International Journal of Antimicrobial Agents 13 (2000) 243 248 www.ischemo.

org

Original article

A comparison of the activities of three amphotericin B lipid formulations against experimental visceral and cutaneous leishmaniasis
Vanessa Yardley, Simon L. Croft *
Department of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK Received 5 August 1999; accepted 15 October 1999

Abstract The polyene antibiotic, amphotericin B, the gold standard for systemic fungal infections is also a recommended second line treatment for visceral, cutaneous and mucocutaneous leishmaniasis. Acute toxicity has limited the use of amphotericin B but less toxic lipid formulations, AmBisome, Amphocil and Abelcet, have shown potential for the treatment of clinical visceral and mucocutaneous leishmaniasis. This study compares the in vitro and in vivo anti-leishmanial activity of Fungizone and the three lipid formulations. AmBisome and Amphocil were more active (ED50 values 0.3 and 0.7mg/kg, respectively) than Abelcet (ED50 2.7mg/kg) against L. dono6ani in a mouse model. Against L. major in vivo, AmBisome at a dose of 25mg/kg was the most successful at reducing lesion size, with Amphocil also showing activity while Abelcet was inactive. In the L. dono6ani peritoneal macrophage (PEM) model Fungizone and Amphocil were signicantly more active (ED50 values 0.013 and 0.02 mg/ml, respectively) than AmBisome and Abelcet (ED50 values 1.5 and 2.6 mg/ml). This trend was similar in the L. major PEM model (Fungizone \Amphocil \AmBisome\Abelcet). THP-1 macrophages infected with L. dono6ani amastigotes showed a different prole with Amphocil= Abelcet\ AmBisome\ Fungizone. Differences could be due to the interaction of the formulations with the biological milieu and uptake into different cell types. 2000 Elsevier Science B.V. and International Society of Chemotherapy. All rights reserved.
Keywords: Amphotericin B; Macrophage; Leishmania; Lipid formulation

1. Introduction The drugs available for the treatment of visceral leishmaniasis (VL), cutaneous leishmaniasis (CL) and mucocutaneous leishmaniasis (MCL) are unsatisfactory. The recommended drugs remain the pentavalent antimonials, sodium stibogluconate and meglumine antimoniate, which have variable efcacy, toxic side effects and require long courses of parenteral treatment [1,2]. The antifungal drug amphotericin B, a polyene antibiotic, was rst shown to have antileishmanial activity in the late 1950s [3,4] and was used in the treatment of clinical leishmaniasis soon after [5]. Am* Corresponding author. Fax: +44-171-636-8739. E-mail address: simon.croft@lshtm.ac.uk (S.L. Croft)

photericin B has long been considered as the alternative treatment for mucocutaneous and visceral leishmaniasis [1,2,5]. The selective activity of amphotericin B against fungi and Leishmania is due to its higher afnity for 24-substituted sterols, for example ergosterol and episterol, that are predominant in the plasmamembrane of these eukaryotic microorganisms, in comparison to cholesterol in the plasmamembranes of mammalian cells [69]. The use of amphotericin B as both an antifungal and antiprotozoal agent has been restricted by toxic side effects, in particular cardiotoxicity and nephrotoxicity [10]. The development of lipid formulations of amphotericin B that have greatly reduced toxicity for the treatment of systemic mycoses [1115] has been exploited for the treatment of leishmaniasis. The unilamellar liposomal formulation AmBisome

0924-8579/00/$20 2000 Elsevier Science B.V. and International Society of Chemotherapy. All rights reserved. PII: S 0 9 2 4 - 8 5 7 9 ( 9 9 ) 0 0 1 3 3 - 8

244

V. Yardley, S.L. Croft / International Journal of Antimicrobial Agents 13 (2000) 243248

(NeXstar, USA), the amphotericin B colloidal dispersion, Amphocil (Sequus, USA), and the amphotericin B lipid complex, Abelcet (Liposome Co., USA) have all been used in the clinical treatment of VL [16 18]. AmBisome has also proved to have activity against experimental cutaneous leishmaniasis [19] as well as being used in clinical mucocutaneous leishmaniasis [20] and a CL infection resistant to antimony treatment [21]. The activities of these three formulations have been previously compared in experimental models of VL against Leishmania dono6ani at a single dose level [22,23] and two of them against Leishmania infantum [24]. In this study we report a comparison of the dose-response effects of the three lipid formulations both in vitro and in vivo against L. dono6ani and Leishmania major, causative organisms of VL and CL, respectively.

2.3. In 6itro assays

Twenty-four hours after starch induction, peritoneal macrophages (PEM) from CD1 mice (Charles Rivers Ltd, UK) were harvested by lavage and dispensed into Lab-tek (Nunc, USA) 16-well tissue culture slides and maintained at 37C, 5% CO2 air mixture. THP-1 monocytes (human-derived, ECACC, UK) were maintained in RPMI 1640 +10% HIFCS. Seventy-two hours prior to L. dono6ani amastigote infection, THP-1 cells were harvested and resuspended in fresh medium with 10 ng/ml PMA (phorbol 12-myristate 13-acetate, Sigma, UK) at a concentration of 3 105/ml. 100 ml of this suspension were dispensed into Lab-tek tissue culture slides. L. dono6ani amastigotes were harvested from an infected hamster spleen and used at a ratio of 10:1 cell. Infected cells, both PEM and THP-1, were exposed to the drug for a total of 5 days and maintained at 37C, 5% CO2. L. major amastigotes were harvested from lesions of infected BALB/c mice and used at a ratio of 5:1 macrophage. The cells were exposed to drug for 5 days and maintained at 34C, 5% CO2. At the end of all experiments slides were xed, Giemsa stained and evaluated microscopically. ED50 and ED90 values were calculated using sigmoidal regression analysis (MS xlt).

2. Materials and methods

2.1. Parasite strains

L. dono6ani (MHOM/ET/67/L82) amastigotes were harvested from the spleens of infected golden hamsters (Mesocricetus auratus). A total of 2107 amastigotes were used to infect female BALB/c mice via the lateral tail vein for in vivo experiments. L. major (MHOM/SA/ 85/JISH118) and L. major (MRHO/SU/59 NEAL-P) amastigotes were harvested from skin lesions and allowed to transform to promastigotes in Schneiders medium (Life Technologies, UK) with 10% heat-inactivated foetal calf serum (HIFCS) at 26C. Low passage, late stage promastigotes were used to infect mice subcutaneously in the shaved rump. All in vivo experiments were conducted, under licence, according to United Kingdom Home Ofce regulations.

2.4. In 6i6o assays 2.4.1. VL model Female BALB/c mice were infected, via the tail vein, with 2107 freshly harvested L. dono6ani amastigotes. After 7 days, patency of infection was checked and dosing commenced. Groups of ve mice received Pentostam (45, 15 and 5 mgSbV/kg) s.c 5 days; Fungizone (1, 0.2, 0.04 mg/kg) i.v. 1 day; AmBisome or Abelcet or Amphocil (5, 1, 0.2 mg/kg) i.v. 1 day. Liver impression smears were made 14 days post infection and activity determined by counting, microscopically, the number of amastigotes per liver cell in treated and untreated mice. ED50 values were calculated by sigmoidal regression analysis. 2.4.2. CL model Female BALB/c mice were infected in the rump with 2107 low passage, stationary phase L. major JISH118 promastigotes. At 14 days post-infection lesions of :5 mm in diameter were visible and dosing commenced. Mice received Pentostam (400 mgSbV/kg) s.c. 10 days; Fungizone (1 mg/kg) i.v.6 days; AmBisome (25 mg/kg) or Abelcet or Amphocil (12.5 mg/kg) i.v.6 days. Lesions were measured, the average of two diameters, every week. The percentage change in lesion size (mm) was calculated for each week and represented histographically. In this way, any differences between treated and untreated groups could be demonstrated. A

2.2. Drugs
Sodium stibogluconate (SbV) (Pentostam) was kindly provided by the GlaxoWellcome, UK. For in vitro experiments, the drug was diluted in RPMI 1640 medium. For in vivo experiments, the drug was dissolved in 0.25% methyl cellulose. Amphotericin B deoxycholate (Fungizone, Squibb) and Amphocil (Zeneca) were purchased commercially in 50 mg vials and were reconstituted as instructed by the manufacturer; 5 mg/ml stock solutions were prepared by the addition of sterile ice-cold water. AmBisome (NeXstar) was kindly provided by Dr R Proftt and Abelcet was generously donated by Mr G McGettigan (The Liposome Co., UK.) All formulations were prepared on the day of the experiment, by diluting 5 mg/ml stock solutions with 5% dextrose and were stored at 4C between doses for a maximum of 1 week.

V. Yardley, S.L. Croft / International Journal of Antimicrobial Agents 13 (2000) 243248 Table 1 In vitro activity of amphotericin B-lipid formulations against L. dono6ani L. dono6ani L82 PEM THP-1 ED50 9s(mg/ml) 50 93.5 0.18 9 0.961 0.2 90.03 1.0 90.05 0.2 90.01 L. major JISH118 PEM L. major NEALP PEM

245

Pentostam (mgSbV/ml) Fungizone Amphocil AmBisome Abelcet

a b

4.991.6 0.0139 0.011 0.029 0.01 1.59 0.12 2.69 0.97

35 95.6 0.6 90.21 0.9 9 0.3 2.0 9 0.5 3.79 1.2

30 96.2 0.2 90.1a 0.6 90.1b 1.89 0.3 2.9 9 1.2

AmBisome and Abelcet were signicantly different from Fungizone-treated cells (P\0.001). Amphocil was signicantly different from Fungizone-treated cells (P\0.1).

high dose of NaSbV was chosen to show that the conventional treatment for CL is not active in this model.

3. Results

3.1. In 6itro
The activities of the two standard drugs, sodium stibogluconate and amphotericin B, against L. dono6ani and L. major amastigotes in murine macrophages (Table 1) were in line with previously reported results with L. dono6ani being signicantly more sensitive to both drugs [25,26]. In all the in vitro tests amphotericin B deoxycholate (Fungizone) was signicantly more effective against intracellular amastigotes than AmBisome and Abelcet, with only Amphocil having a similar level of activity. Fungizone was signicantly more active than AmBisome and Abelcet against Leishmania infected PEM, only AmBisome was signicantly different from Fungizone in the THP-1 assay (P\ 0.001). The differences in activity were more pronounced against L. dono6ani in PEM where Fungizone was B100 times more active than AmBisome or Abelcet. L. dono6ani infected THP-1 cells gave a different prole with only AmBisome being less active than the other formulataions. Overall, Amphocil and Abelcet produced the best activities in comparison with Fungizone. The activity of the control drug, Pentostam, was also signicantly lower.

Abelcet was signicantly different from Fungizone (P\ 0.001). In this BALB/c mouse model of infection the maximum tolerated dose of Fungizone by the intravenous route was 1 mg/kg whereas the lipid formulations could be administered at 50 mg/kg (AmBisome) or 25 mg/kg (Amphocil, Abelcet) without overt signs of toxicity. In studies against established infections of L. major in BALB/c mice both AmBisome and Amphocil showed activity as indicated by a reduction in lesion size. In comparison to untreated control mice, lesion sizes were reduced by 40% (AmBisome, P\0.01)) and 30% (Amphocil, P\ 0.01) (Fig. 1) following 2 weeks of treatment during which both groups received the same total dose of amphotericin B but using different regimes (AmBisome was given at 25 mg/kg i.v. six doses and Amphocil was given at 12.5 mg/kg i.v. 12 doses). In both of these groups complete cure was not observed and lesions relapsed, however, on two of the mice in the AmBisome-treated group lesions disappeared before relapse. Neither Fungizone nor Abelcet were active against L. major as ascertained by either reduction in rate of growth of lesion size or reduction in lesion size at the maximum doses, 1 mg/kg i.v. 6 or 12.5 mg/kg i.v. 12, respectively. The standard antimonial sodium stibogluconate was also inactive at 400 mg SbV/kg in this model.
Table 2 In vivo activity of amphotericin B-lipid formulations against L. dono6ani ED50 9s Pentostam (mgSbV/kg) Fungizone (mg/kg) AmBisome (mg/kg) Amphocil (mg/kg) Abelcet (mg/kg) 13.2 93.75 0.21 90.1a 0.73 90.16 0.3290.12 2.791.2 ED90 9 s \45 \1 3.17 9 1.7 3.64 9 1.1 \5

3.2. In 6i6o
Against L. dono6ani infections all formulations of amphotericin B (single dose, i.v.) were signicantly more active than the standard pentavalent antimonial Pentostam (ve doses, s.c.) as measured by parasite load in the liver (Table 2). In a series of tests Fungizone, AmBisome and Amphocil had a similar range of ED50 values (0.20.8 mg/kg). At the ED50 level, only

a At the ED50 level Fungizone was signicantly more active than Abelcet (P\0.001).

246

V. Yardley, S.L. Croft / International Journal of Antimicrobial Agents 13 (2000) 243248

Fig. 1. In vivo activity of amphotericin B-lipid formulations against L. major JISH118. In comparison to untreated mice, AmBisome signicantly reduced lesions size, 7 (P\ 0.02), 14 (P\ 0.01) and 21 (P\ 0.01) days from start of treatment. Amphocil also had a signicant effect on lesion size (P\ 0.02, P\ 0.01 and P\ 0.001). Abelcet did not signicantly affect lesion size.

4. Discussion The three amphotericin B lipid formulations used in this study have major differences in the size and composition of the lipid vehicles (particles), pharmacokinetics and interaction with both lipoproteins and host cells [27 29]. These characteristics probably contribute to the differential activities observed in both in vitro and in vivo studies. The in vitro series of experiments highlight two points: (i) a signicant difference in activity against intracellular parasites between the amphotericin B formulations and (ii) that in vitro activities are not predictive of in vivo activities. In in vitro studies the micellar Fungizone and the colloidal dispersion Amphocil were 100 and 2 3more active than AmBisome and Abelcet (Table 1) against amastigotes of L. dono6ani in PEM at 37C and against L. major in PEM at 34C. A similar pattern was also observed in vitro against T. cruzi amastigotes where Fungizone and Amphocil were approximately 10-fold more active than AmBisome or Abelcet in the macrophage model whereas there was no difference in activity against T. cruzi amastigotes in Vero cells [30]. Leishmania amastigotes survive and multiply in the macrophage phagolysosome whereas T. cruzi amastigotes inhabit macrophage and Vero cell cytoplasm. Abelcet and Amphocil were the most active formulations in the THP-1 L. dono6ani model, with Fungizone and AmBisome having lower, similar activity. The results suggest that differences in activity of the formulations are at-

tributable to uptake and interaction with host cells rather than intracellular distribution. For example, differences in the numbers of LDL receptors of the cells would affect uptake and hence activity of amphotericin B [31]. On incubation with serum in vitro, Fungizone has been shown to dissociate rapidly releasing monomeric amphotericin B that preferentially binds to low density lipoproteins (LDL) [32]. In contrast it has also been observed that amphotericin B initially binds to high density lipoproteins (HDL) and over time distributes to LDLs, regulated by lipid transfer protein [29]. Amphocil, like Fungizone, is a micellar mixture of amphotericin B and detergent, and is thus likely to dissociate rapidly enabling the drug to bind to LDL and be taken up by infected macrophages. This process of uptake could be assisted by the upregulation of LDL receptors in Leishmania infected macrophages [33]. AmBisome and Abelcet maintain their integrity for longer which would allow them to be phagocytosed resulting in a slower accumulation of drug within the cell [28,34] and a slower intracellular antiprotozoal [35] activity. Amphotericin B associated with LDL increases toxicity to host cells, probably due to increased uptake [9]. In vitro, amphotericin B from a liposomal formulation preferentially binds to HDL [36]. Abelcet, in rodent plasma, initially binds to HDL, but over time, the overall protein binding of Abelcet is not signicantly different than Fungizone [32]. Liposomal amphotericin B (AmBisome), and Abelcet, were the least toxic formulations to cells in vitro. In contrast, AmBisome is the most active formulation in vivo, where for example circulation time (AUC) inuences the in vivo distribution. Three previous papers have reported comparative studies on some or all of these formulations against experimental VL: Gangneux [24] compared Fungizone, AmBisome and Abelcet against L. infantum in BALB/c mice and Mullen [22,23] compared Fungizone, AmBisome, Amphocil and Abelcet against L. dono6ani in BALB/c mice. In all these studies only a single dose level was used with multiple dosing (ve or six doses on alternate days) against established infections, limiting the analysis of relative potencies. Gangneux [24] found little difference between Fungizone, Abelcet and AmBisome in activity against amastigotes in livers of their early treatment groups (days 7 to 17). Mullen [22] reported in a study involving the analysis of liver, spleen and bone marrow infection, an activity ranking of Amphocil= AmBisome\Abelcet; Fungizone was not included. Our results comparing dose-response effects following single dose administration showed that Fungizone, Amphocil and AmBisome had similar potencies in this model and were more active than Abelcet. This is similar to the results of Mullen et al. (1997). The ED50 value of Fungizone is misleading as the dose is close to the maximum tolerated concentration for

V. Yardley, S.L. Croft / International Journal of Antimicrobial Agents 13 (2000) 243248

247

mice. In vivo the activity of Abelcet might be expected to be similar to or better than that of AmBisome as the particle size is larger and hence could be more readily phagocytosed by macrophages. However, AmBisome has a greater AUC [37 39], giving an increased exposure of tissues in general at a therapeutic concentration of drug as well as being eventually phagocytosed by the RES and targeted to the liver and spleen. We have previously reported a dose-response relationship of activity of AmBisome against L. major infections in mice [19]. In L. major infections, amastigote-infected macrophages in the dermis of skin are the target for drug delivery. In this study comparing three formulations, different dosing regimes were used because of tolerance of mice to Abelcet and Amphocil. New [40] rst reported that liposomal formulations of antimonials were active against CL when administered by the i.v. route rather than s.c. into the lesion. This observation was also reported for AmBisome [19]. The results suggest that AmBisome and to a lesser extent Amphocil, gains access to the dermis to kill amastigotes, whereas Abelcet cannot. The large AUC and small particle size of AmBisome might contribute to its ability to extravasate at the site of inammation, i.e. the lesion. Circulating monocytes, which act as reservoirs of infection [41,42] may also take up AmBisome, before migrating to the inammatory site [43]. The presence of amphotericin B in the circulation may also have an immunomodulatory effect, enhancing innate macrophage mechanism [44,45]. Lipid-amphotericin B formulations have been established as a useful treatment for VL and a large multicentre trial for AmBisome has given encouraging results [16]. AmBisome has also been used in HIV-VL cases [46] but is required as maintenance therapy rather than a curative treatment in these patients. The main restriction against the widespread use of the current approved lipid-amphotericin B formulations for leishmaniasis is their high cost. The search remains for a low cost formulation which could have good activity against leishmaniasis [47,48].

Acknowledgements This investigation received nancial support from the UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases (TDR).

References
[1] World Health Organisation. Control of the leishmaniases. Report of a WHO Expert Committee. WHO Tech Rep Ser 1990;793:50 5.

[2] Berman JD. Human leishmaniasis: clinical, diagnostic and chemotherapeutic developments in the last 10 years. Clin Infect Dis 1997;24:684 703. [3] Cappucino EF, Stauber LA. Some compounds active against experimental leishmaniasis. Proc Soc Exp Biol Med 1959;101:742 4. [4] Furtado TA, Cisalpino EO, Santos UM. In vitro studies of the effect of amphotericin B on Leishmania braziliensis. Antibiot Chemother 1960;10:692 3. [5] Jha TK, Giri YN, Singh TK, Jha S. Use of amphotericin B in drug-resistant cases of visceral leishmaniasis in North Bihar, India. Am J Trop Med Hyg 1995;52:536 8. [6] Cohen BE, Ramos H, Gamargo M, Urbina J. The water and ionic permeability induced by polyene antibiotices across plasma membrane vesicles from Leishmania spp. Biochim Biophys Acta 1986;860:57 65. [7] Saha AK, Mukherjee T, Bhadhuri A. Mechanism of action of amphotericin B on Leishmania donovani promastigotes. Mol Biochem Parasitol 1986;19:195 200. [8] Beggs WH. Physicochemical cell damage in relation to lethal amphotericin B action. Antimicrob Agents Chemother 1994;38:363 4. [9] Bolard J, Vertut-Doi A. Internalisation of amphotericin B and other polyene antifungals in mammalian cells: a possible origin of their toxicity. In: Georgopapdakou NH, editor. Drug Transport in Antimicrobial and Anticancer Chemotherapy. NY: Marcel Dekker, 1995:307. [10] Clements JS Jr, Peacock JE Jr. Amphotericin B revisited: reassessment of toxicity. Am J Med 1990;88:22 7. [11] Gregoriadis G. Overview of liposomes. J Antimicrob Chemother 1991;28(SB):39 48. [12] Gregoriadis G. Engineering liposomes for drug delivery: progress and problems. Trends Biotechnol 1995;13:527 37. [13] Bakker-Woudenberg IAJM, Storm G, Woodle MC. Liposomes in the treatment of infection. J Drug Target 1994;2:363 71. [14] Hiemenz JW, Walsh TJ. Lipid formulations of amphotericin B: recent progress and future directions. Clin Infect Dis 1996;22(S1):133 44. [15] Janknegt R, van Etten EWM, de Marie S. Lipid formulations of amphotericin B. Curr Opin Infect Dis 1996;9:403 6. [16] Berman JD, Badaro R, Thakur CP, Wasunna KM, Bebehani K, Davidson R, Kuzoe F, Pang L, Weerasuriya K, Bryceson ADM. Efcacy and safety of liposomal amphotericin B (AmBisome) for visceral leishmaniasis in endemic developing countries. Bull WHO 1998;76:25 32. [17] Dietze R, Fagundes SMS, Brito EF, Feitosa TT, Suassana FAB, Fonschiffrey G, Ksionski G, Dember J. Treatment of Brazilian kala-azar with a short course of Amphocil (amphotericin B cholesterol dispersion). Clin Infect Dis 1995;17:981 6. [18] Sundar S, Murray HW. Cure of antimony-unresponsive Indian visceral leishmaniasis with amphotericin B lipid complex. J Infect Dis 1996;173:762 5. [19] Yardley V, Croft SL. Activity of liposomal amphotericin B against experimental cutaneous leishmaniasis. Antimicrob Agents Chemother 1997;41:752 6. [20] Sampaio SA, Godoy JT, Paiva L, Dillon NL, Lacas C da S. The treatment of American (mucocutaneous)leishmaniasis with amphotericin B. Arch Dermatol Syphilis, Wein 1960;82:627 35. [21] Torre-Cisneros J, Prada JL, Kindelan JM, Jurado R, SanchezGuijo P. Successful treatment of antimony-resistant visceral leishmaniasis with liposomal amphotericin B in patients infected with human immunodeciency virus. Clin Infect Dis 1994;17:625 7. [22] Mullen AB, Carter KC, Baillie AJ. Comparison of the efcacies of various formulations of amphotericin B against murine visceral leishmaniasis. Antimicrob Agents Chemother 1997;41:2089 209.

248

V. Yardley, S.L. Croft / International Journal of Antimicrobial Agents 13 (2000) 243248 [36] Wasan KM, Rosenblum MG, Cheung L, Lopez-Berestein G. Inuence of lipoproteins on renal cytotoxicity and antifungal activity of amphotericin B. Antmicrob Agents Chemother 1994;38:223 7. [37] Janknegt R, de Marie S, Bakker-Woudenberg IAJM, Crommelin DJA. Liposomal and lipid formulations of amphotericin B. Clin Pharmacokinet 1992;23:279 91. [38] Heinemann V, Kahny B, Debus A, Wachholz K, Jehn U. Pharmacokinetics of liposomal amphotericin B (AmBisome) versus other lipid-based formulations. Bone Marrow Transplant 1994;14:S8 9. [39] Heinemann V, Bosse D, Jehn U, Kahny B, Wachholz K, Debus A, Scholz P, Kolb H-J, Wilmanns W. Pharmacokinetics of liposomal amphotericin B (AmBisome) in critically ill patients. Antimicrob Agents Chemother 1997;41:1275 80. [40] New RRC, Chance ML, Heath S. The treatment of experimental cutaneous leishmaniasis with liposome-entrapped pentostam. Parasitology 1981;83:519 27. [41] Mirkovich AM, Galelli A, Allison AC, Moddaber FZ. Increased myelopoiesis during Leishmania major infection in mice: generation of safe targets, a possible way to evade the effector immune system. Clin Exp Immunol 1986;64:1 7. [42] Aebischer T, Moody SF, Hardman E. Persistence of virulent Leishmania major in murine cutaneous leishmaniasis: a possible hazard for the host. Infect Immunol 1993;61:220 6. [43] McDonald PJ, Pruul H. Phagocyte uptake and transport of azithromycin. Eur J Clin Microbiol Infect Dis 1991;10:828 33. [44] Tohyama M, Kawakami K, Saito A. Anticryptococcal effect of amphotericin B is mediated through macrophage production of nitric oxide. Antimicrob Agents Chemother 1996;40:1919 23. [45] Mozaffarian N, Berman JW, Casadevall A. Enhancement of nitric oxide synthesis by macrophages represents and additional mechanism of action for amphotericin B. Antimicrob Agents Chemother 1997;41:1825 9. [46] Russo R, Nigro LC, Minniti S, Montineri A, Gradoni L, Caldeira L, Davidson RN. Visceral leishmaniasis in HIV infected patients: treatment with high dose liposomal amphotericin B (AmBisome). J Infect 1996;32:133 7. [47] Golenser J, Domb A, Ehrenfreund T, Frankenburg S, Rock P, Yardley V, Croft SL. New treatment for hemoagellates using amphotericin B-arabinogalactan conjugates. IX International Congress of Parasitology 1998; Editors Tada I, Komija S, Tsuji M. Publisher Monduzi Editore Sp.A, Bologna, Italy, pp. 1285 1294. [48] Petit C, Yardley V, Gaboriau F, Bolard J, Croft SL. Activity of a heat-induced reformulation of amphotericin B deoxycholate (fungizone) against Leishmania dono6ani. Antimicrob Agents Chemother 1999;43:390 2.

[23] Mullen AB, Baillie AJ, Carter KC. Visceral leishmaniasis in the BALB/c mouse: a comparison of the efcacy of a non-ionic surfactant formulation of sodium stibogluconate with those of three proprietary formulations of amphotericin B. Antimicrob Agents Chemother 1998;42:27225. [24] Gangeux J-P, Sulahian A, Garin Y J-F, Derouin F. Lipid formulations of amphotericin B in the treatment of experimental visceral leishmaniasis due to Leishmania infantum. Trans R Soc Trop Med Hyg 1996;90:5747. [25] Neal RA, Croft SL. An in vitro system for determining the activity of compounds against the intracellular amastigote form of. J Antimicrob Chemother 1984;14:46375. [26] Allen S, Neal RA. In vitro anti-Leishmanial activity of compounds in current clinical use for unrelated diseases. Drugs Exp Clin Res 1989;14:621 8. [27] Brajtburg J, Bolard J. Carrier effects on biological activity of amphotericin B. Clin Microbiol Rev 1996;9:51231. [28] Legrand P, Vertut-Doi D, Bolard J. Comparative internalisation and recycling of different amphotericin B formulations by a macrophage-like cell line. J Antimicrob Chemother 1996;37:519 33. [29] Wasan KM, Lopez-Berestein G. Diversity of lipid-based polyene formulations and their behaviour in biological systems. Eur J Clin Microbiol Infect Dis 1997;16:8192. [30] Yardley V, Croft SL. In vitro and in vivo activity of amphotericin B-lipid formulations against experimental Trypanosoma cruzi infections. Am J Trop Med Hyg 1999;61:1937. [31] Hamanaka R, Seguchi T, Sato Y, Ono M, Kohno K, Kuwano M. Rapid turnover of low density lipoprotein receptor in human monocytic THP-1 cells. FEBS Lett 1991;294:2613. [32] Bhamra R, Saad A, Bolcsak LE, Janoff AS, Swenson CE. Behaviour of amphotericin B lipid complex in plasma in vitro and in the circulation of rats. Antimicrob Agents Chemother 1997;41:886 92. [33] Hart DT. Leishmania host-parasite interactions: the development of chemotherapeutic targets and specic drug delivery systems. I. Lipoprotein mediated anti-leishmania chemotherapy. NATO ASI Series Vol H11. Host-parasite cellular and molecular infections. In: Chang K-P, Snary D, editors. Protozoal Infections. Berlin, Germany: Springer-Verlag, 1987. [34] van Etten EWM, Otte-Lambillion M, van Vianen W, ten Kate MT, Bakker-Woudenberg IAJM. Biodistribution of liposomal amphotericin B (AmBisome) and amphotericin B-desoxycholate (Fungizone) in uninfected immunocompetent mice and leucopenic mice infected with Candida albicans. J Antmicrob Chemother 1995;35:509 19. [35] Legrand P, Cheron M, Leroy L, Bolard J. Release of amphotericin B from delivery systems and its action against fungal and mammalian cells. J Drug Target 1997;4:3119.