Oxygen
Page 41
The idea that science should aim to postpone disabling conditions like Alzheimer’s disease, without necessarily extending life itself, has become something of a mantra. It is a mantra that goes by the name of ‘compression of morbidity’. The aim is to squeeze the bad things that happen to us at the end of life into as short a period as possible. Another way of putting this is that we want to extend the health span, while leaving the life span as it is. On the whole, people seem to find this more reassuring than that scientists want to make us live longer. The trouble is, however, that compression of morbidity makes assumptions about the extent to which we can decouple ageing and disease.
294 • BEYOND GENES AND DESTINY
If we accept that ageing and age-related diseases go hand in hand in nature, then we begin to see why it is proving so dishearteningly difficult to decouple them in medicine. It is like trying to separate the squeal from the pig, or the mind from the brain. Having said that, it is not enough just to denounce the mantra as wrong. If our attempt to extend the ‘health span’ while maintaining the lifespan is misguided, what might we offer instead that is practical and likely to succeed? After all, if nothing else, we know that some genes increase our susceptibility to age-related diseases.
We have had some sort of success in postponing conditions like Alzheimer’s disease, diabetes and cancer by a few years. If we drop this approach, what should we replace it with?
The answer is concealed in an ever-growing pile of data. Three large bodies of work provide the clues, and we have touched on all three in earlier chapters. The problem is that these three fields are largely independent. Few researchers are comfortable in crossing the boundaries of their own field. As medical research grows ever more specialized, it becomes at once less acceptable and more necessary to transgress the limits of personal expertise. Perhaps this is the most valuable role of the science writer: writers must transgress their own expertise as a matter of routine, and flights of fancy can at least be brought down to earth by experts. In my own case, everything I have to say is based on the work of others, but I cannot find a synthesis in the literature that straddles the three fields.
I shall therefore run the risk of error or inadvertent plagiarism and put forward my own ideas.
The first clue comes from the mitochondrial theory of ageing: oxidative stress rises gradually through our lives, especially in the mitochondria, and this rise is the causal basis of ageing (if not of age-related diseases).
There are two objections to this claim. First, the rise in oxidative stress is hard to measure in practice, and some researchers dispute that it happens at all. I hope I convinced you in Chapter 13 that mitochondria alone do cause a rise in oxidative stress, and that this is the causal basis of ageing.
We have yet to establish whether mitochondria can also cause age-related diseases. The second objection is more treacherous: we have had little success in blocking the putative rise in oxidative stress using antioxidant supplements. This failure is often taken as evidence that there is no rise in oxidative stress, or that the rise is inconsequential. This is facile logic. It might simply be that dietary antioxidants are not up to the job — indeed cannot be up to the job. In the last pages of Chapter 10 I noted that dietary antioxidants are far from a panacea. Antioxidants may even be counter-
The Double-Agent Theory of Ageing and Disease • 295
productive in that they can suppress the powerful genetic response to oxidative stress that is produced by proteins such as haem oxygenase and metallothionein. Thus, clue 1 is double: oxidative stress rises as we get older, but we can only gain limited protection from dietary antioxidants.
Clue 2 actually derives from the failure of antioxidant supplements to extend life, and comes from the field of cellular signalling. Signals are as critical to the behaviour of cells as electronic communications are to modern society. Chemical signals control the expression of genes —
switching them on or off — in the same way that information constrains our decisions as individuals in society. Whether cells divide or mature into neurons, die or become cancerous, secrete hormones or absorb salt, does not depend on their genes. All our cells have the same genes. The behaviour of a cell depends on which genes are active at any one time, and this depends on the signals it has received. The signals are converted into an appropriate response through the action of transcription factors, regulatory proteins that bind to DNA and direct the transcription of particular genes into proteins. We saw in Chapter 10 that the activity of several important transcription factors depends on their oxidation state. Many types of physiological stress, such as infection, radiation poisoning and inflammation, cause an increase in oxidative stress. Transcription factors such as NFB and Nrf-2 become oxidized and migrate to the nucleus, where they bind to DNA and coordinate the transcription of ‘stress’
genes.4 The products of these genes muster resistance to the threat. Thus, clue 2 stands opposed to clue 1: some sort of oxidative stress is a necessary signal for cells to marshal their genetic response to physiological stress. If we block oxidative stress, we may make ourselves more vulnerable to infection. Seen in this light, it is quite conceivable that we are ‘refractory’
to large doses of dietary antioxidants because they interfere with our response to stress.
Clue 3 comes from the evolutionary theory of antagonistic pleiotropy — the idea of a trade-off between the detrimental effects of genes in old age and their beneficial effects in youth. The theory was first proposed by George C. Williams in 1957, and has since been accepted by most evolutionary biologists, though it is not in common currency in medical 4 The migration of NFB to the nucleus requires its oxidation. Once in the nucleus, however, it must be restored to its unoxidized state before it can bind to DNA. Thus, the action of NFB is carefully stage-managed, and it only exerts an effect if the cell retains control of the oxidation state of the nucleus. If both nucleus and cytoplasm are oxidized, it is more likely that the cell will fail to muster a response, and will die instead.
296 • BEYOND GENES AND DESTINY
science. The role of pleiotropy in the diseases of old age was discussed by Williams and the physician Randolph Nesse in an enlightening book on the “new science of Darwinian Medicine”, Evolution and Healing, first published in 1994. Much of this book is packed with vivid examples, but I found the section on pleiotropy disappointing. We are offered the examples of Alzheimer’s disease and haemochromatosis (in which heavy iron absorption overcomes the risk of anaemia in youth, but courts misfortune in middle age; see Chapter 10), and one or two others. Tantalizingly, they cite Paul Turke, an evolutionary anthropologist and physician who argues that the whole immune system is age-biased, in that the noxious oxidants released by immune cells to kill invading microbes can also damage the body; but Turke, though quite right, did not close the circle. Such a theory cannot, in itself, explain why mice, which have elaborate immune systems of their own, should die of age-related diseases after four years while we take 70. In a more recent textbook on evolutionary medicine, based on a conference in Switzerland in 1997, the concluding discussion notes tersely that “At this time, few examples of tradeoffs have been established.” Thus, clue 3 is that, despite its theoretical utility, few concrete examples of pleiotropic trade-offs have ever been described. Could something be missing from the equation?
I suggest that there is a trade-off between oxidative stress as a signalling pathway that musters our defences against infection, and oxidative stress as a cause of ageing. In effect, the diseases of old age are the price we pay for the way in which we are set up to handle infections and other forms of stress in our youth. In both cases, the shadowy agent pulling the strings is oxidative stress. The outcomes are diametrically opposed: resistance to disease in youth and vulnerability to disease in old age. The duplicitous role of oxidative stress is central to both, in what I shall call the ‘ double-agent’ theory of ageing and disease (Figure 12).
The problem is that oxidative
stress is a necessary part of our response to infection: without it, we are unable to mount a genetic defence against pathogens. Oxidative stress activates transcription factors such as NFB, which coordinate the broader genetic response promoting inflammation and stress resistance.5 Unfortunately, the rise in oxidative stress during ageing also activates NFB. The pressure to eliminate infection in youth is 5 NFB is not the only transcription factor that responds to oxidative stress; Nrf-2, AP-1, Rel-1, P53 and several others do too. However, in medical circles NFB has become virtually synonymous with the stress response, so I shall focus on it here.
The Double-Agent Theory of Ageing and Disease • 297
far higher than the pressure to eliminate inflammation in old age. We cannot get rid of NFB (we would succumb to infection) and yet its effects in promoting inflammation change the whole balance of the body as we get older. It is this shift in balance, rather than the time that has elapsed, that is responsible for the negative pleiotropic effects of other genes.
Exactly why infections should produce a rise in oxidative stress is uncertain, but it does seem to be the case that they do. In many cases, (a)
(b)
Arthritis
Infection
Heart disease
ation
ation
Cancer
m
hronic
m
C
Inflam
inflam
Oxidative
stress
Oxidative
NFκB
stress
NFκB
Gene
Gene
activation
activation
Young cell
Old cell
(External reversible trigger)
(Internal irreversible trigger)
Figure 12: Schematic representation of the ‘double-agent’ theory of ageing.
In the young cell (a), infections (an external reversible trigger) cause a rise in oxidative stress within the cell. Oxidative stress activates NFB, which migrates to the nucleus, where it orchestrates the transcription of ‘retalia-tory’ genes. These genes code for stress proteins and proteins that mediate inflammation, such as tumour necrosis factor and nitric oxide synthase.
Active inflammation resolves the infection. Once the trigger is removed, oxidative stress returns to normal. In the old cell (b), an equivalent rise in oxidative stress is brought about by leaky mitochondria (an internal irreversible trigger), which also activate NFB and the inflammatory response. In this case, it is impossible to resolve the trigger, so the inflammatory response becomes chronic. This contributes to the diseases of old age, and also attenuates the acute response to infections and other physical stresses. Because oxidative stress is pivotal to our recovery from infections in youth, and therefore affects our likelihood of surviving to have children, it is positively selected for by natural selection to our own detriment in old age.
298 • BEYOND GENES AND DESTINY
oxidative stress is exacerbated by activated immune cells, such as neutrophils (which produce potent oxidants to kill the pathogen); but cell-culture experiments in the absence of immune cells suggest that the mechanism is more subtle and fundamental than this. This is a critical point: infections increase oxidative stress regardless of whether the immune system is involved. In the case of influenza, for example, Heike Pahl and Patrick Baeuerle, at the University of Freiburg in Germany, have shown that a single viral protein — haemagglutinin — induces oxidative stress in cultured cells.6 Pahl and Baeuerle showed that this rise in oxidative stress activated NFB, which in turn coordinated the genetic response of the organism to the infection. Conversely, if the oxidative stress was abolished using an antioxidant such as dithiothreitol, neither NFB nor its subordin-ate genes were activated. A similar dependency on oxidative stress, and abolition by antioxidants, has been demonstrated in infections with many other viruses, including human immunodeficiency virus 1 (HIV-1), hepatitis B and herpes simplex, as well as by components of the bacterial cell wall such as endotoxin and lipopolysaccharide. In each case, the infection produces oxidative stress, which activates NFB, which orchestrates the transcription of numerous other genes. The entire response can be short-circuited by blocking the rise in oxidative stress using antioxidants.
The response to NFB is generally two-pronged: a bolstering of resistance to inflammation — a stress response — coupled with an inflammatory attack on the invading microbe.7 The inflammatory attack can be severe: fever is part of the host’s reaction to infection; it helps us clear the infection. Yet fever is obviously detrimental to our long-term health, insofar as it is disabling. We cannot expect to persist in this state for a prolonged period. Similarly, our response to endotoxin or malaria can be too violent — if we have a serious infection, the ferocity of our own immune response may push us over the brink into septic shock or cerebral malaria, which might well kill us. Some pathogens have learned to modulate the inflammatory response, or even take advantage of it. (For example, HIV
has several defensive genes that are activated by NFB, and it uses inflam-6 Pahl and Baeuerle suggest that the mechanism may depend on the sheer volume of viral protein accumulating in the cell’s export pathway, the endoplasmic reticulum. Overload of the endoplasmic reticulum releases calcium, which in turn activates enzymes such as cyclo-oxygenase and lipoxygenase, which increase the production of oxygen free radicals.
7 There are, in fact, at least seven related forms of NFB, which are activated in slightly different ways, and which activate different selections of genes. One form, for example, increases stress-resistance but does not promote inflammation. When I talk about NFB here I mean the most common, pro-inflammatory form.
The Double-Agent Theory of Ageing and Disease • 299
mation as a signal for proliferation.) In general, though, inflammation is a positive force that has been selected for by evolution because it helps to resolve infections. Once the infection has been resolved, the inflammatory attack dissipates and we regain health. In other words, once the pathogen has gone, oxidative stress falls and NFB is switched off. As a result, the genes controlling the stress and inflammatory responses are switched off.
Normal housekeeping genes are switched back on. The body reverts to its
‘peace-time’ routine. The whole process is reversible.
Now think what happens during ageing. Our mitochondria leak free radicals, bringing about an insidious rise in oxidative stress as we get older. There comes a point when the oxidative stress is severe enough to activate transcription factors like NFB. We begin to go through a low-grade stress response and inflammation. Virtually all diseases of old age are characterized by a chronic activation of stress proteins and persistent inflammation. Because broken mitochondria cannot be repaired, the situation is self-perpetuating — or worse, self-exacerbating. Inflammation damages cells and bodily structures, and so gives the jittery immune system a ‘real’ target. Proteins that are normally hidden inside cells, or behind barriers such as the blood–brain barrier, are exposed to immune surveillance and attack. Any concomitant conditions are exacerbated. We might be able to dampen this attack by using anti-inflammatory drugs, but, unlike infection, we cannot remove the primary cause: we cannot mend broken mitochondria. Nor can dietary antioxidants prevent the mitochondrial leak, so they too cannot counter the oxidation of the cell.
On the contrary, as we have seen, antioxidants may even weaken the stress response — which is, after all, a response to genuine physiological stress.
No gene is an island, any more than a man is an island. If a gene becomes more or less active, the effects are felt by other genes. The activity of all genes depends on their immediate environment, in other words, the chemical balance inside cells. Oxidative stress shifts the spectrum of active genes, regardless of the exact cause of the stress. The rise in oxidative stress over a lifetime means that many genes that are active when we
are 20 are less active when we are 70, and vice versa. Other genes keep working throughout our lives, but their effects shift because their environment changes. In the same way, a song accompanied by solo violin in an intimate setting differs from the same song chanted in a football stadium with a backing rock band. As we age, a pleiotropic gene exerts its negative effects because its environment has become oxidized and pro-inflammatory, not because a particular period of time has elapsed. If we
300 • BEYOND GENES AND DESTINY
wish to overcome the negative effects of pleiotropy, we must therefore prevent the oxidation of cells and tissues with age.
Before we consider some particular examples, we should ask if there is any empirical evidence for concerted changes in gene expression with age. Do cells and tissues really become more oxidized? If so, does this really change the pattern of genes that are switched on? If rhesus monkeys are anything to go by (which share 95–98 per cent of their genes with humans), the answer is almost certainly yes. In Chapter 13, we discussed a study of rhesus monkeys, carried out by Richard Weindruch and his team at the Wisconsin Regional Primate Center, Madison, and published in 2001. In one arm of the study, the team compared the activity of 7000
genes in young animals (aged 8 years) with that of the same genes in ageing animals (aged 26 years; the maximum lifespan of rhesus monkeys is 40 years). The similarities and differences were striking. Of the 7000
genes, about 6 per cent changed their activity by twofold or more in the course of 18 years, some becoming more active (300 genes) and others less active (149 genes).8 Many of the genes that became more active with age were concerned with inflammation and oxidative stress (including a rise in the activity of the NFB gene). The genes that became less active with age were concerned mostly with mitochondrial respiration and cell growth.