While the recently introduced higher throughput electrophysiology technologies offer many advantages compared to conventional screening techniques, throughput still remains significantly lower than established ion flux and fluorescence-based high throughput screening systems.We discuss how the introduction of new technology means that the pharmaceutical industry can now choose from an expanded number of options for their ion channel research strategy.


Ion channels are ubiquitous pore-forming proteins that allow the passive diffusion of ions across cell membranes1. They act as highly selective filters facilitating the movement of a particular ionic species (Na+, K+, Ca2+, Cl-) between cellular compartments, although a number of channels are less discriminative. Functionally, the channels exist as dynamic structures sensing external factors such as voltage gradients, endogenous ligands and mechanical forces, which are able to induce conformational transitions within the channel between non-conducting closed states to an open state where ions can flow according to the electrochemical gradient (gating).

Ion channels are key proteins in all neurones and a variety of other cell types. They underlie electrical signalling in the central and peripheral nervous systems as well as controlling a diverse range of other physiological processes including proprioception, nociception, heart rate, cell volume, secretion and cell membrane potential. The pharmaceutical industry has benefited from a number of therapeutics that modify ion channel function. Examples include the long established first line anti-epileptics such as the Na+ channel blocker carbamazepine (Tegretol, Novartis), the antihypertensive dihydropyridine Ca2+ channel blockers (Norvasc, Pfizer; Adalat LA Bayer) and sulphonylurea potassium channel openers for diabetes (Amaryl, Aventis). Most of these established ion channel modulating drugs were discovered serendipitously at a time when knowledge of the ion channel superfamily was either non-existent or in its infancy. Despite this, recent total annual sales of ion channel targeted drugs are around $20 billion, with Pfizers’ Norvasc alone earning $3.2 billion in the first nine months of 2004.

As the pharmaceutical industry made systematic attempts to exploit the range of ion channels for drug development, it became clear that these complex molecular targets presented their own set of problems. The lack of suitable high throughput screening (HTS) assay formats, the complexity of ion channel biophysics, the range of potential binding sites and binding modes for drugs, combined with serious gaps in our knowledge of the ion channel superfamily, have all held back progress in developing novel targeted therapeutics with the appropriate potency and selectivity profile. Significant progress has been made recently on a number of fronts and many pharmaceutical companies have a renewed interest in ion channels as a target class. One of the most significant advances has been in our understanding of the molecular biology of ion channels. Over the last couple of decades academic institutions have added considerable depth to our understanding of the architecture, distribution and biophysical profile of individual ion channel subtypes. Coupled to the wealth of information available from the human genome project, it is increasingly obvious that ion channels are an underexploited area with immense therapeutic and commercial value (Figure 1).

Other advances have been made in the area of assay technology. Ion channel proteins are extremely complex and dynamic structures. They exist in a number of sub-states (eg open, closed and inactivated) and compounds may bind preferentially to any one of these states. The method of choice for studying ion channels is electrophysiology, which allows resolution of the interactions of candidate drug molecules in real time with very high fidelity. This type of information is essential to allow meaningful progression of hits through to optimised lead candidates. Historically, the inability of electrophysiology to accommodate the large number of compounds typical of a HTS campaign has been a major obstacle to the progress of ion channel research programmes. The established HTS technologies are extremely useful hit finding tools applicable to screening large libraries (>50,000 compounds), but they normally lack the sensitivity, temporal resolution and physiological relevance to steer medicinal chemistry programmes2,3. The recent introduction of high throughput electrophysiology platforms is changing screening strategies for ion channel targets, but before discussing these advances, we will review current HTS systems.

HTS of ion channel targets
The trend in ion channel HTS is not dissimilar to that seen for G-protein coupled receptors but there are several problems that are unique to the target class, particularly for voltage-gated channels. Initially radioligand binding assays were used to identify compounds that bound to specific sites on channels. The value of this type of assay is limited in that there are usually several distinct sites on any channel where a compound could act. Hence important hits could be missed in binding assays by focusing only on the site that interacts with the radioligand2. In the case of voltage-gated channels where there is no endogenous ligand, the physiological relevance of binding assays is even less clear4.

For these reasons, there has been an increasing preference for functional assays. In high throughput assays channel function is either measured through monitoring the movement of specific ions through channels (ion flux methods) or measuring gross changes in the membrane potential of the cell expressing the channel of interest. The former assay format provides a direct and specific measure of channel function, whereas the latter is a more non-specific readout of an event that occurs as a consequence of channel activation but offers a generic assay platform. Several methods are in common usage in HTS laboratories, they are summarised in Tables 1 and 2 for their applicability across channel type and perceived relevance of the data when compared to conventional electrophysiology. It is clear from these tables that it is not possible to identify a single high throughput method that fulfils the needs of screening all ion channel types. However, by focusing on a few technologies most types can be covered, including both ligand and voltage-gated channels (with the caveats discussed below). As methods continue to evolve the number of channel types that are covered by the technologies expands. For example, Aurora Biomed (www.aurorabiomed.com) is currently developing an AAS method utilising silver ions for indirectly studying chloride channels and Invitrogen (www.invitrogen.com) has new fluorescent probes reportedly specific for sodium.

Within a high throughput screen the relevance of the assay format compared to the normal gating of the channel in the physiological setting is a concern, particularly for voltage-gated channels. In most assays, voltage-gated channels are opened either through depolarisation of the cell by elevation of the extracellular potassium concentration or by employing toxins to activate or delay channel inactivation leading to an accumulation of channels in the open state. For example, the application of veratridine or type I pyrethroids to cells expressing voltage-gated sodium channels leads to an influx of extracellular sodium to cause cell depolarisation17. Normal ion channel function is dependent on the resting membrane potential of the cell in which the channel is expressed. In neurones, under normal physiological conditions the membrane potential is close to -70mV whereas most cell lines heterologously expressing cloned channels have a more positive resting potential, typically around -20mV. For voltage-gated channels, the non-physiological mechanism of channel gating coupled with high resting membrane potential is often regarded as the reason for a lack of progression of hits from platebased high throughput functional assays translating to active molecules in conventional electrophysiology. In order to address this several groups are investigating the use of electrical stimulation to gate ion channels via an applied electrical field within assay wells18,19. This advance in technology, while not currently commercialised, combines the proven FRET-based voltage sensor probe technology with the benefits of a more physiologically relevant mechanism for opening voltage-gated channels.

High throughput electrophysiology Electrophysiology allows extremely high quality resolution of the interactions of drug molecules with ion channels. Although providing essential information regarding the potency and mechanism of action of drug binding, the technical complexity of the conventional methodology has meant that the pharmaceutical industry has not been able to obtain the full benefits from this rich source of information. The technique has traditionally required highly skilled personnel providing a limited throughput of around 30 data points per day. A number of companies have been developing new technology to simplify and automate the process of obtaining high quality electrophysiological information. Predominantly based around a technique known as planar patch-clamp it is now possible to perform multiple automated recordings in parallel, vastly increasing throughput compared to conventional methods. The most successful systems on the market achieve this by replacing the glass micropipettes used in conventional electrophysiology with pre-manufactured multi-well substrates. Typically, the plates consist of a flat substrate perforated with a small hole (~1-2μM diameter). The Molecular Devices (www.moleculardevices.com) IonWorks® platform was the first to the market and presently offers the highest throughput of the commercially available systems. Based around a 384-well format, the system uses a 48-channel amplifier to read each screening plate over eight cycles during an experiment. Electrical control of the cell membrane involves use of ‘perforated’ patch-clamp methodology, using an antibiotic solution to permeabilise a small section of cell membrane. While some technical considerations mean that the system does not provide the highest fidelity patch-clamp data, this compromise is acceptable when considering the increase in throughput compared to the conventional technique. With the latest generation instrument, IonWorks® Quattro™, it is possible to obtain around 2,500 data points per eight-hour working day. Furthermore, the system allows the user to gain an insight into details of the mechanism of action of a much larger cross-section of compounds at an early stage. Important considerations such as open channel block and use-dependence are both rapidly and easily resolved, allowing information rich data to be conveyed to medicinal chemists early in the programme.

Stepping down in throughput, the PatchXpress® 7000A (www.moleculardevices.com), QPatch 16 (www.sophion.dk) and NPC©-16 (www.nanion.de) allow higher fidelity recording and have the facility to enable complex recording protocols (including ligand-gated ion channels). Due to the lower throughput, these machines are suited to profiling high ‘value compounds’, making them attractive for lead optimisation or providing high quality safety pharmacology data. First to the market, the PatchXpress®, developed by Axon Instruments (www.axon.com) and Aviva Bioscience (www.avivabio.com), is an increasingly common research tool within the industry. The QPatch has been adopted by fewer companies at the time of writing, however, the 48 channel machine currently in development is a particularly attractive prospect and it could challenge the dominance of the PatchXpress® during 2006. Alternative approaches are also being pioneered by Cytocentrics (www.cytocentrics.com) and FlyIon (www.flyion.de). The Cytocentrics CytoPatch10™ is a 1-3 channel machine using microstructured chips with specialised cell positioning and recording structures. The company claims that the system can be scaled up into 20 parallel recording channels. The Flyion® 8500 (www.flyion.de) is a 2-6 channel system developed using glass micropipettes. The high throughput electrophysiology systems are starting to fill the gap between high throughput, but low resolution screening assays and very low throughput, but high resolution traditional electrophysiology (Figure 2).

Screening strategies
Despite the range of high throughput assay formats and the efficiency and cost-effectiveness with which large numbers of compounds can be tested, electrophysiology remains the bench mark for confirming compound activity5. The traditional approach to ion channel screening has been to test large diverse chemical libraries in high throughput ion flux or fluorescence-based assays. A few of the best compounds are then characterised in detail by conventional electrophysiology (Figure 2, Strategy 1). In this approach, it is not possible to profile large numbers of compounds in detail, so a great deal of important information is missed and data fed into medicinal chemistry programmes are sub-optimal.

The high throughput electrophysiology systems discussed in this review remove the bottleneck associated with the evaluation of large numbers of compounds, but the cost per data point on such systems remains significantly higher than that of the ion flux or fluorescence methodologies. These costs would almost certainly prohibit testing more than a few thousand compounds in an early stage drug discovery programme for most organisations20. Consequently, high throughput methods that do not rely on electrophysiology still have a role in ion channel screening for many groups with large diverse screening libraries2. The high throughput electrophysiology methods are used to obtain high quality information on a significant subset of the hits. Conventional electrophysiology is available, if more detailed characterisation is needed for a small number of compounds (Figure 2, Strategy 2). In this approach high quality data can be obtained for a few thousand compounds to provide a good basis for guiding medicinal chemistry programmes. The starting point is still the large diverse libraries, where hit rates are likely to be low.

An increasingly popular strategy is to screen small ion channel focused compound libraries using high throughput electrophysiology. Here, the emphasis is switched from large diverse library sets to small libraries that are designed specifically for ion channel targets, based on the emerging knowledge of channel structures. This strategy is being pioneered by BioFocus in the UK which has a proprietary ion channel ligand design tool called Helical Domain Recognition Analysis (HDRA™) (www.biofocus.com). With small libraries, screening can be run directly by high throughput electrophysiology, which gives the most appropriate and information-rich readout, while costs are controlled by reducing the number of compounds screened compared to traditional campaigns (Figure 2, Strategy 3). The best compounds can be further profiled by conventional methods. With this approach, hit rates are likely to be relatively high and good quality information is available for all the compounds in the screening set. It is a very attractive option, which combines the latest strategies in ion channel screening and library design to provide the best possible data for guiding chemistry programmes.




Dr Andy Southan has 18 years’ experience in ion channel having been formerly at Ionix, CeNes and Wyeth. He is currently Head of Ion Channel Pharmacology at BioFocus.

Iain James has spent 20 years in drug discovery, in the neuroscience field with Sandoz, Novartis and Ionix. Iain has immense experience in the biology of both receptor and ion channel classes of drug targets. He is currently Director of Biology at BioFocus.

David Cronk has five years’ experience of ionchannel research and 17 years in drug discovery research having previously been with Ionix and GSK (Glaxo-Wellcome and SmithKline Beecham). He is currently Head of Hit Discovery at BioFocus.

References
1 Hille, B (2001). Ion channels of excitable membranes. Sinauer Associates, Massachusetts.

2 Xu, J et al (2001). Ionchannel assay technologies: quo vadis? Drug Discovery Today 6(24):1278-1287.

3 Birch, P et al (2004). Strategies to identify ion channel modulators: current and novel approaches to target neuropathic pain. Drug Discovery Today 9(9): 410-418.

4 Denyer, J et al (1998). HTS approaches to voltage gated ion channel drug discovery. Drug Discovery Today 3(7):323-332.

5 Owen, D and Silverthorne,A (2002). Channeling drug discovery: current trends in ion channel drug discovery research. Drug Discovery World Spring:48-61.

6 Terstappen, G (1999). Functional analysis of native and recombinant ion channels using a high capacity nonradioactive rubidium efflux assay.Anal. Biochem. 272:149-155.

7 Scott, CW et al. (2003).A medium-throughput functional assay of KCNQ2 potassium channels using rubidium efflux and atomic absorption spectroscopy. Anal. Biochem. 319:251-257.

8 Parihar,AS et al (2003). Functional analysis of large conductance Ca2+-activated K+ channels: ion flux studies by atomic absorption spectroscopy. Assay Drug Dev.Tech. 1:647-654.

9 Trivedi, S (2005).A nonradioactive lithium influx assay using atomic absorption spectroscopy for identifying sodium channel modulators. Paper presented at Ion Channel Retreat,Vancouver BC.

10 Weaver, CD et al (2004).A thallium-sensitive, fluorescencebased assay for detecting and characterizing potassium channel modulators in mammalian cells. Journal of Biomolecular Screening 9:671-677.

11 Galietta, I, Jayaraman, S and Verkman,AS (2001). Cell-based assay for high-throughput quantitative screening of CFTR chloride transport channels. Am. J. Physiol: Cell Physiol. 281:C1734-C1742.

12 Tang,W and Wildey, MJ (2004). Development of a colorimetric method or functional chloride channel assay. Journal of Biomolecular Screening 9:607-613.

13 Cronk, D (2001). Cell based high throughput screening of ion channels using FLIPR and VIPR membrane potential assay technology. Paper presented at the annual meeting of the Society for Biomolecular Screening, Baltimore, MD.

14 Wolff, C, Fuks, B and Chatelain, P (2003). Comparative study of membrane potential-sensitive fluorescent probes and their use in ion channel screening assays. Journal of Biomolecular Screening 8:533-543.

15 Gonzalez, J and Tsien, R (1997). Improved indicators of cell membrane potential that use fluorescence resonance energy transfer. Chem Biol 4:269-277.

16 Falconer, M et al (2002). High-throughput screening for ion channel modulators. JBS 7:460-465.

17 Chang, SM et al (2004). Comparison of the pharmacological properties of rat NaV1.8 with rat NaV1.2a and human NaV1.5 voltagegated sodium channel subtypes using a membrane potential sensitive dye and FLIPR. Receptors & Channels 10:11- 23.

18 Maher, M and Gonzalez, J (2002). Ion channel assay methods International patent application WO02/08748.

19 Huang,T et al (2003). Identification of use-dependent blockers of voltage-gated sodium channels with a novel high-throughput optical assay. Annual meeting of the Society for Neuroscience,Washington, DC, Abstract 8.10.

20 Asmild, M et al (2003). Upscaling and automation of electrophysiology:Toward high throughput screening in ion channel drug discovery. Receptors & Channels 9: 49- 58.

with serious gaps in our knowledge of the ion channel
superfamily, have all held back progress in developing
novel targeted therapeutics with the appropriate
potency and selectivity profile. Significant
progress has been made recently on a number of
fronts and many pharmaceutical companies have a
renewed interest in ion channels as a target class.
One of the most significant advances has been in
our understanding of the molecular biology of ion
channels. Over the last couple of decades academic
institutions have added considerable depth to
our understanding of the architecture, distribution
and biophysical profile of individual ion channel
subtypes. Coupled to the wealth of information
available from the human genome project, it is
increasingly obvious that ion channels are an
underexploited area with immense therapeutic and
commercial value (Figure 1).
Other advances have been made in the area of
assay technology. Ion channel proteins are
extremely complex and dynamic structures. They
exist in a number of sub-states (eg open, closed and
inactivated) and compounds may bind preferentially
to any one of these states. The method of
choice for studying ion channels is electrophysiology,
which allows resolution of the interactions of
candidate drug molecules in real time with very
high fidelity. This type of information is essential
to allow meaningful progression of hits through to
optimised lead candidates. Historically, the inability
of electrophysiology to accommodate the large
number of compounds typical of a HTS campaign
has been a major obstacle to the progress of ion
channel research programmes. The established
HTS technologies are extremely useful hit finding
tools applicable to screening large libraries
(>50,000 compounds), but they normally lack the
sensitivity, temporal resolution and physiological
relevance to steer medicinal chemistry programmes2,3.
The recent introduction of high
throughput electrophysiology platforms is changing
screening strategies for ion channel targets, but
before discussing these advances, we will review
current HTS systems.
HTS of ion channel targets
The trend in ion channel HTS is not dissimilar to that
seen for G-protein coupled receptors but there are
several problems that are unique to the target class,
particularly for voltage-gated channels. Initially radioligand
binding assays were used to identify compounds
that bound to specific sites on channels. The
value of this type of assay is limited in that there are
usually several distinct sites on any channel where a
compound could act. Hence important hits could be
missed in binding assays by focusing only on the site