ABSTRACT
The apicoplast housekeeping machinery, specifically apicoplast DNA replication, transcription, and translation, was targeted by ciprofloxacin, thiostrepton, and rifampin, respectively, in the in vitro cultures of four Babesia species. Furthermore, the in vivo effect of thiostrepton on the growth cycle of Babesia microti in BALB/c mice was evaluated. The drugs caused significant inhibition of growth from an initial parasitemia of 1% for Babesia bovis, with 50% inhibitory concentrations (IC50s) of 8.3, 11.5, 12, and 126.6 μM for ciprofloxacin, thiostrepton, rifampin, and clindamycin, respectively. The IC50s for the inhibition of Babesia bigemina growth were 15.8 μM for ciprofloxacin, 8.2 μM for thiostrepton, 8.3 μM for rifampin, and 206 μM for clindamycin. The IC50s for Babesia caballi were 2.7 μM for ciprofloxacin, 2.7 μM for thiostrepton, 4.7 μM for rifampin, and 4.7 μM for clindamycin. The IC50s for the inhibition of Babesia equi growth were 2.5 μM for ciprofloxacin, 6.4 μM for thiostrepton, 4.1 μM for rifampin, and 27.2 μM for clindamycin. Furthermore, an inhibitory effect was revealed for cultures with an initial parasitemia of either 10 or 7% for Babesia bovis or Babesia bigemina, respectively. The three inhibitors caused immediate death of Babesia bovis and Babesia equi. The inhibitory effects of ciprofloxacin, thiostrepton, and rifampin were confirmed by reverse transcription-PCR. Thiostrepton at a dose of 500 mg/kg of body weight resulted in 77.5% inhibition of Babesia microti growth in BALB/c mice. These results implicate the apicoplast as a potential chemotherapeutic target for babesiosis.
INTRODUCTION
Babesia, a tick-born protozoan parasite, is one of the major pathogens that infect erythrocytes in a wide range of wild and economically valuable animals, such as cattle and horses. The clinical symptoms of babesiosis include malaise, fever, hemolytic anemia, jaundice, hemoglobinuria, and edema. Serious economic losses in the livestock industry have been caused by Babesia infections worldwide, mainly in tropical and subtropical areas (41). Babesia is also zoonotic for humans. Babesia microti used to be identified most often as infecting rodents and humans in North America; however, it is now recognized as doing so throughout the world (56, 66). B. microti causes a relatively mild but persistent disease and has served as a useful experimental model for analyses of animal and human babesiosis (23). Analysis of the disease in this model has enabled the search for antibabesial drugs (1, 2, 3, 74). Several babesicidal drugs that have been in use for years have proven to be ineffective owing to problems related to toxicity and the development of resistant parasites (16, 66, 70). Therefore, the development of new drugs that have a chemotherapeutic effect against babesiosis, with a high specificity for the parasites and a low toxicity to the hosts, is desired.
The apicoplast was acquired by horizontal transfer (secondary endosymbiosis) from a eukaryotic alga (40, 47) and has been identified in many apicomplexan parasites, including Toxoplasma, Plasmodium, Eimeria, Theileria, and Babesia (8, 47, 69, 73). The Babesia bovis apicoplast is a circular 33-kbp genome which unidirectionally carries 32 putative protein-encoding genes, a complete set of tRNA genes (71), and a small- and a large-subunit rRNA gene, as observed in other apicoplast genomes (20). The apicoplast is essential for long-term parasite viability (25, 32, 64). Most of the usual cellular processes, such as DNA replication, transcription, translation, posttranslational modification, catabolism, and anabolism, occur inside this organelle (57). The apicoplast has been an attractive target for the development of parasiticidal drug therapies because the biosynthetic pathways represented therein are of cyanobacterial origin and differ substantially from the corresponding pathways in the mammalian host, which are fundamentally eukaryotic (25, 57, 68). The apicoplast has anabolic pathways for the synthesis of fatty acids (FASII), isoprenoid biosynthesis (DOXP), heme biosynthesis, and iron-sulfur biosynthesis (28).
When this organelle is eliminated by either pharmacological or molecular genetic manipulation, parasites are killed through distinctive “delayed-death” kinetics. Plastid-deficient parasites are capable of normal growth within the first host cell and escape from it, but their replication is inhibited immediately upon invasion into a new host cell (25). For Toxoplasma gondii, delayed death refers to the fact that T. gondii parasites treated with an antiapicoplast drug initially show no ill effects, growing and dividing at a normal pace. Remarkably, the drugged parasites produce normal numbers of daughter parasites within the initial host cell, and these tachyzoite progeny emerge and invade new host cells. However, at this stage, having entered a new host, the parasites fail to grow and eventually die (25).
The apicoplasts of Plasmodium falciparum and Toxoplasma gondii have typical bacterial housekeeping machinery, including DNA replication, transcription, and translation pathways. In the past, antibacterials inhibited the growth of apicomplexan parasites but the mechanism was not known; therefore, antibacterials may target this machinery.
Ciprofloxacin blocks prokaryotic DNA replication by inhibiting DNA gyrase, a prokaryotic type II topoisomerase involved in untangling DNA during replication, and results in the linearization of the circular DNA, thus causing the death of prokaryotic organisms (54). DNA gyrase is encoded in the nuclear genome and consists of 2 subunits, namely, DNA gyrase subunit A and DNA gyrase subunit B. Thiostrepton appears to function by binding to the GTPase-binding domain of the large-subunit rRNA, blocking apicoplast translation (22, 46, 60). Recently, thiostrepton was shown to target the proteasome subunits and the mitochondrion translation machinery rather than the apicoplast translation machinery in P. falciparum (5, 63). Thiostrepton is a potent inhibitor of P. falciparum growth in vitro (46, 60) and in vivo (64). Rifampin blocks bacterial-type transcription and abrogates mRNA production by the apicoplast-carried rpoB gene (encoding the DNA-directed RNA polymerase beta subunit) (42, 43, 46). The targets of ciprofloxacin, thiostrepton, and rifampin are reported in the Babesia bovis genome sequence database (20).
The use of drugs targeting the bacterial housekeeping machinery of the apicoplast, i.e., DNA replication, transcription, and protein translation in the apicoplast, may have some effects on the growth cycle of Babesia parasites. Therefore, the aim of the present study was to evaluate the inhibitory effects of ciprofloxacin, thiostrepton, and rifampin on the in vitro growth of Babesia parasites.
MATERIALS AND METHODS
Chemical reagents.Ciprofloxacin, thiostrepton, rifampin, and clindamycin phosphate were purchased from Sigma-Aldrich, and other chemicals were purchased from Wako Pure Chemicals (Osaka, Japan). Stock solutions of 20 mM in ethanol (ciprofloxacin and clindamycin phosphate), dimethyl sulfoxide (DMSO) (thiostrepton), and methanol (rifampin) were prepared and stored at −30°C until use. Diminazene aceturate (Ganaseg) was purchased from Ciba-Geigy Japan Ltd. (Tokyo, Japan) and used as a comparator drug. A working stock solution of 10 mM dissolved in distilled water was prepared and stored at −30°C until required for use. Clindamycin phosphate was used as a control because of its known delayed lethal effect on P. falciparum and T. gondii. For negative control, solvents were dissolved at concentrations similar to the highest concentrations of the drugs in the treated cultures and applied to the in vitro cultures.
Mice.The Munich strain of B. microti was maintained by passaging in the blood of BALB/c mice (1). Forty female BALB/c mice (aged 8 weeks) were purchased from CLEA Japan (Tokyo, Japan) and used for the in vivo studies.
In vitro cultivation of Babesia parasites.Ciprofloxacin, thiostrepton, and rifampin were evaluated for chemotherapeutic effects against B. bovis (Texas strain) (34), B. bigemina (Argentina strain) (36), B. caballi (7), and B. equi (Theileria equi) (U.S. Department of Agriculture) (48). Parasites were grown in bovine and equine red blood cells (RBCs) by use of a continuous microaerophilic stationary-phase culture system (1). The culture medium M199, applied to B. bovis and B. bigemina (obtained from Sigma-Aldrich, Tokyo, Japan), was supplemented with 40% bovine serum, 60 U/ml of penicillin G, 60 μg/ml of streptomycin, and 0.15 μg/ml of amphotericin B (Sigma-Aldrich) and used for culture of the parasites. TES-hemisodium salt (229 mg/ml) {N-Tris-(hydroxymethyl)-methyl-2-aminoethanesulfonic acid; 2-[(2-hydroxy-1,1-bis-[hydroxymethyl]ethyl)amino] ethanesulfonic acid; Sigma-Aldrich} was added to bovine Babesia parasite cultures as a pH stabilizer (pH 7.2) (13). A serum-free GIT medium was used to evaluate the effects of the tested drugs on the growth of B. bovis and B. caballi without serum (17).
In vitro growth inhibition assay.The in vitro growth inhibition assay was adopted from a previous study (3). B. bigemina, B. caballi, B. bovis, and B. equi were obtained from cultures with parasitemias of 5, 7, and 10%. Parasite cultures for drug evaluation were adjusted to 1% for all parasites. The growth inhibition assay was performed in 96-well plates. Twenty microliters of parasite-bovine red blood cell mixture was dispensed per well together with 200 μl of culture medium containing the indicated drug concentration, on the basis of a previous study (28). The concentrations of the three tested drugs were 1, 5, 10, 25, 50, and 100 μM for bovine Babesia and 0.5, 5, 10, 25, 50, and 100 μM for equine Babesia; as controls, similar cultures without drugs and others containing only the solvents at the highest concentrations used for the drugs were prepared. The experiments were carried out in triplicate per drug concentration for each parasite species and in three separate trials. Clindamycin phosphate was used as a control, at concentrations of 0.5, 5, 100, and 500 μM (28, 47). Diminazene aceturate was compared with antibacterials at concentrations of 1, 5, 10, 50, 100, 1,000, and 2,000 nM (45). To evaluate inhibitory effects in the presence of high parasitemia (to resemble acute cases of infection), other parasite cultures were adjusted to 10% initial parasitemia for B. bovis and to 7% initial parasitemia for B. bigemina (10). Cultures were incubated at 37°C in an atmosphere of 5% CO2, 5% O2, and 90% N2. For a period of four (cultures with an initial parasitemia of 1%) or three (cultures with an initial parasitemia of 7 or 10%) days, the culture medium was replaced every day with 200 μl of fresh medium containing the appropriate drug concentration. Parasitemia was monitored on the basis of approximately 1,000 erythrocytes in a Giemsa-stained thin erythrocyte smear. Changes in the morphology of treated Babesia parasites were compared with the control morphology by light microscopy. The 50% inhibitory concentration (IC50) was calculated on the third day of in vitro culture by interpolation using the curve-fitting technique (14).
Viability test.After four (cultures with an initial parasitemia of 1%) or three (cultures with an initial parasitemia of 7 or 10%) days of treatment, 6 μl of parasite-free bovine red blood cells was added to 14 μl of a previously drug-treated culture in 200 μl of fresh growth medium without the drug. The fresh growth medium was replaced every day for the next 10 days, and parasite recrudescence was determined after removal of the drug (2).
Parasite stage inhibition and delayed-death effect.B. bovis cultures at 1% parasitemia were treated with ciprofloxacin, thiostrepton, rifampin, clindamycin, and diminazene at their IC80s or with 0.5% DMSO (negative control) in 200 μl of the culture medium. Smears were made at seven time points between 0 h and 48 h. The percentage of parasites in each stage was calculated for 50 parasites (5, 9). The stages quantified were trophozoites, dividing forms, and single-pear-shaped, double-pear-shaped, and dot-shaped merozoites. The parasitemia percentage was calculated at each time point. To determine a delayed-death effect, B. bovis cultures at 1% parasitemia were treated with drugs at the IC50 or with a 0.5% DMSO negative control in the same volume of culture medium. Smears were made every 24 h for a period of 120 h (5, 9). The percentage of parasites in each stage was calculated for 50 parasites, as mentioned above. The parasitemia percentage was calculated at each time point. B. bovis and B. equi were treated for 24 h with drug concentrations equal to the IC99s (28) of the drugs in the in vitro cultures with 1% initial parasitemia, as reported above, to investigate the delayed-death effect. The medium alone, without any drugs, was used after 24 h of treatment and until 96 h had elapsed. The growth of the treated cultures was calculated as a percentage of the control growth. The IC80 and IC99 were based on the calculated IC50s obtained from the in vitro inhibition assay. Medium M199 was used with horse serum for B. equi culture, while GIT medium alone was used for B. bovis culture.
RNA extraction and RT-PCR.Reverse transcription-PCR (RT-PCR) was used to evaluate the effects of treatment with ciprofloxacin, rifampin, thiostrepton, and clindamycin on the transcription of the target genes, carried either in the nucleus (e.g., DNA gyrase gene) or in the apicoplast (e.g., rpoB), according to previously described methods (46, 75), with some modifications. B. bovis was cultured in bovine RBCs in 24-well culture plates as reported above and was treated with ciprofloxacin, thiostrepton, rifampin, or clindamycin at the IC99 for 8 h. The negative control was used without any drug. RBCs were collected and washed with phosphate-buffered saline (PBS). The total RNA was extracted using Tri reagent (Sigma-Aldrich) according to the supplied protocol. The RNA concentration was measured using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific Inc.). Reverse transcription-PCR was conducted using a PrimeScript One-Step RT-PCR kit, version 2 (Takara, Japan), according to the manufacturer's protocol. Total RNA (150 ng) from the treated cultures or from the control was used for amplification of the DNA gyrase subunit A gene, the DNA gyrase subunit B gene, the thiostrepton interaction site (ribosomal L11 protein) of the large-subunit rRNA, rpoB, and rpoC1 (GenBank accession numbers XM_001609478, XM_001611055, XM_001610852, XM_001609672, and NC_011395 [region 5246.0.6949], respectively). The rpoC1 gene, used as an indicator of mRNA transcription for protein synthesis, was inhibited by thiostrepton and clindamycin treatment (43). Transcription of the B. bovis small-subunit rRNA gene (GenBank accession number L31922) was not affected by inhibitors of protein synthesis in the 8-hour treatment (43); therefore, it was used as a control for the inhibition of protein synthesis by thiostrepton, rifampin, and clindamycin. The B. bovis mitochondrial cytochrome c gene (GenBank accession number XM_001611891) was used to evaluate the effect of thiostrepton on the mitochondria. The B. bovis tubulin beta chain gene (GenBank accession number XM_001611566) was used as a control for transcription of the targeted genes from the ciprofloxacin-, rifampin-, thiostrepton-, and clindamycin-treated and control cultures. Specific forward and reverse primers were used to amplify the DNA gyrase subunit A, DNA gyrase subunit B, ribosomal L11 protein, RpoB, RpoC1, cytochrome c, small-subunit rRNA, and B. bovis tubulin beta chain genes. The reverse transcription reaction was carried out at 50°C for 30 min, with 2 min of denaturation at 94°C, and then PCR was repeated for 30 cycles under the following conditions: 30 s of denaturation at 94°C, 30 s of primer annealing (at 50°C for gyrase A gene, 62°C for gyrase B gene, 59.4°C for L11 protein gene, 56°C for rpoB, 52°C for rpoC1, 53°C for small-subunit rRNA gene, 55°C for cytochrome c gene, and 54°C for tubulin beta chain gene), and elongation at 72°C for 1 (L11 and cytochrome c genes), 2 (small-subunit rRNA, rpoC1, and tubulin beta chain genes), 3 (gyrase A and gyrase B genes), or 4 (rpoB) minutes. The PCR products were subjected to electrophoresis in a 2% agarose gel after staining with ethidium bromide (Sigma-Aldrich, Japan).
Western blotting.To investigate the potential effect of thiostrepton on the Babesia proteasome, B. bovis cultures were treated with thiostrepton or the proteasome inhibitor epoxomicin (1) at the IC80 according to a previous protocol (5). DMSO was used as a negative control. RBCs were collected from cultures and were lysed with 0.15% saponin. Parasite pellets were washed with PBS, resuspended, and sonicated. Parasite proteins were separated by SDS-PAGE and transferred to a Western S membrane (Whatman) according to the manufacturer's protocol. Membranes were blocked for nonspecific binding by incubation in 5% nonfat milk in Tris-buffered saline with Tween 80 (TBST) at 4°C overnight. Blots were washed and incubated for 2 h with the primary antibody, mouse anti-β SU type 5 (1:200) (Santa Cruz), diluted in 0.5% nonfat milk. The blots were subsequently washed and incubated for 1 h with secondary goat anti-mouse antibody conjugated to horseradish peroxidase (Bethyl), diluted 1:2,000 in 0.5% nonfat milk in TBST. Membranes were washed and developed in diaminobenzidine (DAB).
In vivo growth inhibition assay.The in vivo growth inhibition assay for thiostrepton was performed twice with BALB/c mice by use of a previously described method (2, 64), with some modifications. Twenty 8-week-old female BALB/c mice were divided into 4 groups, each containing 5 mice, and inoculated intraperitoneally with 1 × 107 B. microti-infected RBCs. In the first group, thiostrepton was administered at a dose of 500 mg/kg of body weight after being dissolved in N,N-dimethylacetamide (DMA) in 0.5 ml PBS (58). DMA in 0.5 ml PBS was administered to the control group. Clindamycin (Dalacin; Upjohn Ltd.) at a dosage of 500 mg/kg was administered intraperitoneally in 0.5 ml PBS to the second experimental group (6). Diminazene aceturate (Ganaseg; Japan Cieba-Geigy Ltd.) was administrated subcutaneously at a dosage of 25 mg/kg in 0.15 ml double-distilled water to the third experimental group (4). When the infected mice showed approximately 1% parasitemia, mice in the experimental groups were administered daily injections from day 1 to day 5 postinoculation (p.i.). The levels of parasitemia in all mice were monitored daily until 20 days postinfection, by examination of stained thin blood smears prepared from venous tail blood. All animal experiments were conducted in accordance with the Standard Relating to the Care and Management of Experimental Animals set by the National Research Center for Protozoan Diseases, Obihiro University of Agriculture and Veterinary Medicine, Hokkaido, Japan (1, 14).
Statistical analysis.The differences in the percentages of parasitemia for the in vitro cultures and among groups for the in vivo studies were analyzed with JMP statistical software (SAS Institute Inc.), using independent Student's t test (1). P values of <0.05 and <0.01 were considered statistically significant for the in vitro and in vivo studies, respectively.
RESULTS
In vitro growth inhibition assay of cultures with 1% initial parasitemia.The growth of cultured parasites from an initial parasitemia of 1% was significantly inhibited (P < 0.05) by ciprofloxacin at 10 μM (B. bovis [see Fig. S1A in the supplemental material] and B. bigemina [see Fig. S1B]) or 5 μM (B. caballi [see Fig. S4A] and B. equi [see Fig. S4B]) and was significantly suppressed by ciprofloxacin at 5 μM (B. caballi), 10 μM (B. equi), 50 μM (B. bovis), or 100 μM (B. bigemina). Thiostrepton significantly (P < 0.05) inhibited the growth of the parasites from an initial parasitemia of 1% at 5 μM (B. bigemina [see Fig. S1D], B. caballi [see Fig. S4C], and B. equi [see Fig. S4D]) or 10 μM (B. bovis [see Fig. S1C]) and significantly suppressed growth at 5 μM (B. caballi), 25 μM (B. bigemina), 50 μM (B. bovis), or 100 μM (B. equi). The growth of rifampin-treated parasites from an initial parasitemia of 1% was significantly (P < 0.05) inhibited at 5 μM (B. caballi and B. equi [see Fig. S4E and F] and B. bovis [see Fig. S1E]) or 10 μM (B. bigemina [see Fig. S1F]) and significantly suppressed in the presence of 100 μM rifampin. Clindamycin significantly inhibited the growth of parasites at 5 μM (B. caballi and B. equi) or 100 μM (B. bovis and B. bigemina) and suppressed the growth of parasites at 100 μM (B. caballi) or 500 μM (B. bovis, B. equi, and B. bigemina) (see Fig. S7). The in vitro growth of the four Babesia species was significantly inhibited (P < 0.05) by 5 nM diminazene aceturate treatment. Complete suppression of diminazene aceturate-treated parasites was observed at a concentration of 1,000 nM, while a concentration of 50 nM was required to suppress the growth of B. caballi (data not shown).
Ciprofloxacin completely cleared the parasites at 100 μM, as early as the second (B. bigemina and B. caballi) or third (B. bovis and B. equi) day of drug treatment. Subsequent cultivation of the parasites without the drug for a 10-day period showed no regrowth at 5 μM for all parasites, except for B. bigemina (see Fig. S1 and S4 in the supplemental material). Complete clearance of the parasites was observed at 50 μM on the third day (B. caballi) and 100 μM on the fourth day (B. bovis, B. bigemina, and B. equi) of thiostrepton treatment. Subsequent cultivation of the parasites without the drug for a 10-day period showed no regrowth of the parasites at 5 μM (B. caballi), 10 μM (B. bigemina and B. equi), or 25 μM (B. bovis) (see Fig. S1 and S4). Parasites exposed to lower drug concentrations started to grow again, as shown by light microscopy. Complete clearance of the parasites was observed at a concentration of 100 μM rifampin, as early as the second (B. bovis), third (B. caballi), or fourth (B. bigemina) day of rifampin treatment. However, B. equi parasites were still detected at all concentrations during the period of treatment. Subsequent cultivation of the parasites without the drug for a 10-day period showed no regrowth of the parasites at 25 μM (B. bovis), 50 μM (B. bigemina), or 100 μM (B. equi and B. caballi) (see Fig. S1 and S4). Parasites exposed to lower drug concentrations started to grow again, as shown by light microscopy. Clindamycin completely cleared the parasites at a concentration of 500 μM, as early as the first (B. bovis) or third (B. bigemina and B. caballi) day of drug treatment (see Fig. S7), while in B. equi cultures, no clearance was observed during the 4 days of treatment. After removal of the drug, the parasites could not grow in subsequent cultures at 5 μM (B. bigemina and B. caballi) or 100 μM (B. bovis and B. equi) (see Fig. S7). There was no regrowth of the diminazene aceturate-treated parasites in the subsequent viability test at a concentration of 25 nM (B. caballi) or 1,000 nM (B. bovis, B. bigemina, and B. equi) (data not shown). The IC50s of ciprofloxacin, thiostrepton, rifampin, clindamycin, and diminazene for different Babesia species are shown in Table 1. The addition of only ethanol, methanol, or DMSO to the culture had no influence on growth. There were no differences in the effects of the drugs on B. bovis and B. caballi in the cultures with serum and the cultures with a serum-free GIT medium (data not shown).
IC50s of ciprofloxacin, thiostrepton, and rifampin for B. bovis, B. bigemina, B. caballi, and B. equi
The morphological changes observed in treated cultures and damaged parasites were compared with the morphology of intact Babesia parasites. The parasites in ciprofloxacin-treated cultures appeared degenerate (see Fig. S2B, S3D, S5C, and S6C in the supplemental material). Thiostrepton treatment resulted in the loss of typical parasitic shapes, and pycnosis occurred in all thiostrepton-treated cultures. Furthermore, large numbers of abnormally dividing forms were observed (see Fig. S2C, S3C, S5D, and S6D). In rifampin-treated cultures, degenerate parasites, parasites with obtuse angles between pairs of B. bigemina organisms, and larger numbers of abnormally dividing parasites in B. bovis cultures than in the untreated control cultures were seen with a light microscope (see Fig. S2D, S3B, S5B, and S6B). The inhibitors were not toxic to the bovine and equine erythrocytes, as erythrocytes pretreated with the highest concentration (100 μM) of the inhibitors for 3 h at 37°C and then washed 3 times with normal medium and used for cultivation of Babesia parasites for 72 h showed no differences in the pattern of parasites compared to nontreated erythrocytes (data not shown).
In vitro growth inhibition assay of B. bovis and B. bigemina cultures with 7 and 10% initial parasitemias.The three antibacterials inhibited the growth of Babesia species in cultures with 1% initial parasitemia; therefore, we attempted to use B. bovis and B. bigemina cultures with 7 and 10% initial parasitemias as an in vitro model of acute disease, during which parasitemia is high. The pattern of parasite growth inhibition by ciprofloxacin in cultures with an initial parasitemia of 10% was similar to that in cultures with an initial parasitemia of 1% for B. bovis (see Fig. S8A in the supplemental material), while for B. bigemina cultures with an initial parasitemia of 7%, the growth was significantly (P < 0.05) inhibited at 5 μM (see Fig. S8B). Complete clearance was observed on day 2 (B. bigemina) or day 3 (B. bovis) of treatment. The parasite could not regrow at 5 μM (B. bovis) or 25 μM (B. bigemina) in the subsequent cultivation after removal of ciprofloxacin. The pattern of parasite growth inhibition by thiostrepton in cultures with an initial parasitemia of 7% was similar to that in cultures with an initial parasitemia of 1% for B. bigemina (see Fig. S8D), while for B. bovis cultures with an initial parasitemia of 10%, the growth inhibition was significantly (P < 0.05) inhibited at 5 μM (see Fig. S8C). Complete clearance was observed on day 3. Parasite growth was inhibited at 25 μM after the removal of thiostrepton. With an initial parasitemia of 10%, parasite growth inhibition by rifampin was significant (P < 0.05) at 5 and 50 μM for B. bovis and B. bigemina, respectively (see Fig. S8E and F). Rifampin completely cleared the parasites from cultures of B. bovis on day 2, while it could not clear B. bigemina from the cultures during the 3 days of treatment. Parasite growth after the removal of rifampin was inhibited in the subsequent cultivation, at 25 μM for B. bovis and 50 μM for B. bigemina.
Parasite stage inhibition and delayed-death effect.To investigate the parasite stages affected by the antibacterials, B. bovis cultures were treated with ciprofloxacin, thiostrepton, rifampin, clindamycin, and diminazene at their IC80s for 48 h. The numbers of parasites in different stages were counted at seven time points between 0 h and 48 h. The percentage of parasites in each stage was calculated for 50 parasites. The stages quantified were trophozoites, dividing forms, and single-pear-shaped, double-pear-shaped, and dot-shaped merozoites. The parasitemia percentage was calculated at each time point. The trophozoite-stage percentages at different time points for thiostrepton-, ciprofloxacin-, rifampin-, and diminazene-treated parasites were lower than those for the control and clindamycin-treated parasites (Fig. 1). The percentages of dot-shaped merozoites at 6 h in thiostrepton- and diminazene-treated cultures were significantly higher than those for the other drugs. Furthermore, the parasitemias of ciprofloxacin-, diminazene-, thiostrepton-, and rifampin-treated cultures were lower than those of clindamycin-treated cultures and the DMSO control. Ciprofloxacin and diminazene were faster acting than the other drugs, and the parasitemia percentages with these drugs, i.e., 1.4% and 2.4%, respectively, were lower than the control level of 7.7%. On the other hand, clindamycin-treated cultures had a parasitemia of 7%, which was similar to that of the control. To investigate the delayed-death effect, B. bovis cultures were treated at the IC50s of the drugs or with a 0.5% DMSO negative control in the same volume of culture medium. Parasite stages were quantified every 24 h for a period of 120 h. Parasitemia percentages were calculated at each time point. The percentages of trophozoite-stage parasites were very low compared with that of the control in cultures treated with ciprofloxacin, diminazene, thiostrepton, and rifampin, while the percentage of trophozoite-stage parasites in clindamycin-treated cultures was similar to that in the DMSO-treated control (Fig. 2). The percentage of dot-shaped parasites was significantly higher in ciprofloxacin-treated cultures at 24 h than in the control cultures and cultures treated with the other drugs. This percentage increased with time, reaching 100% at 72 h and remaining there until 120 h (Fig. 1). The percentage of dot-shaped parasites was also high in diminazene-, thiostrepton-, and rifampin-treated cultures and reached a peak at 120 h. Dot-shaped parasites were observed at 24 h and 72 h and reached a peak at 96 h in the control and clindamycin-treated cultures, but the percentage was higher for clindamycin-treated cultures than for the control. The parasitemia percentages were very low for ciprofloxacin (1%)- and diminazene (1.6%)-treated cultures, followed by thiostrepton (6.3%)- and rifampin (6.2%)-treated cultures, compared with the control (27%), while the parasitemia percentage in clindamycin-treated cultures reached a peak (8.2%) at 96 h and decreased to 2.5% at 120 h. The calculated IC50s for the inhibitors at 48 h were 4.7, 14.4, 14, 224.14, and 0.26 μM for ciprofloxacin, thiostrepton, rifampin, clindamycin, and diminazene, respectively (Table 2). The IC50s at 96 h were 4, 8.16, 8.6, 102.1, and 0.126 μM for ciprofloxacin, thiostrepton, rifampin, clindamycin, and diminazene, respectively. There was little difference between the IC50s at 48 h and those at 96 h for ciprofloxacin, thiostrepton, and rifampin, while the IC50s of diminazene and clindamycin decreased 1-fold (Table 2). The IC50s at 120 h were slightly decreased compared to those at 96 h and were 4.2, 7.42, 7.87, 69.42, and 0.12 μM for ciprofloxacin, thiostrepton, rifampin, clindamycin, and diminazene, respectively (Table 2). Cultures of B. bovis and B. equi were treated with ciprofloxacin, thiostrepton, rifampin, diminazene, and clindamycin at their IC99s for 24 h. In B. bovis cultures treated with ciprofloxacin, the IC99 (16.4 μM) resulted in 0.3% of the control growth at 48 h, and complete death of the parasites was observed within 72 h of treatment (Fig. 3A). Thiostrepton (22.8 μM)- and rifampin (23.8 μM)-treated B. bovis showed 35% and 28% of control growth, respectively, at 48 h, 26.5% and 24% of control growth at 72 h, and 23.3% and 31.2% of control growth at 96 h (Fig. 3A). The diminazene IC99 (0.67 μM) resulted in 25% of control growth at 48 h, 18% of control growth at 72 h, and 20% of control growth at 96 h. On the other hand, the clindamycin IC99 (250.7 μM) resulted in 103.1% of control growth at 48 h, 94.9% of control growth at 72 h, and 45% of control growth at 96 h of treatment (Fig. 3A). In B. equi cultures treated with ciprofloxacin, the IC99 (about 4.95 μM) resulted in 10.4% of control growth at 48 h, 3.2% of control growth at 72 h, and complete death at 96 h (Fig. 3B). Thiostrepton (12.67 μM)- and rifampin (8.1 μM)-treated B. equi had 23% and 27.2% of control growth, respectively, at 48 h, 20.8% and 25.9% of control growth at 72 h, and 12.6% and 12.9% of control growth at 96 h (see Fig. 9B in the supplemental material). The diminazene IC99 (1.2 μM) resulted in 52% of control growth at 48 h, 46% of control growth at 72 h, and 32% of control growth at 96 h. On the other hand, the clindamycin IC99 (53.8 μM) resulted in 100.3% of control growth at 48 h, 95.4% of control growth at 72 h, and 55% of control growth at 96 h of treatment (Fig. 3B).
Stage of growth inhibition of B. bovis during 48 h of compound treatment. Compounds at their IC80s or 0.5% DMSO was added to the parasites. Giemsa-stained blood smears were prepared at seven time points between 0 and 48 h of compound incubation, and the numbers of trophozoites, dividing forms, single-pear-shaped merozoites, double-pear-shaped merozoites, and dot-shaped merozoites were counted. Histograms indicate the percentages of developmental-stage parasites present in the respective blood smears. The respective parasitemia (P%) is indicated above each column. Fifty parasites were counted for each condition.
Delayed-death effect on B. bovis. For the delayed-death test, compounds at their IC50s or 0.5% DMSO was added to the parasites. Giemsa-stained blood smears were made at 0, 24, 48, 72, 96, and 120 h of incubation with the inhibitors, and the developmental stages were quantified as described in the legend for Fig. 1. The respective parasitemia (P%) is indicated above each column. A total of 50 parasites were counted for each condition.
Antibabesial activities of compounds over time (delayed-death test)a
Drug kinetics of ciprofloxacin, thiostrepton, rifampin, diminazene aceturate, and clindamycin on in vitro cultures of Babesia bovis and Babesia equi. (A) Babesia bovis; (B) Babesia equi. Parasitemia was calculated as a percentage of the control growth.
Reverse transcription-PCR.Reverse transcription-PCR showed that treatment of cultured B. bovis with thiostrepton at the IC99 for 8 h inhibited the transcription of the ribosomal L11 protein (Fig. 4A, lane 2), mitochondrial cytochrome c (Fig. 4A, lane 8), rpoB (Fig. 4A, lane 10), and rpoC1 (Fig. 4A, lane 12) genes. Rifampin treatment resulted in inhibition of the transcription of the rpoB gene (Fig. 4B, lane 1). Ciprofloxacin treatment for 8 h resulted in inhibition of transcription of the mRNAs of DNA gyrase subunit A (Fig. 4C, lane 2) and DNA gyrase subunit B (Fig. 4A, lane 8). Clindamycin treatment inhibited the transcription of the ribosomal L11 protein gene (Fig. 4D, lane 2) but could not inhibit the transcription of the rpoB and rpoC1 genes (Fig. 4D, lanes 4 and 6). The transcription of the small-subunit rRNA gene was not affected by treatment with thiostrepton (Fig. 4A, lanes 4 and 10), rifampin (Fig. 4B, lane 3), ciprofloxacin (Fig. 4A, lanes 4 and 10), or clindamycin (Fig. 4D, lane 8). The transcription of the B. bovis tubulin beta chain gene was not affected by treatment with thiostrepton (Fig. 4A, lanes 6 and 16), rifampin (Fig. 4B, lane 5), ciprofloxacin (Fig. 4A, lanes 6 and 12), or clindamycin (Fig. 4D, lane 10) any more than it was in the control cultures.
Reverse transcription-PCR analysis of DNA gyrase subunit A, DNA gyrase subunit B, ribosomal L11 protein, mitochondrial cytochrome c, rpoB, rpoC1, small-subunit rRNA, and tubulin beta chain genes from B. bovis cultures treated with ciprofloxacin, thiostrepton, rifampin, and clindamycin at their IC99s for 8 h. (A) Thiostrepton treatment. Lane 1, L11 gene from control culture; lane 2, L11 gene from treated culture; lane 7, mitochondrial cytochrome c gene from control culture; lane 8, mitochondrial cytochrome c gene from treated culture; lane 9, rpoB from control culture; lane 10, rpoB from treated culture; lane 11, rpoC1 from control culture; lane 12, rpoC1 from treated culture; lanes 3 and 13, small-subunit rRNA gene from control culture; lanes 4 and 14, small-subunit rRNA gene from treated culture; lanes 5 and 15, tubulin beta chain gene from control culture; lanes 6 and 16, tubulin beta chain gene from treated culture. (B) Rifampin treatment. Lane 1, rpoB from treated culture; lane 2, rpoB from control culture; lane 3, small-subunit rRNA gene from treated culture; lane 4, small-subunit rRNA gene from control culture; lane 5, tubulin beta chain gene from control culture; lane 6, tubulin beta chain gene from treated culture. (C) Ciprofloxacin treatment. Lane 1, gyrase A gene from control culture; lane 2, gyrase A gene from treated culture; lane 7, gyrase B gene from control culture; lane 8, gyrase B gene from treated culture; lanes 3 and 9, small-subunit rRNA gene from control culture; lanes 4 and 10, small-subunit rRNA gene from treated culture; lanes 5 and 11, tubulin beta chain gene from control culture; lanes 6 and 12, tubulin beta chain gene from treated culture. (D) Clindamycin treatment. Lane 1, ribosomal L11 protein gene from control culture; lane 2, ribosomal L11 protein gene from treated culture; lane 3, rpoB from control culture; lane 4, rpoB from treated culture; lane 5, rpoC1 from control culture; lane 6, rpoC1 from treated culture; lane 7, small-subunit rRNA gene from control culture; lane 8, small-subunit rRNA gene from treated culture; lane 9, tubulin beta chain gene from control culture; lane 10, tubulin beta chain gene from treated culture. M, molecular size marker.
Western blotting.We investigated the effect of thiostrepton on the Babesia proteasome by monitoring ubiquitinated proteins. B. bovis cultures were treated with thiostrepton and the proteasome inhibitor epoxomicin at their IC80s for 6 h and then harvested, and lysates were screened by Western blotting using mouse anti-ubiquitin antibody. Treatment with epoxomicin was used as the positive control in the assay. Western blotting revealed an accumulation of ubiquitinated proteins in parasites treated with epoxomicin (Fig. 5B, lane 2) and thiostrepton (Fig. 5B, lane 3) compared to the results for the 0.5% DMSO-treated control (Fig. 5B, lane 1). Coomassie blue staining of protein gels was used to show equal loading (Fig. 5A).
Accumulation of ubiquitinated blood-stage proteins following compound treatment. B. bovis was incubated for 6 h with 0.5% DMSO (lanes 1) or with the IC80 of epoxomicin (lanes 2) or thiostrepton (lanes 3). Parasite extracts were separated by polyacrylamide gel electrophoresis and screened via Western blot analysis. (A) Coomassie blue staining of protein gels was used to demonstrate equal loading of parasite protein extracts. (B) Ubiquitinated parasite proteins were detected with a mouse antibody against ubiquitin.
Effect of thiostrepton on B. microti infection in vivo.Thiostrepton had inhibitory effects on Babesia parasites in in vitro cultures; it was therefore used for the treatment of B. microti in the mouse model. In the thiostrepton-treated group, the levels of parasitemia increased at a significantly lower rate than that of the control group (P < 0.01) from days 3 to 9 p.i. Peak parasitemia levels in the treated groups reached averages of 6.2% in the presence of 25 mg/kg diminazene aceturate at 5 days p.i., 10.5% in the presence of 500 mg/kg thiostrepton at 7 days p.i., and 14.7% in the presence of 500 mg/kg clindamycin at 7 days p.i., relative to 46.8% in the control group (DMA) at 7 days p.i. (Fig. 6).
Inhibitory effects of thiostrepton given intraperitoneally at 500 mg/kg, clindamycin given intraperitoneally at 500 mg/kg, and diminazene aceturate given subcutaneously at 25 mg/kg on the in vivo growth of Babesia microti for observations of 5 mice per experimental group. Each value represents the mean ± standard deviation for 2 experiments. Asterisks indicate a significant difference (Student's t test; P < 0.01) from days 3 to 9 postinoculation between the thiostrepton-treated and DMA control groups.
DISCUSSION
In the present study, the inhibitory effects of ciprofloxacin, thiostrepton, and rifampin on the in vitro growth of B. bovis, B. bigemina, B. equi, and B. caballi were revealed. The presence of higher concentrations of the drugs in cultures completely suppressed the growth of the parasites tested in this study. Since the presence of solvents (ethanol, DMSO, and methanol) did not affect the growth of the parasites, the growth inhibition observed in this study was due to the effects of the drugs used. B. bovis and B. bigemina were most sensitive to ciprofloxacin, followed by thiostrepton and rifampin, while B. caballi and B. equi were sensitive to thiostrepton and ciprofloxacin, followed by rifampin, but clindamycin was the least effective for all four parasites. The IC50s of clindamycin were very high compared with those of the three tested antibacterials for bovine Babesia and B. equi, while a similar range to that of rifampin was observed for B. caballi. The inhibition was significant at the lowest concentration for the three drugs used in cultures with initial parasitemias of 7 and 10% for B. bovis and B. bigemina. Ciprofloxacin and thiostrepton were more effective than rifampin in the cultures with a high initial parasitemia.
The IC50s of the three apicoplast inhibitors for Babesia parasites were lower than those of other drugs used in previous studies, including ketoconazole (13), gossypol (15), heparin (14), EGTA (52), clindamycin phosphate, and chloroquine diphosphate (45). The IC50s for Babesia parasites were in a similar range to those for other drugs tested as babesicidal drugs, including triclosan (11), clotrimazole (12, 13), tetracyclines (45, 51), staurosporine (18), purvalanol A (49), N-acetyl-leucinyl-leucinyl-norleucinal (ALLN) (53), (−)-epigallocatechin-3-gallate (EGCG) (2), and nerolidol (3). The IC50s of the three inhibitors were higher than the IC50s of other babesicidal drugs, including quinuronium sulfate (21), imidocarb dipropionate (19, 59), clindamycin phosphate (19), atovaquone (45, 55), and epoxomicin (1). The IC50s of clindamycin phosphate reported in the present study were in a similar range to those for B. gibsoni (44). The IC50s of the antibacterials for Babesia parasites were higher than those of diminazene aceturate reported in this study.
The IC50s of ciprofloxacin for Babesia parasites were lower than the IC50s of 50 μM for P. falciparum (24) and 30 μM for T. gondii (29). Babesia parasites appeared to be more sensitive to ciprofloxacin than P. falciparum and T. gondii. The IC50s of thiostrepton for B. bovis and B. bigemina were in a similar range to the IC50 of 8.9 μM for P. falciparum (5), and they were higher than the IC50 of 2 μM for P. falciparum reported in another study (46). The IC50s for B. equi and B. caballi were similar to that for P. falciparum. The IC50s of rifampin for B. bovis and B. bigemina were higher than the IC50 of 3 μM for both P. falciparum (63) and T. gondii (57), and the IC50s for B. equi and B. caballi were in a similar range to those for P. falciparum and T. gondii. Bovine Babesia parasites were less sensitive to rifampin than P. falciparum and T. gondii. The IC50s of clindamycin were higher than those for P. falciparum (11.6 nM) (33) and T. gondii (20 nM) (26). The IC50 of ciprofloxacin is 129 μM for human fibroblast cells (31) and >188.5 μM for the mammalian Vero cell line (62); therefore, the IC50s of ciprofloxacin for Babesia parasites are not toxic to mammalian cells. The IC50 of rifampin is >75.9 μM for the mammalian Vero cell line (62); therefore, rifampin IC50s for Babesia parasites are not toxic to mammalian cells. The IC50 of thiostrepton is 27.8 μM for HeLa cells (5) and about 28 μM for several tumor cell lines (50); therefore, the selectivity index is 2.42, which is >1, indicating that the drug is more active than toxic at a concentration equivalent to the IC50 for parasite growth.
The Babesia erythrocytic life cycle starts after infection with a merozoite. The intraerythrocytic merozoite grows and rounds up to form a trophozoite. The trophozoite divides to form 2 merozoites by simple binary fission. The merozoites are released from the cells after its rupture. The time from the entrance to the release of merozoites is about 6 h (49). We observed dot-shaped merozoites under normal culture conditions when merozoites were overgrown or were going to die, while in the drug treatment experiments, dot-shaped merozoites were indicative of the degeneration and death of merozoites. Based on the quantification of parasites in the different developmental stages, the delayed-death test, and in vitro drug kinetics, the DNA gyrase inhibitor ciprofloxacin showed an immediate effect against Babesia parasites, and this effect is consistent with the immediate lethal effect in P. falciparum (28) and in contrast to the delayed-death effect in T. gondii (25). These effects demonstrate a fundamental difference between parasite responses to drug inhibition. The kinetics of growth inhibition resulting from treatment with the RNA polymerase inhibitor rifampin was similar to that seen with ciprofloxacin, as were those of the translation inhibitors thiostrepton and diminazene, but ciprofloxacin was more potent. In contrast, clindamycin, the drug known to cause delayed-death effects in P. falciparum and T. gondii, caused an immediate effect on B. bovis, but the effect was less potent than that of ciprofloxacin. Differences in the effects of antibacterials are attributed to several factors, such as target accessibility, which depends on the physical characteristics of the cells and organelles. Differences in host cell and lifestyle and in developmental processes may also be reflected in differing drug responses (28).
Reverse transcription-PCR showed that treatment of B. bovis for 8 h with ciprofloxacin at the IC99 inhibited the transcription of mRNAs of the nucleus-encoded DNA gyrase subunits A and B. Therefore, ciprofloxacin inhibited not only the DNA gyrase, which is responsible for apicoplast replication, but also the transcription of its nuclear genes. This result is in agreement with those of Xie et al. (75), who reported the interaction of ciprofloxacin with cytochrome enzymes (mitochondrially targeted) and the suppression of relevant nuclear cytochrome P450 (cytochromes P2C11 and P3A1) genes at the transcriptional level. Effects of ciprofloxacin on mRNA expression of some genes may be due to the regulatory effects of these genes on some transcription factors, for instance, ciprofloxacin decreased the nucleus-encoded interleukin-6 (IL-6) mRNA expression level in the EAhy926 endothelial cell line, and this was accompanied by a decreased level and binding of nuclear factor IL-6 (NF-IL-6) (27). On the other hand, ciprofloxacin increased nuclear IL-2 mRNA gene transcription, mediated through enhanced levels of the nuclear factor of activated T cells (NFAT-1) and, to a lesser extent, activator protein 1 (AP-1) in human peripheral blood lymphocytes (58). Rifampin treatment inhibited the transcription of mRNA of the rpoB gene (DNA-directed RNA polymerase gene). Thiostrepton inhibited the transcription of the ribosomal L11 protein gene (thiostrepton interaction site) and the rpoB and rpoC1 genes after treatment for 8 h, in agreement with the work of McConkey and colleagues (46). In addition, thiostrepton inhibited the transcription of mitochondrial cytochrome c; therefore, thiostrepton affects mitochondrial translation, in agreement with the work of Tarr et al. (65), who showed that thiostrepton affected mitochondrial translation in P. falciparum. Clindamycin inhibited the transcription of the ribosomal L11 protein gene, indicating that the B. bovis apicoplast is the target of clindamycin. On the other hand, it could not inhibit the transcription of rpoB and rpoC1 after the same period. The mRNA transcripts of the B. bovis tubulin beta chain and small-subunit rRNA genes were not affected by the treatment. Ciprofloxacin inhibited the DNA gyrase in B. bovis as well as in P. falciparum (43). Rifampin inhibited the transcription of rpoB (DNA-directed RNA polymerase beta subunit gene) in Babesia parasites as well as in P. falciparum (46). The inhibition of transcription of the rpoB and rpoC1 genes by thiostrepton in Babesia parasites indicates an inhibition of protein synthesis, the process targeted by thiostrepton, which is similar to its effect on P. falciparum (46). The Western blot results showed that thiostrepton targets the B. bovis proteasome subunits, and this may explain its immediate lethal effect. This effect of thiostrepton was reported previously for P. falciparum by another research group (5).
Thiostrepton was chosen due to its inhibitory effects in vitro in the present study and in vitro and in vivo on malaria parasites (46, 60, 64); thus, we were encouraged to evaluate the in vivo inhibitory effects of thiostrepton on B. microti, the rodent Babesia parasite, which is also known to be infective to humans, in a mouse model. The inhibitory effect of thiostrepton on the growth of B. microti was evident. The difference in growth inhibition between the control and drug-treated groups was significant on days 3 to 9 p.i. B. microti-infected mice that were treated with 500 mg/kg of thiostrepton did not show signs of toxicity and were alive during and after the experiment, in good agreement with the findings of Sullivan et al. (64), who showed that thiostrepton at 500 mg/kg potently inhibited P. falciparum in a mouse model, without toxic side effects. Furthermore, the dose used is four times lower than the 50% lethal dose (LD50) of thiostrepton for mice by the same route of administration (2,000 mg/kg) (10). In the current study, thiostrepton at 500 mg/kg effectively suppressed B. microti in mice treated for 5 days, with 77.5% inhibition on day 7 p.i. Clindamycin treatment for 5 days at 500 mg/kg resulted in 68.5% inhibition, and diminazene aceturate treatment for 5 days at 25 mg/kg resulted in 86.7% inhibition. Epoxomicin effectively suppressed B. microti in mice treated with 0.05 and 0.5 mg/kg for 10 days, with inhibition of 36.3 and 47.6%, respectively (1). EGCG effectively suppressed B. microti in mice treated for 10 days at 5 and 10 mg/kg, with 83.3 and 84% inhibition, respectively, on day 11 p.i. (2). B. microti was suppressed in mice treated with nerolidol for 10 days at 10 and 100 mg/kg, with 29.2 and 53.7% inhibition, respectively, on day 10 p.i. (3). On the other hand, mice treated with 2.5 mg heparin for 10 days had 86% inhibition on day 8 p.i. (14). In the hamster model, atovaquone at doses of 300, 150, and 80 mg/kg/day for 8 days was effective in the recovery of all animals; on the other hand, 50% of those receiving 10 mg/kg/day (37) recovered, and cases of recrudescence occurred at the dosage of 100 mg/kg/day (73). Clindamycin plus quinine was used at 150 mg/kg for 7 days, and the inhibition was 70% on day 7 posttreatment (44). Azithromycin, alone (150 and 300 mg/kg) or in combination (150 mg/kg) with quinine (250 mg/kg), was effective at reducing the levels of parasitemia during the period of drug administration, but parasitemia rapidly returned to the pretreatment levels when the drug was stopped (71). In the gerbil model, atovaquone at 50 mg/kg for 10 days resulted in a rapid reduction of parasitemia but did not eliminate the infection (30). Further research is also required to evaluate the in vivo inhibitory effects of rifampin and ciprofloxacin on B. microti in a mouse model.
In conclusion, the antibacterials targeting the apicoplast inhibited the growth of Babesia species in in vitro cultures, with an immediate lethal effect, and thiostrepton inhibited the in vivo growth of B. microti in BALB/c mice. Furthermore, thiostrepton affected the proteasomes and mitochondria of B. bovis. The present study implicates the apicoplast as a potential chemotherapeutic target for the development of drugs for babesiosis.
ACKNOWLEDGMENTS
This study was supported by the Japan Society for the Promotion of Science (JSPS). Mahmoud AbouLaila was supported by a research grant fellowship for young scientists from the JSPS.
FOOTNOTES
- Received 9 August 2011.
- Returned for modification 27 October 2011.
- Accepted 26 February 2012.
- Accepted manuscript posted online 5 March 2012.
Supplemental material for this article may be found at http://dx.doi.org/10.1128/AAC.05488-11.
- Copyright © 2012, American Society for Microbiology. All Rights Reserved.
