Ion Channels and Relevant Drug Screening Approaches

In this issue of SLAS Discovery, we present a special collection of manuscripts, including three original research papers and one review, that reflect recent advances and continuing challenges in the development and application of assay technologies to drug discovery for ion channel targets. First, though, we provide our perspectives on the specific challenges and opportunities in this field.

Transport proteins, which are embedded in and span most cell membranes, play essential roles in regulating ionic homeostasis in all mammalian cell types. It is well established that the electrolyte compositions of intracellular and extracellular fluids in the human body are very different, with K⁺ (~140 mM) being the dominant cation intracellularly and Na⁺ (~140 mM) and Cl^– (~100 mM) being the dominant ions extracellularly. In fact, most ions have multiple-fold differences in concentration across the plasma membrane and across the membranes of intracellular organelles. Ionic concentration gradients are established and maintained by active, energy-requiring pumps, including the electrogenic Na⁺/K⁺-ATPase in the plasma membrane and SERCA, a Ca²⁺-ATPase that is present in the membrane of the sarcoendoplasmic reticulum. These concentration gradients are possible because lipid-based membranes are impermeable to ions. Ion channels constitute a class of pore-forming proteins that provide regulated permeability pathways that allow the passive flow of ions from one side of the membrane to the other. Ion channels can be grouped into families that are defined by their structural features, tissue localization, and functional and pharmacological characteristics. Typically, they are classified according to their gating mechanism, for example, voltage-gated or ligand-gated and, in some cases, by their permeant ion(s): ¹ cation channels are permeable either selectively or nonselectively to K⁺, Na⁺ and Ca²⁺, whereas anion channels are permeable to Cl^–.

In most mammalian cell types, leaky (constitutively open) K⁺-selective channels allow K⁺ to flow down their concentration gradient from the interior of the cell to the exterior. The outward flow of positively charged ions establishes an electric potential across the membrane, whereby the inside of the cell becomes negative with respect to the outside. Eventually, the buildup of negative charge within the cell opposes the continued efflux of K⁺ until an equilibrium is reached. In this way, the K⁺ concentration gradient along with the presence of open K⁺ channels dominates the resting membrane potential of the cell, setting it anywhere between –30 and –90 mV, depending on the cell type. Without exception, all permeant ions will flow in a direction that is dictated by their individual electrochemical gradients: Na⁺ and Ca²⁺ generally flow into the cell and exert a depolarizing influence under normal circumstances, whereas K⁺ flows out of the cell and Cl^– flows into the cell, with both exerting a repolarizing or hyperpolarizing influence. The regulated movement of ions across cell membranes through selective and nonselective ion channels underlies many cellular physiological processes, including establishment of the resting membrane potential, generation and propagation of action potentials in excitable cells, muscle contraction, neurotransmitter release, hormone secretion, signal transduction, cell volume regulation, and cell proliferation. Furthermore, mutations that disrupt the normal function of ion channels (channelopathies) underlie many diseases of the nervous system, cardiovascular system, skeletal muscle, kidney, and the endocrine system. Channelopathies provide important genetic validation of ion channels such that they are the subject of immense interest in the medical field where scientists seek not only to understand their complex roles in both healthy and disease states, but also to identify and develop drugs to modulate their function.

Due to their complexity and diverse physiological roles, ion channels are both interesting and challenging to scientists and drug hunters alike. Ion channels are highly dynamic proteins that are capable of transitioning between several conformational states. Most ion channels have an inherently low probability of opening spontaneously and so, under resting conditions, they are primarily in a closed, nonconducting state. However, in response to an activating stimulus (e.g., the binding of a neurotransmitter or a change in either membrane potential, temperature, light, or mechanical stretch), the probability of channel opening increases and the flow of ions is initiated. In the continued presence of the activating stimulus, open channels tend to close (inactivate or desensitize) and do not become available for reopening until the stimulus is removed and the channels have time to return to their resting state. The kinetics of transitions vary widely; rapidly activating and inactivating channels such as voltage-gated Na⁺ channels and ionotropic receptors such as α7-nicotinic and GABA_A receptors open and close on a millisecond timescale, whereas certain voltage-gated Ca²⁺ and delayed-rectifier K⁺ channels as well as some P2X receptors have slower kinetics, opening and closing over the course of many seconds or not inactivating at all.

At a structural level, most ion channels are protein complexes containing one or more large (up to 4000 amino acids), pore-forming α-subunits in combination with one or more auxiliary or accessory subunits. The pore-forming subunits tend to have multiple membrane-spanning segments joined by intracellular and extracellular loops of varying length. Prototypical voltage-gated ion channels can be envisioned as having a tetrameric architecture consisting of 4 α-subunits, where each subunit has six transmembrane segments forming the voltage sensor (S1–S4) and the pore (S5 and S6). In the case of voltage-gated K⁺ channels, individual α-subunits (of which there are 40 subtypes) may mix and match largely, but not exclusively, within families (of which there are 12, i.e., K_V1–K_V12) to form homotetrameric and heterotetrameric channel assemblies; this vastly increases the range of functional diversity that is possible. In contrast, voltage-gated Na⁺ channel α-subunits (of which there are 9 subtypes in a single family, i.e., Na_V1.1–Na_V1.9) and Ca²⁺ channel α-subunits (of which there are 10 subtypes in three families, i.e., Ca_V1–Ca_V3) are encoded by single genes where four homologous, nonidentical domains (with topography that resembles the monomeric subunit of the voltage-gated K⁺ channel) are linked by intracellular loops. In addition to containing the elements that are necessary for the coordinated arrangement of the transmembrane segments to form a central pore, ion channel α-subunits also contain elements that determine ionic selectivity or that can sense changes in the channel’s local environment, including the transmembrane potential, the metabolic state of the cell, or the extracellular concentration of endogenous chemicals, such as neurotransmitters. On the other hand, the smaller auxiliary subunits serve to modulate the expression, trafficking, assembly, subcellular localization, and/or kinetic, gating, and conducting properties of the channel complex. Auxiliary subunits can be either cytosolic in nature (e.g., K_Vβ) or anchored to the membrane (e.g., Ca_V α₂δ). Broad structural, functional, and pharmacological heterogeneity also exists within the various families of ligand-gated ion channels. These channels are typically composed of assemblies of ligand-sensing α-subunits in association with auxiliary subunits and may be trimeric (P2X purinergic receptors and the ASIC acid-sensing receptors), tetrameric (glutamate receptors), or pentameric (Cys-loop superfamily of ionotropic receptors) in nature. ²

Up to 18% of all approved small-molecule drugs work by modulating the activity of ion channels ³ and, for the most part, they do so by binding to α-subunits. Modulators tend to be either small-molecule chemicals or synthetic peptides that may act from the outside of the channel to occlude the pore or from within the plane of the membrane to interfere with or enhance the movement of the voltage sensor or other elements of the gating machinery. Interestingly, many blockers of voltage-gated Na⁺ and Ca²⁺ channels have greater affinity for open and/or inactivated states of the channels, and so they tend to become more potent when the cell membrane is depolarized or hyperexcitable. Additional mechanisms of ion channel modulation could include disruption of the protein–protein interactions between α- and auxiliary subunits. Notable examples of ion channel-modulating drugs include:

1. Blockers of voltage-gated K⁺ channels delay the repolarization phase of cardiac and neuronal action potentials and are used in the treatment of atrial and ventricular fibrillation (amiodarone and dofetilide) and to improve motor function in patients with multiple sclerosis (dalfampridine) or Lambert–Eaton myasthenic syndrome (amifampridine).

2. Openers of ATP-sensitive K⁺ channels hyperpolarize the cell membranes of vascular smooth muscle cells and pancreatic β-cells, inhibiting the activation of Ca²⁺ channels and reducing Ca²⁺ influx leading to vasodilation in the vasculature and to reduced insulin secretion from the pancreas. Such K⁺ channel openers can be used in the treatment of angina (nicorandil) or hypoglycemia (diazoxide).

3. Moderately selective blockers of voltage-gated Na⁺ channels inhibit action potential generation and propagation in excitable cells. Such drugs are used clinically as local anesthetics (lidocaine and tetracaine) and class I antiarrhythmics (quinidine, mexiletine, and flecainide).

4. Subtype-selective blockers of voltage-gated, L-type Ca²⁺ channels relax vascular smooth muscle and are used to treat hypertension and angina (amlodipine and nifedipine). Blockers of voltage-gated, N-type Ca²⁺ channels inhibit Ca²⁺ entry into presynaptic nerve terminals in the dorsal horn of the spinal cord, inhibiting excitatory neurotransmission, and exert analgesic effects (ziconotide).

5. Nonselective blockers of voltage-gated Na⁺ and Ca²⁺ channels modulate neuronal firing patterns in the central nervous system that are used to treat epilepsy and convulsions (phenytoin, carbamazepine, and ethosuximide).

6. Modulators of the Cys-loop superfamily of ligand-gated ion channels have also been found useful in the treatment of several diseases and disorders. Activators of nicotinic acetylcholine receptors (nicotine and varenicline) are stimulants of the central nervous system and are used to aid smoking cessation. Allosteric modulators of GABA_A receptors (diazepam) enhance the inhibitory effects of endogenous GABA in the nervous system and are clinically useful in the management of anxiety and insomnia. Inhibitors of the 5-HT₃ serotonin receptor (ondansetron) are used for the prevention and treatment of nausea and vomiting that can result from chemotherapy or radiation therapy.

7. Blockers of the N-methyl-d-aspartate (NMDA) subtype of ionotropic glutamate receptor are approved for use in the treatment of mental confusion associated with Alzheimer’s disease (memantine) and as general anesthetics (ketamine).

8. Modulators of the cystic fibrosis transmembrane conductance regulator (CFTR) anion channel help maintain salt–water balance in the lungs by increasing the open-state probability of mutant channels (ivacaftor) or by rescuing their surface expression (lumacaftor and tezacaftor) and are approved (either alone or in combination) for the treatment of cystic fibrosis in certain patients.

9. Antagonists of the purinergic P2X3 receptor (gefapixant) have shown promise in alleviation of chronic cough.

A smaller number of ion channel-modulating drugs seem to act by binding to auxiliary subunits. Clinically useful examples include the antiepileptic and analgesic drugs gabapentin and pregabalin, which bind to the α₂δ subunit of voltage-gated Ca²⁺ channels, and the sulfonylurea class of type II antidiabetics (exemplified by glibenclamide) that bind to the regulatory sulfonylurea subunit (SUR) of ATP-sensitive K⁺ channels. Blockers of ATP-sensitive K⁺ channels depolarize the membrane of pancreatic β-cells, leading to the activation of voltage-gated Ca²⁺ channels, increased Ca²⁺ influx, and increased insulin secretion.

Despite the large number of ion channel-modulating drugs on the market, great opportunity remains to discover and develop better ion channel modulators. There are many reasons for this, including the fact that many of the ion channel modulators mentioned above were discovered and developed before their molecular targets were known. These older drugs do not tend to be very potent, nor are they sufficiently selective for their primary target. As a result, effective human doses can be on the high side and associated with undesirable side effects. Thus, there is a need for improved molecules that are more potent and more selective for their primary target with favorable pharmacokinetic profiles. The amino acid sequence homology among ion channel subfamily members can be quite high (>70%), and this can make it challenging to identify and develop the subtype-selective drugs that are needed. Furthermore, some of the most interesting ion channels remain undrugged, so there is ample opportunity to discover and develop first-in-class molecules that target them. Much of the potential opportunity exists because significant scientific and technical hurdles remain in this field. On the basic biology side, our understanding of the exact subunit composition of ion channels in different cell types and tissues is incomplete, and this can be complicated further when the expression of ion channel subunits changes in disease. Thus, recapitulation of disease phenotypes is challenging because we cannot reliably reconstitute the native channel with the correct combination of α- and auxiliary subunits. On the drug discovery side, assay developers often do not have access to the desired biological reagents and pharmacological tools to develop and validate screening and profiling assays. Furthermore, despite the recent resolution of some ion channel structures that have advanced our understanding of ion channel function,^4,5 the relative paucity of co-crystal structures of channels in complex with chemical ligands of known mechanism precludes rational drug design and forces a reliance on traditional structure–activity relationships (SARs) to improve and optimize the activity profile of lead molecules. Absent suitable pharmacological tools, we often rely on human genetic data to help validate such channels. Thus, we can look for associations between loss- or gain-of-function mutations in ion channel subunits and disease phenotypes when selecting which targets to pursue. However, genetic association of a mutation with a disease does not prove causation of the disease, and sometimes the direction of the phenotype is opposite to what one might desire in an efficacious drug.

Consequently, ion channel drug discovery tends to proceed in a somewhat conventional manner, that is, generate qualified reagents, build and execute screening and profiling assays to support the identification, optimization, and characterization of lead series of molecules, and then test the most advanced molecules in models of disease. The preferred qualified reagent for an ion channel assay is often a cell line that stably expresses the appropriate combination of α- and auxiliary subunits and that yields relevant functional and pharmacological responses on high-throughput assay platforms. Ideally, the cellular reagent will perform well in assays when used directly from frozen aliquots as this reduces the amount of time spent maintaining the cells in culture; alternatively, the stable cell line should perform consistently over many passages. The gold standard approach for studying ion channels remains the patch-clamp method. Historically a laborious, manual, low-throughput technique where single cells are studied one at a time, the last couple of decades have seen the emergence and evolution of automated, 384-well plate-based electrophysiology platforms ranging from the perforated-patch-based IonWorks Barracuda (Molecular Devices, Sunnyvale, CA) to the latest-generation, giga-seal-based instruments such as Qube (Sophion Bioscience, Copenhagen, Denmark) and SyncroPatch (Nanion Technologies, Munich, Germany). ⁶ These whole-cell patch-clamp platforms offer much higher throughput than manual methods and are capable of recording currents from populations of individual cells, which increases the likelihood of obtaining useful data from each well of an assay plate. They have been used in screening campaigns for libraries approaching 200,000 compounds. ⁷ These systems can be used to study all classes of ion channels but are particularly advantageous for voltage-gated channels as they afford the means to control and change the membrane potential of the cells. With their advanced microfluidics, test compounds can be added to each well and washed off, so the reversibility of effects can be assessed. Although automated electrophysiology systems can be operated by many laboratory staff, they still require an advanced understanding of ion channel biology and the principles of electrophysiology to design relevant experimental protocols and to interpret the results correctly. Other hurdles to their widespread implementation include the high initial expense of the instrumentation and the specialized consumables that they need on a continual basis.

Consequently, kinetic fluorescence and luminescence-based cellular assays remain popular for plate-based screening campaigns and for supporting medicinal chemistry efforts on ion channels. These orthogonal, functional assays are usually developed using engineered, ion channel-expressing cell lines that also express an ion-sensitive luminescent protein (such as aequorin) or that have been loaded with a membrane potential- or ion-sensitive fluorescent dye. Such assays can be developed in 384- or 1536-well plates for truly high-throughput screening of compounds. They are particularly suited to the study of ligand-gated channels as the operator has control over ligand and test compound addition, but are less satisfactory for voltage-gated channels as they do not permit fine control over the transmembrane potential. For ligand-gated channels, assays can be configured in activator, inhibitor, and allosteric modulator modes, assuming availability of an agonist, preferably the endogenous ligand. For instance, a test compound that activates the channel may induce ion flux that will change the fluorescence of an ion-sensitive or a membrane potential-sensitive dye, whereas a test compound that inhibits the channel will prevent the fluorescence change in response to added agonist. Ideally, these assays will be sensitive to the actions of both competitive inhibitors and allosteric modulators. Molecules that bind to allosteric sites may offer certain advantages over orthosteric ligands, such as greater subtype selectivity and alternative mechanisms of channel modulation. For voltage-gated channels, assay protocols often involve activating the channels in a nonphysiological way such as by adding a high concentration of K⁺ to the extracellular buffer to evoke a graded yet rather irreversible depolarization of the cell membrane or by using a toxin that modifies the gating characteristics of the channel, perhaps causing it to remain open for longer than normal. These sorts of measures surely bias the assays in ways that are likely to make them sensitive to molecules with obscure mechanisms of action (MoAs), but one hopes that with rigorous and appropriate assay validation, the assays will remain sensitive to molecules that display the desired MoA. An additional drawback of the plate-based, fluorescence, and luminescence assays is that the agonist or activating agent, once added, cannot be removed easily, so it is difficult to study the effects of test compounds on intricate gating and permeation mechanisms of the channel or the reversibility of the compound’s effects on channel activity.

Screening Technologies for Inward Rectifier K⁺ (K_ir) Channels

K_ir ion channels are a family of K⁺-permeable channels that conduct inward K⁺ currents under physiological conditions. The K_ir family of channels consists of seven subfamilies (K_ir1–K_ir7) and they have been identified in a wide variety of cells, including cardiomyocytes, neurons, blood cells, and osteoclasts, as well as endothelial, glial, and epithelial cells. K_ir channels play diverse physiological roles depending on their properties, including cellular and tissue localization, and dysfunction of the K_ir channels has been implicated in the pathophysiology of many diseases. These considerations have made K_ir channels attractive therapeutic targets. In this special collection, Walsh provides a review of K_ir channel drug discovery technologies and challenges. ⁸ The structure, function, and pharmacology of K_ir channels, with a focus on K_ir1.1, K_ir2.x, K_ir3.x, K_ir4.1, and K_ir6.x channels, are introduced and current screening technologies for K_ir channels are evaluated. The author analyzes the pros and cons of a number of assay technologies that have been used for screening K_ir channels, including membrane potential-based fluorescent dye assays, using either fast-response fluorescence resonance energy transfer (FRET) probes or slow-response probes, thallium-sensitive fluorescent assays, radiometric and nonradiometric Rb⁺ flux assays, automated patch-clamp assays, a cell-free ion flux assay that uses K_ir channel-containing liposomes, and a K⁺-dependent yeast growth assay. Although these technologies have been successfully used for screening different K_ir channels, each technology has its own limitations in one or several of the following parameters: sensitivity, specificity, throughput, robustness, flexibility, cost, content information, and/or physiological relevance. Therefore, there is a need for continued development of ion channel screening technologies.

Development of an Optogenetic Ion Channel Screening Technology

Chronic pain remains a major unmet medical need due to failure of standard pain management approaches for patients. Key genetic evidence supporting Na_V1.7 as a pain target derives from the identification of patients with congenital indifference to pain (CIP) who harbor recessive loss-of-function mutations in SCN9A. So far, small-molecule inhibitors of Na_V1.7 have shown limited clinical success, possibly due to an unfavorable MoA or suboptimal pharmacokinetic properties, providing a rationale to continue the search for novel Na_V1.7 inhibitors with unique chemical properties and channel blocking mechanisms. The manuscript by Zhang et al. in this special collection introduces a novel, all-optical screening platform (the spiking HEK Optopatch) for identifying and characterizing inhibitors of voltage-gated Na⁺ channels. ⁹ The key cellular reagent in this assay platform is an HEK293 cell line co-expressing Na_V1.7 along with a K_ir channel and a modified light-gated channelrhodopsin. The assay protocol involves depolarizing the cells with light, through activation of the channelrhodopsin, and then detecting changes in membrane potential with a voltage-sensitive dye. Briefly, the cells are grown in 384-well plates and imaged with a high-speed, automated fluorescence microscope. Illumination of the cells with blue light produces action potential-like waveforms that can be recorded by far-red fluorescence emitted from the voltage-sensitive dye. The optical action potential-like waveforms are exquisitely sensitive to inhibition by Na_V1.7 blockers. The authors further developed a stimulation paradigm that can provide mechanistic information regarding channel block in a single scan. The performance of the spiking HEK Optopatch assay was compared directly to an automated electrophysiology assay (IonWorks Barracuda). Agreement between the two platforms was demonstrated in a pilot screen of a small-molecule library (consisting of 3520 Amgen compounds) with subsequent characterization of the hits in follow-up experiments. The authors conclude that the Optopatch platform compares favorably to the IonWorks Barracuda in terms of sensitivity and ability to characterize a compound’s MoA while offering the expected advantages of a multiwell optical assay, including higher throughput and lower cost, when executed on an automated plate reader, currently under development at Q-State Biosciences.

In summary, no universal ion channel screening strategy can satisfy all needs at this time. It is important to understand the advantages and disadvantages of different screening technologies and to choose the ones that are fit for purpose when designing and executing assays for primary and secondary screening, selectivity, and safety profiling, as well as SAR and MoA studies.

Orthogonal Studies of Ion Channels in Maturing Cortical Neurons

Mechanistic and pharmacological assays of neuronal ion channels are often used in early drug discovery not only to elucidate the role of a target in biology and disease but also to understand and characterize the MoA of screening hits or new lead molecules on the target. Investigators typically conduct these more physiological studies either using tissue slices, where intracellular or extracellular recordings can be made, or using primary cells that have been isolated acutely or cultured in vitro. In this special collection, Toh et al. demonstrate the use of a multielectrode array device to measure firing patterns in developing cultures of embryonic (E18) rat cortical neurons in vitro. ¹⁰ They observed increasing spontaneous electrical activity in the cultured neurons over time (comparing days 9, 14, and 21), as evidenced by higher firing rates and burst frequencies at the later time points. Furthermore, coordinated network bursting increased over the same time frame, suggesting that the neurons were organizing into functional circuits as they matured. These electrophysiological phenomena were correlated with increased densities of the synaptic markers, synapsin and postsynaptic density protein 95, over time, along the dendrites of the neurons in culture. In parallel, whole-cell patch-clamp studies were conducted with dissociated neurons using the automated electrophysiology platform, SyncroPatch 384PE, to shed light on developmental changes in the expression of voltage-gated K⁺ and Na⁺ channels. Robust currents were recorded at all time points, although there was a trend for larger K⁺ and Na⁺ currents to be observed at day 14 compared with days 9 and 21. In addition, there appeared to be a rightward shift in the voltage dependence of activation for both the whole-cell K⁺ and Na⁺ currents over time. Furthermore, time-dependent changes in the sensitivity of the currents to inhibition by a number of ion channel blockers were evident. The whole-cell Na⁺ currents were almost fully inhibited by tetracaine and tetrodotoxin (TTX) at all time points, but the IC₅₀ of tetracaine was lower at day 14 compared with days 9 and 21, whereas TTX tended to become progressively less potent (>2-fold) over time. The whole-cell K⁺ currents were partially inhibited by tetraethylammonium (TEA) and 4-aminopyridine (4-AP) at day 9, with the extent of inhibition increasing over time; interestingly, the IC₅₀ of TEA dropped significantly from >300 µM at day 9 to only 16 µM at day 21. Taken together, the changes in current amplitudes, shifts in voltage dependence of channel activation, and changes in the pharmacological sensitivities of both current types over time may reflect developmental changes in the expression patterns of different subtypes of each ion channel family. In addition, the differences may also reflect the heterogeneity of cell types in mixed co-cultures of neuronal and glial cells. The electrophysiology assays described in this paper offer the potential advantage of being more physiologically relevant than assays built with recombinant expression systems as the ionic currents under investigation are more likely to be composed of subunits in their native complexes. The overall success rate for recording currents from cortical neurons on the SyncroPatch 384PE was around 40%, and this was obtained with a cell density of 4000 cells/well, that is, around 1.5 million cells required per 384-well plate. Although there is room for improvement, this still represents a significant throughput advantage over manual patch-clamp methods. However, due to the requirement for large numbers of cells that need to be carefully maintained in culture, the automated patch-clamp assay with cortical neurons is unlikely to be used for screening large libraries. Nevertheless, it will likely find its place in the postscreening assessment of hits and characterization of new lead molecules from medicinal chemistry.

Development of Label-Free Cellular Imaging as an Ion Channel Screening Technology

Label-free whole-cell screening technologies using biosensors to detect changes in morphology, adhesion, volume, or cell–cell contacts have been developed and applied in drug discovery. These technologies use either optical- or impedance-based biosensors for noninvasive quantification of integrated cellular responses. The development and application of label-free technologies to cell proliferation, cytotoxicity, cell migration, G-protein-coupled receptor (GPCR), ion channel, and other receptor screening have demonstrated label-free technologies as attractive tools for drug discovery. In this special collection, Rappaz et al. describe the development of a label-free, image-based cellular assay for ion channel screening that relies on quantitative phase digital holographic microscopy (QP-DHM). ¹¹ This approach has been validated previously for a number of other cellular phenotypic applications including drug toxicity screening and the detection of modulators of adipocyte differentiation. The label-free image-based QP-DHM assay is executed by acquiring live-cell phase images on a DHM T-1001 system followed by a quantitative analysis of the optical path difference (OPD) signal (calculated from the cell thickness and the difference between intracellular refractive index and refractive index of the surrounding culture medium). The OPD signal, which is proportional to both the cell thickness and the intracellular refractive index (a property linked to the protein and water content of the cell), can be used for quantifying the activity of ion channels. The QP-DHM assay also allows for measurement of time-lapse OPD signals. The method was validated by screening the ionotropic GABA_A receptor using a selection of known GABA_A active compounds, followed by a blind screen of a library of more than 3000 compounds, including 2627 natural products, in 384-well format. The results presented in this article demonstrate that the label-free image-based QP-DHM assay is suitable for identifying agonists, antagonists, and allosteric modulators of GABA_A. Furthermore, confirmed hits from the DHM agonist screen show good agreement with potency estimates obtained by electrophysiology methods. The authors also point out the potential for extending this technology to screen electroneutral transporters that are difficult to study using electrophysiology. In addition, it offers a direct visual quality control of the cells by allowing identification of compounds that interfere with cell growth. As this is a label-free image-based technology, compounds inducing nonspecific effects on cell morphology without activation of the GABA_A receptor were picked up as hits in the screen, which highlights the need to run a counterscreen with parental cells that lack the target of interest to exclude such false positives.