. is also the proud owner of a heartbeat working under the control of a number of membrane spanning proteins known as voltage dependent ion channels. Andy Extance gets to the heart of the matter.

. is also the proud owner of a heartbeat working under the control of a number of membrane spanning proteins known as voltage dependent ion channels. Andy Extance gets to the heart of the matter.

In regulating the heart’s electrical currents, as well as many other biological functions, voltage dependent ion channels are crucial to all of us, a fact reflected by the recent award of the Nobel prize for chemistry to Roderick MacKinnon for his research in this field. Recently these channels have come under suspicion as the medical community investigates genetic mutations responsible for increasing the incidence of heart disease, and the pharmaceutical industry examines various medicines displaying adverse cardiac side effects which are all related to potassium ion channels in the heart.

The working of the heart is very easy to take for granted. Except for those occasions when it beats much faster and stronger than normal, or when something goes wrong, it seldom makes its presence felt. The heartbeat is however a complex entity. Periodic contractions of the heart muscle are triggered by electrical excitations, controlled by the influx and efflux of sodium, calcium and potassium ions. This action forces blood through the heart and to the rest of the body, carrying oxygen to our organs and other tissues, and deoxygenated blood to the lungs for replenishing.

While such phenomena as the quickening and slowing of the heartbeat are under the autonomic nervous system’s control, the pacemaker driving the heart is otherwise independent. If the controlling nerves were to die, the heart would continue to beat. This is largely due to the manner by which electrical excitation is transmitted within the heart. Gap junctions - junctions between two cells that have many pores - permit the rapid spread of excitation between the heart’s cells. The speed of propagation allows muscle cells to function in a synchronised manner, resulting in the pulsating effect we are all familiar with. Within these gap junctions the different ions carry charge between cells through their ion channels, establishing the electrical currents that are the driving force behind the heartbeat.

The change in voltage across the heart’s cell membranes that elicits a muscular contraction is known as the ’action potential’. At the start of the heartbeat there is a resting negative potential across the cell membrane wall, caused by potassium ions effluxing from heart cells through potassium ion channels that sit open at this voltage. Sodium channels quickly open and allow a rapid sodium ion influx, changing the membrane potential from negative to positive, a process known as depolarisation, which is accompanied by the triggering of slower acting calcium channels. After a short delay, these channels open prompting an inrush of calcium ions that keep the membrane potential positive. The calcium ion channels are also slow to close, so the positive potential persists. Finally, as both the calcium and sodium channels close, the cell is repolarised back to the negative resting membrane potential by the delayed activation of potassium channels. This cyclical ’action potential’ triggers calcium movement to the heart cell’s interior, helping releases from intracellular calcium stores to activate myofibrils, the cellular substructures responsible for the heart’s muscular walls contracting.

Clearly, any problems with the ion channels could have very serious consequences for this complex yet essential process. The most serious disorder associated with ion channels in the heart is considered to be long QT syndrome (LQTS), which may cause cardiac arrhythmias and sudden death, even in otherwise young and healthy individuals. LQTS is characterised by a prolonged cardiac action potential, corresponding to a lengthened QT interval (the QT interval is a feature of the standard human heartbeat’s electrocardiogram). This prolongation is caused by delayed heart repolarisation; a consequence of potassium currents with reduced magnitudes, most specifically a rapid repolarising current known as I Kr. LQTS may be an inherited disorder, caused by mutation in the genes coding for ion channels, or an acquired one, caused by the ion channels being blocked by drugs.

As an inherited disorder the potential fatality of LQTS is serious, but for drugs, sudden death is obviously a completely unacceptable side effect, usually causing any such drug to be immediately withdrawn from the market. As Fabrizio de Ponti, professor of pharmacology at the University of Bologna, Italy, explains, ’in the recent past, QT prolongation was still considered as a pharmacological curiosity with little clinical impact, but the number of fatalities associated with these drugs has rapidly changed this view’.

For example, Janssen Pharmaceutica’s gastrointestinal drug indicated for oesophageal reflux, Cisapride, was voluntarily withdrawn in 2000. Janssen cited 80 deaths amongst users. Other drugs that have been withdrawn include Hoechst-Marion-Roussel’s antihistamine Terfenadine, which was withdrawn in the US in 1998, and Glaxo-Wellcome’s antibacterial Grepafloxacin, which was taken off the market in 1999. At least 140 other compounds also show similar QT prolongation properties.

Fear of disastrous side effects has encouraged the pharmaceutical industry to add its might in researching acquired LQTS to the efforts of those investigating the congenital disorder. Both sets of research point a finger at one major culprit. In approximately half the inherited cases, and virtually all the acquired cases of LQTS the same ion channel is implicated; a potassium channel encoded for by the improbably named Human-ether-a-go-go-related-gene, or the Herg ion channel for short.

The study of Herg and its effect upon LQTS has only been under way since the mid-1990s and despite the realisation very early on that both acquired and congenital LQTS operated through the Herg channel, questions about congenital LQTS seem to have been answered a lot more thoroughly than the drug induced variety.

Herg channels have been described as an assembly of four identical Herg protein subunits into a protein tetramer that is transported to the cell membrane to allow the passage of potassium ions. As with all proteins, a Herg protein possesses an amino- and a carboxy-terminus at opposite ends of its length, and it seems that each is responsible for a proposed mechanism to reduce the I Kr current in patients with Herg mutations. The carboxy-terminus is implicated in coassembly or trafficking abnormalities, in which mutant subunits either do not coassemble with normal subunits, or if they do, are not transported to the cell membrane to fulfil their channelling function. Mutations at the amino-terminus result in defective channels that prevent the passage of potassium and hence stop the I Kr current taking part in repolarising the heart.

While the link between Herg and LQTS was published in 1995, and Terfenadine’s interaction with Herg in 1996, the full extent to which drugs and potential drugs interact with Herg has only been realised in the last three to four years. A consequence, or possibly a cause, of this is a lack of quantitatively accurate tests to examine to what extent medicines and potential medicines interact with Herg.

Tools useful for examining the function of Herg are unreliable under conditions required for accurately analysing the effects of different chemical structures. For example, in tests where Herg systems are modelled in frog egg cells, problems are often experienced distributing the drug molecule into the egg yolk. Standard ’patch clamp’ techniques used to measure ion currents in mammalian cells have also presented problems, mainly related to how long it takes to obtain results.

Recently it has been accepted that blocking the Herg channel can cause problems in the heart. Uncertainty about the significance of this presents still further obstacles for the pharmaceutical industry.

Gail Robertson at the University of Wisconsin, US, one of the first Herg behaviour researchers, explains: ’Another challenge is that Herg block does not necessarily predict prolonged action potential duration and LQTS. Some drugs, probably because of compensatory blocking of other channels as well, block Herg but have low risk for inducing cardiac side effects. Nonetheless, in the current regulatory climate that regards Herg block as a leading concern in drug development, drugs that block Herg are highly unlikely to be brought to clinical trials. The result? Many drugs with a negligible risk and potentially high therapeutic value will never be available, unless we can identify a better preclinical test or find a way to screen out individuals at risk for drug-induced LQTS.’

As the financial burden of drug withdrawals and difficulties with drugs in clinical trials has made itself felt in the pharmaceutical industry, significant efforts have been made to improve the situation. MacKinnon’s Nobel prize winning crystal structure of bacterial potassium channels used in combination with knowledge about the Herg protein sequence, as determined by those studying congenital LQTS, has significantly improved the situation. ’The more we learn about the precise molecular composition of ion channels, the more predictive our drug counterscreening technologies may become,’ Robertson explains.

Several computational Herg models have been developed in conjunction with a database of structures known to have affinity for the Herg potassium channel. These include ’Pharmacophore’ models of generalised structural criteria required for binding to the Herg channel, which allow an early and relatively inexpensive method to assess whether new drugs match these criteria for unwanted interaction. De Ponti, who participated in a project developing one such model, describes this method’s uses and limitations. ’In the near future, it will be possible to predict interaction with Herg potassium ion channels with acceptable sensitivity and specificity in the preclinical development of new chemical entities. However, in my opinion, computational testing cannot fully replace other existing tests. Several tests should be used to rule out interaction with Herg channels: no single test has an absolute predictive value.’

The knowledge acquired since the Herg channel was discovered represents a significant achievement for the pharmaceutical and medical communities, one that may be used to great effect moving forward in treating LQTS. While identifying the genetic principles behind Herg-derived congenital LQTS currently stops short of providing treatment, advances in DNA sequencing may soon enable patients with a disposition towards the disease to be identified, then drugs and other substances that may exacerbate it can be avoided.

It must be said that not all drug interactions with Herg are completely undesirable. The class III anti-arrhythmic drugs have been shown to bring about their effects through that very mechanism. This being the case, genetic LQTS resulting from malformed ion channels could perhaps be treated with small molecule drugs that hold existing channels open, allowing a repolarising current to be re-established.

The pharmaceutical industry is now treating the part its drugs play in LQTS onset much more seriously. The knowledge obtained over the past decade has enabled molecular models to be used to investigate the problem, testing against the Herg channels to be carried out much earlier in the drug discovery process and has encouraged more accurate tests to probe the behaviour of drugs at these ion channels to be developed. All those at the Herg channel’s mercy, be it those whose DNA codes for an LQTS inducing mutation, those who must take drugs which put them at risk of LQTS, or the pharmaceutical industry, from whom the Herg channel claims millions of dollars each year, truly have reason to take heart.

Acknowledgements

Andy Extance is a medicinal chemist working at Tripos Discovery research centre, Bude, UK