Reaction hijacking inhibition of Plasmodium falciparum asparagine tRNA synthetase

Malaria poses an enormous threat to human health. With ever increasing resistance to currently deployed drugs, breakthrough compounds with novel mechanisms of action are urgently needed. Here, we explore pyrimidine-based sulfonamides as a new low molecular weight inhibitor class with drug-like physical parameters and a synthetically accessible scaffold. We show that the exemplar, OSM-S-106, has potent activity against parasite cultures, low mammalian cell toxicity and low propensity for resistance development. In vitro evolution of resistance using a slow ramp-up approach pointed to the Plasmodium falciparum cytoplasmic asparaginyl tRNA synthetase (PfAsnRS) as the target, consistent with our finding that OSM-S-106 inhibits protein translation and activates the amino acid starvation response. Targeted mass spectrometry confirms that OSM-S-106 is a pro-inhibitor and that inhibition of PfAsnRS occurs via enzyme-mediated production of an Asn-OSM-S-106 adduct. Human AsnRS is much less susceptible to this reaction hijacking mechanism. X-ray crystallographic studies of human AsnRS in complex with inhibitor adducts and docking of pro-inhibitors into a model of Asn-tRNA-bound PfAsnRS provide insights into the structure activity relationship and the selectivity mechanism.

. Effects of compounds on protein translation inhibition and eIF2α phosphorylation. (A, B). P. falciparum cultures (Cam3.II-rev; trophozoite stage; 30-35 h p.i.) were exposed to PfDHFR inhibitor, WR99210 (A) or protein translation inhibitor, cycloheximide (B) for 6 h. Protein translation was assessed in the last two hours of the incubation, via the incorporation of OPP. Aliquots of inhibitor-exposed cultures were washed and returned to cultures, and viability was estimated at the trophozoite stage of the next cycle. WR99210: IC50 (Translation) > 1 μM, IC50 (Viability) = 0.9 nM. Cycloheximide: IC50 (Translation) = 331 nM, IC50 (Viability) > 1 μM. Error bars correspond to SEM of three independent experiments. (C) Trophozoite stage Cam3.II_rev parasites (30-35 h p.i.) were incubated with 0.05% DMSO (Mock), 50 nM borrelidin (BOR) or 2.5 µM OSM-S-106 or 2.5 µM OSM-S-137 for 3 h. Western blots of lysates were probed for phosphorylated-eIF2α with PfBiP as a loading control. The data represent additional blots related to data presented in Fig.  2B.   :** of these motifs. One Plasmodium-specific feature of interest is the presence of a large insert (highlighted in black, white font). R487 (Pf) and the equivalent residue in other species are indicated (bold, underline type). The flipping loop as defined by (Schmitt et al., 1998) is highlighted in yellow. The mobile residues that are stabilised upon ligand binding in the active site are boxed. The flipping loop that has previously been shown to undergo dynamic motions that facilitate tRNA binding (Schmitt et al., 1998) is highlighted in khaki green. . Density for some residues in these loops was insufficient to allow modeling, and missing residues are represented as dashed yellow lines.
Alternatively, sorbitol-synchronsied parasites (3D7 strain, ring stage) (Lawrence et al., 2000) were incubated with OSM-S-106 and other inhibitors for 72 h. Viability was assessed in the second cycle by flow cytometry, following labelling with 2 µM Syto-61 (Thermo Fisher Scientific) (Dogovski et al., 2015;Xie et al., 2014). Viability represents the parasitemia normalized to untreated and "kill" controls that were treated with 2 µM dihydroartemisinin (DHA; Sigma-Aldrich) for 48-72 h. For drug pulse assays, tightly synchronized Cam3.II-rev  parasites (1-1.5% parasitemia, 0.2% final hematocrit) were added to the plates and incubated for 6 h. Drugs were removed and the parasitemia assessed in the trophozoite stage of the next cycle.
Activity against HepG2 and P. berghei Hepatic human transformed cells (12 x 10 3 ; HepG2-A16-CD81-EGFP), stably transformed to express a GFP-CD81 fusion), were pretreated for 18 h with decreasing concentrations of the compounds of interest, over the range 50 µM to 0.85 nM. The cells were then infected with freshly dissected luciferase-expressing P. berghei (PbLuc) (4 x 10 3 ) sporozoites, as previously described (Swann et al., 2016). After 48 h of incubation with the compound, the viability of P. berghei exoerythrocytic forms (EEF) was measured by bioluminescence using Bright Glow reagent (Promega). HepG2 cytotoxicity was assessed by adding CellTiterGlo reagent (Promega). The plates were read in a PHERAstar FSX reader (BMG LABTECH).

Metabolic stability study using human liver microsomes
A solution of the test compounds in phosphate buffer solution (1 µM) was incubated in pooled human liver microsomes (0.5 mg/mL) for 0, 5, 20, 30, 45 and 60 minutes at 37°C in the presence and absence of an NADPH regeneration system (NRS). The reaction was terminated with the addition of ice-cold acetonitrile, containing a system suitability standard, at designated time points. The sample was centrifuged (4200 rpm) for 20 minutes at 20°C and the supernatant was diluted by half in water and then analysed by LC-MS/MS. The % parent compound remaining, half-life (T1/2) and clearance (CLint,app) were calculated using standard methodology. The experiment was carried out in duplicate. Verapamil, diltiazem, phenacetin and imipramine were used as reference standards.

Minimum inoculum of resistance
Minimum Inoculum of Resistance (MIR) studies were conducted for OSM-S-106 using a modified "Gate Keeper assay" (Duffey et al., 2021). The IC50 was determined to be 88.9 nM (N,n=3,2), and the IC90 was determined to be 169.2 nM (N,n=3,2) in the P. falciparum Dd2-B2 clone. A single-step selection was set up by exposing P. falciparum cultures (Dd2-B2, 3% hematocrit; 1E7 Dd2-B2 parasites in each well of a 24-well plate) to 3 x IC90 (507.7 nM) of OSM-S-106 over 60 days. Wells were monitored daily by smear during the first seven days to ensure parasite clearance, during which media was changed daily. Thereafter, cultures were screened three times weekly by flow cytometry and smearing, and the selection maintained a consistent drug pressure of 3 x IC90 over 60 days. No recrudescence was observed over the course of this selection. Control selections with DSM265 (at 58 nM, corresponding to 5×IC50), yielded 14/96 recrudescent wells, consistent with earlier reports (Phillips et al., 2015) (Duffey et al., 2021). Whole-genome sequencing analysis employed MiSeq data from libraries of 2 x 300 bp paired end reads (Murithi et al., 2021).
In vitro evolution of P. falciparum with reduced sensitivity to OSM-S-106. P. falciparum Dd2 was selected for resistance to OSM-106 over a period of two months, starting at the IC50 and increasing to 4x IC50. Two independent selections were performed, and two or four clones were isolated from each of the selection flasks by limiting dilution, yielding a total of 6 resistant Dd2 clones. Whole-genome sequencing was applied to an average coverage of 117. Reads were mapped to the 3D7 reference genome.
Mutations that were present in both the resistant clones and their isogenic parent were removed. In addition, the genomes were analysed for potential copy number variation with the GATK4 CNV pipeline using panels of controls developed for the Dd2 genetic background (McKenna et al., 2010) (Miles et al., 2016).

Whole genome sequencing and analysis of OSM-S-106-resistant parasites
The sequencing library for parasite genomic DNA was prepared with the Nextera XT kit (Cat. No. FC-131-1024, Illumina) following the standard dual index protocol. The library was sequenced at the UC San Diego IGM Genomics Center on the Illumina HiSeq 2500 in RapidRun mode to generate 100bp paired-end reads. Fastq files were aligned to the P. falciparum 3D7 reference genome (PlasmoDB v13.0) using the previously described Platypus pipeline (Manary et al., 2014). The seven clones generated in the study (one parent clone and six OSM-S-106-resistant clones) were sequenced to an average depth of 132x.
SNVs and INDELs were called against the 3D7 reference genome using GATK HaplotypeCaller and filtered according to GATK recommendations (McKenna et al., 2010). Briefly, SNVs were retained if they met the following filter criteria: ReadPosRankSum >8.0 or <-8.0, QUAL<500, Quality by Depth (QD) <2.0, Mapping Quality Rank Sum <-12.5, and filtered depth (DP) <7. INDELs were retained if they passed ReadPosRankSum <-20, QUAL<500, QD<2, and DP<7. SnpEff was used to annotate variants in the resulting VCF file (Cingolani et al., 2012). Variants with passing quality metrics and ≥90% allele frequency were further filtered to remove mutations that were also present in the Dd2 parent clone, as these would not have evolved over the course of OSM-S-106 selection. Each resistant clone contained 3-6 SNVs or INDELs that met all filtering criteria. CNVs were identified by differential Log2 copy ratio as previously described (Summers et al., 2022).

Generation of PfAsnRSR487S transfectant cell line
A single CRISPR/Cas9 plasmid was used to generate parasites encoding the R487S mutation in PfAsnRS, as shown in Suppl Figure 3. Two guide RNAs were designed using Benchling (benchling.com). The gRNA1 (5'-CATTCGAAGTGAAAGTTGAA-3') and gRNA2 (AGTGAAAGTTGAATGGGGAA) were located near the mutation site. Both gRNAs and their complementary sequences were synthesized as primers by IDT. Each gRNA was cloned into the pDC2-coCas9-gRNA plasmid essentially as described previously (Adjalley and Lee, 2022). A donor template of 780 bp, encompassing coding nucleotide sequences 1045-1824, was synthesized (Thermo Fisher Scientific) and assembled at the AatII and EcoRI sites using NEBuilder HiFi DNA Assembly. In addition to the R487S mutation, additional silent shield mutations that prevent Cas9 binding were introduced, as shown in Suppl Figure 3. Transfections were performed on ring-stage Dd2 parasites using a BioRad Gene Pulser II as described (Adjalley and Lee, 2022), with 5 nM WR99210 drug pressure applied for 8 days. Edited clones were isolated by limiting dilution and validated by Sanger sequencing.

Generation of conditional knockdown parasite lines
Conditional knockdown (cKD) P. falciparum lines were generated for the cytosolic AsnRS (PF3D7_0211800), cytosolic AlaRS (PF3D7_1367700), cytosolic GlyRS (GlyRS; PF3D7_1420400), PfNT4 ( PF3D7_0103200), P. falciparum glutamate dehydrogenase (GDH3; PF3D7_0802000), and P. falciparum carbonic anhydrase (CA, PF3D7_1140000) by fusing the coding sequences and non-coding RNA aptamer sequences in the 3'-UTR, permitting translation regulation using the TetR-DOZI system (Ganesan et al., 2016;Nasamu et al., 2021). Gene editing was achieved by CRISPR/SpCas9 using the linear pSN054 vector that contains cloning sites for the left homology region (LHR) and the right homology region (RHR) as well a target-specific guide RNA under control of the T7 promoter. Cloning into the pSN054 donor vector was carried out following previously described procedures (Ganesan et al., 2016;Nasamu et al., 2021). The vector includes V5-2xHA epitope tags, a 10x tandem array of TetR aptamers upstream of an Hsp86 3'UTR, and a multicistronic cassette for expression of TetR-DOZI (translation regulation), blasticidin S-deaminase (selection marker) and a Renilla luciferase (RLuc) reporter. All primer and synthetic fragment sequences that were generated using the BioXP™ system and IDT gBlocks™ are included in Supplementary Table S9. The final constructs were sequence-verified and further confirmed by restriction digests.
Transfection into Cas9-and T7 RNA polymerase-expressing NF54 parasites was carried out by pre-loading erythrocytes with the donor vector as previously described (Deitsch et al., 2001). Parasite cultures were maintained continuously in 500 nM anhydrotetracycline (aTc, Sigma-Aldrich 37919) and drug selection with 2.5 µg/mL of Blasticidin S (RPI Corp B12150-0.1) was initiated four days after transfection. Cultures were monitored by Giemsa smears and RLuc measurements.

Growth assay for knockdown parasite lines
Assessment of parasite viability during target protein perturbations were carried out using luminescence as a readout of growth. Synchronous ring-stage parasites, cultured in the presence (50 nM) and absence of aTc, were set up in triplicate in a 96-well U-bottom plates (Corning ® 62406-121). Luminescence signals were taken at 0 and 72 h post-invasion using the Renilla-Glo(R) Luciferase Assay System (Promega E2750) and the GloMax® Discover Multimode Microplate Reader (Promega). The luminescence values in the knockdown conditions were normalized to aTc-treated (100% growth) and dihydroartemisinin-treated (500 nM, no growth) samples and results were visualized using GraphPad Prism (version 9; GraphPad Software).

OSM-S-106 susceptibility assays for knockdown parasite lines
The stock solution of OSM-S-106 was dispensed into 96-well (BD Falcon™ 62406-121) and 384-well (Corning ® MPA-3656) U-bottom microplates and serially diluted in complete medium to yield a final concentration in the assay ranging from 0.8-0.003 µM. Synchronous ring-stage PfAsnRS, PfAlaRS, PfGlyRS, PfNT4, PfGDH3, and PfCA cKD parasites, as well as a control line expressing a fluorescent protein under the control of the TetR/DOZI module (Ganesan et al., 2016), were maintained in 0.5 µM aTc to achieve wild-type protein levels, and 0.001 or 0.0015 µM aTc for knockdown of PfAsnRS, PfAlaRS and PfGlyRS, and no aTc for knockdown of PfNT4, PfGDH3 and PfCA. DMSO-and dihydroartemisinin-treatment (0.5 µM) served as reference controls. Luminescence was measured after 72 h as described above and IC50 values were obtained from corrected dose-response curves using GraphPad Prism.

Protein translation assay.
Highly synchronous P. falciparum Cam3.II rev  infected RBCs (30-35 h post-invasion) were exposed to OSM-S-106, cycloheximide, and WR99210 for 4 h. O-propargyl-puromycin (OPP) (Abcam) was added to the culture and incubated for a further 2 hr. Parasites were washed three times in 1x PBS (Gibco™) and fixed with 4% formaldehyde (Polysciences) and 0.02% glutaraldehyde (Sigma) in 1x PBS for 20 min at room temperature (RT). Cells were washed two times with buffer A (3% human serum in 1x PBS). Pellets were permeabilized in buffer A containing 0.05% Triton ® X-100 and washed two times with buffer A. Fixed-permeabilized cells were subjected to copper-catalyzed azide-alkyne cycloaddition (CuAAC) at 37 o C for 1 h in the presence of 0.1 mM CuSO4, 0.5 mM THPTA, 5 mM sodium ascorbate, and 0.1 μM Alexa Fluor 488 azide in buffer A. Pellets were washed four times in buffer A and resuspended in buffer A containing 25 μg/ml propidium iodide (Invitrogen ™). Cells were interrogated by flow cytometry (FACS Canto II; BD Biosciences, San Jose, CA) using FITC and Cy™5.5 channels.

Mass spectrometry to identify and quantify the OSM-S-106-aspargine conjugate
In vitro AsnRS reactions were set up with the following components: 1 μM PfAsnRS, 20 μM L-asparagine, 10 μM ATP, 10 μM OSM-S-106 and 2.5 mg/mL E. coli tRNA (Merck). The reaction buffer consists of 100 mM HEPES pH 7.5 (KOH), 160 mM KCl, 3.5 mM MgCl2, 1 mM DTT. The mixture was incubated at 37°C for 1 h. After that, an equal volume of 8 M urea was added to the mixture. Finally, trifluoroacetic acid was added to a final concentration of 1%. The sample was centrifuged at 15,000 g for 10 min and the supernatant was used for LCMS analysis. Synthetic Asn-OSM-S-106 standards were processed in the same way.
For identification of conjugates in cell cultures, a late trophozoite stage P. falciparum (3D7 strain) culture was exposed to 1 μM or 10 μM OSM-S-106 for 3 h. Following drug treatment, parasite-infected RBCs were lysed with 0.1% saponin in PBS and the parasite pellet was washed 3 times with ice-cold PBS. Cell pellets were kept on ice and resuspended in water as one volume, followed by the addition of five volumes of cold chloroform-methanol (2:1 [vol/vol]) solution. Samples were incubated on ice for 5 min, subjected to vortex mixing for 1 min and centrifuged at 12,000 rpm for 10 min at 4°C to form 2 phases. The top aqueous layer was transferred to a new tube and subjected to LCMS analysis.

High-performance liquid chromatography (HPLC) and mass spectrometric (MS) analyses
Samples were analysed by reversed-phase ultra-high performance liquid chromatography (UHPLC) coupled to tandem mass spectrometry (MS/MS) employing a Vanquish UHPLC linked to an Orbitrap Fusion Lumos mass spectrometer (Thermo Fisher Scientific, San Jose, CA, USA) operated in positive ion mode. Solvent A was 0.1% formic acid/10 mM ammonium acetate in water and solvent B was 0.1% formic acid/10 mM ammonium acetate in acetonitrile. 10 μL of each sample was injected into an RRHD Eclipse Plus C18 column (2.1 × 1000 mm, 1.8 μm; Agilent Technologies, USA) at 50 °C at a flow rate of 350 μL/min for 3 min using 0% solvent B. During separation, the percentage of solvent B was increased from 0% to 25% in 7 min. Subsequently, the percentage of solvent B was increased to 99% in 0.1 min and then maintained at 99% for 0.9 min. Finally, the percentage of solvent B was decreased to 0% in 0.1 min and maintained for 3.9 min. MS experiments were performed using a Heated Electrospray Ionization (HESI) source. The spray voltage, flow rate of sheath, auxiliary and sweep gases were 3.5kV, 20, 6, and 1 'arbitrary' unit(s), respectively. The ion transfer tube and vaporizer temperatures were maintained at 350°C and 400°C, respectively, and the S-Lens RF level was set at 50%. A full-scan MS spectrum and targeted MS/MS for proton adduct of Asn-OSM-S-106 or 20 possible common amino acid-containing inhibitor adducts were acquired in cycles throughout the run. The full-scan MS-spectra were acquired in the Orbitrap at a mass resolving power of 120,000 (at m/z 200) across an m/z range of 200-1500 using quadrupole isolation and the targeted MS/MS were acquired using higher-energy collisional dissociation (HCD)-MS/MS in the Orbitrap at a mass resolving power of 7500 (at m/z 200), a normalized collision energy (NCE) of 20% and an m/z isolation window of 1.6.

Analytical Ultracentrifugation
PfAsnRS, PfAsnRSR478S, CDHsAsnRS and HsAsnRS samples were diluted to 2.8 µM in 25 mM Tris-HCl, pH 7.4, 150 mM NaCl and 0.5 mM TCEP. 400 μL aliquots were loaded into double-channel quartz window cells (Beckman Coulter), with the above buffer in the reference compartment. Cells were centrifuged at 50,000 rpm or 40,000 rpm at 20°C using an XL-I analytical ultracentrifuge (Beckman Coulter) or an Optima analytical ultracentrifuge (Beckman Coulter). Radial absorbance data were acquired at a wavelength of 238 or 280 nm (as indicated), with radial increments of 0.003 cm, in continuous scanning mode. The sedimenting boundaries were fitted to a model that describes the sedimentation of a distribution of sedimentation coefficients with no assumption of heterogeneity (c(s)) using the program SEDFIT (Schuck and Rossmanith, 2000). Data were fitted using a regularization parameter of p = 0.95, floating frictional ratios, and 250 sedimentation coefficient increments.

ATP consumption assay
The consumption of ATP by wildtype PfAsnRS, PfAsnRSR487S and HsAsnRS was determined using a luciferasebased assay as per the manufacturer's instructions (Kinase-Glo Luminescent Kinase Assay, Promega). Reactions were conducted in 100 mM HEPES pH 7.5, 160 mM KCl, 3.5 mM MgCl2, 0.1 mg/mL BSA, 1 mM DTT, with 200 µM L-asparagine, 10 µM ATP, 1 unit/mL inorganic pyrophosphatase and 2.5 mg/mL E.coli tRNA if present. Enzyme concentration and incubation time for each experiment are described in the figure legends.
Reactions were incubated at 37°C, followed by addition of the Kinase Glo reagent and incubation for 10 minutes at room temperature. Luminescence output was measured using a Clariostar plate reader, and the concentration of ATP quantified by linear regression using an ATP standard curve (Microsoft Excel). Data are normalised to the ATP consumption by DMSO (0.5%) treated AsnRSs as a positive control (100% activity). Samples with no enzyme served as negative controls. Dose-response curves and IC50 values were obtained using GraphPad Prism.
Expression and purification of His tagged human AsnRS catalytic domain (His-CDHsAsnRS).
The expression and purification of His tagged human AsnRS catalytic domain, residues A98-P548 (His-CDHsAsnRS) has been described previously (Park et al., 2018). Briefly, the amino acid sequence comprising residues A98-P548 with N-terminal His6-tag was expressed via pET-28a in E. coli strain Solu_BL21 (Genlantis). Cells were cultivated in 1 L LB media supplemented with 50 µg/mL ampicillin in a shaker-incubator at 37°C to OD600 0.5. Recombinant protein expression was induced by addition of 0.5 mM isopropyl β-D-1thiogalactopyranoside (IPTG). Cultures was further incubated for 4 h at 37°C and cells harvested by centrifugation (6,000 g). Pelleted cells were resuspended in lysis buffer containing 0.5 M NaCl, 20 mM Tris-HCl (pH 7.5), 35 mM imidazole, and 1 mM β-mercaptoethanol, lysed with an ultrasonic processor (Cole-Parmer), and centrifuged at 35,000 ×g for 30 min. The supernatant was filtered with 0.45-μm syringe filter device (Sartorius) and loaded onto a HisTrap chelating 5-mL HP column (Cytiva). The loaded column was washed with lysis buffer, and retained His-CDHsAsnRS was eluted with an increasing gradient of lysis buffer containing 1 M imidazole. Prior to ion-exchange chromatography, fractions containing CDHsAsnRS were buffer-exchanged with binding buffer; 100 mM NaCl, 20 mM Tris-HCl pH 7.5, and 5 mM dithiothreitol using a HiPrep desalting 26/10 column (Cytiva) and loaded onto a HiTrap Q 5-mL HP column (Cytiva). CDHsAsnRS was eluted with an increasing gradient of binding buffer containing 1 M NaCl and finally subjected to a HiLoad 16/600 Superdex 200 pg column (Cytiva) equilibrated with the buffer containing 200 mM NaCl, 10 mM HEPES-NaOH (pH 7.0).

Expression and purification of native PfAsnRS, PfAsnRSR478S, HsAsnRS and CDHsAsnRS recombinant proteins.
Plasmid vectors were designed to express recombinant PfAsnRS, PfAsnRSR478S, HsAsnRS (residues M1-P548) and CDHsAsnRS (residues A98-P548) comprising a hexa-histidine tag at the N-terminus, an intervening TEV cleavage sequence and C-terminal AsnRS sequence (His-TEV-AsnRS). Open reading frames were codon optimized for expression in E. coli, synthesized and cloned into the pET11a expression vector (GeneScript). E. coli BL21(DE3) containing expression vector was cultivated in 2 L LB media containing 100 µg/mL ampicillin in a shaker-incubator at 37°C. The culture was transferred to a 16°C shaker-incubator when the cell density approached mid log phase (OD600~0.6). Recombinant His-TEV-AsnRS expression was induced by addition of 0.1 mM IPTG to culture media and cells incubated for an additional 16 hours. Cells were harvested by centrifugation (6,000 g) and resuspended in 40 mL lysis buffer containing 50 mM Tris-HCl, pH 7.4, 350 mM NaCl, 40 mM imidazole, 0.5 mM TCEP, 1 mg/mL lysozyme and 1x protease inhibitor cocktail (Roche). Cells were lysed by sonication (Microtip, QSonica) and the lysate clarified by centrifugation at 30,000 g for 25 min at 4 °C and passage through 0.8/0.2 µm (Pall) syringe filter. The supernatant was applied to a 5 mL HisTrap HP column (GE Healthcare) and washed with 50 mL binding buffer containing 50 mM Tris-HCl, pH 7.4, 350 mM NaCl, 40 mM imidazole, and 0.5 mM TCEP. His-TEV-AsnRS enzyme was eluted using a 0-500 mM imidazole gradient in binding buffer over 100 mL. His-tagged TEV protease (L56V/S135G/S219V triplemutant (Cabrita et al., 2007) was added to His-TEV-AsnRS (mass ratio 1:100, His-TEV-AsnRS:His-TEV protease) and dialyzed overnight at 4°C against 50 mM Tris-HCl, pH 7.4, 350 mM NaCl, 40 mM imidazole, 0.5 mM TCEP. The resultant native AsnRS enzyme was isolated from cleaved His tag and His-TEV protease by passage of sample through a 5 mL HisTrap HP column and collection of flow-through material. Native AsnRS enzyme was further purified by gel filtration using a HiLoad 16/600 Superdex 200 column (GE Healthcare), pre-equilibrated in 25 mM Tris, pH 7.4, 150 mM NaCl and 0.5 mM TCEP.
Crystallisation and X-ray diffraction data collection For crystallization of Asn-AMP-bound CDHsAsnRS, purified apo His-CDHsAsnRS was concentrated to 10 mg/mL. Crystals of His-CDHsAsnRS were first obtained with a solution containing 20% (v/v) glycerol, 40 mM potassium phosphate, and 16% (w/v) polyethylene glycol 8,000 using the hanging drop vapor diffusion method at 295 K. The drops containing crystals were mixed with their reservoir solutions supplemented with 10 mM adenylyl imidodiphosphate lithium salt hydrate, 10 mM L-asparagine, and 20 mM MgCl2 with a 1:1 molar ratio. The crystals were further incubated for 8 h at 295 K, flash-cooled in a 100 K nitrogen stream, and subjected to X-ray diffraction. The collected data were processed with HKL2000. Initial phase estimates were obtained by molecular replacement with PHASER using the previous apo-His-CDHsAsnRS structure (PDB ID: 5XIX) as a template. Automated structure refinement using phenix.refine (Adams et al., 2010) was followed iteratively by manual model building in COOT (Emsley et al., 2010). The statistics for the His-CDHsAsnRS structure are shown in Supplementary Table 8.
For Asn-OSM-S-106 bound, Asn-AMS bound, and apo CDHsAsnRS, purified apo CDHsAsnRS (native sequence) was concentrated to 10 mg/mL. Crystals were obtained in a solution containing 20% (v/v) glycerol, 40 mM potassium phosphate and 14% polyethylene glycol 8,000, and 100-mM Tris pH 7.6 using the sitting drop vapour diffusion method at 295 K. Drops containing crystals were mixed with their reservoir solutions supplemented with 350 µM Asn-OSM-S-106 or 350 µM Asn-AMS. The crystals were further incubated for 24 h at 295K. Crystals were flash-cooled in liquid nitrogen directly from the crystallization drop, and X-ray diffraction data were collected at 100 K and a wavelength of 0.9537 Å using the Eiger 16M detector at the MX2 beamline of the Australian Synchrotron (Aragão et al., 2018). Diffraction data were indexed and integrated using XDS (Kabsch, 2010) and analysed using POINTLESS (Evans, 2011), prior to merging by AIMLESS (Evans and Murshudov, 2013) from the CCP4 software suite (Winn et al., 2011). Initial phase estimates were obtained by molecular replacement in PHASER (Mccoy et al., 2007) using modified coordinates of our Asn-AMP-bound CDHsAsnRS as the search model. Automated structure refinement using phenix.refine (Adams et al., 2010) was followed iteratively by manual model building in COOT (Emsley et al., 2010). Structure refinement was performed using translation/libration screw (TLS) refinement with each chain comprising a single TLS group. Restraints for Asn-OSM-S-106 and Asn-AMS were generated using phenix.elbow (Moriarty et al., 2009). Final data collection and refinement statistics are shown in Supplementary Table 8.

Modelling of the P. falciparum AsnRS/ Asn-tRNA complex
A model of the PfAsnRS/ Asn-tRNA complex was generated by combining a modified version of the AlphaFold model for PfAsnRS bound to Asn-AMP with the tRNA from the structure of the E. coli aspartyl-tRNA synthase/ tRNA complex, 1C0A (Eiler et al., 1999). The catalytic domain of the PfAsnRS model was aligned to the equivalent region of 1C0A using PyMOL (Schrödinger, 2022) and visual inspection showed an extremely good match for the local structure, with the tRNA from 1C0A positioned appropriately across both the active site and onto the anticodon domain. The only significant clash was of the acceptor stem with residues of the flipping loop adjacent to the active site, due to the PfAsnRS model having these in the closed conformation seen in the tRNA-free structures of class II tRNA synthase enzymes (Eiler et al., 1999). The conformation of the flipping loop in the PfAsnRS model was manually corrected to the open position using Coot (Emsley et al., 2010), and the PfAsnRS/ Asn-AMP / tRNA complex model was minimised to remove any minor steric overlaps using SybylX2.1 (Certara, NJ, USA). To generate the PfAsnRS/ AMP/ Asn-tRNA complex, the bond between the asparagine residue and AMP was manually broken and a new bond to the 3'OH oxygen of the acceptor stem terminal adenine was added using SybylX2.1. The modified complex was minimised to correct any errors in bond lengths or angles.

Chemical abbreviations
Asn

General information 1
Reagents were purchased from Sigma-Aldrich, Merck, Fisher Scientific, Apollo Scientific, and Fluorochem and were used without further purification unless specified. Anhydrous conditions: glassware was dried at >130 °C for >12 h, assembled hot, and allowed to cool under a high vacuum or purged with inert gas where suitable. Nitrogen (N2) and argon (Ar) gas were used as obtained, from a cylinder. On a Schlenk line, the phrase in vacuo equates to 10 mbar. Reduced pressure means 900 to 50 mbar under rotary evaporation at 40 °C. Davisil Grace Davison 40 -63 μm (230 -400 mesh) silica gel and a Biotage Isolera One or Biotage Selekt were used for automated flash chromatography. Analytical thin-layer chromatography (TLC) was carried out on Merck Silica Gel 60 F254-precoated aluminum plates (0.2 mm) and observed using UV irradiation (254 nm and 280 nm) and staining with potassium permanganate, anisaldehyde, or ninhydrin. High-temperature reactions were carried out in either temperature-controlled silicone oil baths or DrySyn heating blocks.
Melting points (M.P.) were measured using a Stanford Research Systems OptiMelt instrument at 1 °C min -1 (capillaries = 1.5 -1.6 mm, 90 mm). Without atmospheric adjustment, infrared spectroscopy was performed using a Bruker Alpha-E (attenuated total reflectance) and analysed with Microlab PC software. The samples were examined neat. Bruker spectrometers were used for nuclear magnetic resonance spectroscopy at 300 K: AVANCE III 400 (  High resolution mass spectrometry was carried out on Agilent 6545XT AdvanceBio LC/Q-TOF with ESI ionisation. The charge of the ion specifies whether the detection is positive or negative; for instance, [M+H] + denotes positive-ion detection. Analytical liquid chromatography-mass spectrometry (LCMS) was performed on an Agilent Infinity 1290 II system consisting of a quaternary pump (G7111A) and a diode array detector WR (G7115A) coupled to a InfinityLab LC/MSD (G6125B) using ESI. An Agilent Poroshell 120 EC-C18 column (2.7 µm, 4.6 x 50 mm) was eluted at a flow rate of 1.5 mL/min with a mobile phase of 0.05 % formic acid in H2O and 0.05 % formic acid in MeCN. Preparative LCMS was performed on a combined Agilent Infinity 1260 II and Infinity 1290 II system consisting of a preparative binary pump (G7161A) and a multiple wavelength detector (G7165A) coupled to a InfinityLab LC/MSD (G6125B) using ESI and a preparative open-bed fraction collector (G7159B). An Agilent Eclipse XDB-C18 column (5 µm, 9.4 x 250 mm) was eluted at a flow rate of 5 mL/min with a mobile phase of 0.05% formic acid in H2O and 0.05% formic acid in MeCN. As determined by NMR spectroscopy, the purity of all substances exceeded 95%.

General information 2 (for Asn-AMS synthesis)
Reagents: Reagents were obtained from Aldrich Chemical (www.sigma-aldrich.com) or Acros Organics (www.fishersci.com) and used without further purification. Optima or HPLC grade solvents were obtained from Fisher Scientific (www.fishersci.com), degassed with Ar, and purified on a solvent drying system unless otherwise indicated.
Reactions: All reactions were performed in flame-dried glassware under positive Ar pressure with magnetic stirring unless otherwise noted. Liquid reagents and solutions were transferred through rubber septa via syringes flushed with Ar prior to use.Cold baths were generated as follows: 0 °C, wet ice/water; -10 °C, wet ice/brine; -20 °C, dry ice/isopropanol monitored with a thermometer; -44 °C, dry ice/CH3CN; -63 °C, dry ice/chloroform; -78 °C, dry ice/acetone; -100 °C, dry ice/Et2O. Nomenclature: N.B.: Atom numbers in chemical structures herein refer to the standard nucleoside numbering system used in the text of the article and Supporting Information and not to IUPAC nomenclature, which was used solely to name each compound. Compounds not cited in the paper are numbered herein from S1.

General Synthetic Procedure
General procedure 1: Suzuki reaction between thienopyrimidine core and functionalised aryl halides.
Organohalide (1 equiv.), pinacol boronate (1.1 equiv.) and PdCl2(dppf)·DCM (0.1 equiv.) were combined. i-PrOH and 1 M aq. K2CO3 (1 -3 equiv. for the stated condition) were added and the reaction mixture was heated using conventional heating or microwave irradiation for the stated time after being degassed with Ar. The reaction mixture was diluted with MeOH, filtered through celite, and concentrated under reduced pressure to give a residue that was purified by automated flash chromatography on silica to give the coupled product.
compound as a white powder (480 mg, 40% over two steps). Analytical data agreed with those reported previously (Ishikawa and Kakeya, 2014).