Optimization of a Fluorescence-Based Assay for Large-Scale Drug Screening against Babesia and Theileria Parasites

A rapid and accurate assay for evaluating antibabesial drugs on a large scale is required for the discovery of novel chemotherapeutic agents against Babesia parasites. In the current study, we evaluated the usefulness of a fluorescence-based assay for determining the efficacies of antibabesial compounds against bovine and equine hemoparasites in in vitro cultures. Three different hematocrits (HCTs; 2.5%, 5%, and 10%) were used without daily replacement of the medium. The results of a high-throughput screening assay revealed that the best HCT was 2.5% for bovine Babesia parasites and 5% for equine Babesia and Theileria parasites. The IC50 values of diminazene aceturate obtained by fluorescence and microscopy did not differ significantly. Likewise, the IC50 values of luteolin, pyronaridine tetraphosphate, nimbolide, gedunin, and enoxacin did not differ between the two methods. In conclusion, our fluorescence-based assay uses low HCT and does not require daily replacement of culture medium, making it highly suitable for in vitro large-scale drug screening against Babesia and Theileria parasites that infect cattle and horses.


Introduction
Babesiosis is a tick-transmitted disease caused by hemoprotozoan parasites and known to bring great economic losses in the bovine and equine industries worldwide. Bovine babesiosis caused by B. bovis and B. bigemina have a considerable impact on cattle health and productivity [1]. Equine piroplasmosis, caused by T. equi [2] and B. caballi [3], is considered one of the most important protozoan diseases affecting horses, mules, and donkeys [3]. Clinical manifestations of babesiosis include fever, hemolytic anemia, hemoglobinuria, and sometimes death [1,4]. To date, chemotherapy by imidocarb dipropionate and diminazene aceturate is the most common strategy for controlling infection in the field [5]. However, concerns about the toxicity and resistance of these drugs have emerged [5]. Therefore, there is an urgent need to develop more effective and safer antibabesial drugs.
Microscopic examination of stained blood smears is the gold standard for direct detection of parasites. However, this method is influenced by the quality of the blood smears as well as the skill and experience of experimentators and is not suitable for large-scale drug screening [6]. A new, alternative strategy that offers accuracy, simplicity, and automatic analysis is absolutely necessary [7]. To that end, a fluorescence-based technique using SYBR Green I (SGI) has been proposed as it provides more reliable results in a short time without being influenced by experimentator variations. In fact, this assay has been employed to evaluate the efficacies of antimalarials in vitro [8]. In brief, the assay relies on the high throughput screening of red blood cells to detect parasite DNA using a fluorescent spectrophotometer [8]. High-throughput screening (HTS) assay is a well-established process for drug discovery in biotech companies [9]. During the last 2 decades, there has been a dramatic increase in the number of available compounds, leading to a fundamental change in the drug discovery process applied in research units. Strikingly, HTS assay has the ability to test 10,000 to 100,000 compounds per day [9]. Therefore, this assay is one suitable approach to mass screening of compounds.
Although we have recently developed a novel fluorescence-based assay in our laboratory for drug evaluations against B. bovis [10], the assay is not suitable for large-scale screening of drugs due to the need for daily replacement of the medium, which is time consuming and requires great effort. Moreover, the previous assay has been shown to lower coefficients of variation at the maximal signal (% CV max ) and the minimal signal (% CV min ) percentages in comparison with the malaria parasites [11]. Therefore, in the present study, we optimized the assay to be suitable for large-scale screening of drugs and evaluated its usefulness in monitoring the in vitro growth of B. bovis, B. bigemina, T. equi, and B. caballi treated with diminazene aceturate, luteolin, pyronaridine tetraphosphate, nimbolide, gedunin, or enoxacin. The study revealed the promising use of our optimized fluorescence based assay for high throughput screening of compounds against in vitro growth of Babesia and Theileria parasites.

Ethics statement
Experiments described in this article were conducted in accordance with the Guiding Principles for the Care and Use of Research Animals promulgated by Obihiro University of Agriculture and Veterinary Medicine, Japan. The protocol was approved by the Committee on the Ethics of Animal Experiments of the Obihiro University of Agriculture and Veterinary Medicine (Permit numbers [26][27].

Assessment of SYBR Green I fluorescence linearity
A 96-well plate was used to assess the linearity between the fluorescent values and parasitemia values as determined by microscopy. Briefly, B. bovis, B. bigemina, T. equi, and B. caballi pRBCs were serially diluted with non-parasitized bovine or equine RBCs to parasitemia ranging from 0.25% to 8% in 100 uL [18]. Non-parasitized bovine or equine RBCs were used as a blank control. Bovine RBCs were prepared in M199 medium; equine RBCs were prepared in M199 medium for T. equi and RPMI 1640 medium for B. caballi in triplicate at three different hematocrit (HCT) (2.5%, 5%, and 10%). One hundred μl of a lysis buffer was mixed with a 2× SGI (10,000x) nucleic acid stain and added directly to each dilution by gentle mixing. Plates were incubated for 24 hours in a dark place at room temperature, and fluorescence values determined using a fluorescence plate reader (Fluoroskan Ascent, Thermo Labsystems, USA) at 485 nm and 518 nm excitation and emission wavelengths, respectively. Gain values were set to 100. The Parasitemia values were plotted against RFU values, after substraction of non-parasitized bovine or equine RBCs, and analyzed by linear regression.
Determination of statistical parameters for high-throughput screening (HTS) assays B. bovis, B. bigemina, T. equi, and B. caballi pRBCs at 1% parasitemia were cultivated in 96-well plates with non-parasitized bovine or equine RBCs at three different HCT percentages (2.5%, 5%, and 10%). Culture medium was M199 or RPMI 1640 alone (positive growth, 100% growth) or supplemented with a supralethal dose (20 uM) of diminazene aceturate (negative control). Then, the cultures of pRBCs were incubated without daily replacement of the medium, and fluorescence values were determined after the fourth day as previously mentioned. Nine samples were used for each medium in triplicate.
Optimization of the fluorescence-based assay B. bovis and T. equi pRBCs at 1% parasitemia were cultivated in 96-well plates at different HCT values (2.5%, 5%, and 10%) in M199 medium alone or with the indicated concentrations of diminazene aceturate (10, 250, 500, 1000, 5000, and 10000 nM) in 100 μl final volume. Non-parasitized RBCs were loaded into each well in triplicate and used as a blank control. RBC cultures were maintained for 4 days without replacement of the medium. On the fourth day, thin blood smears from each well were prepared and stained with Giemsa stain to be observed under the microscope to calculate parasitemia. Next, 100 μl of a lysis buffer containing 2× SGI nucleic acid stain was directly added to each well by gentle mixing. The plates were then incubated for 24 hours in a dark place at room temperature, and the fluorescence values were determined using a fluorescence plate reader as previously mentioned. Gain values were set to 100. The mean fluorescence values (after background subtraction of uninfected bovine and equine RBCs) were then plotted against the logarithm of drug concentrations. The 50% inhibitory concentration (IC 50 ) values were calculated on the fourth day of culture using both the fluorescence-based assay and microscopic examination. In a separate experiment, two 96-well plates were used for the cultivation of B. bovis and T. equi pRBCs at 1% parasitemia at 2.5% and 5% HCTs, respectively. The pRBCs were cultivated for 4 days with replacement of the medium in triplicate wells for each concentration of the drug. On the fourth day, IC 50 values of diminazene aceturate were determined as reported above. Each experiment was repeated three times.
Antibabesial drug screening by fluorescence-based assay B. bovis, B. bigemina, T. equi, and B. caballi pRBCs were cultivated at 1% parasitemia in 96-well plates using 2.5% and 5% HCTs for bovine and equine parasites, respectively, and M199 or RPMI 1640 media alone or with the indicated concentrations of drugs, including diminazene aceturate, luteolin, pyronaridine tetraphosphate, nimbolide, gedunin, and enoxacin. The pRBCs were cultivated for 4 days without replacement of the medium in triplicate wells for each concentration of the drug. Four plates were used in quadruplicate for each drug experiment in bovine Babesia parasites and in diminazene aceturate only for equine Babesia and Theileria parasites, and growth inhibition assays [20] between 1 and 4 days post-drug addition were conducted to evaluate whether the drug-induced growth inhibition is reflected by fluorescence signals in a dose-dependent manner. With other drugs for equine Babesia and Theileria parasites, only one plate was used in each drug experiment for 50% inhibitory concentration (IC 50 ) value calculation on the fourth day of culture after adding lysis buffer containing 2× SGI. The IC 50 values were calculated on the fourth day of culture using both the fluorescencebased assay and microscopic examination. Each experiment was repeated three times.

Statistical analysis
Data analysis was performed using GraphPad Prism ver. 5 (GraphPad Software, Inc., USA) using the one-way ANOVA. The IC 50 values were calculated using GraphPad Prism ver. 5. The mean IC 50 values by fluorescence assay and microscopic examination for each tested drug were analyzed using an unpaired t-test with the GraphPad software available online (http://www.graphpad.com/quickcalcs/ttest1.cfm?Format=SD). P < 0.05 was considered to be statistically significant.

Assessment of SYBR Green I fluorescence linearity
The correlation between the fluorescence values and microscopy values was assessed in this assay using three different HCTs (2.5%, 5%, and 10%). Correlating relative fluorescence units (RFUs) with parasitemias for bovine Babesia parasites showed significant linear relationships with R 2 values of 0.9945, 0.9328, and 0.9930 for B. bovis and R 2 values of 0.9842, 0.9637, and 0.9387 for B. bigemina with HCTs of 2.5%, 5%, and 10%, respectively (Fig 1A and 1B). For equine Babesia and Theileria parasites, significant linear relationships between the fluorescence and parasitemia values were only seen for 5% and 10% HCTs with R2 values of 0.9764 and 0.9946 for T. equi and R 2 values of 0.9786 and 0.9315 for B. caballi (Fig 1C and 1D). In summary, strong correlations (R 2 > 0.97) were obtained at 2.5% and 10% HCTs for B. bovis, 2.5% HCT for B. bigemina, 5% and 10% HCTs for T. equi, and 5% HCT for B. caballi.

Validation of modified high-throughput screening (HTS) assays
To evaluate the quality of the assay, the Z' factor, S/N ratio, % CV max , and % CV min were calculated for each of the three HCTs (2.5%, 5%, and 10%). For bovine Babesia parasites, the Z' factors were similar for all HCTs within a range of 0.60 to 0.97 (S1 Table). On the other hand, S/N ratios showed the highest values at 2.5% HCT (S1 Table). Regarding the equine parasites, the Z' factor values were greater than 0.5 at 5% HCT but lower than 0.5 for the other two HCTs (S1 Table). In addition, S/N ratios had the highest values at 5% HCT for both parasites. These results reveal the robustness and accuracy of the fluorescence-based assay with 2.5% and 5% HCTs without daily replacement of the medium for bovine and equine parasites, respectively. Optimization of the fluorescence-based assay Optimization of the fluorescence-based assay required determination of the IC 50 values of diminazene aceturate for B. bovis and T. equi by fluorescence-and microscopy-based methods in cultured parasites after 4 days of treatment without daily replacement of the medium at the optimal HCTs (2.5% and 5% for B. bovis and T. equi, respectively) in three separate experiments. The results obtained by the two assays at 2.5% and 5% HCTs, for B. bovis and T. equi, respectively were not significantly different (Table 1). Next, to exclude the death of parasites due to the non-daily replacement of the medium, the same experiment was applied with 2.5% HCT for B. bovis and 5% HCT for T. equi with daily replacement of the medium, and the IC 50 value was calculated by the fluorescence-based method. The results revealed no significant difference between the IC 50 values obtained from the experiment with daily replacement of the medium and that without daily replacement of the medium ( Table 2; Fig 2A and 2B). The absolute values obtained from the fluorescence-based assay after background subtraction of uninfected bovine RBCs for both cultures of B. bovis are also reported (S4A Fig). These results revealed the usefulness of the fluorescence-based assay for mass drug screenings in in vitro cultures of bovine Babesia parasites and equine Babesia and Theileria parasites with 2.5% and 5% HCTs, respectively, without daily replacement of the medium.

Fluorescence-based assay for evaluation of the in vitro growth inhibitory effects
We assessed parasitemia using the fluorescence based assay in day 1 to day 4 RBC cultures infected with four hemoparasites and exposed to six drugs (Figs 3 and 4; S1 and S2 Figs). Data showed that fluorescence values were inversely correlated with drug concentrations.   The maximum parasitemias of all parasites in this study in the control cultures were observed on the fourth day. Therefore, the fluorescence values of the experiments were determined on the fourth day of treatment.
The IC 50 values for the six drugs commonly used in the field, such as diminazene aceturate, luteolin, pyronaridine tetraphosphate, nimbolide, gedunin, and enoxacin were determined and compared to those obtained by the microscopy method. These drugs were used with the indicated concentrations: 5, 10, 100, 250, 500, 1000, 2500, 5000, 10000, and 15000 nM for diminazene aceturate; 10, 250, 500, 1000, 5000, 10000, 25000, 50000, 100000, 200000, and 400000 nM for luteolin, pyronaridine tetraphosphate, nimbolide, and gedunin; and 10, 250, 500, 1000, 5000, 10000, 25000, 50000, and 100000 nM for enoxacin. Strikingly, the IC 50 values obtained from the current assay were significantly concordant with those obtained by microscopy (P > 0.05) in all tested drugs for the four parasites (Table 3; Fig 5). Additionally, the absolute values obtained from the fluorescence-based assay after background subtraction of uninfected bovine RBCs and that obtained by the microscopy method for the culture of B. bovis without daily replacement of the medium are also reported (S4B and S4C Fig). These results confirmed the successful application of our fluorescence-based assay for mass drug screenings against bovine and equine Babesia and Theileria parasites in in vitro cultures.  The differences between the IC 50 values calculated using fluorescence-and microscopy-based methods were not statistically significant for all six drugs analyzed.
To determine the best effective drug against the in vitro growth of bovine and equine hemoparasites and examine whether IC 50 values of compounds differ from each other, ranking of drug potencies were assessed. The IC 50 of the diminazene aceturate determined by the fluorescence-based assay of each parasite was used as a control. The obtained IC 50 s from all treated groups showed different levels of statistical significant differences (P < 0.05) in comparison with diminazene aceturate treated group in vitro cultures of four parasites (Table 3). For B. bovis, luteolin and pyronaridine tetraphosphate treated groups showed the lowest statistical significant difference, followed by nimbolide, gedunin and enoxacin (Table 3). For B. bigemina, luteolin and nimbolide treated groups showed the lowest statistical difference in comparison with diminazene aceturate treated group followed by pyronaridine tetraphosphate, gedunin and enoxacin  (Table 3). While, luteolin and gedunin showed the lowest statistical significant difference in comparison with diminazene aceturate treated group in vitro culture of T. equi, followed by pyronaridine tetraphosphate, nimbolide and enoxacin (Table 3). Finally, for B. caballi, pyronaridine tetraphosphate, nimbolide and enoxacin showed very statistical significant difference in comparison with diminazene aceturate treated group, while, other treated groups, exhibited an extremely significant difference (Table 3). These results suggest that luteolin has the best activity against the in vitro growth of bovine Babesia and T. equi parasites next to diminazene aceturate. While, for B. caballi pyronaridine tetraphosphate showed the best activity after diminazene aceturate. Other drugs including, nimbolide, gedunin and enoxacin were effective but with lower activity against the in vitro growth of bovine and equine hemoparasites.

Discussion
In recent years, antibabesial drugs commonly used in field, such as diminazene aceturate and imidocarb dipropionate [21], have shown some drawbacks, including toxicity and resistance [5]. Therefore, developing safer antibabesial medicaments merits priority in veterinary treatment research. However, several studies regarding the mode of action and target of new candidates that require more time, effort, and money should be applied. Therefore, the quest for future babesiosis drugs should be shifted to the in vitro screening of large-scale chemicals either commercially available or non-commercially available compounds. Then, the compounds that show potent antibabesial effects in vitro should be examined for their inhibitory effects in vivo before application in the field. In the current study, we attempted to establish an assay using SGI stain suitable for mass drug screening in vitro for B. bovis, B. bigemina, B. caballi, and T. equi.
The fluorescence-based assay has been approved for in vitro evaluation and drug screening for certain hemoprotozoan parasites [8]. The assay depends on DNA detection using SGI stain. This stain is an asymmetrical cyanine dye that has been used as a substitute for ethidium bromide in molecular biology for several years. Its mode of action depends on the interaction with double-stranded DNA, preferring G and C base pairs [22]. This interaction yields a highly fluorescent, absorbing light at a wavelength between 390 and 505 nm, with a peak at 497 nm and a secondary peak near 254 nm. It emits light at 505 to 615 nm, with a peak at 520 nm [http:// www.sigmaaldrich.com/sigma-aldrich/datasheet/s9430dat.pdf]. The assay quality is determined by statistical parameters including the Z'-factor, S/N ratio, % CV max , and % CV min [19,23,24]. The Z'-factor represents the separation of the distributions between the positive and negative controls, and its value should be ! 0.5 [19]. Low values of the Z' factor indicate that the assay format is not effective for generating useful data. However, higher values of the S/N ratio and % CV min revealed the strength and accuracy of the assay in generating reliable data [19].
In the current study, we evaluated the usefulness of our modified fluorescence-based assay for mass drug screening for Babesia and Theileria parasites. The modifications, including the final volume of the medium, hematocrit, and the non-daily replacement of the medium, improved the performance of the assay and made it more practical for the large-scale screening of drugs. The results of the HTS assay for B. bovis with 2.5% HCT generated S/N ratios 1-10 and 12-203 times higher than those obtained from our previous laboratory assay [10] and those reported in malaria research [11], respectively. Meanwhile, the S/N ratios obtained from the assay with 5% HCT for equine Babesia and Theileria parasites were 6-16 times higher than those obtained in malaria research [11].
Furthermore, the IC 50 values obtained from the fluorescence assay without daily replacement of medium at 2.5% and 5% HCTs for bovine Babesia and equine Babesia and Theileria parasites, respectively, for diminazene aceturate, luteolin, pyronaridine tetraphosphate, nimbolide, gedunin, and enoxacin were similar to those obtained by microscopy in our study and those obtained in previous reports [10,20,25]. Taken together, the consistency in IC 50 values calculated by these two methods highlights the robustness of the current assay as an accurate, simple, and rapid method for mass drug screening in an in vitro culture.
The drugs luteolin, pyronaridine tetraphosphate, and enoxacin had not yet been tested against Babesia and Theileria species but have been previously documented to be effective against the growth of Plasmodium falciparum [26,27,28]. Nimbolide was previously used only against B. bovis [10], but its inhibitory effect against other Babesia and Theileria parasites had not yet been evaluated. The IC 50 values of nimbolide for B. bovis and T. equi parasites were higher than the value calculated for P. falciparum 2000 nM (0.95 microgram/ml) [29]. In contrast, the IC 50 values of nimbolide for B. bigemina and B. caballi parasites were, respectively, similar to and lower than that calculated for P. falciparum [29]. Luteolin is one of the flavonoids present in fruits, vegetables, wine, tea, and coffee [30]. In addition, this plant product has antioxidant, antitumor, anti-inflammatory, antimicrobial, antiviral, and antiprotozoal activities against Toxoplasma [31], Trypanosoma [32], Leishmania [33], and malaria [26]. The IC 50 values of luteolin against the growth of 3D7 and 7G8 P. falciparum have been reported to be 11000 ± 1000 nM and 12000 ± 1000 nM, respectively [26]. These values appeared to be higher than those for bovine and equine Babesia and equine Theileria parasites determined by the current assay, which might indicate the susceptibility of Babesia parasites to this drug. Pyronaridine tetraphosphate is a Chinese drug that has a strong inhibitory effect against chloroquinesensitive and-resistant strains of P. falciparum [34,35]. The IC 50 value of pyronaridine has been reported to be 2 nM against the growth of KT1 and KT3 P. falciparum [27], which is lower than the values obtained for Babesia and equine Theileria parasites in the current study. Enoxacin is one of the fluoroquinolones, a potent DNA gyrase inhibitor that had the lowest IC 50 values against FCC1 and VNS strains of P. falciparum [28,36]. The IC 50 values of enoxacin for Babesia and Theileria parasites were significantly higher than the IC 50 values for P. falciparum, which ranged from 2300 to 4400 nM [28,37]. Finally, the obtained IC 50 values of diminazene aceturate for B. caballi were lower than and nearly similar to those previously calculated by microscopy [38,20], respectively.
In summary, our optimized fluorescence-based assay with 2.5% and 5% HCTs without daily replacement of the medium for bovine and equine Babesia and Theileria parasites, respectively, offers a new approach for accurate, simple, and rapid detection of Babesia and Theileria parasites and large-scale screening of antibabesial drugs in an in vitro culture. In addition, luteolin and pyronaridine tetraphosphate drugs exhibited the best growth inhibition of bovine and equine hemoparasites next to diminazene aceturate, fellowed by nimbolide, gedunin and enoxacin. These drugs might be more effective if used as part of a combination therapy rather than a single therapy. Further studies are required for analyzing the synergistic or an antagonistic effect of these drugs when used in combination with each other, and to determine the best effective composition ratio for the growth inhibition of bovine and equine hemoparasites for clinical application. bovis (x-axis) and log concentration of diminazene aceturate (nM) (y-axis) by the fluorescence-based method without daily replacement of the medium (black) and with daily replacement of the medium (gray). (B) Growth inhibition of B. bovis by diminazene aceturate on the fourth day estimated by the fluorescencebased method. (C) Growth inhibition of B. bovis by diminazene aceturate on the fourth day estimated by the microscope-based method. Each value is presented as the mean of three triplicate wells ± SD after subtraction of the background fluorescence for non-parasitized RBCs. (PDF) S1 Table. Statistical parameters for determining the quality of the high-throughput screening (HTS) assay in B. bovis, B. bigemina, T. equi, and B. caballi parasites with different percentages of HCTs. S/N ratio = Signal to noise, % CV max = coefficient of variation at the maximum signal and % CV min = coefficient of variation at the minimum signal. (DOCX)