Classification of Polyene Antibiotics According to Chemical Structure and Biological Effects

Fourteen polyene antibiotics and six of their semisynthetic derivatives were compared for their effects on potassium (K(+)) leakage and lethality or hemolysis of either Saccharomyces cerevisiae or mouse erythrocytes. These polyene antibiotics fell into two groups. Group I antibiotics caused K(+) leakage and cell death or hemolysis at the same concentrations of added polyene. In this group fungistatic and fungicidal levels were indistinguishable. Group I drugs included one triene (trienin); tetraenes (pimaricin and etruscomycin); pentaenes (filipin and chainin); one hexaene (dermostatin); and one polyene antibiotic with unknown chemical structure (lymphosarcin). Group II antibiotics caused considerable K(+) leakage at low concentrations and cell death or hemolysis at high concentrations. The fungistatic levels were clearly separable from fungicidal. This group included the heptaenes (amphotericin B, candicidin, aureofungin A and B, hamycin A and B), and five of their semisynthetic derivatives (amphotericin B methyl ester, N-acetyl-amphotericin B, hamycin A and B methyl esters, and N-acetyl-candicidin). Nystatin, classified as a tetraene, and its derivative, N-acetyl nystatin, also were in this group.     

Potentiation of antiviral activity of acyclovir by polyene macrolide antibiotics.

The potentiation of the antiviral activity of acyclovir [9-[(2-hydroxyethoxy)methyl]guanine] by polyene macrolide antibiotics has been studied as a function of the macrolide structure. The 12 polyenes chosen for this study represented the major structural groups of these antibiotics and induced in mammalian cells repairable membrane alterations or irreversible cell damage. The potentiating activity of the polyene macrolides was determined based on the differential decrease of in vitro production of infectious virions in the presence of acyclovir alone or in combination with the polyene. Pseudorabies virus, a representative herpesvirus susceptible to acyclovir, was replicated in BHK-21 cells grown in serum-free medium to avoid the interference of serum factors in the polyene macrolide-cell interaction. The potentiation activity of the polyene antibiotics was concentration dependent. The enhancement of the antiviral activity of acyclovir was observed at polyene concentrations which had no direct effect on pseudorabies virus replication in BHK-21 cells. The optimal potentiating concentrations of polyenes were 2 to 15 times lower than that inducing 50% of potassium efflux from BHK-21 cells. The highest potentiating activity was observed for the methyl ester of the trimethylammonium derivative of aureofacin B, which reduced the pseudorabies virus titer by two orders of magnitude. Potentiation by polyene macrolides appeared to coincide with the K+-dependent membrane repair process. The acyclovir potentiating activity was associated with polyene macrolide antibiotics having a large and rigid macrolide ring (amphotericin B and aureofacin). Polyene antibiotics with small and rigid (pimaricin and filipin) or large but flexible (nystatin A1 and lienomycin) macrolide rings showed no potentiation of the antiviral effect of acyclovir.     

Measurement of Polyene Antibiotic-Mediated Erythrocyte Damage by Release of Hemoglobin and Radioactive Chromium

Polyene antifungal antibiotics produce various degrees of membrane damage in sheep erythrocytes in vitro. Mediocidin, filipin, amphotericin B, and candicidin were found to result in greater damage than nystatin, pimaricin, and amphotericin B methyl ester. The degree of sensitivity of the cells varied by 100-fold for mediocidin verus amphotericin B methyl ester as measured by curves of hemoglobin release versus drug concentration. In erythrocytes prelabeled with radioactive chromium, release of the isotope through polyene-damaged cell membranes was found to occur at lower drug concentrations than measurable hemoglobin release, and the percentage of isotope released at the highest drug dose was consistently greater than the percentage of hemoglobin released. Thus, the isotope assay is a more sensitive indicator of polyene-induced membrane damage in the test system. These significant differences in release of molecules through polyene-induced membrane lesions indicate the complex nature of the binding and further interactions of this class of drugs with the plasma membrane.     

Effect of fatty acids on action of polyene antibiotics

Fatty acids cause a decrease in the absorption spectra of the antifungal polyene macrolide antibiotics nystatin, filipin, candicidin, and amphotericin B. For nystatin, filipin, and candicidin, this decrease in absorption could be correlated with the activity of the fatty acids in protecting the yeast Saccharomyces cerevisiae against the action of these antibiotics. With amphotericin B a correlation was observed between the decrease in absorption caused by certain derivatives of fatty acids and the protective action of these derivatives against the activity of amphotericin B on yeast. It is concluded that, like the sterols, fatty acids also interact with the polyene antibiotics and thereby reduce their effective concentrations.     

Dissociation between the induction of potassium efflux and cytostatic activity of polyene macrolides in mammalian cells.

The paper contains data on the induction of K+ efflux and viability of baby hamster kidney (BHK-21) cells after their treatment with macrolide antibiotics inducing specific pores in membrane. New water-soluble semisynthetic derivatives of amphotericin B and aureofacin (N-glycosyl and trimethylammonium methyl ester derivatives) as well as the parent compounds were used to compare the concentration of antibiotics inducing permeabilizing and cytostatic effects. We found that a two- to eight-times-higher concentration of polyene antibiotic was required to observe a cytostatic effect than for release of 50% of the cellular potassium (K50 concentration) from BHK-21 cells. These differences were larger for water-soluble derivatives than for the parent compounds. The amount of intracellular potassium in treated cells incubated under optimal growth conditions was higher than in cells which had been further washed with K+-free maintenance medium. The membrane permeability changes induced by low concentrations of specific polyenes were observed to be reversible. BHK-21 cells were able to repair polyene-induced membrane permeability within 3 to 12 h under optimal growth conditions, after cell treatment with K50 concentration of specific macrolide antibiotics. The repair phenomenon is postulated as an explanation for the dissociation observed between permeabilizing and cytostatic effect of specific polyenes in BHK-21 cells.     

Classification of polyene antibiotics according to their synergistic effect in combination with bleomycin A2 or fusidic acid

Five polyene antibiotics were compared for their effects on colony formation of either Chinese hamster V79 or Saccharomyces cerevisiae cells. A 10 to 40 times higher concentration of amphotericin B (heptaene) or nystatin (degenerated heptaene) was necessary to inhibit colony formation of hamster cells than that needed to inhibit colony formation of yeast cells. In contrast, colony formation of both hamster and yeast cells was inhibited to the same extent by similar concentrations of filipin (pentaene), pentamycin (pentaene), or pimaricin (tetraene). The five polyene antibiotics were also compared for their effects on colony formation of either V79 or S. cerevisiae cells when combined with a nonpolyene antibiotic, fusidic acid or bleomycin A2. Amphotericin B or nystatin could augment the cytocidal effect of fusidic acid but not that of bleomycin A2, whereas pentamycin or pimaricin could augment the cytocidal effect of both fusidic acid and bleomycin A2 against hamster and yeast cells. Filipin was found to enhance the action of fusidic acid and bleomycin upon growth of mammalian cells, whereas the pentaene polyene significantly potentiated the action of fusidic acid, but not that of bleomycin A2, against S. cerevisiae. It was therefore suggested that these polyene antibiotics be classified into two groups: group 1 (pimaricin, pentamycin, and filipin) and group 2 (amphotericin B and nystatin).     

Response of gene expression in Saccharomyces cerevisiae to amphotericin B and nystatin measured by microarrays

The response of the yeast Saccharomyces cerevisiae to two polyene antibiotics, amphotericin B and nystatin, was studied by genomic expression profiling. The two agents produced highly similar expression pattern changes, which was consistent with their known identical mechanisms of action on cell membranes. Detailed analysis was focused on the amphotericin B-treated sample in this study. Our data showed that genes involved in mitochondrial ribosomal protein synthesis were more severely repressed than those in the cytoplasm, which might contribute to the cytotoxicity of amphotericin B. To counteract the leakage of intracellular nutrients and ions from the pores in the cell membrane caused by amphotericin B, c. 17 genes involved in transport facilitation were induced, presumably to allow more efficient uptake of nutrients and ions. The expression level of five genes involved in ergosterol synthesis dropped and three genes related to cell wall biogenesis were induced, indicating that the cell membrane and cell wall were also affected by the presence of polyene antibiotics. It was observed that the pleiotropic drug resistance network in yeast cells was activated after exposure to amphotericin B, possibly contributing to the acquisition of amphotericin B resistance. Part of the gene expression alteration measured by microarray was confirmed by quantitative RT-PCR.     

Incidence of polyene-resistant yeasts recovered from clinical specimens.

The development of resistance to amphotericin B and nystatin in yeast isolates was determined. Organisms recovered from patients on the oncology service, undergoing extensive chemotherapy for acute leukemia and bone marrow transplantation, were compared with yeasts recovered from patients on other services in the same hospital over a 7-month period. An agar dilution method was used to assay the susceptibility for each antibiotic; resistance was defined as a minimal inhibitory concentration of greater than or equal to 2 micrograms/ml for amphotericin B and greater than or equal to 16 micrograms/ml for nystatin. None of 625 isolates from 238 patients on non-oncology services demonstrated polyene resistance. Resistance only occurred in a subpopulation of oncology patients, in which 55 isolates (7.4%) from six patients (8.6%) exhibited polyene resistance. Resistance yeasts included Candida albicans (three strains), Candida tropicalis (one strain), and Torulopsis glabrata (two strains). All of the patients from whom resistant yeasts were recovered had experienced extensive chemotherapy with cytotoxic agents, granulocytopenia, and long-term treatment with both antibacterial and polyene antibiotics. Resistance to 2 micrograms of amphotericin B per ml and to 16 micrograms of nystatin per ml was associated with loss or marked depression of ergosterol in the cell membrane as measured by ultraviolet spectra. A significant incidence of polyene resistance in an oncology subpopulation was documented, suggesting a need for susceptibility testing in patients who are at high risk for development of drug-resistant fungal pathogens.     

Bioassay Method for Polyene Antibiotics Based on the Measurement of Rubidium Efflux from Rubidium-Loaded Yeast Cells

A bioassay method for the polyene antibiotics nystatin and amphotericin B is proposed based on the measurement of the efflux of rubidium ions from a rubidium-loaded yeast culture challenged with the antibiotics. For this purpose a major proportion of the intracellular K(+) ions in a Saccharomyces cerevisiae culture has been substituted by Rb(+) ions. The rubidium leakage is measured by atomic absorption spectrophotometry, and a straight-line, dose-response correlation has been obtained for both antibiotics.     

In-vitro antifungal activity of liposomal nystatin in comparison with nystatin, amphotericin B cholesteryl sulphate, liposomal amphotericin B, amphotericin B lipid complex, amphotericin B desoxycholate, fluconazole and itraconazole.

The in-vitro susceptibilities of 120 clinical isolates of yeasts to liposomal nystatin were compared with those to amphotericin B lipid complex (ABLC), liposomal amphotericin B (LAB), amphotericin B cholesteryl sulphate (ABCD), amphotericin B desoxycholate, nystatin, fluconazole and itraconazole. Yeast isolates examined included strains of Candida albicans, Candida parapsilosis, Candida glabrata, Candida krusei, Candida guilliermondii, Candida tropicalis, Candida kefyr, Candida viswanathii, Candida famata, Candida rugosa, Rhodotorula rubra, Trichosporon spp., Cryptococcus laurentii and Cryptococcus neoformans. The mean MICs for all strains examined were: liposomal nystatin 0.96 mg/L; nystatin 0.54 mg/L; ABLC 0.65 mg/L; LAB 1.07 mg/L; ABCD 0.75 mg/L; amphotericin B 0.43 mg/L; fluconazole 5.53 mg/L; and itraconazole 0.33 mg/L. No significant differences were seen between the activity of liposomal nystatin and the polyene drugs or itraconazole, but liposomal nystatin was more active than fluconazole. MICs were lower than the reported blood concentrations following therapeutic doses of this drug, indicating the potential for a therapeutic use of liposomal nystatin in humans. These results indicate good activity in vitro against medically important yeasts, which compares favourably with the activities of other currently available antifungal drugs. Liposomal nystatin may have a role in the treatment of disseminated and systemic mycoses.     

Itraconazole versus amphotericin B plus nystatin in the prophylaxis of fungal infections in neutropenic cancer patients.

The efficacy and safety of itraconazole oral solution and a combination of amphotericin B capsules plus nystatin oral suspension were compared in the prophylaxis of fungal infections in neutropenic patients. In an open, randomized, multicentre trial, 144 patients received itraconazole oral solution 100 mg bd, and 133 patients received amphotericin B 500 mg tds plus nystatin 2 MU qds. Overall, 65% of itraconazole-treated patients were considered to have had successful prophylaxis, compared with 53% in the polyene group. Proven deep fungal infections occurred in 5% of patients in each group. Fewer patients receiving itraconazole than amphotericin plus nystatin had superficial infections (3 versus 8%; P = 0.066). This trend in favour of itraconazole was seen in patients with profound neutropenia (neutrophil count <0.1 x 10(9) cells/L at least once) or prolonged neutropenia (neutrophil count <1.0 x 10(9) cells/L for >14 days). The median time to prophylactic failure was longer in the itraconazole group (37 days) than in the polyene group (34 days). The number of patients with fungal colonization (nose, sputum, stool) changed more favourably from baseline to endpoint in the itraconazole group than in the polyene group. Both treatments were safe and well tolerated; however, patients receiving amphotericin plus nystatin had a higher incidence of nausea and rash. In conclusion, itraconazole oral solution at doses of 100 mg bd and oral amphotericin B plus nystatin have similar prophylactic efficacy against fungal infections in neutropenic patients. On the basis of reduced incidence of superficial fungal infections, fungal colonization and specific adverse events, itraconazole may be the preferred treatment.     

Carbon Catabolite Regulation of Cephalosporin Production in Streptomyces clavuligerus

The production of interferon by polyriboinosinic-polyribocytidylic acid [poly(I) . poly(C)] and poly(I) . poly(C)-diethylaminoethyl-dextran in L929 cells was enhanced from 10 to 100 times by polyene macrolides, including amphotericin B (AmB), AmB methyl ester, nystatin, and filipin. AmB and its water-soluble methyl ester were the most effective; retinol, a nonmacrolide polyene, was ineffective. Interferon induction by Newcastle disease virus was not enhanced by AmB. The kinetics of interferon production were not markedly altered by AmB. Polyenes and poly(I) . poly(C)-diethylaminoethyl-dextran did not need to be present on cells simultaneously to enhance interferon production. Pretreatment with polyenes was as effective as simultaneous addition. Even treatment of washed cells, several hours after removal of poly(I) . poly(C)-diethylaminoethyl-dextran, resulted in enhancement of interferon production. AmB did not appear to form a macromolecular complex with poly(I) . poly(C) in that neither the ultraviolet absorption spectrum nor the melting point of poly(I) . poly(C) was altered by mixing with AmB. Isotopic studies indicated that AmB did not enhance binding of poly(I) . poly(C) to cells. Since the macrolide polyenes have been demonstrated to bind to cell membrane sterols with subsequent alterations in membrane permeability barriers, they may enhance interferon production by increasing cell penetration of poly(I) . poly(C).     

Comparative Toxicological Studies of Amphotericin B Methyl Ester and Amphotericin B in Mice, Rats, and Dogs

In acute and subacute toxicological studies, amphotericin B methyl ester was shown to be much less toxic than the parent antibiotic. As a single intravenous dose in mice, the methyl ester was approximately 20 times less toxic than amphotericin B. Also, the acute toxicity of the methyl ester in mice was not enhanced by the presence of chemically induced hepatic or renal damage or by the concurrent administration of amphotericin B or flucytosine. In a 1-month intraperitoneal study in rats, the methyl ester was about one-fourth as nephrotoxic as amphotericin B. In a 1-month intravenous study in dogs, the methyl ester was about one-eighth as nephrotoxic and one-fourth to one-half as hepatotoxic as the parent compound. In addition, the methyl ester, unlike amphotericin B, produced minimal renal effects, which did not increase in severity with increasing dosage. Based on the results of these studies, it is concluded that amphotericin B methyl ester has the potential for an improved therapeutic ratio in the treatment of systemic mycoses.     

Comparative Susceptibility of Four Kinds of Pathogenic Fungi to Amphotericin B and Amphotericin B Methyl Ester

The activity of amphotericin B methyl ester was compared with that of amphotericin B, using Candida albicans (34 isolates), Torulopsis glabrata (12 isolates), Filobasidiella neoformans (stat. conid. Cryptococcus neoformans) (14 isolates), and Coccidioides immitis (37 isolates) and tube dilution in a totally synthetic, completely defined medium (SAAMF) with inocula of 10(4) colony-forming units per ml. Minimal inhibitory concentrations were read after 24 h at 34 degrees C for C. albicans and T. glabrata, and after 48 h at 34 degrees C for C. immitis and F. neoformans. Minimal lethal concentrations were determined by subculture of 10% of the volume of the cultures without evident growth onto Sabouraud glucose agar medium. Overall, amphotericin B methyl ester was slightly less active than amphotericin B, with the differences attaining statistical significance for: (i) inhibition of C. albicans and T. glabrata and (ii) killing of T. glabrata.     

Induction of amphotericin B-specific antibodies for use in immunoassays.

High-titered antisera specific for amphotericin B (AmB) were induced by immunization with a protein conjugate of the D-lysyl AmB methyl ester. These polyclonal anti-AmB antibodies reacted preferentially with AmB or the AmB methyl ester and discriminated sharply between nystatin and AmB. A solid-phase radioimmunoassay was developed with radioiodinated immunoglobulin G fractions derived from the anti-AmB antisera. This assay was capable of detecting AmB in the sera in the same concentration range that is regularly achieved during AmB treatment of systemic fungal infections. This study demonstrated the feasibility of immunoassays in measuring the concentration of AmB in blood and tissue fluids.     

Molecular basis for the selective toxicity of amphotericin B for yeast and filipin for animal cells

Among the polyene antibiotics, many, like filipin, cannot be used clinically because they are toxic; amphotericin B, however, is useful in therapy of human fungal infections because it is less toxic. Both the toxicity of filipin and the therapeutic value of amphotericin B can be rationalized at the cellular and molecular level by the following observations: (i) these polyene antibiotics showed differential effects on cells; filipin was more potent in lysing human red blood cells, whereas amphotericin B was more potent in inhibiting yeast cell growth; and (ii) the effects of filipin were more efficiently inhibited by added cholesterol, the major membrane sterol in human cells, whereas the effects of amphotericin B were more efficiently inhibited by ergosterol, the major membrane sterol in yeast. The simplest inference is that the toxicity and effectiveness of polyenes are determined by their relative avidities for the predominant sterol in cell membranes.     

Comparison of in vitro activity of liposomal nystatin against Aspergillus species with those of nystatin, amphotericin B (AB) deoxycholate, AB colloidal dispersion, liposomal AB, AB lipid complex, and itraconazole

We compared the in vitro activity of liposomal nystatin (Nyotran) with those of other antifungal agents against 60 Aspergillus isolates. Twelve isolates were itraconazole resistant. For all isolates, geometric mean (GM) MICs (micrograms per milliliter) were 2.30 for liposomal nystatin, 0.58 for itraconazole, 0.86 for amphotericin B (AB) deoxycholate, 9.51 for nystatin, 2.07 for liposomal AB, 2.57 for AB lipid complex, and 0.86 for AB colloidal dispersion. Aspergillus terreus (GM, 8.72 micrograms/ml; range, 8 to 16 micrograms/ml) was significantly less susceptible to all of the polyene drugs than all other species (P = 0.0001).     

Comparison of amphotericin B and N-D-ornithyl amphotericin B methyl ester in experimental cryptococcal meningitis and Candida albicans endocarditis with pyelonephritis.

Amphotericin B and N-D-ornithyl amphotericin B methyl ester were compared for therapeutic efficacies against experimentally induced cryptococcal meningitis and Candida albicans endocarditis with pyelonephritis in rabbits. Antifungal activity of the two polyenes in vitro was similar for the yeasts used in these experiments. N-D-ornithyl amphotericin B methyl ester gave a slightly higher concentration in serum than amphotericin B did, but both drugs had similar elimination curves, and penetration into the cerebrospinal fluid was poor for both. Despite these similarities between the two polyenes, amphotericin B was much more effective than N-D-ornithyl amphotericin B methyl ester in the treatment of cryptococcal meningitis in rabbits. For C. albicans endocarditis, both polyenes had similar cure rates, but in vitro measurement of fungicidal activity in serum did not predict treatment outcome. For C. albicans pyelonephritis, both polyenes showed efficacy; because higher doses of the less toxic methyl ester could be used, it sterilized the urinary tract more often than amphotericin B. These studies indicate that in vivo and in vitro experiments may be needed to predict the results of treatment with polyenes.     

Exposure to subtherapeutic concentrations of polyene antifungals suppresses the adherence of Candida species to denture acrylic

BACKGROUND: The adherence of Candida species to denture acrylic is the initial event leading to Candida-associated denture stomatitis, with Candida albicans being the main aetiological agent. However, the increased incidence of immunocompromised patients in the community has resulted in the emergence of a number of non-albicans Candida species as causative agents of this disease, which is commonly managed by topically delivered polyene antifungals. Hence, we investigated the effect of the exposure of denture acrylic surfaces to nystatin and amphotericin B on the subsequent adhesion of six different Candida species. METHODS: Acrylic strips were exposed to subtherapeutic concentrations of the two polyenes for 30 min, and the adhesion of 4 isolates each of C. albicans, Candida glabrata, Candida guilliermondii, Candida krusei, Candida parapsilosis and Candida tropicalis was assessed using a previously described in vitro method with slight modifications. RESULTS: Overall, the results indicated a 35.9% (p < 0.01) and 63.1% (p < 0. 01) reduction, respectively, in yeast adhesion to denture acrylic following exposure to nystatin and amphotericin B, although this effect was not uniform for all the tested isolates. Thus, all C. glabrata, 3 C. guilliermondii and a single isolate each of C. krusei, C. parapsilosis and C. tropicalis were not significantly affected by nystatin exposure, and a single isolate each of C. glabrata and C. guilliermondii were not significantly affected by amphotericin B. CONCLUSIONS: The present data, the first on the effect of polyenes on a wide range of Candida species, indicate that the in vitro exposure of denture acrylic to subtherapeutic concentrations of nystatin and amphotericin B suppresses the adherence of pathogenic Candida species in general.     

Activities of the modified polyene N-D-ornithyl amphotericin methyl ester and the azoles ICI 153066, Bay n 7133, and Bay l 9139 compared with those of amphotericin B and ketoconazole in the therapy of experimental blastomycosis.

We studied the efficacy of new experimental antifungal drugs, which represent molecular modifications of present active agents, in a murine model of blastomycosis. Ketoconazole, previously the best azole drug studied and which is protective when administered orally, was superior to a new oral imidazole, Bay l 9139, and a new oral triazole, Bay n 7133. A new oral triazole, ICI 153066, was markedly more effective than ketoconazole and is the only oral drug studied which came close to producing complete sterilization of all visceral infection in all animals treated. Amphotericin B, a polyene given parenterally, was shown to be more efficacious than any drug studied. It completely sterilized the infection. A modified polyene, N-D-ornithyl amphotericin methyl ester, was only slightly less effective on a milligram-per-kilogram basis.