What on Earth is Nitric Oxide Doing Here?

The year 1952 was a bad one for Londoners. The city was frequently shrouded in smog – something between a smoke and a fog – causing traffic chaos and considerable physical distress, particularly amongst the more elderly members of the population. Smog forms when the moisture in a fog, for which London had long been famous, condenses around tiny particles from industrial emissions. Then sulphur dioxide, coming from the burning of coal rich in sulphur in open fires, dissolves in the moisture to give the acrid taste so characteristic of smog. The authorities took rapid action to reduce the amount of smoke and sulphur dioxide released into the air in urban areas. London's smog episodes were quickly eradicated.

London was by no means the only city with smog problems. For some years similar measures had been taken in Los Angeles to reduce smoke and particulate emissions, but the chronic smog problem had failed to ease. It transpired, much to the surprise of the local residents, that the Los Angeles smog had a different cause. Car exhaust fumes, in which the air above Los Angeles abounded, contained not only carbon dioxide and water but also traces of two oxides of nitrogen (nitric oxide, NO, and nitrogen dioxide, NO2) as well as volatile organic compounds. To complicate matters further, the strong Californian sunlight provided suitable energy for a complex series of reactions involving NO which gave a 'photochemical smog'. This was just as distressing as London's industrial smog. Exposure caused eye and bronchial irritation in humans, it blanched the leaves of trees and it accelerated the corrosion of rubber. This is the context in which, before 1987, NO was usually mentioned.

NO is a simple gas and each molecule contains one atom of nitrogen and one atom of oxygen. It just gets mentioned in a school chemistry syllabus. Its most well-known property is that when it is released into the atmosphere it reacts with oxygen to form a brown gas, nitrogen dioxide NO2. This means that if NO forms in the atmosphere the result is a mixture of NO and NO2. Formation of NO does occur naturally, but only in the extreme conditions of a lightning strike. For convenience the mixture of NO and NO2 is known as NOx, pronounced to rhyme with 'socks'. Areas with high NOx concentrations (and hence photochemical smog) sometimes appear on weather charts to warn people to keep away if they can.

Without NOx photochemical smog does not form and so cars in Los Angeles, and elsewhere, are now fitted with catalytic converters to destroy the NOx before it enters the atmosphere. This process is described in more detail later (Chapter 7). With increasing understanding of air pollution NO and NO2 were seen as major villains in the saga and great efforts were made to banish them. However, in 1987 scientists were astonished to learn of another side of NO's complex character: human life depends upon it.

Vessels that carry blood around the body can enlarge, or dilate, and this fixes the amount of blood that is delivered to specific tissues or organs. To see how this happens we have to look at the structure of a blood vessel. The hole in the middle, down which the blood flows, is called the lumen. On the inside of the lumen is a single layer of cells known as the endothelium (endothelial cells) and the wall of the vessel consists largely of smooth muscle (Figure 1.1). It is called smooth because of its appearance under the microscope. When the muscles of the artery wall relax the lumen enlarges and more blood flows through the vessel, provided the heart is pumping properly. If the muscles contract the lumen decreases in size. Then, either less blood flows along the vessel or the heart has to work harder to maintain the flow of the same amount of blood. This is why 'hardening' of the arteries, which prevents muscle relaxation, puts strain on the heart.

It had been known for many years that certain substances, such as acetylcholine, bring about vascular muscle relaxation and some of these substances are released into the bloodstream when relaxation is required. During the late 1970s this matter was under investigation by one of the world's leading muscle physiologists, Robert Furchgott (Figure 1.2), working at the Downstate Medical Center in New York City. He wanted to know how and why substances like acetylcholine affected muscles in this way and he was working with small segments of artery taken from a rabbit. He and his student (John Zawadzki) encountered the most trying of situations in scientific research; their results were not reproducible. In one set of experiments acetylcholine was an active muscle relaxant but in another set of apparently identical experiments it had hardly any relaxing effect at all. The most natural thing to do when this happens is to give up and go fishing, but Furchgott and Zawadzki were made of sterner stuff and they eventually concluded that the result they obtained depended on whether or not the endothelium was intact. It is extremely easy to damage the endothelium when the experiment is being set up.

If the endothelium was undamaged acetylcholine worked fine; remove the endothelium by rubbing, either intentional or accidental, and acetylcholine had no effect at all. Furchgott concluded, quite rightly it is now clear, that acetylcholine was not acting directly on the muscle cells but on the endothelial cells which, in turn, produced another chemical species that diffused into the surrounding muscle and began the process of relaxation. Such a substance is called a messenger molecule as it tells the muscles what to do. Furchgott named this particular messenger molecule the 'endothelium-derived relaxing factor'. It was seen as important enough to warrant a set of initials (but not an acronym): EDRF. The chemical identity of the EDRF was a matter of intense study by many scientists during the 1980s. It was assumed to be a complex organic molecule, like most messenger molecules in the body, but it stubbornly resisted identification. You can see the magnitude of the challenge. The EDRF is produced, along with hundreds of other chemicals, by endothelial cells in quantities around a thousand millionth of a gram, well below the limits of detection by normal chemical means.

The problem was solved by an inspired guess and, although the first person to make the suggestion in print was Furchgott himself, the same idea occurred to others. As is described in more detail in Chapter 5, for many years drugs have been available for bringing about relaxation of the muscle in blood vessels. One, glyceryl trinitrate (1.1), is used either as a lozenge or in a puffer by people suffering from angina. Another one, sodium nitroprusside (1.2), is used in the management of cardiac crises to relax vascular muscle and so lessen the workload on the heart. The action of both drugs was discovered by accident and both now have valued places in medical practice. Why they act as muscle relaxants was, in the 1980s, not known but it was generally assumed that, along with many clinical drugs, their mode of action was completely different from that of naturally occurring vascular muscle relaxants, but perhaps this was not correct. Possibly both glyceryl trinitrate and sodium nitroprusside are transformed within the body into something that is also the naturally occurring messenger molecule for effecting vascular muscle relaxation, the mysterious EDRF. Both drugs contain a group of atoms that includes nitrogen and oxygen and so, the speculation went, the EDRF could be NO. With hindsight the idea is fairly obvious but at the time it was considered absurd. No-one had ever detected NO, or anything like it, in a living system. However, once the suggestion had been made several groups of researchers set about testing it. Two groups were successful in showing that the EDRF is indeed NO. One was led by Louis Ignarro (Figure 1.3), working at the University of California Medical School in Los Angeles and the other was led by Salvador Moncada (Figure 1.4), then at the Wellcome Research Laboratories in Beckenham, England.

They both used a bioassay to establish the identity of the EDRF. What this amounts to is a series of experiments in which a solution of NO and a solution of the EDRF were shown to have identical characteristics with respect to the relaxation of vascular muscle. Thus the evidence, initially, was circumstantial but quite quickly Moncada and his group produced direct evidence for the production of NO from cultured endothelial cells. This is described in more detail later (Chapter 3). Accounts of the work were published simultaneously by the two groups in January 1987. One paper appeared in the journal Nature and the other in Proceedings of the National Academy of Sciences of the USA. The publications were greeted with excitement and some incredulity. Excitement because of their importance in understanding muscle physiology, and incredulity because the findings were was so unexpected. The EDRF was not the complex organic molecule we had expected but a small, diatomic, inorganic molecule. The unexpected nature of the discovery led many to express doubt. NO is produced in lightning strikes and in internal combustion engines, not in mammals at body temperature. Anyway, it is a harmful substance, responsible for photochemical smog, not a benign substance keeping our cardiovascular system in a healthy state. Also, many said that NO is too reactive to function in a living system. It reacts with oxygen to produce NO2, which is highly toxic. None of these objections was sustained on further investigation and it is now firmly established that NO is the EDRF and responsible for vascular muscle relaxation. When sufferers take glyceryl trinitrate for relief of the symptoms of angina the drug is converted into NO, which causes the cardiac arteries to enlarge and more blood to flow. The significance of the discovery was recognized by the award of the 1998 Nobel Prize for Physiology or Medicine, the highest award in the world of science, to three scientists involved in advancing our understanding of muscle relaxation, Ferid Murad, Robert Furchgott and Louis Ignarro. The last two were directly involved in the discovery of NO. Astonishingly the award did not include Salvador Moncada, a matter that has been commented upon publicly on a number of occasions. The papers of the Nobel Committee are not made public for 50 years, long after the deaths of most of those who are so puzzled and disappointed by this omission.

Had that been the end of the NO story it would have been a remarkable discovery, adding another piece to the jigsaw puzzle that makes up the chemistry of life. However, it was only the beginning. This small, diatomic, inorganic molecule turns out to have so many functions in living processes that we wonder how we missed it for so long. At the same time, NO is still polluting the atmosphere and more and more cars are being fitted with catalytic converters to rid the air of this dangerous chemical. Unlike almost any other molecule NO has two diametrically opposed characteristics, giving us both life and death.

It is now firmly established that NO is a cellular messenger molecule bringing about vascular smooth muscle relaxation, but exactly how does it do so? Muscle relaxation is a positive process rather than merely the absence of contraction. It is accompanied by the conversion, within the cell, of guanosine-5-triphosphate (GTP) into cyclic guanosine-3,5-monophosphate (cGMP), shown in Scheme 1.1. This conversion initiates a whole cascade of reactions, leading to protein phosphorylation (attachment of phosphate to a hydroxyl group) and muscle relaxation, processes well understood but too complex to be detailed here. It suffices to know that conversion of GTP into cGMP is catalysed by the enzyme-soluble guanylate cyclase (sGC, also known as guanylyl cyclase), an enzyme activated by NO.

The process observed by Furchgott is shown in Figure 1.5. Acetylcholine binds to a receptor on the surface of an endothelial cell and, in a manner described in more detail later, stimulates the cell to produce NO, which then diffuses into the underlying muscle cell and activates sGC to bring about relaxation. Destroy or damage the endothelium and acetylcholine will have no effect.

sGC is found in the liquid part of almost all mammalian cells, with the highest concentration in the cells of the brain and lungs. It is described as soluble GC because it is in solution and there is another form, particulate GC, that is found in membranes. At the time of writing the three-dimensional structure of sGC has not been determined but the enzyme is known to contain a haem component. Haem is the nonprotein part of haemoglobin and consists of an iron atom held in the centre of a porphyrin ring (1.3 in Scheme 1.2). The protein part of sGC is a dimer of two similar but not identical subunits, named α and β. It is the haem component that makes sGC sensitive to NO. If the haem component is removed, as may happen during isolation and purification, the sensitivity of sGC to NO is lost. The haem component is bound to the enzyme via an imidazole ligand (part of histidine at position 105) in the β subunit. Spectral evidence suggests that the iron is present as Fe2+ bound to imidazole, as in nonoxygenated haemoglobin. The latter readily binds both oxygen and NO. In contrast, sGC, because of structural features in the α subunit, has a remarkably low affinity for oxygen yet readily binds NO. If it were not for this distinction, sGC in cells would be oxygenated and useless as a receptor for NO.

When NO binds to the haem iron of sGC, the bond to the histidine at position 105 (which is trans to the incoming NO) is broken (Scheme 1.2). This results in the formation of a pentaco-ordinated nitrosyl haem complex; the reason for this is explained in Chapter 9. At the same time the iron moves out of the plane of the porphyrin ring, which exposes the catalytic site of the enzyme and allows GTP to enter. Catalytic conversion of GTP into cGMP is dependent on the presence of Mg2+ ions as this cation binds to a number of negatively charged groups in the enzyme and gives the required rigidity to the three-dimensional structure. The other metal known to be necessary for sGC activity is Cu+. Animals fed on a copper-deficient diet exhibit marked loss of endothelium-dependent smooth muscle relaxation. The role of Cu+ ions is unclear but, as described in Chapter 4, Cu+ ions strongly catalyse the release of NO from S-nitrosothiols and so nitrosation of thiols on sGC may be a step in the activation of the enzyme. From this brief account of just one enzyme it is apparent how much inorganic chemistry there is in the chemistry of living processes.

sGC is a constitutive enzyme in that it is present nearly everywhere in animal cells all of the time. In conditions such as septic shock (see Chapter 11), where there is overproduction of NO, it is important to know how far elevated levels of sGC are also responsible for the massive arterial dilation that leads to life-threatening loss of blood pressure. Drugs to inactivate sCG, as well inhibitors of the enzyme responsible for the production of NO (see Chapter 3), may be relevant to the treatment of this condition. The same consideration applies to other faults in the cardiovascular system. Raynaud's phenomenon ('cold finger syndrome') is a troublesome condition that is caused by very restricted flow in peripheral blood vessels, particularly in the fingers. In extreme cases the sufferer cannot go out in cold weather without electrically heated gloves. It is, for some reason, much more common amongst young women than men and tends to lessen with age. The simplest explanation of the effect is that it is due to a shortfall in the production of NO and there is insufficient dilation of minor arteries to allow warming blood to reach the fingers. This condition could be cured by applying NO-releasing drugs to the skin (transdermal delivery of NO), and this is quite feasible. However, experimental evidence suggests that there is also a shortage of sGC in the fingers of sufferers, a situation far more difficult to relieve by means of drugs. Much more research on sGC is required if we are to appreciate its role not only in vascular smooth muscle relaxation, but also in the other living processes in which it plays a part. The elucidation of its activation by NO has done much to stimulate interest in sGC and surely it must be one of the next enzymes to yield its secrets.