Characterization of the substituted N-triazole oxindole TROX-1, a small-molecule, state-dependent inhibitor of Ca(V)2 calcium channels.

Biological, genetic, and clinical evidence provide validation for N-type calcium channels (Ca(V)2.2) as therapeutic targets for chronic pain. A state-dependent Ca(V)2.2 inhibitor may provide an improved therapeutic window over ziconotide, the peptidyl Ca(V)2.2 inhibitor used clinically. Supporting this notion, we recently reported that in preclinical models, the state-dependent Ca(V)2 inhibitor (3R)-5-(3-chloro-4-fluorophenyl)-3-methyl-3-(pyrimidin-5-ylmethyl)-1-(1H-1,2,4-triazol-3-yl)-1,3-dihydro-2H-indol-2-one (TROX-1) has an improved therapeutic window compared with ziconotide. Here we characterize TROX-1 inhibition of Cav2.2 channels in more detail. When channels are biased toward open/inactivated states by depolarizing the membrane potential under voltage-clamp electrophysiology, TROX-1 inhibits Ca(V)2.2 channels with an IC(50) of 0.11 µM. The voltage dependence of Ca(V)2.2 inhibition was examined using automated electrophysiology. TROX-1 IC(50) values were 4.2, 0.90, and 0.36 µM at -110, -90, and -70 mV, respectively. TROX-1 displayed use-dependent inhibition of Ca(V)2.2 with a 10-fold IC(50) separation between first (27 µM) and last (2.7 µM) pulses in a train. In a fluorescence-based calcium influx assay, TROX-1 inhibited Ca(V)2.2 channels with an IC(50) of 9.5 µM under hyperpolarized conditions and 0.69 µM under depolarized conditions. Finally, TROX-1 potency was examined across the Ca(V)2 subfamily. Depolarized IC(50) values were 0.29, 0.19, and 0.28 µM by manual electrophysiology using matched conditions and 1.8, 0.69, and 1.1 µM by calcium influx for Ca(V)2.1, Ca(V)2.2, and Ca(V)2.3, respectively. Together, these in vitro data support the idea that a state-dependent, non-subtype-selective Ca(V)2 channel inhibitor can achieve an improved therapeutic window over the relatively state-independent Ca(V)2.2-selective inhibitor ziconotide in preclinical models of chronic pain.

Identification of novel and selective Kv2 channel inhibitors.

Identification of selective ion channel inhibitors represents a critical step for understanding the physiological role that these proteins play in native systems. In particular, voltage-gated potassium (K(V)2) channels are widely expressed in tissues such as central nervous system, pancreas, and smooth muscle, but their particular contributions to cell function are not well understood. Although potent and selective peptide inhibitors of K(V)2 channels have been characterized, selective small molecule K(V)2 inhibitors have not been reported. For this purpose, high-throughput automated electrophysiology (IonWorks Quattro; Molecular Devices, Sunnyvale, CA) was used to screen a 200,000-compound mixture (10 compounds per sample) library for inhibitors of K(V)2.1 channels. After deconvolution of 190 active samples, two compounds (A1 and B1) were identified that potently inhibit K(V)2.1 and the other member of the K(V)2 family, K(V)2.2 (IC(50), 0.1-0.2 µM), and that possess good selectivity over K(V)1.2 (IC(50) >10 µM). Modeling studies suggest that these compounds possess a similar three-dimensional conformation. Compounds A1 and B1 are >10-fold selective over Na(V) channels and other K(V) channels and display weak activity (5-9 µM) on Ca(V) channels. The biological activity of compound A1 on native K(V)2 channels was confirmed in electrophysiological recordings of rat insulinoma cells, which are known to express K(V)2 channels. Medicinal chemistry efforts revealed a defined structure-activity relationship and led to the identification of two compounds (RY785 and RY796) without significant Ca(V) channel activity. Taken together, these newly identified channel inhibitors represent important tools for the study of K(V)2 channels in biological systems.

A KV2.1 gating modifier binding assay suitable for high throughput screening.

Gating modifier peptides alter gating of voltage-gated potassium (KV) channels by binding to the voltage sensor paddle and changing the energetics of channel opening. Since the voltage sensor paddle is a modular motif with low sequence similarity across families, targeting of this region should yield highly specific channel modifiers. To test this idea, we developed a binding assay with the KV2.1 gating modifier, GxTX-1E. Monoiodotyrosine-GxTX-1E (125I-GxTX-1E) binds with high affinity (IC50 = 4 nM) to CHO cells stably expressing hKV2.1 channels, but not to CHO cells expressing Maxi-K channels. Binding of 125I-GxTX-1E to KV2.1 channels is inhibited by another KV2.1 gating modifier, stromatoxin (IC50 = 30 nM), but is not affected by iberiotoxin or charybdotoxin, pore blocking peptides of other types of potassium channels, or by ProTx-II, a selective gating modifier peptide of the voltage-gated sodium channel NaV1.7. Specific 125I-GxTX-1E binding is not detectable when CHO-KV2.1 cells are placed in high external potassium, suggesting that depolarization favors dissociation of the peptide. The binding assay was adapted to a 384-well format, allowing high throughput screening of large compound libraries. Interestingly, we discovered that compounds related to PAC, a di-substituted cyclohexyl KV channel blocker, displayed inhibitory binding activity. These data establish the feasibility of screening large libraries of compounds in an assay that monitors the displacement of a gating modifier from the channel's voltage sensor. Future screens using this approach will ultimately test whether the voltage sensor of KV channels can be selectively targeted by small molecules to modify channel function.

A high-throughput assay for evaluating state dependence and subtype selectivity of Cav2 calcium channel inhibitors.

Cav2.2 channels play a critical role in pain signaling by controlling synaptic transmission between dorsal root ganglion neurons and dorsal horn neurons. The Cav2.2-selective peptide blocker ziconotide (Prialt, Elan Pharmaceuticals, Dublin, Ireland) has proven efficacious in pain relief, but has a poor therapeutic index and requires intrathecal administration. This has provided impetus for finding an orally active, state-dependent Cav2.2 inhibitor with an improved safety profile. Members of the Cav2 subfamily of calcium channels are the main contributors to central and peripheral synaptic transmission, but the pharmacological effects of blocking each subtype is not yet defined. Here we describe a high-throughput fluorescent assay using a fluorometric imaging plate reader (FLIPR [Molecular Devices, Sunnyvale, CA]) designed to quickly evaluate the state dependence and selectivity of inhibitors across the Cav2 subfamily. Stable cell lines expressing functional Cav2 channels (Ca(V)alpha, beta(3), and alpha(2)delta subunits) were co-transfected with an inward rectifier (Kir2.3) so that membrane potential, and therefore channel state, could be controlled by external potassium concentration. Following cell incubation in drug with varying concentrations of potassium, a high potassium trigger was added to elicit calcium influx through available, unblocked channels. State-dependent inhibitors that preferentially bind to channels in the open or inactivated state can be identified by their increased potency at higher potassium concentrations, where cells are depolarized and channels are biased towards these states. Although the Cav2 channel subtypes differ in their voltage dependence of inactivation, by adjusting pre-trigger potassium concentrations, the degree of steady-state inactivation can be more closely matched across Cav2 subtypes to assess molecular selectivity.

An automated electrophysiology serum shift assay for K(V) channels.

The presence of serum in biological samples often negatively impacts the quality of in vitro assays. However, assays tolerant of serum are useful for assessing the in vivo availability of a small molecule for its target. Electrophysiology assays of ion channels are notoriously sensitive to serum because of their reliance on the interaction of the plasma membrane with a recording electrode. Here we investigate the tolerance of an automated electrophysiology assay for a voltage-gated potassium (K(V)) channel to serum and purified plasma proteins. The delayed rectifier channel, K(V)2.1, stably expressed in Chinese hamster ovary cells produces large, stable currents on the IonWorks Quattro platform (MDS Analytical Technologies, Sunnyvale, CA), making it an ideal test case. K(V)2.1 currents recorded on this platform are highly resistant to serum, allowing recordings in as high as 33% serum. Using a set of compounds related to the K(V) channel blocker, 4-phenyl-4-[3-(2-methoxyphenyl)-3-oxo-2-azaprop-1-yl]cyclohexanone, we show that shifts in compound potency with whole serum or isolated serum proteins can be reliably measured with this assay. Importantly, this assay is also relatively insensitive to plasma, allowing the creation of a bioassay for inhibitors of K(V)2.1 channel activity. Here we show that such a bioassay can quantify the levels of the gating modifier, guangxitoxin-1E, in plasma samples from mice dosed with the peptide. This study demonstrates the utility of using an automated electrophysiology platform for measuring serum shifts and for bioassays of ion channel modulators.

An unstructured initiation site is required for efficient proteasome-mediated degradation.

The proteasome is the main ATP-dependent protease in eukaryotic cells and controls the concentration of many regulatory proteins in the cytosol and nucleus. Proteins are targeted to the proteasome by the covalent attachment of polyubiquitin chains. The ubiquitin modification serves as the proteasome recognition element but by itself is not sufficient for efficient degradation of folded proteins. We report that proteolysis of tightly folded proteins is accelerated greatly when an unstructured region is attached to the substrate. The unstructured region serves as the initiation site for degradation and is hydrolyzed first, after which the rest of the protein is digested sequentially. These results identify the initiation site as a novel component of the targeting signal, which is required to engage the proteasome unfolding machinery efficiently. The proteasome degrades a substrate by first binding to its ubiquitin modification and then initiating unfolding at an unstructured region.

Protein unfolding by the mitochondrial membrane potential.

Mitochondria can unfold importing precursor proteins by unraveling them from their N-termini. However, how this unraveling is induced is not known. Two candidates for the unfolding activity are the electrical potential across the inner mitochondrial membrane and mitochondrial Hsp70 in the matrix. Here, we propose that many precursors are unfolded by the electrical potential acting directly on positively charged amino acid side chains in the targeting sequences. Only precursor proteins with targeting sequences that are long enough to reach the matrix at the initial interaction with the import machinery are unfolded by mitochondrial Hsp70, and this unfolding occurs even in the absence of a membrane potential.