 100 µl as previously described (12, 13). Microsomes (30 µg) were incubated in 80 mM Tris-HCl (pH 7.75) containing 1 mM EDTA, 8% (vol/vol) glycerol, 20 µM GTPS [guanosine-5'-0-(3-thiotriphosphate], 0.2 mM DTT, 1.6 mg of BSA/ml, and 1.0 mM UDP-[¹⁴C]glucose (0.125 mCi/mmol). Assays were initiated by substrate, mixtures were incubated for 45 min at 30°C, and reactions were stopped by addition of 100 µl of ethanol. Samples were transferred to a 96-well MilliBlot-D filtration apparatus (Millipore, Bedford, Mass.) containing a type G-10 glass fiber filter (Inotech, Lansing, Mich.). Each well was washed once with 100 µl of distilled deionized water followed by 100 µl of 10% trichloroacetic acid (TCA). Samples were then washed three times with 100 µl of distilled deionized water. The filters were removed, air dried, and counted on a Topcount scintillation counter (Packard Instruments, Downers Grove, Ill.). One unit of activity is defined as one nanomole of glucose incorporated into TCA-insoluble glucan per minute. An estimated 50% inhibitory concentration (IC₅₀) was determined by using the logistic function y = [(a  d)/1 + (I/c)^b] + d (from reference 9), restated as c = [(a/y)  1] × [I], where y is the inhibitor response, a is the response when I = 0, I is the concentration of inhibitor, and c is IC₅₀.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

MICs. The MICs of A172013 for C. albicans CCH442, LP3-1, and HOG301, S. cerevisiae SEY2101, and A. niger ATCC 16404 were 0.39, 50, 0.78, 0.78, and 6.25 µg/ml, respectively.

Cellular accumulation of A172013. Cellular accumulation of the cyclopeptamine (0.1 µg/ml) was rapid, reaching maximum levels after 5 min postaddition in most fungi tested (Fig. 2). Cell-associated drug levels peaked after 5 min at 24.3 and 18.1 pmol/mg of cells (dry weight) in sensitive (CCH442) and resistant (LP3-1) C. albicans strains, respectively. Over the next 55 min, the drug level in strain CCH442 decreased by 9.2 pmol, to 15.1 pmol/mg of cells (dry weight), while in strain LP3-1 a smaller drop of 3.4 pmol, to 14.7 pmol/mg of cells (dry weight), was observed. Since CCH442 cells containing tritiated RNA did not lose any significant cellular label when treated with the cyclopeptamine (0.1 µg/ml) for 60 min, drug loss due to cell lysis was ruled out. In addition, the initial accumulation of drug (0 to 5 min) in strain CCH442 was not inhibited by metabolic inhibitors (Table 1).

View larger version (19K): [in this window] [in a new window]   FIG. 2.   Cellular accumulation of A172013 in several fungi. Concentrated cells (7.5 × 107 CFU/ml) in transport buffer were incubated with 0.1 µg of [3H]A172013/ml at 23°C for the times indicated. Duplicate samples were spun through silicone oil as described in Materials and Methods. Results are presented as averages of two values per time point after subtraction of a zero time point. Symbols: , C. albicans CCH442; , C. albicans LP3-1; , C. albicans HOG301; , A. niger ATCC 16404; , S. cerevisiae SEY2101.

View larger version (17K): [in this window] [in a new window]   FIG. 3.   Displacement of [3H]A172013 from C. albicans. Concentrated cells (7.5 × 107 CFU/ml) were incubated with 0.1 µg of [3H]A172013/ml at 23°C for 5 min and centrifuged; then the cell pellet was resuspended in medium containing no drug (, sensitive strain CCH442) or 1.0 µg of unlabeled drug/ml (, resistant strain LP3-1). Triplicate samples were taken at the times indicated and processed as for Fig. 1. Results are presented as means ± standard deviations for the triplicate samples.

12

7

8

View larger version (14K): [in this window] [in a new window]   FIG. 4.   Saturation curve of A172013 uptake in C. albicans CCH442. A range of 0.05 to 10 µg of [3H]A172013/ml was incubated at 23°C for 5 min with concentrated cells, and triplicate samples were processed as for Fig. 1. Results are presented as means ± standard deviations for the triplicate samples.

Drug localization in cell wall, protoplasts, and microsomes. When 200 ml of C. albicans CCH442 (1.5 × 10⁷ CFU/ml) was incubated with 1.0 µg of [³H]A172013/ml at 30°C for 5 min, 68.8% of the label (0.59 µCi/ml) was associated with the cells. Treatment of the labeled cells with various protocols (exposure to proteases, Novozyme 234, salt washes, sulfhydryl, and chelating agents) designed to remove the cell wall and produce protoplasts resulted in the recovery of 55 and 22.9% of the cell-associated label in the cell wall (14 × 10⁶ molecules) and protoplast (6 × 10⁶ molecules), respectively (Table 2). After disruption of the protoplasts, about 21.7% of the cell-associated label was recovered in the microsomal fraction. Based on the specific activity (910 mCi/mmol) of the label, and assuming no metabolism of the drug, the drug concentration in the microsome preparation after resuspension was calculated to be 1.69 µg/ml. A total of 77.9% of the cell-associated label was accounted for, and the balance was presumably lost during manipulation of the samples.

Microsomal glucan synthesis activity. The labeled microsomes recovered in the localization study were compared to drug-free microsomes by using the glucan synthesis assay. The labeled microsomes (1.69 µg of drug/ml), with a glucan synthase activity of 0.21 ± 0.04 nmol/min/mg, demonstrated 80.7% inhibition of glucan production compared to control microsomes, with a glucan synthase activity of 1.09 ± 0.15 nmol/min/mg. An IC₅₀ of 0.40 µg/ml was estimated from these data by using the formula IC₅₀ = [(100/%I)  1] × [I], where %I is 80.7 and [I] is 1.69 µg/ml. The actual IC₅₀ of 0.48 ± 0.18 µg/ml was determined from a dose response by using A170213 and unlabeled microsomes (data not shown).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The cyclopeptamine A172013 accumulated rapidly in several fungi, including a lipopeptide/cyclopeptamine-resistant strain of C. albicans (LP3-1), with the greatest accumulation occurring in C. albicans CCH442 and HOG301. The latter two strains also exhibited a similar decrease in cell-associated drug after peaking at 5 to 20 min after drug addition. This decrease in accumulation, which was observed to a smaller extent in strain LP3-1, was not due to cell lysis and remains unexplained. The sensitive strain CCH442 accumulated 25% more drug on a per-cell-weight basis than the resistant strain LP3-1. Since strain LP3-1 had an altered membrane component of -(1,3)-glucan synthase (14), the additional drug accumulation observed with strain CCH442 may have been facilitated by an unaltered glucan synthase enzyme. Also, drug accumulation in C. albicans CCH442 and HOG301 demonstrated the ability of the compound to penetrate both the yeast and hyphal forms equally, an important asset in the treatment of pathogenic fungi.

The difference in drug MICs for the resistant and sensitive C. albicans strains was not apparent from the small difference observed in drug accumulation. The resistance mechanism in strain LP3-1 had been attributed to an altered membrane component of -(1,3)-glucan synthase and not to membrane changes, since a profile of phospholipids, neutral lipids, and fatty acids from this strain was normal (14). Our accumulation data support the conclusion that the cell wall/membrane composition was not a major factor in the resistance mechanism of strain LP3-1.

Lipopeptides/cyclopeptamines weaken the structural integrity of the fungal cell wall by inhibiting the synthesis of -(1,3)-glucan (3, 15, 16, 23). A second mechanism of action for this class of drug was thought to be cell leakage and lysis caused by the penetration of the lipophilic side chain into the membrane (3, 8, 15). However, our study showed that the cell wall/membrane of the resistant organism, C. albicans LP3-1, did not provide an effective barrier against drug accumulation, and therefore the cells would have lysed if drug penetration into the membrane provided a significant disruptive effect. Moreover, investigators using stereoisomers of a semisynthetic pneumocandin analog have demonstrated enhanced antifungal activity attributable to the specific inhibition of glucan synthesis and not to nonspecific membrane effects (22).

The lower drug accumulation and higher MICs for A. niger may reflect a cell wall/membrane composition which was relatively less conducive to cyclopeptamine penetration compared to Candida. Aspergillus species were reported to be less susceptible to lipopeptides based on MIC data; however, drug effectiveness against Aspergillus was better demonstrated by morphological changes or in vivo efficacy (15, 22, 24). In fact, -(1,3)-glucan synthase isolated from A. fumigatus demonstrated a 10-fold lower K_i for cilofungin than enzyme isolated from C. albicans (4, 15).

The accumulation of A172013 in C. albicans CCH442 was not dependent on energy, indicating that passive diffusion and partitioning into the membrane were the driving forces. Inhibiting the electron transport system (NaN₃), glycolysis (NaF), or phosphorylation (NaAsO₂) or dissipating the membrane potential (CCCP) had no effect on drug uptake during the initial 5-min exposure. The partitioning of drug into cells was unsaturable and not readily displaced, indicating that the amphiphilic nature of the compound was driving drug-cell interactions. The negative surface charge of C. albicans cells (18, 19) was not a deterrent to drug entry due to the electrostatic interaction of the positively charged amine groups in the cyclic peptide moiety with the cell surface. This initial interaction could facilitate the introduction of the lipophilic side chain to hydrophobic regions of the cell wall/membrane or the formation of drug aggregates at the cell surface which would lead to cell entry (1).

Localization studies demonstrated that the drug was primarily distributed between the cell wall and protoplast, the former accounting for a majority of the cell-associated compound. The drug recovered from the protoplasts was nearly all associated with the microsomes and was fully active in vitro. Our estimated IC₅₀ for the concentration of drug recovered (1.69 µg/ml) from labeled microsomes was within the standard deviation of the actual IC₅₀ for this compound tested in microsome preparations isolated in parallel from control protoplasts. Although similar compounds were thought to exert their inhibition of glucan synthesis by attachment to the FKS-encoded integral membrane fraction of glucan synthase (23), further investigations are needed to identify the exact site of drug-enzyme interaction.

Cells incubated with 1.0 µg of drug/ml (2.5× MIC) for 5 min at 30°C accumulated 2.57 × 10⁷ molecules of drug/cell, and 55% (1.4 × 10⁷ molecules) was associated with the cell wall. Our estimate of the volume of the wall space per cell was 2.95 × 10¹² cm³, based on measurements taken from thin-section electron micrographs. From these data, the spatial concentration of drug in the wall was estimated to be 7.82 × 10³ M, or about 8,500 times the original extracellular concentration. Thus, it appears that some mechanism for the selective partitioning of drug into the wall area exists. The 6 × 10⁶ molecules localized in the microsomal membrane fraction are likely to be present at an even greater concentration, given the expected smaller volume of membrane versus wall. Based on the above calculations, exposure of cells to the MIC of the drug (0.4 µg/ml) would produce 3.1 × 10³ M concentrations in the cell wall and some value greater than this in the membrane fraction. Therefore, the dynamics of drug interaction with the glucan synthase target (i.e., drug entry into the cell wall, followed by diffusion into and within the plasma membrane, followed by association with the target) are obviously complex for this class of drug, requiring more than traditional interpretations to integrate drug potency against the isolated target enzyme versus the whole cell.

The diffusion of A172013 from medium to membrane-bound target or receptor may initially involve electrostatic interactions at the cell surface followed by hydrophobic and electrostatic interactions with the plasma membrane. Amphiphilic compounds are known to position at the interface between the lipophilic interior and the aqueous head groups of lipid bilayers (17). The difficulty in displacing A172013 from cells suggested that membrane partitioning was involved. The partitioning process resulted in the concentrating and possibly positioning of the drug for interaction with the membrane-bound receptor (26, 29). There may also be lateral two-dimensional diffusion within the membrane which could be advantageous to the overall drug entry rate if the compound is properly oriented for receptor recognition and binding (26, 27).

In summary, the cyclopeptamine A172013 rapidly partitioned into the cell wall of C. albicans by passive diffusion. Drug accumulation was unsaturable up to 10 µg/ml, and drug was not readily displaced from the cells into drug-free medium or medium containing 10 times the concentration of unlabeled drug. Protoplast- or microsome-associated drug was fully active against glucan synthesis in vitro.