Babesia divergens strain 2210A G2 was cultivated in vitro in the suspension of commercially available bovine erythrocytes (BioTrading) under constant conditions of 37°C and 5% CO₂ by a previously described procedure (Jalovecka et al., 2018). The cultivation media was composed of RPMI 1640 (Lonza), amphotericin B (Merck, c=250µg/ml), gentamicin sulfate (Merck, c=10mg/ml) and supplemented with heat-inactivated fetal bovine serum (Capricorn) in 20% volume. Culture growth was monitored on thin blood smears stained by Diff-Quik (Siemens, Germany) under the BX53F light microscope (Olympus) at a magnification of 1000×.

To introduce a large-scale parasitemia monitoring method, we employed nuclear staining followed by a flow cytometry analysis. Infected red blood cells were pelleted, washed with 1× phosphate buffered saline (PBS), and fixed in a solution of 4% paraformaldehyde and 0.025% glutaraldehyde in PBS for 30 min at room temperature (RT). The fixed cells were subsequently washed twice with PBS (600×g, 3 min) and stored at 4°C for up to three weeks. To analyze parasitemia, the cells were stained with 0.02mM Ethidium Homodimer 1 (EthD-1, Biotinum) diluted in PBS for 30 min at RT. Afterwards, the stained samples were washed twice (600×g, 3 min) with PBS and analyzed using a FACS CantoII flow cytometer and Diva software provided by BD Biosciences, including the determination of Median Fluorescence Intensity (MFI) values for the GFP signal. When needed, flow cytometry was utilized to determine the GFP signal from transgenic parasites. Fixed parasites samples stained with EthD-1 were incubated with an Anti-GFP Polyclonal Antibody conjugated with Alexa Fluor™ 488 (Thermofisher), diluted 100× in PBS for 30 min. The samples were then washed twice with PBS (600×g, 3 min) and analyzed in the same manner as stated above.

The inhibitory effect of two standard selection drugs, WR99210 (selection marker hdhfr - human dihydrofolate reductase - gene, kindly provided by Jacobus Pharmaceutical) and blasticidin S (BSD, selection marker blasticidin S deaminase gene, Merck) was tested in B. divergens in vitro system. Inhibitory assays were conducted in 96-well plate format in biological triplicates for 8 consecutive days. Media supplemented either with WR99210 (concentration ranging from 20 nM to 0.019 nM) or BSD (concentration ranging from 16 µg/ml to 0.0156 µg/ml) were exchanged in an interval of 48 hours, DMSO diluted in cultivation media served as solvent control. Culture samples were collected at two-day intervals, fixed, and analyzed using flow cytometry as stated in chapter 2.1.

To construct B. divergens-specific plasmids for both episomal and intragenomic gfp expression, we first identified, PCR amplified, and sequenced B. divergens-specific untranslated regions (UTRs): the 5’ UTR of the actin gene (Bdiv_007890; length: 2002 bp), the 5’ UTR of the elongation factor tu gtp binding domain family protein (ef-tgtp) gene (Bdiv_030590; length: 696 bp), and the 3’ UTR of the chloroquine resistance transporter (CRT) gene (Bdiv_036760; length: 1496 bp). The coding sequences of gfp, hdhfr, and bsd genes were obtained from B. bovis-specific plasmids (Asada et al., 2015). The target locus for intragenomic insertion of the GFP-expressing cassette, the B. divergens 6-cys-e gene (6-cysteine (E), Bdiv_004560c), and its 5’ and 3’ UTRs, were PCR amplified and sequenced. The final plasmid included two homology regions (HRs): HR1 (length: 817 bp) spanning the 5’ UTR and the beginning of the gene CDS, and HR2 (length: 802 bp) spanning the end of the gene CDS and the 3’ UTR to allow homologous recombination.

The plasmids were assembled in several subsequent steps using NEBuilder® HiFi DNA Assembly Kit (New England BioLabs) following the manufacturer’s protocols. The procedure involved the linearization of the plasmid’s backbone using specific restriction enzymes and PCR amplification of the desired loci. Detailed information about individual plasmid parts and restriction sites are depicted in plasmid maps ( Supplementary Data ; https://doi.org/10.6084/m9.figshare.24433384.v1). After assembly, the plasmids were transformed into NEB 5-alpha Competent E. coli cells (New England Biolabs) and grown on LB agar plates supplemented with ampicillin, following the manufacturer’s protocols. Individual colonies obtained from the plates were grown in LB medium supplemented with ampicillin for 12-16 hours, and the plasmid DNA was extracted using the NucleoSpin® Plasmid kit (Macherey-Nagel) and verified by sequencing. To isolate and purify plasmid DNA on a large scale for transfections, the NucleoBond® Xtra Midi kit (Macherey-Nagel) was employed.

The plasmids designed for episomal GFP expression were transfected in circular form, as detailed in the subsequent chapter. The plasmid intended for intragenomic gfp expression was linearized using the FseI restriction enzyme before the transfection.

Three different protocols were employed for the transfection of B. divergens parasites: electroporation of infected RBCs (iRBCs), electroporation of free merozoites, and electroporation of uninfected bovine RBCs (uRBCs). All transfection protocols included a “plasmid solution”: 10 µg of the plasmid that was precipitated in ethanol, resuspended in 20 µl of filtered milliQ water, and mixed with 100 µl of nucleofector solution from Basic Parasite Nucleofector® Kit 2 (Lonza). Transfections were performed in a device ‘Amaxa’ Nucleofector 2b (Lonza) and an electric pulse using the V-024 program was applied to the mixture of plasmid solution and iRBCs/free merozoites/uRBCs solution. Transfected samples were then transferred into a mixture of 2 ml cultivation media and 100 µl of uRBCs/uRBCs/iRBCs, respectively. Twenty-four hours post-transfection, the medium was replaced, and the 5 nM WR99210 selection drug was added. The medium containing the drug was changed daily during the experiments.

Transfection of B. divergens iRBCs was adapted from the previously described protocol for B. bovis (Suarez et al., 2007; Asada et al., 2012b): 100 µl of iRBCs with ~15% parasitemia were washed in pre-warmed PBS and mixed with the plasmid solution. Transfection of free merozoites included 100 μl of iRBCs with ~15% parasitemia that were washed in pre-warmed PBS and filtered twice with a 1.2 µm Acrodisc® syringe filter unit (Pall corporation) as described in (Hakimi et al., 2021a). The filtered suspension was then centrifuged (600×g, 3 min) to remove the residues of RBCs, and free merozoites resuspended in 100 μl of PBS were mixed with the plasmid solution. The transfection of uRBCs (also referred to as DNA or plasmid pre-loading of RBCs) was adapted from P. falciparum protocol (Deitsch et al., 2001). Briefly, 100 μl of uRBCs were washed with PBS and mixed with the plasmid solution. After the transfection, the mixture of DNA-loaded uRBCs was mixed with 2 ml cultivation media supplemented with 100 µl of iRBCs. Three independent experiments were conducted to evaluate the efficiency of the above presented different transfection protocols.

To obtain clonal populations with correctly integrated plasmid, the parasites transfected with the linearized plasmid were subjected to serial dilutions: 0.3 parasites were seeded per well in a 96-well plate and grown for 10 days with medium exchanges every second day. Obtained clones were expanded for DNA extraction and the correct integration of the plasmid was confirmed by PCR using the primers listed in Supplementary Table 1 .

To analyze the growth pattern of GFP-expressing and knockout (6-cys-e gene disruption) parasites, their intraerythrocytic development was monitored for 96 hours and compared to a wild-type parasite lineage. All experiments were conducted in biological triplicates, and parasitemia was determined every 12 hours through blood smears and flow cytometry.

B. divergens gDNA was isolated from iRBCs (~10% parasitemia) using the NucleoSpin® Blood DNA isolation kit (Macherey-Nagel) according to the manufacturer’s instructions and stored at -20°C. PCR reactions were performed in the T100 Thermal Cycler (BioRad) following the protocol for Q5® High-Fidelity 2X Master Mix (New England BioLabs) and visualized on agarose gel using Gel logic 112 transilluminator (Merck).

Total RNA from B. divergens in vitro cultures was extracted using TRI Reagent® (Merck) followed by chloroform phase separation and isolation with the NucleoSpinRNA II kit (Macherey-Nagel). To eliminate gDNA residues, the RNA samples were treated with the TURBO DNA-free™ Kit (Invitrogen) and stored at -80°C. Reverse transcription was performed with 1 μg of RNA using the Transcriptor First Strand cDNA Synthesis Kit (Roche), following the manufacturer’s protocol, and stored at -20°C. Transcript levels of gfp and hdhfr genes were quantified in biological as well as technical triplicates using LightCycler® 480 SYBR Green I Master (Roche) with specific primers for each transcript ( Supplementary Table 1 ) in CFX96 Touch Real-Time PCR Detection System (BioRad). The relative gfp and hdhfr genes expression was normalized to the B. divergens gapdh gene (Bdiv_010720) using the ΔΔCt method (Jalovecka et al., 2016).

GFP-expressing B. divergens in vitro cultures were smeared, air-dried and fixed with 4% paraformaldehyde and 0.075% glutaraldehyde in PBS for 30 min at RT on Superfrost Plus™ Adhesion Microscope Slides (Thermo Fisher Scientific). After fixation, the samples were washed with PBS (3 × 5 min) and incubated with 0.01% Triton X-100 (Merck) diluted PBS for 30 min at RT to permeabilize cell membranes. Anti-GFP Polyclonal Antibody conjugated with Alexa Fluor™ 488 (Thermofisher) was diluted 100× in 0.01% Triton X-100 and applied to samples for 1 hour at RT. Subsequently, the samples were washed in PBS (3 × 5 min) and stained with 300 nM DAPI diluted in PBS for 10 min at RT. After another round of washing with PBS (3 × 5 minutes), the samples were mounted in DABCO (Merck). GFP signal was examined by BX53F fluorescence microscope (Olympus) and processed in Fiji (ImageJ) software.

The statistical analyses, IC50 calculations, and graphs were created in GraphPad Prism software (version 7.0a). The inhibitory effect of selection drugs was evaluated by one-way analysis of variance (ANOVA), Kolmogorov-Smirnov test, and the Bartlett test passed. The expression of gfp and hdhfr genes was compared with the non-parametric Mann-Whitney test. * = p < 0.05, ** = p < 0.01, *** = p < 0.001.

Since the establishment of stable transfection system requires a reliable drug selection, we conducted tests to assess the impact of two commonly used selection drugs, WR99210 and BSD, on the growth of B. divergens. Both drugs demonstrated inhibitory activity against B. divergens in our in vitro system. WR99210 was found to be effective at a concentration of 5 nM ( Figure 1A ), while BSD exhibited inhibitory effects at a concentration of 4 μg/ml (equivalent to 8.72 μM) ( Supplementary Figure 1A ).

At 5 nM, WR99210 significantly inhibited parasite growth by 2 days post infection (DPI), reducing parasitemia to 0.1% by 8 DPI compared to the initial 1% (0 DPI, Figure 1A ), whereas the non-treated culture (NC) grew up to a parasitemia of 25.35 ± 2.33% on 8 DPI ( Figure 1B ). Inhibitory effects were also observed at 10 and 20 nM concentrations, while concentrations of 1.25, 0.31, 0.07, and 0.02 nM did not exhibit significant growth inhibition. The IC₅₀ of WR99210 in B. divergens bovine RBCs, determined from parasitemia values measured on 8 DPI, was calculated 2.50 nM with an R² value of 0.89 ( Figure 1B ).

BSD at 4 μg/ml (8.72 μM) significantly decreased B. divergens growth: on 8 DPI, parasitemia decreased to 0.4% ± 0.1 from an initial 1% (0 DPI, Supplementary Figure 1A ), while the NC group showed 23.85% ± 2.48% parasitemia ( Supplementary Figure 1B ). BSD also exhibited inhibitory effects at concentrations of 8 and 16 μg/ml (17.43 and 34.86 μM, respectively), but no impact on growth was observed with concentrations of 1, 0.25, 0.0625, and 0.015 μg/ml (equivalent to 2.18, 0.54, 0.14 and 0.03 μM, respectively). The IC₅₀ of BSD in B. divergens in vitro system was calculated as 2.61 μg/ml (5.69 μM), with an R² value of 0.90, based on parasitemia values measured on 8 DPI ( Supplementary Figure 1B ).

With a reliable selection system, we further proceeded to develop a B. divergens GFP-expressing reporter line. We designed and de novo constructed a circular plasmid “act-gfp-ef-tgtp-hdhfr” containing the GFP reporter cassette, where the gfp gene expression was driven by the actin gene promoter. The hdhfr gene, conferring resistance to the drug WR99210, was under the control of the promoter of the Elongation Factor Tu GTP binding domain family protein (ef-tgtp) ( Figure 1C ; Supplementary Data ; https://doi.org/10.6084/m9.figshare.24433384.v1). The plasmid was delivered into B. divergens-infected bovine RBCs, and WR99210-resistant parasites emerged after 12 to 14 days post transfection (DPT). By 16 to 18 DPT, parasitemia reached 5%. Throughout this study, the GFP fluorescence signal was consistently detected ( Figure 1D ) under WR99210 pressure. No GFP signal was detected in wild-type parasites ( Figure 1E ).

Similarly, the BSD-resistant parasites appeared between 12 to 15 DPT after delivery of the modified plasmid in which the same promoter was used to drive the expression of the blasticidin S deaminase gene as in the above-mentioned validated hdhfr-WR99210 selection system. However, despite several attempts to stabilize the culture, the parasitemia remained low, and the parasites were no longer detectable after 24 DPT (data not shown). Due to the low parasitemia, the GFP signal could not be adequately confirmed. Consequently, in the following experiments, the hdhfr-WR99210 selection system was used.

Since the GFP-hdhfr expression plasmid was verified, we proceeded to optimize the transfection protocol, with the aim of establishing a versatile platform for B. divergens-specific transgenic techniques. To achieve this, we explored three different methods for delivering the circular plasmid “act-gfp-ef-tgtp-hdhfr” into parasite cells: the widely used and well-established direct iRBCs transfection ( Figure 2A ) (Suarez and McElwain, 2010; Asada et al., 2012b), the less common but validated method of electroporation of Babesia free merozoites ( Figure 2D ) (Suarez and McElwain, 2008), and lastly, the plasmid delivery into uRBCs before their infection ( Figure 2G ), followed by their coculturing with wild-type (WT) parasites. The latter method of erythrocytes pre-loading has not been previously validated in any Babesia-transgenic system, but in P. falciparum, it has demonstrated higher efficiency when compared to other methods (Deitsch et al., 2001; Skinner-Adams et al., 2003; Hasenkamp et al., 2012).

After comparing the three transfection methods ( Figure 2 ), the plasmid pre-loading into uRBCs proved to be the most efficient transfection protocol in B. divergens in vitro system ( Figures 2H, I ) in all three independent experiments. The observed GFP signal from parasites confirmed the spontaneous uptake of plasmid DNA from the bovine host cell cytoplasm into B. divergens intraerythrocytic stages. Additionally, parasitemia of nearly 3% of GFP-expressing parasites was detected 1 DPT ( Supplementary Figure 2C ) and gradually increased up to 17%. However, the parasites showed signs of stress due to overgrowth, likely associated with the decreased ratio of GFP expression to total parasitemia observed from 8 DPT. Thus, the cultures were diluted on 10 DPT, when the MFI of GFP signal was measured as 1411, to avoid stressful conditions that could interfere with parasite growth. Following dilution, the parasites quickly re-emerged with MFI value on 11DPT 1419, and parasitemia exceeded 3% by 13 DPT. Throughout the experiment, when comparing total parasitemia with the number of GFP-expressing parasites, the same trend of growth was observed. This confirms the successful transfection and plasmid maintenance by the parasites. Moreover, this technique demonstrated the highest viability compared to the other tested methods, as evidenced by a blood smear taken on 1 DPT ( Supplementary Figure 2F ). After the direct electroporation of iRBCs, a harmful effect on parasites was observed, as evidenced by the presence of multiple dead parasites on 1 DPT and a high rate of RBC lysis ( Supplementary Figures 2A, D ). Despite this, the GFP-expressing parasite population gradually increased over time, reaching over 5% parasitemia by 8 DPT, but remained at lower levels compared to the pre-loaded transfectants ( Figures 2D, E ). Signs of overgrowth were detected from 9 DPT, necessitating dilution of the culture on 10 DPT, when MFI reached 626. Following dilution, the parasite population re-emerged with detected MFI 826 on 11 DPT and reached a stable parasitemia with sustained GFP expression. In contrast, the electroporation of B. divergens free merozoites displayed delayed appearance of transgenic parasites ( Figures 2F, G ). Despite a few viable parasites being detected on 1 DPT ( Supplementary Figures 2B, E ), the first GFP-expressing parasites only emerged around 10 DPT. This delay may be attributed to the fragility of the free merozoite population, resulting in reduced merozoite viability after electroporation, or a low efficiency of the transfection protocol, or a combination of both factors. By 12 DPT, the parasitemia of transgenic parasites exceeded 7% with MFI value 967, and the observed majority of GFP-expressing parasites indicates successful transformation. We then proceeded with culture dilution to observe whether parasites would be able to re-emerge. Indeed, after the dilution, GFP-expressing parasites exponentially multiplied with MFI value 1726 detected on 13 DPT, indicating the continuous growth of a successfully transfected population.

In most Babesia species, the promoters controlling the expression of elongation factor-1 alpha (ef1-α) exhibit robust, constitutive, and bidirectional activity in transgenic applications (Liu et al., 2018b). Consequently, when introducing transgenic applications into our B. divergens system, we selected the promoter of Bdiv_030590, a gene annotated in PiroplasmaDB as “the translation elongation factor EF-1, subunit alpha protein, putative”. However, a more comprehensive in silico analysis of the Bdiv_030590 gene revealed distinct differences from the typical structure of well characterized ef1-α promoters in other Babesia species (Suarez et al., 2006; Silva et al., 2016; Liu et al., 2018b; Wang et al., 2022). Instead, the arrangement of Bdiv_030590 gene closely resembles the locus of syntenic gene BBOV_II000640, which codes for Elongation factor Tu GTP binding domain family protein (Dr. Carlos Suarez, personal communication). Therefore, we designated the promoter of Bdiv_030590 gene as the “ef-tgtp”. Based on its position between two genes (Bdiv_030590 and Bdiv_030580c) in opposite direction, we hypothesized that it might possess bidirectional activity, allowing us to reduce the size of the original plasmid “act-gfp-ef-tgtp-hdhfr” to enhance its versatility. By removing the actin promoter sequence from the original plasmid, we created a circular plasmid “gfp-ef-tgtp-hdhfr” which contained both the GFP reporter cassette and the hdhfr gene-based resistance cassette under the control of the ef-tgtp promoter in a head-to-head orientation ( Figure 3A ; Supplementary Data ; https://doi.org/10.6084/m9.figshare.24433384.v1). The stable GFP expression in episomally transfected parasites and their continuous growth under drug selection pressure confirmed a bidirectional activity of the ef-tgtp promoter ( Figures 3B, C ). Moreover, the promoter activity was symmetrical, as both transcripts of the gfp and hdhfr genes exhibited expression levels with no significant difference ( Figure 3D ), which makes this promoter a widely adaptable tool for other transgenic applications.

To complete the development of a B. divergens-specific transgenic platform, we proceeded with the direct insertion of a foreign gene into the parasite genome. In already established Babesia-transgenic systems, this process is commonly achieved through homologous recombination mechanisms (Suarez et al., 2017). To target a dispensable locus/gene for parasite blood stages, we chose the B. divergens 6-cys-e gene (Bdiv_004560c). The ortholog of this gene in B. bovis has been demonstrated to be dispensable for the intraerythrocytic life cycle of this parasite. To introduce the GFP reporter cassette into the selected locus, we designed the plasmid “6-cys-e-gfp-ef-tgtp-hdhfr” ( Supplementary Data ; https://doi.org/10.6084/m9.figshare.24433384.v1). This plasmid contains two homology regions (HRs) that are homologous to the 5’ and 3’ UTRs of 6-cys-e gene. Within these HRs, the GFP reporter cassette and the hdhfr gene-based resistance cassette were positioned in a head-to-head orientation, separated by the ef-tgtp promoter ( Figure 4A ). The pre-loading of the linearized plasmid into bovine uRBCs resulted in the emergence of WR99210-resistant parasites on 12 DPT that gradually increased, surpassing 5% parasitemia by 15 DPT. Following clonal selection, the successful intragenomic insertion was confirmed ( Figure 4B ), and GFP signal was detected ( Figure 4C ). Furthermore, the symmetrical activity of ef-tgtp bidirectional promoter was reconfirmed ( Figure 4E ). The continuous growth of the obtained clonal population under the drug pressure with no significant differences ( Figure 4D ) confirms that the expression of the 6-cys-e gene is not essential for B. divergens intraerythrocytic multiplication. The dispensability of the 6-cys-e gene makes it a suitable recipient locus for more advanced B. divergens-specific transgenic systems requiring target integration of the parasite genome. However, the inserted sequences should be driven by external and constitutive promoters to assure stable expression of the transgenes.