Natamycin Inhibits Vacuole Fusion at the Priming Phase via a Specific Interaction with Ergosterol

doi:10.1128/AAC.01794-09

Antimicrob Agents Chemother. 2010 June; 54(6): 2618–2625.

Published online 2010 April 12. doi: 10.1128/AAC.01794-09

PMCID: PMC2876427

Natamycin Inhibits Vacuole Fusion at the Priming Phase via a Specific Interaction with Ergosterol

Yvonne Maria te Welscher,¹^* Lynden Jones,² Martin Richard van Leeuwen,³ Jan Dijksterhuis,³ Ben de Kruijff,¹ Gary Eitzen,² and Eefjan Breukink¹

Department of Biochemistry of Membranes, Bijvoet Center, Institute of Biomembranes, Utrecht University, Utrecht 3584 CH, Netherlands,¹ Department of Cell Biology, University of Alberta, Edmonton, Alberta T6G 2H7, Canada,² Applied and Industrial Mycology, CBS/Fungal Biodiversity Centre, Uppsalalaan 8, Utrecht 3584 CT, Netherlands³

^*Corresponding author. Mailing address: Department of Biochemistry of Membranes, Bijvoet Center, Institute of Biomembranes, Utrecht University, Padualaan 8, Utrecht 3584 CH, Netherlands. Phone: 31 30 2533345. Fax: 31 30 2533969. E-mail: y.m.tewelscher/at/uu.nl

Received December 18, 2009; Revised January 29, 2010; Accepted April 1, 2010.

This article has been cited by other articles in PMC.

Abstract

The antifungal antibiotic natamycin belongs to the family of polyene antibiotics. Its antifungal activity arises via a specific interaction with ergosterol in the plasma membrane (te Welscher et al., J. Biol. Chem. 283:6393-6401, 2008). However, this activity does not involve disruption of the membrane barrier function, a well-known property of other members of the polyene antibiotic family, such as filipin and nystatin. Here we tested the effect of natamycin on vacuole membrane fusion, which is known to be ergosterol dependent. Natamycin blocked the fusion of isolated vacuoles without compromising the barrier function of the vacuolar membrane. Sublethal doses of natamycin perturbed the cellular vacuole morphology, causing the formation of many more small vacuolar structures in yeast cells. Using vacuoles isolated from yeast strains deficient in the ergosterol biosynthesis pathway, we showed that the inhibitory activity of natamycin was dependent on the presence of specific chemical features in the structure of ergosterol that allow the binding of natamycin. We found that natamycin inhibited the priming stage of vacuole fusion. Similar results were obtained with nystatin. These results suggest a novel mode of action of natamycin and perhaps all polyene antibiotics, which involves the impairment of membrane fusion via perturbation of ergosterol-dependent priming reactions that precede membrane fusion, and they may point to an effect of natamycin on ergosterol-dependent protein function in general.

The increase in invasive fungal infections, especially in persons whose immune systems are compromised, is a growing threat to human health. Only a few antifungal agents have proven to be effective, including the polyenes, the fluorocytes, and the azole derivatives, but an increase in resistance has been observed for several members (14). Polyene antibiotic resistance is still a rare occurrence, which makes these antibiotics particularly useful as antifungal agents. In the past, convincing evidence has been presented that this class of antibiotics targets sterols, in particular ergosterol, the abundant and main sterol of fungal membranes. The interaction of these antibiotics with ergosterol leads to changes in the membrane that ultimately cause the destruction of the membrane barrier (3, 10, 11). Natamycin (also called pimaricin) is a very effective member of the polyene antibiotic family, with a large record of applications. Natamycin is produced by Streptomyces natalensis and is used for the topical treatment of fungal infections, and it is also widely utilized in the food industry. For many years, people have believed that the polyene antibiotic natamycin would kill fungi by permeabilizing the plasma membrane. Only recently have we discovered that in marked contrast to amphotericin B, filipin, or nystatin, the polyene antibiotic natamycin does not act via membrane permeabilization (32). And yet, its activity is strongly ergosterol dependent and requires a specific sterol structure (32). We aim to elucidate the mode of action of natamycin, and through a detailed understanding of its mechanism, new and improved antifungal formulations may be developed. Because of the specific interaction with ergosterol, natamycin may act via excluding ergosterol from performing important functions in the membrane.

Besides important roles in modulating membrane fluidity, regulatory processes, and domain formation, sterols also have been shown to be important during membrane fusion and fission events (6, 28, 30). Both fusion and fission are similar processes that rely on the central event of a merger or separation of two membranes. This requires a transient reorganization of membrane lipids into highly curved fusion intermediates (7). Both endocytic and exocytic pathways are dependent on the fusion and fission of membranes in which sterols have been shown to be important (12, 29). For example, by deleting different ERG genes in Saccharomyces cerevisiae, strains with altered sterol compositions are formed (17, 24). These strains show deficiencies in the endocytic process and in plasma membrane fusion (17, 19, 24). This implies that these processes are dependent on ergosterol and have specific structural requirements for the sterols present.

The fusion reaction of isolated vacuoles from yeast can be studied via a content mixing assay and has been used as a model system to examine membrane fusion reactions in general, particularly because it uses many of the same mechanisms as other fusion reactions (15, 37). Ergosterol has been shown to be required for the fusion of vacuoles, indicating the importance of ergosterol in vacuolar fusion in yeast (21, 31). Here we have used this model system to decipher the mode of action of natamycin. Natamycin was able to inhibit the vacuolar homotypic fusion. Like the overall inhibitory effect of natamycin on yeast cells, the inhibition on vacuolar fusion was not due to membrane permeabilization. Natamycin acted at an early stage of the fusion process, even before membrane contact. This activity was dependent on the presence of specific chemical features in the structure of ergosterol and may involve an effect on protein functions that are ergosterol dependent.

MATERIALS AND METHODS

Chemicals.

The polyene antibiotics nystatin and filipin were dissolved in pure dimethyl sulfoxide (DMSO), and natamycin was dissolved in DMSO-H₂O (85:15 [vol/vol]); all were obtained from Sigma Chemical (St. Louis, MO). The concentrations of the polyene antibiotics were determined spectrophotometrically on a Perkin Elmer UV/visible (UV/Vis) spectrometer (Lambda 18). The molar extinction coefficients and corresponding wavelengths of the polyene antibiotics in methanol were 7.6 × 10⁴ M⁻¹ cm⁻¹ (318 nm), 6.7 × 10⁴ M⁻¹ cm⁻¹ (318 nm), and 8.5 × 10⁴ M⁻¹ cm⁻¹ (356 nm) for natamycin, nystatin, and filipin, respectively.

Poly-l-lysine, N-(2-acetamido)-2-aminoethanesulfonic acid (ACES), DEAE-dextran, Ficoll (molecular weight, 400,000), para-nitrophenylphosphate (pNPP), quinacrine, neomycin, ATP, creatine kinase, creatine phosphate, leupeptin, pepstatin, o-phenanthroline, Pefabloc SC, and apyrase (VI and VII) were obtained from Sigma Chemical (St. Louis, MO). MDY-64 was purchased from Molecular Probes (Eugene, OR). Antibodies against Sec18p or Vam3p were purified as IgG fractions from rabbit sera as previously described (13). All protein concentrations were measured using Bio-Rad protein assay reagents from Bio-Rad Laboratories (Richmond, CA) using bovine serum albumin as a standard.

Strains, growth conditions, and genetic modifications.

Strains used for vacuole staining and isolation are listed in Table Table1.1. Yeast cells were grown at 30°C in 10 g/liter yeast extract, 20 g/liter Bacto peptone, and 20 g/liter dextrose without (YPD) or with (YPDUAT) supplementation with 2 g/liter uracil, 1 g/liter adenine, and 1 g/liter tryptophan. ERG gene deletions were performed in strains KTY1 and KTY2 by homologous recombination of PCR products using primers with ~40 nucleotides of homology to the 5′ and 3′ ends of the gene of interest and 20 nucleotides of homology to the pRS 403 vector as the template (Table (Table2)2) (4).

TABLE 1.

Strains used in this study

TABLE 2.

Primers used in this study

Vacuole isolation and fusion reactions.

Vacuoles were isolated, and their fusion was tested as previously described (15). Standard fusion reaction mixtures contained vacuoles isolated from two strains (3 μg protein each) either with alkaline phosphatase deleted (pho8Δ KTY2 parental strains) or with proteinase A and proteinase B deleted (pep4Δ prb1Δ KTY1 parental strains), 30 μl of PSS buffer [20 mM piperazine-N,N′-bis(2-ethanesulfonic acid) (PIPES)-KOH, pH 6.8, 0.2 M sorbitol, 125 mM KCl, 5 mM MgCl₂] supplemented with 10 μM coenzyme A, a protease inhibitor cocktail (6.6 ng/ml leupeptin, 16.6 ng/ml pepstatin, 16.6 μM o-phenanthroline, 3.3 μM Pefabloc SC), and an ATP regenerating system (ATP_reg; 1 mM Mg-ATP, 0.5 mg/ml creatine kinase, 40 mM creatine phosphate) and 1 mg/ml cytosol (isolated as described previously [18]). Reactions were incubated for 90 min at 27°C and then assayed for alkaline phosphate activity. For this, 470 μl of developer solution (250 mM Tris-Cl, pH 8.5, 0.4% Triton X-100, 10 mM MgCl₂, 1.5 mM pNPP) was added to the reactions and incubated for 5 min at 30°C. The reaction was stopped by addition of 500 μl 1 M glycine-KOH, pH 11.5, and the absorption at 400 nm was determined. Fusion reactions with the strains containing ergΔ that are based on KTY1 and KTY2 (Table (Table1)1) were performed in a similar manner, with one exception, that no cytosol was added to these reactions.

Quinacrine uptake assay.

Reaction mixtures contained vacuoles (60 μg) freshly isolated from strain KTY2 in 150 μl of PSS buffer. To have an active H⁺-pumping system that allows for acidification of the vacuoles, the reactions were supplemented with ATP_reg. Quinacrine (200 μM) was added from a 10 mM stock in water, and polyene antibiotics were added in a concentration range from 0 to 400 μM. The reactions were incubated for 20 min at 27°C, after which the reaction mixtures were placed on ice, 1 ml of PSS buffer was added, and the vacuoles were spun down for 4 min, 14,000 rpm at 4°C (26). The pellet was resuspended in 150 μl of 0.4% Triton X-100, and the fluorescence of the quinacrine was determined (excitation at 421 nm/emission at 496 nm) using a QM-4SE spectrofluorometer with a four-position sample holder (Photon Technologies Inc., London, United Kingdom). The percentage of uptake of quinacrine at a given polyene antibiotic concentration was determined by comparing the fluorescence of vacuoles supplemented with ATP_reg (100%) to that of vacuoles without ATP_reg (0%).

Staging assay.

Staging was performed as described previously (9, 26). Standard fusion reactions were started at 27°C as described above. At different time points (0, 10, 20, 30, 45, or 60 min), inhibitors of the specific stages or PS buffer (control; 20 mM PIPES-KOH, pH 6.8, 0.2 M sorbitol) were added. The inhibitor used was 4 μl (250 μg/ml) anti-Sec18 (priming) or 4 μl (150 μg/ml) anti-Vam3 (docking) or placement on ice (fusion) at the indicated times. The inhibitor effects were compared with the effects of the polyene antibiotic natamycin (200 μM), nystatin (200 μM), or filipin (100 μM). After 90 min, the amount of fusion was determined by measuring the alkaline phosphatase activity.

Vacuole staining and observation of live yeast cells.

A small single colony of strain RH448 (Table (Table1)1) was grown aerobically for 16 h at 30°C in 50 ml YPUADT. Cultures were diluted to an optical density at 600 nm (OD₆₀₀) of 0.3, and after 1.5 h of growth, the cells were inoculated on poly-l-lysine-coated coverslips as described in reference 35. Concentrations of 0, 0.5, 1.0, and 1.5 μM natamycin were used, and coverslips were incubated for 5 h at 25°C. Yeast cells were stained with 10 μM MDY-64 dissolved in ACES buffer (10 mM ACES, 0.02% Tween 80, pH 6.8) and incubated for 2 min, followed by an ACES buffer wash step. After removal of ACES buffer, the glass cover slides with the immobilized cells were put upside-down on top of a thin layer (<0.5 mm) of 2% agar. Images were acquired by automatic exposure at a magnification of 100 × 2.0 with a Zeiss Axioplan II microscope equipped with a Plan-ApoChromat 100×/1.4 oil objective and Zeiss filter set 09 (450 to 490, FT510, LP520). Images were acquired by automatic exposure at a magnification of 100 × 2.0 with a Zeiss Axioplan II microscope equipped with a Plan-Apochromat 100×/1.4 oil objective, an additional 2× slider, and Zeiss filter set 09.

RESULTS

Effect of natamycin on fusion of isolated vacuoles.

Vacuole fusion can be assayed using a content mixing assay. Vacuoles are isolated from two strains; one strain contains normal vacuole proteases but is deleted for alkaline phosphatase (ALP, the PHO8 gene product), and the other strain is deleted for vacuolar proteases and hence bears catalytically inactive pro-ALP (Table (Table1).1). Neither population of purified vacuoles has phosphatase activity. Vacuole-to-vacuole fusion allows the proteases to gain access to the pro-ALP and convert it to the catalytically active form, which can be assayed by a colorimetric enzyme assay (15, 36). The effect of natamycin on vacuolar fusion was compared to that of two other polyene antibiotics, filipin and nystatin, and the chemical structures are given in Fig. Fig.1A.1A. Different concentrations of these antibiotics were added to standard fusion reactions, and the amount of fusion signal was compared to that for controls with no antibiotics (100% fusion) and incubation on ice (0% fusion) (Fig. (Fig.1B).1B). Filipin was most efficient in inhibiting vacuole fusion, with a half-maximal inhibitory concentration (IC₅₀) of 14 μM. This was followed by nystatin, with an IC₅₀ of 36 μM. The inhibition profiles of filipin and nystatin are in accordance with the profiles observed by Kato and Wickner (21). Natamycin was also able to inhibit the fusion of vacuoles, with an IC₅₀ of 56 μM. The maximal amount of inhibition caused by natamycin (71% ± 2.0%) is lower than those of nystatin (96% ± 0.5%) and filipin (90% ± 7.5%).

FIG. 1.

Effect of the polyene antibiotics natamycin, nystatin, and filipin on the fusion of isolated vacuoles. (A) Structures of natamycin (I), filipin (II), and nystatin (III). (B) Vacuoles isolated from yeasts with a wild-type sterol composition (3 μg (more ...)

Effect of the polyene antibiotics on vacuole permeability.

The permeabilization of the vacuolar membrane could potentially explain the observed inhibition on fusion by the polyene antibiotics. Vacuolar fusion requires an intact membrane to maintain a required electrochemical potential as well as the ability to release calcium (9, 27). Although natamycin is unable to permeabilize model membranes or the plasma membrane of yeast cells (32), this does not rule out the possibility that natamycin may permeabilize the vacuolar membrane. To determine the effect of the polyene antibiotics on the permeability of the vacuole, a quinacrine assay was performed (33). Quinacrine is a fluorescent compound, known to accumulate in acidic compartments such as vacuoles. If the pH gradient of the vacuole is compromised (e.g., by permeabilization of the membrane), quinacrine will be unable to accumulate in the vacuole, resulting in a reduced fluorescence. Vacuole acidification is maintained by the vacuolar type H⁺-ATPase (V-ATPase), which requires ATP (20). In Fig. Fig.2A,2A, the accumulation of quinacrine in purified vacuoles with or without an active H⁺-pumping system, +ATP_reg or −ATP_reg respectively, is compared. Extended incubation with the ATP-degrading enzyme apyrase results in no uptake of quinacrine due to inhibition of V-ATPase function (Fig. (Fig.2A).2A). Both filipin and nystatin used at 100 μM inhibited quinacrine accumulation. The quinacrine uptake is even less that that for vacuoles incubated without ATP_reg or added apyrase, indicating that intact vacuoles are still able to accumulate some quinacrine while vacuoles treated with filipin and nystatin do not. This is likely the result of a total loss of the membrane barrier function caused by these polyene antibiotics. Natamycin had no effect on quinacrine accumulation at this concentration. A similar picture emerged when a broader range of polyene concentrations was used. The percentage of uptake of quinacrine in treated vacuoles was determined via normalization to the quinacrine uptake in vacuoles with or without supplementation of ATP_reg (Fig. (Fig.2B).2B). The results clearly show that both filipin and nystatin were able to cause membrane permeabilization at the same concentrations that inhibit fusion. Natamycin, however, did not disrupt the vacuolar membrane at any of the concentrations used. This is in agreement with the results found previously that natamycin does not permeabilize model membranes nor the yeast plasma membrane (32). These results suggest that the mode of inhibition of vacuolar fusion for natamycin is fundamentally different from that of the other polyenes.

FIG. 2.

Quinacrine uptake of isolated vacuoles after incubation with the polyene antibiotics. Vacuoles from the KTY2 strain (60 μg protein) were incubated with quinacrine (200 μM) and different antibiotics. After 20 min of incubation, vacuoles (more ...)

Sterol dependency of inhibition of fusion by polyene antibiotics.

Binding of natamycin to membranes is highly dependent on the presence and chemical structure of sterol molecules (32). To test whether the inhibition of vacuolar fusion by natamycin is related to the chemical structure of ergosterol, we studied the effect of natamycin on the vacuole fusion of different ERG mutant strains. This was achieved by deleting the genes of specific sterol biosynthesis proteins in the parental strains KTY1 and KTY2 (Table (Table1),1), resulting in different sterol profiles. From these strains, the vacuoles were isolated, and their fusion abilities were tested in the content mixing assay as described above. The biosynthesis proteins and their functions together with the structure of ergosterol are shown in Fig. Fig.3A.3A. Deletion of ERG4 leads to changes in the tail part of the sterol, while deletion of ERG3 or ERG2 causes a loss of double bonds in the B-ring. We found that vacuoles isolated from the erg4Δ strain pairs fuse at levels comparable to those of the wild type (WT); however, vacuoles isolated from the erg3Δ and erg2Δ strain pairs showed a significant reduction in fusion (Fig. (Fig.3B).3B). These results confirm that the fusion of vacuoles is dependent on the chemical structure of sterols present in the vacuolar membrane (21). In addition, the isolated vacuole fusion results correlate well with the extent of vacuole fragmentation previously shown by morphological analyses in the intact yeast strains (17, 21). The relatively small but reproducible amount of fusion of the isolated vacuoles from the erg3Δ and erg2Δ strain pairs allowed us to determine the sterol dependency of the polyene antibiotic fusion inhibiting activity. The results are presented as the percentage of fusion relative to the specific amount of fusion obtained in the absence of polyene antibiotics of that particular erg deletion strain pair (Fig. (Fig.3C).3C). Filipin did not show any dependence on sterol structure for its inhibition of fusion. This is in accordance with the lack of dependence on sterol structure for its binding to membranes or inhibition of yeast growth (32). Natamycin and nystatin show similar inhibition patterns. A loss of the double bonds in the B-ring by deletions of ERG3 (5,6 position) and especially ERG2 (7,8 position) showed a loss of inhibition caused by natamycin and nystatin (Fig. (Fig.3C).3C). Changes in the lipid-embedded tail from deletion of ERG4 did not have a significant effect on natamycin or nystatin inhibition. Together, these results indicate that the presence of double bonds in the B-ring of the sterol, specifically at the 7,8 position, are essential for the ability of natamycin and nystatin to inhibit vacuole fusion. Similar structural requirements for sterols have been observed for natamycin and nystatin in their membrane binding and inhibition activity toward yeast cells (32). Therefore, the inhibition of vacuole fusion caused by natamycin and nystatin is most likely directly related to their binding of sterols in the membrane.

FIG. 3.

Ergosterol dependency of the polyene antibiotics for vacuole fusion. (A) The structure of ergosterol together with different proteins in the ergosterol biosynthesis pathway and their functions as indicated (17). (B and C) The effect of the polyene antibiotics (more ...)

Stage-specific inhibition of fusion by the polyene antibiotics.

To determine how natamycin inhibits fusion, we examined its effect on the different stages in the fusion process (9, 36). Homotypic yeast vacuole fusion occurs in three different stages: priming, docking, and fusion (36). Because different stages are dependent on different proteins, it is possible to examine fusion reactions of isolated vacuoles by using specific inhibitors. For example, the priming and docking stages can be inhibited by antibodies against Sec18p and Vam3p, respectively, proteins that are essential for these steps of the fusion process (9, 13). To inhibit the fusion stage, a vacuole fusion reaction is placed on ice. In this manner, a staging assay can be performed, where the specific inhibition profiles of the polyene antibiotics can be compared to the inhibition profiles of the known inhibitors. Figure Figure4A4A shows the inhibition profiles of the controls, anti-Sec18 (priming), anti-Vam3 (docking), and ice (fusion). Vacuole priming occurs within the first 30 min of fusion reactions. Thus, compounds that inhibit this stage of the fusion process show inhibition only within this time span. Indeed, the inhibition profile of anti-Sec18 fits this criterion. Conversely, compounds that inhibit the final stage, membrane bilayer mixing, will show inhibition throughout the whole time span of the reaction. Such an inhibition profile shows the least amount of fusion (Fig. (Fig.4A,4A, Ice). Inhibitors of the docking stage will display their activity between the profiles of the priming and the fusion (Fig. (Fig.4A,4A, anti-Vam3). For clarity, the inhibition profiles of the polyene antibiotics are compared separately to those of the controls for filipin, nystatin, and natamycin (Fig. 4B to D, respectively).

FIG. 4.

Inhibitory staging assay of vacuole fusion with the polyene antibiotics filipin, nystatin, and natamycin. Vacuoles isolated from yeasts with a wild-type sterol composition (3 μg protein of KTY1 and KTY2 each) were incubated with different inhibitors (more ...)

The inhibition profile of filipin lies in between the docking and fusion profiles, indicating it most likely inhibits between these stages (Fig. (Fig.4B).4B). This fits with the membrane-permeabilizing activity of filipin, because a pH gradient is necessary for the docking stage (9, 33). The profile of nystatin inhibition overlaps with that of the control for the priming, anti-Sec18, which indicates it acts on the priming (Fig. (Fig.4C).4C). Natamycin also showed an inhibition profile similar to the profile of priming (Fig. (Fig.4D).4D). Given its inability to cause membrane permeabilization and the specific interaction of natamycin with ergosterol, this indicates that the effect of natamycin is related to an ergosterol-dependent function in the priming stage of vacuole fusion.

Effect of natamycin on cellular morphology of vacuoles.

To determine if natamycin also has an effect on vacuole fusion in intact yeast cells, cells were incubated with different concentrations of natamycin for 5 h and stained with the vacuolar membrane marker, MDY-64 (Fig. (Fig.5).5). Filipin and nystatin were not included in this assay, because their permeabilizing effect on the yeast membrane may cause a free entry of the dye into the cell (34). Most control cells, untreated with natamycin, have more than one vacuole per cell, and treatment with natamycin resulted in a fragmentation of the vacuoles, already visible after treatment with 0.5 μM, which is about 30% of the MIC of this strain for natamycin (Fig. (Fig.5A)5A) (32). A quantification of the number of vacuoles per yeast cell shows that approximately 98% of untreated cells had two vacuoles maximally (Fig. (Fig.5B).5B). However, after incubation with natamycin, this number dropped to 35% owing to an increase in the number of vacuoles per cell to a maximum of 6 in 5% of the cases. These results show that natamycin treatment of whole yeast cells results in the fragmentation of the vacuoles.

FIG. 5.

The effect of natamycin on the cellular morphology of vacuoles in yeast. (A) Yeast cells were incubated with 0, 0.5, 1.0, or 1.5 μM natamycin for 5 h at 25°C. This was followed by staining with MDY-64 and fluorescence microscopy. Bar = (more ...)

DISCUSSION

Although the biological consequences of the action of the polyene antifungal compound natamycin are not known, the mode of action is thought to arise via a specific interaction with ergosterol but does not involve membrane permeabilization. In this study, we have demonstrated that natamycin is able to interfere in the process of vacuole fusion in a sterol-dependent manner. This inhibition also did not involve membrane permeabilization and seemed to take place early in the fusion mechanism, even before any membrane contact had occurred.

Ergosterol is known to be important during fusion and fission processes, including vacuole fusion (12, 19, 31). To determine if natamycin was able to act on these processes via its specific interaction with ergosterol, the effects of this antibiotic on the fusion of isolated yeast vacuoles were studied using a content mixing assay (15, 37). Indeed, natamycin was shown to inhibit the fusion process of isolated vacuoles. In addition, this inhibition was not related to a permeabilizing effect, similar to natamycin's inability to permeabilize model membranes or yeast cells (32). The sterol structure dependency of the vacuolar fusion inhibition by natamycin was almost identical to the sterol structure dependency for its activity toward yeast cells and its binding to sterols in model membranes (32). All were dependent on the presence of sterols containing double bonds in the B-ring, most importantly at the 7,8 position (32). Therefore, we conclude that natamycin inhibits vacuolar fusion through the specific interaction with ergosterol.

Treatment of yeast cells with natamycin led to a fragmented vacuolar morphology that is characteristic of a defect in vacuole fusion (1, 2). A similar vacuolar morphology has been observed in conidia of Penicillium discolor upon natamycin treatment (M. R. van Leeuwen and J. Dijksterhuis, unpublished observations). We therefore conclude that natamycin is able to inhibit vacuole fusion both in purified vacuoles and in intact yeast cells. Besides this inhibition of vacuolar fusion, natamycin may act on more ergosterol-dependent membrane fusion and fission processes through its interaction with ergosterol (23, 37). Indeed, natamycin has been shown to inhibit the early stages of endocytosis in the fungus P. discolor (34), an ergosterol-dependent fission process (17, 24). This suggests that the basis of the toxicity of natamycin could be the inhibition of fusion and fission processes. To act on vacuole fusion in an intact yeast cell most likely requires natamycin to enter this cell. This could be either via permeation across the plasma membrane or in an early stage via endocytosis. Currently we have no information on whether natamycin enters the cell and if so via which mechanism this occurs.

The polyene antibiotics nystatin and filipin were shown to be more efficient in inhibiting the fusion of isolated vacuoles. These differences are probably directly related to the relative affinity of the polyenes for ergosterol and their differences in membrane-permeabilizing activity. Nystatin and natamycin had binding affinities similar to those of ergosterol (32), yet nystatin is more efficient in its inhibition of vacuole fusion. This is best explained by the ability of nystatin to permeabilize the vacuole membrane, thereby increasing its efficacy of vacuole fusion inhibition. Filipin displayed the highest affinity for ergosterol, and it severely damages the membrane barrier (10, 32). Altogether, this likely explains why filipin was the most efficient inhibitor of vacuole fusion in our assays.

What would be the mechanism behind the inhibition of fusion caused by natamycin? We have observed that through the specific interaction with ergosterol, natamycin was able to act on the early priming stage of fusion. During this phase, no actual contact between the vacuolar membranes has taken place (25), making it unlikely that natamycin will act on lipid reorganization. The priming phase consists solely of the rearrangements of different protein complexes (for reviews, see references 25 and 36). Thus, the most straightforward conclusion is that natamycin is able to disturb these rearrangements as a result of its binding to ergosterol, and this suggests a more general mode of action, namely, to disturb ergosterol-dependent protein functions.

This immediately poses the question of whether the other members of the family of polyene antibiotics, which all bind to ergosterol (3, 8, 11, 32), are also able to act on the priming stage through their interaction with ergosterol. Indeed, we have shown that nystatin is able to act on the priming stage, as was observed previously as well, and the same is true for amphotericin B (21). The effect of filipin is less clear, because we found it to act in between the docking and fusion stages, while in a different study filipin was shown to act on the priming stage (21). The differences in results are likely best explained by different assay conditions. These findings points to a dual mode of action for some members of the polyene antibiotic family, where all members have the basic ability to act through the inhibition of ergosterol-dependent protein functions, while the additional ability is to permeabilize the membrane. This relates to a freeze fracture electron microscopy study, where natamycin, nystatin, and filipin all produced distinct morphological effects on the fungal membrane, indicating the different end results according to the mechanisms involved in polyene-sterol interactions (22). Because natamycin has only the basic ability to bind ergosterol, it is the ideal candidate for studying the basic mode of action of the polyenes. In addition, this makes natamycin an interesting tool for cell biology when analyzing ergosterol-dependent protein functions.

Interestingly, there is another naturally produced family of antibiotics where several members are known to have a dual mode of action. This is the antibacterial lantibiotic family, a group of small antimicrobial peptides, a large part of which are known to bind the bacterial cell wall component lipid II and through this interaction block cell wall synthesis (5). In the group of lantibiotics that are able to bind lipid II, several members are long enough to span the lipid bilayer and have an additional ability to form pores (5, 16). This striking parallel shows how nature has repeatedly used dual modes of action for membrane-active antibiotics, and this might be applicable to other families of membrane active antibiotics as well.

Acknowledgments

We thank M. Logan for valuable research discussions and K. Tedrick for technical support in constructing the erg knockout strains (University of Alberta, Edmonton, Canada).

This work was supported by the Technology Foundation (STW) and Applied Science division of NWO (UBC.6524) and an operating grant to G.E. from the Canadian Institutes of Health Research (no. 53068). Y.M.T.W. received a short-term EMBO fellowship (ASTF 60-2008) to visit the laboratory of G.E. in Edmonton, Canada.

Footnotes

Published ahead of print on 12 April 2010.

REFERENCES

1. Baars, T. L., S. Petri, C. Peters, and A. Mayer. 2007. Role of the V-ATPase in regulation of the vacuolar fission-fusion equilibrium. Mol. Biol. Cell 18:3873-3882. [PMC free article] [PubMed]

2. Banta, L. M., J. S. Robinson, D. J. Klionsky, and S. D. Emr. 1988. Organelle assembly in yeast: characterization of yeast mutants defective in vacuolar biogenesis and protein sorting. J. Cell Biol. 107:1369-1383. [PMC free article] [PubMed]

3. Bolard, J. 1986. How do the polyene macrolide antibiotics affect the cellular membrane properties? Biochim. Biophys. Acta 864:257-304. [PubMed]

4. Brachmann, C. B., A. Davies, G. J. Cost, E. Caputo, J. Li, P. Hieter, and J. D. Boeke. 1998. Designer deletion strains derived from Saccharomyces cerevisiae S288C: a useful set of strains and plasmids for PCR-mediated gene disruption and other applications. Yeast 14:115-132. [PubMed]

5. Breukink, E., I. Wiedemann, C. van Kraaij, O. P. Kuipers, H. Sahl, and B. de Kruijff. 1999. Use of the cell wall precursor lipid II by a pore-forming peptide antibiotic. Science 286:2361-2364. [PubMed]

6. Brown, D. A., and E. London. 1998. Functions of lipid rafts in biological membranes. Annu. Rev. Cell Dev. Biol. 14:111-136. [PubMed]

7. Burger, K. N. 2000. Greasing membrane fusion and fission machineries. Traffic 1:605-613. [PubMed]

8. Clejan, S., and R. Bittman. 1985. Rates of amphotericin B and filipin association with sterols. A study of changes in sterol structure and phospholipid composition of vesicles. J. Biol. Chem. 260:2884-2889. [PubMed]

9. Conradt, B., A. Haas, and W. Wickner. 1994. Determination of four biochemically distinct, sequential stages during vacuole inheritance in vitro. J. Cell Biol. 126:99-110. [PMC free article] [PubMed]

10. de Kruijff, B., and R. A. Demel. 1974. Polyene antibiotic-sterol interactions in membranes of Acholeplasma laidlawii cells and lecithin liposomes. 3. Molecular structure of the polyene antibiotic-cholesterol complexes. Biochim. Biophys. Acta 339:57-70. [PubMed]

11. de Kruijff, B., W. J. Gerritsen, A. Oerlemans, P. W. van Dijck, R. A. Demel, and L. L. van Deenen. 1974. Polyene antibiotic-sterol interactions in membranes of Acholeplasma laidlawii cells and lecithin liposomes. II. Temperature dependence of the polyene antibiotic-sterol complex formation. Biochim. Biophys. Acta 339:44-56. [PubMed]

12. D'Hondt, K., A. Heese-Peck, and H. Riezman. 2000. Protein and lipid requirements for endocytosis. Annu. Rev. Genet. 34:255-295. [PubMed]

13. Eitzen, G., N. Thorngren, and W. Wickner. 2001. Rho1p and Cdc42p act after Ypt7p to regulate vacuole docking. EMBO J. 20:5650-5656. [PubMed]

14. Ghannoum, M. A., and L. B. Rice. 1999. Antifungal agents: mode of action, mechanisms of resistance, and correlation of these mechanisms with bacterial resistance. Clin. Microbiol. Rev. 12:501-517. [PMC free article] [PubMed]

15. Haas, A. 1995. A quantitative assay to measure homotypic vacuole fusion in vitro. Methods Cell Sci. 17:283-294.

16. Hasper, H. E., N. E. Kramer, J. L. Smith, J. D. Hillman, C. Zachariah, O. P. Kuipers, B. de Kruijff, and E. Breukink. 2006. An alternative bactericidal mechanism of action for lantibiotic peptides that target lipid II. Science 313:1636-1637. [PubMed]

17. Heese-Peck, A., H. Pichler, B. Zanolari, R. Watanabe, G. Daum, and H. Riezman. 2002. Multiple functions of sterols in yeast endocytosis. Mol. Biol. Cell 13:2664-2680. [PMC free article] [PubMed]

18. Isgandarova, S., L. Jones, D. Forsberg, A. Loncar, J. Dawson, K. Tedrick, and G. Eitzen. 2007. Stimulation of actin polymerization by vacuoles via Cdc42p-dependent signaling. J. Biol. Chem. 282:30466-30475. [PubMed]

19. Jin, H., J. M. McCaffery, and E. Grote. 2008. Ergosterol promotes pheromone signaling and plasma membrane fusion in mating yeast. J. Cell Biol. 180:813-826. [PMC free article] [PubMed]

20. Kane, P. M. 2006. The where, when, and how of organelle acidification by the yeast vacuolar H+-ATPase. Microbiol. Mol. Biol. Rev. 70:177-191. [PMC free article] [PubMed]

21. Kato, M., and W. Wickner. 2001. Ergosterol is required for the Sec18/ATP-dependent priming step of homotypic vacuole fusion. EMBO J. 20:4035-4040. [PubMed]

22. Kitajima, Y., T. Sekiya, and Y. Nozawa. 1976. Freeze-fracture ultrastructural alterations induced by filipin, pimaricin, nystatin and amphotericin B in the plasma membranes of Epidermophyton, Saccharomyces and red complex-induced membrane lesions. Biochim. Biophys. Acta 455:452-465. [PubMed]

23. Mayer, A. 2002. Membrane fusion in eukaryotic cells. Annu. Rev. Cell Dev. Biol. 18:289-314. [PubMed]

24. Munn, A. L., A. Heese-Peck, B. J. Stevenson, H. Pichler, and H. Riezman. 1999. Specific sterols required for the internalization step of endocytosis in yeast. Mol. Biol. Cell 10:3943-3957. [PMC free article] [PubMed]

25. Ostrowicz, C. W., C. T. Meiringer, and C. Ungermann. 2008. Yeast vacuole fusion: a model system for eukaryotic endomembrane dynamics. Autophagy 4:5-19. [PubMed]

26. Peters, C., M. J. Bayer, S. Buhler, J. S. Andersen, M. Mann, and A. Mayer. 2001. Trans-complex formation by proteolipid channels in the terminal phase of membrane fusion. Nature 409:581-588. [PubMed]

27. Peters, C., and A. Mayer. 1998. Ca2+/calmodulin signals the completion of docking and triggers a late step of vacuole fusion. Nature 396:575-580. [PubMed]

28. Rodriguez, R. J., C. Low, C. D. Bottema, and L. W. Parks. 1985. Multiple functions for sterols in Saccharomyces cerevisiae. Biochim. Biophys. Acta 837:336-343. [PubMed]

29. Salaun, C., D. J. James, and L. H. Chamberlain. 2004. Lipid rafts and the regulation of exocytosis. Traffic 5:255-264. [PMC free article] [PubMed]

30. Simons, K., and E. Ikonen. 1997. Functional rafts in cell membranes. Nature 387:569-572. [PubMed]

31. Tedrick, K., T. Trischuk, R. Lehner, and G. Eitzen. 2004. Enhanced membrane fusion in sterol-enriched vacuoles bypasses the Vrp1p requirement. Mol. Biol. Cell 15:4609-4621. [PMC free article] [PubMed]

32. te Welscher, Y. M., H. H. ten Napel, M. Masia Balague, C. M. Souza, H. Riezman, B. de Kruijff, and E. Breukink. 2008. Natamycin blocks fungal growth by binding specifically to ergosterol without permeabilizing the membrane. J. Biol. Chem. 283:6393-6401. [PubMed]

33. Ungermann, C., W. Wickner, and Z. Xu. 1999. Vacuole acidification is required for trans-SNARE pairing, LMA1 release, and homotypic fusion. Proc. Natl. Acad. Sci. U. S. A. 96:11194-11199. [PubMed]

34. Van Leeuwen, M. R., E. A. Golovina, and J. Dijksterhuis. 2009. The polyene antimycotics nystatin and filipin disrupt the plasma membrane, whereas natamycin inhibits endocytosis in germinating conidia of Penicillium discolor. J. Appl. Microbiol. 106:1908-1918. [PubMed]

35. Van Leeuwen, M. R., W. Smant, W. de Boer, and J. Dijksterhuis. 2008. Filipin is a reliable in situ marker of ergosterol in the plasma membrane of germinating conidia (spores) of Penicillium discolor and stains intensively at the site of germ tube formation. J. Microbiol. Methods 74:64-73. [PubMed]

36. Wickner, W. 2002. Yeast vacuoles and membrane fusion pathways. EMBO J. 21:1241-1247. [PubMed]

37. Wickner, W., and A. Haas. 2000. Yeast homotypic vacuole fusion: a window on organelle trafficking mechanisms. Annu. Rev. Biochem. 69:247-275. [PubMed]

Articles from Antimicrobial Agents and Chemotherapy are provided here courtesy of
American Society for Microbiology (ASM)