P2X7 receptors (P2X7Rs) belong to the P2X receptor family of ligand-gated ion channels activated by extracellular ATP. The human P2X7R (hP2X7R) is implicated in numerous debilitating disease conditions and thus represents a promising therapeutic target. However, P2X7R structure-function relationships remain less well understood. The study presented in this thesis used electrophysiology in conjunction with structural modelling, molecular docking and site-directed mutagenesis to better understand the structural basis for ligand-receptor interactions at P2X7Rs and used such structural information to identify novel hP2X7R antagonists.
Initially, P2X7R homology models were produced based on the crystal structures of the zebrafish P2X4R (zfP2X4R) in closed and ATP-bound states and validated through docking and biochemical approaches. First of all, molecular docking showed that ATP binds to the zfP2X4R and hP2X7R in a strikingly similar configuration to the crystal structure. Secondly, docking of the antagonists AZ11645373, KN62 and SB203580 revealed a specific interaction with Phe95 in the hP2X7R that is absent in the rat P2X7R (rP2X7R), providing structural insight into their preferential hP2X7R antagonism. Thirdly, replacing Asp48 and Ile331 with cysteine resulted in disulfide bonding that impaired hP2X7R-mediated currents, which was reversibly restored by dithiothreitol. These results are consistent with the transmembrane domains moving substantially apart, as predicted by the closed and open state models. Overall, these experiments show that homology models can yield meaningful structural information in terms of P2X7R interactions with ATP, antagonists and receptor activation.
The second part of the study searched for P2X7R antagonists using a structure-based approach. Virtual screening of ~100,000 compounds in the ZINC library against the ATP-binding pocket in the hP2X7R model identified C23, C40 and C60 as structurally novel antagonists of the hP2X7R but not the rP2X7R. These compounds inhibited the agonist-evoked increase in intracellular Ca2+ concentration ([Ca2+]i) with IC50 values in the micromolar range. C23 and C40 also inhibited agonist-induced currents with similar potency, but C60 did not. All three compounds suppressed large pore formation with micromolar potency. While C23 inhibited agonist-induced [Ca2+]i increase mediated by the hP2X4R and rP2X3R, C40 and C60 were more selective towards the hP2X7R. In conclusion, these results show C23, C40 and C60 as novel hP2X7R antagonists. Such structure-based approaches should aid novel P2XR antagonist identification.
Finally, the models were used in combination with site-directed mutagenesis to investigate residues influencing hP2X7R-agonist interactions. The first set of experiments examined four residues implicated in interactions with ATP, and only the mutation of Tyr288 to various residues significantly affected agonist sensitivity. This Tyr288 mutation abolished receptor function, which was mainly due to impairment in protein surface expression as shown by immunofluorescent imaging. The second set of experiments examined several residues for their contribution in the difference in functional expression and agonist sensitivity. Substitution of Val87 in the hP2X7R for Ile in the rP2X7R increased the maximal agonist-induced currents.
Overall, the present study provides structural insights into ligand-receptor interactions at the P2X7Rs. It demonstrates that structure-based approaches are feasible in identifying novel antagonists, and this has wider implications for the P2XR family and membrane proteins as a whole.
