The longevity of the long distance runner, part 2: the basic science.

In my previous blog post I had posed the question: What determines the rate at which a runner’s performance declines with age?   As a prelude to addressing the scientific evidence, I had discussed anecdotal evidence gleaned from the family history, lifestyle and training of the two greatest veteran distance runners of all time: Derek Turnbull and Ed Whitlock. The anecdotal evidence suggested that genes, life-style and training all played a role. Especially in the case of Ed Whitlock, it is probable that having long-lived forebears; deferring very high volume training until after his retirement from work; and adopting a training program designed to minimise stress all contributed to his extraordinary longevity as a world-record breaking marathoner into his mid-eighties.   However, anecdotal evidence provides little basis for drawing general conclusions. What does science tell us?

At first sight, the answer appears to be that science provides a lot of obscure information that in practice offers us little guidance as to how we might adjust our life-style or training to maximise longevity, either as functioning living creatures or more particularly, as athletes. However, if we do not allow ourselves to be put-off by the apparent complexity of the story, it is possible to establish the basis for some simple speculations that might be useful in practice.

Although my primary focus is on longevity as a runner, longevity as a runner is very closely linked to healthy aging.   Healthy aging is not merely freedom from identified illnesses, though many illnesses are common in the elderly and unhealthy elderly people are often afflicted by multiple illnesses.  In fact it is probably more appropriate to consider that healthy aging is a state characterised continued good functioning of all systems of the body, that creates a low vulnerability to illness and is also a requirement for longevity as a runner.

Are there genes for longevity?

There have been several large studies of genes associated with longevity in the general population. These indicate that many genes contribute a small amount to longevity but few contribute an appreciable amount. In fact only one gene has emerged as a significant predictor of longevity in genome-wide association studies: the gene for apolipoprotein E (APOE).   Apolipoprotein E is a protein involved in the transport and metabolism of cholesterol and in several other metabolic functions. The E4 variant of the gene for APOE is associated with substantially increased risk of Alzheimer’s disease and also of heart disease and of increased rate of shortening of telomeres – the protective caps on the ends of chromosome that protect them from damage. Rapid shortening of telomeres is associated with decreased longevity.

We have two copies of each gene (apart from genes on the sex chromosomes), one copy inherited from each parent. In individuals in whom both copies of the APOE gene are the E4 variant, the risk of Alzheimer’s disease is around 15 times greater than in individuals who have two copies of the ‘neutral’ E3 variant, but fortunately very few individuals carry two copies of the E4 variant. However, almost 14% of the population carry one E4 variant. An individual with one copy of E4 together with a copy of E3 has a risk of Alzheimer’s that is about 3 times greater than that of a person with two copies of E3.   Similarly, carrying the unfavourable E4 variant of the gene for APOE does have an apprecibale effect on life expectancy, but even this ‘unfavourable’ gene accounts for only small amount of the variation in longevity in the population. It should also be noted that in contrast, the unfavourable E4 variant is associated with potentially beneficial higher levels of vitamin D which might explain why the gene has persisted in the population despite its unfavourable effects.

But the gene for APOE is the exception. Other genes that appear to contribute to variation in longevity in the population account for a much smaller proportion of the variation than the APOE gene. One other gene that warrants a passing acknowledgement is a gene with the whimsical name, FOXO3. It is a gene that plays a role in regulating gene transcription: the process by which the genetic code specified in our DNA is transcribed onto a temporary RNA copy in preparation for translation into the structure of the proteins that are the building blocks of our bodies. FOXO3 influences the process by which cells die naturally and also plays a role in defence against oxidative damage – a topic we shall return to later.   Its function suggests that FOXO3 is a candidate for an important role in determining longevity, but in fact its influence is not large enough to be discernible above the noise in the data obtained in large (‘genome wide’) studies of the association between genes and longevity.

Genetic variants with small effect

Most of the variants of the many genes that are associated with small alterations in longevity occur commonly in the population. Individually these genetic variants produce a slight perturbation of the structure or function of the body. The very fact that these variants are common demonstrates that individually they cannot have a devastating effect on structure or function, as variants with devastating effects are unlikely to get handed down through many generations.

At this stage it is worth pausing to look briefly at the nature of genetic variation and the mechanism by which it can affect the body’s structure or function. The genetic code is specified by the sequence of the molecular units that a strung together to form the double helical chains of DNA. There are only four of these elementary molecular units, which are assigned the labels A, T, G and C. (These labels are the first letters of the names of the purine and pyrimidine molecules that from part of these units.) DNA consists of a pair of intertwined chains, linked by the bonds that form between A and T or between G and C, at corresponding locations on the two chains Thus each element in the code is either an A-T pair or a G-C pair.  During the preparation for translation, the twinned DNA strands get copied as a single-stranded RNA molecule where each A,T,G, or C unit in one of the DNA chains is copied as a U,A,C or G.   Note that the elementary unit labelled as T (representing the pyrimidine, thymine) in DNA has been replaced by a slightly different molecular unit labelled U (representing the pyrimidine, uracil) in RNA. The crucial thing is that each possible triplet of three sequential units in an RNA chain is the code for a particular amino acid. Amino acids are the basic units that are assembled to form proteins. Proteins are the basic building blocks of the body, serving many specific purposes. Many are enzymes that catalyse the various metabolic processes in the body. Others, such as collagen, are structural elements. The contractile proteins, actin and myosin, enable muscles to do work.

The mechanism by which the genetic code gets transcribed and translated into protein is known as gene expression.  It is gene expression that shapes the structure and function of the body.  As we shall discuss later, many things can influence gene expression.


Figure 1: schematic illustration of gene expression. An extra-cellular signalling molecule (eg an inflammatory cytokine) binds to a specific receptor embedded in the membrane of the cell , triggering a cascade of signalling within the cell. This cascade involves messenger molecules such as cAMP and various effector proteins, including kinase enzymes which activate other proteins by attaching a phosphate group (‘phosphorylation’). When the CREB protein is activated it initiates transcription of DNA, producing an RNA molecule in which the order of the A, U, C & G units is the code for a specific protein. Each triplet of A, U, C & G units represents a particular amino acid. The code specified by the RNA template is translated into the sequence of amino acids that are assembled to make the specified protein. The process of assembling the protein is performed by a molecular construction device called a ribosome.



During the rough and tumble process of cell duplication that occurs regularly in living tissues, one letter of the code might get changed (‘mutated’). This is known as a point mutation, and the resulting variation is known as a Single Nucleotide Polymorphism (SNP). The mutation might be triggered by irradiation by radioactive materials, chemical assault by disruptive chemicals in the environment or the diet, or merely by random jiggling of units making up DNA as it is duplicated during cell division.   As a result of the change in one of the letters, a particular triplet in the code is likely to specify a different amino acid. When the mutant DNA is transcribed into RNA and subsequently translated into a protein, one amino acid will be replaced by another. Just as when a particular footballer is substituted during a football match, the substitution might have a dramatic effect, for better or worse, or alternatively, the team might continue to function with little overall change in effectiveness, in the case of amino acid substitution, there might be either a dramatic change in function of the protein if the substituted amino acid plays a cardinal role or merely a slight change in effectiveness of the protein. Because the sequence of amino acids in proteins has been shaped though many generations, most proteins in the body are well honed to fit their particular role. In the absence of major environmental change, mutations that substantially enhance the fitness of the body for its survival are extremely rare.   On the other hand, mutations that result in serious disruption of the function of the protein diminish fitness for survival, and therefore disappear from the population. The mutations that survive to become common in the population usually have only small effects on the function of the specified protein. In addition to the point mutations that generate SNPs, other types of variation are possible, but these are beyond the scope of this discussion

In summary, the commonly occurring variations in the genes that code for particular proteins usually have only minor effects on the function of those proteins. These functional effects might be helpful or helpful depending on circumstance. But the crucial thing is that it is likely that in most instances various other circumstances including life-style factors might over-ride the relatively minor effect of a specific genetic variant on the structure or function of the body. For most of us, our fate is not pre-ordained by these genes.

One might expect to find that that among the minority of exceptional individuals who live to a great age, the co-existence of many favourable genes each contributing a little, might make an appreciable contribution to their extraordinary longevity. While twin studies demonstrate that the genes contribute only about 25% to the probability of survival to age 85, studies of extremely elderly individuals, such as the study of 801 centenarians (with median age at death of 104 years) by Sebastian and colleagues,, demonstrate that genes play a substantially greater role in the longevity of these exceptional individuals. Similarly, for individuals who exhibit extraordinary longevity as athletes, it is probable that the co-existence of many favourable genes plays an appreciable role. When a large number of small nudges all push in the same direction, their combined effect is appreciable. However for the majority of us, who carry a mixed selection of mildly favourable and unfavourable genetic variants, it is plausible that if we could adopt a range of life-style choices (including appropriate training) that tend to enhance longevity in a consistent manner, we could engineer our fate in a way that swamps the potpourri of random minor influences arising from our genetic endowment.

Gene expression does matter

While the selection of minor genetic variants we happen to have been born with plays only a small part in life-expectancy for most of us, the manner in which our genes are expressed nonetheless plays a crucial role in determining how long we live and how well we continue to function in old age. Unlike an inanimate machine, such as a bicycle with parts that become abraded or degraded by friction and/or corrosion as it grows old, living creatures have inbuilt mechanisms for repair and for correcting internal imbalances that threaten their well-being. A bicycle eventually ceases to function because the abrasion or degradation causes a component to break or jam unless maintained and repaired by an external agency.   However, when a human is subject to wear and tear an elaborate self-repair mechanism is mobilised. The occurrence of damage triggers the release of signalling molecules, which travel via the blood stream to remote regions of the body to mobilise defences. The signalling molecules bind to specific receptors on the surface of the target cell, initiating a series of steps leading to the transcription and translation of DNA to produce proteins that replace or augment the existing proteins as required to repair or even enhance the functions of the body.

After the arrival at the cell surface of a signalling molecule indicating the need for repair or some other response to the external environment, the next step is a cascade of internal signalling within the cell which initiates the transcription of the DNA code onto a temporary RNA template (as discussed above in the review of the process by which the genetic code is expressed, and illustrated in figure 1).

There is a unique RNA template for each protein that is to be constructed. Therefore at any time, the profile of RNA in the cells of a particular tissue indicates which particular proteins are under construction at that time.   The RNA profile of a particular tissue at a particular time is in effect a snap-shot of the multiple building, repair and maintenance processes underway in that tissue at that time.

Environmental factors including life-style and training work in synergy with genes to maintain the body in good working order.   The expression of genes in muscle is not only of particular importance for athletes whose activities are depend on well-functioning muscles, but growing evidence indicates that the expression of genes in muscles is a marker for heathy aging throughout the body. Recent studies indicate that the RNA profile of muscle in late middle age might be a good predictor of the fitness not only of muscle but of other body tissues in subsequent decades.

For example, Sood and colleagues from Kings College, London, demonstrated that a particular RNA profile initially identified in muscle biopsies from a small sample of healthy individuals at age 65, could be used to predict subsequent health of kidneys and brain in several independent samples of elderly people. In one sample followed for 20 years, this profile proved to be a significant predictor of overall survival. Sood proposes that this RNA profile, initially identified in muscle, is a robust marker of healthy aging.

The finding that the state of gene expression in muscle at age 65 can be a good predictor of subsequent overall health is consistent with the observation that self-selected walking speed in late middle age is a strong predictor of survival, and is perhaps of special interest to dedicated runners, though we should not read too much into the fact that the investigators chose to examine muscle tissue. At this stage many question remain unanswered. Two key questions are: what does the set of proteins that are specified by the RNA profile identified by Sood tell us about the molecular processes that characterise healthy aging; and what are the factors that determine the RNA profile in muscle in middle age.

Examination of the list of proteins specified by the identified RNA profile provides few strong clues regarding the molecular processes that characterise healthy aging. Some of the proteins have a known role in cell survival. Perhaps disappointingly for anyone dedicated to running, none of the proteins are those known to be produced in response to vigorous exercise.   However, I am not greatly surprised by this. Although the sample of individual in who the RNA profile was initially identified were healthy and active, none were athletes.   Nonetheless even as a dedicated runner, I find it intriguing that there are features in the current internal state of muscle fibres, other than (or perhaps in addition to) the recognised consequences of vigorous exercise, that indicate current good health and predict future well-being. Vigorous exercise is not all that matters.

Furthermore, the identified RNA profile does not include diminished amounts of the RNA associated with the known risks for diabetes and cardiovascular disease, suggesting that there are aspects s of healthy aging that are not specifically associated with low risk of heart disease. This implies that current guidelines for a healthy lifestyle, which focus largely on known factors associated with cardiovascular health, might fail to include other important aspects of healthy aging.

Perhaps the most important practical issue is whether we can do anything to promote the development of a healthy RNA profile. In general, RNA profile is determined by a combination of genetic and environmental factors.   The fact that genes themselves are not a strong predictor of longevity (except in the small group of exceptional individuals who reach extreme old age) makes it plausible that environmental factors play a large role in promoting the identified healthy RNA profile in middle age. At this stage there is little reason to propose that these factors are uniquely related to muscle. It is possible that influences from elsewhere in the body, such as neural regulation by the brain, the action of hormones or effects produced by other signalling molecules circulating in the blood stream, might shape the RNA profile in muscle.

It is likely that the challenge of remaining a healthy athlete into old age is a ‘whole body’ challenge, and I therefore look forward to future studies that might indicate what can be done to promote the development of a healthy RNA profile in muscle in middle age, irrespective of whether the direct site of action is in muscle or elsewhere in the body.

What can we do now?

Studies such as that of Sood and colleagues provide a fascinating pointer towards future investigations that might enable us to improve our chances of aging in a healthy manner, but it is reasonable to ask what guidance science provides now. In fact there is a substantial body of existing evidence about the mechanisms of cellular repair, protection and maintenance that allows us to make intelligent guesses about what might be helpful.

At the heart of this self-repair mechanism is the process of inflammation. This mechanism is not only responsible for repair of overt damage, but is also the mechanism by which training makes an athlete stronger and fitter. But the mechanism for self-repair does not confer immortality for two reasons. First, inflammation itself can leave a trail of debris in the tissues of the body. The debris is at least partially removed by the crucial scavenging process known as autophagy, but ultimately the residual junk gums up the works. Secondly, it appears that there is a limit to the number of times that the cells of the body can divide to generate new cells to replace those that are worn out. The gradual shortening of the protective telomeres on the ends of chromosomes is a crucial factor in limiting the number of times that cells can divide.

Cardinal among the processes that regulate the maintenance of living tissues are processes mediated by hormones. In particular achieving a balance between catabolic hormones that promote the break-down of tissues, including the process of autophagy, and anabolic hormones that promote the building of tissues, is crucial.

Finally, in light of the fact that gene expression matters throughout life, the cellular mechanism for protecting and repairing DNA itself, are likely to play an important role in life-expectancy and in our longevity as runners.


Figure 2: Schematic illustration of the mechanisms involved in cellular repair. These mechanisms are central to the response to training and also to the responses various other types of cellular damage that are crucial for healthy aging.

Although there is still much to learn about all of these processes, there are things that we can do now that help harness inflammation constructively, achieve a good balance between catabolism and anabolism and perhaps even promote the protection and repair of DNA. But this post has already grown long. I will address these issues in greater detail in future posts


3 Responses to “The longevity of the long distance runner, part 2: the basic science.”

  1. Ewen Says:

    Thanks Canute. I’m interested in living a healthy life into old age and hope the combination of activity/healthy eating has some bearing on longevity, but I guess genes are what matters most. I’m lucky to have long-lived relatives, so I guess that’s a good sign.
    You mention the balance of catabolic and anabolic hormones — how much does training/lifestyle matter to this balance?

    On another point to do with genetics, I’ve been reading Charles Wegelius’s book (a former professional cyclist) and he mentions his naturally high hematocrit level, which at one point caused him to fail the arbitrary 50% health limit that was the rule for racing. I think Paula Radcliffe also has a high hematocrit level. I was wondering if hematocrit is genetically determined and is it part of the so-called “natural talent” of some distance runners? Are there any studies of elite distance runners that show some with lower than average hematocrit doing well, in a similar way that some with lower VO2max have done well?

  2. canute1 Says:

    Thanks for your comment. It is certainly reassuring to have long lived relatives. Genes play a part, in both athletic performance and in longevity, though as I outlined above, I think that genes usually only have a crucial role to play in exceptional athletic performance or exceptional longevity. For most of us, who aspire to a pleasingly good level of performance, and/or to healthy longevity for a decade or so beyond our Biblical allocation of three score and ten years without coveting the longevity of Methuselah, genes only determine the starting point and there is a lot we can do to shape the eventual outcome.

    Haematocrit is certainly important for endurance athletes, and can be influenced by genes. There are genetically-determined forms of polycythaemia (excess production of red cells). For example, Finnish skier, Eero Mäntyranta, who won seven Olympic medals, had familial polycythaemia arising from a mutation of the EPO receptor. While a mild degree of polycythaemia is beneficial, an excess of red blood cells results in sludgy blood and a high risk of stroke. But an excess of red cells can be produced by non-genetic means – infamously by doping with EPO as in the case of Lily Shobukova , or by legal means such as altitude training and/or sleeping in an reduced-oxygen tent, as Paula Radcliffe did.

    I do not know of any explicit evidence that athletes with low haematocrit have performed exceptionally well. However, ‘athlete’s anaemia’ is common. This is characterised by decreased haematocrit due to increased blood volume. The total red cell mass in the body is not decreased. The greater blood volume produces stronger heart contraction on account of greater filling of the heart during diastole, thereby stretching the heart muscle, producing more powerful contraction. There are very many factors that influence the amount of oxygen delivered to the muscles. In general you need many of the relevant variables to be near optimal to become a world champion, but it is possible to compensate for a deficit in any one variable. I would be surprised if there were not at least some high-performing athletes with low haematocrit.

  3. Ewen Says:

    Thanks Canute, that’s interesting. I hadn’t heard of Mäntyranta. I guess the first runner to break 2 hours for the marathon (if it ever happens) will have a perfect allocation of all the variables needed for ultimate performance. As well as racing on the perfect course, with perfect pacing, weather etc.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: