In vitro and in vivo safety and efficacy studies of amphotericin B on Babesia gibsoni

Veterinary Parasitology 205 (2014) 424–433

Contents lists available at ScienceDirect

Veterinary Parasitology journal homepage: www.elsevier.com/locate/vetpar

In vitro and in vivo safety and efficacy studies of amphotericin B on Babesia gibsoni Masahiro Yamasaki a,∗ , Eriko Harada a , Yu Tamura a , Sue Yee Lim a , Tatsuyuki Ohsuga a , Nozomu Yokoyama a , Keitaro Morishita b , Kensuke Nakamura b , Hiroshi Ohta a , Mitsuyoshi Takiguchi a a Laboratory of Veterinary Internal Medicine, Department of Veterinary Clinical Sciences, Graduate School of Veterinary Medicine, Hokkaido University, Sapporo 060-0818, Japan b Veterinary Teaching Hospital, Graduate School of Veterinary Medicine, Hokkaido University, Sapporo 060-0818, Japan

a r t i c l e

i n f o

Article history: Received 27 December 2013 Received in revised form 4 September 2014 Accepted 7 September 2014 Keywords: Babesia gibsoni Amphotericin B Liposomal amphotericin B Anti-babesial drug

a b s t r a c t Babesia gibsoni is a causative pathogen of canine babesiosis, which is commonly treated with anti-babesial drugs; however, the development of novel, more effective anti-babesial drugs is necessary because the currently used drugs cannot remove the parasites from dogs. Therefore we investigated the anti-babesial effect of amphotericin B (AmB), a membraneactive polyene macrolide antibiotic. The interaction of such compounds with sterols in bilayer cell membranes can lead to cell damage and ultimately cell lysis. AmB exhibits in vitro activity against B. gibsoni in normal canine erythrocytes within 12 h. We also studied liposomal AmB (L-AmB), a liposomal formulation of AmB that required a longer incubation period to reduce the number of parasites. However, L-AmB completely inhibited the invasion of free parasites into erythrocytes. These results indicated that free parasites failed to invade erythrocytes in the presence of L-AmB. Both AmB and L-AmB induced mild hemolysis of erythrocytes. Moreover, the methemoglobin level and the turbidity index of erythrocytes were significantly increased when erythrocytes were incubated with AmB, suggesting that AmB induced oxidative damage in erythrocytes. Finally, the anti-babesial activity of AmB in vivo was observed. When experimentally B. gibsoni-infected dogs were administered 0.5 and 1 mg/kg AmB by the intravenous route, the number of parasites decreased; however, recurrence of parasitemia was observed, indicating that AmB did not eliminate parasites completely. Blood urea nitrogen and creatinine of dogs were abnormally elevated after the administration of 1 mg/kg AmB. These results indicate that AmB has in vivo activity against B. gibsoni; however, it does not eliminate parasites from infected dogs and affects kidney function at a high dose. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Babesia gibsoni is a causative pathogen of canine babesiosis, which is commonly treated with diminazene

∗ Corresponding author. Tel.: +81 11 706 5224; fax: +81 11 706 5223. E-mail address: [email protected] (M. Yamasaki). http://dx.doi.org/10.1016/j.vetpar.2014.09.005 0304-4017/© 2014 Elsevier B.V. All rights reserved.

aceturate. Other anti-babesial drugs such as atovaquone, clindamycin, metronidazole, doxycycline, and pentamidine are also used for the treatment of canine babesiosis; however, these drugs do not eliminate the parasites from the infected dogs (Fowler et al., 1972; Farwell et al., 1982; Wulansari et al., 2003; Suzuki et al., 2007; Sakuma et al., 2009). Accordingly, the development of novel, more effective anti-babesial drugs is necessary.

M. Yamasaki et al. / Veterinary Parasitology 205 (2014) 424–433

We previously reported that nystatin, an ionophorous antibiotic, exhibited in vitro effects against B. gibsoni (Yamasaki et al., 2011). Nystatin modifies the intracellular concentrations of monovalent cations and can markedly injure cells that maintain their intracellular potassium concentration via active transporters such as Na,K-ATPase (Yamasaki et al., 2011). Nystatin can destroy B. gibsoni through its ionophorous activity (Yamasaki et al., 2011). Some dogs have erythrocytes containing high concentrations of potassium (high kalium, HK), GSH and free amino acids as a result of inherited high Na,K-ATPase activity (i.e., canine high kalium [HK] erythrocytes) (Inaba and Maede, 1984). Maede et al. (1983) reported that some dogs have canine HK erythrocytes containing high concentration of potassium. In addition, canine HK erythrocytes are useful host cells for in vitro culture of B. gibsoni because the parasites proliferate better in canine HK erythrocytes than in canine normal erythrocytes with low potassium (LK) concentrations (i.e., LK erythrocytes) (Yamasaki et al., 2000). Nystatin cannot reduce the number of B. gibsoni in HK erythrocytes, although it exhibits in vitro anti-babesial effects on the parasites in LK erythrocytes (Yamasaki et al., 2011). The anti-babesial activity of nystatin seems to be counteracted by Na,K-ATPase activity in HK erythrocytes. In addition, canine HK erythrocytes are mildly hemolysed by nystatin (Yamasaki et al., 2011); therefore, it is possible that nystatin could cause hemolytic anemia in dogs with HK erythrocytes, although it does not affect LK erythrocytes or peripheral polymorphonuclear leukocytes (Yamasaki et al., 2011). Moreover, nystatin cannot currently be intravenously administered to dogs. It is therefore not useful as a therapeutic treatment against B. gibsoni infection. Amphotericin B (AmB) is a membrane-active polyene macrolide antibiotic and an antifungal compound that is similar in structure to nystatin; therefore, it might also induce hemolysis of canine HK erythrocytes. However, it can be intravenously administered to dogs. AmB and nystatin are membranolytic due to their lipid-binding activities and the target membrane is thought to be osmotically fragile by binding AmB to beta-ergosterol, the principal fungal and protozoal sterol (Wiehart et al., 2006). Wiehart et al. (2006) also found that erythrocytes infected with the trophozoite stage of Plasmodium falciparum were particularly susceptible to lysis by AmB and nystatin; however, AmB and nystatin can also bind cholesterol in mammalian cell membranes, albeit with lower affinity. Thus AmB and nystatin induce tissue injury to the kidney and erythrocytes. Recently, therefore, a liposomal formulation of AmB was developed for intravenous use. Liposomal AmB (L-AmB) displays specificity for P. falciparum-infected erythrocytes, but complete lysis requires a longer incubation period than for AmB (Wiehart et al., 2006). These results imply that AmB and L-AmB may be effective in the treatment of severe malaria caused by P. falciparum (Wiehart et al., 2006). The piroplasms Babesia rodhaini and Theileria parva are inhibited by nystatin and AmB in vitro (McColm and McHardy, 1984), suggesting that AmB and LAmB might be useful as therapeutic agents against Babesia spp. parasites. However, it is possible that AmB and LAmB could cause hemolysis of canine HK erythrocytes

425

as described above; therefore, the effects of AmB and L-AmB on canine erythrocytes should be examined. In addition, their effects on canine leukocytes should be determined. In the present study, we examined the effects of AmB and L-AmB on B. gibsoni, canine erythrocytes, and peripheral polymorphonuclear leukocytes in vitro. Additionally, we treated experimentally B. gibsoni-infected dogs with AmB in vivo to evaluate its therapeutic activity. 2. Materials and methods 2.1. Preparation of canine erythrocytes Canine HK erythrocytes containing a high concentration of potassium (high kalium, HK) and a low concentration of sodium as a result of inherited high Na,K-ATPase activity (Inaba and Maede, 1984) were obtained from three mongrel male dogs that had inherited high kalium (HK) erythrocytes. Dogs with HK erythrocytes have been maintained since 1986 in our laboratory. Canine normal erythrocytes having low potassium (low kalium, LK) and high sodium concentrations (i.e., low kalium [LK] erythrocytes) were obtained from three genetically unaffected male beagle dogs. Peripheral polymorphonuclear leukocytes were also prepared from peripheral blood of three beagle dogs with LK erythrocytes. Canine LK and HK erythrocytes were identified by measuring the intracellular concentrations of potassium and sodium (Yamasaki et al., 2005). The dogs used had body weights of 8–12 kg and were 2–3 years old. In the experimental protocols for animal care and handling, the investigators adhered to the guidelines of Hokkaido University, which basically conform to those of the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC). The present study was approved by the Committee for Laboratory Animals, Graduate School of Veterinary Medicine, Hokkaido University (approval number: 1022). 2.2. Cultivation of B. gibsoni The B. gibsoni used in the present study originated from a naturally infected dog from Nagasaki, Japan in 1973. Since then this isolate has been maintained in experimentally infected dogs and in culture (Yamasaki et al., 2003). For in vitro assay, the parasites maintained in culture with LK erythrocytes were utilized. To prepare erythrocytes for culture, blood samples were collected into sterile disposable syringes, immediately moved into blood collection tubes with ethylenediaminetetraacetic acid disodium salt (EDTA-2Na; Wako Pure Chemical Co., Osaka, Japan), and washed according to the method of Yamasaki et al. (2000). Briefly, blood samples were centrifuged at 900 g for 5 min at room temperature (ca. 25 ◦ C). After removal of the plasma and buffy coat, packed cells were resuspended in dog plasma to yield a packed cell volume (PCV) of 50% (v/v) and filtered through an alpha-cellulose/microcrystalline cellulose column to remove leukocytes and platelets. Filtered cells were washed three times with 10 mM phosphate-buffered saline (PBS, pH 7.4) and washed twice with RPMI-1640 with l-glutamine and 25 mM HEPES

426

M. Yamasaki et al. / Veterinary Parasitology 205 (2014) 424–433

(Invitrogen, Carlsbad, CA, U.S.A.) supplemented with sodium pyruvate (0.11 mg/mL), glutamine (0.3 mg/mL), penicillin (100 units/mL) and streptomycin (100 ␮g/mL). After washing, they were resuspended in culture medium consisting of 80% (v/v) of the RPMI-1640 described above and 20% (v/v) dog serum, to yield a PCV of 5% (v/v) as uninfected fresh erythrocyte suspension. For the preparation of dog sera, blood was collected from three beagle dogs described above. Blood drawing was performed on a different day of the preparation of erythrocytes. Collected sera were frozen and stored −30 ◦ C, and thawed once for use. The parasites were routinely cultured at 38 ◦ C in a humidified atmosphere of 5% (v/v) CO2 , 5% (v/v) O2 , and 90% (v/v) N2 in the culture medium described above and canine LK erythrocytes sufficient to yield a PCV of 5% (v/v). Every 24 h, 60% (v/v) of the culture supernatant was removed and replaced with an equal volume of fresh culture medium. Every 7 days, a half volume of cultured infected erythrocyte suspension was removed and replaced with an equal volume of uninfected fresh erythrocyte suspension as a subculture. The antibabesial activities of AmB (Fungizone; BristolMyers Squibb Company, NY) and L-AmB (Ambisome; Astellas Pharma Inc., Tokyo, Japan) against B. gibsoni were determined. AmB was dissolved in 5% (v/v) glucose solution (Otsuka Glucose Injection 5%, Otsuka Pharmaceutical Co., Ltd., Tokyo, Japan) to a concentration of 5 mg/mL, and serially diluted to 100, 50, 25, and 12.5 ␮g/mL in culture medium. L-AmB was dissolved in distilled water to a concentration of 4 mg/mL, and serially diluted to 400, 200, 100, 50, and 25 ␮g/mL in culture medium. One-tenth of the total volume of the culture was replaced by the diluted drug solutions described above. Then, B. gibsoni cultured in LK erythrocytes were incubated in culture media containing 0, 1.25, 2.5, 5, or 10 ␮g/mL AmB (the initial level of parasitemia was 4.5 ± 0.63%) and 0, 2.5, 5, 10, 20, or 40 ␮g/mL L-AmB for 24 h (the initial level of parasitemia was 2.7 ± 0.64%). In addition, the parasites were incubated in culture medium containing 5 ␮g/mL AmB for 0, 4, 8, 12, 20, and 24 h, and incubated in culture medium containing 40 ␮g/mL L-AmB for 0, 12, 24, 48, 72, and 96 h. A Giemsa stained thin smear sample was made and the percentage of parasitemia was calculated by counting the number of parasitized cells per 2000 erythrocytes. These experiments were repeated three times. The 50% inhibitory concentration (IC50 ) of AmB at 24 h was calculated using probit analysis. In the analysis, all data from three experiments were used in one analysis. In addition, B. gibsoni cultured as described above was isolated from infected host cells by the method of Sugimoto et al. (1991) with some modifications. For the preparation of free parasites, hemolysin purified from Aeromonas hydrophila strain Ah-1 was used. The purified hemolysin was a gift from Prof. C. Sugimoto, Department of Collaboration and Education, Research Center for Zoonosis Control, Hokkaido University. A suspension of infected erythrocytes was centrifuged at 800 × g for 5 min at 4 ◦ C. After removal of the culture supernatant, erythrocytes were resuspended in an equal volume of Tris buffer (10 mM Tris–HCl, 150 mM NaCl, pH 7.4). After two washes, the erythrocytes were

resuspended with the Tris buffer at a concentration of 50% (v/v). Ah-1 hemolysin was then added to a final concentration of 300 hemolytic units (HU)/mL. After incubation at 37 ◦ C for 10 min, the erythrocyte lysate was cooled on ice. EDTA-2Na solution (0.5 M, pH 9.0) was added to a final concentration of 5 mM. The erythrocyte lysate was then centrifuged at 5000 × g for 5 min at 4 ◦ C. After removal of the erythrocyte lysate and erythrocyte ghosts, the gray pellet was collected as free parasites and resuspended in RPMI-1640. After two washes with RPMI-1640, the number of parasites was counted using a Burker-Turk counting chamber. Then 1 × 105 mL−1 free parasites were cultured with fresh uninfected erythrocytes (PCV 5% [v/v]) in culture medium containing 40 ␮g/mL L-AmB for 7 days. The culture condition was the same as described above. In addition, free parasites were cultured with fresh uninfected erythrocytes without L-AmB as a control. The level of parasitemia was measured every 24 h as described above. This experiment was repeated three times.

2.3. Incubation of canine erythrocytes in media containing AmB and L-AmB Canine LK and HK erythrocytes were collected from uninfected dogs and washed as described above. They were resuspended in culture medium to yield a PCV of 5% (v/v) as uninfected erythrocyte suspension. AmB dissolved in 5% (v/v) glucose solution was prepared, and L-AmB dissolved in distilled water was prepared as described above. They were serially diluted to 200, 100, 50, and 25 ␮g/mL in culture medium. One-tenth of the total volume of uninfected erythrocyte suspension was replaced by the diluting drug solutions. Then, uninfected LK and HK erythrocytes suspensions were incubated in media containing 0, 2.5, 5, 10, or 20 ␮g/mL AmB or L-AmB at 37 ◦ C for 24 h in the same condition as the culture of B. gibsoni to observe the effects of AmB and L-AmB on canine erythrocytes. The extracellular hemoglobin (Hb) concentrations in canine erythrocyte suspensions with AmB and L-AmB were measured as previously described (Yamasaki et al., 2011). Briefly, after incubation, erythrocyte suspensions were centrifuged at 4000 × g for 5 min at room temperature. The hemoglobin concentrations of the resultant supernatant were measured using the hemoglobin test (Wako Pure Chemical Industries, Osaka, Japan). Additionally, to induce complete hemolysis, an uninfected erythrocyte suspension was incubated in culture medium with 0.75% (w/v) saponin for 10 min, and the extracellular Hb concentration was measured as a control. For evaluation of the oxidative damage to hemoglobin in canine erythrocytes cultured with AmB, the methemoglobin (metHb) level and turbidity index were also measured. After 24 h incubation, canine erythrocytes were harvested and washed with 154 mM NaCl three times. For the measurement of metHb and the turbidity index, the washed erythrocytes were resuspended in 154 mM NaCl to yield a PCV of 10% (v/v). MetHb was assayed as described by Hegesh et al. (1970). The turbidity index was measured as described by Winterbourn (1979). These experiments were repeated three times.

M. Yamasaki et al. / Veterinary Parasitology 205 (2014) 424–433

2.4. Incubation of canine peripheral polymorphonuclear leukocytes in media containing AmB and L-AmB Peripheral polymorphonuclear leukocytes (PMNs) were prepared according to the method of Yamamori et al. (2000). PMNs were prepared from peripheral blood of three beagle dogs with LK erythrocytes. The isolated PMNs were resuspended in culture medium to yield a cell count of 1500/␮L and incubated in culture medium containing 0, 1.25, 2.5, 5, or 10 ␮g/mL AmB and 0, 2.5, 5, 10, 20, or 40 ␮g/mL L-AmB for 24 h to observe the effects of AmB and L-AmB on canine PMNs. After incubation, dead PMNs were stained with 0.3% (w/v) trypan blue solution (Wako Pure Chemical Co.) and the percentage of live PMNs was calculated by counting the unstained cells per 500 PMNs. This experiment was repeated three times.

427

107–120 mEq/L) were measured using a Dry-Chem 7000V just before (day 9) and after (day 9 and 10) administration of AmB (FUJIFILM Corporation, Tokyo, Japan). In dog 4, the same biochemical examination of blood was additionally performed just before and after administration of AmB (day 16). This study was also approved as described above. 2.6. Administration of AmB to normal uninfected dogs Three normal uninfected beagle dogs with LK erythrocytes were given 0.5 mg/kg on the first day and 1 mg/kg AmB on the second day intravenously. It was administered as the same procedure as B. gibsoni-infected dogs. Just before and after administration, peripheral blood of these dogs was collected, and the concentrations of BUN,

2.5. Administration of AmB to experimentally B. gibsoni-infected dogs An adult female dog was chronically infected with B. gibsoni in our laboratory (Yamasaki et al., 2011). B. gibsoniinfected blood was collected in a heparinized syringe. The percentage of parasitemia in this blood was calculated by counting the number of parasitized cells as described above, and the erythrocyte count was measured using a Celltac-alpha (Nihon Kohden Corporation, Tokyo, Japan). The number of parasites in 1 mL of blood was calculated using the percentage of parasitemia and erythrocyte count. Then whole blood, which contained the equivalent of 1 × 109 B. gibsoni particles was used to inoculate four adult mongrel dogs (2 male and 2 female) with LK erythrocytes intravenously to create acute B. gibsoni-infected dogs. Collection of blood samples was performed before and after inoculation with Babesia-infected blood. EDTAtreated blood specimens were used for CBC. The percentage of parasitemia was determined using Giemsa stained thin smear sample as described above. They showed a mild degree of anemia and recovered without treatment. The dogs were splenectomized at day 30 post-inoculation. Two of the four dogs (one male and one female, dogs 1 and 2) were not given AmB as controls. The other two dogs (dogs 3 and 4) were AmB-treated dogs. Both dogs were given 0.5 mg/kg AmB on day 9 and 1 mg/kg AmB on day 10 post-splenectomy. Briefly, AmB was dissolved in 5% (v/v) glucose solution to a final concentration of 0.01 mg/mL. AmB in 5% (v/v) glucose solution was administered to the dogs for 4 h by infusion. Since dog 4 showed a significantly increased level of parasitemia on day 16 post-splenectomy, it was given 1 mg/kg AmB on day 16. When the PCV of the peripheral blood of both control and AmB-treated dogs was less than 20%, 2 mg/kg diminazene diaceturate (Ganaseg; Novartis Animal Health K.K., Tokyo, Japan) was administered (endpoint of this study). After the experiments, all infected dogs were maintained as chronically infected dogs in our laboratory. In dog 3 and 4, the concentrations of blood urea nitrogen (BUN; reference range 17.6–32.8 mg/dL), creatinine (Cr; reference range 0.8–1.8 mg/dL), sodium (reference range 147.0–156.0 mEq/L), potassium (reference range 3.40–4.60 mEq/L), and chloride (reference range

Fig. 1. Low-potassium erythrocytes infected with Babesia gibsoni were incubated in medium containing amphotericin B (AmB). (A) The parasitized erythrocytes were incubated in medium containing 0–10 ␮g/mL AmB for 24 h. The level of parasitemia before the incubation was described as Pre. *Significantly (P < 0.05) different from the value without AmB. (B) The parasitized erythrocytes were incubated in medium containing 5 ␮g/mL AmB for 24 h. The percentages of parasitized cells with AmB (closed circles) and without AmB (open circles) were measured at 4, 8, 12, 20 and 24 h. † Significantly (P < 0.05) different from the values without AmB.

428

M. Yamasaki et al. / Veterinary Parasitology 205 (2014) 424–433

Cr, sodium, potassium, and chloride were measured using a Dry-Chem 7000V (FUJIFILM Corporation, Tokyo, Japan).

2.7. Statistical analysis Results are shown as the mean ± standard deviation. The statistical significance of the differences in the results for the level of parasitemia, Hb concentration, metHb, turbidity index, BUN, Cr, sodium, potassium, and chloride between with and without AmB or L-AmB were evaluated using Student’s t-test. Results were considered significant when P < 0.05. These analyses were carried out on a computer using the JMP® 8 statistical software package, (SAS Institute Japan Ltd., Tokyo, Japan).

3. Results 3.1. Antibabesial activities of AmB and L-AmB in vitro B. gibsoni in LK erythrocytes were incubated with AmB for 24 h. The level of parasitemia before the incubation was 4.5 ± 0.63%, and those with 0 and 1.25 ␮g/mL AmB were not different form that before the incubation (Fig. 1A). When the parasites were incubated in media containing 5 and 10 ␮g/mL AmB for 24 h, the levels of parasitemia were significantly (P < 0.05) decreased compared to that without AmB (Fig. 1A). The IC50 of AmB calculated from these results was 3.08 ␮g/mL. Moreover, when the parasites were incubated in medium containing 5 ␮g/mL AmB for 4, 8, 12, 20 and 24 h, the number of parasitized erythrocytes gradually decreased with time. It was significantly (P < 0.05) lower at 12, 20 and 24 h than cultures without AmB, although the number of parasitized erythrocytes without AmB gradually increased (Fig. 1B). B. gibsoni in LK erythrocytes were incubated with L-AmB for 24 h. The level of parasitemia before the incubation was 2.7 ± 0.64%, and those with 0, 2.5, 5, 10, 20, and 40 ␮g/mL L-AmB were not different from that before the incubation (Fig. 2A). Even 40 ␮g/mL L-AmB could not decrease the number of parasitized erythrocytes within 24 h (Fig. 2A). Subsequently, the parasites were incubated in medium containing 40 ␮g/mL L-AmB for 0, 12, 24, 48, 72, and 96 h. The number of parasitized erythrocytes significantly (P < 0.05) decreased at 48, 72, and 96 h, although that gradually increased until 24 h (Fig. 2B). When the free parasites were cultured in medium containing 40 ␮g/mL L-AmB, they could not invade fresh erythrocytes, and the level of parasitemia was zero during the culture period (Fig. 2C). In contrast, when they were cultured without L-AmB, the parasites were able to invade erythrocytes and proliferate well after day 3 of culture (Fig. 2C). Fig. 2. Low-potassium erythrocytes infected with Babesia gibsoni were incubated in medium containing liposomal amphotericin B (L-AmB). (A) The parasitized erythrocytes were incubated in medium containing 0–40 ␮g/mL L-AmB for 24 h. The level of parasitemia before the incubation was described as Pre. (B) The parasitized erythrocytes were incubated in medium containing 40 ␮g/mL L-AmB for 96 h. The percentages of parasitized cells with L-AmB (closed squares) and without L-AmB (open squares) were measured at 12, 24, 48, 72 and 96 h. *Significantly (P < 0.05)

different from the values without L-AmB. (C) Direct effect of L-AmB on free parasites in vitro. The free parasites were cultured with fresh uninfected low-potassium erythrocytes in culture medium with 40 ␮g/mL L-AmB (closed squares) and without L-AmB (closed circles).

M. Yamasaki et al. / Veterinary Parasitology 205 (2014) 424–433

429

L-AmB. Moreover, the extracellular Hb concentration with 2.5 ␮g/mL AmB was significantly higher than that with 2.5 ␮g/mL L-AmB (Fig. 3A). When uninfected HK erythrocytes were incubated in media containing 0, 2.5, 5, 10, and 20 ␮g/mL AmB or L-AmB at 37 ◦ C for 24 h, the extracellular Hb concentrations in the HK erythrocyte suspensions were significantly (P < 0.05) increased at doses of 5 ␮g/mL AmB or more, and at doses of 5 ␮g/mL L-AmB or more compared to cultures without AmB or L-AmB (Fig. 3B). Approximately 90% and 25% of HK erythrocytes were hemolysed by 20 ␮g/mL AmB and L-AmB, respectively. Additionally, the extracellular Hb concentrations at 2.5, 5, 10 and 20 ␮g/mL AmB were significantly higher than those with L-AmB (Fig. 3B). The extracellular Hb concentrations with AmB in the HK erythrocyte suspension were higher than in the LK erythrocyte suspension (Fig. 3). When uninfected LK and HK erythrocytes were incubated in medium containing 5 ␮g/mL AmB at 37 ◦ C for 24 h, the metHb levels and turbidity indices of both LK and HK erythrocytes were significantly (P < 0.05) elevated compared to those without AmB (Table 1). Moreover, the metHb level in LK erythrocytes incubated with AmB was significantly (P < 0.05) higher than that in HK erythrocytes incubated with AmB, although the turbidity index did not significantly differ between LK and HK erythrocytes incubated with AmB (Table 1). 3.3. Effects of AmB and L-AmB on canine peripheral polymorphonuclear leukocytes

Fig. 3. Hemolytic effects of amphotericin B (AmB, open circles) and liposomal amphotericin B (L-AmB, open squares) on uninfected low- and high-potassium erythrocytes. The extracellular hemoglobin concentrations in erythrocyte suspensions were measured after the incubation with 0, 2.5, 5, 10, or 20 ␮g/mL AmB or L-AmB at 37 ◦ C for 24 h. (A) Low-potassium erythrocytes were incubated in media containing AmB or L-AmB. (B) High-potassium erythrocytes were incubated in media containing AmB or L-AmB. *Significantly (P < 0.05) different from the values without AmB and L-AmB, respectively. † Significantly (P < 0.05) different from the values at the same dose of AmB.

3.2. Effects of AmB and L-AmB on canine erythrocytes When uninfected LK erythrocytes were incubated in media containing 0, 2.5, 5, 10, and 20 ␮g/mL AmB or LAmB at 37 ◦ C for 24 h, the extracellular Hb concentrations in the LK erythrocyte suspension were significantly (P < 0.05) increased with 5, 10, and 20 ␮g/mL AmB and with 2.5, 5, and 20 ␮g/mL L-AmB compared to those without AmB or L-AmB (Fig. 3A). The extracellular Hb concentration in the LK erythrocyte suspension treated with saponin was 1.2 ± 0.1 g/dL. Because those treated with 20 ␮g/mL AmB and L-AmB were around 0.3 g/dL, nearly 25% of the LK erythrocytes in suspension were hemolysed by AmB and

When canine PMNs were incubated in media containing 0, 1.25, 2.5, 5, or 10 ␮g/mL AmB for 24 h, the percentage of live cells were 97.7 ± 1.2, 98.3 ± 0.9, 98.4 ± 0.8, 98.0 ± 0.2, and 98.1 ± 0.3%, respectively. When canine PMNs were incubated in media containing 0, 2.5, 5, 10, 20, and 40 ␮g/mL L-AmB for 24 h, the percentage of live cells were 96.5 ± 0.1, 96.2 ± 0.0, 95.4 ± 1.4, 95.5 ± 1.2, 96.6 ± 1.4, and 96.7 ± 0.7, respectively. The percentage of live cells did not decrease by both AmB and L-AmB. 3.4. Effect of AmB on experimentally B. gibsoni-infected dogs Because AmB can be intravenously administered to dogs, the in vivo antibabesial activity of AmB alone was determined using dogs having LK erythrocytes. The four experimentally B. gibsoni-infected dogs were splenectomized (day 0). In dogs 1 and 2, the levels of parasitemia gradually increased, and were more than 2% on day 11 (dog 1) and day 13 (dog 2), respectively (Fig. 4B). The PCVs of these dogs were mostly stable until day 15, and suddenly decreased after day 16 (Fig. 4A). When the PCVs of dogs 1 and 2 were less than 20% at day 18 and day 24, respectively (Fig. 4A), these dogs were treated with diminazene diaceturate. In dogs 3 and 4, the levels of parasitemia also gradually increased, and were more than 2% on day 9, respectively (Fig. 4D and 4F). Therefore, dogs 3 and 4 were given 0.5 mg/kg AmB at day 9, and those were given 1 mg/kg AmB at day 10 as AmB-treated dogs (Fig. 4D and 4F). The levels of parasitemia in both dogs

430

M. Yamasaki et al. / Veterinary Parasitology 205 (2014) 424–433

Table 1 Effects of amphotericin B on the intracellular concentrations of methemoglobin and turbidity indices in uninfected canine normal erythrocytes and erythrocytes containing a high concentration of potassium. Amphotericin B (5 ␮g/mL)

LK erythrocytes

Methemoglobin (%) Turbidity index

HK erythrocytes

−

+

−

+

0.57 ± 0.07 0.13 ± 0.01

11.87 ± 0.29* 8.69 ± 0.99*

0.75 ± 0.15 0.15 ± 0.02

10.21 ± 0.78* , † 10.09 ± 0.71*

Canine normal erythrocytes having a low potassium (LK erythrocytes), and canine erythrocytes containing a high potassium concentration as a result of inherited high Na,K-ATPase activity (HK erythrocytes) were incubated with or without 5 ␮g/mL amphotericin B for 24 h. The methemoglobin level and turbidity index were measured for evaluation of the oxidization of the hemoglobin in canine erythrocytes with amphotericin B. Data are expressed as the mean ± SD (n = 3). * Values are significantly (P < 0.05) different from each value without AmB. † Significantly (P < 0.05) different from value of LK erythrocytes.

Table 2 Effect of amphotericin B on serum chemistry of experimentally Babesia gibsoni-infected dogs. Dog No.

Days after splenectomy 9

BUN (mg/dL) Cr (mg/dL) Sodium (mEq/L) Potassium (mEq/L) Chloride (mEq/L)

3 4 3 4 3 4 3 4 3 4

10

16

Before administration

AmB (0.5 mg/kg)

AmB (1 mg/kg)

Before administration

AmB (1 mg/kg)

18.0 15.0 1.6 1.0 147 147 3.9 3.9 111 108

18.7 15.9 1.7 1.2 145 147 3.7 4.0 107 112

41.6* 35.4* 3.5* 2.8* 149 147 4.4 3.9 116 116

– 31.4 – 1.8 – – – – – –

– 35.0* – 2.5* – 148 – 4.4 – 114

The dogs 3 and 4 were splenectomized 30 days after inoculation of B. gibsoni (day 0). After splenectomy, canine babesiosis developed. Those dogs were given 0.5 mg/kg amphotericin B (AmB) on day 9 and 1 mg/kg AmB on day 10 post-splenectomy, respectively. Only dog 4 was given 1 mg/kg AmB on day 16. Concentrations of blood urea nitrogen (BUN; reference range 17.6–32.8 mg/dL), creatinine (Cr; reference range 0.8–1.8 mg/dL), sodium (reference range 147.0–156.0 mEq/L), potassium (reference range 3.40–4.60 mEq/L), and chloride (reference range 107–120 mEq/L) were measured using a Dry-Chem 7000V (FUJIFILM Corporation, Tokyo, Japan) on day 9, 10 and 16. * Higher than the upper limit of the reference range.

decreased after the administration of AmB; however, the PCV gradually decreased (Fig. 4C and 4E). Because early recrudescence was observed in dog 4, it was also given 1 mg/kg AmB on day 16 (Fig. 4F). After the administration of AmB, the PCV in dog 4 continuously decreased (Fig. 4E), although the level of parasitemia decreased once again (Fig. 4F). Eventually, recrudescence was observed in both dogs (Fig. 4D and 4F). When the PCVs of dogs 3 and 4 were less than 20% at day 29 and day 27, respectively (Fig. 4D and 4F), these dogs were treated with diminazene diaceturate. The serum chemistry of dogs 3 and 4 showed that BUN and Cr in both dogs increased abnormally after the administration of 1 mg/kg AmB (Table 2). Moreover, the increased BUN level in dog 4 was not improved at 6 days after administration (Table 2). 3.5. Effect of AmB on normal uninfected dogs Normal uninfected dogs having LK erythrocytes were intravenously administered either 0.5 mg/kg AmB at the first day and 1 mg/kg AmB at the second day. The counts of erythrocytes, leukocytes, and thrombocytes were not affected (data not shown). The results of serum chemistry were not changed after the administration of

Table 3 Effect of amphotericin B on serum chemistry of normal dogs. Days of administration test of AmB 1

2

Before administration BUN (mg/dL) Cr (mg/dL) Sodium (mEq/L) Potassium (mEq/L) Chloride (mEq/L)

22.2 1.5 147.0 4.0 113.0

± ± ± ± ±

1.6 0.4 1.0 0.3 1.0

AmB (0.5 mg/kg) 19.6 0.8 146.0 3.7 112.7

± ± ± ± ±

3.5 0.1 0.0 0.2 1.5

AmB (1 mg/kg) 31.6 1.4 146.0 4.0 111.7

± ± ± ± ±

6.4* 0.2 2.0 0.2 2.5

Three normal beagle dogs were given 0.5 mg/kg amphotericin B (AmB) at the first day and 1 mg/kg AmB at the second day intravenously. Before and after administration of AmB, concentrations of blood urea nitrogen (BUN; reference range 17.6–32.8 mg/dL), creatinine (Cr; reference range 0.8–1.8 mg/dL), sodium (reference range 147.0–156.0 mEq/L), potassium (reference range 3.40–4.60 mEq/L), and chloride (reference range 107–120 mEq/L) were measured using a Dry-Chem 7000V (FUJIFILM Corporation, Tokyo, Japan). Data are expressed as the mean ± SD (n = 3). * Significantly (P < 0.05) different from value without AmB.

0.5 mg/kg AmB (Table 3). Only BUN was significantly (P < 0.05) increased after the administration of 1 mg/kg AmB, although its level was not higher than the upper limit of the reference range (Table 3).

M. Yamasaki et al. / Veterinary Parasitology 205 (2014) 424–433

431

Fig. 4. The effects of amphotericin B (AmB) on experimentally Babesia gibsoni-infected dogs. Hematocrit values (A, C, E) were measured for observing the development of anemia. The level of parasitemia (B, D, F) were calculated for observing the proliferation of the parasites and the effect of AmB on the parasites in vivo. As a control, two experimentally B. gibsoni-infected dogs (dogs 1 and 2 [A and B]) were not treated with AmB. Hematocrit values (A) and the levels of parasitemia (B) of these control dogs were observed. As AmB-treated dogs (dogs 3 [C and D] and 4 [E and F]), two experimentally B. gibsoni-infected dogs were treated with 0.5 (closed arrowheads) or 1 mg/kg AmB (open arrowheads). Hematocrit values (C and E) and the levels of parasitemia (D and F) of these AmB-treated dogs were observed. *Dogs were given 2 mg/kg diminazene diaceturate (endpoint of this study).

4. Discussion AmB exhibited in vitro effects against B. gibsoni in LK erythrocytes within 12 h. The calculated IC50 value of AmB (3.08 ␮g/ml) was less than that of nystatin (31.96 ␮g/mL)

(Yamasaki et al., 2011). Moreover, L-AmB needed more than 48 h for the decrease of B. gibsoni in vitro, although it completely suppressed the invasion of free parasites into fresh erythrocytes. These results indicated that the liposomal formulation of AmB acted on the free parasite, and

432

M. Yamasaki et al. / Veterinary Parasitology 205 (2014) 424–433

that free parasites did not invade erythrocytes; therefore, it took a longer time to exhibit effects against B. gibsoni. The reason why the parasites could not invade erythrocytes was not proven in this study. Similarly, L-AmB displayed an anti-malarial effect against erythrocytes infected with the trophozoite stage of P. falciparum, but complete lysis of the parasitized erythrocytes required a longer incubation period (Wiehart et al., 2006). It is speculated that the liposomal formulation of AmB attacks the parasite membrane or alters erythrocytes infected with the parasites. Babesia sp. BQ1 (Lintan) were cultured in RPMI-1640 medium supplemented with 0.5 ␮g/mL AmB (Guan et al., 2010). In the study, AmB was used for the control of contamination by other microorganisms. The results of their study do not appear to agree with the results of the present study. Sheep also have either HK or LK erythrocytes, although the type of erythrocytes of sheep used in the study by Guan et al. (2010) was not identified. As described above, the antibabesial activity of nystatin seemed to be counteracted by Na,K-ATPase activity in HK erythrocytes (Yamasaki et al., 2011). Similarly, it is considered that the activity of AmB might be counteracted by HK erythrocytes, and AmB might not affect the parasites in HK erythrocytes. If HK erythrocytes of sheep are utilized for the cultivation of Babesia sp. BQ (Lintan), the anti-babesial activity of AmB would be masked by Na,K-ATPase activity in HK erythrocytes; however, it is possible that the parasites could proliferate well when the infected-erythrocytes were cultured in medium without AmB. The influences of AmB and L-AmB on LK and HK erythrocytes were observed. After 24 h incubation in medium containing either AmB or L-AmB, mild hemolysis was observed in both LK cultures. The degree of hemolysis induced by AmB in HK erythrocytes was higher than in LK erythrocytes, although that induced by L-AmB in HK erythrocytes was almost the same as that in LK erythrocytes. Previously, nystatin was found to induce hemolysis in HK erythrocytes, but could not lyse LK erythrocytes (Yamasaki et al., 2011). It is considered that nystatin affects cells with Na,K-ATPase via its ionophorous activity (Yamasaki et al., 2011). It was hypothesized that the ionophorous activity of AmB results in hemolysis of HK erythrocytes. In addition, LK erythrocytes were mildly hemolysed by both AmB and L-AmB, suggesting that these drugs induced hemolysis in LK erythrocytes via a mechanism of action different from ionophorous activity. Brajtburg et al. (1985) reported that oxidative damage to human erythrocytes was involved in the lytic, but not the permeabilizing, action of AmB. The MetHb levels and the turbidity indices of both canine LK and HK erythrocytes increased when they were incubated with AmB, indicating that the hemoglobin in erythrocytes were oxidized. Accordingly, it is hypothesized that AmB could induce oxidative damage on membrane of canine erythrocytes. AmB acts as an ionophorous antibiotic and oxidation agent. These characteristics of AmB might induce hemolysis of canine erythrocytes, especially canine HK erythrocytes. In addition, the degree of hemolysis induced by AmB was significantly greater than that induced by L-AmB, suggesting that the toxicity of AmB against erythrocytes was reduced by the liposomal formulation.

Moreover, the percentage of live PMNs did not decrease when PMNs were incubated in medium including either AmB or L-AmB, suggesting that AmB and L-AmB did not affect the survival of canine leukocytes. Nystatin does not affect canine leukocytes either (Yamasaki et al., 2011). Considering the previous report and the present results, it appears that these polyene macrolide antibiotics do not damage canine leukocytes. In addition, AmB was utilized for the treatment of experimental canine babesiosis. When the experimentally B. gibsoni-infected dogs were given 0.5 mg/kg AmB, BUN and Cr were not changed. After the administration of 1 mg/kg AmB, both BUN and Cr in B. gibsoni-infected dogs abnormally increased, and had not normalized 6 days after administration. When normal uninfected dogs were given 1 mg/kg AmB, the BUN level was not higher than the upper limit of the reference range, and that of creatinine was not changed. These results showed that the kidney injury in B. gibsoni-infected dogs was greater than in normal dogs. Generally, clinicians can administer AmB intravenously to dogs at a dose of up to 0.5 mg/kg because high-dose AmB might induce tissue injury to the kidney (Plumb, 2011); therefore, it should be diluted in 500 or 1000 mL of 5% (v/v) glucose solution and slowly administered over 3–6 h in dogs with compromised renal function regardless of age. In canine babesiosis, evidence of renal damage is common in complicated and uncomplicated cases (Taboada and Lobetti, 2006); therefore, AmB was administered using the protocol for dogs with compromised renal function, although the BUN and Cr in dogs used were normal. Moreover, renal damage in canine babesiosis might result in the apparent increase of BUN and Cr in B. gibsoni-infected dogs treated with AmB. After the administration of AmB, the level of parasitemia in the experimentally B. gibsoni-infected dogs decreased but AmB did not eliminate the parasites, which proliferated again in a short time. Thus AmB exhibited in vivo effect against B. gibsoni, but could not sterilize B. gibsoni infection. B. gibsoni-infection in dogs is characterized by fever, lethargy, anemia (Conrad et al., 1991). In the present study, anemia gradually developed in all experimentally infected dogs (control dogs and AmB-treated dogs) as stated above. No normal uninfected dog given AmB exhibited anemia, although AmB induced hemolysis in canine erythrocytes in vitro, suggesting that it could not affect the erythrocytes in vivo. It was considered that blood concentration of AmB might be less than the concentration tested in vitro in the present study; therefore, anemia in experimentally infected dogs would be the common clinical sign of B. gibsoni-infection. Since AmB induced hemolysis in HK erythrocytes in vitro, anemia might be observed in dogs with HK erythrocytes in vivo. The effects of AmB in dogs having HK erythrocytes thus need to be examined further. From the present results, it was considered that AmB could not be utilized for single-agent treatment of canine babesiosis, and that we should be aware of its adverse effects. It is reported that diminazene aceturate- and atovaquone-resistant B. gibsoni isolates have developed (Hwang et al., 2010; Matsuu et al., 2004); therefore, AmB might be useful for the treatment of

M. Yamasaki et al. / Veterinary Parasitology 205 (2014) 424–433

canine babesiosis induced by these drug-resistant parasites. The effect of L-AmB against B. gibsoni in vivo was not observed in the present study. Based on the present results, we speculate that the anti-babesial activity of L-AmB is weak in vivo, although it has few side effects. Accordingly, the in vivo effect of L-AmB against B. gibsoni-infected dogs should be determined in the future. In addition, the liposomal formulation of AmB is an expensive drug; therefore, it is costly to treat canine babesiosis with L-AmB. 5. Conclusion AmB exhibited in vitro and in vivo activities against B. gibsoni. The anti-babesial activity of the liposomal formulation of AmB was reduced, although its adverse effects on canine erythrocytes were also reduced. Moreover, kidney function was adversely affected in dogs experimentally infected with B. gibsoni that were treated with AmB. Therefore we should give AmB to B. gibsoni-infected dogs cautiously, especially dogs with HK erythrocytes, when we use it as an anti-babesial drug. Acknowledgments This work was supported in part by a Grants-in-Aid for Scientific Research (KAKENHI) from the Science Research Fund of the Ministry of Education, Culture, Sports, Science and Technology of Japan (grant number: 20688014). References Brajtburg, J., Elberg, S., Schwartz, D.R., Vertut-Croquin, A., Schlessinger, D., Kobayashi, G.S., Medoff, G., 1985. Involvement of oxidative damage in erythrocyte lysis induced by amphotericin B. Antimicrob. Agents Chemother. 27, 172–176. Conrad, P., Thomford, J., Yamane, I., Whiting, J., Bosma, L., Uno, T., Holshuh, H.J., Shelly, S., 1991. Hemolytic anemia caused by Babesia gibsoni infection in dogs. J. Am. Vet. Med. Assoc. 199, 601–605. Farwell, G.E., LeGrand, E.K., Cobb, C.C., 1982. Clinical observations on B. gibsoni and Babesia canis infections in dogs. J. Am. Vet. Med. Assoc. 180, 507–511. Fowler, J.L., Ruff, M.D., Fernau, R.C., Fursusho, Y., 1972. Babesia gibsoni: chemotherapy in dogs. Am. J. Vet. Res. 33, 1109–1114. Guan, G., Moreau, E., Brisseau, N., Luo, J., Yin, H., Chauvin, A., 2010. Determination of erythrocyte susceptibility of Chinese sheep (Tan mutton breed) and French sheep (Vendeen breed) to Babesia sp. BQ1 (Lintan) by in vitro culture. Vet. Parasitol. 170, 37–43. Hegesh, E., Gruener, N., Cohen, S., Bochkovsky, R., Shuval, H.I., 1970. A sensitive micromethod for the determination of methemoglobin in blood. Clin. Chim. Acta 30, 679–682. Hwang, S-J., Yamasaki, M., Nakamura, K., Sasaki, N., Murakami, M., Wickramasekara Rajapakshage, B.K., Ohta, H., Maede, Y., Takiguchi, M., 2010.

433

Development and characterization of a strain of Babesia gibsoni resistant to diminazene aceturate in vitro. J. Vet. Med. Sci. 72, 765–771. Inaba, M., Maede, Y., 1984. Increase of Na+ gradient-dependent lglutamate and l-aspertate transport in high K+ dog erythrocytes associated with high activity of (Na+ , K+ )-ATPase. J. Biol. Chem. 259, 312–317. Maede, Y., Inaba, M., Taniguchi, N., 1983. Increase of Na-K-ATPase activity, Glutamate, and aspartate uptake in dog erythrocytes associated with hereditary high accumulation of GSH, glutamate, glutamine, and aspartate. Blood 61, 493–499. Matsuu, A., Koshida, Y., Kawahara, M., Inoue, K., Ikadai, H., Hikasa, Y., Okano, S., Higuchi, S., 2004. Efficacy of atovaquone against Babesia gibsoni in vivo and in vitro. Vet. Parasitol. 124, 9–18. McColm, A.A., McHardy, N., 1984. Evaluation of a range of antimicrobial agents against the parasitic protozoa, Plasmodium falciparum, Babesia rodhaini and Theileria parva in vitro. Ann. Trop. Med. Parasitol. 78, 345–354. Plumb, D.C., 2011. Plumb’s Veterinary Drug Handbook, 7th edition. PharmaVet Inc., Stockholm, Wisconsin, pp. 62–66. Sakuma, M., Setoguchi, A., Endo, Y., 2009. Possible emergence of drugresistant variants of Babesia gibsoni in clinical cases treated with atovaquone and azithromycin. J. Vet. Intern. Med. 23, 493–498. Sugimoto, C., Sato, M., Kawazu, S., Kamio, T., Fujisaki, K., 1991. Purification of merozoites of Theileria sergenti from infected bovine erythrocytes. Parasitol. Res. 77, 129–131. Suzuki, K., Wakabayashi, H., Takahashi, M., Fukushima, K., Yabuki, A., Endo, Y., 2007. A possible treatment strategy and clinical factors to estimate the treatment response in Babesia gibsoni infection. J. Vet. Med. Sci. 69, 563–568. Taboada, J., Lobetti, R., 2006. Babesiosis. In: Greene, C.E. (Ed.), Infectious Diseases of the Dog and Cat. , 3rd edition. Elsevier, St. Louis, pp. 722–736. Wiehart, U.I.M., Rautenbach, M., Hoppe, H.C., 2006. Selective lysis of erythrocytes infected with the trophozoite stage of Plasmodium falciparum by polyene macrolide antibiotics. Biochem. Pharmacol. 71, 779–790. Winterbourn, C.C., 1979. Protection by ascorbate against acetylhydrazineinduced Heinz body formation in glucose-6-phosphate dehydrogenase deficient erythrocytes. Br. J. Haematol. 41, 245–252. Wulansari, R., Wijaya, A., Ano, H., Hori, Y., Nasu, T., Yamane, S., Makimura, S., 2003. Clindamycin in the treatment of Babesia gibsoni infections in dogs. J. Am. Anim. Hosp. Assoc. 39, 558–562. Yamamori, T., Inanami, O., Nagahata, H., Cui, Y., Kuwabara, M., 2000. Roles of p38 MAPK, PKC and PI3-K in the signaling pathways of NADPH oxidase activation and phagocytosis in bovine polymorphonuclear leukocytes. FEBS Lett. 467, 253–258. Yamasaki, M., Asano, H., Otsuka, Y., Yamato, O., Tajima, M., Maede, Y., 2000. Use of canine red blood cell with high concentrations of potassium, reduced glutathione, and free amino acid as host cells for in vitro cultivation of Babesia gibsoni. Am. J. Vet. Res. 61, 1520–1524. Yamasaki, M., Hossain, M.A., Jeong, J.R., Chang, H.S., Satoh, H., Yamato, O., Maede, Y., 2003. Babesia gibsoni-specific isoenzymes related to energy metabolism of the parasite in infected erythrocytes. J. Parasitol. 89, 1142–1146. Yamasaki, M., Takada, A., Yamato, O., Maede, Y., 2005. Inhibition of Na, K-ATPase activity reduces Babesia gibsoni infection of canine erythrocytes with inherited high K, low Na concentrations. J. Parasitol. 91, 1287–1292. Yamasaki, M., Tamura, N., Nakamura, K., Sasaki, N., Murakami, M., Bandula Kumara, W.R., Tamura, Y., Lim, S.Y., Ohta, H., Takiguchi, M., 2011. Effects and mechanisms of action of polyene macrolide antibiotic nystatin on Babesia gibsoni in vitro. J. Parasitol. 97, 1190–1192.