June 1, 2008 (Vol. 28, No. 11)

David Daniels, Ph.D.

Subtype Selectivity Is Being Developed to Improve Efficacy and Reduce Side Effects

Maintenance of cardiac rhythm and pain sensation require the flow of ions across the membranes of all cells. Ion channels, which facilitate this movement, are protein assemblies that penetrate the membrane and catalyze passage of specific ions such as sodium (Na+), potassium (K+), or calcium (Ca2+) into or out of the cell down the electrochemical gradient.

Ion channels are as diverse as they are numerous (more than 400 channel genes have been described) and they are grouped into different families based on which specific ion flows through them and how they are gated. That is whether they open or close in response to a chemical (ligand gated), electrical signal (voltage gated), pH, temperature, or mechanical force.

Ion channels play a role in a wide variety of biological processes that involve rapid signaling such as neural signal transduction; cardiac, skeletal, and smooth muscle contraction; epithelial transport of nutrients and ions; T-cell activation; and pancreatic beta-cell insulin release. It is not surprising then that ion channels are favorite targets of pharmaceutical companies and also hot topics at recent and upcoming conferences including Select Biosciences’ “Ion Channel Targets” and “Target Opportunities”, and Informa’s “Ion Channel Targets.”

Basic Biology

At Stanford University School of Medicine, Merritt Maduke, Ph.D., assistant professor in the department of molecular and cellular physiology, has built a lab focused on elucidating the molecular mechanisms underlying the function of the CLC chloride channels, the largest molecular family of mammalian anion channels.

There are nine mammalian members, CLC-1 to 7 and Ka and Kb for the kidney-specific type. Four family members are expressed on the plasma membrane of various cell types; five are expressed on intracellular membranes. Dr. Maduke’s interest in the CLC family is that half of the members are chloride channels, while the other half are chloride-proton antiporters.

“Historically, ion channels and antiporters have been thought of as two distinct classes of membrane proteins. Channels allow only passive movement of ions; antiporters can be used to pump ions. The finding of both of these classes of membrane proteins within a single gene family means that the mechanisms are likely not as distinct as had been previously supposed,” indicated Dr. Maduke. “Study of these mechanisms will help us understand how similar genes have evolved to carry out these different functions and will thus provide fundamental insight into how membrane proteins work.”

The CLC-1 channels found in skeletal muscle are closed at rest and upon excitation, i.e., membrane depolarization, the chloride channels open allowing chloride to enter into the cell. The negative charge entering in repolarizes the membrane, which allows the muscle to relax. Mutations in the CLC-1 channels cause myotonia in humans, a state where fewer CLC channels open, impairing membrane repolarization and muscle relaxation. Individuals with myotonia have problems with activities that involve skeletal muscle.

The physiological role for CLC-7 antiporters, one of the members expressed on intracellular membranes, is quite interesting. In osteoclasts, CLC-7 channels are localized to a specialized membrane called the ruffled border. Osteoclasts adhere to the bone surface and secrete protons and hydrolytic enzymes to degrade and absorb the attached bone.

Throughout our lives, bone is continually chewed up by osteoclasts and built up by osteoblasts. CLC-7 is essential for this process because it allows chloride to also be secreted into the bone matrix. This also balances the charge built up from proton secretion, enabling further proton movement.

Since CLC-7 is essential for the process of bone degradation, the prevailing thought is that by knocking down CLC-7 function one would limit the ability of osteoclasts to degrade bone.

For Dr. Maduke, the question that keeps her focused is how are the mechanisms of CLC chloride channels and CLC chloride/proton antiporters related. To get at this answer, x-ray crystallographic structural analysis of the CLC-ec1 homolog in E. coli provides some basic information. But this data only gives a single conformational view of the protein.

Dr. Maduke and her team are looking to understand how the proteins move as they open and close to allow regulated flux of ions. The team is looking to NMR and deuterium exchange detected by mass spec to obtain snapshots of the proteins in action.

Mechanosensitive Channels

Frederick Sachs, Ph.D., distinguished professor at the Center for Single Molecule Biophysics in the physiology and biophysical sciences department at SUNY Buffalo, is working on perhaps the oldest ion channel.

Dr. Sachs’ serendipitous discovery of mechanosensitive ion channels (MSC) in skeletal muscle led to experiments linking cell mechanics and the mechanisms of mechanical signal transduction. The Sachs lab uses patch-clamp, high-resolution microscopy, molecular biology, and structural NMR to study these ion channels.

MSCs are ubiquitous. While most familiar in the sensory cells of touch and hearing, they are found in a similar density in all cells. Probably arising as a method to balance cell volume under the osmotic drive of metabolism, they represent an evolutionary history from bacteria to humans. As opposed to many other channels, their activity is not regulated by channel density but by regulating the mechanical shielding of the channels by the cytoskeleton.

The Sachs lab has discovered a specific inhibitor of MSC—GsMtx-4, a small peptide isolated from the venom of Grammostola spatulata, a tarantula. The tarantula peptide has a unique pharmacology since the D- and the L-forms are equally active, but the site of action is specific to cation-selective mechanosensitive ion channels.

The peptide may have significant clinical applications since it inhibits cardiac arrhythmias, calcium influx in dystrophic muscle, spontaneous channel activation (and hence contraction) of human bladder smooth muscle, and suppresses stretch-induced endothelia production by glial cells that may play a role in brain tumor development. Furthermore, the peptide is nontoxic in mice.

“I have a recurring pharmaceutical dream for the millions of Americans with atrial fibrillation,” Dr. Sachs said. “If we provide the peptide in an inhaler when a patient feels an atrial fibrillation episode coming on, they can take a hit on their inhaler to stop the acute attack.

“Since atrial fibrillation begets atrial fibrillation, acute inhibition should improve long-term prospects for the patient. Since the peptide doesn’t affect the ion channels responsible for the normal cardiac-action potential such as Kv channels, there are minimal side effects. The peptide only acts on channels activated by abnormal mechanical stress such as overinflation of the atria. And with acute dosing toxicity is minimized.”

The dream aside, despite their ubiquity, the breadth of knowledge about MSCs, and the availability of a chemically synthesized, specific inhibitor, pharma companies have shied away from tackling this new array of targets, perhaps from the uncertainty of utilizing unfamiliar stimuli.

Drug Discovery Targets

The real challenge in developing therapeutics for ion-channel targets is the size of ion-channel families and their widespread distribution. The ongoing focus for the development of therapeutics with channel specificity is to develop subtype selectivity to improve efficacy and reduce the impact of side effects.

Within the pharmaceutical division of Devgen (www.devgen.com), the research focus is on treatment for inflammatory and metabolic diseases and arrhythmia. Petra Blom, Ph.D., section leader within the medicinal chemistry group, and Titus Kaletta, Ph.D., head of preclinical development, are working on a project targeting the Kv4.3 ion channel. The goal of their work is to develop a first-in-class oral treatment for acute and chronic atrial fibrillation.

Inhibition of Kv4.3 and Kv1.5 prolongs the action potential of the atria and the atrial refractory period. This mechanism is predicted to cardiovert patients and to keep them in sinus rhythm. Devgen identified inhibitors with excellent availability that are atrial selective and show no propensity to cause life-threatening ventricular arrhythmias in animal models, according to Dr. Blom.

Kv4.3, a voltage-gated potassium channel, functions in the early stage of the cardiac action potential. Channel activity is difficult to screen given the rapid polarization and depolarization cycle. “To solve this problem,” Dr. Kaletta noted, “the team developed an HTS assay based on the expression of the cloned human Kv4.3 channel in C. elegans.

“This whole-animal screening system combines the simplicity and convenience of C. elegans as a research tool and its close resemblance to human systems.” In the test, the action potential of the pharynx muscle, which shares comparable features to the action potential of the human heart, is used. Ingestion of fluorescently labeled food is used to measure the Kv4.3 channel activity in the pharynx of C. elegans.

“When the Kv4.3 channel is on, the pharynx pumps food poorly, generating a low fluorescent signal in the gut. When the channel is inhibited by, for example, a small molecule compound or is off, the pharynx activity changes, which results in a significant change in the fluorescent signal. The measured fluorescent signal directly correlates with ion-channel activity.”

With about 100 worms per well, the assay is performed in 96-well plates and enables the collection of 5,000 data points per day. The Devgen team obtained a 2% hit rate from the screen of its compound library using the C. elegans HTS assay. They verified selected hits in mammalian cells by means of patch-clamp electrophysiology; 83% of the hits were true human Kv4.3 inhibitors.

“Kv4.3 is a promising drug target for treatment of atrial fibrillation,” indicated Dr. Blom. “We identified two lead series that are potent human Kv4.3 blockers, which have proven efficacy in mice and dogs without hERG, Na+, or Ca2+ channel activity. With no safety issues encountered to date, we are proceeding with the preclinical development.”

Adrian Mason, Ph.D., senior scientist, and a team of scientists at Organon Laboratories (www.organon.co.uk; a part of Schering-Plough) are focusing on a different set of voltage-gated ion channels called HCN channels (hyperpolarization-activated cyclic nucleotide-gated channels). HCN channels, members of the voltage-gated cation-channel superfamily, selectively enable the flow of K+ and Na+ ions across the membrane.

Based on in situ hybridization studies, immunocytochemistry, and electrophysiological recordings, the four subtypes of HCN channel have been found to be widely distributed in humans. They are particularly plentiful in the brain, heart, and retina, but there is also evidence for expression in peripheral tissues such as kidney and testis. To a pharmaceutical company, the presence of HCN channels in many organs suggests therapeutic opportunities in a variety of human disorders but also flags the danger of side effects as a result of drugs acting at these targets.

Fortunately, there is some differential distribution of the HCN-channel subtypes that might be exploited to circumvent this problem. For example, HCN1 is highly expressed in the brain but restricted mainly to the cortex, hippocampus, and cerebellum. HCN4 is expressed prominently in the sino-atrial node of the heart, where it helps to control the cardiac rhythm, although it is also present in thalamic brain regions.

Although Organon has a successful history of developing and marketing ion-channel blocker drugs that are used during surgery to produce muscle relaxation, this current project takes the company in a new direction. “To be honest, the focus on HCN channels as a target for drug development started out of a bit of serendipity,” Dr. Mason said. “We had a drug candidate that was being developed for a psychiatric indication but found that it had a high specificity for HCN channels and blocked their activity. We followed up on this observation by testing related compounds using patch-clamp techniques on cloned HCN channels that had been expressed in HEK293 and CHO cell lines.”

Several years ago, evidence began to accumulate in the scientific literature that HCN channels might play a role in pain. Not only are the channels expressed in dorsal root ganglion cells, but knowledge of the basic biology of the channels suggested that they might have a pivotal influence on impulse activity in sensory neurons.

Building on the work of others, the Organon team showed that this was indeed the case and also demonstrated that their compounds were effective in animal models of pain. The team’s current focus is on finding therapeutic solutions to neuropathic pain.

In an attempt to exploit the well-established role of HCN4 in the heart, several companies have tried to develop HCN modulators for cardiac indications. Unfortunately, most of these compounds have failed due to side effects. Just one drug of this type has made it to market—Procoralan® (ivabradine from Servier), which is authorized in the EU for symptomatic treatment of stable angina.

“The challenge now is to find drugs acting at HCN channels that are effective in treating neuropathic pain but don’t have unwanted effects on other physiological functions,” said Dr. Mason. “It won’t be an easy task, but with our increasing knowledge of the basic biology of HCN channels together with recent developments in ion-channel screening technology, we are confident that we can give it our best shot.”

How Do They Do That?

The gold standard for measuring ion-channel function is the patch clamp. This method of electrophysiological discovery enables the skilled researcher to measure the flow of ions through the channel of an intact cell. It is a high-resolution technique supplying both high-quality data and high information content.

In manual patch clamp, a glass microelectrode is pressed against a cell membrane, forming a tight seal known as a gigaseal. By breaking the patch, you gain electrical access to the entire cell. This is a hit-and-miss approach. This variability coupled with the time required for constructing pipettes and finding and approaching a suitable cell greatly limits its throughput. A skilled researcher can look at maybe 20–30 cells in a day. The other limitation is the nonparallel nature of manual patch clamp.

Automated patch clamp has emerged as a way to increase the throughput of manual patch clamp by a factor of 1,000 to 10,000. Pipettes are replaced by glass chips with small holes, and the process of sealing and “going whole cell” has been replaced with sophisticated automated suction and software control systems.

Cells are automatically and blindly positioned at each recording well. The first automated patch-clamp system on the market was the IonWorks by MDS’s Molecular Devices. Others include PatchXpress from Axon Instruments (now part of MDS), Sophion Bioscience’s Qpatch, Nanion Technologies’ Patchliner, and Flyion’s Flyscreen.

Despite these advances, the automated platforms have one major disadvantage; they are only suitable for recording signals from cell lines with stable, high expression of the ion channel of interest. This is because the quality of the cell suspension is crucial for a reasonable success rate. Channel expression must be relatively homogenous and cells must be healthy.

Primary cell cultures such as neuronal or cardiac preparation do not fit these criteria, and also their lack of uniform shape may render a blind-capture system rather useless. The only practical use of automated patch-clamp systems is for secondary screening efforts and safety profiling of hERG ion-channel blockers.

True high-throughput screens that enable the screening of thousands of compounds for blockers or inhibitors of ion-channel function depend upon indirectly measuring ion-channel activity using membrane potential-sensitive fluorescent probes in a fluorescent detector, i.e., FLIPR from Molecular Devices, or FRET dyes in VIPR from Invitrogen.

Resources and Services

Founded to provide pharmaceutical companies with a cardiotoxicity screening service, ChanTest (www.chantest.com) is focused on hERG ion channels. These channels are activated at the late stage of the action potential following electrical stimulation in the human heart. ChanTest scientists were reportedly the first to prove that hERG was the target for adverse cardiac events linked to noncardiac drugs.

“In the last five years, we’ve grown from 15 to 75 employees,” reported Arthur Brown, president and CEO of ChanTest, “largely due to our screening service business. But more recently, we’ve undertaken a huge project to build an ion-channel reference library. Of the 400 genes encoding human ion channels, we’ve selected the top 80 to clone and express in each of two cell lines, HEK293 and CHO cells.”

“The cloning and cell line construction has been an involved process,” said Brown. Each clone has to be verified by DNA-sequence analysis and protein-sequence verification. Following transfection, the ability of the cloned channel to assemble in the cell membrane is confirmed by Western blot if a good antibody is available. The next challenge is the optimization of the culture conditions for patch-clamp measurements using the manual method and each of the different automated patch-clamp platforms.

Finally, the cloned channels must demonstrate the same pharmacological behavior when expressed in the cell lines as reported in the literature. ChanTest completed construction of 62 of the 160 cell lines to date and is on track to complete 104 by the end of the year, Brown noted. With completion of the reference library, ChanTest will make their cell line clones available to researchers through Molecular Devices and Sophion.

Beyond hERG, ChanTest also provides screening from a number of fully validated ion-channel panels. Report outcomes from screening using its Cardiac Channel Panel (11 channels) complements S7B-integrated cardiac risk assessments, provides mechanistic interpretation of the activity, and predicts proarrhythmia potential, added Brown. Recently, drugs that do not block hERG yet have QT risk have been identified using the Cardiac Channel Panel. “Now that chemists are learning how to design out hERG risk, other cardiac ion channels will be more important for off-target effects and QT risk,” Brown said.

Previous articleInvestigators Uncover Role of Two Tumor Suppressor Genes in Aging
Next articleApplied Biosystems and BioTrove to Offer Synergistic Mass Spec Technologies