Anti-Candida activity and in vitro toxicity screening of antifungals complexed with β-cyclodextrin

The emergence of resistant fungal species and the toxicity of currently available antifungal drugs are relevant issues that require special consideration. Cyclodextrins inclusion complexes could optimize the antimicrobial activity of such drugs and create a controlled release system with few side effects. This study aimed to assess the in vitro toxicity and antifungal effectiveness of nystatin (Nys) and chlorhexidine (Chx) complexed or not with β-cyclodextrin (βCD). First, a drug toxicity screening was performed through the Artemia salina bioassay. Then, the minimum inhibitory concentrations (MICs) against Candida albicans were determined with the broth microdilution test. After MICs determination, the cytotoxicity of the drugs was evaluated through the methyl-thiazolyl-tetrazolium (MTT) and neutral red (NR) assays and through cell morphology analysis. The PROBIT analysis was used to determine the median lethal concentration (LC50), and the cell viability values were submitted to one-way analysis of variance(ANOVA)/Tukey (α = 0.05). Overall, the βCD-complexed antifungals were less toxic against A. salina than their raw forms, suggesting that inclusion complexes can reduce the toxicity of drugs. The MICs obtained were as follows: Nys 0.5 mg/L; Nys:βCD 4 mg/L; Chx 4 mg/L; and Chx:βCD 8 mg/L. Chx showed significant cytotoxicity (MTT: 12.9 ± 9.6%; NR: 10.6 ± 12.5%) and promoted important morphological changes. Cells exposed to the other drugs showed viability above 70% with no cellular damage. These results suggest that antifungals complexed with βCD might be a biocompatible option for the treatment of Candida-related infections.

The global effect of exposing bakers' yeast to 5-fluoruracil and nystatin; a view to Toxichip

A genome-wide screen of a haploid deletion library of bakers' yeast (Saccharomyces cerevisiae) was conducted to document the phenotypic and transcriptional impact of exposure to each of the two pharmaceutical products 5-fluorouracil (an anti-tumor agent) and nystatin (an anti-fungal agent). The combined data set was handled by applying a systems biology perspective. A Gene Ontology analysis identified functional categories previously characterized as likely targets for both compounds. Induced transcription profiles were well correlated in yeast and human HepG2 cells. The identified molecular targets for both compounds were used to suggest a small set of human orthologues as appropriate for testing on human material. The yeast system developed here (denoted "Toxichip") has likely utility for identifying biomarkers relevant for health and environmental risk assessment applications required as part of the development process for novel pharmaceuticals.

Novel Nystatin A₁ derivatives exhibiting low host cell toxicity and antifungal activity in an in vitro model of oral candidosis

Opportunistic oral infections caused by Candida albicans are frequent problems in immunocompromised patients. Management of such infections is limited due to the low number of antifungal drugs available, their relatively high toxicity and the emergence of antifungal resistance. Given these issues, our investigations have focused on novel derivatives of the antifungal antibiotic Nystatin A1, generated by modifications at the amino group of this molecule. The aims of this study were to evaluate the antifungal effectiveness and host cell toxicity of these new compounds using an in vitro model of oral candidosis based on a reconstituted human oral epithelium (RHOE). Initial studies employing broth microdilution, revealed that against planktonic C. albicans, Nystatin A1 had lower minimal inhibitory concentration than novel derivatives. However, Nystatin A1 was also markedly more toxic against human keratinocyte cells. Interestingly, using live/dead staining to assess C. albicans and tissue cell viability after RHOE infection, Nystatin A1 derivatives were more active against Candida with lower toxicity to epithelial cells than the parent drug. Lactate dehydrogenase activity released by the RHOE indicated a fourfold reduction in tissue damage when certain Nystatin derivatives were used compared with Nystatin A1. Furthermore, compared with Nystatin A1, colonisation of the oral epithelium by C. albicans was notably reduced by the new polyenes. In the absence of antifungal agents, confocal laser scanning microscopy showed that C. albicans extensively invaded the RHOE. However, the presence of the novel derivatives greatly reduced or totally prevented this fungal invasion.

[Comparative analysis of in vitro antifungal activity and in vivo acute toxicity of the nystatin analogue S44HP produced via genetic engineering]

New polyene macrolide S44HP was purified from the culture of recombinant Streptomyces noursei strain with engineered nystatin polyketide synthase. S44HP, nystatin (NYS), and amphotericin B (Amph-B) were tested against 19 clinical fungal isolates in agar diffusion assay, which demonstrated clear differences in antifungal activities of these antibiotics. Sodium deoxycholate suspensions of all three antibiotics were subjected to acute toxicity studies in vivo upon intravenous administration in mice. NYS exhibited the lowest acute toxicity in mice in these experiments, while both Amph-B and S44HP were shown to be 4 times more toxic as judged from the LD50 values. While the acute toxicity of S44HP was higher than that of Amph-B, the data analysis revealed a significantly increased LD10 to LD50 dose interval for S44HP compared to Amph-B. The data revealed structural features of polyene macrolides, which might have an impact on both the activity and toxicity profiles of these antibiotics. These results represent the first example of preclinical evaluation of an "engineered" polyene macrolide, and can be valuable for rational design of novel antifungal drugs with improved pharmacological properties.

Liposomal Polyene Antibiotics

Polyene antibiotics (i.e., amphotericin B and nystatin) have been incorporated into lipid-based delivery systems to decrease their toxicity and enhance their therapeutic index, the most common being liposomes. This chapter describes the protocols for preparing liposomal amphotericin B and determining the efficacy and toxicity of the formulations in animals. Furthermore, methods for determining the pharmacokinetics and drug distribution after administration of amphotericin B in lipid-based delivery systems are discussed. Procedures for comparing the toxicity of different amphotericin B formulations in cell culture studies are also elucidated.

Protection by tetracyclines against ion transport disruption caused by nystatin in human airway epithelial cells

Polyene antifungal antibiotics like nystatin form monovalent cation pores on the plasma membrane that perturb the intracellular electrolyte milieu, resulting in cell damage. In the present study, we investigated the effects of tetracyclines (minocycline and tetracycline) on ion transporters disrupted by nystatin in cultured human airway Calu-3 cells. Apical application of nystatin (50 microM) on a monolayer of the cells stimulated Na(+)-K(+) pump activity as estimated by ouabain (1 mM)-sensitive short-circuit current (I(sc)). The nystatin-potentiated I(sc) was inhibited by minocycline (IC(50) = 25 microM) or tetracycline (IC(50) = 150 microM) applied only from the apical (nystatin-treated) side. Nystatin increased monolayer conductance that was reversed by the application of tetracyclines. In contrast, ouabain potentiated the nystatin-induced change in the conductance. Further, Na(+)-glucose transport affected by nystatin was also normalized by tetracyclines from the nystatin-treated side of the membrane. These data suggest that tetracyclines may lower the cell permeability potentiated by nystatin, protecting cells against damage.

Polyene antibiotics: relative degrees of in vitro cytotoxicity and potential effects on tubule phospholipid and ceramide content

Polyene antibiotic administration is limited by dose-dependent nephrotoxicity. The latter is believed to be mediated by polyene anchoring to plasma membrane cholesterol, resulting in pore formation, abnormal ion/solute flux, adenosine triphosphate (ATP) declines, and, ultimately, a loss of tubule viability. The relative nephrotoxicity of these agents and their liposomal preparations has remained poorly defined. Thus, freshly isolated mouse proximal tubules or cultured human proximal tubule (HK-2) cells were exposed to either nystatin, amphotericin B, or three different polyene liposomal preparations (Nyotran, AmBisome, or Abelcet; 4 to 64 microg/mL). The impact of these agents on (1) plasma membrane injury (sodium-driven ATP consumption, assessed by ATP-adenosine diphosphate [ADP] ratios); (2) cellular susceptibility to superimposed injury (chemical hypoxia or ferrous ammonium sulfate-mediated oxidative stress; assessed by lactate dehydrogenase release); and (3) membrane cholesterol, phospholipid, and ceramide expression was assessed. Amphotericin B was more cytotoxic than nystatin (approximately 25% to 50% greater ATP-ADP ratio declines). Most of this toxicity could be eliminated by polyene liposomal formulation. Nevertheless, the liposomal polyenes still fully sensitized tubule cells to superimposed chemical hypoxic (antimycin/deoxyglucose), but not oxidant, attack. Nystatin and amphotericin B caused acute increments in tubule sphingomyelin-phosphatidylcholine ratios and ceramide content (indicating an impact on the plasma membrane extending beyond the classic view of pore formation with ion flux). In conclusion, (1) nystatin is seemingly less cytotoxic than amphotericin B (in contrast to the prevailing clinical view); (2) liposomal formulation markedly decreases this cytotoxicity; (3) despite this reduced toxicity, liposomal polyenes are still able to render tubule cells more vulnerable to selected forms of superimposed injury; and (4) acute alterations in plasma membrane phospholipid and ceramide expression are previously unrecognized consequences and potential mediators of polyene-mediated tubular cell attack.

Ototoxicity of common topical antimycotic preparations

OBJECTIVE

To determine the ototoxic effects of five commonly used topical antimycotic agents-clotrimazole, miconazole, nystatin, tolnaftate, and gentian violet-in the guinea pig.

DESIGN

A controlled animal study in which the ototoxicity of commonly used topical antifungal agents was investigated by measurement of hair cell loss.

METHODS

Several readily available topical antimycotic preparations were instilled into the middle ears of female Hartley guinea pigs over a 1-week period. Two weeks after the last instillation, the animals were euthanized. An active control group was treated with neomycin to confirm the adequacy of the treatment in delivering a known ototoxin; an untreated control group defined the normal distribution of hair cells. The temporal bones were removed, and the cochleas were fixed and dissected. The basilar membranes were examined under the scanning electron microscope. A map of hair cell survival was made for each row in segments of each turn.

RESULTS

The untreated control animals had no discernible hair cell loss in the two lower turns. In the apical turn and sometimes the third turn, loss of hair cells was a common finding, this is a known effect of aging in this species. The animals treated with neomycin had damage consistently in the basal turn, sometimes extending into the second turn, as well as the expected hair cell loss in the apical turn. Clotrimazole, miconazole, or tolnaftate did not cause any hair cell loss in the first two turns. Hair cell loss in the third and fourth turns was similar to that of the untreated control group. Likewise, nystatin exhibited no evidence of ototoxicity. Of note, however, the preparation used in this study left a persistent residue in the round window niche. Of the first four animals treated with gentian violet, three developed pronounced behavioral signs of vestibular damage, and three demonstrated extensive middle ear inflammation and extensive new bone growth. Hair cell counts were not attempted because the extreme bone growth interfered with successful perfusion and dissection.

CONCLUSIONS

Extrapolating from guinea pigs to humans requires caution. However, it is likely that guinea pigs are, if anything, more susceptible to topical ototoxins than are humans. The specific antimycotics clotrimazole, miconazole, and tolnaftate appear to be safe. Gentian violet has the potential for severe damage. The persistent residue left by the nystatin preparation is cause for concern and is a reminder that both the active ingredient and vehicle must be considered in evaluation of safety.

The reproductive and developmental toxicity of the antifungal drug Nyotran (liposomal nystatin) in rats and rabbits

Nyotran is a liposomally encapsulated i.v. formulation of the antifungal polyene nystatin. This drug was evaluated in a series of reproductive toxicity studies, according to the guidelines outlined by the International Conference on Harmonization (ICH). A fertility and early embryonic development study (SEG I) and a prenatal and postnatal development (SEG III) study were conducted in rats, and embryo-fetal development (SEG II) studies were conducted in rats and rabbits. Nyotran was administered iv in all studies. In SEG I and SEG III, rats were administered daily doses of 0.5, 1.5, or 3.0 mg/kg Nyotran. In both studies, parental mortality and toxicity in the 3.0 mg/kg dose group necessitated the lowering of the high dose to 2.0 mg/kg/day. Parental toxicity, in the form of decreased body weights, decreased food consumption, and piloerection were also observed at the 1.5 mg/kg/day dose level in the SEG I and SEG III studies. Despite the parentally toxic doses in the SEG I study, there was no effect of Nyotran on F0 male or female fertility or early embryonic development of F1 offspring. In the SEG III study, lactational body weights of the F1 generation were decreased at all Nyotran dose levels. There was no effect on pre-wean developmental landmarks, but post-wean development was affected by Nyotran administration at all dosage levels. Preputional separation was delayed in the 1.5 and 3.0/2.0 mg/kg/day F1 offspring, auditory startle function was decreased in F1 females at all dose levels, and motor activity was decreased in male F1 offspring at all dose levels. However, there were no treatment-related effects on the subsequent mating of the F1 generation and resulting F2 offspring. In SEG II studies, rats and rabbits were also administered 0.5, 1.5, or 3.0 mg/kg/day of Nyotran during gestation. The high dose in these SEG II studies was not lowered, as the maternal animals were able to tolerate the shorter duration of dosing. Maternal effects in rabbits were observed only in the high-dose group and were limited to decreased food consumption and decreased absolute and relative liver weight. Decreased food consumption in high-dose dams and clinical weight loss in some animals at the mid- and high-dose levels evidenced maternal toxicity in rats. Nyotran did not have any effect on Caesarian section parameters in either rats or rabbits and no effect on the incidence of fetal malformations in rabbits. A statistically significant increase in mild hydrocephaly, observed in 4 rat fetuses, was seen at the highest dose level of 3.0 mg/kg/day. The biological significance and relationship to Nyotran treatment of this finding is not clear. This finding may represent a change in the background incidence or a change in the pattern of responsiveness of this strain of rat fetus to the test chemical. Toxicokinetic data were also collected in the SEG II rabbit and rat studies for comparison to human exposures. In both species, systemic exposure to the nystatin at effective antifungal concentrations was demonstrated. The systemic exposures in rats and rabbits were, however, considerably less than have been reported in humans administered clinical doses of 2 or 4 mg/kg/day Nyotran. Thus, humans tolerate higher dosages and systemic exposures of Nyotran relative to rats and rabbits and there is no margin of safety in either dosage level or systemic exposure to drug. Given this lack of a margin of safety and the effects on postnatal development in F1 rats, caution should be exercised when using this drug in females of childbearing potential.

Safety and efficacy of multilamellar liposomal nystatin against disseminated candidiasis in persistently neutropenic rabbits

The activity of liposomal nystatin (L-Nys) against subacute disseminated candidiasis was investigated in persistently neutropenic rabbits. Antifungal therapy was administered for 10 days starting 24 h after intravenous inoculation of 10(3) blastoconidia of Candida albicans. Responses to treatment were assessed by the quantitative clearance of the organism from blood and tissues. Treatments consisted of L-Nys at dosages of 2 and 4 mg/kg of body weight/day (L-Nys2 and L-Nys4, respectively) amphotericin B deoxycholate at 1 mg/kg/day (D-AmB), and fluconazole at 10 mg/kg/day (Flu). All treatments were given intravenously once daily. Compared to the results for untreated but infected control animals, treatment with L-Nys2, L-Nys4, D-AmB, and Flu resulted in a significant clearance of the residual burden of C. albicans from the kidney, liver, spleen, lung, and brain (P < 0.0001 by analysis of variance). When the proportion of animals infected at at least one of the five tissue sites studied was evaluated, a dose-dependent response to treatment with L-Nys was found (P < 0.05). Compared to D-AmB-treated rabbits, mean serum creatinine and blood urea nitrogen levels at the end of therapy were significantly lower in animals treated with L-Nys2 (P < 0.001) and L-Nys4 (P < 0.001 and P < 0.01, respectively). L-Nys was less nephrotoxic than conventional amphotericin B and had dose-dependent activity comparable to that of amphotericin B for the early treatment of subacute disseminated candidiasis in persistently neutropenic rabbits.

Liposomal nystatin against experimental pulmonary aspergillosis in persistently neutropenic rabbits: efficacy, safety and non-compartmental pharmacokinetics

The activity of liposomal nystatin against invasive pulmonary aspergillosis was investigated in persistently neutropenic rabbits. Treatment groups included liposomal nystatin at dosages of 1, 2 and 4 mg/kg/day intravenously, or amphotericin B deoxycholate 1 mg/kg/day administered intravenously after normal saline loading. As compared with untreated controls, liposomal nystatin administered at 2 and 4 mg/kg/day prolonged survival and reduced fungus-mediated tissue injury and excess lung weight at post-mortem in a similar manner to amphotericin B. Although amphotericin B was superior in clearing infected lung tissue, treatment with all regimens of liposomal nystatin led to a significant reduction in pulmonary fungal tissue burden. During treatment, ultrafast CT-scan demonstrated ongoing resolution of pulmonary lesions at 2 and 4 mg/kg/day, but not at 1 mg/kg/day. With the exception of mild increases in blood urea nitrogen (BUN) and serum creatinine values during treatment at 2 and 4 mg/kg/day, which were similar to those found in amphotericin B-treated rabbits, liposomal nystatin was well tolerated. Preliminary pharmacokinetic studies in non-infected animals established linear drug disposition of liposomal nystatin in plasma over the investigated dosage range and peak plasma levels above the MIC for the test strain after multiple daily dosing for 7 days. Liposomal nystatin increased survival and provided reduced tissue injury, effective microbiological clearance and tolerable side effects in experimental pulmonary aspergillosis in persistently neutropenic rabbits, thus providing a rational basis for further investigations in clinical trials.

Noncytotoxic combinations of topical antimicrobial agents for use with cultured skin substitutes

Cultured skin grafts are destroyed more easily than split-thickness skin grafts by common burn wound organisms, including gram-negative and gram-positive bacteria and fungi. To increase the survival and engraftment of cultured skin grafts, formulations of antimicrobial agents were tested for cytotoxicity to cultured human keratinocytes and fibroblasts and for activity against common organisms from burn wounds. On the basis of previous studies, a base formulation containing neomycin (40 micrograms/ml), polymyxin B (700 U/ml), and mupirocin (40 micrograms/ml) was prepared, to which ciprofloxacin (20 micrograms/ml) or norfloxacin (20 micrograms/ml) and amphotericin B (0.25 microgram/ml) or nystatin (100 U/ml) were added. Toxicity to cultured human cells was determined by the growth response of cell cultures (n = 6) to each drug combination over 4 days. Activity against clinical isolates (n = 40) of Staphylococcus aureus, Pseudomonas aeruginosa, other gram-negative bacteria, and Candida spp. was determined by the wet disc assay. Analysis of variance testing showed no significant differences in the growth of keratinocytes or fibroblasts under control or experimental conditions. Medium without antimicrobial agents was not effective against any of the 40 microbial strains tested. The base formulation was effective against all bacterial strains tested but against none of the fungi, while all experimental formulations were effective against all microbial strains tested. These findings suggest that neomycin, mupirocin, and polymyxin B may be combined with a quinolone and an antimycotic agent to provide broad antimicrobial activity for a formulation for topical use with cultured skin on burns. However, the formulations described here are strictly experimental and are not recommended for clinical use without further evaluation.

Toxicity and therapeutic effects in mice of liposome-encapsulated nystatin for systemic fungal infections

The therapeutic activity of nystatin (NYS) incorporated in multilamellar liposomes (L-NYS) was studied in vivo. Hale-Stoner mice injected intravenously with various doses of L-NYS and free NYS showed a significant reduction in toxicity of NYS after the NYS was incorporated into liposomes (maximal tolerated doses, 16 and 4 mg/kg of body weight, respectively). The maximal tolerated dose of free NYS had no effect in the treatment of mice infected with Candida albicans, whereas L-NYS at an equivalent dose improved the survival of mice. A marked increase in survival was observed when L-NYS was administered in higher and multiple doses (total doses up to 80 mg/kg). Liposome encapsulation thus provided a means for intravenous administration of NYS, reducing its toxicity and making it an active systemic antifungal agent.

[Prostaglandin synthetase activity in the different layers of the kidneys of young rats exposed to polyene antibiotics]

Prostaglandin (PG)-synthestase activity was studied in the cortical, medullary and papillary kidney layers in young rats subjected to prolonged administration of polyene antibiotics (amphotericin B, levorin and nystatin). This activity was markedly increased during the first few hours after the administration of amphotericin B. At later terms a pronounced decline in the enzyme activity was observed. The changes were most prominent in the medullary and papillary layers. The other two antibiotics were less potent. The experimental results have shown that amphotericin B had maximal effect on renal PG-synthetase activity, while the sodium salt of nystatin was least effective.

Membrane potential and cation permeability. A study with a nystatin-resistant mutant of Rhodotorula gracilis (Rhodosporidium toruloides)

Cells of a nystatin-resistant mutant of the obligatory aerobic yeast Rhodotorula gracilis displayed an electrical potential difference, delta psi, across the plasma membrane which was, in contrast to the wild-strain cells, virtually independent of the pH of cell suspensions down to 4.5. In addition, the delta psi in mutant cells was insensitive to extracellular K+ concentrations. The mutant cells failed to cotransport measurable amounts of H+ by the onset of monosaccharide transport and to take up K+ in exchange for H+. Taking into account the lower passive permeability of the mutant plasma membrane for cations, it has been concluded that the pH dependency of delta psi in wild-strain cells is correlated with the electrogenic leak of H+ back into the cells in course of increasing delta pH across the plasma membrane.

Passive fluxes of K+ and H+ in wild strain and nystatin-resistant mutant of Rhodotorula gracilis (ATCC 26194)

The passive fluxes of protons and potassium ions have been studied in the obligatory aerobic yeast Rhodotorula gracilis. The cellular energy metabolism was suspended by introducing anaerobic conditions. The H+-permeability of the plasma membrane was modified by adding an uncoupler under both aerobic and anaerobic conditions. Unfortunately, the plasma membrane of R. gracilis was insensitive to K+-ionophores. The passive flows of H+ and K+ under anaerobic and/or uncoupled conditions were electrically coupled and exhibited a constant stoicheiometry of 1:1. The H+ permeability of the plasmalemma was shown to determine the velocity of the passive K+-H+ exchange. The nystatin-resistant mutant M 67 displayed distinctly lower permeability for both H+ and K+, which can explain the observed differences in some transport characteristics of the two strains. In order to account for the properties of passive K+ flows, a membrane-potential-gated channel for K+ has been proposed. Evidence is presented that the inhibitor of the plasmalemma-bound H+-ATPase, N,N'-dicyclohexylcarbodiimide (DCCD), reduced at first the permeability for both K+ and H+ and only upon prolonged incubation the ATPase itself. Since DCCD effected an immediate hyperpolarization of the membrane potential, it has been concluded that the H+ does not slip through the H+-ATPase under deenergized conditions.

Increased toxicity of oral antibiotics in sodium-deficient rats

The toxicity of an oral antibiotic mixture used to decontaminate the gastrointestinal tract of experimental animals was compared in rats with a normal sodium intake to rats on a sodium-deficient diet. Sodium-depleted rats are quite sensitive to the oral antibiotic mixture. The antibiotic mixture was nephrotoxic, resulting in necrosis of the proximal tubules. Therefore, since the parenteral administration of antibiotics also produced necrosis of the proximal tubules, the mechanism of antibiotic toxicity in sodium-deficient rats is not influenced by the route of antibiotic administration.

[Changes in ATPase activity of dog kidney nuclear membrane under the effect of polyenic antibiotics]

Effect of amphotericin B and nistatin on ATPase activity of dog kidney nuclear membranes is studied in vivo and in vitro. Long-term intravenous injections of the antibiotics do not change the ATPase activity of kidney nuclear membranes. However, short-term injections of polyenic antibiotics have some effect on ATPase activity on nuclear membranes: amphotericin B, considerably activates the enzyme . In vitro incubation of isolated dog kidney nuclei with amphotericin and nistatin at concentrations of 1 and 10 mcg/ml does not affect the activity of nuclear membrane ATPase, while increased concentrations of polyenic antibiotics, (up to 200 mcg/ml) results in a slight inhibition of the enzyme activity. The role of the data obtained for solving molecular basis of the toxic effect of polyenic antibiotics.

Relationship between acute toxicity in mice and polymorphic forms of polyene antibiotics

Marked differences in LD50 values were found between various batches of polyene antibiotics (mepartricin and nystatin) after intraperitoneal administration in suspension to mice. Such differences do not seem attributable to the particle size but to different molecular spatial arrangements of the products, which modify their solubility and biological availability. In the samples obtained by chemical treatments of some batches of the polyene substances, it has been possible to change the toxicity drastically and then to bring it back to its original value.

[Study of the effect of nystatin and mycoheptin on the intrauterine development of the rat fetus]

The effect of nystatin and mycoheptyne on pregnancy and fetus of rats of Wister line was studied after single and multiple oral administrations of the drugs. Nystatin was used in single doses of 3000, 500 and 100 mg/kg. In chronic experiments the dose was 500 mg/kg. It was found that nystatin had a slight abortive effect when used during the whole period of pregnancy. Mycoheptyne was used in single doses of 1000, 100 and 20 mg/kg. In chronic experiments the dose was 100 mg/kg. It was noted that mycoheptyne had a higher toxic effect on the pregnant rats than nystatin. The animal death and abortus were observed more often. The level of the changes depended on the dose, time and period of the antibiotic use. No defects in the rat fetus were found.

Selective membrane toxicity of the polyene antibiotics: studies on lecithin membrane models (liposomes)

In the absence of sterol, amphotericin B at 5 x 10(-6) M caused maximum marker release from the saturated dipalmitoyl lecithin liposomes, minimum release from the unsaturated dioleoyl lecithin liposomes, and an in-between response from egg lecithin liposomes. Nystatin at 2.5 to 4.0 x 10(-5) M induced appreciable marker release from all three types of sterol-free liposomes. The amphotericin B- and nystatin-induced permeability changes in dipalmitoyl lecithin liposomes were drastically suppressed by the incorporation of cholesterol or stigmasterol (with identical Delta5 sterol nuclei), but were unaffected by the incorporation of ergosterol or 5,7-cholestadien-3beta-ol (with identical Delta5,7 sterol nuclei). The nystatin sensitivity of dioleoyl lecithin liposomes remained low after the incorporation of cholesterol or stigmasterol, but was greatly enhanced by the incorporation of ergosterol or 5,7-cholestadien-3beta-ol. Digitonin, a compound known to interact specifically with membrane sterol, induced marker release from liposomes in proportion to the amount of either cholesterol or ergosterol incorporated; epicholesterol did not sensitize to digitonin. These results lead to the following conclusions: (i) polyene-induced permeability alteration in model membrane systems is effected by the composition of membrane phospholipid fatty acyl chains; (ii) the distribution of double bonds in the sterol nucleus is related to the selective toxicity of the polyenes toward natural sterol-containing membranes; and (iii) polyenes differ in membrane selectivity.

Selective membrane toxicity of the polyene antibiotics: studies on natural membranes

The effect of polyene antibiotics on Candida albicans, human erythrocytes, and Acholeplasma laidlawii was studied. The results sustain the observations made with lecithin-sterol liposomes. The distribution of double bonds in the membrane sterol nucleus appears to be of major importance in conferring polyene susceptibility; those sterols with the ergosterol nucleus are far more effective than those with a nucleus similar to cholesterol. Different polyenes vary in their membrane selectivity. The clinical implications of these observations are discussed.