June 1, 2009 (Vol. 29, No. 11)
New Screening Methods Could Increase the Utility of These Abundant Proteins
Arthur Brown, M.D., Ph.D., president and CEO of ChanTest, summed up the sentiments of a number of speakers at the recent “SBS Conference” in Lille, France, when he said, “Ion channels are as abundant as GPCRs in the known druggable genome and have as much potential as drug targets for pain, hypertension, and inflammation. Yet they still don’t receive the time and attention they should from drug developers.”
Kim Dekermendjian, Ph.D., senior research scientist in molecular pharmacology at AstraZeneca, agreed, adding “only seven percent of current drug targets are ion channels while 30 percent are GPCRs, so ion channels are totally underexploited and this is because historically it has been so difficult to screen them in a rational way.”
Since ion channels are a diverse family of pore-forming proteins that control the voltage gradient and ionic flow across the plasma membrane, it is well established that the most accurate method of measuring a drug’s effects on an ion channel is via patch clamping.
This works by forming a patch through which intracellular access is gained to measure and control membrane voltage and current flowing through the cell membrane. In the past five years relatively low-throughput automated patch clamps such as the QPatch from Sophion Bioscience, the Patchliner from Nanion Technologies, and the PatchXpress® and IonWorks® Quattro™ systems from Molecular Devices, a MDS Analytical company, have emerged. The cost per data point, however, has made these automated methods too expensive to use in primary screening.
According to Dr. Dekermendjian, sometimes the screening costs of using these systems, even in secondary screens, is referred to at AstraZeneca as “tears per data point”. As a result, surrogate measures of ion-channel activity such as ligand displacement, ionic flux, and membrane potential are used as primary screens with equipment including the Molecular Devices FLIPRTetra®, the most commonly used.
The disadvantage of using these assays is they cannot maintain a population of ion channels in a particular conformational state due to the lack of voltage control and so need to be used as methods to weed out the best drug candidates for a confirmatory patch-clamp screen. “The surrogate measures are nonfunctional and cannot identify allosteric modulators,” noted Dr. Brown. “Ion-channel sites for allosteric modulators are enormous and relatively untapped.”
Integrated Screening
Despite the shortcomings of the screening technologies available, many pharma and biotech companies are investigating ion channels as drug targets for some interesting applications. An example of this is ChanTest’s combination of surrogate assays and patch-clamping technologies for repurposing old drugs.
“With more than 86 cell lines of HEK and CHO cells expressing ion channels screened by fluorescence assay and patch clamp we provide the largest collection of validated ion channels for screening,” explained Dr. Brown.
“Using these cells we screened a library of compounds that had been withdrawn, suspended, or discontinued at late-stage clinical trials. Many of these were potent hERG blockers, something that sends regulatory authorities into seizure mode but because they showed no adverse human cardiac events we decided these could be a useful group of compounds to test as potential treatments for, in this example, cardiac diseases.
“From primary screens using the FLIPRTetra, we produced 10,000 data points per day. Any promising compounds we screened using IonWorks Quattro to generate 2,000 data points per day, followed by a PatchXpress or QPatch screen to produce 100 data points per day,” Dr. Brown said.
“Finally, those compounds that produced the best results were screened again by manual patch clamp which, because it is so difficult and time consuming, only produces seven data points per day. From this funneled screening approach, we identified a compound we call CT-1, and its ion-channel profile showed it would be an appropriate drug candidate for atrial fibrillation.”
Dr. Brown presented data demonstrating that CT-1, in both IV and oral dosage form, is well tolerated in canine models and Phase 1 dose-escalating studies for atrial fibrillation. He also showed that in a Phase II study of 27 patients with atrial fibrillation 12/20 receiving the drug converted to having normal sinus heart (NSR) rhythms within two hours of having CT-1 treatment.
“We’re pleased with the Phase II trial results because the usual rate of patient conversion from atrial fibrillation to NSR is 30 percent and we’re achieving 60 percent,” Dr Brown concluded. “CT-1 is going to be out-licensed soon, and we’re using the same integrated screening methodology to repurpose two other drugs because we estimate it can reduce the pre-clinical development time and Phase I work by several years.”
High-Throughput Patch Clamping
Since patch clamping is the gold standard for screening ion channels, there is a distinct need for high-throughput systems for primary and secondary screening. As a result, three new high-throughput systems were introduced at the conference.
Fluxion Biosciences showed the IonFlux-HT, which looks like a traditional plate reader. Fluxion claims this system can deliver between 8,000 and 10,000 assays per day at a cost of $0.30 per data point, making it suitable for primary or secondary screening.
“The system uses a standard 384 microplate into which cells, compounds, and reagents are loaded,” Jeff Jensen, CEO, explained. “A polymer microfluidic channel network replaces the well plate bottom, with interconnected channels forming a fluidic network. By applying pressure, cells and compounds are introduced. By applying vacuum, the cells are trapped at channel junctions and their changes in currents can be measured by the system’s integrated electrodes,” Jensen added.
“We have tested the system with voltage-gated ion channels such as Kv2.1, hERG, and Nav1.7, and ligand-gated channels such as P2X3, TRPV1, and GABAA. We are providing the system with or without liquid-handling capability, and this means researchers can either manually pipette or integrate their own liquid handler in if they want, or purchase a fully integrated solution from Fluxion.”
Cellectricon has gone a different route to develop its high-throughput patch clamper. The company has teamed up with The Automation Partnership and AstraZeneca to produce Dynaflow®HT for a fully automated workflow. The system contains cell and compound storage, as well as robotic liquid and plate handling, all of which allow the cells to be prepared in one place. The cells are then sucked from 96 well plates into a silica microfluidic chip. Here the cell membrane breaks and it forms a seal with the chip where the current changes are monitored.
“This system has been custom built over the past three years, and the heart of it is the way the microfluidics work,” said Mattias Karlsson, Ph.D., CTO. “The system can currently achieve over 7,500 data points per day with competitive running costs and when we introduce our new chip later this year it will be able to generate 15,000 data points per day at a running cost of $0.10 per data point.”
The performance of this system was validated by Dr. Dekermendjian who presented data to show that the dose-response curves were comparable to a manual patch clamp of a GlyRa ligand-gated ion channel treated with a number of different compounds. According to Dr. Karlsson, Dynaflow HT has been used with cell lines including WSS, CHO, LTK, and HEK with voltage- and ligand-gated ion channels such as GABAA and hERG, and the system is now being further tested by BioFocus DPI to screen its SoftFocus® ion channel library of compounds.
The Sealing Advantage
The Fluxion and Cellectricon platforms allow the patch clamped cells to produce mega-Ohm seals, while Nanion Technologies’ SyncroPatch 96 generates giga-Ohm seals. It is this sealing that the company claims provides higher quality data recordings.
Like Cellectricon’s system, Nanion Technologies’ is also fully automated and records from 96 cells at a time, using a glass chip to replace the glass pipette used in manual patch clamping. Cells are captured by applying suction and when they reach the opening of the chip they form a tight contact known as a gigaseal with the sensor for detection.
“To complement our product portfolio of automated patch-clamp systems, we have developed the SyncroPatch to match the standards of industrial ion-channel screening,” explained Niels Fertig, Ph.D., CEO of Nanion Technologies. “Based on our success using glass chips, the SyncroPatch 96 obtains success rates of approximately 60 percent for stable, whole-cell recordings, with giga-Ohm seals. Giga-Ohm seals are vital for high quality recordings and for obtaining reliable compound data. At the same time, throughput is equally important for efficient ion-channel screening.
“With the SyncroPatch 96, we want to combine the best of both worlds. We have, so far, tested cell lines including HEK293, CHO, RBL, and Jurkat with voltage- and ligand-gated ion channels such as Nav1.5, hERG, GABAA, and TRPV1. The throughput is around 5,000 data points per day at a cost per data point of $0.60, so we see this as a good secondary screening system rather than one for primary screens,” Dr. Fertig added.
When Patch Clamping Won’t Work
“Although patch clamping is the standard for measuring ion channels, there are some ion-conducting proteins that it simply cannot be used for,” explained Petr Obrdlik, Ph.D., head of assay development at IonGate Biosciences. “You cannot measure the effects of drugs on the native intracellular organelles because standard patch-clamping equipment cannot access these membranes. I have never seen any publications on patch clamping of respiratory chain complexes and oxidative phosphorylation, for example, so these are valid yet hidden targets for many potential drugs.”
To overcome this problem, IonGate has developed SURFE2R (surface electrogenic event reader) technology that doesn’t use whole cells, but instead, reads the charge displacement from membrane preparations isolated from cell culture. The membranes are coupled via an alkaline thiol and lipid bilayer to gold-coated electrodes, and it is the electrode that detects changes in the electrical signal without the need to use reporter molecules. Beside mitochondrial membranes, the technique is also suited for native membranes from different tissues and different species such as transport proteins of heart plasma membranes, synaptic vesicles, and sarcoplasmic reticulum.
At the “SBS” meeting, an interesting application of the SURFE2R was presented by Victoria Balannik, Ph.D., of Northwestern University. Dr. Balannik showed that, by using the system, it was possible to measure the activity of the anti-influenza drug amantadine on the M2 ion channel protein from influenza A. “It is extremely difficult and time consuming to perform high-throughput screening of the activity of novel anti-viral agents on M2 by conventional electrophysiology,” Dr. Balannik said.
“This technology is complementary to patch clamping,” Stefan Schork, vp of sales at IonGate, concluded. “Since it is now available in a higher throughput format as the SURFE2R Workstation 5000, it will open up the possibility of screening drugs that target transporters and channels in their native sub-cellular surroundings.”
Bright Future?
New high-throughput screening technologies being introduced could make working on ion-channel targets more attractive. According to Dr. Dekermendjian, the ideal scenario would be able to screen 100,000 data points per day at a cost of $0.10 per data point so it is clear that many of the high-throughput electrophysiology systems shown at the “SBS Conference” are still around an order of magnitude away in terms of throughput and cost. As there have been such vast improvements in both throughput and cost in the past three years it cannot be long before high-throughput patch clamping for primary screening becomes a reality.
With the improvement in automated patch-clamping methods, however, comes the difficulty of finding appropriate cell lines to use in the screens. Companies such as ChanTest and Genionics with its offering of around 30 ChanClone™ ion-channel cell lines, as well as larger players like PerkinElmer, which introduced new ion-channel expressing cell lines at the conference, are beginning to recognize this gap.
Andrew Southan, director of ion-channel biology at BioFocus DPI, feels there is still something missing. “If we are going to use automated patch clamping as a primary screening tool then, in addition to clonal cell lines, we should also seek to incorporate more physiological models such as immortalized cell lines from disease-related tissue and stem cells.
“This approach would clearly include experimental challenges to reduce off-target background currents, but undoubtedly creates an opportunity for pharmaceutical researchers and stem cell manufacturers to seriously consider the development of robust lines which work well on automated patch-clamp systems.”