Proton-Pumping-ATPase-Targeted Antifungal Activity of a Novel Conjugated Styryl Ketone

Abstract

NC1175 (3-[3-(4-chlorophenyl)-2-propenoyl]-4-[2-(4-chlorophenyl)vinylene]-1-ethyl-4-piperidinol hydrochloride) is a novel thiol-blocking conjugated styryl ketone that exhibits activity against a wide spectrum of pathogenic fungi. Incubation of NC1175 with various concentrations of cysteine and glutathione eliminated its antifungal activity in a concentration-dependent fashion. Since NC1175 is a lipophilic compound that has the potential to interact with cytoplasmic membrane components, we examined its effect on the membrane-located proton-translocating ATPase (H⁺-ATPase) of yeast (Candida albicans, Candida krusei, Candida guilliermondii, Candida glabrata, and Saccharomyces cerevisiae) and Aspergillus (Aspergillus fumigatus, Aspergillus niger, Aspergillus flavus, and Aspergillus nidulans) species. The glucose-induced acidification of external medium due to H⁺-ATPase-mediated expulsion of intracellular protons by these fungi was measured in the presence of several concentrations of the drug. NC1175 (12.5 to 50 μM) inhibited acidification of external medium by Candida, Saccharomyces, and Aspergillus species in a concentration-dependent manner. Vanadate-inhibited hydrolysis of ATP by membrane fractions of C. albicans was completely inhibited by 50 μM NC1175, suggesting that the target of action of NC1175 in these fungi may include H⁺-ATPase.

Fungal diseases in humans have increased significantly with the advent of an expanding population of immunosuppressed patients and with the introduction of sophisticated life-saving medical procedures. Microorganisms once considered to be commensals have become opportunistic pathogens responsible for severe and often fatal infections in humans (2, 5, 7, 15). The prevalence of hospital-acquired fungal infections has almost doubled in recent years and now accounts for as many as 10 to 15% of such infections (6). Among the nosocomial fungal infections, pathogenic yeasts account for up to 70% of the such infections (6). Equally alarming is the rapid increase in the incidence of invasive aspergillosis in AIDS patients (27, 28, 34, 53, 62), solid-organ (45, 59) and bone marrow transplant (25, 40, 44, 47, 61) patients, leukopenic compromised hosts receiving chemotherapy (19, 24, 67), and patients on corticosteroid therapy (34). Although fungal infections in humans have increased significantly in recent years, no major improvement in the treatment and management of fungal infections has occurred. In fact, the management of fungal infections has become more complicated now due to the emergence of resistance to commonly used antifungal agents (12, 43, 54–56, 63–66). Thus, the need for new antifungal agents directed to novel fungal targets is greater than ever before.

Recently, we examined various Mannich bases of a series of cyclic conjugated styryl ketones for their antifungal activities (32). These compounds were designed as thiol-alkylators and had two centers for attack by cellular thiols. They contained a conjugated styryl keto moiety and were chosen because of their affinity for thiols but not amino or hydroxy groups, which are found in nucleic acids (4, 14). Thus, the potential problems of mutagenicity and carcinogenicity (10, 18) may be avoided. Since the conversion of certain α,β-unsaturated ketones into the corresponding Mannich bases increased the rates of thiol alkylation considerably (13), Mannich bases of enones were prepared for antifungal evaluation. 3-[3-(4-Chlorophenyl)-2- propenoyl]-4-[2-(4-chlorophenyl)vinylene]-1-ethyl-4-piperidinol hydrochloride (NC1175) was the most potent member of this class of compounds, and it possessed a hydrophobic, electron-attracting substituent in the aryl rings. In vitro susceptibility studies (32) showed that NC1175 possessed fungicidal activity against a wide spectrum of pathogenic fungi including azole- and polyene-resistant isolates of Candida and Aspergillus species. Exposure of Candida and Aspergillus cells to MICs (1.56 to 6.25 μM) and super-MICs (25 to 100 μM) of NC1175 killed greater than 99.9% of the cells within 8 h. NC1175 showed promising activity against Candida albicans and Aspergillus fumigatus infections in murine vaginitis (unpublished data) and pulmonary aspergillosis (33) models as determined by the fungal burdens of infected animals. These encouraging results prompted us to study the mode of action of this conjugated styryl ketone in fungi.

The proton-translocating ATPase (H⁺-ATPase) of fungi is a plasma membrane-located ATP-driven proton pump belonging to the P-type ATPase superfamily. To date, 211 members (ranging in size from 646 to 1,956 amino acids) of the P-type ATPase have been identified in a wide spectrum of organisms ranging from archaebacteria to humans (3, 35). The charged substrates that the P-type ATPases translocate include Na⁺, K⁺, Ca²⁺, Mg²⁺, Cd²⁺, H⁺, and phospholipids. On the basis of their substrate specificities, the ion-translocating P-type ATPases are grouped into five (types I to V) families (3). The distinguishing feature of the P-type ATPases is the formation of a phosphorylated enzyme intermediate during the reaction cycle (hence, they are called P-type ATPases). The phosphorylation of the enzyme invariably involves an aspartic acid residue of a highly conserved motif consisting of DKTGT (3, 35, 58). The number of P-type ATPases present in an organism is highly variable, ranging from either a few (pathogenic bacteria), 7 to 9 (free-living bacteria), or as many as 16 in Saccharomyces cerevisiae to probably more than 30 members in plants and animals.

The H⁺-ATPase of fungi is encoded by the PMA1 gene (3, 58). Comparison of the predicted amino acid sequences revealed that the PMA1 gene products of S. cerevisiae (57), C. albicans (36), and Neurospora crassa (48) showed a high degree of relatedness (3). Detailed molecular and genetic studies (52) with the S. cerevisiae H⁺-ATPase revealed that it is an integral protein of which greater than 80% is exposed to the cytoplasmic side of the cell. Approximately 15% of the polypeptide is estimated to be associated with the lipid bilayer, forming 10 membrane-spanning α-helical regions, while the remaining 5% of the protein is exposed to the extracytoplasmic phase of the cell. Previous studies (39, 52) have shown that mutational alteration of the transmembrane segments of the enzyme affects its function carried out by the distal region of the enzyme exposed to the cytoplasmic side of the cell. The plasma membrane H⁺-ATPase plays an essential role in fungal cell physiology (58). This ion-translocating enzyme is mainly responsible for maintaining the electrochemical proton gradient necessary for nutrient uptake and the regulation of the intracellular pH of the fungal cell (58). Interference of the function of H⁺-ATPase in fungi by antagonists will lead to cell death. Thus, use of the plasma membrane H⁺-ATPase as a molecular target for antifungal drug therapy is an attractive possibility, provided that inhibition of the enzyme activity correlates with the cessation of cell growth. Monk and Perlin (38) previously reported that the anti-gastric ulcer drug omeprazole (a cysteine-modifying agent which inhibits the gastric K⁺ H⁺- ATPase) inhibited C. albicans growth, although high concentrations of the drug were needed. They further demonstrated that the inhibition of C. albicans growth was correlated with the inhibition of the H⁺-ATPase of this organism (37). Since the conjugated styryl ketone NC1175 showed fungicidal activity against a broad spectrum of fungi and was designed to react preferentially with thiols (41), including those associated with membrane proteins, we postulated that the fungal H⁺-ATPase may be a potential target for conjugated styryl ketones. We therefore investigated the effect of NC1175 on the H⁺-ATPases of Candida, Saccharomyces, and Aspergillus species.

MATERIALS AND METHODS

Organisms and culture conditions.

A. fumigatus W73355, Aspergillus niger S11335, and Aspergillus flavus I65850 were clinical isolates obtained from the Microbiology Laboratory, Detroit Medical Center, Wayne State University. Aspergillus nidulans A767 was obtained from the Fungal Genetics Stock Center, University of Kansas Medical Center, Kansas City. The primary cultures obtained from the respective sources were subcultured on peptone yeast extract glucose (PYG; peptone [1 g], yeast extract [1 g], and glucose [3 g] per liter of distilled water) agar to ensure the purity of the cultures. Working cultures were maintained on PYG agar at room temperature. Long-term storage of the cultures was done as conidial suspensions in 25% glycerol at −70°C.

C. albicans 90028, Candida glabrata 33554, Candida guilliermondii 9390, and Candida krusei 6258 were obtained from the American Type Culture Collection, Manassas, Va. Working cultures of Candida species were grown for 48 h at 30°C on Sabouraud dextrose agar from stock cultures stored at −70°C in litmus milk (Becton Dickinson Microbiology Systems, Cockeysville, Md.). Single colonies from 2-day-old cultures were used as the source of the inoculum for all subsequent experiments.

S. cerevisiae GW201 and the pma1 cysteine mutants (strains pma1.1 to pma1.5) derived from S. cerevisiae GW201 were kindly provided by David Perlin (Public Health Research Institute, New York, N.Y.). The cultures obtained on yeast-peptone-dextrose agar were maintained on Ura⁻ dropout plates (Clontech, Palo Alto, Calif.). Single colonies from 2-day-old cultures grown on Ura⁻ dropout plates were used as the sources of inoculum for all the subsequent experiments.

MIC determination. (i) Yeasts.

The MICs of various antifungal agents for C. albicans, C. glabrata, C. guilliermondii, C. krusei, and S. cerevisiae were determined by the broth microdilution method as recommended by the National Committee for Clinical Laboratory Standards (42) with RPMI 1640 as the growth medium. The MIC was defined as the lowest concentration of the drug that inhibited growth by 80% compared to the growth of the drug-free control after 48 h of incubation at 35°C. Determination of the MIC for each organism was repeated at least once, and the data were within ±1 dilution.

(ii) Aspergillus.

Conidial suspensions of A. fumigatus, A. niger, A. flavus, and A. nidulans were prepared and the MICs of various antifungal agents were determined as described previously (16, 17, 31). We used PYG broth instead of RPMI 1640 for MIC studies since the latter medium failed to discriminate between clinical isolates of microorganisms with reduced susceptibility to amphotericin B and those that were highly susceptible to the drug. Briefly, fresh conidia were resuspended in PYG medium at a density of 2 × 10⁴ conidia/ml. Twice the required concentrations of the drugs were prepared in PYG medium (0.5 ml) by serial dilution in sterile 6-ml polystyrene tubes (Falcon 2054; VWR Scientific, Philadelphia, Pa.), and the tubes were inoculated with an equal volume (0.5 ml) of the conidial suspension. The tubes were incubated at 35°C for 48 h and scored for visible growth after gentle vortexing of the tubes or scraping of the walls of the tube followed by vortexing. The MIC was defined as the lowest concentration of the drug in which no visible growth occurred. Determination of the MIC for each isolate was repeated at least once, and the data were within ±1 dilution.

NC1175 inactivation assay.

The inactivation of the antifungal activity of NC1175 by various thiol reagents such as cysteine, glutathione, dithiothreitol (DTT), and mercaptoethanol was examined by a C. albicans growth inhibition assay. Briefly, a series of concentrations of thiol reagents ranging from 0.0195 to 20 mM was prepared by twofold serial dilution in 0.5 ml of PYG broth in 6-ml polystyrene tubes (Falcon 2058). To each tube 5 μl of a 5 mM stock of NC1175 was added to obtain a final concentration of 25 μM, and each tube was then inoculated with 0.495 ml of C. albicans 90028 cell suspension from an appropriately diluted (≈2 × 10⁴ cells/ml) culture grown overnight in PYG broth. The tubes were incubated at 35°C for 24 h, and the C. albicans growth in each tube was determined by measuring the absorbance at 595 nm. An identical set of tubes with dimethyl sulfoxide (DMSO) instead of NC1175 was used as the control. The A₅₉₅ values obtained for the NC1175-treated and the control groups were plotted against various concentrations of thiol reagents. If the antifungal activity of NC1175 is nullified in the presence of a thiol reagent, then C. albicans will grow in NC1175-plus-thiol combinations in which the effective concentration of the drug falls below its MIC for C. albicans.

Measurement of acidification of external medium.

The proton-pumping activities of Candida, Saccharomyces, and Aspergillus species were determined by monitoring glucose-induced acidification of the external medium by measuring the pH with an electrode as described previously (39), with modifications as described below.

(i) Yeasts.

One-liter cultures of various Candida species and S. cerevisiae strains were grown in PYG broth for 18 h at 30°C on a gyratory shaker at 160 rpm. The cells were collected by centrifugation at 3,500 × g for 10 min at 4°C and washed with 1,000 ml each of sterile distilled water and 50 mM KCl (pH 6.5). The washed cells were resuspended in 500 ml of 50 mM KCl (pH 6.5) and incubated at 4°C overnight to deplete their carbon reserves. The carbon-starved cells were harvested by centrifugation, and the pellet was resuspended in 500 ml of 50 mM KCl (pH 6.5) at a cell density of ≈2 × 10⁸ cells/ml. To a 40-ml aliquot of the cell suspension, the inhibitor was added to obtain the required concentration and mixed well, and the volume was adjusted to 45 ml with 50 mM KCl. The cell suspension was incubated with the inhibitor at room temperature with gentle stirring for 10 min, and then 5 ml of 20% glucose (final concentration, 55 mM) was added and the pH of the external medium was monitored at regular intervals for 60 min. The experiment was performed in the presence of a comparable concentration of DMSO (control) to measure the extent of acidification of the external medium in the absence of the inhibitor.

(ii) Aspergillus.

One-liter cultures of various Aspergillus species were grown from conidial suspensions in PYG medium for 24 h at 35°C. The resulting mycelia were harvested by filtration and washed with distilled water and then with 50 mM KCl (1,000 ml each). Approximately 0.5-g (wet weight) amounts of the washed mycelia were resuspended in 40-ml aliquots of 50 mM KCl, and the suspensions were incubated at 4°C overnight (18 h) for glucose starvation. An inhibitor was added to the glucose-starved mycelial suspension to obtain the required concentration, and the volume was adjusted to 45 ml with the addition of 50 mM KCl. The mycelial suspension was then incubated at room temperature for 10 min, and glucose-induced acidification of the external medium was measured as described above. When applicable, DMSO was used as the control.

Preparation of C. albicans membrane fraction.

The plasma membrane fraction containing proton-pumping ATPase of C. albicans was prepared basically by a previously described procedure (50). Briefly, C. albicans 90028 was grown in PYG broth (10 liters) for 18 h (mid-logarithmic phase) at 30°C with vigorous aeration. The cells were harvested by centrifugation (3,500 × g, 10 min, 4°C) and washed extensively with sterile distilled water. The washed cells were collected by centrifugation, and the resulting cell pellet (approximately 50 g [wet weight]) was either used immediately for the preparation of the membrane fraction or was stored at −80°C until further use.

To isolate the membrane fraction enriched with proton-pumping ATPase, approximately 50 g of cells (wet weight) was resuspended in 200 ml of homogenization buffer containing 50 mM Tris-HCl (pH 7.5), 0.3 M sucrose, 5 mM disodium EDTA, 1 mM EGTA, 5 mg of bovine serum albumin per ml, 2 mM DTT, 1 mM phenylmethylsulfonyl fluoride (PMSF), and 2.5 μg of chymostatin per ml. Since PMSF and DTT are unstable, the homogenization buffer was prepared fresh, before use, from appropriate stock solutions. The cell suspension thus prepared was passed through a French pressure cell at 20,000 lb/in², and the lysate was collected. The process was repeated a second time to obtain adequate breakage of the cells.

The pH of the lysate was adjusted to 7.25 with 1 M Tris, and the lysate was then centrifuged at 3,500 × g for 5 min to remove any remaining intact cells as well as cell debris. The supernatant obtained by low-speed centrifugation was collected and further centrifuged at high speed (14,000 × g for 20 min at 4°C) to obtain a clarified homogenate. The resulting clear supernatant was centrifuged at 105,000 × g for 2 h at 4°C with a Beckman L8M ultracentrifuge and a Ti 70 rotor. The sticky pellet obtained from the ultracentrifugation containing membrane fragments was resuspended in 100 ml of membrane wash buffer containing 10 mM Tris-HCl (pH 7.0), 1 mM EGTA, 1% (wt/vol) glycerol, and 0.1 mM PMSF with a glass homogenizer and a tight-fitting Teflon plunger. The resulting preparation of the membrane fraction was centrifuged a second time at 105,000 × g for 2 h, and the pellet was resuspended in membrane wash buffer (≈10 mg of protein/ml) (9) and stored at −80°C. The purity of the membrane fraction was examined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (PAGE). Approximately 10 to 12 detectable protein bands were seen on the sodium dodecyl sulfate-polyacrylamide gel after Coomassie blue staining. This preparation was highly enriched with plasma membrane which contained very little mitochondrial ATPase activity, as determined by its susceptibility to potassium azide, an inhibitor of mitochondrial ATPase (48), and was subsequently used for the H⁺-ATPase assay.

Measurements of ATP hydrolysis.

The proton-pumping ATPase activities of the C. albicans membrane preparations were assayed by measuring the inorganic phosphate liberated by the hydrolysis of ATP as described previously (8, 50). Briefly, a 1-ml reaction mixture containing 10 mM morpholineethanesulfonic acid (MES)-Tris (pH 6.5), 5 mM MgSO₄, 5 mM ATP, 25 mM NH₄Cl, and 10 μl of the membrane preparation (≈100 μg of protein) (9) was assembled and was incubated at 30°C for 10 min. The reaction was stopped by the addition of 0.2 ml of 50% trichloroacetic acid, and the components were mixed well and stored on ice for 10 min. The ice-cold acidified mixture was centrifuged at top speed in an Eppendorf microcentrifuge for 5 min. The clear supernatant was transferred to a clean tube, and a portion of it was used for the measurement of the inorganic phosphate liberated by the hydrolysis of ATP.

Measurement of inorganic phosphate was performed as described previously (1). A reaction solution containing 1.42% ascorbic acid and 0.36% ammonium molybdate was prepared from appropriate stock solutions (10% ascorbic acid and 0.42% ammonium molybdate in 1 N H₂SO₄) by combining the stocks at a ratio of 1:6, respectively. The reaction was initiated by the addition of 0.7 ml of the reagents to 0.3 ml of the phosphate solution in a small test tube. The reaction mixture was incubated at 37°C for 1 h, and the absorbance at 490 nm was determined spectrophotometrically. An equal amount (0.3 ml) of sterile distilled water was used for the blank. The readings were proportional to the phosphate concentrations to an optical density of approximately 2. All glassware used was acid washed to minimize inorganic phosphate contamination from extraneous sources.

The procedure of Ames (1) for measurement of inorganic phosphate is highly sensitive, and any contaminating P_i from extraneous sources will greatly obscure the rate of ATP hydrolysis as well as the effect of antagonists on the enzyme activity. We therefore included several controls to examine the specific effect of NC1175 on the ATP hydrolysis by the membrane fraction. An enzyme-minus control was used to determine the base level of P_i in the assay mixture in the absence of ATP hydrolysis. Since DMSO was used as the solvent for NC1175, a control with a comparable amount of DMSO was used to monitor the effect of DMSO on C. albicans plasma membrane-bound ATPase. Vanadate is a known inhibitor of P-type ATPases, including plasma membrane ATPases from fungi (8, 49, 50). Thus, vanadate was used as a positive control to inhibit the activity of the plasma membrane ATPase of C. albicans. An inhibitor-free enzyme control was used to obtain the maximal level of H⁺-ATPase in the membrane preparation. A comparison of the levels of P_i in the controls with those obtained in the presence of NC1175 enabled us to determine the effect of the conjugated styryl ketone on the H⁺-ATPase of C. albicans.

Chemicals and reagents.

NC1175 was obtained from J. R. Dimmock. Itraconazole (R51 211; batch no. STAN-9304-005-1) was obtained from Janssen Pharmaceutica, Beerse, Belgium. Fluconazole and voriconazole were from Pfizer Pharmaceuticals, New York, N.Y. Amphotericin B (batch no. 20-914-29670) was obtained from Squibb Institute for Medical Research, Princeton, N.J. Nystatin was obtained from Sigma Chemical Company (St. Louis, Mo.). All the antifungal agents except fluconazole were dissolved in DMSO at a concentration of 1 mg/ml and were stored as 0.25-ml aliquots at −20°C. The frozen stock was thawed at room temperature and was gently vortexed several times to ensure that any remaining crystals were completely dissolved before use. Drug concentrations ranging from 0.0625 to 16 μg/ml were used for MIC determinations. For fluconazole, however, concentrations of 0.0625 to 512 μg/ml were used. When applicable, comparable concentrations of DMSO were tested to examine its effect on the growth of the organisms. All the other chemicals and reagents were obtained from Sigma Chemical Company and were either reagent or molecular biology grade.

RESULTS

Susceptibility studies.

The antifungal susceptibilities of the various strains of Candida, Saccharomyces, and Aspergillus species used in our investigation are shown in Table 1. All the strains used in our investigation were susceptible to commonly used polyenes and azoles to various degrees except in the case of fluconazole, which possesses no significant activity against Aspergillus species and C. krusei. The other notable exception was A. flavus I65850 and A. niger S11335, which showed reduced susceptibilities to amphotericin B (MIC, 5.33 ± 2.30 μg/ml) and itraconazole (MIC, 3.33 ± 1.15 μg/ml), respectively. Nonetheless, all fungal isolates that we used were susceptible to NC1175, and the MICs ranged from 0.83 to 2 μg/ml. This finding is in agreement with our previous studies (32, 33), which showed that both drug-resistant and -susceptible organisms are equally susceptible to the styryl ketone NC1175.

TABLE 1.

In vitro susceptibilities of study strains of microorganisms to various antifungal agents^a

Organism	Source	MIC (μg/ml)
		NC1175	AMB	FLZ	ITZ	VCZ
C. albicans 90028	ATCC	1.66 ± 0.57	0.18 ± 0.08	0.20 ± 0.07	0.16 ± 0.07	0.16 ± 0.07
C. glabrata 33554	ATCC	1.33 ± 0.57	0.10 ± 0.03	2 ± 0.0	0.25 ± 0.0	0.41 ± 0.14
C. guilliermondii 9390	ATCC	1.66 ± 0.57	1.33 ± 0.57	0.25 ± 0.0	0.33 ± 0.14	0.25 ± 0.0
C. krusei 6258	ATCC	1.66 ± 0.57	0.20 ± 0.07	53.3 ± 18.4	0.33 ± 0.14	1.66 ± 0.57
S. cerevisiae GW201	PHRI	0.83 ± 0.28	0.12 ± 0.0	0.25 ± 0.0	0.125 ± 0.0	0.20 ± 0.07
S. cerevisiae pma1.1	PHRI	0.83 ± 0.28	0.10 ± 0.03	0.10 ± 0.03	0.10 ± 0.03	0.16 ± 0.07
S. cerevisiae pma1.2	PHRI	1 ± 0.0	0.12 ± 0.0	0.16 ± 0.07	0.20 ± 0.07	0.20 ± 0.07
S. cerevisiae pma1.3	PHRI	0.83 ± 0.28	0.08 ± 0.03	0.20 ± 0.07	0.20 ± 0.07	0.20 ± 0.07
S. cerevisiae pma1.4	PHRI	0.83 ± 0.28	0.12 ± 0.0	0.12 ± 0.0	0.12 ± 0.0	0.25 ± 0.0
S. cerevisiae pma1.5	PHRI	1 ± 0.0	0.12 ± 0.0	0.16 ± 0.07	0.20 ± 0.07	0.20 ± 0.07
A. fumigatus W73355	DMC	1.66 ± 0.57	0.33 ± 0.14	426 ± 147	0.41 ± 0.14	0.5 ± 0.0
A. niger S11335	DMC	2 ± 0.0	0.83 ± 0.28	512 ± 0.0	3.33 ± 1.15	0.83 ± 0.28
A. flavus I65850	DMC	1.66 ± 0.57	5.33 ± 2.30	341 ± 147	0.41 ± 0.14	1 ± 0.0
A. nidulans A767	FGSC	1.33 ± 0.57	0.66 ± 0.28	256 ± 0.0	0.20 ± 0.07	0.41 ± 0.14

^aAbbreviations: DMC, Detroit Medical Center; ATCC, American Type Culture Collection; PHRI, Public Health Research Institute; FGSC, Fungal Genetics Stock Center; AMB, amphotericin B; FLZ, fluconazole; ITZ, itraconazole; VCZ, voriconazole. Each value represents the mean ± standard deviation of three experiments. S. cerevisiae pma1.1 to pma1.5 were strains carrying mutant PMA1 gene whose several cysteine coding triplets were altered by site-directed mutagenesis. The genotypes of the pma1 mutants were as follows: pma1.1, C148S, C867A; pma1.2, C148S, C221S, C867A; pma1.3, C148S, C221S, C312A, C569A, C867A; pma1.4, C148S, C221S, C312A, C867A; pma1.5, C532S, C867A. 

Inactivation of antifungal activity of NC1175.

The effects of various thiols on the anticandidal activity of NC1175 (Fig. 1) is shown in Fig. 2. The naturally occurring cellular thiols such as cysteine (Fig. 2A) and glutathione (Fig. 2B), but not DTT (Fig. 2C) or mercaptoethanol (Fig. 3D), eliminated the inhibitory effect of NC1175 on C. albicans. In the presence of relatively high concentrations (0.625 to 2.5 mM) of cysteine and glutathione, NC1175 at a concentration of 25 μM (about eightfold higher than the MIC) failed to inhibit the growth of C. albicans. In the presence of cysteine or glutathione at less than 0.625 mM, NC1175 at 25 μM completely inhibited the growth of C. albicans. On the other hand, at high concentrations (5 to 10 mM) the thiol reagents by themselves inhibited C. albicans growth, perhaps due to general cytotoxicity. The exception was glutathione, which showed little inhibition of C. albicans growth at a concentration as high as 10 mM.

FIG. 1.

Chemical structures of four analogues of NC1175. The proposed sites of thiolation are indicated by the arrows.

FIG. 2.

Effect of cysteine (A), glutathione (B), DTT (C), and mercaptoethanol (D) on the antifungal activity of NC1175. Symbols: ●, growth control; ■, NC1175 25 μM.

FIG. 3.

Effect of NC1175 on proton pumping (as measured by the acidification of the external medium) of C. albicans (A), C. glabrata (B), C. guilliermondii (C), and C. krusei (D). Each value represents the mean ± standard deviation of two experiments. The pH of the external medium in the presence of various concentrations of NC1175 was plotted against time by linear regression (95% confidence level) third-order curve fitting (SigmaPlot 3.0; Jandel Scientific Software, San Rafael, Calif.). The concentrations of NC1175 used were 0 (●), 6.25 (▾), 12.5 (■), 25 (⧫) and 50 (▴) μM.

Inhibition of proton pumping.

The proton-pumping ability of fungi mediated by the H⁺-ATPase at the expense of energy is crucial for the regulation of the internal pH of a fungal cell. When fungal cells depleted of their carbon sources are exposed to glucose, the sugar is rapidly taken up by the cells by the proton motive force generated by the proton gradient due to the pumping out of intracellular protons. The extrusion of intracellular protons to the surrounding medium will acidify it, and the resulting alteration of the pH of the external medium can be measured with the help of a pH electrode.

The effect of NC1175 on the proton-pumping abilities of various Candida species as determined by the glucose-induced acidification of external medium is shown in Fig. 3. NC1175 inhibited the glucose-induced acidification of the external medium by C. albicans (Fig. 3A), C. glabrata (Fig. 3B), C. guilliermondii (Fig. 3C), and C. krusei (Fig. 3D) in a partly concentration-dependent manner. Although all four Candida species were susceptible to NC1175, there appeared to be variations between different species. Both C. albicans and C. glabrata had the highest sensitivity to NC1175 inhibition of proton pumping, whereas C. guilliermondii and C. krusei had moderate and low sensitivities, respectively. We used concentrations of NC1175 ranging from 6.25 to 50 μM (≈2- to 16-fold higher than the MICs). At 50 μM, the rate of proton pumping was reduced almost to zero within 60 min. The exact reason why such a high concentration of the drug (compared to the MICs) was required to inhibit the proton pumping of Candida species is not understood. It is likely that it has to do with the number of cells used in the two assay systems. For MIC studies approximately 1 × 10³ cells/ml were commonly used, whereas for the in vivo proton pumping assay approximately 2 × 10⁸ cells/ml were commonly used. The use of high cell density was essential to obtain measurable pH changes in the external medium. A 2 × 10⁵-fold increase in cellular targets will reduce the effective concentration of the drug.

In addition to Candida species, we investigated the effect of NC1175 on the proton-pumping activities of clinical and laboratory isolates of Aspergillus species. The glucose-induced acidification of the external medium by A. fumigatus (Fig. 4A), A. niger (Fig. 4B), A. flavus (Fig. 4C), and A. nidulans (Fig. 4D) was inhibited by NC1175 in a partly concentration-dependent fashion. Unlike the Candida species, the Aspergillus species appeared to be more susceptible to the drug. At 50 μM the proton-pumping ability of Aspergillus species was completely inhibited within 20 min, whereas approximately 60 min was required to obtain the same level of inhibition in the case of various Candida species.

FIG. 4.

Effect of NC1175 on proton pumping (as measured by the acidification of the external medium) of A. fumigatus (A), A. niger (B), A. flavus (C), and A. nidulans (D). Each value represents the mean ± standard deviation of two experiments. The pH of the external medium in the presence of various concentrations of NC1175 was plotted against time by linear regression (95% confidence level) third-order curve fitting (SigmaPlot 3.0; Jandel Scientific Software). The concentrations of NC1175 used were 0 (●), 6.25 (▾), 12.5 (■), 25 (⧫) and 50 (▴) μM.

Not only the proton-pumping ability of pathogenic fungi such as Candida and Aspergillus species but also that of a nonpathogenic yeast such as S. cerevisiae was highly susceptible to NC1175 (Fig. 5A). At 50 μM NC1175 almost completely inhibited the glucose-induced acidification of the external medium by S. cerevisiae within 10 min. We also examined the effects of known membrane-acting antifungal agents (polyenes) on the glucose-induced acidification of the external medium by S. cerevisiae. Neither nystatin nor amphotericin B at 12.5 μM (≈50- to 100-fold higher than the MICs) inhibited the proton-pumping ability of this yeast (Fig. 5B), indicating that other membrane-acting antifungal agents (26, 60) failed to interfere with the proton-pumping activity of S. cerevisiae. Also, members of the azole (fluconazole, itraconazole, and voriconazole) family of antifungal agents failed to inhibit (data not shown) the proton pumping of S. cerevisiae.

FIG. 5.

(A) Effect of NC1175 on proton pumping (as measured by the acidification of the external medium) of S. cerevisiae. Each value represents the mean ± standard deviation of two independent experiments. The pH of the external medium in the presence of various concentrations of NC1175 was plotted against time by linear regression (95% confidence level) third-order curve fitting (SigmaPlot 3.0; Jandel Scientific Software). The concentrations of NC1175 used were 0 (●), 6.25 (▾), 12.5 (■), 25 (⧫) and 50 (▴) μM. (B) A comparison of the effect of NC1175 on the proton pumping of S. cerevisiae with those of amphotericin B and nystatin. Symbols: ●, control; ■, amphotericin B (12.5 μM); ⧫, nystatin (12.5 μM); ▾, NC1175 (50 μM).

In addition to S. cerevisiae GW201, we examined the effect of NC1175 on the proton-pumping ability of pma1 mutant isolates of GW201. In these mutants one or more cysteine residues of the H⁺-ATPase was altered by site-directed mutagenesis. If the inhibitory action of NC1175 on the S. cerevisiae H⁺-ATPase is mediated by one or more critical thiol groups on the enzyme, then the mutant isolate(s) carrying the altered cysteine(s) will be relatively resistant to the inhibitory action of NC1175 against H⁺-ATPase. The proton-pumping ability of none of the pma1 mutants that we examined in our experiment was resistant to the inhibitory action of NC1175 (data not shown). This result is not surprising considering the fact that the MICs (Table 1) of NC1175 for the pma1 mutants were almost identical to the MIC obtained for the parent strain GW201, suggesting that none of the altered cysteine residues affected the susceptibilities of the pma1 mutants to NC1175. Our results are different from those of Monk et al. (39) obtained with omeprazole, in which the inhibitory action of this cysteine-modifying compound was greatly reduced in pma1 mutants, in which specific cysteines are altered.

Inhibition of C. albicans plasma membrane H⁺-ATPase.

Plasma membrane fractions of C. albicans 90028 were rich in ATPase activity. The ATPase activity of the membrane fraction was directly proportional to the amount of the membrane fraction added to the reaction mixture in the presence of an excess amount of ATP (Fig. 6A). The effect of NC1175 on ATP hydrolysis catalyzed by the membrane fraction of C. albicans is shown in Fig. 6B. When NC1175 and the substrate were added to the reaction mixture simultaneously, the drug had very little effect on the H⁺-ATPase. However, when the membrane preparation was preincubated with NC1175 on ice for 30 min, the effect was greatly augmented. Therefore, in all subsequent experiments hydrolysis of ATP in the presence of NC1175 was measured after 30 min of preincubation of the enzyme with the drug on ice. The amounts of inorganic phosphate liberated in the presence of enzyme and enzyme plus DMSO, vanadate, or NC1175 were compared with those obtained in the absence of the enzyme. The use of an enzyme-free control was essential to obtain the background level of P_i in the reaction mixture. High base levels of P_i due to contamination from external sources and by nonenzymatic chemical breakdown of ATP will mask the amount of P_i produced by enzymatic hydrolysis of ATP. As shown in Fig. 6B, the enzyme-free control, vanadate (10 μM), and NC1175 (50 μM) treatments produced approximately the same amounts of P_i. On the other hand, the enzyme alone and enzyme plus DMSO produced approximately 2.5-fold higher amounts of P_i than the background amount, suggesting that vanadate and NC1175 inhibited C. albicans plasma membrane-bound H⁺-ATPase completely. These results suggest that the C. albicans H⁺-ATPase is susceptible to the inhibitory action of NC1175, and the lack of acidification of the external medium in the presence of this styryl ketone is most likely due to the inhibition of H⁺-ATPase.

FIG. 6.

Effect of NC1175 on ATP hydrolysis by membrane fractions of C. albicans. (A) Relationship between the amount of membrane fraction and H⁺-ATPase activity. Each value represents the mean ± standard deviation of two experiments. The enzyme activity (A₄₉₀) associated with various amounts of membrane fraction was plotted by linear regression (95% confidence level) third-order curve fitting (SigmaPlot 3.0; Jandel Scientific Software). (B) Effects of NC1175 and vanadate on ATP hydrolysis by membrane fractions of C. albicans. Each value represents the mean ± standard deviation of two experiments. Treatments were as follows: A, no enzyme; B, enzyme; C, enzyme plus DMSO; D, enzyme plus vanadate (10 μM); E, enzyme plus NC1175 (50 μM).

DISCUSSION

Representatives of the conjugated styryl ketones preferentially react reversibly with the sulfhydryl groups of free small molecules (e.g., cysteine and glutathione) and irreversibly with the protein-associated sulfhydryl groups (41). Thus, thiols would be predicted to react at the electron-deficient centers of NC1175 indicated in Fig. 1. Nucleophilic attack would be expected to occur preferentially at the carbon atom beta to the carbonyl group due to the electron-withdrawing influences of both the carbonyl function and the quadrivalent nitrogen atom. A subsequent thiol attack could occur at the isolated olefinic double bond of the 4-β-arylvinyl group. Theoretically, 2 mol of cysteine, glutathione, and mercaptoethanol and 1 mol of DTT should abolish the antifungal activity of 1 mol of NC1175 if the reaction went to completion and the resultant adduct was stable. The data in Fig. 2 revealed that stoichiometrically an approximately 100-fold excess of cysteine or glutathione was required to obtain maximum inhibition of the antifungal activity of NC1175.

Thus, explanations are required to account for, first, why such large excesses of cysteine and glutathione were necessary to eliminate the antifungal activity of NC1175 and, second, why DTT and mercaptoethanol did not antagonize the bioactivity of this novel antifungal agent. The requirement for a large excess of thiol to react with the Mannich base of an α,β-unsaturated ketone has been noted previously (41). In this case, incubation of equimolar quantities of 1-(4-chlorophenyl)- 4,4-dimethyl-5-diethylamino-1-penton-3-one hydrobromide and cysteine led to a small reduction in the UV absorptivity at the λ_max, whereas an absorptivity decrease of approximately 84% was noted when a 40-fold molar excess of cysteine was used; the situation was unchanged when a 100-fold excess of cysteine was used. This result may have been due to the thiol adduct undergoing a reverse Michael reaction, regenerating the thiol and unsaturated ketone. A similar explanation may be invoked to explain the data presented in Fig. 2. In regard to the reaction of NC1175 with only cysteine and glutathione, the increased reactivity and stability of the adducts of these two thiols may contribute to this observation. Thus, considering the three reagents which contain only one thiol group, namely, cysteine, glutathione, and mercaptoethanol, the fraction of the compounds in the deprotonated sulfhydryl form have been calculated to be 9:5:1 (41). Hence, greater nucleophilic attack for the olefinic carbon atoms in NC1175 would be predicted for cysteine and glutathione. In addition, it is likely that the thiolate anionic species obtained from the dithiothreitol and mercaptoethanol adducts are better leaving groups than the thiolate anions released from the cysteine and glutathione adducts.

The need for the discovery and development of new antifungal agents directed to novel cellular targets has been recognized in recent years with the emergence of clinical as well as laboratory isolates of pathogenic yeasts and filamentous fungi resistant to commonly used antifungal agents such as azoles and polyenes (11, 12, 20–23, 29, 30, 43, 46, 54–56, 63–66). The proton-translocating ATPase of fungi has been considered by several investigators to be a possible target in the development of antifungal agents (39, 51). There are several advantages to the development of chemical reagents that inhibit the activity of this enzyme. The noteworthy characteristics of the enzyme as a novel target include the following: (i) it is an essential enzyme for the survival of the cell, and any interference of its function either fully or partially will eventually lead to cell death; (ii) since this enzyme is a membrane-bound integral protein spanning the membrane extending to the cytoplasmic as well as to the external medium, nonpenetrating chemical agents can be developed to effectively interfere with its function; (iii) an obvious advantage of such nonpenetrating antibiotics is that the emergence of drug resistance due to efflux is unlikely to be developed; and (iv) it is imperative that a chemical agent that inhibits the function of H⁺-ATPase be, with all likelihood, a fungicidal agent as opposed to a fungistatic one, which is crucial for the possible eradication of fungal infection in the absence of a competent immune system of the host.

A major concern for the development of antifungal agents is their lack of specificity since fungi and mammalian host cells often share substantially similar metabolic pathways. Molecular genetic analysis shows that the mammalian ion-transporting ATPases and the H⁺-ATPases of fungi share low-level (≤30%) similarities (52). The mammalian and the fungal enzymes are divergent enough so that it is possible to develop a specific chemical reagent(s) that preferentially inhibits the fungal H⁺-ATPase. By the same token, all the fungal H⁺-ATPases studied so far have shown high degrees of similarity (≥50%), so that it is feasible that a single chemical agent capable of inhibiting enzymes from various fungi can be developed, thereby maintaining the broad spectrum of activity. The fact that NC1175 has activity against all fungal species that we have examined so far provides encouragement for achieving this prospect.

The majority of the antifungal agents currently licensed for use as antifungal agents (and those under development) belong to either the polyene or the azole family. The cellular targets for both groups of compounds depends either directly (azoles) or indirectly (polyenes) on the sterol synthetic pathway. Even several antifungal agents not belonging to the azole or the polyene families of compounds have the sterol synthetic pathway as their target (e.g., allylamines, thiocarbamates, and morpholines). The main drawback of several agents directed to a single metabolic pathway is that the emergence of resistance to one agent will result in a greater likelihood of cross-resistance to related agents. Development of chemical reagents directed to additional fungal targets unrelated to those commonly interfered with is useful even for the design of combination therapy with more than one drug. The design of such combination therapy is attractive for therapeutic agents with different targets of action. Since NC1175 acts by interference with H⁺-ATPase, which is different from the mechanisms of action of commonly used antifungal agents on the market to date, this drug may have the potential to be used not only alone but also in combination with other currently used antifungal agents.

Our in vitro and in vivo susceptibility studies show that NC1175 has good activity against a wide spectrum of pathogenic fungi including C. albicans and A. fumigatus. A comparison of the MICs of various azoles and polyenes showed that they are significantly lower than that of NC1175 for susceptible organisms. Since very little toxicity data for NC1175 are available so far, the efficacy and the therapeutic index of the drug are not known. Similarly, little information is available with regard to the pharmacokinetics of the compound. However, it is important that NC1175 is active against organisms that are resistant to the polyenes and azoles. Being a fungicidal agent, NC1175 has the potential to be used as a topical agent against dermatophytes and other fungi that cause superficial infections. In summary, further investigation of the toxicity and pharmocokinetic properties of the compound would be worthwhile, as would additional experiments that would study efficacy in animal models.