Structural basis of the substrate recognition and inhibition mechanism of Plasmodium falciparum nucleoside transporter PfENT1

Abstract

By lacking de novo purine biosynthesis enzymes, Plasmodium falciparum requires purine nucleoside uptake from host cells. The indispensable nucleoside transporter ENT1 of P. falciparum facilitates nucleoside uptake in the asexual blood stage. Specific inhibitors of PfENT1 prevent the proliferation of P. falciparum at submicromolar concentrations. However, the substrate recognition and inhibitory mechanism of PfENT1 are still elusive. Here, we report cryo-EM structures of PfENT1 in apo, inosine-bound, and inhibitor-bound states. Together with in vitro binding and uptake assays, we identify that inosine is the primary substrate of PfENT1 and that the inosine-binding site is located in the central cavity of PfENT1. The endofacial inhibitor GSK4 occupies the orthosteric site of PfENT1 and explores the allosteric site to block the conformational change of PfENT1. Furthermore, we propose a general “rocker switch” alternating access cycle for ENT transporters. Understanding the substrate recognition and inhibitory mechanisms of PfENT1 will greatly facilitate future efforts in the rational design of antimalarial drugs.

Putative purine nucleoside interacting residues in the malaria parasite purine uptake transporter PfENT1 are critical for transporter function

“…This may be due to the fact that hENT1 L26, aligned with L50, has hydrophobic interactions with the purine ring [ 15 ]. In the PfENT1 cryo-EM structure, L50 is close to but does not directly interact with inosine [ 17 ]. This difference from the hENT1 structure might be due to the fact that the hENT1 structure is open outward and the PfENT1 structure is open inward.…”

“…This difference from the hENT1 structure might be due to the fact that the hENT1 structure is open outward and the PfENT1 structure is open inward. In contrast, the other three purine binding residues form H-bond interactions with either the purine ring, Q158, or the ribosyl hydroxyls, D341 and R345 in the hENT1 structure [ 15 ] and with inosine in the PfENT1 structure [ 17 ].…”

“…We did not expect that the mutations W53C and S290C would have such a dramatic impact on PfENT1 function because in the hENT1 structure, which is outward facing, the aligned residues did not directly interact with the NBMPR purine nucleobase or ribose moieties [ 15 ]. In contrast, in the PfENT1 structure, which is inward open with a nanobody bound in the cytoplasmic entrance to the permeation pathway, W53 interacts with the hydroxyl on the ribose 5’ carbon residue [ 17 ]. They also report that the tryptophan limits the rotation of the inosine in the binding site.…”

“…However, the uridine affinity was altered in the N51C and V52C mutants ( Fig 5D ). Neither of these residues interacts with the ribose sugar or the nucleobase in the PfENT1 cryo-EM structure [ 17 ]. It is possible that these mutations alter the local structure and the molecular interactions with the pyrimidine nucleoside.…”

“…While this manuscript was in preparation, a paper was published that reported high resolution cryo-electron microscopy structures of PfENT1 with inosine and with GSK-4 bound in the transport pathway [ 17 ]. They mutated to alanine five of the 16 residues mutated to cysteine in this work and functionally characterized the mutants by inosine binding isothermal calorimetry and transport.…”