Reaction hijacking of tyrosine tRNA synthetase as a new whole-of-life-cycle antimalarial strategy

Abstract

Aminoacyl transfer RNA (tRNA) synthetases (aaRSs) are attractive drug targets, and we present class I and II aaRSs as previously unrecognized targets for adenosine 5'-monophosphate-mimicking nucleoside sulfamates. The target enzyme catalyzes the formation of an inhibitory amino acid-sulfamate conjugate through a reaction-hijacking mechanism. We identified adenosine 5'-sulfamate as a broad-specificity compound that hijacks a range of aaRSs and ML901 as a specific reagent a specific reagent that hijacks a single aaRS in the malaria parasite Plasmodium falciparum, namely tyrosine RS (PfYRS). ML901 exerts whole-life-cycle-killing activity with low nanomolar potency and single-dose efficacy in a mouse model of malaria. X-ray crystallographic studies of plasmodium and human YRSs reveal differential flexibility of a loop over the catalytic site that underpins differential susceptibility to reaction hijacking by ML901.

Figure 1. AMS-treated infected RBCs reveals aaRSs as potential targets.

(A) E1 enzymes can catalyze attack of the sulfamate nitrogen on the carbonyl carbon of the thioester bond between the UBL and the E1 to form a UBL conjugate. aaRSs could catalyze nucleoside sulfamate attack on activated amino acids to form an amino acid adduct. (B) Structure of 5′-adenylate sulfamate. (C) Trophozoite stage parasites were incubated with DMSO (Mock), different concentrations of AMS, or, borrelidin (BOR; a threonyl-tRNA synthetase inhibitor). Western blots of lysates were probed for phosphorylated-eIF2α with PfBiP as a loading control. The blot is typical of data from three independent experiments. (D) P. falciparum-infected RBCs were treated with 10 μM AMS for 3 h. Extracts were subjected to LCMS analysis identifying the Tyr-AMS conjugate. The profile is typical of data from three independent experiments.

Figure 2. ML901 exhibits potent activity against P. falciparum in vivo.

(A) Structure of pyrazolopyrimidine ribose sulfamate, ML901. (B) Pharmacokinetics profile (in blood) over the first day for SCID mice engrafted with human RBCs infected with P. falciparum following treatment with ML901 at 50 mg/kg i.p. (C) Therapeutic efficacy of ML901 in the SCID mouse P. falciparum model, dosed with ML901 at 50 mg/kg i.p. in comparison with gold standard antimalarial, chloroquine, dosed at 50 mg/kg p.o..

Figure 3. ML901 targets PfYRS and inhibits protein translation.

(A,B) Sensitivity to ML901 exposure (72-h) for a cloned wildtype line (Dd2) and 3 CRISPR-edited clones harboring PfYRS_S234C (A) or an aptamer-regulatable PfYRS line upon addition of aTc, with data normalized to a no drug control (B). See table S5 for data values. (C) RBCs infected with schizont stage (43-46 h p.i.) P. falciparum (Cam3.II-rev) were exposed to ML901 for 3 h. Protein translation was assessed in the second 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. IC₅₀ (Translation) = 65 nM, IC₅₀ (Viability) = 56 nM. Data are representative of three independent experiments. Error bars correspond to the range of technical duplicates. (D) Schizont stage Cam3.II_rev parasites were incubated with DMSO (Mock), 1 μM DHA, 200 nM borrelidin (BOR) or 200 nM ML901 for 3 h and Western blots of lysates were probed for phosphorylated-eIF2α with PfBiP as a loading control. The blot is typical of data from three independent experiments.

Figure 4. ML901 inhibits PfYRS by a reaction-hijacking mechanism.

(A) The apparent melting temperature (T_m) of PfYRS after incubation at 37°C for 3 h with the indicated reactants: ML901 (50 μM), ATP (50 μM), tyrosine (100 μM), PftRNA^Tyr (4 μM). Data represent the average of three independent assays and error bars correspond to SD. (B) First derivatives of melting curves for PfYRS and HsYRS with or without pre-incubation with ML901 (50 μM), ATP (50 μM), tyrosine (100 μM) and PftRNA^Tyr (4 μM). Data is representative of three independent assays. (C) Effects of increasing concentrations of ML901 on tyrosine acylation of the cognate tRNA^Tyr by PfYRS and HsYRS with YRS (0.25 μM), ATP (10 μM), tyrosine (100 – 200 μM), cognate tRNA^Tyr (24 μM) and pyrophosphatase (1 unit/mL), at 37°C for 1 h. IC₅₀ (PfYRS) = 53 μM; IC₅₀ (HsYRS) > 500 μM. Data represent the average of 8 independent assays and error bars correspond to SEM. (D) Structure of ML901-Tyr. (E) PfYRS was incubated with ML901 (50 μM), ATP (10 μM), tyrosine (20 μM) and PftRNA^Tyr (8 μM). Following urea denaturation and TFA precipitation, the supernatant was subjected to LCMS analysis, revealing the expected protonated ML901-Tyr ion. The profile is typical of data from three independent experiments. (F) Schematic of reaction-hijacking mechanism.

Figure 5. Structural analysis of YRSs reveals the determinants of potency and specificity.

(A) The structure of the dimeric PfYRS/ML901-Tyr complex showing chain A (green), chain B (blue), and bound ML901-Tyr (red stick representation). (B) Inhibitor/active site interactions for the B chain. (C) (i) The PfYRS chain B active site highlighting the “HIGH” (₇₀HIAQ₇₃; light purple) and “KMSKS” (₂₄₇KMSKS₂₅₁; light brown) motifs with bound ML901-Tyr (colored by atom type). M248 and H70 are positioned to interact. (ii) Active site of HsYRS with bound ML901-Tyr highlighting the “HIGH” motif (₄₉HVAY₅₂; light pink). (iii) Active site of PfYRS_S234C with bound ML901-Tyr highlighting the “HIGH” (₇₀HIAQ₇₃; light green). Unmodelled loops are shown in Cii and Ciii as dashed lines. (D) Overlay of PfYRS (B chain), HsYRS, and PfYRS_S234C showing the different configurations adopted by His70/His49. PfYRS His70, purple; HsYRS His49, pink; PfYRS_S234C His70, green.