Identification of selective inhibitors of the potassium channel Kv1.1-1.2((3)) by high-throughput virtual screening and automated patch clamp

Abstract

Two voltage-dependent potassium channels, Kv1.1 (KCNA1) and Kv1.2 (KCNA2), are found to co-localize at the juxtaparanodal region of axons throughout the nervous system and are known to co-assemble in heteromultimeric channels, most likely in the form of the concatemer Kv1.1-1.2((3)) . Loss of the myelin sheath, as is observed in multiple sclerosis, uncovers the juxtaparanodal region of nodes of Ranvier in myelinated axons leading to potassium conductance, resulting in loss of nerve conduction. The selective blocking of these Kv channels is therefore a promising approach to restore nerve conduction and function. In the present study, we searched for novel inhibitors of Kv1.1-1.2((3)) by combining a virtual screening protocol and electrophysiological measurements on a concatemer Kv1.1-1.2((3)) stably expressed in Chinese hamster ovary K1 (CHO-K1) cells. The combined use of four popular virtual screening approaches (eHiTS, FlexX, Glide, and Autodock-Vina) led to the identification of several compounds as potential inhibitors of the Kv1.1-1.2((3)) channel. From 89 electrophysiologically evaluated compounds, 14 novel compounds were found to inhibit the current carried by Kv1.1-1.2((3)) channels by more than 80 % at 10 μM. Accordingly, the IC(50) values calculated from concentration-response curve titrations ranged from 0.6 to 6 μM. Two of these compounds exhibited at least 30-fold higher potency in inhibition of Kv1.1-1.2((3)) than they showed in inhibition of a set of cardiac ion channels (hERG, Nav1.5, and Cav1.2), resulting in a profile of selectivity and cardiac safety. The results presented herein provide a promising basis for the development of novel selective ion channel inhibitors, with a dramatically lower demand in terms of experimental time, effort, and cost than a sole high-throughput screening approach of large compound libraries.

Figure 1

ROC curves a) with respect to different putative binding sites based on Vina Scores; b) ROC curves with respect to the inner cavity of the main scores; and c) scoring function sub-terms, as well as the ROC curves from the respective consensus scores. ROC curves of random distributions are expected in the grey area around the diagonal dashed lines. A perfect ranking corresponds to the red dashed lines.

Figure 2

Average Vina Scores over the AROC of different putative binding sites. Error bars reflect the standard deviation of the scores.

Figure 3

Structures of the 14 confirmed novel Kv1.1–1.2₍₃₎ active compounds. Larger repeated motifs are highlighted.