It is impossible (at present) to predict the three-dimensional structure of a protein simply from its amino acid sequence. Yet to fully understand how a channel operates it is important to have some idea of what it looks like. It was knowledge of the DNA sequence that provided the first step towards understanding the relationship between structure and function. Once the genetic code of a protein is known it is possible to change it, and to produce designer channels tailored to the question you wish to address. Wonder what a particular amino acid does? Simple: change it to a different one and see what happens. It sounds straightforward, and these days it is. Now that we know the complete sequence of the human genome (and that of many other species) it is possible to look up the DNA sequence of your favourite protein in an online database and order it up from a commercial company at a cost of around a thousand pounds: it arrives within a few days, an invisible drop on a tiny piece of filter paper. Back in the 1980s, however, things were not so easy. It was usually necessary to figure out the DNA sequence yourself and that could take time – many, many years in some cases.
The Needle’s Eye
Nevertheless, the marriage of molecular biology and new electrical recording methods gradually began to address the problem of ion channel selectivity – of how channels choose between ions. It turns out that because similar charges repel one another and opposite charges attract, many channels use rings of charge at their entrances to exclude or enhance ion entry. For example, by using negative charges – which will encourage cations to enter, but repel anions – a channel can permit the passage of all positively charged cations, but exclude negatively charged anions. The crucial problem most ion channels must solve, however, is how to produce high selectivity without slowing down the rate at which ions move through the pore. And one of the most difficult questions to answer was how potassium channels allow potassium ions to permeate but not the much smaller sodium ions, which are also positively charged. It mystified scientists for years. The field had a skeletal idea, a cartoon model of how things worked, based on a plethora of functional experiments, but what was really needed was to couple this information with a structural understanding. Just what did a potassium channel actually look like? That puzzle was finally solved in 1998, when Rod MacKinnon achieved a stunning breakthrough: by growing crystals of a potassium channel protein and shooting X-rays at them, he was able to see each atom of the potassium channel for the first time. And potassium ions were caught ‘in flagrante’, trapped at various locations within the pore, so that the way they traverse the membrane could be seen in exquisite detail.
A slight figure with an elfin face, MacKinnon is one of the most talented scientists I know. He was determined to solve the problem of how channels worked and he appreciated much earlier than others that the only way to do so was to look at the channel structure directly, atom by atom. This was not a project for the faint-hearted, for nobody had ever done it before, no one really knew how to do it and most people did not even believe it could be done in the near future. The technical challenges were almost insurmountable and at that time he was not even a crystallographer. But MacKinnon is not only a brilliant scientist; he is also fearless, highly focused and extraordinarily hard-working (he is famed for working around the clock, snatching just a few hours’ sleep between experiments). Undeterred by the difficulties, he simultaneously switched both his scientific field and his job, resigning his post at Harvard and moving to Rockefeller University because he felt the environment there was better. Some people in the field wondered if he was losing his mind. In retrospect, it was a wise decision. A mere two years later, MacKinnon received a standing ovation – an unprecedented event at a scientific meeting – when he revealed the first structure of a potassium channel. And ion channels went to Stockholm all over again.2
The X-ray structure showed in dazzling detail how the potassium channel works and how it is able to support a very rapid throughput of potassium ions, so fast it seems there is no barrier to ion movement at all, while at the same time excluding the smaller sodium ions. Potassium channels, it turns out, have evolved specialized ‘selectivity filters’, short regions where the pore narrows so much that permeating ions must interact with its walls. Simply put, this region is just wide enough for a potassium ion to squeeze through, but nothing larger can pass. The passage is so small, in fact, that potassium has to shed its coat of water molecules to squeeze through. In solution, all ions are clothed in bulky coats of water and it takes a lot of effort to remove them. Potassium is happy to shrug off its coat because the selectivity filter mimics the embrace of its watery jacket. Not so, sodium. Although sodium is small enough to slip through the pore when dehydrated, the effort required to strip off its water shell is too much – much greater than the energy supplied by the clinch of the selectivity filter – so it remains fully clothed. And with its coat on, sodium is simply too big to enter.
An Open and Shut Case
Ion channels are the gatekeepers of the cell. Arguably their most important property is that they open and close, so regulating ion movements, and, crucially, this gating is tightly controlled – by the binding of intracellular or extracellular chemicals, mechanical stress or changes in the voltage difference that exists across the cell membrane.
Nerve cells talk to one another by means of chemical messengers, known as transmitters, which interact with specialized ion channels in the membrane of the target cell. The transmitter binds to a specific site on the channel protein, fitting snugly into its receptor like a key in a lock. When it does so, it triggers a conformational change in the channel protein that opens the pore and enables ion flow. We still know little about how such shape shifting takes place, or how binding of a chemical at one site leads to a structural change in another part of the protein, which may be far distant. But this type of gating is very important, not just for transmitting information between cells but also because many medicinal drugs and many poisons influence channel activity (and thereby cellular functions) by binding to the same site as the native transmitter and either blocking, or mimicking, its action.
The South American Indian arrow toxin curare, for example, binds to ion channels involved in nerve-muscle transmission and prevents the action of the native transmitter, so producing paralysis. Conversely, the psychedelic drug LSD mimics the action of the transmitter serotonin, leading to overstimulation of certain neurones in the brain. My own particular favourite, the KATP channel, is closed by binding a breakdown product of glucose known as ATP; this is how glucose metabolism leads to the closing of the channel, and thus to insulin secretion. If the binding site is altered – for example, by a mutation like that which James carries – then ATP cannot bind, the KATP channel does not close, and insulin is not secreted. The result is diabetes.
‘Voltage-dependent’ gating requires that the channel is able to sense a change in the voltage field across the membrane. All cells have a potential difference across their membranes, the inside of the cell being about 70 millivolts more negative than the outside. When a nerve fires an electrical impulse this potential suddenly alters by about 100 millivolts, the inside of the cell briefly becoming positive with respect to the outside. A hundred millivolts may not sound a lot, but in fact it is, because the membrane is very thin. When the thickness of the membrane is taken into account the electric field experienced by the channel can be colossal: of the order of 100,000 volts per centimetre. Mains electricity in the UK is supplied at 240 volts and if you have ever been unfortunate enough to get a mains shock (I hope you never have, or will) you will appreciate the enormity of the electric shock that an ion channel experiences when a nerve fires an impulse. When put like this, it is less surprising that a voltage change can alter the protein conformation, switching its shape from one state to another. How the channel senses the voltage field has only been discovered in the last twenty-five years and the precise details are still the subject of heated debate.
In resting nerve and muscle cells, the voltage-gated sodium and
potassium channels are held firmly shut by the negative membrane potential. They open only when the membrane potential becomes more positive and when this happens it triggers an electrical impulse. How this is achieved and the long and intricate work needed to unravel the story of how nerves and muscles work is considered in the next few chapters.
3
Acting on Impulse
I cannot see her tonight.
I have to give her up
So I will eat fugu.
Yosano Buson
During his voyage in the South Seas in 1774, Captain James Cook wrote the following account of the peculiar symptoms he experienced after sampling a strange and ugly fish: ‘The operation of drawing and describing took up so much time, that it was too late, so that only the liver and roe were dressed, of which the two Mr. Forsters and myself did but taste. About three o’clock in the morning, we found ourselves seized with an extraordinary weakness and numbness all over our limbs: I had almost lost the sense of feeling, nor could I distinguish between light and heavy bodies, of such as I had strength to move; a quart-pot full of water and a feather being the same in my hand. We each of us took an emetic, and after that a sweat, which gave us much relief. In the morning, one of the pigs which had eaten the entrails was found dead. When the natives came on board, and saw the fish hang up, they immediately gave us to understand it was not wholesome food, and expressed the utmost abhorrence of it.’
It is possible that Cook and his crew had inadvertently eaten puffer fish. The liver, intestines, skin and ovaries of this fish contain a virulent poison known as tetrodotoxin, which acts by blocking the sodium channels in nerve and muscle cells. Consequently, nerve impulses and muscle contraction are inhibited. The victim typically dies of suffocation caused by paralysis of the respiratory muscles. Cook was very fortunate that the amount he ate was insufficient to kill him.
Wiring the Body
Nerve fibres are used to transmit electrical signals around the body. What we generally refer to as a nerve is in fact a collection of many nerve fibres bundled together within a protective outer sheath, rather like a cable containing thousands of different telephone wires. Most nerves are located deep within our tissues to guard them against injury. The exceptions are the ends of the sensory nerves, which ramify throughout the outer layers of the skin, and the ulnar nerve that comes close to the surface of the skin at the elbow. This explains why a sharp blow to the elbow (the funny bone) produces a peculiar tingling pain that shoots down your arm: knocking the nerve excites it in the same way as a small electric shock.
Nerve cells are the building blocks of the nervous system, including the brain. They come in many shapes and sizes, but all consist of a cell body from which extend a number of fine, branched processes. Usually one of these processes is much longer than the others and is known as the nerve fibre or axon. It can be extremely long. The axons in your ulnar nerve, for example, run from your spinal cord to your fingers. The vagus nerve – the longest of the cranial nerves – runs from the brain to the stomach, and that of the giraffe can be more than 15 feet long. Yet despite its length, a single nerve fibre is very thin, with a diameter less than a tenth of that of a human hair.
Although nerve fibres are capable of conducting impulses in either direction, they usually only transmit them in one direction. Motor nerves conduct signals outwards from the brain and spinal cord to direct muscle contraction, whereas sensory nerves conduct information in the opposite direction, from our sense organs to the brain.
A typical neurone showing the axon and numerous delicate branching dendrites that arise from the cell body, and the multiple finger-like processes at the axon terminal.
The cell body is the control centre of the nerve cell: it houses the nucleus where the genetic material (the DNA) is stored. Multiple short processes branch off the nerve cell body like the limbs of a tree, hence they are named dendrites, from the Greek dendron, meaning ‘tree’. The dendrites receive numerous signals from other nerve cells and serve as first-line information processing centres, integrating all incoming information before passing it on to the cell body. Nerve cell bodies lie almost exclusively within the brain and spinal cord, where they are protected by a ‘blood–brain barrier’ which separates the blood from the cerebrospinal fluid bathing the brain and spinal cord. The brain serves as the command centre of the entire nervous system. It contains millions of nerve cells, each of which has multiple processes and multitudinous connections to other brain cells.
Acting on Impulse
Nerve cells transmit information by means of electrical signals known as nerve impulses or action potentials. These race along the nerve fibre at speeds of up to 400 kilometres per hour (250 miles per hour). The fastest nerves of all are those that are enveloped in an insulating myelin sheath. This is formed from layer upon layer of membranes of a specialized cell (the Schwann cell) that wraps itself tightly around the axon like the layers of a Swiss roll, or the paper layers enveloping a toilet roll tube. This insulating myelin sheath enables nerve fibres to conduct electrical impulses more rapidly. When it is damaged, nerve conduction is disrupted.
A myelinated nerve, showing the layers of insulating myelin wrapped around the nerve axon. The small organelle in the centre of the nerve is a mitochondrion, one of the cell’s power plants.
Multiple Schwann cells are strung out along the length of the axon. Every few micrometres, adjacent Schwann cells are separated by small gaps known as the nodes of Ranvier, which allow the naked nerve membrane to contact the extracellular fluid. Because the myelin sheath is such a good insulator, it is only at the nodes that electric current can flow from the nerve cell to the extracellular fluid. The nodes thus serve as repeater stations, boosting the action potential and enhancing its speed. In effect, the nerve impulse travels faster in myelinated nerves because its leading edge leaps forward one node at a time. This explains why myelinated nerves conduct action potentials much faster than unmyelinated nerve fibres.
A dramatic example of the crucial importance of myelin is afforded by Guillain-Barré syndrome. This rare autoimmune disease usually begins with tingling and weakness in the feet, and is followed with frightening speed by paralysis of the lower limbs, then the hands and arms, and subsequently the chest muscles, so that the victim is unable to breathe and must be kept alive by artificial respiration. Ultimately, almost all the nerves may be affected, including those of the face, so that the person may be unable to speak and can only communicate by eye blinks. In the worst case you can go from normal nerve function to near-total paralysis within a day.
Guillain-Barré syndrome is caused by antibodies produced by the body against foreign proteins that for unknown reasons also attack its own tissues, in a form of cellular friendly fire. This leads to loss of myelin and destruction of the nerve sheath, which prevents impulse conduction. The brain and spinal cord are spared because the antibodies cannot cross the protective blood–brain barrier that surrounds them, and are thus barred from reaching the myelinated fibres within the brain. Fortunately the paralysis is usually not permanent and once the antibodies have been cleared from the system, the myelin grows back. But it is a slow process, taking about a centimetre a day, and in a tall person it can be well over a year before some muscles are fully reinnervated. In many cases, full function is never regained.
Similarly, multiple sclerosis is caused by a gradual and inexorable autoimmune attack on the myelin sheath, which results in progressive impairment of nerve conduction and eventually loss of coordination and difficulty in walking. It can also cause blindness, due to damage to the optic nerves. One of its most celebrated victims was the gifted and charismatic young British cellist, Jacqueline du Pré. When she was only twenty-six, she started to lose sensitivity in her fingertips and soon she was unable to feel the strings of her cello at all. She ceased performing two years later.
Listening to Nerves Talk
We humans have been digital for years. Long before computers were even dreamt of, impulses wer
e being sent in digital code along our nerve fibres. Action potentials are said to be ‘all-or-none’, as their amplitude is constant and independent of the strength of the stimulus that evokes it: instead, increasing the stimulus intensity provokes a higher frequency of action potentials. A good analogy is with a machine gun. Press the trigger hard enough and the gun fires, but if the stimulus fails to reach a certain threshold level no bullets (or action potentials) are fired. Furthermore, much like a stream of bullets fired from a machine gun, information is transmitted along nerves as a volley of identical action potentials, with stronger stimuli eliciting a greater number of spikes. This frequency coding has significant advantages. It ensures, for example, that electrical impulses are transmitted over long distances without the information becoming garbled or the signal strength decaying.
In order to study how nerve impulses are generated and propagated, detectors sensitive enough to pick up the tiny brief electrical signals are needed. What tantalized early investigators like Galvani was that although they could easily detect the result of the nerve impulse – the twitch of a frog’s muscle – they were quite unable to record it electrically. By the middle of the nineteenth century specialized instruments known (in a nod to Galvani) as galvanometers had been developed. Using such equipment, many investigators deduced that nerves and muscles do indeed generate their own electrical signals, but they were still unable to measure them accurately. Frustratingly, if their instruments were sensitive enough they were too slow, and if they were fast enough they were not sufficiently sensitive. Neuroscientists had to await the invention of the thermionic (triode) valve, originally developed for radio communication, in order to build amplifiers capable of detecting nerve impulses accurately.
The Spark of Life: Electricity in the Human Body Page 6