The neuronal calcium ion channel activity of constrained analogues of MONIRO-1

Abstract

Structural modifications of the neuronal calcium channel blocker MONIRO-1, including constraining the phenoxyaniline portion of the molecule and replacing the guanidinium functionality with tertiary amines, led to compounds with significantly improved affinities for the endogenously expressed Ca_V2.2 channel in the SH-SY5Y neuroblastoma cell line. These analogues also showed promising activity towards the Ca_V3.2 channel, recombinantly expressed in HEK293T cells. Both of these ion channels have received attention as likely targets for the treatment of neuropathic pain. The dibenzoazepine and dihydrobenzodiazepine derivatives prepared in this study show an encouraging combination of neuronal calcium ion channel inhibitory potency, plasma stability and potential to cross the blood–brain-barrier.

Introduction

Neuropathic pain is a type of chronic pain that results from nerve damage caused by trauma, surgery or diseases, including diabetes, shingles (herpes zoster) and cancer, and is often characterized by a complex array of unpleasant sensory experiences. The estimated prevalence within the general population ranges from 1 to 10%, and this number is likely to increase with the aging of the population.1, 2 Unlike nociceptive pain, neuropathic pain can occur without stimuli, and although non-prescription pain medications such as NSAIDs are often widely used by patients suffering from this condition, there is a lack of strong evidence supporting this use.³ Instead, patients must often rely on the off-label prescription of antidepressants, such as amitriptyline and nortriptyline,⁴ anticonvulsants like carbamazepine, and in more severe cases, opioids.1, 2 The Ca_V2.2 (N-Type) voltage-gated calcium channel has been shown to be a promising target for the treatment of neuropathic pain.5, 6 The gabapentinoids, gabapentin and pregabalin, are also widely prescribed pain medications, and have been shown to interact with the auxiliary α₂δ subunit of the Ca_V2.2 channel.⁷ Many neuropathic pain patients, however, obtain only limited relief from these available therapies.

The selective peptidic Ca_V2.2 inhibitor, ziconotide, a synthetic version of ω-conotoxin MVIIA, can be used as a last-resort treatment for different types of neuropathic pain in the US and Europe, with some guidelines now beginning to recommended it as a first-line agent.⁸ However, its side effects are often severe at therapeutic doses (therapeutic window 1.5–2.1) and needs to be delivered by injection into the spinal canal via an intrathecal catheter. Importantly, ziconotide is about 10-times more potent than intrathecally administered morphine, and unlike with morphine, its administration does not appear to induce tolerance, addiction or respiratory depression.6, 8, 9, 10, 11, 12 The clinical results obtained with ziconotide provide some of the strongest validation of the N-type neuronal calcium channel as target for the treatment of intractable neuropathic pain. The limitations associated with ziconotide treatments, however, mean that there is still a strong need to develop small molecule drugs that selectively inhibit Ca_V2.2 channels and can cross the blood–brain barrier following oral or intravenous delivery. Despite concerted efforts, both academic and commercial, no drugs of this type have so far successfully passed through Phase III clinical trials.¹³

One approach is to engineer low molecular weight compounds that mimic key features of pain-blocking bioactive peptides.¹⁴ Following this strategy, efforts initially aimed to develop type III peptide mimetics of ω-conotoxin GVIA,15, 16^–17 a 27-residue peptide with greater selectivity and inhibitory potency for Ca_V2.2 than ω-conotoxin MVIIA. In an attempt to engender more drug-like characteristics into these early, relatively high molecular weight and polar compounds, one program of work led to a series of phenoxyanilides, which retained good levels of Ca_V2.2 inhibition.18, 19 It was later discovered that these compounds are also antagonists of the T-type Ca_V3.2 calcium channel, inhibitors of which have also been shown to have analgesic effects in animal models.20, 21 While these compounds are distant from ideal drugs, particularly because they are readily metabolised to toxic phenoxyanilines,²² they are of interest because one of the first to be investigated, MONIRO-1 (1, Fig. 1), was found in whole-cell patch clamp electrophysiology experiments to inhibit Ca_V2.2 channels with an IC₅₀ = 34 µM and Ca_V3 channels at 2–7 µM, while showing selectivity for these channels over other neuronal calcium channels.²³ More interestingly, MONIRO-1 exhibited state- and use-dependent inhibition of Ca_V2.2 channels and stronger inhibition at high stimulation frequencies with Ca_V3.1 channels, properties which, if reproduced in more drug-like compounds, are likely to provide clinically-relevant functional selectivity.

In the absence of a crystal structure or a well-developed in silico model for these neuronal calcium channels, improved knowledge of an optimum pharmacophore continues to be attained from bioassay-guided development of compound libraries that embody iterative structural modifications around promising leads. With the phenoxyaniline series, it was of interest to understand if constraining the relatively free rotation of the two phenyl rings about the bonds attached to the bridging oxygen in MONIRO-1 would lead to improved inhibitory potency for the target N-type and T-type neuronal calcium ion channels. In addition, it was hoped that activity would be retained if the guanidinium group was replaced with less hydrophilic tertiary amine functionalities, as this would provide more drug-like compounds with greater potential to cross the blood–brain barrier.