Targeting prolyl-tRNA synthetase via a series of ATP-mimetics to accelerate drug discovery against toxoplasmosis

The prolyl-tRNA synthetase (PRS) is a validated drug target for febrifugine and its synthetic analog halofuginone (HFG) against multiple apicomplexan parasites including Plasmodium falciparum and Toxoplasma gondii. Here, a novel ATP-mimetic centered on 1-(pyridin-4-yl) pyrrolidin-2-one (PPL) scaffold has been validated to bind to Toxoplasma gondii PRS and kill toxoplasma parasites. PPL series exhibited potent inhibition at the cellular (T. gondii parasites) and enzymatic (TgPRS) levels compared to the human counterparts. Cell-based chemical mutagenesis was employed to determine the mechanism of action via a forward genetic screen. Tg-resistant parasites were analyzed with wild-type strain by RNA-seq to identify mutations in the coding sequence conferring drug resistance by computational analysis of variants. DNA sequencing established two mutations, T477A and T592S, proximal to terminals of the PPL scaffold and not directly in the ATP, tRNA, or L-pro sites, as supported by the structural data from high-resolution crystal structures of drug-bound enzyme complexes. These data provide an avenue for structure-based activity enhancement of this chemical series as anti-infectives.

The prolyl-tRNA synthetase (PRS) is a validated target for antimalarial drug development [26][27][28][29][30][31]. It has also been structurally and biochemically dissected in context of halofuginone (HFG) and several quinazolinone-based inhibitors (QBI) [5,8,14,26,28]. HFG and QBIs act through a proline-competitive binding mode, with more effective binding to PRS with ATP [27]. Despite the high homology between the parasite and host PRSs and comparable biochemical sensitivity, HFG is significantly more potent against P. falciparum at the asexual blood stage (EC 50 of 1 ± 0.5 nM) than the mammalian cell lines (EC 50 of 150 ± 9 nM) [6]. However, clinical development of HFG has been held back due to side effects and a deleterious phenotypic drug resistance mechanism (within five generations) via the accumulation of L-proline (~20-fold upregulation)-called the Adaptive Proline Response (APR) [32].
In 2017, Takeda Pharmaceutical Company Limited disclosed a new class of HsPRS inhibitors to treat fibrosis [33]. Unlike HFG and analogs which span the A76 and proline-binding sites and thus interact in an ATP-uncompetitive manner, these new 1-(pyridin-4-yl) pyrrolidin-2-one (PPL) derivatives target the ATP-binding pocket. Some of these compounds, e.g., T-3767758, displayed proline-uncompetitive steady-state kinetics for HsPRS [29]. Based on this class of inhibitors, Okaniwa et al. identified PPL derivatives as a new class of antimalarials that bind to the PRS ATP-site with double-digit nanomolar activity against Plasmodium falciparum (Pf) and Plasmodium vivax (Pv) strains [31]. In this study, five compounds centered on the PPL scaffolds were investigated for their therapeutic efficacy in killing both the Toxoplasma parasites and the Toxoplasma encoded PRS enzyme (Fig 1). We provide cellular, enzymatic and structural data to validate PRS as the target of this new series of ATP-mimetics that hold promise as future agents against toxoplasmosis.

Compound synthesis and cellular assays
Compounds were synthesized (HPLC purity > 95%) according to reported procedures [31, [33][34][35][36]. A confluent Human Foreskin Fibroblast (HFF) monolayer was infected with 2,000 tachyzoites of RH parasites expressing the NLuc luciferase (RH NLuc) for 2 h to allow parasite invasion. Each compound was added to the culture medium, and infected cells were incubated at 37˚C for 48 h. To measure luminescence, the medium was replaced by 50 μL of PBS, and the luminescence assay was performed using Nano-Glo Luciferase Assay System according to the manufacturer's instructions (Promega). After 3 min of incubation, the luminescence was measured using the CLARIOstar (BMG Labtech) plate reader. EC 50 was determined using a non-linear regression analysis of normalized data assuming a sigmoidal dose-response. EC 50 values for each compound represent an average of three independent biological replicates. Statistical analyses were performed using a one-way ANOVA test with GraphPad software.

Measurement of CC 50 for human cells and determination of selectivity index
The medium cytotoxic concentrations for mammalian cells were determined as follows.

PLOS PATHOGENS
Targeting prolyl-tRNA synthetase in toxoplasmosis metabolic activity as opposed to dividing cells. It was chosen to be done so as ARPE-19 and MDA-231 are more relevant for testing drug biological activity due to their sensitivity to drug treatment, in contrast to HFFs-which are quite robust and would overestimate the selectivity indices.

Toxoplasma gondii random mutagenesis
Parasites were chemically mutagenized as previously described, with the following modifications [37]. Briefly,~10 7 tachyzoites (RH strain) growing intracellularly in HFF cells in a T25 flask were incubated at 37˚C for 4 h in 0.1% Fetal Bovine Serum (FBS) DMEM growth medium containing either 2.5 mM ethyl methane sulphonate (EMS) or the appropriate vehicle controls. After exposure to the mutagen, parasites were washed three times with 1XPBS. The mutagenized population was allowed to recover in a fresh T25 flask containing an HFF monolayer without the drug for 3-5 days. Released tachyzoites were then inoculated into fresh cell monolayers in a medium containing 100 nM L35 and incubated until viable extracellular tachyzoites emerged 8-10 days later. Surviving parasites were passaged once more under continued L35 treatment and cloned by limiting dilution. Four cloned mutants were isolated, each from 6 independent mutagenesis experiments. Thus, each flask contained unique SNV pools.

RNA-seq, sequence alignment and variant calling
For each biological assay, a T175 flask containing a confluent monolayer of HFF was infected with RH wild-type or L35-resistant strains. Total RNA was extracted and purified using TRIzol (Invitrogen, Carlsbad, CA, USA) and RNeasy Plus Mini Kit (Qiagen). RNA quantity and quality were measured using NanoDrop 2000 (Thermo Scientific). RNA sequencing was performed as previously described [37], following standard Illumina protocols, by GENEWIZ (South Plainfield, NJ, USA). Briefly, RNA quantity and integrity were determined using the Qubit Fluorometer and the Fragment Analyzer system with the PROSize 3.0 software (Agilent Technologies, Palo Alto, California, USA). The RQN ranged from 8.6 to 10 for all samples, which was considered sufficient. Illumina TruSEQ Stranded RNA library prep and sequencing reagents were used following the manufacturer's recommendations using polyA-selected transcripts (Illumina, San Diego, CA, USA). The samples were sequenced on the Illumina NovaSeq platform (2 x 150 bp, single index) and generated~20 million paired-end reads for each sample (S1 Table). The quality of the raw sequencing reads was assessed using FastQC [38] and MultiQC [39]. The RNA-Seq reads (FASTQ) were processed and analyzed using the Lasergene

PLOS PATHOGENS
Targeting prolyl-tRNA synthetase in toxoplasmosis

Protein purification
PRSs from TgPRS: length 334-830 (TGME49_219850 S8G8I1_TOXGM) and HsPRS: length 1015-1506 (EPRS (EARS, PARS, PIG32) Q3KQZ8_HUMAN) were purchased as a gBlock (Integrated DNA Technologies, Leuven, Belgium). The ORF of the TgPRS and HsPRS was optimized for expression in the E. coli BL-21 strain. The transformed E. coli BL-21 strain was grown in an LB medium containing 50 μg ml −1 kanamycin to an A 600 of 0.6-0.8 at 37˚C. Expression of both His6-tagged recombinant proteins was induced by adding 0.65 mM isopropyl β-D-thiogalactopyranoside (IPTG) to cells grown at 37˚C for 6 h, followed by incubation at 18˚C for 16-18 hr. Bacterial cells were lysed by sonication in buffer containing 50 mM Tris-HCl pH 8.0, 200 mM NaCl, 3 mM ßME, 15% v/v glycerol, 0.1 mg ml -1 lysozyme, and EDTA free protease inhibitor cocktail (Roche). The lysed cells were cleared by centrifugation at 20,000 g for 45 min, and the supernatant was affinity captured using nickel-nitrilotriacetic acid-agarose beads (Ni-NTA) (GE Healthcare), followed by elution with buffer containing 50 mM Tris-HCl pH 8.0, 200 mM NaCl, 10 mM ßME and 250 mM Imidazole. The eluted protein fractions were dialyzed against 30 mM HEPES pH 7.5, 20 mM NaCl, 1 mM DTT and 0.5 mM EDTA (buffer A). The protein was purified by heparin chromatography (GE Healthcare) using NaCl gradients with buffer B containing 30 mM HEPES pH 7.5, 500 mM NaCl, 1 mM DTT, and 0.5 mM EDTA. The protein peak was found at 40% buffer B. The 6xHis tag was removed by incubating with TEV protease at 20˚C for 24 h. The cleaved TgPRS and HsPRS proteins were concentrated with a 30-kDa cut-off Centricon centrifugal device (Millipore) followed by gel permeation chromatography on a Superdex 200 column 16/60 column (GE Healthcare) in a buffer containing 20 mM HEPES pH 7.5, 200 mM NaCl, and 2 mM DTT. Bovine serum albumin (66 kDa, Sigma) was used as a standard for molecular mass estimation. The eluted fractions were checked by SDS-PAGE, and the pure fractions were pooled, concentrated in 50 mM Tris-HCl pH 8.0, 200 mM NaCl, 10 mM ßMe and stored at -80˚C.

Thermal shift assays
Fluorescence-based thermal shift assays were performed to assess the binding potencies of the five 1-(pyridin-4-yl) pyrrolidin-2-one (PPL)-based derivatives for TgPRS and HsPRS in the presence or absence of substrates (L-Pro and ATP). Purified PRS enzymes in the presence and absence of their substrates and inhibitors were heated from 25 to 99˚C at a rate of 1˚C min -1 , and fluorescence signals of the SYPRO orange dye were monitored by a quantitative real-time PCR system (Life Technologies). Proteins were used at 1 μM, the drugs at 50 μM (for TgPRS) and 100 μM (for HsPRS), and the substrates at saturating concentrations of 2 mM. The melting temperature is an average of three measurements, and data were analyzed using Protein Thermal shift software (v1.3, Thermofisher). The inhibitors and substrates alone in assay buffers and no PRS enzyme controls were used, and flat lines were observed for these fluorescence readings across the temperatures. The derivative T m was used for analysis.

Enzyme inhibition assays
Aminoacylation inhibition assays were performed according to the previously published report [7,40]. Briefly, 25 μM ATP, 25 μM L-pro, and 400 nM recombinant purified protein (TgPRS and HsPRS) in a buffer containing 30 mM HEPES (pH 7.5), 150 mM NaCl, 30 mM KCl, 50 mM, MgCl 2 , 1 mM DTT and 2 U/ml E. coli inorganic pyrophosphatase (NEB) at 37˚C was used to perform the assay. The assay was carried out in a transparent, flat-bottomed, 96-well plate (Costar 96-well standard microplates) for 1.6 h at 37˚C. Malachite green solution was added to stop the enzymatic reaction. Absorbance was measured at 620 nm using a Spectramax M2 (Molecular Devices). The background controls were set up without the TgPRS and

PLOS PATHOGENS
Targeting prolyl-tRNA synthetase in toxoplasmosis HsPRS enzymes. The values obtained were deducted from the enzymatic reaction values. The five compounds were added to the aminoacylation assay reaction buffer in concentrations of 0.00005 to 50 μM. The IC 50 values for the data are shown for three replicates as the mean ± SD. All competitive assays were performed in the presence of both L-pro and ATP at 1.5X concentrations of their K m values.

Crystallization
Highly purified HsPRS (12-15 mg mL -1 ) and TgPRS (10-14 mg mL -1 ) enzymes were used for crystallization via the hanging-drop vapor-diffusion method at 20˚C using commercially available crystallization screens (Hampton Research and Molecular Dimensions). Compounds were initially in 100 mM stocks prepared in 100% DMSO-which were then added to the protein whilst crystallization with the final DMSO concentration being 2.5-7.5%. Initial screening was performed in 96-well plates using a nanodrop dispensing Mosquito robot (TTP Labtech).
Three different drop ratios of purified protein and reservoir (i.e., 1:1, 2:1, and 1:2 drop ratios) were used for the crystallization trials. Each drop was equilibrated against 100 μl of the corresponding reservoir solution. Before crystallization, 1 to 3 mM compounds and 2 mM L-pro were added to PRS enzymes, and the mixtures were incubated at 4˚C for 10 min. Diffractionquality crystals were obtained at 20˚C by the hanging-drop vapor diffusion method. The crystallization conditions for each enzyme-inhibitor complex are listed in S2 Table. Diffraction data collection and structure determination The X-ray diffraction data sets were collected on beamlines at Diamond Light Source (DLS), United Kingdom, PROXIMA 1 and PROXIMA 2A, SOLEIL, France. The data were processed by the auto-processing pipelines using DIALS and XDS for integration [41]. The initial models were determined by the molecular-replacement (MR) method using Phaser [42] and the available host (PDB ID: 4K86) and parasite PRS (PDB ID: 5XIF) enzyme structures as the template. The structures were further refined by iterative refinement cycles with Phenix [43] and model building with COOT [44]. Map interpretations and model building were based on electron densities in difference Fourier (F o − F c ) and 2F o − F c maps. In all stages, model building was guided by manual inspection of the model and R free . The substrate/inhibitor and water molecules were added based on the difference Fourier maps (F o − F c ). The occupancies of the ligand molecules were refined, and highly disordered loop regions were not included in the final model. The stereo-chemical quality of the models was assessed and corrected using MolProbity [45]. The summary of the collection and refined parameters are given in S3 and S4 Tables. The figures were prepared using Chimera [46] and PyMOL [47]. Structural interaction analyses were performed using the PLIP server [48].

Accession numbers
The atomic coordinates and structure factors for the TgPRS and HsPRS complex with inhibitor-AMPNP/L-pro are deposited in the RCSB Protein Data Bank (PDB), and the accession codes are listed in S2 Table. Results

T. gondii growth inhibition assays
1-(Pyridin-4-yl) pyrrolidin-2-one (PPL) derivatives are ATP mimetics that were designed as anti-fibrosis scaffolds that selectively target Plasmodium prolyl-tRNA synthetase (PRS) [29,31,49]. Efficient in vitro inhibition of T. gondii growth was repeatedly confirmed by  (Figs 2A and S1A). Among these five PPL derivatives, the EC 50 of L35 with 27 nM was 14X lower than pyrimethamine with an EC 50 of 396 nM-the standard care for toxoplasmosis. Therefore, we further investigated L35 to assess its efficacy. The growth of the type I RH strain was monitored using human foreskin fibroblasts (HFFs) treated with L35; pyrimethamine and vehicle (DMSO) were used as positive and negative controls respectively. Complete and sustained growth inhibition was observed at 100 nM of L35 with no adverse effects on host cells (Fig 2B). The selectivity index (SI) of the five PPL derivatives was determined using ARPE19 (epithelial cells) and MDA321 (breast cancer) cell lines (Figs 2C, 2D, S1D and S1E). Cells were incubated with increasing concentrations of drugs for 72 h. The viability of the human cell lines was determined to calculate the cells' cytotoxicity concentration, CC 50 . SI values presented in the tables are based on the CC 50 of the human cells divided by the EC 50 of T. gondii. The PPL derivative-L36, with the highest EC 50 value of 5800 nM, showed no inhibition for the human cell lines at 1000 nM (Figs 2A and S1A).

Target validation of the PPL derivatives via random T. gondii mutagenesis
We then investigated the mechanism of action of L35 as it displayed the lowest EC 50 at 27 nM. This was done via a forward genetic screen using chemical (ethyl methane sulphonate (EMS)) mutagenesis to isolate L35-resistant parasites. Resistant parasites were analyzed with the wildtype strain by RNA-seq to identify mutations in the coding sequence conferring drug resistance by computational analysis of the variants (Fig 2E). To map the EMS-induced mutations conferring L35 resistance, Illumina sequencing reads were aligned to the~65-Mb T. gondii GT1 reference genome. The assembled sequences were analyzed to identify single nucleotide variations (SNVs), small insertions, or short deletions using the parental strain as a reference, as described previously (Fig 2F) [37]. By focusing on mutations in coding sequences, we identified a single gene, PRS, that contained two SNVs, resulting in amino acid substitutions at the protein residues numbered 477 and 592 (S1 Table). The mutations T477A and T592S in the L35-resistant lines were absent in the parental strain. To confirm that the PRS mutations were sufficient to confer resistance to L35, we reconstructed each of the mutations identified in L35-resistant parasites into the susceptible parental wild-type strain using the CRISPR/Cas9 system coupled with homology-directed repair for gene editing in T. gondii (S2A Fig). After selection with L35, emerging resistant parasites were cloned, and DNA sequencing established that the mutations were correctly inserted into TgPRS (Fig 3A). In the genetically modified parasites, the PRS mutations T477A and T529S substantially reduced susceptibility to L35 compared to wild-type parasites (Figs 3B-3D and S2B and S5 Table). Notably, the plaque size of the T592S mutant was somewhat higher than that of the T477A mutant, however, upon treatment with L35, the trend was reversed (although the differences were non-significant). Furthermore, we would like to point out that the mutant T477A was~3X more resistant towards L35 (186 nM for T477A vs 76 nM for T592S) than the T592S mutant. However, both these mutants (T477A and T592S) are equivalently susceptible to pyrimethamine as wild-type parasites (as is expected by virtue of the mutual exclusion of their mechanisms of action). The activity of the other four compounds was tested against the PRS-edited parasites (Figs 4, S1B, and S1C and S5 Table). Although there are some differences between the EC 50 values for the five compounds in the L35-resistant parasites-T477A and T592S, the EC 50 values were higher than in the wild-type parasites confirming that these five PPL-derivatives are active against PRS (Fig 4A-4E and S5 Table). Moreover, TgPRS mutation T477A substantially reduced the sensitivity to L36 in the engineered compared to wild-type parasites (Fig 4D and S5 Table).

PLOS PATHOGENS
Targeting prolyl-tRNA synthetase in toxoplasmosis  50 . SI values, presented in the tables, are based on the human cells CC 50 divided by the T. gondii EC 50 . No inhibition was detected for L36 compound. n.d., not detected. Associated dose-response curves are showed in S1 Fig. E) The diagram shows the strategy for L35 target deconvolution. The workflow is based on forward genetic screen using chemically induced resistant mutant lines. EMS-mutagenized parasites were then selected with lethal concentrations of L35. The resistant parasites along with the wild-type strain were analyzed by RNA-seq for computational analysis of the variants to

PLOS PATHOGENS
Targeting prolyl-tRNA synthetase in toxoplasmosis Our group recently reported the simultaneous binding of L95 and HFG in complex with TgPRS covering all three of the enzyme-substrate subsites, with HFG occupying the 3'-end of tRNA and the L-pro binding sites, and L95 resident in the ATP pocket (Fig 4F). Structural data analysis for the available three-dimensional crystal structure of the complex TgPRS-L95 +HFG (PDB: 7EVU) revealed that the two L35-resistant mutations-T477A and T592S -are proximal to the two ends of the scaffold (Fig 4F). Interestingly, the mutation T592S made the parasite considerably more susceptible towards L36 (See S5 Table). Although the variant residues Tg-T477 and T592 were not directly in the ATP, tRNA, or L-pro site, these were located at the edge of the ATP pocket. This indicates that both the methyl-pyrazole and hydroxymethyl terminals may play critical roles in the selectivity of the PPL scaffolds. Based on these observations, we biochemically and structurally characterized each of these five PPL derivatives in TgPRS and HsPRS enzymes via enzyme-based aminoacylation inhibition assays and high-resolution crystal structures of drug-bound enzymes, respectively.

The five PPL-derivatives inhibit aminoacylation via competition with ATP
The potency of the five PPL derivatives against TgPRS and HsPRS was evaluated using enzyme-inhibitor binding and inhibition assays. We examined the first step of the aminoacylation reaction, i.e., the formation of the aminoacyl-adenylate complex, and measured the release of pyrophosphate (PPi) using a malachite green dye-based assay. The measured half-maximal inhibitory concentration (IC 50 ) values for each inhibitor L95, L96, L97, L35, and L36 are shown in Fig 5. The high potency of these inhibitors was evident from their nanomolar IC 50 values for TgPRS, i.e., L95: 139.9 nM, L96: 79.7 nM, L97: 50 nM, L35: 9.2 nM, and the exception of L36 with IC 50 of 2409 nM (Fig 5A and 5B). Interestingly, the ATP-mimetic PPL scaffolds were originally designed as antifibrosis scaffolds against HsPRS. However, three out of five PPL derivatives, i.e., L97, L96, and L35, were found to be selective against TgPRS as opposed to its human counterpart in vitro, with enzymatic selectivity indices of 27.9, 13.5, and 8.2, respectively (Fig 5C).
Enzyme-ligand interactions are critical for stabilizing the enzyme conformation during the thermal denaturation process and increasing the melting temperature (Tm). Our analysis of the thermal shift data for enzyme-inhibitor binding was consistent with the inhibition data. It showed that L97 and L35 were bound more efficiently to TgPRS in the presence of L-Pro compared with HsPRS (Fig 5D). For TgPRS, we observed ΔTm (˚C) values of 22, 23, 20, 26, and 14 for L95, L96, L97, L35, and L36, in the presence of L-Pro, respectively, compared to the apo-TgPRS. Similarly, the ΔTm (˚C) values for HsPRS were 19, 21, 16, 23, and 10 for L95, L96, L97, L35, and L36, in presence of L-pro, respectively, compared to the apo-HsPRS (Fig 5D). Notably, the five PPL-derivatives stabilized the PRSs more effectively in the presence of L-Pro, with L35 being the most stable (ΔTm of 26 and 23 (˚C) for Tg-and Hs-PRSs), suggesting a cooperative binding that assists the five inhibitors in binding the enzyme more efficiently in the presence of the L-amino acid, a natural substrate (Fig 5E). Next, co-crystallization was performed to elucidate the structural basis of activity of the five PPL derivatives and PRS interactions. identify the mutations on coding sequence conferring drug-resistance. F) Circos plot showing the single nucleotide variants (SNVs), insertions, and deletions detected by transcriptomic analysis of the T. gondii L35-resistant lines, grouped by chromosome (numbered in Roman numerals with size intervals given outside). Each dot in the 6 innermost gray tracks corresponds to a scatter plot of the mutations identified in the coding regions of the 6 drugresistant strains, with each ring representing one of the 6 drug-resistant lines (35-1 to 35-6). In the second outermost track, lines depicting whole-genome RNA-seq data of the T. gondii parental strain (RPKM values of genes are shown). Each bar in the outermost track represents locations of selected essential genes. (n.d-not detected). https://doi.org/10.1371/journal.ppat.1011124.g002

PLOS PATHOGENS
Targeting prolyl-tRNA synthetase in toxoplasmosis

PLOS PATHOGENS
Targeting prolyl-tRNA synthetase in toxoplasmosis

PLOS PATHOGENS
Targeting prolyl-tRNA synthetase in toxoplasmosis

Structural overview of PRS with the five PPL-derivatives
To decipher the structural basis of the selectivity of the inhibitors by the host (Hs) and parasite (Tg) PRSs, and to address the apparent higher affinity of the drug in the presence of L-pro, we attempted to co-crystallize the Tg-and Hs-PRSs with the five PPL derivatives. We successfully co-crystallized all five compounds with TgPRS (L95, L96, L97, L35, and L36) and three with HsPRS (L95, L96, and L97), all in complex with L-pro (S2 Table). Interestingly, co-crystals of HsPRS with these compounds were obtained in two forms, namely forms 1 and 2, in 0.1 M HEPES pH 7.5, 20% PEG 3350 and 0.5 or 1.5 M CaCl 2 (S2 Table). Form 1 crystals belonged to the monoclinic space group P2 1 (a = 71, b = 91, c = 83 Å and β = 110˚) with two molecules per asymmetric unit, and form 2 crystals belonged to the orthorhombic space group P2 1 Table) revealed that most structures are well ordered and have low mobility. The RMSD between the empty and L-pro bound TgPRS and with ligand-bound TgPRS is 0.6-0.8 Å and 0.2-0.5 Å, respectively. The overall folding of the host (Hs) and parasitic (Tg) PRSs was similar to the previously reported structures (PDB ID: 4K86, 5XIF). Minimal but significant conformational changes were observed for the active site residues and a shift in the active site loops. These are described and discussed in the following sections. The bound inhibitor molecules were verified using a difference Fourier electron density map at 3 σ levels (Fig 6). Because the RMSD between the present enzyme-inhibitor complex structures with known structures was <1 Å, we considered only chain A for all structural descriptions.

Toxoplasma gondii PRS (TgPRS)
The three-dimensional crystal structures of all the five PPL derivatives in the holo-TgPRSs were probed (Fig 7A). The five PPL derivatives were buried in a ligand-induced fit model between the ATP pocket (470-483) and the loop (528-538) (Fig 7B). Superposition of apo-(PDB: 5XIF) (in grey) and the holo-TgPRSs (in purple) and structural analysis revealed remarkable plasticity in the ATP pocket (470-483) of TgPRS to accommodate these PPL-derivatives in a groove forming the binding pocket that is composed of Arg470, Glu472, Lys474, Arg481, Thr482 (Fig 7C). The second half of the binding pocket consists of Phe485, Phe 534, Gln555, Thr592, and Arg594 (Fig 7D and 7E). The phenyl ring of Phe485 and the guanidinium moiety of Arg594 support the adenine equivalent of the 6-methylpyridine core via a π-π stacking interaction. The compounds are stabilized by hydrogen bonds via the O atom of the phenylacetamide group and the keto O atom of oxopyrrolidine interaction with a water molecule. Furthermore, in L35 and L95, the 6-methylpyridine N atom interacts with the mainchain N atom of Thr482, Benzyl-formamide (L96, L97, L35, L36) or phenylacetamide (L95) Natom interacts with the OG1 of side-chain Thr482. The oxopyrrolidine keto O atom interacts with the OG1 of side-chain Thr592 in L95, L96, and L97. Additionally, the O atom of response curves are shown in S1 Fig. E) The EC 50 (nM) values for the engineered PRS mutant strains (T477A and T592S) are shown in comparison with the TgPRS wild type. F) The catalytic cavity of TgPRS is shown with two inhibitors along with their electron densities-HFG (green) and L95 (red)-bound to the tRNA and L-pro binding sites and the ATP binding site respectively. The ribbon of PRS is in cyan whilst the resistance-conferring mutating residues T477 and T592 to L35 are shown in purple. (n.d.-not detected). https://doi.org/10.1371/journal.ppat.1011124.g004

PLOS PATHOGENS
Targeting prolyl-tRNA synthetase in toxoplasmosis

PLOS PATHOGENS
Targeting prolyl-tRNA synthetase in toxoplasmosis Hydroxymethyl attached to the pyrrolidine group interacts with OG1 of Thr558 and the side chain of Gln555 in L97 (S3 Fig). The superposition of holo-TgPRS revealed no striking steric or sidechain rotameric conformational differences among the five PPL derivatives bound structures of TgPRS complexes. In contrast to L35, L95, L96, and L97, compound L36 showed poor enzyme inhibition as indicated by an IC50 value in the μM range compared to the other four compounds analyzed in the current study (Fig 5A and 5C). This is attributed to preferential binding favoring the (S) enantiomer (L35) and not the (R) enantiomer (L36). Structurally, the stacking of 4-(3-Fluorophenyl)-1-methylpyrazole moiety allows rotameric conformational changes in Glu472, showed similar trends based on the melting temperature shifts. Thermal shift profiles for the TgPRS and HsPRS enzymes with the five PPL derivatives are shown. https://doi.org/10.1371/journal.ppat.1011124.g005

PLOS PATHOGENS
Targeting prolyl-tRNA synthetase in toxoplasmosis pyrrolidine and methyl-pyridine groups. Regardless, the 6-methylpyridine core is stacked between the phenyl ring of Phe485 and the guanidinium moiety of Arg594. Whereas the sig-nificant~90˚bent in oxo-pyrrolidine exposes its hydrophobic core to the active site, displacing the oxo-pyrrolidine keto O atom away from Thr592 and surrounding water molecules. Interestingly, the stereoisomeric distortions in L36 result in a slight lateral shift in the neighboring L-pro, preventing its OXT atom from forming a fourth H-bond with NE2 of His560. This absence of the fourth H-bond may lead to loose packing in the active site.
The active site residues between HsPRS and TgPRS are highly conserved with a few exceptions-His1157 (Gln475), Gln1159 (Thr477), Gly1238 (Ala556), and Gly1239 (Ala557). Based on the TgPRS enzyme data, L97 showed the highest selectivity index of 27.9 compared with the other four PPL derivatives. Structurally, in both TgPRS and HsPRS complex, L97 shared conserved interaction points except the H-bond between the O atom of hydroxymethyl attached to the pyrrolidine group with OG1 of Thr558, which is missing in HsPRS (S3 and S4 Figs). This is due to a slight displacement of the hydroxyl group and, in turn, a slight downward displacement of the entire cyclopropyl-oxo-pyrrolidine group, sufficient enough to overcome the hydrophobic environment resulting from the exposed cyclopropyl core. This allows the accommodation of a water molecule and the restoration of its H-bond with L-pro. In addition, the guanidinium moiety of Arg1163 is slightly displaced away from the pyrazole ring to form an H-bond with the OE1 atom of neighboring Gln1159 resulting in loose stacking of the entire 4-(3-Fluorophenyl)-1-methylpyrazole moiety in the ATP pocket. This is also evident in the rotameric conformations in Lys1156 and Phe1155. Furthermore, between the two variant residues, T477 is located at the edge of the ATP pocket, while the other variant residue T592 lies on the other end facing the oxo-pyrrolidine moiety. This observation indicates that both the methyl-pyrazole terminal and hydroxy-methyl end might play a crucial role in the selectivity of L35.

PLOS PATHOGENS
Targeting prolyl-tRNA synthetase in toxoplasmosis

PLOS PATHOGENS
Targeting prolyl-tRNA synthetase in toxoplasmosis against T. gondii. In contrast, the enzymatic SI of L35 was 8.2 versus the enzymatic SI of L97 which was 27.9 (Fig 5). Further, the data revealed that except for L36, the other four PPLderivatives are effective at nM concentrations against TgPRS enzyme. This is attributed to preferential binding favoring the (S) enantiomer (L35) and not the (R) enantiomer (L36), as supported by the crystal structures of drug-bound enzyme complexes. To determine the mechanism of action using L35, cell-based chemical mutagenesis was employed via a forward genetic screen. Tg-resistant parasites were analyzed with the wild-type strain by RNA-sequencing to identify mutations in the coding sequence conferring drug resistance by computational analysis of the variants (S1 Table). Further, DNA sequencing established two mutations, i.e., T477A and T592S in TgPRS, which substantially reduced susceptibility to L35 compared to wild-type parasites (Fig 4). Interestingly, both mutations are proximal to the two terminals of the PPL scaffold and not directly in the ATP, tRNA, or L-pro sites (Fig 4). This observation indicates that both the methyl-pyrazole terminal and hydroxy-methyl moieties may play a crucial role in the selectivity of L35. As for thermal stability, the TgPRS-L35 and HsPRS-L35 had a higher ΔTm value than the other four PPL derivatives. In addition, the thermal stability was higher with L-pro and the compound combination with ATP, or L-pro + ATP (Figs 2 and 5). While the possibility of off-site targeting by ATP-mimetics cannot be dismissed, genetic selection data shown here provide evidence these compounds target only PRS. These ATP mimetics and their stereoisomers (e.g., L35 and L36), while occupying the same catalytic pocket, exhibit specific and distinct inhibition profiles. We recently showed that HFG and L95 can indeed bind together to TgPRS, covering all three of the enzyme-substrate subsites [49]. HFG and L95 compounds act as a triple-site inhibitor set forming an unusual ternary complex wherein HFG occupies the 3'-end of tRNA and the L-pro binding sites and L95 occupies the ATP pocket. These data provide an avenue for structure-based activity enhancement of this chemical series as anti-infectives. Further characterization of the PPL derivatives described in this study will support drug tailoring and development against toxoplasma and related parasites.

S1 Fig. Activity of MMV drugs on Toxoplasma gondii parasite and human cells. A-C)
Graphs showing the dose-response curves of T. gondii parasites in the presence of the MMV compounds indicated. Confluent HFFs were infected with wild-type (A), T447A (B) or T592S (C) edited parasites expressing the Nanoluc luciferase. After 48h of incubation, parasite proliferation was quantified to calculate the IC50 by non-linear regression analysis. The graphs represent the mean ±SD of 3 technical replicates from one experiment. Shaded error envelopes depict 95% confidence intervals. D-E) Dose-response curves of ARPE-19 and MDA231 cell lines in presence of different MMV drugs. Human cells were plated on 96 wells plates and incubated with growing concentrations of drugs. After 72h, the cells' viability was revealed using the CellTiter-Blue assay kit (Promega) and the CC 50 was calculated. The graphs represent the mean ± SD of 3 technical replicates from one experiment.

PLOS PATHOGENS
Targeting prolyl-tRNA synthetase in toxoplasmosis