WO2007023370A1 - Compositions and methods for treating malaria - Google Patents

Abstract

Polyene macrolide compounds, analogues thereof and compositions containing polyene macrolides or analogues are described for use in treating malaria infections in humans and animals. Examples of polyene macrolide compounds are amphotericin B, nystatin A, natamycin (pimaricin), filipin, rimocidin, candidin and vacidin, and in particular, formulations similar to those described for amphotericin B, such as amphotericin B deoxycholate, amphotericin B colloidal dispersion, amphotericin B lipid complex and liposomal amphotericin B. The polyene macrolide may be produced by a microorganism, by chemically modifying a polyene macrolide or via de novo synthesis. The analogue may be chemically modified to improve solubility, bio-availability and/or bio-activity or limit toxicity. The polyene macrolide or analogue may be administered together with one or more other antimalarial compounds, such as chloroquine, quinine, quinidine, mefloquine, halofantrine, sulfonamides, tetracyclines, atovaquone, artemisinin compounds, rimaquine, proguanil or pyrimethamine, so as to increase the efficacy of the other antimalarial compound.

Description

COMPOSITIONS AND METHODS FOR TREATING MALARIA

BACKGROUND OF THE INVENTION

THIS invention relates to a method of treating malaria using polyene macrolide antibiotics and their derivatives/analogues, and polyene macrolides for use as anti-malarial (anti-Plasmodium) drugs, with particular emphasis on cerebral and severe malaria in humans.

Following the initial success of global control programs, malaria resurged as the most devastating tropical disease in the latter part of the 20th century. There are presently 300 - 500 million new malaria cases per year, of which 90% occur in sub-Saharan Africa, leading to 1.53 million fatalities/year (4-5% of all fatalities in the world). Nevertheless, malaria is one of the neglected diseases, with 50 million deaths due to malaria over the last 15 years. Reasons for this resurgence are varied, but a major contributing factor is the rapid emergence and spread of parasite resistance to front-line drugs, which has provoked an urgent search for novel or improved anti-malarial compounds (22).

The disease malaria is caused by protozoans of the genus Plasmodium, which are transmitted to humans by Anopheles mosquitoes as sporozoites that are the target of the drug primaquine. The sporozoites infect the liver and this stage of the parasite can be treated with tissue schizontocides, i.e. primaquine, proguanil and pyrimethamine. Following an asymptomatic period of growth and multiplication in liver cells, the parasites rupture into the circulation in the form of merozoites that rapidly invade erythrocytes. Immediately following invasion, the merozoite differentiates into a ring form that ingests erythrocyte cytoplasm and grows into a trophozoite. In the subsequent schizont stage, growth ceases, the nucleus undergoes multiple divisions and the plasma membrane invaginates to give rise to several daughter merozoites that escape from the infected host cell and invade fresh erythrocytes. This cycle is completed every 48 hours and is responsible for the pathology of malaria. The erythrocytic stages of the parasite are the most prominent drug targets. These drugs fall in the group of blood schizontocides and examples are chloroquine, quinine, quinidine, mefloquine, halofantrine, sulfonamides, tetracyclines, atovaquone and artemisinin compounds, but widespread resistance against these drugs exists (39,40).

There is thus a need for new compounds for treating malaria infections.

SUMMARY OF THE INVENTION

According to a first embodiment of the invention, there is provided a method of treating malaria infections in animals and/or humans, the method comprising administering a therapeutically effective amount of a polyene macrolide compound, or a chemically modified analogue thereof, to the animal or human.

The malaria infection may be a new or severe infection.

The polyene macrolide compound may be produced by a suitable microorganism, by chemically modifying a polyene macrolide compound or via de novo synthesis. Examples of polyene macrolide compounds are amphotericin B, nystatin A, natamycin (pimaricin), filipin, rimocidin, candidin and vacidin.

Suitable formulations of the polyene macrolide compound may be similar to formulations described for amphotericin B, which are amphotericin B deoxycholate, amphotericin B colloidal dispersion, amphotericin B lipid complex and liposomal amphotericin B, or a formulation to improve bio-availability and limit toxicity. The analogue may be chemically modified to improve solubility, bio-availability and/or bio-activity or limit toxicity. Modification methods may include oxidation, hydroxylation, acylation, amidation, coupling of an organic moiety, biosynthetic modification and hydroxyl, carbonyl, carboxyl, amino, methyl or sugar group substitution.

The polyene macrolide compound or analogue may be administered together with one or more other antimalarial compounds, such as chloroquine, quinine, quinidine, mefloquine, halofantrine, sulfonamides, tetracyclines, atovaquone, artemisinin compounds, rimaquine, proguanil or pyrimethamine, so as to increase the efficacy of the other antimalarial compound.

The polyene macrolide compound or analogue may form part of a pharmaceutical composition that is formulated to improve bio-availability and/or limit toxicity of the compound or analogue.

According to a second embodiment of the invention, there is provided a polyene macrolide compound or analogue thereof for use in treating malaria infections in an animal and/or human, and in particular, for treating cerebral or severe malaria in humans.

The polyene macrolide or analogue thereof may be substantially as described above.

According to a third embodiment of the invention, there is provided the use of a polyene macrolide compound or analogue thereof for use in a method of making a medicament for treating malaria infection in a human and/or animal.

In particular, the medicament may be for use in treating cerebral or severe malaria in humans.

The polyene macrolide or analogue thereof may be substantially as described above. BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 Growth inhibition dose-responses, obtained with trophozoite- infected erythrocytes (filled circles) and haemolytic dose-responses of normal erythrocytes (unfilled circles) as measured after 48 hours.

Haemolysis was determined by measuring the released haemoglobin in the supernatant at 412 nm and growth inhibition by a lactase dehydrogenase assay. Cells were treated with serial dilutions of filipin (A), natamycin (B), saponin (C) nystatin (D), amphotericin B (E) or liposomal amphotericin B. The average of at least 3 determinations of each data point ± standard error of the mean (SEM) is shown. Between 30 and 80 data points were used to generate the sigmoidal dose-response curves from which the HC₅₀ and IC₅₀ values were calculated.

Fig. 2 Giemsa-stained blood smears of treated cultures. Parasite cultures were left untreated (A, C, E), or incubated for 2 hours with 50 μM nystatin (B, F) or 1.3 μM amphotericin B (D, G). A-D; trophozoite- infected cultures. E-G; ring-infected cultures.

Fig. 3 Phase-contrast microscopy of trophozoite-stage cultures. Cultures were incubated with 50 μM nystatin (A) or 1.3 μM amphotericin B (B) and the percentage of parasites found inside intact erythrocytes (white bars), surrounded by a red blood cell ghost membrane (grey bars), or free of surrounding host membranes (black bars), was determined at various time-points. Representative phase-contrast microscopy images of the three conditions are shown on the right.

Fig. 4 Fluorescence microscopy of drug-treated cultures incubated with trypan blue. Trophozoite-infected cultures were left untreated (A), or incubated with 50 μM nystatin (B), 1.3 μM amphotericin B (C) or 8 μM gramicidin S (D), resuspended in PBS containing trypan blue and viewed by fluorescence microscopy. In A: arrowhead indicates trypan blue staining of the red blood cell membrane; small arrow, parasite-derived organelles in red blood cell cytoplasm; larger arrow, parasite plasma membrane. B: arrowhead indicates red blood cell ghost membrane; arrow, parasite plasma membrane. D: arrowhead indicates staining of the erythrocyte ghost membrane; larger arrow, accumulation of trypan blue around the hemozoin crystal; smaller arrow, staining of the nuclear membrane. Ring- infected cultures were also left untreated (F), or incubated for 2 hours with amphotericin B (E, H) or nystatin (G). Following incubation, erythrocytes were either resuspended in trypan blue and viewed directly (E), or resuspended in trypan blue following aldehyde fixation and rinsing in saponin (F-H). In E, the red blood cell membrane is strongly stained with trypan blue, but the dye has failed to penetrate the erythrocyte and stain the intracellular ring- stage parasite (denoted by the arrow in the phase-contrast image). In F, the arrow indicates control, untreated ring-stage parasites surrounded by red blood cell ghost membranes. In H, the arrow indicates the nuclear remnant of a pyknotic, amphotericin B-treated ring. In each case, the left-hand panel presents the fluorescence microscopy image and the right-hand panel the corresponding phase-contrast microscopy image.

DETAILED DESCRIPTION OF THE INVENTION

Polyene macrolide compounds, analogues thereof and compositions containing polyene macrolides or analogues are described for use in treating malaria infections in humans and animals, and in particular, severe or cerebral malaria infections in humans.

Initial symptoms of malaria are varied and are often described as "flu-like", but among the most life-threatening consequences of an infection by the most prevalent species, Plasmodium falciparum, is when capillaries and postcapillary venules are obstructed and local tissue anoxia results. Infected erythrocytes have an altered surface composition, including the presence of parasite proteins, which results in their aggregation and binding to non-infected cells to yield red blood cell clumps or rosettes (7). Binding of the rosettes to endothelial cells in brain capillaries could reduce perfusion in critical areas and contribute to the pathogenesis of cerebral malaria, leading to convulsions, coma and death of the infected individual (1,33). Rosette-induced obstructions can lead to acute tubular necrosis and renal failure in the kidneys, as well as ischemia and ulceration in the intestines, which in turn can lead to gastrointestinal bleeding and systemic bacteremia (41). Intravenous infusion of quinine is used to treat such severe malaria cases, but is often ineffective due to the prevalence of quinine resistance (42).

During the parasite's residence inside the red blood cell it extensively customises the host cell to meet its requirements. Several parasite proteins associate with the erythrocyte membrane and contribute to rosetting (32). To facilitate transport of the proteins to the membrane, the parasite constructs novel membrane-bound organelles in the erythrocyte cytoplasm (29) and the proteins may be delivered to the erythrocyte plasma membrane by vesicular trafficking (20,36). Vesicular fusion could conceivably alter the lipid composition and properties of the latter membrane. In addition to the secretory traffic, detergent resistant erythrocyte membrane lipid rafts and associated proteins are imported to the parasite, further modifying the host membrane (12). To acquire vital nutrients, the permeability of the erythrocyte membrane to low molecular weight compounds is increased (30), while neoantigens are created on the erythrocyte surface by the modification of native integral erythrocyte membrane proteins during parasite infection (33). The result of these activities is a marked alteration of the membrane protein and lipid composition and distribution in erythrocytes infected with mature malaria parasites (9, 14, 23, 24, 27, 37).

The parasite-induced alteration of the infected red blood cell membrane raises the possibility that the latter may be vulnerable to selective recognition and perturbation by membrane-active drugs and compounds. The search for new drugs led the inventors to investigate membrane active amphipathic compounds with known antifungal activity from the group of polyene macrolides. The interaction of these compounds with bilayer cell membranes increases the membrane permeability, which can lead to cell damage and ultimately cell lysis. Due to this mechanism of action it is extremely difficult for pathogens to develop resistance to these drugs. The polyene macrolides are considered membranolytic due to their lipid, and in particular sterol, binding activity. For example, filipin is widely used in cell biological studies of cholesterol dynamics due to its affinity for the sterol and its inherent fluorescent properties (4). Natamycin (pimaricin) is used as a food preservative and in the treatment of fungal keratitis due to its ability to bind sterols and disrupt fungal membranes (19, 28, 34). Amphotericin B (AmpB) and nystatin are commonly used for the treatment of topical and systemic (in the case of the former) fungal infections (10) and are thought to permeabilise target membranes by binding to β-ergosterol, the principal fungal sterol (3, 18, 34, 38).

Amphotericin B was first isolated from Streptomyces nodosus in 1955. It is an amphoteric compound composed of a hydrophilic polyhydroxyl chain along one side and a lipophilic polyene hydrocarbon chain on the other (see structure below). Amphotericin B is available in four formulations. The classic amphotericin B deoxycholate (Fungizone™) formulation has been available since 1960 and is a colloidal suspension of amphotericin B. A bile salt, deoxycholate, is used as the solubilizing agent.

amphotericin B

Other examples of polyene macrolides are:

Rimocidin (C₃₉H_6INO₁₄):

Candidin (C₄₇H_2INO₁₇):

Toxicity problems encountered with AmpB and nystatin are attributed to the ability of the polyene macrolides to also bind cholesterol in human cell membranes, albeit with a lower affinity, which has led to the development of liposomal formulations of the compounds for intravenous use (8, 26, 35). The licensed lipid formulations are Amphotericin B Colloidal Dispersion (ABCD; Amphocil™ or Amphotec™), Amphotericin B Lipid Complex (ABLC; Abelcet™) and Liposomal Amphotericin B (L-AMB; Ambisome™).

The polyene macrolide can be chemically modified to form an analogue with improved solubility, bio-availability, bio-activity and/or limited toxicity. Typical modification methods include oxidation, hydroxylation, acylation, amidation, coupling of an organic moiety, biosynthetic modification and hydroxyl, carbonyl, carboxyl, amino, methyl or sugar group substitution.

The polyene macrolide compound or analogue can be administered on its own or together with one or more other antimalarial compounds, such as chloroquine, quinine, quinidine, mefloquine, halofantrine, sulfonamides, tetracyclines, atovaquone, artemisinin compounds, rimaquine, proguanil or pyrimethamine, so as to increase the efficacy of the other antimalarial compound(s) or to obtain a synergistic effect.

Materials and methods

Drugs tested

Filipin complex from Streptomyces filipinesis, nystatin dihydrate, saponin and natamycin (pimaricin) were obtained from Sigma-AIdrich, an amphotericin B- deoxycholate preparation (Fungizone™) from Bristol-Myers Squibb and liposomal amphotericin B (AmBisome™) from Key Oncologics. Filipin and nystatin stock solutions were prepared in dimethyl sulfoxide (DMSO; 25 mg/ml and 10 mM, respectively) and stored frozen. Saponin was stored at 4⁰C as a 50 mg/ml solution in water. Amphotericin B-deoxycholate was dissolved in water to 0.5 mM according to manufacturers specifications and stored frozen. Liposomal amphotericin B was also reconstituted to 0.5 mM in water, according to manufacturers specifications immediately before use. Natamycin was dissolved in a minimum amount of acetic acid, diluted in culture medium and neutralised with NaOH to yield a 3 mM stock solution shortly before use.

Parasite culture

The D10 strain of P. falciparum was cultured in RPMI-1640 medium supplemented with 50 mM glucose, 0.65 mM hypoxanthine, 25 mM Hepes, 0.2% (w/v) NaHCO₃, 0.048 mg/ml gentamicin, 0.5% (w/v) Albumax II, and 2-4% (v/v) human O⁺ erythrocytes, under an atmosphere of 3% CO₂, 4% O₂, balance N₂.

Parasite viability and haemolysis dose-response assays

Culture-derived parasitised erythrocytes were mixed with fresh culture medium and erythrocytes to yield a 2% parasitemia, 2% haematocrit suspension and distributed in microtitre plates at 90 μl/well. Serial dilutions of test drug in culture medium was prepared in quadruplicate wells in a separate plate and transferred to the parasite plate to yield a final volume of 100 μl/well. The plates were incubated at 37⁰C for 48 hours and parasite viability in each well measured by the colorimetric determination of lactate dehydrogenase activity (25).

To determine haemolytic activity of the drugs, parallel plates were prepared containing 90 μl/well of a 2% haematocrit suspension of uninfected erythrocytes. After the 48-hour incubation, intact erythrocytes were sedimented in the microtitre plate wells by centrifugation at 20Og for 3 minutes in a swing-out rotor. Aliquots of the supernatants were removed and diluted 1 :8 in water in separate microtitre plates. Haemoglobin content in the supernatant dilutions was determined by absorbance at 405 nm in a microtitre plate spectrophotometer. Absorbance readings were converted to percentage parasite viability (in the case of the lactate dehydrogenase assays) or percentage haemolysis.

Data Processing All data was analysed using Graphpad Prism version 3.01 for Windows, GraphPad Software, San Diego California USA (www.graphpad.com). Non-linear regression was performed on the dose-response data obtained from the lactate dehydrogenase and haemolysis assays. A sigmoidal curve with variable slope and constant difference of 100±5 between the top bottom was fitted to each of the data sets using the following equation:

For curve fitting only the mean value of each data point, without weighting, was considered. The 50% inhibitory concentration (IC₅₀) was calculated from the x- value of the response halfway between top and bottom plateau. Experiments, with three to eight determinations at each concentration, were performed for each compound against each target cell. Between 30 and 80 data points were considered in each dose-response curve to calculate the IC₅₀ (concentration that leads to 50% growth inhibition), and HC₅₀ (concentration that leads to 50% erythrocyte lysis).

ELISA-based haemolysis assay

Parasite-infected erythrocytes were separated from uninfected erythrocytes by centrifugation through a step-wise Percoll gradient containing 3% alanine (11). Serial dilutions of test drugs were added to suspensions of the enriched pRBC or uninfected erythrocytes in Albumax-free culture medium in microtitre plates (0.2% final hematocrit, 50 μl final volume/well). After a 40-minute incubation at 37⁰C, intact erythrocytes were sedimented by centrifugation at 20Og for 3 minutes in a swing-out rotor. Aliquots of the supernatants were diluted 50 times in PBS, transferred to ELISA plates (50 μl/well) and incubated for 20 minutes at room temperature. The plates were washed in phosphate-buffered saline (PBS)₁ blocked in PBS containing 1% (w/v) bovine serum albumin and 0.1% (v/v) Tween 20 (blocking buffer) for 20 minutes and incubated for an additional 20 minutes with rabbit anti-human haemoglobin antiserum (Sigma-Aldrich) diluted 1 :2000 in blocking buffer. Following four washes in PBS-0.1% (v/v) Tween 20 (washing buffer), plates were incubated with peroxidase-conjugated goat anti-rabbit antiserum (KPL Laboratories) diluted 1:2000 in blocking buffer. Following washes in washing buffer, bound peroxidase activity was determined colourimetrically by adding 0.1 M phosphate-citrate buffer (pH 4.8) containing 1 mg/ml o-phenylenediamine and 0.015% H₂O₂. Colour development was terminated by adding 2.5N sulphuric acid and quantified by absorbance at 450 nm using a microtitre plate reader. Duplicate experiments with each measurement in triplicate were performed to determine the HC₅₀-values from dose-response curves (analysed as described above). Selectivity was calculated from the ratio between the minimum concentration of each compound causing 100±5% lysis (HC₁₀0) of infected and uninfected cells (Selectivity = HC₁₀₀(normal CeIIs)/HC₁₀₀(infected cells)).

Light microscopy

Drugs were added to parasite cultures (2% haematocrit) and incubated at 37°C. At various time-points, blood smears were prepared on microscope slides, stained with Giemsa solution and viewed by light microscopy. Alternatively, aliquots were removed from the cultures, directly mounted on microscope slides under glass cover slips and examined by phase-contrast light microscopy. For fluorescence microscopy analysis, aliquots removed from the drug-treated cultures were mixed with a solution of trypan blue and 4',6-diamidino-2- phenylindole (DAPI) in PBS (final concentrations 0.5% and 1 μg/ml, respectively), mounted on microscope slides under cover slips and viewed by epifluorescence illumination using tetramethyl-rhodamine and ultraviolet filters. Alternatively, erythrocytes from treated cultures were centrifuged through 60% Percoll in RPMI medium for 10 minutes to remove lysed cells and debris, immobilized on poly- lysine coated glass cover slips, rinsed in 0.5 mg/ml saponin in PBS and fixed for 10 minutes in PBS containing 3% para-formaldehyde and 0.2% glutaraldehyde. Cover slips were subsequently incubated for 5 minutes in PBS containing DAPI and trypan blue, inverted on microscope slides and viewed by fluorescence microscopy. All microscopy assays were performed on a Nikon Eclipse E600 fluorescence microscope fitted with a 10Ox Apochromat objective and images were captured with a Media Cybernetics CoolSNAP-Pro monochrome cooled CCD camera.

Results

Determination of parasite-inhibitory and haemolvtic concentrations The anti-malarial activity of formulations of the polyene macrolides filipin (complex from S. filipinensis), natamycin (pimaricin), nystatin (nystatin dihydrate), amphotericin B (amphotericin B-deoxycholate, Fungizone™) and liposomal amphotericin B (AmBisome™) was determined by 48-hour incubation of parasite cultures treated with serial dilutions of the compounds. Parasite viability was assessed by a colorimetric assay for parasite lactate dehydrogenase activity. Given that the parasites invade and grow in erythrocytes and that the polyene macrolides are haemolytic at high concentrations, haemolytic activity of the compounds was tested in parallel by spectrophotometrically measuring haemoglobin release from uninfected erythrocytes incubated under identical conditions. Since polyene macrolides disrupt membranes by interaction with sterols, the cholesterol-binding and haemolytic agent saponin was also tested. As shown in Figures 1A-C, the concentrations at which 50% parasite inhibition (IC₅₀) and haemolysis (HC₅₀) were achieved were similar in the cases of filipin and saponin, and slightly less so for natamycin, indicating that these compounds inhibit parasite viability by a general lysis of erythrocytes. In contrast, the IC₅₀S obtained with nystatin and especially AmpB were markedly lower than their HC₅₀S, with >10 and >2000 fold higher concentration of nystatin and AmpB respectively required for lysis of normal erythrocytes (Figures 1 D, E). Similarly, liposomal amphotericin B yielded a markedly increased anti-parasitic activity compared to erythrocyte lysis (Figure 1F). The IC₅₀ obtained with liposomal amphotericin B was 5.4 μM, while a maximum red blood cell lysis of only 15 % was found over the concentration range tested (0.5-511 μM). The observed differences between haemolysis and growth inhibition for nystatin and the AmpB preparations indicate a parasite-inhibitory mechanism different from non-specific red blood cell lysis. Evaluation of treated cultures by Giemsa-stained blood smears In order to gauge the effect of nystatin, amphotericin B and liposomal amphotericin B on cultured parasites and assess their modes of action, parasite cultures were incubated with the drugs at non-haemolytic concentrations, 3-9 fold above compound IC₅₀S, and Giemsa-stained blood smears were prepared at various time-points for evaluation by light microscopy. With trophozoite-infected cultures, 25-minute incubations with nystatin and amphotericin B caused most of the parasites to appear free of their host erythrocytes, but otherwise morphologically indistinguishable from untreated trophozoites (Figures 2A-D). Consequently, parasitemia decreased significantly, with 50 μM nystatin leading to a 92% and 1.3 μM amphotericin B to a 97% reduction within 60 minutes (Figure 2, see graph insert). Treatment with 51 μM liposomal amphotericin B had a similar effect, however the appearance of significant numbers of red blood cell- free trophozoites required much longer incubation times. Thirteen percent of the trophozoites in the culture appeared to be released from their host erythrocytes after 2 hours and 35 % after 9 hours, resulting in a concomitant parasitemia decrease from 3.4% (at time 0) to 2.3% (after 9 hours; not shown). As with nystatin and amphotericin B, the extra cellular parasites appeared morphologically intact. In addition, the trophozoites still contained in erythrocytes developed normally in the presence of the liposomal amphotericin B, entering the schizont stage and completing merozoite formation and subsequent invasion of erythrocytes to yield ring-stage parasites after 21 hours, at 3.3% parasitemia, compared to 8.5% for untreated cultures.

The same concentrations of nystatin, amphotericin B and liposomal amphotericin B were also used to assess the effect of the compounds on ring-stage cultures by light microscopy of Giemsa-stained blood smears. In the case of both nystatin and amphotericin B, no apparent red blood cell-free parasites were visible over a 2-hour treatment time (Figures 2E-G). However, a more gradual reduction in parasitemia was observed with a 60% reduction for nystatin, and 63% for amphotericin B after 60 minutes (Figure 2, see graph insert). In addition, although the remaining rings were present in seemingly intact erythrocytes, their morphology had deteriorated (Figures 2E-G). Following nystatin treatment, rings appeared more insubstantial and often had unusual structures (Figure 2F), while AmpB-treated rings were pyknotic (Figure 2G). Liposomal amphotericin B, by contrast, did not alter ring morphology or decrease parasitemia (results not shown) with parasitemia remaining between 1.6 and 1.7% over the first 9 hours. However, the rings failed to develop into trophozoites and after 21 hours, no parasites were visible in the culture, while untreated cultures contained normally developed trophozoites.

Evaluation of treated cultures by phase-contrast light microscopy The results obtained with the Giemsa-stained smears suggested that nystatin and amphotericin B exert their anti-parasitic activity at sub-haemolytic concentrations by specifically lysing erythrocytes infected with trophozoites. However, cell membranes are not detectable in stained smears and the process of preparing the thin blood smears, fixing in methanol and Giemsa staining may disrupt fragile cells. We therefore assessed the effect of nystatin and amphotericin B on trophozoite-infected cells at various time-points by directly mounting treated cultures under cover slips and immediately viewing the cells by phase-contrast microscopy. Treated cultures contained a mixture of trophozoites inside intact erythrocytes, trophozoites in lysed erythrocytes but surrounded by a erythrocyte ghost membrane and "free" trophozoites with no discemable surrounding host membrane (Figure 3). During a 40-minute incubation with 50 μM nystatin, there was a gradual decrease in the percentage of parasites contained in intact erythrocytes, with a concomitant increase in parasites in erythrocyte ghosts and free parasites (Figure 3). After 40 minutes, 67% of the parasites were located in lysed red blood cell ghosts and 25% were free of visible erythrocyte membranes. The same trend was observed by incubating parasite cultures in 1.3 μM amphotericin B, except that erythrocyte lysis was more gradual (Figure 3). Uninfected erythrocytes appeared normal at all time-points.

Determination of haemolvtic activity by ELISA

The results obtained with the light microscopy studies suggested that nystatin and amphotericin B have the ability to selectively lyse trophozoite-infected erythrocytes. To allow the direct comparison of the dose-dependent haemolysis of trophozoite-infected vs. uninfected erythrocytes we developed a sensitive ELISA-based haemolysis assay. Uninfected erythrocytes and trophozoite- infected erythrocytes isolated from parasite cultures by density centrifugation were incubated in serial dilutions of the test compounds for 40 minutes and haemoglobin released into the supernatant was detected with anti-haemoglobin antiserum (Table 1). The results showed the same trend as that found previously by comparing haemolytic activity (assessed by spectrophotometric detection of released haemoglobin) and parasite-inhibitory activity (compare Figure 1 and Table 1). Filipin, natamycin and saponin were virtually unselective and did not distinguish between uninfected and parasitised erythrocytes (Table 1). Nystatin again showed an intermediate haemolytic selectivity (5-fold), while amphotericin B lysed parasitised erythrocytes at sub-micromolar concentrations 65-fold lower than the concentrations required for 100% haemolysis of normal erythrocytes (Table 1). The pronounced selectivity of nystatin and especially AmpB indicated that the selective lysis of trophozoite-infected erythrocytes contributes significantly to their anti-parasitic mechanism of action. In the case of liposomal amphotericin B₁ no detectable haemolysis was obtained over the concentration range used during the 40-minute incubation with both uninfected and parasitised erythrocytes (results not shown). This agrees with the results obtained by light microscopic evaluation of liposomal AmpB-treated cultures by Giemsa-stained blood smears in which a low percentage of red blood cell-free trophozoites were observed only after incubations of two hours or longer.

Table 1. Summary of the dose-dependent haemolysis of normal vs. trophozoite- infected erythrocytes, as determined by ELISA. The concentrations given ,are the minimum concentration for each compound needed to cause 100+5% haemolysis (HC₁₀o) and those in brackets are the HC₅₀ values as calculated from the dose-response curves. Selectivity is defined as HCioo(normal cells)/HC₁₀₀(infected cells)

Evaluation of parasite morphology by trypan blue fluorescence microscopy As described above, the Giemsa-stained blood smears of parasite cultures treated with nystatin and AmpB suggested that these compounds lyse the plasma membranes of trophozoite-infected erythrocytes, but leave the parasites relatively intact (Figures 2B, D). To evaluate the membrane integrity of trophozoites, following nystatin and AmpB treatment, a trypan blue exclusion assay was performed. Parasite cultures were treated with 50 μM nystatin and 1.3 μM AmpB for 40 minutes, suspended in a trypan blue solution and viewed by fluorescence microscopy. Using TRITC excitation filters, trypan blue yields a bright red fluorescence. In control cultures (Figure 4A), trypan blue was bound to the erythrocyte plasma membrane (arrowhead). In addition, the increased permeability of erythrocytes infected with mature stage parasites to diverse low molecular weight compounds (17) resulted in trypan blue leaking into the red blood cell. Here it also bound to parasite-derived membranous organelles in the erythrocyte cytoplasm (Figure 4A, small arrow) and to the parasite plasma membrane (Figure 4A, large arrow). No conspicuous trypan blue staining of intracellular parasite structures was found. In nystatin-treated cultures, parasites were surrounded by a iysed red blood cell ghost membrane (Figure 4B, right panel), which also bound trypan blue (Figure 4B, arrowhead). Although the parasite plasma membrane was brightly stained with trypan blue (Figure 4B, arrow), no trypan blue fluorescence was visible inside the parasite, suggesting that the impermeability of the parasite plasma membrane to certain small molecules was retained. Similar results were obtained with AmpB-treated cultures (Figure 4C). As a positive control for membrane disruption, cultures were treated with 8 μM gramicidin S (Figure 4D). In addition to staining the red blood cell ghost membrane (arrowhead), trypan blue was also found inside the parasite cytoplasm where it showed discernable labelling of internal parasite structures, including the nuclear membrane (small arrow) and around the haemozoin crystal (large arrow).

Trypan blue exclusion was also used to determine the effect of nystatin and

AmpB on membrane permeability in ring-stage cultures. In contrast to trophozoite-infected cells, ring-infected erythrocyte membranes appear to largely retain their native impermeability to small molecules. Consequently, trypan blue brightly stained the red blood cell membranes in control cultures, but failed to penetrate this barrier and stain the parasite membrane or parasite-derived structures (not shown). Similar results were obtained with ring-stage cultures treated for 90 minutes with nystatin or AmpB (the latter is shown in Figure 4E), suggesting that these compounds do not significantly disrupt ring-infected erythrocyte permeability.

Giemsa-stained blood smears suggested that nystatin and AmpB alter ring shapes following 1-2 hours of incubation (Figures 1 F₁G). Since stained smears inaccurately reflect ring structure, a modified trypan blue fluorescence microscopy assay was used to further assess the effect of these compounds on ring morphology following a 90-minute incubation. Erythrocytes from treated cultures were centrifuged through Percoll to remove lysed cells and debris, immobilized on glass cover slips, briefly rinsed in saponin to lyse erythrocyte membranes and remove excess haemoglobin, fixed in paraformaldehyde/glutaraldehyde, immersed in trypan blue and viewed by fluorescence microscopy. In contrast to the well defined, rounded rings in control cells (Fig. 4F₁ arrow), fixed and permeabilised erythrocytes from nystatin-treated cultures contained irregularly shaped parasites considerably reduced in size (Figure 4G). AmpB-treated rings were reduced to the nucleus (Figure 4H, arrow; discemable by DAPI-staining not shown in these images) and a few membrane remnants.

Discussion

In this study we have investigated the possibility that the extensive modifications wrought on erythrocytes by infecting malaria parasites renders the host cell membrane vulnerable to specific lysis by membrane-active agents. We have focused on several polyene macrolide formulations, since these compounds have membrane permeabilising activity, are widely used in cell biological studies, are clinically important in treating topical and systemic fungal infections and have been reported to have plasmodicidal properties.

Using a sensitive ELISA haemolysis assay we found that amphotericin B and, to a lesser extent, nystatin, permeabilise trophozoite-infected erythrocytes at concentrations significantly below those required for normal erythrocytes. The selective lysis of infected cells by these two preparations was further confirmed by Giemsa-stained blood smears and direct phase-contrast light microscopy. Our results therefore demonstrated clearly that selective lysis of trophozoite- infected erythrocytes contributes significantly to the plasmodicidal activity of the nystatin and amphotericin B preparations. The molecular basis for the lytic selectivity of these two compounds, however, is uncertain. A simple possibility might be that alterations or damage caused by infecting parasites renders erythrocyte membranes more fragile and innately labile when challenged with membrane-active agents. However, an argument against this explanation is the finding that virtually identical concentrations were required to lyse both uninfected and trophozoite-infected erythrocytes in the case of filipin, saponin and natamycin. Moreover, the 50% parasite-inhibitory and haemolytic concentrations were similar for each of these compounds. It is envisaged, however, that modifications to the structures of filipin, saponin and natamycin will allow these compounds to better distinguish between normal and parasite-infected erythrocytes.

Polyene macrolides have a known affinity for cholesterol and other sterols and binding to β-ergosterol in fungal membranes mediates the activity of nystatin and AmpB (3, 18, 19, 28, 34, 38). However, the cholesterol:phospholipid ratio in infected erythrocytes is in fact reduced by up to 55% in infected red blood cell membranes (24). In addition, the inability of the cholesterol-specific compounds saponin and filipin to distinguish haemolytically between infected and uninfected erythrocytes suggests that cholesterol binding per se may not explain the selectivity of nystatin and AmpB. However, AmpB has been reported to have an enhanced affinity for oxidized forms of cholesterol (6). Conceivably the presence of oxidised cholesterol moieties in the infected erythrocyte membrane, caused by parasite-induced oxidation (2, 15), could allow amphotericin B and possibly nystatin to discriminate between infected and normal cells. In addition, the cholesterol-rich lipid rafts found in erythrocyte membranes appear to be disrupted during parasite infection, likely due to the reduction in cholesterol and sphingomyelin levels (27, 31), which could result in an altered distribution and organization of cholesterol in the lipid bilayer and enhance nystatin or AmpB- mediated permeabilisation. The latter may be further modulated by the altered phospholipid bilayer distribution and composition of the infected erythrocyte membrane (23). The possibility remains, however, that a non-sterol parasite- derived lipid/protein modification of the erythrocyte membrane forms the basis of AmpB and nystatin selectivity. Noteworthy in this regard is the fact that the trophozoites themselves seem not to be significantly compromised by the AmpB and nystatin concentrations that lyse infected host cells, correlating with the reported low cholesterol levels in parasite membranes (16, 37). Following treatment, the parasites appear morphologically intact in Giemsa-stained smears and by phase-contrast microscopy and remain impermeable to trypan blue. This suggests that the molecular target of the polyene macrolides in the infected erythrocyte membrane is not appreciably present in trophozoite plasma membranes.

Our results further indicate a second mode of action that contributes, during the ring stage of the parasite life cycle, to the plasmodicidal activity of the nystatin and amphotericin B preparations. Ring morphology degenerated significantly after 1-2 hours of treatment as judged by Giemsa-stained smears and by trypan blue membrane staining and fluorescence microscopy. However, the rings were present in erythrocytes that appeared intact in blood smears, by light microscopy and trypan blue exclusion. This suggests that the polyene macrolide preparations do not destroy the ring-stage parasites by outright lysis of the infected erythrocytes, in contrast to what was found with trophozoite-infected cells. Conceivably, ring-infected host cell membranes have not yet accumulated the modifications required for selective lysis. Alternatively, AmpB and nystatin may permeabilise the infected erythrocyte membranes to an extent that does not result in haemoglobin release, but is sufficient to disrupt ion gradients and small molecule homeostasis and affect ring development and survival. The inability of trypan blue to penetrate ring-infected erythrocytes after 90 minutes of incubation with the polyene macrolide preparations, however, implies that the permeabilisation, if present, is not extensive. It has been reported that ring-stage parasites import and accumulate cholesterol-rich red blood cell membrane lipid rafts (12, 21). This produces an avenue by which amphotericin B and nystatin may bind to raft cholesterol on the red blood cell surface, find their way to ring membranes and disrupt the latter. Such a mechanism would further imply that ring-stage parasite membranes may be significantly more prone to permeabilisation by amphotericin B and nystatin than later trophozoite membranes or uninfected red blood cell membranes.

The liposomal amphotericin B preparation, AmBisome™, appeared to have a less rapid plasmodicidal mode of action, but similar to that of nystatin and non- liposomal AmpB (amphotericin B-deoxycholate). A significant number of apparent red blood cell-free trophozoites were found after 9 hours of incubation, as opposed to 30 minutes for nystatin and AmpB. In fact, a number of trophozoites continued development into the schizont stage and completed subsequent invasion, yielding rings. Ring-stage parasite morphology or parasitemia was not affected over a 9 hour period, compared to the considerable effects found with the other two macrolide preparations in 2 hours or less, but the rings failed to develop into trophozoites. The 50% parasite-inhibitory concentration of liposomal amphotericin B (5.4 μM) falls within the range of peak plasma concentrations found in clinical applications (7.9-90 μM; AmBisome™ package insert [http://www.fujisawa.com/medinfo/pi/pi_page_amb.htm]) and is considerably lower than the haemolytic concentration (>500 μM). However, the prolonged incubation times necessary to produce notable effects on parasites suggest that repeated dosages might be required to sustain effective drug levels, making it less attractive in the clinical setting as an antimalarial drug despite the benefits offered by the liposomal formulation in reducing in vivo amphotericin B toxicity (35).

The parasite-inhibitory concentration found for the amphotericin B preparation in this study (0.22 μM) agrees with the IC₅₀S reported elsewhere (13) and also falls within the peak plasma concentration range obtained in clinical applications (0.5 -

2.2 μM; Fungizone™ package insert [http://home.intekom.com/ pharm/bm_squib/fungiz-i.html]). Fungizone™ has been approved for systemic intravenous us against systemic fungal infections. It has thus potential application in critical care cases of severe malaria, where parasitised red blood cell sequestration and resetting in blood capillaries result in a reduction of circulation in vital organs. The rapid lysis of trophozoite-infected erythrocytes could conceivably destabilise the obstructing rosettes and facilitate the restoration of perfusion to ischemic areas. In summary, this invention highlights the capacity of membrane-active compounds to selectively lyse malaria parasite-infected erythrocytes, with potential clinical application in cases of severe malaria caused by sequestration of Plasmodium falciparum-infected cells.

While the invention has been described in detail with respect to specific embodiments thereof, it will be apparent to those skilled in the art that various alterations, modifications and other changes may be made to the invention without departing from the spirit and scope of the present invention.

REFERENCES

1. Adams, S., H. Brown and G. Turner. 2002. Breaking down the blood-brain barrier: signalling a path to cerebral malaria? Trends. Parasitol. 18:360-366.

2. Atamna, H. and H. Ginsburg. 1993. Origin of reactive oxygen species in erythrocytes infected with Plasmodium falciparum. MoI. Biochem. Parasitol. 61:231-241.

3. Baginski, WI., H. Resat and J.A. McCammon. 1997. Molecular properties of amphotericin B membrane channel: a molecular dynamics simulation. MoI. Pharmacol. 52:560-570.

4. Behnke, O., J. Tranum-Jensen and B. van Deurs. 1984. Filipin as cholesterol probe. II. Filipin-cholesterol interaction in red blood cell membranes. Eur. J. Cell Biol. 35:200-215.

5. Berestein. 1987. Toxicity and therapeutic effects in mice of liposome-encapsulated nystatin for systemic fungal infections. Antimicrob. Agents Chemother. 31 :1901-1903.

6. Charbonneau, C, I. Fournier, S. Dufresne, J. Barwicz and P. Tancrede. 2001. The interactions of amphotericin B with various sterols in relation to its possible use in anticancer therapy. Biophys. Chem. 91:125-133.

7. Craig, A. and A. Scherf. 2001. Molecules on the surface of the Plasmodium falciparum infected erythrocyte and their role in malaria pathogenesis and immune evasion. MoI. Biochem. Parasitol. 115:129-143.

8. Deray, G. 2004. Amphotericin B nephrotoxicity. J. Antimicrob. Chemother. 49:37-41.

9. Eda, S. and I.W. Sherman. 2002. Cytoadherence of malaria-infected red blood cells involves exposure of phosphatidylserine. Cell. Physiol. Biochem. 12:373-384.

10. GaIHs, H.A., R.H. Drew and W. W. Pickard. 1990. Amphotericin B: 30 years of clinical experience. Rev. Infect. Dis. 12:308-329.

11. Ginsburg, H., O. Famin, J. Zhang and M. Krugliak. 1998. Inhibition of glutathione- dependent degradation of heme by chloroquine and amodiaquine as a possible basis for their antimalarial mode of action. Biochem. Pharmacol. 56:1305-1313.

12. Haldar, K., N. Mohandas, B.U. Samuel, T. Harrison, N.L. Hiller, T. Akompong and P. Cheresh. 2002. Protein and lipid trafficking induced in erythrocytes infected by malaria parasites. Cell. Microbiol. 4:383-395.

14. Hsiao, L.L., RJ. Howard, Nl. Aikawa and T.F. Taraschi. 1991. Modification of host cell membrane lipid composition by the intra-erythrocytic human malaria parasite. Biochem. J. 274:121-132.

15. Huber, S.M., A.-C. Uhlemann, N. L. Gamper, C. Duranton, P.G. Kremsner and F. Lang. 2002. Plasmodium falciparum activates endogenous Cl^" channels of human erythrocytes by membrane oxidation. EMBO J. 21:22-30.

16. Jackson, K.E., N. Klonis, DJ.P. Ferguson, A. Adisa, C. Dogovski and L. Tilley. 2004. Food vacuole-associated lipid bodies and heterogeneous lipid environments in the malaria parasite, Plasmodium falciparum. MoI. Microbiol. 54:109-122.

17. Kirk, K. 2001. Membrane transport in the malaria-infected erythrocyte. Physiol. Rev. 81:495-537.

18. Weinberg, M.E. and A. Finkelstein. 1984. Single-length and double-length channels formed by nystatin in lipid bilayer membranes. J. Membr. Biol. 80:257-269.

19. Koontz, J.L and J.E. Wlarcy. 2003. Formation of natamycin:cyclodextrin inclusion complexes and their characterization. J. Agric. Food Chem. 51:7106-7110.

20. Kriek, N., L Tilley, P. Horrocks, R. Pinches, B.C. Elford, D.J.P. Ferguson, K. Lingelbach and C.I. Newbold. 2003. Characterization of the pathway for transport of the cytoadherence-mediating protein, PfEMPI , to the host cell surface in malaria parasite- infected erythrocytes. MoI. Microbiol. 50:1215-1227.

21. Lauer, S., J. VanWye, T. Harrison, H. McManus, B.U. Samuel, N.L. Hiller, N. Mohandas and K. Haldar. Vacuolar uptake of host components, and a role for cholesterol and sphingomyelin in malarial infection. EMBO J. 19:3556-3564.

22. Macreadie, I., H. Ginsburg, W. Sirawaraporn and L Tilley. 2000. Antimalarial drug development and new targets. Parasitology Today 16:438-444.

23. Maguire, P., J. Prudhomme and I.W. Sherman. 1991. Alterations in erythrocyte membrane phospholipid organization due to the intracellular growth of the malaria parasite, Plasmodium falciparum. Parasitol. 102:179-186.

24. Maguire, P.A. and I.W. Sherman. 1990. Phospholipid composition, cholesterol content and cholesterol exchange in Plasmodium ra/c/pan/m-infected red cells. MoI. Biochem. Parasitol. 38:105-112.

25. Makler, M.T. and DJ. Hinrichs. 1993. Measurement of the lactate dehydrogenase activity of Plasmodium falciparum as an assessment of parasitemia. Am. J. Trop. Med. Hyg. 48:205-210.

26. Mehta, R.T., R. L. Hopfer, T. McQueen, R.L. Juliano and G. Lopez-Berestein. 1987. Toxicity and therapeutic effects in mice of liposome-encapsulated nystatin for systemic fungal infections. Antimicrob. Agents Chemother. 31 :1901-1903.

27. Nagao, E., K.B. Seydel and J.A. Dvorak. 2002. Detergent-resistant membrane rafts are modified by a Plasmodium falciparum infection. Exp. Parasitol. 102:57-59.

28. O'Day, D.M., W.A. Ray, R.D. Robinson and W.S. Head. 1987. Correlation of in vitro and in vivo susceptibility of Candida albicans to amphotericin B and natamycin. Investigative Opthalmology and Visual Science 28:596-603.

29. Przyborski, J.M., H. Wickert, G. Krohne and M. Lanzer. 2003. Maurer's clefts - a novel secretory organelle? MoI. Biochem. Parasitol. 132:17-26.

30. Saliba, K.J. and K. Kirk. 2001. Nutrient acquisition by intracellular apicomplexan parasites: staying in for dinner. Int. J. Parasitol. 31:1321-1330.

31. Samuel, B. U., N. Mohandas, T. Harrison, H. McManus, W. Rosse, M. Reid and K. Haldar. 2001. The role of cholesterol and glycosylphosphatidylinositol-anchored proteins of erythrocyte rafts in regulating raft protein content and malarial infection. J. Biol. Chem. 276:29319-29329.

32. Sherman, I.W., S. Eda and E. Winograd. 2003. Cytoadherence and sequestration in Plasmodium falciparum: defining the ties that bind. Microbes and Infection 5:897-909.

33. Sherman, I.W. and E. Winograd. 1990. Antigens on the Plasmodium falciparum infected erythrocyte surface are not parasite derived. Parasitology Today 6:317-320.

34. Teerlink, T., B. De Kruijff and R.A. Demel. 1980. The action of pimaricin, etruscomycin and amphotericin B on liposomes with varying sterol content. Biochem. Biophys. Acta 599:484-492.

35. Tiphine, WI., V. Letscher-Bru and R. Herbrecht. 1999. Amphotericin B and its new formulations: pharmacologic characteristics, clinical efficacy, and tolerability. Transpl. Infect. Dis. 1:273-283.

36. Trelka, D.P., T.G. Schneider, J.C. Reeder and T.F. Taraschi. 2000. Evidence for vesicle-mediated trafficking of parasite proteins to the host cell cytosol and erythrocyte surface membrane in Plasmodium falciparum infected erythrocytes. MoI. Biochem. Parasitol. 106:131-145.

37. Vial, H.J., P. Eldin, A.G. Tielens and J.J. van Hellemond. 2003. Phospholipids in parasitic protozoa. MoI. Biochem. Parasitol. 126:143-154.

38. Volpon, L. and J.-WI. Lancelin. 2002. Solution NMR structure of five representative glycosylated polyene macrolide antibiotics with a sterol-dependent antifungal activity. Eur. J. Biochem. 269:4533-4541.

39. Winstanley, P.A. 2000. Chemotherapy for falciparum malaria: the armoury, the problems and the prospects. Parasitol. Today 16:146-153.

40. Foley, WI. and L. Tilley. 1998. Quinoline antimalarials: mechanisms of action and resistance and prospects for new drugs. Pharmacol. Ther. 79:55-87.

41. Trampuz, A., Wl. Jereb, I. Wluzlovic and R.WI. Prabhu. 2003. Clinical review: Severe malaria. Critical Care 7:315-323.

42. White, NJ. 1992. Antimalarial drug resistance - the pace quickens. J. Antimicrob. Chemother. 30:571-585.

Claims

CLAIMS:

1. The use of a polyene macrolide compound or analogue thereof in a method of making a medicament for use in treating malaria infection in a human and/or animal.

2. The use according to claim 1 , wherein the medicament is for use in treating cerebral malaria in humans.

3. The use according to either of claims 1 or 2, wherein the malaria infection is a new infection.

4. The use according to either of claims 1 or 2, wherein the malaria infection is a severe infection.

5. The use according to any one of claims 1 to 4, wherein the polyene macrolide compound is produced by Streptomyces species.

6. The use according to any one of claims 1 to 5, wherein the polyene macrolide compound is selected from the group consisting of amphotericin B, nystatin A, natamycin (pimaricin), filipin, rimocidin, candidin and vacidin.

7. The use according to any one of claims 1 to 6, wherein the polyene macrolide is selected from the group consisting of amphotericin B deoxycholate, amphotericin B colloidal dispersion, amphotericin B lipid complex and liposomal amphotericin B.

8. The use according to any one of claims 1 to 7, wherein the analogue of the polyene macrolide is chemically modified to improve solubility, bioavailability and/or bio-activity.

9. The use according to any one of claims 1 to 7, wherein the analogue of the polyene macrolide is chemically modified to limit toxicity.

10. The use according to either of claims 8 or 9, wherein the analogue is chemically modified by a method selected from the group consisting of oxidation, hydroxylation, acylation, amidation, coupling of an organic moiety, biosynthetic modification and hydroxy!, carbonyl, carboxyl, amino, methyl or sugar group substitution.

11. The use according to any one of claims 1 to 10, wherein the medicament is intended to be administered together with one or more other antimalarial compounds so as to increase the efficacy of the other antimalarial compound.

12. The use according to any one of claims 1 to 10, wherein the medicament includes one or more other antimalarial compounds.

13. The use according to either of claims 11 or 12, wherein the other antimalarial compound is selected from the group consisting of chloroquine, quinine, quinidine, mefloquine, halofantrine, sulfonamides, tetracyclines, atovaquone, artemisinin compounds, rimaquine, proguanil and pyrimethamine.

14. A method of treating malaria infections in animals and/or humans, the method comprising administering a therapeutically effective amount of a polyene macrolide compound, or a chemically modified analogue thereof, to an animal or human.

15. A method according to claim 14, wherein the malaria infection is cerebral malaria.

16. A method according to either of claims 14 or 15, wherein the malaria infection is a new infection.

17. A method according to either of claims 14 or 15, wherein the malaria infection is a severe infection.

18. A method according to any one of claims 14 to 17, wherein the polyene macrolide compound is produced by Streptomyces species.

19. A method according to any one of claims 14 to 18, wherein the polyene macrolide compound is selected from the group consisting of amphotericin B, nystatin A, natamycin (pimaricin), filipin, rimocidin, candidin and vacidin.

20. A method according to any one of claims 14 to 19, wherein the polyene macrolide is selected from the group consisting of amphotericin B deoxycholate, amphotericin B colloidal dispersion, amphotericin B lipid complex and liposomal amphotericin B.

21. A method according to any one of claims 14 to 20, wherein the analogue of the polyene macrolide is chemically modified to improve solubility, bioavailability and/or bio-activity.

22. A method according to any one of claims 14 to 20, wherein the analogue of the polyene macrolide is chemically modified to limit toxicity.

23. A method according to either of claims 21 or 22, wherein the analogue is chemically modified by a method selected from the group consisting of oxidation, hydroxylation, acylation, amidation, coupling of an organic moiety, biosynthetic modification and hydroxyl, carbonyl, carboxyl, amino, methyl or sugar group substitution.

24. A method according to any one of claims 14 to 23, wherein the polyene macrolide compound or analogue is administered together with one or more other antimalarial compounds so as to increase the efficacy of the other antimalarial compound.

25. A method according to claim 24, wherein the other antimalarial compound is selected from the group consisting of chloroquine, quinine, quinidine, mefloquine, halofantrine, sulfonamides, tetracyclines, atovaquone, artemisinin compounds, rimaquine, proguanil and pyrimethamine.

26. A method according to any one of claims 14 to 25, wherein the polyene macrolide compound or analogue is included in a pharmaceutical composition that is formulated to improve bio-availability and/or limit toxicity of the compound or analogue.

27. A polyene macrolide compound, analogue thereof or composition containing a polyene macrolide or analogue for use in treating malaria infections in an animal and/or human.

28. A compound or composition according to claim 27, wherein the malaria infection is a cerebral malaria infection in humans.

29. A compound or composition according to either of claims 27 or 28, wherein the malaria infection is a new infection.

30. A compound or composition according to either of claims 27 or 28, wherein the malaria infection is a severe infection.

31. A compound or composition according to any one of claims 27 to 30, wherein the polyene macrolide or analogue is produced by Streptomyces species.

32. A compound or composition according to any one of claims 27 to 31 , wherein the polyene macrolide or analogue is selected from the group consisting of amphotericin B, nystatin A, natamycin (pimaricin), filipin, rimocidin, candidin and vacidin.

33. A compound or composition according to any one of claims 27 to 32, wherein the polyene macrolide or analogue is selected from the group consisting of amphotericin B deoxycholate, amphotericin B colloidal dispersion, amphotericin B lipid complex and liposomal amphotericin B.

34. A compound or composition according to any one of claims 27 to 33, wherein the analogue of the polyene macrolide is chemically modified to improve solubility, bio-availability and/or bio-activity.

35. A compound or composition according to any one of claims 27 to 33, wherein the analogue of the polyene macrolide is chemically modified to limit toxicity.

36. A compound or composition according to claim 35, wherein the analogue is chemically modified by a method selected from the group consisting of oxidation, hydroxylation, acylation, amidation, coupling of an organic moiety, biosynthetic modification and hydroxyl, carbonyl, carboxyl, amino, methyl or sugar group substitution.

37. A compound or composition according to any one of claims 27 to 36, which is administered together with one or more other antimalarial compounds so as to increase the efficacy of the other antimalarial compound.

38. A compound or composition according to claim 37, wherein the other antimalarial compound is selected from the group consisting of chloroquine, quinine, quinidine, mefloquine, halofantrine, sulfonamides, tetracyclines, atovaquone, artemisinin compounds, rimaquine, proguanil and pyrimethamine.