A potent and selective reaction hijacking inhibitor ofPlasmodium falciparumtyrosine tRNA synthetase exhibits single dose oral efficacyin vivo

41 The Plasmodium falciparum cytoplasmic tyrosine tRNA synthetase ( PfTyrRS) is an attractive drug 42 target that is susceptible to reaction -hijacking by AMP -mimicking nucleoside sulfamates. We 43 previously identified an exemplar pyrazolopyrimidine ribose sulfamate, ML901, as a potent pro-44 inhibitor of PfTyrRS. Here we examined the stage specificity of action of ML901, showing very good 45 activity against the schizont stage, but lower trophozoite stage activity. We explored a series of ML901 46 analogues and identified ML471, which exhibits improved potency against trophozoites and enhanced 47 selectivity against a human cell line. Additionally, it has no inhibitory activity against human ubiquitin-48 activating enzyme (UAE) in vitro. ML471 exhibits low nanomolar activity against asexual blood stage 49 P. falciparum and potent activity against liver stage parasites, gametocytes and transmissible gametes. 50 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 2 It is fast-acting and exhibits a long in vivo half-life. ML471 is well-tolerated and shows single dose 51 oral efficacy in the SCID mouse model of P. falciparum malaria. We confirm that ML471 is a pro-52 inhibitor that is converted into a tight binding Tyr-ML471 conjugate by the PfTyrRS enzyme. A crystal 53 structure of the PfTyrRS/ Tyr-ML471 complex offers insights into improved potency, while molecular 54 docking into UAE provides a rationale for improved selectivity. 55 56 Short title: Reaction hijacking inhibitor of P. falciparum tyrosine tRNA synthetase 57 58

Plasmodium falciparum, aminoacyl tRNA synthetase, reaction-hijacking, antimalarial 59 60

61 Malaria is a debilitating disease caused by protist parasites of the genus Plasmodium that places an 62 enormous health burden on the world’s poorest communities. In 2022, more than 200 million people 63 were infected with P . falciparum , resulting in more than 600,000 deaths [1]. T he burden was 64 exacerbated by disruptions to services during the COVID pandemic [2]. Unfortunately, the past 15 65 years have seen the emergence of P . falciparum parasites that exhibit partial resistance to artemisinin 66 and partner drugs, such as piperaquine and mefloquine , resulting in ~50% treatment failure with 67 standard artemisinin combination therapies in some regions of Southeast Asia [3, 4] . The recent 68 emergence in Africa of parasites harbouring artemisinin resistance-conferring K13 mutations [5-7] is 69 of great concern, and the Medicines for Malaria Venture (MMV) not-for-profit partnership has declared 70 that new antimalarial therapies and prophylaxis regimens need to be developed as a failsafe [8]. 71 Certain AMP-mimicking n ucleoside sulfamates act as pro -inhibitors of E1 enzymes, i.e. ubiquitin/ 72 ubiquitin-like protein (UBL) -activating enzymes [9-13]. The UBL-bound form of these Adenylate -73 Forming Enzymes (AFEs) is susceptible to attack, leading to the formation of an inhibitory sulfamate-74 UBL adduct within the active site. This unusual reaction hijacking mechanism has been exploited to 75 generate new clinical candidates, such as Pevonedistat [14, 15]. 76 We previously screened a Takeda Pharmaceuticals nucleoside sulfamates library (Cambridge, MA, 77 USA) and identified ML901 as an exemplar pyrazolopyrimidine sulfamate, with potent activity against 78 P . falciparum [16]. O ur group showed that, surprisingly, this AMP-mimicking nucleoside sulfamate 79 uses a related reaction hijacking mechanism to target another AFE subclass, the amino acyl tRNA 80 synthetases. ML901 binds the P . falciparum cytoplasmic tyrosine tRNA synthetase ( PfTyrRS), and 81 then reacts with the bound activated amino acid, resulting in the synthesis of an inhibitory sulfamate-82 amino acid adduct within the active site of the enzyme. By contrast, the equivalent human enzyme is 83 not susceptible to reaction hijacking. That finding was the first demonstration of reaction hijacking of 84 an enzyme class other than the E1 enzymes. More recently, we have identified an 85 aminothienopyrimidine sulfonamide, OSM -S-106, with selective reaction hijacking activity against 86 the P . falciparum asparagine tRNA synthetase [17]. 87 ML901 exhibits low, but measurable toxicity against a mammalian cell line [16], which is potentially 88 due to cross-inhibition of UBLs [12]. Thus, we explored a range of ML901 derivatives from the Takeda 89 nucleoside sulfamate library, with different substitutions at the 7 -position of the pyrazolopyrimidine 90 ring system, to identify compounds with improved selectivity. We identified ML471 as a compound 91 with enhanced potency against P . falciparum and decreased activity against human ubiquitin-activating 92 enzyme (UAE). ML471 exhibits enhanced cellular and biochemical selectivity. It also exhibits activity 93 against plasmodium liver stages and sexua l transmissible stages. Importantly, ML471 exhibits rapid 94 killing kinetics and demonstrates single dose oral efficacy against P . falciparum in an in vivo model. 95

and Discussion 96 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 3 97 Potency and selectivity of ML901 derivatives 98 We examined the activity of a series of pyrazolopyrimidine sulfamates with different substitutions at 99 the 7-position (Fig. 1A-I), from the Takeda Pharmaceuticals Library, against the growth of asexual 100 blood stage P . falciparum (3D7 strain) in a 72-h exposure assay. Consistent with our previous report 101 [16], ML901 exhibits potent activity with a 50% Inhibitory Concentration (IC50_72h) of 2.8 nM (Table 102 1). Indeed, most of the compounds show excellent potenc y (Table 1), demonstrating that different 103 substitutions at the 7-position are well -tolerated, except for the bulky phenoxy substituent ( ML470) 104 (IC50_72h = 44.8 nM; Table 1). The non-specific inhibitor, adenosine 5’-monosulfamate (AMS, Fig. 1J 105 [16]), which has a different heterocyclic base and lacks the 7-position substituent, also exhibits potent 106 activity (IC50 = 1.8 nM; Table 1). ML471, which bears an isopropyl group at the 7-position, exhibits 107 very potent activity (IC50_72h = 1.5 nM, Table 1). 108 109 Short duration pulse exposure to antimalarial drugs can be used to dissect differences in potency of 110 compounds against different stages of development that may not be evident in standard 72-h exposure 111 assays [18]. Here we subjected tightly synchronised blood stage P . falciparum cultures to different 112 duration pulses of ML901 and measured parasitemia in the next cycle [19] . Schizont stage parasites 113 are efficiently killed by ML901, even when exposed to pulses as short as 3 h or 6 h (Supplementary 114 Fig. 1B , Table S1). This may reflect the need for synthesis of daughter merozoite proteins during 115 schizogony. By contrast, trophozoite stage parasites are 10 to 20 -fold less sensitive (Supplementary 116 Fig. 1A, Table S1). 117 118 To determine whether ML471 and other selected pyrazolopyrimidine sulfamates exhibit enhanced 119 potency, we exposed synchronised trophozoite stage cultures to 6- h pulses and measured growth 120 inhibition by quantifying the SYBR Green I fluorescence signal in the next cycle. ML471 exhibits 121 enhanced potency (IC 50_6h = 29.1 nM, Fig. 2A, Table 1) compared with ML901 and the other analogues 122 tested (Fig. 2A, Table 1, IC50_6h values ranging from 135 nM to 220 nM). 123 124 We examined the toxicity of the compounds against the human HepG2 cell line , employing a 72-h 125 exposure period. ML901 inhibits the growth of HepG2 cells with an IC 50_72h of 4.65 μ M (Table 1). 126 Some of the compounds from the pyrazolopyrimidine sulfamate series, including ML471 exhibited 127 markedly improved selectivity, with IC50 values above the range of the assay (>50 μM, Table 1). Other 128 compounds such as ML107, which has a trifluoro methyl substituent, and ML681 and ML864 which 129 have slightly smaller substituents as compared to an isopropyl group, show a higher level of toxicity 130 against the mammalian cell line. These data suggest that the size of the 7-position substituent impacts 131 selectivity. Our previous report [16] showed that AMS is also toxic to mammalian cells lines, such as 132 HCT116 (IC50_72h = 26 nM), in agreement with previous reports [20, 21], and, as expected, given its 133 broad inhibitory activity. 134 135 ML471 exhibits low activity against critical E1 enzymes 136 ML901 was originally investigated as an inhibitor of Atg7, an E1 that activates the Autophagy-related 137 protein 8 (Atg8), a ubiquitin-like protein involved in the formation of autophagosomal membranes [12, 138 22]. However, Atg7 is not essential for cell survival in vitro [23] and is unlikely to underpin the 139 observed mammalian cell toxicity [12]. By contrast, loss-of-function of other E1 enzymes, in particular 140 UAE, is known to be deleterious to the survival and growth of cells [10, 11, 24, 25]. 141 142 To explore the molecular basis of the enhanced cellular selectivity of ML471 and other derivatives 143 compared with ML901, we assessed inhibitory activity against a range of E1 enzymes. ML901 exhibits 144 strong inhibition of Atg7 (IC 50 = 33 nM) and clinically relevant activity against UAE (IC50 = 5.39 μM), 145 with lower-level activity against NEDD8 Activating Enzyme (NAE IC50 = 28 μ M) and no activity 146 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 4 against SUMO Activating Enzyme (SAE) (Table 2). ML676, ML723, ML681, ML107 and ML111 147 also exhibit relevant activity against one or more of UAE, NAE or SAE, consistent with their weak to 148 moderate cellular toxicity. Such off-target activity could limit the development of these compounds. 149 As previously reported [16], AMS is a potent inhibitor of each of the E1s tested (Table 2), likely 150 contributing to the high cellular toxicity. ML471 inhibits the activity of human Atg7 (IC50 = 22 + 9 151 nM), but exhibits no or very little activity against UAE , NAE and SAE, consistent with low 152 mammalian cell cytotoxicity. 153 154 Parasitological properties of ML471 155 Given its enhanced potency and selectivity, ML471 was selected for further characterisation. In 156 addition to potent and selective activity against laboratory asexual blood stage P . falciparum, new 157 antimalarial compounds should exhibit activity against clinical str ains of P . falciparum and P . vivax 158 and, preferably, exhibit activity against liver and transmissible stages. ML471 exhibits potent activity 159 against South American clinical isolates of P . falciparum and P . vivax, including chloroquine-resistant 160 strains, with median IC50 values of 4.0 nM (Pf) and 6.7 nM ( Pv), respectively, similar to artesunate 161 (Table S2). ML471 exhibits improved potency compared with ML901 against gametocytes at both the 162 early (IC50 = 112 nM) and mature (IC50 = 392 nM) stages of development, with potencies similar to 163 those for Methylene B lue and the Plasmodium phosphatidylinositol 4-kinase (PI4K) inhibitor, 164 MMV390048 (Table S3). ML471 prevents development of both P . falciparum NF175 and NF135 165 schizonts in primary human hepatocytes with high potency ( IC50 = 2.8 nM for NF175 and IC 50 = 5.5 166 nM for NF135, Table S4), while exhibiting no toxicity against the primary hepatocyte host cells (Table 167 S4). ML471 potently inhibits the fertility of transmissible male (IC50 = 49 nM) and female (IC50 = 260 168 nM) gametocytes (Table S5). The positive control, Cabamiquine, exhibited potent activity against both 169 gametocyte sexes, consistent with a previous report [26]. In each of these assays, ML471 exhibits 170 similar or improved potency compared with ML901 (Table S3-S5). 171 The Parasite Reduction Rate (PRR) was assessed using a standardized method [27] and compared with 172 compounds exhibiting very fast (artemisinin), fast (chloroquine), moderate (pyrimethamine) or slow 173 (atovaquone) killing profiles, at 10 times their respective IC50_48h values. The Log PRR for ML471 of 174 4.1 is considered fast, and is similar to chloroquine (Fig. 2B, Table S6). 175 176 Pharmacological properties of ML471 177 To meet MMV candidate selection criteria n ew antimalarial compounds for treatment indications 178 would minimally need to have an oral dose of <500 mg to achieve a 12-log kill in a 55 kg adult. ML471 179 exhibits a favourably low molecular weight ( MW = 3 88) and good solubility (Table S 7). It has a 180 predicted AlogP of -1.18 and a Topological Polar Surface Area (TSPA) of 189 Å 2, which suggest this 181 compound may have difficulty being absorbed but should be metabolically stable. As expected, rat 182 oral bioavailability needs optimisation (%F (blood) = 8.72 to 9.56, n = 3) and renal clearance of the 183 parent compound was evident (9-38% of the dose recovered in 0-24 h urine as parent). Despite the sub-184 optimal oral absorption, the rat pharmacokinetic profile of ML471 (25 mg/kg p.o.; Fig. 3B; Table S 7) 185 exhibits excellent duration of absorbed drug exposure. The area under the curve is 30 µM.h, reflecting 186 the low blood clearance (~4% of liver blood flow after an IV dose of 1 mg/kg) and the long terminal 187 half-life in blood (T1/2∞ = 30.5 h) (Table S7). Acting in its favour, ML471 shows high retention in red 188 blood cells (RBCs) in the i.v. PK study, with blood to plasma ratios around 1 at the initial sampling 189 times, but increasing greatly over time due to slower clearance from the RBC compartment (Table S7; 190 and compare Fig. 3A and B). ML471 contains a sulf amate group that is predicted to bind tightly to 191 carbonic anhydrase [28, 29], which likely explains accumulation of ML471 into RBCs (where carbonic 192 anhydrase is abundant and where the asexual stage parasites are located). 193 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 5 Efficacy of ML471 in a SCID model of P . falciparum 194 Single, low-dose oral efficacy is a key requirement for new antimalarial treatments to be used as part 195 of the MMV’s Single Encounter Radical Cure and Prophylaxis (SERCAP) target product profile 196 (TPP1) [8]. We determined the in vivo antimalarial efficacy of ML471 in severe combined immune 197 deficient (SCID) mice, engrafted with human RBCs and infected with P. falciparum [30, 31], which 198 is the gold standard for testing in vivo efficacy of malaria drug candidates. Oral dosing results in good 199 exposure. Single doses of 100 mg/kg p.o. and 200 mg/kg p.o. resulted in AUC0-120h values of 640 µM.h 200 (250,000 h.ng/mL) and 550 µM.h (215,000 h.ng/mL), respectively (Fig. 3C; Table S8), while a dosing 201 regimen of 4 x 50 mg/kg p.o. resulted in an area under the curve (AUC 0-24h) of 54 µM.h (21,100 202 h.ng/mL), assessed at 24 h, after the first dose) ( Fig. S2A; Table S8). All doses were well tolerated , 203 and the long half-life and high exposure are encouraging as these are important properties for single 204 dose antimalarials. 205 206 Mice were infected intravenously with 2 x 10 7 P . falciparum ( Pf3D70087/N9) and ML471 was 207 administered on day 3 post -infection. The single dose regimen (either 100 or 200 mg/kg p.o.) w as 208 sufficient to achieve reduction of 3D7 parasitemia to baseline, with a parasite clearance rate similar to 209 that of chloroquine (CQ; 50 mg/kg p.o.) (Fig. 3D) and no evidence of toxicity. Similarly, the dosing 210 regimen of 4 x daily doses of 50 mg/kg p.o. reduced the 3D7 parasitemia to baseline with a clearance 211 rate similar to chloroquine (CQ; 4 x 50 mg/kg p.o.) (Supp Fig. 2B). 212 213 ML471 and ML901 selection leads to amplification of the PfTyrRS locus 214 In vitro evolution of resistance, under a standardized protocol, has been used to assess the propensity 215 for the development of resistance [32-34]. Here, we examined the resistance potential of both ML901 216 and ML471, employing a single-step selection with Dd2-B2 parasites. For ML901, the parasites were 217 subjected to pressure at 3 x IC 50, while for ML471, parasites were subjected to 10 x IC 50. With both 218 compounds, parasites were retrieved and IC 50 shifts were observed (ranging from two- to 16-fold; 219 Table S9). The Minimum Inoculum for Resistance values for ML901 and ML471 were estimated to 220 be 107 and 7.1 x105, respectively (Table S9). These values are at or below the preferred threshold for 221 further development, making this a parameter of concern. Use of these sulfamates in a drug 222 combination could suppress the evolution of resistant mutants. 223 We performed whole -genome sequencing of parasite lines selected for resistance. Copy number 224 variations (CNVs) were found in flasks selected with either ML901 or ML471, with amplification s 225 always containing the PfTyrRS gene located within amplicons of varying sizes (Table S10, S11). This 226 gene was present in 2 -4 copies in the amplified lines. This finding is consistent with our earlier 227 identification of PfTyrRS as the target of ML901 [16]. No SNPs were found in any of the samples, in 228 contrast to a previous report [16]. This may be due to the slow ramp-up exposure method employed in 229 the earlier study. 230 ML471 targets P . falciparum tyrosine tRNA synthetase via a reaction hijacking mechanism 231 Re action hijacking inhibition of PfTyrRS is expected to lead to the formation of Tyr-ML471 adducts 232 (Fig. 1K) in the active site. We treated P . falciparum infected RBCs for 2 h with 1 μM ML471 and 233 subjected extracts to LC-MS to search for amino acid -ML471 conjugates. An LC -MS peak 234 corresponding to Tyr-ML471 precursor ion ( m/z 552.1871) was detected at the retention time of 3.0 235 min (Fig. 4A). Synthetic Tyr-ML471 was generated as a standard to confirm the peak assignment (Fig. 236 4A, Supplementary Fig. 3A). None of the other 19 possible amino acid conjugates were detected. 237 Recombinant PfTyrRS was produced in Escherichia coli as previously described [16]. To examine the 238 ability of Pf TyrRS to generate the Tyr-ML471 conjugate, the enzyme was incubated with ATP, Tyr, 239 tRNATyr and ML471. Following sample extraction, LC-MS analysis revealed a peak at m/z 552.1868 240 with a retention time of 3.0 min, consistent with that of the Tyr-ML471 precursor ion (Supplementary 241 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 6 Fig. 3B). MS/MS analysis further confirmed the identity of the conjugate (Supplementary Fig. 3C). 242 Recombinant P . falciparum tyrosine tRNA synthetase is thermally stabilised upon formation of the Tyr-243 ML471 adducts 244 When ML471 was incubated with Pf TyrRS in the presence of all other substrates ( i.e., Tyr, ATP and 245 PftRNATyr), the apparent protein melting point (T m), measured by differential scanning fluorimetry 246 (DSF), increased by a remarkable 18°C (Fig. 4B , Table 3). The increase in thermal stability is even 247 greater than that induced by the Tyr-ML901 adduct (Table 3), suggesting higher affinity binding, which 248 is consistent with the enhanced antimalarial activity of ML471 compared with ML901. Importantly, 249 when incubated in presence of ML901 or ML471, and all substrates, recombinant HsTyrRS was not 250 stabilised (Fig. 4C, red and orange curves). This shows that the human enzyme is not susceptible to 251 hijacking by ML471. By contrast, incubation of HsTyrRS with the broad specificity compound, AMS, 252 and all substrates, leads to substantial thermal stabilization, consistent with efficient reaction hijacking 253 by AMS (Fig. 4C, Table 3). 254 ML471 inhibits ATP consumption by PfTyrRS 255 Recombinant PfTyrRS consumes ATP at a moderate level, even in the absence of tRNA , due to the 256 generation and release of AMP-Tyr in the initial reaction phase (Fig. 4D). The rate of ATP consumption 257 is increased 6- fold upon addition of tRNA Tyr, consistent with productive aminoacylation. ML471 258 inhibited ATP consumption by Pf TyrRS when added in the presence of PftRNATyr (IC50 = 1.4 μ M) 259 much more potently than ML901 (IC50 = 13.4 μM) (Fig. 4E, Table 3). Indeed, ML471 is 6 to 30 times 260 more effective than the other ML901 analogues examined ( Supplementary Fig. 4, Table 3). AMS is 261 also significantly less potent in this assay (IC50 = 51.7 μM) (Fig. 4E, Table 3). 262 Docking of ML471 into UAE reveals differential interactions within the active site 263 Susceptibility to reaction hijacking depends on the ability of the pro -inhibitor to bind in the ATP -264 binding pocket of the relevant adenylate -forming enzyme, in a pose that is suitable for reaction with 265 the relevant enzyme- bound product. To probe the molecular basis for the enhanced selectivity of 266 ML471 compared with ML901, we used the Surflex -Dock molecular docking module in SybylX2 to 267 dock ML901 and ML471 into the ATP-binding site of human UAE (6DC6) [35] . For comparison, 268 ML901 and ML471 were docked into the binding pockets of the A- and B-chains of the PfTyrRS/ Tyr-269 ML901 complex (7ROS) [16], noting that the two chains of the dimeric Pf TyrRS structure in 7ROS 270 show differences in the position of key residues around the Tyr-ML901 ligand that are thought to relate 271 to altered mobility of the KMSKS loop (Xie et al., 2022). 272 As described above, ML901 and ML471 bear, respectively, difluoromethoxy and isopropyl groups at 273 the 7-postion of the pyrazolopyrimidine ring. Both ML901 and ML471 can be docked into the active 274 sites of the A- and B-chains of PfTyrRS, with the 7-position substituent located in a solvent accessible 275 pocket. Fig. 5A illustrates the B-chain with docked ML901 (aqua carbons), overlaid with the pose 276 adopted when ML901 (yellow carbons) is docked into the A-chain. In both cases the difluoromethoxy 277 group is positioned away from His70 (red arrow). By contrast, when ML471 (aqua carbons) is docked 278 into the B-chain and overlaid with the pose adopted when ML471 (yellow carbons) is docked into the 279 A-chain, the isopropyl group is positioned closer to His70 (Fig. 5B), indicative of a clash. 280 Overlay of the docked c onformations of the two pro-inhibitors with the published PfTyrRS/ Tyr-281 ML901 structure (7ROS) shows close alignment of the nucleoside sulfamate regions (7ROS) (Fig. 282 5C,D). Again, the different poses of the difluoromethoxy and isopropyl groups are evident (red 283 arrows). Of interest, the ribose ring of Tyr-ML901 in the crystal structure is twisted relative to the 284 docked pro-inhibitors (purple arrows). This may arise from the conjugation of the ML901 sulfamate 285 to tyrosine, which repositions the ligand (Fig. 5C,D). 286 When ML901 and ML471 are docked into the ATP -binding site of human UAE, the oxygen of the 287 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 7 ML901 difluoromethoxy group makes a H -bond interaction with Asn577 (Fig. 5E) that cannot exist 288 for the isopropyl group in ML471 (Fig. 5F). Instead, the hydrophobic isopropyl group must pack into 289 a polar space flanked by Asn577 and Arg551, an unfavourable interaction that is likely to reduce the 290 affinity of ML471 for the site. The sulf amate groups are positioned slightly differently (Fig. 5E,F). 291 However, as the chemical structure of this region of the two compounds is identical, this difference is 292 regarded as an artefact of the docking rather than a significant alteration of the probable binding pose. 293 Thus, differences in the interactions made by the 7 -position substituent appear likely to underpin the 294 improved selectivity of ML471. 295 Crystal structure of PfTyrRS -Tyr-ML471 provides insights into the molecular basis for enhanced 296 potency 297 The complex of Pf TyrRS with synthetic Tyr -ML471 was crystalised using previous ly established 298 conditions [16] and the structure refined at 1.8 Å resolution, revealing a homodimer organization (Fig. 299 6A) with clear density for the Tyr-ML471 ligand. As described previously [16], PfTyrRS is a Class I 300 aaRS, characterized by a catalytic domain with Rossmann fold structure (residues 18–260) linked to a 301 C-terminal anti-codon binding domain (residues 261–370) involved in recognition of tRNA Tyr. 302 PfTyrRS contains “HIGH” and “KMSKS” (70HIAQ74 and 247KMSKS251 in PfTyrRS) motifs that are 303 characteristic of the catalytic domain of Class I tRNA synthetases (sub-class c). 304 305 Tyr-ML471 makes many interactions with active site residues, involving the pyrazolopyrimidine 306 amine, ribose hydroxyls, sulfamate, and tyrosine (Fig. 6B; Supp Fig. 5A,B). The 7-position isopropyl 307 group is oriented away from the binding pocket and is partially solvent exposed, consistent with the 308 docking study. The isopropyl group of ML471 is oriented differently in the A - and B-chains and the 309 adjacent His70 (of 70HIAQ73) adopts different side chain rotamers in each chain (Fig. 6C; Supp Fig. 310 5C). The majority of the KMSKS loop was poorly defined in the electron density in both chains, 311 suggesting flexibility. 312 313 Comparison of the B chains of the Tyr-ML901-bound and Tyr-ML471-bound PfTyrRS structures 314 highlights a difference in the KMSKS loop organisation. The difluoromethoxy group of ML901 is 315 oriented away from His70 ( Fig. 6D, red arrow) , allowing His70 to adopt a configuration that make s 316 close contact with Met248 in the 247KMSKS251 loop, thereby stabilising the loop (Fig. 6D,E; aqua 317 backbone). By contrast the isopropyl group of ML471 in chain B is oriented with one methyl group 318 towards His70 (Fig. 6C, blue arrow), and the His70 side chain adopts a rotamer that is incompatible 319 with the Met248 interaction observed for Tyr-ML901. Thus, for chain B, Tyr-ML471 binding appears 320 to be associated with loop destabilisation while Tyr-ML901 binding is associated with loop 321 stabilisation. Interestingly, in the A-chain, His70 adopts a similar pose that does not support interaction 322 with Met248 in both Tyr-ML471- and Tyr-ML901-bound structures, leading to destabilisation of the 323 loop (Supp Fig. 5C-E). 324 325 Movement of the KMSKS loop is required for access to the active site. For example, in tRNA Trp-bound 326 human TrpRS, the equivalent KMSAS loop adopts a semi -open conformation, that is intermediate 327 between the open conformation observed in the unliganded enzyme and the closed conformation 328 observed in the Trp-AMP complex [36]. The pose adopted by the 7-position isopropyl group of ML471 329 appears to re-position His70, leading to increased flexibility of the KMSKS loop. This may, in turn, 330 enhance the binding or re-binding of the Tyr-tRNA product, positioning the Tyr-tRNA carbonyl carbon 331 for attack by the sulfamate nitrogen of ML471. This may underpin the higher potency of ML471 as a 332 hijacking inhibitor of recombinant PfTyrRS and as an inhibitor of the growth of P . falciparum. 333 In conclusion, we have identified ML471 as a pyrazolopyrimidine sulfamate with improved potency 334 and selectivity compared with ML901. The improved potency derives from improved effi ciency of 335 reaction hijacking inhibition of PfTyrRS and/ or tighter binding of the Tyr-ML471 adduct, as indicated 336 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 8 by enhanced thermal stability. The enhanced potency of ML471 i s associated with repositioning of 337 His70 in the active site of chain B when Tyr-ML471 is bound. T he enhanced cellular selectivity of 338 ML471 may be due to its decreased activity against human UAE, which is associated with a lack of 339 interaction between the 7-position substituent of ML471 and active site residues. ML471 exhibits a 340 rapid mode of action and a long in vivo half-life underpinning its single dose efficacy in a mouse model 341 of P. falciparum malaria. With further improvement of the oral bioavailability , ML471 represents a 342 very interesting compound for prophylaxis, treatment and blocking of transmission of deadly malaria 343 infections. 344 345

346 347 Inhibition of growth of P . falciparum cultures 348 Routine analyses of antimalarial activity against P . falciparum 3D7 was tested by TCGLS (Kolkata, 349 India) using the lactate dehydrogenase (PfLDH) growth inhibition assay as previously described [37]. 350 For assays readout, 70 μ L of freshly prepared reaction mix containing 100 mM Tris -HCl pH 8, 143 351 mM sodium L -lactate, 143 μ M 3-acetyl pyridine adenine dinucleotide (APAD), 179 μ M Nitro Blue 352 tetrazolium chloride (NBT), diaphorase (2.8U/mL) and 0.7% Tween 20 was added into each well of 353 the assay plate. Plates were shaken to ensure mixing and were placed in the dark at 21⁰C for 20 min. 354 Data were normalized to the percentage of growth inhibition with respect to positive (0.2% DMSO, 355 0% inhibition) and negative (mixture of 100 μM chloroquine and 100 μ M atovaquone, 100% 356 inhibition) controls. P . falciparum strain (3D7) was obtained from BEI Resources. 357 358 Analysis of inhibition of growth and viability of P . falciparum cultures followed pulsed compound 359 exposure 360 Assessment of the killing activity of ML901 in pulsed exposure assays was performed using a 361 modification of a previously described procedure [38]. Briefly, cultures of Cam3.II rev [39] trophozoites 362 (1.5% final hematocrit; 1.4% final parasitemia) were pre-synchronized to a 5-h window at trophozoite 363 or schizont stage [40] . C ompounds were serially diluted in complete medium in v -bottomed 364 microplates. Parasites were exposed to ML901 for 3 h, 6 h, 9 h or 24 h before washing 5 times with 365 200 µl of complete medium, then returned to culture conditions. For the no wash samples, ML901 was 366 left in the culture until the assay point. Growth inhibition was assessed in the second cycle by labelling 367 with the DNA-binding dye, SYBR Green I. Quantification of total DNA level reflects cytostatic effects 368 as well as cytocidal effects [41] . Old media (140 µl) was firstly replaced with fresh media, followed 369 by the addition of lysis buffer (20 mM Tris, pH 7.5, 5 mM EDTA, 0.008% w/v saponin, 0.08% v/v 370 Triton X-100) containing SYBR Green I. Plates were incubated at room temperature for 2 h and 371 fluorescence readings were taken using a microplate reader (BMG LABTECH). Unwashed samples 372 containing compounds at >10 times the IC50 values were used as background controls. 373 374 Activity against HepG2 cell cultures 375 The HepG2 (Human Caucasian hepatocyte carcinoma) cell line was procured from ATCC (American 376 Type Culture Collection, Manassas, USA; HB -8065) and viability assessed using CellTiter -Glo 377 luminescent cell viability assay reagents (Promega). For the assay, 2,000 cells/well were plated in 384-378 well plates 24 h prior to the experiment and incubated at 37°C in a CO 2 incubator. The medium was 379 removed; and cells were treated with fresh medium containing either vehicle (0.5% DMSO) or serially 380 diluted compounds or doxorubicin (1.3 nM to 25 μ M) in a final volume of 50 μ L/well and further 381 incubated for 72 h at 37°C in a CO 2 incubator. In the positive control wells (100% inhibition), cells 382 were treated with 5 μ L of 1% Triton X -100. Following incubation, 25 μ L of medium was discarded 383 and 25 μl of CellTiter-Glo reagent was added to each well and the plate was kept on a plate shaker for 384 15 min at 25°C with shaking at 300 rpm. Luminescence signals were measured in a PHERAstar FSX 385 reader (BMG LABTECH). 386 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 9 387 E1-E2 Transthiolation Assays 388 An Homogeneous Time -Resolved Fluorescence (HTRF) enzyme assay was employed to evaluate 389 compound activity against ATG7 as previously described [12]. In this assay, a Flag -tagged ATG8 390 homolog (GABA type A receptor -associated protein; GABARAP) is activated by ATG7 and then 391 transthiolated to a GST -tagged E2 (ATG3). The product of the enzyme reaction, Flag -GABARAP-392 ATG3-GST, is quantified by measuring FRET between Europium-Cryptate labelled monoclonal anti-393 Flag M2 ( FLAG M2-Eu cryptate; Revvity, Cat# 61FG2KLB) and goat polyclonal antibody against 394 mouse IgG conjugated to allophycocyanin (APC) ( Anti-GST IgG conjugated to SureLight ™-395 Allophycocyanin) (Revvity, Cat# AD0059G)). The activation and transthiolation of ubiquitin by UAE, 396 activation and transthiolation of NEDD8 by NAE and activation and transthiolation of SUMO by SAE 397 were all assayed in a similar fashion with appropriately tagged ubiquitin- like proteins and E2 398 conjugating enzymes as described [9, 25]. 399 400 Activity against panels of ex vivo field isolates of P . falciparum and P . vivax. 401 Compounds were assayed on ten P . vivax isolates and seven P . falciparum Brazilian isolates collected 402 from mono-infected patients, who had signed a written informed consent form to participate in the 403 study, using previously described methods [42] . The initial parasitemia ranged from 2,100-8,000 404 parasites/μL for P . vivax and 3,500-9,000 parasites/μL for P . falciparum isolates. Artesunate and 405 chloroquine were assayed in parallel as standard compounds. The analyses included only the isolates 406 that were incubated for ≥ 40 h with the compounds. 407 408 Parasite Reduction Rate (PRR) estimation 409 PRR was assessed using a standardized method [27]. P. falciparum (strain 3D7A, MRA -151), 410 contributed by David Walliker, was obtained through BEI Resources, NIAID, NIH. Cultures of 411 parasites (~90% ring stage) were treated with compounds for 120 h, with daily renewal. Samples of 412 parasites were taken at 0, 24, 48, 72, 96 and 120 h. Compounds were washed out and four independent 413 3-fold serial dilutions were established in 96-well plates, with fresh RBCs and culture medium. After 414 18 days, and again at 22 days, samples were taken to examine growth using SYBR Green I in an 415 EnVision Multilabel Plate Reader (Perkin Elmer) and analysed using Excel and Grafit 5.0 software. 416 The human biological samples were sourced ethically, and their research use was in accord with the 417 terms of the informed consents under an IRB/EC approved protocol. 418 419 Minimum inoculum of resistance 420 Minimum Inoculum of Resistance (MIR) studies were conducted for ML901 and ML471 using a 421 modified “Gate keeper assay” [34]. The in-house IC 50 for ML901 2.6 ± 0.05 nM (mean ± SEM; N,n = 422 2,2), while for ML471 the mean IC50 and IC90 values were determined as 1.45 and 1.99 nM, 423 respectively. For ML901, the parasites (starting parasite inoculums of 1x10^7 or 1x10^8 in triplicate) 424 were subjected to pressure at 3 x IC50, with recrudescence on days 12 -14. For ML471, parasites were 425 subjected to 10 x IC 50. Wells were monitored daily by smear during the first seven days to ensure 426 parasite clearance, during which media was changed daily. Thereafter, cultures were screened three 427 times weekly by flow cytometry and smearing, and the selection maintained a consistent drug pressure 428 over 60 days. In both cases, IC50 shifts were observed (ranging from two- to 16-fold). Whole-genome 429 sequencing analysis identified CNVs in chromosome 8 segments, always containing the Pf TyrRS 430 locus, consistent with our earlier evidence of this as the target of ML901 [16]. Single nucleotide 431 polymorphism (SNP) filtering was lowered to 0.5 allelic balance (AB) but yielded no point mutations 432 in the core genome. 433 434 Gametocyte killing assays 435 Gametocytogenesis was induced on a tightly synchronised (>97% rings) asexual parasite culture 436 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 10 (Pf3D7-pfs16-CBG99 (kind gift from Pietro Alano), 0.5% parasitemia and 6% hematocrit) with a 437 combination of nutrient starvation and a decrease in hematocrit, as previously described [43] . For 438 immature gametocytes (>90% stage II/III), cultures were exposed to 50 mM N -acetyl glucosamine 439 (NAG) on days 1–4 to eliminate residual asexual parasites and harvested at days 5 –6. For mature 440 (>95% stage V) gametocytes, NAG treatment occurred from days 3–7 and harvested at day 13. Stage 441 II/III and V gametocyte cultures (2 % gametocytaemia 1.5 % haematocrit, 150 μL/well) were exposed 442 to compounds and incubated at 37°C for 48 h under hypoxic conditions [44] , after which luciferase 443 activity was determined with a non-lysing D-luciferin substrate (1mM in 0.1M citrate buffer, pH 5.5, 444 100 µL) and bioluminescence was detected with a 2 s integration time with a GloMax®-Multi Detection 445 System with Instinct® software. 446 447 Activity against P . falciparum male and female gamete formation 448 Inhibition of the viability of stage V male and female gametocytes was assessed in the P . falciparum 449 Dual Gamete Formation Assay (PfDGFA) as described previously [45]. Briefly, mature stage V NF54 450 strain gametocytes were incubated with test molecules for 48 h in 384- well plates at 37°C. 451 Gametogenesis was then triggered by cooling the plates to room temperature and addition of 452 xanthurenic acid -containing activating medium (also containing anti -Pfs25 antibody (Mab 4B7) 453 conjugated to a Cy3 fluorophore ). Twenty minutes after activation, exflagellation was imaged in all 454 wells of the plate using a x4 objective and automated brightfield microscopy. The plate was then 455 incubated at 20°C for a further 24 h in the dark. Thereafter, female gamete formation was quantified 456 by automated identification of Pf s25-positive cells. Automated counts were transformed to percent 457 inhibition values with reference to positive 100 nM Cabamiquine (DDD498) and negative (DMSO) 458 controls. Data represent the means of multiple independent biological repeats. 459 460 Activity against liver stage P . falciparum 461 Activity against liver stage parasites was performed using a modification of published procedures [46, 462 47]. Briefly, cryopreserved human primary hepatocytes (NF175: H1500.H15B+ Lot No. HC0-6, 463 TebuBio or NF135: F00995-P Lot No. IRZ, BioIVT) were thawed and seeded at 18,000 cells per well 464 in collagen-coated 384-well microtiter plates in medium containing 10% heat inactivated fetal bovine 465 serum (hiFBS). Cells were cultured at 37°C in 5% CO 2. For NF175, medium was replaced by fresh 466 medium containing 10% hiFBS, 5 h post plating. For NF135, medium was replaced by medium 467 containing 0.2% BSA, 24 h post plating. 48 h post plating, salivary glands from Plasmodium -NF175 468 or NF135-infected Anopheles stephensi mosquitoes were dissected, added to the hepatocytes (10,000 469 per well/ NF175; or 12,500 per well/ NF135) and allowed to infect for 3 hours. Sporozoites were then 470 aspirated, and compounds diluted in medium containing 10% hiFBS or 0.2% BSA, were added to the 471 hepatocytes. Medium containing 10% hiFBS or 0.2% BSA and compounds was refreshed daily for 472 four days. Hepatocytes were fixed with ice-cold methanol and samples were blocked with 10% hiFBS 473 in PBS. Schizonts were stained with rabbit anti-HSP70 in 10% hiFBS for 1-2 h followed by incubation 474 with a mixture of secondary goat anti -rabbit AlexaFluor 594 antibody and 4′,6-diamidine -2′-475 phenylindole dihydrochloride (DAPI) in 10% hiFBS for 30 min. Samples were washed with PBS 476 containing 0.05% Tween 20 between different steps. Cells were imaged on a PicoExpress high content 477 imager and images were analy sed automatically using CellReporterXpress software. Data were 478 analysed by logistic regression using a four -parameter (Hill equation) model and a least -squares 479

to find the best fit. 480 481 Rat pharmacokinetics (PK) analyses 482 Sprague-Dawley rats (11 weeks old) were sourced from Hilltop Lab Animals, Inc (Scottdale, 483 Pennsylvania, USA). ML901 was formulated in ethanol: dimethyl acetamide (DMAc): PEG400: H2O 484 at 1:1:4:4 (v/v) and 10% captisol in 50 mM citrate (pH 3.3) for i.v. (1 mg/kg) and p.o. (10 mg/kg) 485 administration to male Sprague-Dawley rats (n = 3 per route of administration). Blood was collected 486 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 11 from a jugular cannula at 0.083, 0.25, 0.5, 1, 2, 4, 8 and 24 h post i.v. dosing, and at the same times 487 (except the 0.083- h sample) following oral administration. A portion of the blood samples were 488 processed into plasma. Samples were precipitated with 0.5% formic acid in methanol and the 489 supernatants were analysed by positive ion electrospray LC-MS for the administered compound. Non-490 compartmental pharmacokinetic parameters were calculated from individual concentration vs time 491 profiles using Phoenix 64 (WinNonlin) Version 8.1 Certara, Princeton NJ. 492 493 Permeability analysis 494 Permeability studies were performed as described previously [48]. In brief, Caco-2 cells were cultured 495 for 21-25 days to differentiate them into enterocyte-like cells. The transepithelial electrical resistance 496 (TEER) was measured to ensure tight junction formation and cells with TEER value more than 250 497 ohms.cm2 were used in the study. On the day of the transport study, cells were washed with warm 498 HBSS buffer and equilibrated with buffer for 60 min. ML471 was added at a concentration of 5 µM 499 (containing 50 µM Lucifer Yellow) into a 24 Transwell cell plate (apical 210 µ L and basal 1000 µL) 500 and buffer was added in the receiver side. Cells were incubated at 37°C for 60 min and 120 µL aliquots 501 were taken from the receiver side after 30 and 60 min. Samples were mixed with 100 nM carbutamide 502 in acetonitrile (ACN) containing 0.1% formic acid (internal standard). Samples were centrifuged at 503 2,056 x g for 10 min and the supernatant was collected and analyzed for quantification of the test 504 article by LC-MS [48]. 505 506 P . falciparum humanized NOD-scid IL2Rnull mouse model 507 The model using P . falciparum Pf3D7 0087/N9 in NODscidIL2Rγnull mice engrafted with human RBCs 508 was adapted from a previously described procedure [49]. Female NODscidIL2R γnull mice were 509 purchased from Charles River (Germany). Briefly, two engrafted mice/dosing group (females, 20 - 22 510 g) were infected intravenously with 2 x 10 7 P . falciparum (Pf3D70087/N9) on day 0. The antimalarial 511 efficacy was assessed following administration (p.o.) of 100 or 200 mg/kg of compound or of 4 daily 512 doses of 50 mg/kg, initiated on day 3 post-infection. The effect on blood parasitemia was measured by 513 microscopic analysis of Giemsa-stained blood smears (on days 3, 4, 5, 6 and 7 post -infection). Mice 514 were euthanized on day 7. 515 516 Plasma exposure in the infected mouse model 517 Compound was administered orally to two mice at 25 mg/kg on days 3, 4, 5, 6 after infection. On day 518 3, blood samples (20 μL) were obtained at time points up to 24 h after the first administration. Protein 519 was precipitated with acetonitrile and the remaining compound was assessed by LC-MS/MS in the selected 520 reaction monitoring mode using HESI ionization in positive ion mode. 521 522 Preparation of P . falciparum TyrRS 523 PfTyrRS was expressed and purified as previously reported [16]. Briefly, the vectors were transformed 524 into E. coli BL21(DE3) cells and induced for 3 h at 37°C with 0.1- 0.5 mM IPTG. Cell pellets were 525 resuspended in the lysis buffer containing 50 mM Tris-HCl, pH 8, 500 mM NaCl, 50 mM imidazole, 526 1 mM TCEP, 1 mg/mL lysozyme and 1x protease inhibitor cocktail (Roche). Cells were lysed by 527 sonication (Microtip, QSonica) and the lysate was clarified by centrifugation. The supernatant was 528 applied to a HisTrap HP column (GE Healthcare), washed , and eluted using a 0- 500 mM imidazole 529 gradient. The eluted His -PfTyrRS was dialyzed overnight with the addition of His -tagged TEV 530 protease (L56V/S135G/S219V triple-mutant). Cleaved His tags and TEV protease were removed by 531 running the dialyzed protein through a HisTrap HP column. Pf TyrRS was further purified by gel 532 filtration using a HiLoad 16/600 Superdex 200 column (GE Healthcare). 533 534 In vitro transcription/ translation of PftRNA Tyr 535 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 12 A T7 RNA polymerase promoter sequence was added to the 5’ end of the DNA sequence of PftRNATyr. 536 This DNA template and its complementary strand were custom- synthesised by Sigma-Aldrich. Two 537 oligonucleotides were annealed at 95°C for 5 min and the double -stranded DNA template was used 538 for in vitro transcription. The transcription reaction was incubated at 37°C overnight. The reaction 539 mixture consists of template DNA, T7 RNA polymerase and NTP mix as per manufacturer's 540 instructions (HiScribe™ T7 Quick High Yield RNA Synthesis Kit, NEB). On the following day, the 541 reaction mixture was treated with DNase I at 37°C for 15 min. tRNA was purified using Phenol: 542 Chloroform: Isoamyl alcohol (25:24:1, v/v), followed by 1M LiCl precipitation and isopropanol 543 precipitation. Purified PftRN ATyr was subjected to NAP -25 desalting columns (Cytiva) to remove 544 floating NTPs. The obtained solutions were concentrated with isopropanol precipitation and the 545 PftRNATyr pellets were dissolved in DEPC-treated water. 546 547 ATP consumption assays 548 Consumption of ATP was measured using a luciferase -based assay as per the manufacturer’s 549 instructions (Kinase-Glo Luminescent Kinase Assay, Promega). Reactions were conducted in 50 mM 550 Tris-HCl pH 7.6, 50 mM KCl, 25 mM MgCl2, 0.1 mg/mL BSA, 1 mM DTT, with 200 μM L-tyrosine, 551 48 μM PftRNATyr, 10 μM ATP, 25 nM PfTyrRS and 1 unit/mL inorganic pyrophosphatase (yeast) in 552 the presence or absence of 0 – 200 μM of inhibitors. Reactions were incubated at 37°C for 1 h. 553 554 Identification of amino acid- ML471 conjugates generated by ML471 -treated PfTyrRS and P. 555 falciparum cell culture 556 In vitro recombinant PfTyrRS reactions were set up with 2 μM enzyme, 20 μM tyrosine, 10 μM ATP, 557 10 μM ML471 and 4 μM PftRNATyr. The reaction buffer consisted of 25 mM Tris, pH 8, 150 mM 558 NaCl, 5 mM MgCl2 and 1 mM TCEP. The mixture was incubated at 37°C for 1 h. An equal volume 559 of 8 M urea was added to the mixture after the incubation, followed by trifluoroacetic acid to a final 560 level of 1%. The sample was centrifuged at 15,000 g for 10 min and the supernatant was used for mass 561 spectrometry analysis. 562 563 For the identification of adducts in parasite culture, aliquots of late trophozoite stage P. falciparum 564 (3D7 strain) culture were exposed to 1 μM ML471 for 2 h. Following treatment, parasite -infected 565 RBCs were lysed with 0.1% saponin and the pellet was washed 3 times with ice -cold PBS. P. 566 falciparum cell pellets were resuspended in one volume of water, followed by the addition of five 567 volumes of cold chloroform -methanol (2:1 [vol/vol]) solution. Samples were incubated on ice for 5 568 min, subjected to vortex mixing for 1 min and centrifuged at 13,500 x g for 10 min at 4°C to form 2 569 phases. The top aqueous layer was transferred to a new tube and subjected to LCMS analysis. 570 571 High-performance liquid chromatography (HPLC) and mass spectrometric (MS) analyses 572 Samples were analysed by reversed -phase ultra-high performance liquid chromatography (UHPLC) 573 coupled to tandem mass spectrometry (MS/MS) (Q Exactive, ThermoFisher Scientific). Samples (5 574 μL) were injected onto a Dionex Ultimate 3000 UHPLC system (ThermoFisher Scientific) and 575 analytical separation was performed with a RRHD Eclipse Plus C8 column (2.1 × 100 mm, 1.8 μm; 576 Agilent Technologies). The system was run at a flow rate of 300 μL/min using a binary gradient solvent 577 system consisting of 0.1% formic acid in water (solvent A) and 0.1% formic acid in acetonitrile 578 (solvent B). The gradient profile was as follows: 0–4.5 min, 3- 40% B; 4.5–4.6 min, 40–95% B; 4.6–579 5.5 min, 95% B; 5.5–5.8 min, 95–3% B and 5.8–8 min, 3% B. Full scan MS acquisition was performed 580 in polarity switching mode, with the following settings: resolution 35,000, 900 AGC target 1 × 10 6, 581 m/z range 85–1275, sheath gas 50, auxiliary gas 20, sweep gas 2, probe temperature 120°C, capillary 582 temperature 300°C and S-Lens RF level was set to 50. The spray voltage was set at 3.5 kV for positive 583 and negative ionization modes. All data shown for the Tyr-ML471 adduct were collected using positive 584 mode ionisation. 585 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 13 586 Differential scanning fluorimetry (DSF) 587 The effect on ML471 and analogues on the thermal stability of Pf TyrRS was assayed as previously 588 described [16]. Briefly, PfTyrRS (2.3 μM) was incubated with 10 μ M ATP, 20 μM L-tyrosine, 4 μM 589 PftRNATyr and various inhibitors at 37°C for 2 h. SYPRO Orange (Sigma-Aldrich; 5,000X concentrate 590 in DMSO) was added to the reaction mixture at a final concentration of 5X. 25 μ L of the sample was 591 added into each well of a 96-well qPCR plate (Applied Biosystems). The plate was sealed and analysed 592 using StepOnePlus Real-Time PCR system (Applied Biosystems). The samples were heated from 20°C 593 to 90°C with a 1% continuous gradient. The thermal unfolding curve was plotted as the first derivative 594 curve of the raw fluorescence values. The melting temperature (T m), defined as the peak of the first 595 derivative curve, was used to assess the thermal stability of protein-ligand complexes. 596 597 Crystallography 598 Recombinant PfTyrRS in complex with synthetic Tyr-ML471 was crystallised using the sitting drop 599 vapour diffusion technique at 20°C. C rystals were formed in 2.25 M sodium malonate, pH 6. Drops 600 contained 1.5 μL of protein-ligand solution (10 mg/mL PfTyrRS in Tris-HCl (25 mM, pH 8)), 100 mM 601 NaCl, 10 mM MgCl 2, 1 mM TCEP, 500 μM Tyr-ML471 synthetic ligand) and 1.5 μ L of crystallant 602 solution (2.25 M sodium malonate, pH 6). 603 604 Crystals were flash -cooled in liquid nitrogen directly from the crystallisation drop, and X -ray 605 diffraction data were collected at the MX2 beamline of the Australian Synchrotron [50] . Diffraction 606 data were indexed and integrated using XDS and analysed using POINTLESS, prior to merging by 607 AIMLESS [51] from the CCP4 software suite [52]. Initial phase estimates for Pf TyrRS in complex 608 with Tyr-ML471 were obtained by molecular replacement in PHASER [53] using modified crystal 609 structure coordinates of Pf TyrRS/ Tyr-ML901 as the search model (PDB ID: 7ROS, Xie 2022). 610 Automated structure refinement using phenix.refine [54] was followed iteratively by manual model 611 building in COOT [55]. Structure refinement was performed using non- crystallographic torsion 612 restraints and translation/libration screw (TLS) refinement with each chain comprising a single TLS 613 group. Restraints for Tyr-ML471 were generated using phenix.elbow [56]. Final data collection and 614 refinement statistics are shown in Supplementary Table 12. 615 616 Docking 617 Docking was carried out with the Surflex-Dock molecular docking module in SybylX2.1 (Certara, NJ, 618 USA). Docks were performed both with and without protein flexibility. Docking poses with the best 619 Surflex scores were inspected in SybylX2.1 and figures generated with PyMOL. 620 621 Chemistry 622 The synthes es of the compounds from the series have been reported previously [12, 22] . MMV 623 designations for the compounds are listed in Table 1. Methods used for resynthesis of ML471; and for 624 synthesis of Tyr-ML471 are presented in the Supplementary Material. Adenosine 5'-sulfamate (AMS) 625 [57] and was kindly provided by Dr Derek Tan, Memorial Sloan Kettering Cancer Center. 626 627 Ethics statement 628 Human biological samples were sourced ethically; and their research use was in accord with the terms 629 of the informed consent. Animal studies were ethically reviewed by the Institutional Animal Care and 630 Use Committee at GSK or by the ethical review process at the institution where the work was 631 performed and carried out in accordance with relevant countries' directives, European Directive 632 2010/63/EU, and institution's and GSK's Policy on Care, Welfare and Treatment of Animals. 633 Parasitology work and volunteer human blood donation (from healthy adult consenting volunteers) at 634 the University of Pretoria is covered under ethical approval from the Research Ethics Committee from 635 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 14 Health Sciences (506/2018) and Natural and Agricultural Sciences (180000094). Studies of P. v i v a x 636 isolates and P . falciparum Brazilian isolates were approved by the Ethics Committee of the Tropical 637 Medicine Research Center - CEPEM (CAAE 61442416.7.0000.0011). 638 639

640 We thank the following colleagues for technical contributions: Liver Stage Assay: Marloes de Bruijni 641 and Rob Henderson, TropIQ, Netherlands ; Caco2 Assays : Bei-Ching Chuang, Takeda 642 Pharmaceuticals, USA; SCID mouse assay : Ursula Lehmann, Swiss Tropical and Public Health 643 Institute, Switzerland, Christoph Siethoff, Swiss BioQuant; 3D7 parasite assays: TCG LifeSciences, 644 Kolkata, India; Assay coordination: Delphine Baud and Anna Adam, Medicines for Malaria Venture, 645 Switzerland; Mass Spectrometry: Shuai Nie and Nick Williamson, Melbourne Mass Spectrometry and 646 Proteomics Facility ; Crystallisation: Roxanne Smith, Bio21- WEHI Crystallisation facility ; Protein 647 Purification: Yee-Foong Mok, Melbourne Protein Facility , Bio21 Institute . We thank Hirotake 648 Mizutani, Takeda Pharmaceuticals and Winnie Ye, University of Melbourne, for technical help. Thanks 649 also extended to Heekuk Park, Sachel Mok and Anne-Catrin Uhlemann for whole-genome sequencing 650 and analysis at the Columbia University Irving Medical Center. We thank Dr Derek Tan, Memorial 651 Sloan Kettering Cancer Center , for supplying adenosine 5'- sulfamate. We thank the Australian Red 652 Cross for supply of blood products. GSK acknowledges the Centro de Hemoterapia y Donación de 653 Valladolid, Castilla y León, and the Centro de Transfusiones de la Comunidad de Madrid for the supply 654 of blood samples. This research was partly undertaken at the Australian Synchrotron, part of the 655 Australian Nuclear Science and Technology Organization, and made use of the ACRF Detector on the 656 MX2 beamline. We thank the beamline staff for their assistance. 657 658 Grant funding 659 We would like to acknowledge funding from the Global Health Innovative Technology Fund, Japan 660 (H2019-104; LT, LD, SL, AEG ), the Australian National Health and Medical Research Council 661 (APP2022075; to LT), the Medicines for Malaria Venture (LMB: RD-19-001; DAF: RD-08-0015), the 662 Foundation for Research Support of the State of São Paulo (FAPESP; 2019/19708-0 and 2013/07600-663 3), the South African Medical Research Council , the Department of Science and Innovation South 664 African Research Chairs Initiative Grants managed by the National Research Foundation (LMB UID: 665 84627) and a Medical Research Council Career Development Award (MR/V010034/1) awarded to 666 MJD. MTF is supported by an MMV grant (RD-21-1003) awarded to MJD. The University of Pretoria 667 Institute for Sustainable Malaria Control acknowledges the South African Medical Research Council 668 as a Collaborating Centre for Malaria Research. We acknowledge support from Millennium 669 Pharmaceuticals, a wholly owned subsidiary of Takeda Pharmaceuticals Company Limited. 670 Data Availability 671 Additional data are available in Supplementary Information. Source data are provided. The following 672 structures have been deposited in the PDB: PfTyrRS/Tyr-ML471 - PDB ID 9CLL. 673 Author Contributions 674 Conceptualisation: S.C.X., C-W.T., C.J.M., L.M., S.W., C.D., F.J.G, C.H., D.A.F., L.R.D., S.L.B., 675 A.E.G., S.L., M.D.W.G., L.T.; Investigation: S.C.X., C-W.T., C.J.M., L.M., S.H., S.W., Y.D., Y.H., 676 C.D., R.G., D.E., E dl C., I.D., T.Y., A.Y.B., J.S., K.A.S., B.C., Y.K., M.S.S., T.R., M.F., M.D., J.B., 677 K.M.J.K., R.vdL., A.C.C.A., D.B.P.; Analysis: S.C.X., C-W.T., C.J.M., L.M., S.H., S.W., Y.D., Y.H., 678 C.D., R.G., D.E., E.dl C., I.D., T.Y., A.Y.B., J.S., K.A.S., B.C., Y.K., M.S.S., T.R., L -M.B., M.F., 679 M.D., J.B., K.M.J.K., R.vdL., A.C.C.A., D.B.P., R.V.C.G., D.A.F., L.R.D., S.L.B., A.E.G., S.L., 680 M.D.W.G., L.T.; Funding acquisition: S.C.X., S.W., F.J.G, C.H., L -M.B., R.V.C.G,D.J.C., D.A.F., 681 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 15 L.R.D., S.L.B., A.E.G., S.L., M.D.W.G., L.T.; Writing: S.C.X., C-W.T., C.J.M., L.M., S.W., M.S.S., 682 C.H., D.A.F., L.R.D., S.L.B., A.E.G., S.L., M.D.W.G., L.T. 683 684 Competing interests. The authors have no competing interests to declare, while noting that employees 685 of Takeda Pharmaceuticals are owners of Takeda stock. 686 687 Figure Legends 688 689 Figure 1. Structures of ML901 and derivatives and adenosine 5’-sulfamate (AMS). 690 (A) ML901, (B) ML471, (C) ML676, (D) ML681, (E) ML723, (F) ML107, (G) ML470, (H) ML864, 691 (I) ML111, (J) AMS, (K) Tyr-ML471. 692 693 Figure 2. ML471 exhibits improved short -exposure activity against P . falciparum cultures, 694 associated with rapid parasite killing 695 (A) Synchronized Cam3.IIrev parasite cultures were subjected to 6-h pulses of ML901, ML471, ML107 696 and ML723, at the trophozoite (25- 30 h.p.i.) stage. Growth inhibition was determined in the cycle 697 following treatment. Data represent the mean of three independent experiments and error bars 698 correspond to SEM. (B) 3D7 parasite cultures were treated for 0 to 120 h with ML471 or compounds 699 with fast (artemisinin, chloroquine), moderate (pyrimethamine) or slow (atovaquone) killing profiles, 700 at 10 times their respective IC50_48h values. Following removal of inhibitor, serial dilutions of cultures 701 were established, and assessed after 18 days of culturing. 702 703 Figure 3. Pharmacokinetics profiles and in vivo efficacy of ML471 704 (A, B) Rat pharmacokinetics for ML471. Rats were dosed with ML471 at 1 mg/kg i.v. (blue) or 1, 10 705 or 25 mg/kg p.o. (green, red, orange) and plasma (A) and blood (B) samples were collected for analysis. 706 See Supplementary Table S7 for pharmacokinetics values. (C) Pharmacokinetics profile (in blood), for 707 SCID mice engrafted with human RBCs infected with P . falciparum, over the first day following 708 treatment with ML471 at 100 or 200 mg/kg p.o.. See Supplementary Table S8 for pharmacokinetics 709 values. (D) Therapeutic efficacy of ML471 in the SCID mouse P . falciparum model, dosed with 710 ML471 at 100 or 200 mg/kg p.o. on Day 3 post -infection (arrowed). The chloroquine data are from 711 [16]. 712 713 Figure 4. Identification of ML471 conjugates in P . falciparum and effects of pro -inhibitors on 714 enzyme stability and activity. P. falciparum-infected RBCs were treated with 1 µ M ML471 for 2 h. 715 Extracts were subjected to LCMS and the expected mass for amino acid-ML471 conjugates searched. 716 (A) The extracted ion chromatograms of the Tyr-ML471 adduct made by a P . falciparum culture (upper 717 panel) and the synthetic conjugate at 0.2 μ M (lower panel). The inset shows the MS analysis of the 718 parasite-generated Tyr-ML471, and the structure of Tyr -ML471. Profiles are typical of data from 3 719 independent experiments. (B,C) First derivatives of melting curves for PfTyrRS (B) and HsTyrRS (C) 720 (2.3 μM) in the apo form or after incubation at 37°C with ML901, ML471 or AMS, in the presence of 721 10 μ M ATP and 20 μ M tyrosine. For Pf TyrRS, 50 μ M pro -inhibitor and 4 μ M PftRNA Tyr were 722 incubated with substrates for 2 h. For HsTyrRS, 200 μM pro-inhibitor and 8 mg/mL yeast tRNA were 723 incubated with substrates for 4 h. Data are representative of three independent experiments. ( D) AT P 724 consumption by PfTyrRS in the presence and absence of the cognate tRNATyr. ATP consumption in the 725 absence of tRNA Tyr derives from turnover of Tyr- AMP generated in the initial phase of the TyrRS 726 reaction. The reaction component concentrations are: Pf TyrRS (25 nM), ATP (10 μM), tyrosine (200 727 μM), pyrophosphatase (1 unit/mL) and cognate tRNATyr (4.8 μM), if present; and incubations were at 728 37°C for 1 h. Data are the average of three independent experiments and error bars correspond to SEM. 729 (E) Effects of increasing concentrations of ML471, ML901 and AMS on ATP consumption by 730 PfTyrRS. Assay conditions are the same as in (D), with cognate tRNATyr. Data represent mean + SEM 731 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 16 from three or four independent experiments. 732 733 Figure 5. Docking of ML901 and ML471 into structures of PfTyrRS and UAE provides insights 734 into selectivity. (A) Active site of Pf TyrRS/Tyr-ML901 (7ROS) B-chain (His70 depicted in green) 735 with docked ML901 (aqua carbons). The model is overlayed with ML901 (depicted with yellow 736 carbons) with the pose adopted upon docking into the A-chain. (B) Active site of PfTyrRS/Tyr-ML901 737 (7ROS) B-chain (His70 depicted in green) with docked ML471 (aqua carbons). The model is overlayed 738 with ML471 (depicted with yellow backbone) with the pose adopted upon docking into the A -chain. 739 The red arrow illustrates the different conformations adopted by the difluoromethoxy and isopropyl 740 groups. (C, D) The structure of 7ROS B-chain with bound Tyr-ML901 is overlayed with B -chain-741 docked ML901 (C) and ML471 (D). The red arrows illustrate the different conformations adopted by 742 the difluoromethoxy and isopropyl groups. The purple arrows illustrate the twisted ribose group in the 743 Tyr-ML901 conjugate. By contrast in the docked pro-inhibitors, the rings systems are co-planar. (E,F) 744 ML901 (E) and ML471 (F) were docked into the ATP-binding site of human UAE (6DC6). A H-bond 745 made by ML901 with residue Arg 551 is indicated with a red arrow. Asn577 and Arg551 (blue arrows) 746 flank the hydrophobic isopropyl group in the ML471 dock. 747 748 Figure 6. Comparison of the crystal structures of Tyr-ML471- and Tyr-ML901-bound PfTyrRS 749 reveals differential mobility of the “KMSKS” loop. 750 (A) Crystal structure of t he dimeric PfTyrRS/Tyr-ML471 complex showing chain A (green), chain B 751 (blue), and bound Tyr-ML471 (red, stick representation). (B) Architecture of the B-chain of PfTyrRS 752 with bound Tyr-ML471, showing direct interactions with active site residues. (C) B -chain of Tyr-753 ML471-bound PfTyrRS showing the poses adopted by the ML471 isopropyl group (blue arrow) and 754 His70 (H70), which are incompatible with a structured KMSKS loop. (D) B- chain of Tyr-ML901-755 bound PfTyrRS (7ROS). The conformation of the ML901 difluoromethoxy group (red arrow) allows 756 His70 to interact with Met248 of the KMSKS loop, leading to stabilisation. (E) Overlay of the B -757 chains of Tyr-ML471- and Tyr-ML901-bound PfTyrRS. 758 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 17 Tables 759 760 Table 1. Activities of pyrazolopyrimidine sulfamates as inhibitors of parasite growth. AMS = Adenosine 761 5'-sulfamate. n = Number of biological repeats. Data values represent mean ± SEM. Medicines for Malaria 762 Venture (MMV) designations are in brackets. 763 Compound P. falciparum (3D7) H. sapiens (HepG2) 72-h IC50 (nM) 6-h Trophozoite Stage IC50 (nM) 72-h IC50 (nM) ML901 (MMV1581329) 2.8 + 0.2 (n = 3) 135 + 14 (n = 3) 4,650 + 1,390 (n = 3) ML471 (MMV1793207) 1.5 + 0.2 (n = 3) 29.1 + 3.0 (n = 3) > 50,000 (n = 3) ML676 (MMV1793313) 2.8 + 0.2 (n = 3) N/A 22,400 + 7,300(n = 3) ML681 (MMV1793314) 3.4 + 0.1(n = 3) N/A 7,530 + 1,250 (n = 3) ML723 (MMV1793208) 2.5 + 0.6 (n = 3) 220 + 31 (n = 3) > 50,000 (n = 3) ML107 (MMV1793318) 4.2 + 0.5 (n = 3) 150 + 33 (n = 3) 2,520 + 820 (n = 3) ML470 (MMV1793342) 44.8 + 8.2 (n = 4) N/A > 50,000 (n = 3) ML864 (MMV1793301) 2.5 + 0.6 (n = 3) N/A 21,300 + 1,700 (n = 3) ML111 (MMV1793328) 4.3 + 0.7 (n = 3) N/A > 50,000 (n = 3) AMS 1.8 ± 0.6 (n = 3)* N/A N/A * Data from [16] 764 765 Table 2. Inhibitory activity of selected pyrazolopyrimidine sulfamates in E1 enzymes assays. ATG7 = 766 autophagy-related protein-7. NAE = NEDD8 -activating enzyme. UAE = ubiquitin activating enzyme. SAE = 767 SUMO-activating enzyme. GABARAP = GABAA receptor-associated protein. HTRF = Homogeneous Time-768 Resolved Fluorescence. Data represent mean + SEM. n = Number of independent experiments. 769 Compound ATG7 IC50 HTRF RH-GABARAP (µM) NAE IC50 HTRF (µM) UAE IC50 HTRF (µM) SAE IC50 HTRF (µM) ML901 0.033 ± 0.003 (n = 57)* 28.0 ± 0.6 (n=3) 5.39 ± 0.160 (n = 3) >100 (n = 3) ML471 0.022 ± 0.009 (n = 6) >100 (n=3) 85.7 ± 6.5 (n=3) >100 (n = 3) ML681 N.A. 35.3 ± 0.8 (n=3) 8.7 ± 0.6 (n=3) 22.4 ± 3.1 (n=3) (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 18 ML676 N.A. 66.6 ± 1.1 (n=3) 73.4 ± 3.5 (n=3) 31.4 ± 5.7 (n=3) ML723 N.A. >100 (n=3) >100 (n=3) >100 (n=3) ML470 N.A. >100 (n=3) >100 (n=3) >100 (n=3) ML107 N.A. 75.3 ± 0.8 (n=3) 22.9 ± 0.7 (n=3) 78.5 ± 6.2 (n=3) AMS* 410 ± 20 (n = 90) 0.006 ± 0.001 (n = 9) 0.006 ± 0.003 (n = 3) 0.006 ± 0.002 (n = 7) *Data from [16] 770 771 Table 3. Activities of pyrazolopyrimidine sulfamates in selected biochemical assays. AMS = Adenosine 772 5'-sulfamate. n = Number of biological repeats. Data values represent mean ± SEM. 773 The Tm values for PfTyrRS and HsTyrRS (2.3 μM) was measured in the apo form or after incubation 774 at 37°C for 2 h with the pro-inhibitors (50 μ M with PfTyrRS and 200 μM with Hs TyrRS) in the 775 presence of 10 μM ATP, 20 μM tyrosine, 4 μM cognate tRNATyr (PfTyrRS) or 8 mg/mL yeast tRNA 776 (HsTyrRS). KD values are estimated from differential scanning fluorimetry (DSF) analysis using an 777 irreversible protein thermal unfolding model that has been described previously [58]. 778 779 Compound Inhibition of PfTyrRS (Kinase Glo) PfTyrRS binding (DSF) (n = 3) HsTyrRS binding (DSF) (n = 3) IC50 (µM) Delta Tm (°C) apparent KD* (x10-9) (M) Delta Tm (°C) apparent KD (x10-9) (M) ML901 13 ± 3 (n = 3) 15.2 ± 0.1 0.7 0.4 ± 0.1 N/A ML471 1.4 ± 0.2 (n = 3) 18.0 ± 0.1 0.2 0.06 ± 0.04 N/A ML676 48 ± 16 (n = 5) 15.69 ± 0.01 0.6 N/A N/A ML681 23 ± 8 (n = 3) 16.4 ± 0.1 0.4 N/A N/A ML723 14 ± 5 (n = 3) 16.2 ± 0.1 0.5 N/A N/A ML107 24 ± 5 (n = 3) 13.6 ± 0.6 1.4 N/A N/A ML470 41 ± 23 (n = 4) 15.7 ± 0.3 0.6 N/A N/A ML864 8.5 ± 4.5 (n = 4) 17 ± 0.1 0.3 N/A N/A ML111 34 ± 14 (n = 3) 15.1 ± 0.2 0.7 N/A N/A AMS 52 ± 16 (n = 4) 13.6 ± 0.7 1.3 9.1 ± 0.1 8.0 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 19 780 781 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 20

782 1. World_Health_Organisation. WHO World Malaria Report 2023. Geneva: World Health 783 Organization; Licence: CC BY-NC-SA 30 IGO. 2023. 784 2. Rogerson SJ, Beeson JG, Laman M, Poespoprodjo JR, William T , Simpson JA, et al. Identifying 785 and combating the impacts of COVID-19 on malaria. BMC Medicine. 2020;18(1):239. 786 3. van der Pluijm RW , Imwong M, Chau NH, Hoa NT , Thuy-Nhien NT , Thanh NV, et al. 787 Determinants of dihydroartemisinin-piperaquine treatment failure in Plasmodium falciparum 788 malaria in Cambodia, Thailand, and Vietnam: a prospective clinical, pharmacological, and genetic 789 study. Lancet Infect Dis. 2019;19(9):952-61. 790 4. Imwong M, Dhorda M, Myo Tun K, Thu AM, Phyo AP , Proux S, et al. Molecular epidemiology 791 of resistance to antimalarial drugs in the Greater Mekong subregion: an observational study. Lancet 792 Infect Dis. 2020;20(12):1470-80. 793 5. Straimer J, Gandhi P , Renner KC, Schmitt EK. High prevalence of Plasmodium falciparum K13 794 mutations in Rwanda is associated with slow parasite clearance after treatment with artemether-795 lumefantrine. The Journal of infectious diseases. 2021;225(8):1411-4. 796 6. Balikagala B, Fukuda N, Ikeda M, Katuro OT , Tachibana S-I, Yamauchi M, et al. Evidence of 797 artemisinin-resistant malaria in Africa. New England Journal of Medicine. 2021;385(13):1163-71. 798 7. Mihreteab S, Platon L, Berhane A, Stokes BH, Warsame M, Campagne P , et al. Increasing 799 prevalence of artemisinin-resistant HRP2-negative malaria in Eritrea. N Engl J Med. 800 2023;389(13):1191-202. 801 8. Burrows JN, Duparc S, Gutteridge WE, Hooft van Huijsduijnen R, Kaszubska W , Macintyre F, 802 et al. New developments in anti-malarial target candidate and product profiles. Malaria Journal. 803 2017;16(1):26. 804 9. Soucy TA, Smith PG, Milhollen MA, Berger AJ, Gavin JM, Adhikari S, et al. An inhibitor of 805 NEDD8-activating enzyme as a new approach to treat cancer. Nature. 2009;458(7239):732-6. 806 10. Brownell JE, Sintchak MD, Gavin JM, Liao H, Bruzzese FJ, Bump NJ, et al. Substrate-assisted 807 inhibition of ubiquitin-like protein-activating enzymes: the NEDD8 E1 inhibitor MLN4924 forms a 808 NEDD8-AMP mimetic in situ. Mol Cell. 2010;37(1):102-11. 809 11. Hyer ML, Milhollen MA, Ciavarri J, Fleming P , Traore T , Sappal D, et al. A small-molecule 810 inhibitor of the ubiquitin activating enzyme for cancer treatment. Nat Med. 2018;24(2):186-93. 811 12. Huang S-C, Adhikari S, Brownell JE, Calderwood EF , Chouitar J, D'Amore NR, et al. Discovery 812 and optimization of pyrazolopyrimidine sulfamates as ATG7 inhibitors. Bioorganic & medicinal 813 chemistry. 2020;28(19):115681. 814 13. Milhollen MA, Thomas MP , Narayanan U, Traore T , Riceberg J, Amidon BS, et al. Treatment-815 emergent mutations in NAEβ confer resistance to the NEDD8-activating enzyme inhibitor MLN4924. 816 Cancer Cell. 2012;21(3):388-401. 817 14. Ferris J, Espona-Fiedler M, Hamilton C, Holohan C, Crawford N, McIntyre AJ, et al. 818 Pevonedistat (MLN4924): mechanism of cell death induction and therapeutic potential in colorectal 819 cancer. Cell Death Discovery. 2020;6(1):61. 820 15. Fu D-J, Wang T. Targeting NEDD8-activating enzyme for cancer therapy: developments, 821 clinical trials, challenges and future research directions. Journal of Hematology & Oncology. 822 2023;16(1):87. 823 16. Xie SC, Metcalfe RD, Dunn E, Morton CJ, Huang S-C, Puhalovich T , et al. Reaction hijacking of 824 tyrosine tRNA synthetase as a new whole-of-life-cycle antimalarial strategy. Science. 825 2022;376(6597):1074-9. 826 17. Xie SC, Wang Y , Morton CJ, Metcalfe RD, Dogovski C, Pasaje CFA, et al. Reaction hijacking 827 inhibition of Plasmodium falciparum asparagine tRNA synthetase. Nat Commun. 2024;15(1):937. 828 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 21 18. Dogovski C, Xie SC, Burgio G, Bridgford J, Mok S, McCaw JM, et al. Targeting the cell stress 829 response of Plasmodium falciparum to overcome artemisinin resistance. PLoS Biol. 830 2015;13(4):e1002132. 831 19. Fu Y , Tilley L, Kenny S, Klonis N. Dual labeling with a far red probe permits analysis of growth 832 and oxidative stress in P . falciparum-infected erythrocytes. Cytometry A. 2010;77(3):253-63. 833 20. Bloch A, Coutsogeorgopoulos C. Inhibition of protein synthesis by 5'-sulfamoyladenosine. 834 Biochemistry. 1971;10(24):4395-8. 835 21. Florini JR, Bird HH, Bell PH. Inhibition of protein synthesis in vitro and in vivo by nucleocidin, 836 an antitrypanosomal antibiotic. J Biol Chem. 1966;241(5):1091-8. 837 22. Adhikari S, Calderwood EF, England DB, Gould AE, Harrison SJ, Huang S-C, et al. Atg7 838 inhibitors and the uses thereof. World Intellectual Property Organization. 2017;WO/2018/089786 839 PCT/US2017/061094. 840 23. Mandelbaum J, Rollins N, Shah P , Bowman D, Lee JY , Tayber O, et al. Identification of a lung 841 cancer cell line deficient in atg7-dependent autophagy. Autophagy. 2015;(in press):00-. 842 24. Schulman BA, Harper JW. Ubiquitin-like protein activation by E1 enzymes: the apex for 843 downstream signalling pathways. Nat Rev Mol Cell Biol. 2009;10(5):319-31. 844 25. Chen JJ, Tsu CA, Gavin JM, Milhollen MA, Bruzzese FJ, Mallender WD, et al. Mechanistic 845 studies of substrate-assisted inhibition of ubiquitin-activating enzyme by adenosine sulfamate 846 analogues. J Biol Chem. 2011;286(47):40867-77. 847 26. Baragaña B, Hallyburton I, Lee MC, Norcross NR, Grimaldi R, Otto TD, et al. A novel multiple-848 stage antimalarial agent that inhibits protein synthesis. Nature. 2015;522(7556):315-20. 849 27. Sanz LM, Crespo B, De-Cozar C, Ding XC, Llergo JL, Burrows JN, et al. P . falciparum in vitro 850 killing rates allow to discriminate between different antimalarial mode-of-action. PloS one. 851 2012;7(2):e30949. 852 28. Supuran CT. Carbonic anhydrase inhibitors. Bioorg Med Chem Lett. 2010;20(12):3467-74. 853 29. Boddy A, Edwards P , Rowland M. Binding of sulfonamides to carbonic anhydrase: influence 854 on distribution within blood and on pharmacokinetics. Pharm Res. 1989;6(3):203-9. 855 30. Angulo-Barturen I, Jimenez-Diaz MB, Mulet T , Rullas J, Herreros E, Ferrer S, et al. A murine 856 model of falciparum-malaria by in vivo selection of competent strains in non-myelodepleted mice 857 engrafted with human erythrocytes. PloS one. 2008;3(5):e2252. 858 31. Jimenez-Diaz MB, Mulet T , Gomez V, Viera S, Alvarez A, Garuti H, et al. Quantitative 859 measurement of Plasmodium-infected erythrocytes in murine models of malaria by flow cytometry 860 using bidimensional assessment of SYTO-16 fluorescence. Cytometry A. 2009;75(3):225-35. 861 32. Cowell AN, Istvan ES, Lukens AK, Gomez-Lorenzo MG, Vanaerschot M, Sakata-Kato T , et al. 862 Mapping the malaria parasite druggable genome by using in vitro evolution and chemogenomics. 863 Science. 2018;359(6372):191-9. 864 33. Luth MR, Gupta P , Ottilie S, Winzeler EA. Using in vitro evolution and whole genome analysis 865 to discover next generation targets for antimalarial drug discovery. ACS Infect Dis. 2018;4(3):301-14. 866 34. Duffey M, Blasco B, Burrows JN, Wells TNC, Fidock DA, Leroy D. Assessing risks of 867 Plasmodium falciparum resistance to select next-generation antimalarials. Trends in Parasitology. 868 2021;37(8):709-21. 869 35. Lv Z, Williams KM, Yuan L, Atkison JH, Olsen SK. Crystal structure of a human ubiquitin E1-870 ubiquitin complex reveals conserved functional elements essential for activity. J Biol Chem. 871 2018;293(47):18337-52. 872 36. Shen N, Guo L, Yang B, Jin Y , Ding J. Structure of human tryptophanyl-tRNA synthetase in 873 complex with tRNATrp reveals the molecular basis of tRNA recognition and specificity. Nucleic Acids 874 Res. 2006;34(11):3246-58. 875 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 22 37. Gamo FJ, Sanz LM, Vidal J, de Cozar C, Alvarez E, Lavandera JL, et al. Thousands of chemical 876 starting points for antimalarial lead identification. Nature. 2010;465(7296):305-10. 877 38. Klonis N, Xie SC, McCaw JM, Crespo-Ortiz MP , Zaloumis SG, Simpson JA, et al. Altered 878 temporal response of malaria parasites determines differential sensitivity to artemisinin. Proc Natl 879 Acad Sci U S A. 2013;110(13):5157-62. 880 39. Straimer J, Gnadig NF, Witkowski B, Amaratunga C, Duru V, Ramadani AP , et al. K13-propeller 881 mutations confer artemisinin resistance in Plasmodium falciparum clinical isolates. Science. 882 2014;Epub ahead of print. 883 40. Klonis N, Crespo-Ortiz MP , Bottova I, Abu-Bakar N, Kenny S, Rosenthal PJ, et al. Artemisinin 884 activity against Plasmodium falciparum requires hemoglobin uptake and digestion. Proc Natl Acad 885 Sci U S A. 2011;108(28):11405-10. 886 41. Johnson JD, Dennull RA, Gerena L, Lopez-Sanchez M, Roncal NE, Waters NC. Assessment and 887 continued validation of the malaria SYBR green I-based fluorescence assay for use in malaria drug 888 screening. Antimicrob Agents Chemother. 2007;51(6):1926-33. 889 42. Aguiar AC, Pereira DB, Amaral NS, De Marco L, Krettli AU. Plasmodium vivax and Plasmodium 890 falciparum ex vivo susceptibility to anti-malarials and gene characterization in Rondonia, West 891 Amazon, Brazil. Malar J. 2014;13:73. 892 43. Reader J, Botha M, Theron A, Lauterbach SB, Rossouw C, Engelbrecht D, et al. Nowhere to 893 hide: interrogating different metabolic parameters of Plasmodium falciparum gametocytes in a 894 transmission blocking drug discovery pipeline towards malaria elimination. Malar J. 2015;14:213. 895 44. Reader J, van der Watt ME, Birkholtz LM. Streamlined and robust stage-specific profiling of 896 gametocytocidal compounds against Plasmodium falciparum. Front Cell Infect Microbiol. 897 2022;12:926460. 898 45. Delves MJ, Straschil U, Ruecker A, Miguel-Blanco C, Marques S, Dufour AC, et al. Routine in 899 vitro culture of P . falciparum gametocytes to evaluate novel transmission-blocking interventions. 900 Nat Protoc. 2016;11(9):1668-80. 901 46. Miglianico M, Bolscher JM, Vos MW , Koolen KJM, de Bruijni M, Rajagopal DS, et al. 902 Assessment of the drugability of initial malaria infection through miniaturized sporozoite assays and 903 high-throughput screening. Commun Biol. 2023;6(1):216. 904 47. Yang ASP , van Waardenburg YM, van de Vegte-Bolmer M, van Gemert GA, Graumans W , de 905 Wilt JHW , et al. Zonal human hepatocytes are differentially permissive to Plasmodium falciparum 906 malaria parasites. Embo J. 2021;40(6):e106583. 907 48. Yang JJ, Milton MN, Yu S, Liao M, Liu N, Wu JT , et al. P-glycoprotein and breast cancer 908 resistance protein affect disposition of tandutinib, a tyrosine kinase inhibitor. Drug Metab Lett. 909 2010;4(4):201-12. 910 49. Jimenez-Diaz MB, Mulet T , Viera S, Gomez V, Garuti H, Ibanez J, et al. Improved murine 911 model of malaria using Plasmodium falciparum competent strains and non-myelodepleted NOD-scid 912 IL2Rgammanull mice engrafted with human erythrocytes. Antimicrob Agents Chemother. 913 2009;53(10):4533-6. 914 50. Aragão D, Aishima J, Cherukuvada H, Clarken R, Clift M, Cowieson NP , et al. MX2: a high-flux 915 undulator microfocus beamline serving both the chemical and macromolecular crystallography 916 communities at the Australian Synchrotron. J Synchrotron Radiat. 2018;25(Pt 3):885-91. 917 51. Evans PR, Murshudov GN. How good are my data and what is the resolution? Acta 918 Crystallogr D. 2013;69:1204-14. 919 52. Winn MD, Ballard CC, Cowtan KD, Dodson EJ, Emsley P , Evans PR, et al. Overview of the CCP4 920 suite and current developments. Acta Crystallogr D. 2011;67:235-42. 921 53. Mccoy AJ, Grosse-Kunstleve RW , Adams PD, Winn MD, Storoni LC, Read RJ. Phaser 922 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint 23 crystallographic software. Journal of Applied Crystallography. 2007;40:658-74. 923 54. Adams PD, Afonine PV, Bunkoczi G, Chen VB, Davis IW , Echols N, et al. PHENIX: a 924 comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr D. 925 2010;66:213-21. 926 55. Emsley P , Lohkamp B, Scott WG, Cowtan K. Features and development of Coot. Acta 927 Crystallographica Section D-Biological Crystallography. 2010;66:486-501. 928 56. Moriarty NW , Grosse-Kunstleve RW , Adams PD. Electronic ligand builder and optimization 929 workbench (eLBOW): A tool for ligand coordinate and restraint generation. Acta Crystallographica 930 Section D: Biological Crystallography. 2009;65:1074-80. 931 57. Mujumdar P , Bua S, Supuran CT , Peat TS, Poulsen SA. Synthesis, structure and bioactivity of 932 primary sulfamate-containing natural products. Bioorg Med Chem Lett. 2018;28(17):3009-13. 933 58. Hall J. A simple model for determining affinity from irreversible thermal shifts. Protein Sci. 934 2019;28(10):1880-7. 935 936 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint Figure 1 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint Figure 2 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint Figure 3 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint Figure 4 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint Figure 5 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint Figure 6 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 23, 2024. ; https://doi.org/10.1101/2024.07.22.604682doi: bioRxiv preprint