Article Text


Is lifespan determined in utero?
  1. A Aihie Sayer,
  2. C Cooper,
  3. D J P Barker
  1. MRC Environmental Epidemiology Unit, University of Southampton, Southampton General Hospital, Southampton SO16 6YD
  1. Dr Aran Aihie Sayer. Email: aas{at}

Statistics from

Low rates of growth in early life are now known to be associated with an increase in age related disease in later life.1 This is thought to reflect programming, the process whereby a stimulus or insult acting at a critical period of development in early life, has lasting or lifelong importance.2 In animal experiments that have examined the mechanisms responsible for programming, attention has focused on prenatal undernutrition and its influence on gene expression, cell division and differentiation, and tissue structure.

Fetal growth and adult disease

Epidemiological studies have shown that markers of poor fetal growth, including low early weight, thinness, and shortness at birth, are associated with increased mortality and morbidity from cardiovascular disease in later life.1 3 Correlations have also been shown between poor early growth and the major cardiovascular risk factors—raised blood pressure,4raised plasma fibrinogen5 and serum cholesterol concentrations,6 impaired glucose tolerance7and reduced arterial compliance.8 These associations were first established by studies in Hertfordshire, where from 1911 to 1948, every newborn baby was weighed and followed up to the age of 1 year. The prevalence of non-insulin diabetes and impaired glucose tolerance, for example, falls threefold between men who weighed 5.5 pounds at birth and those who weighed 9.5 pounds.7 These associations have been replicated in several different countries including the United States9 and India10 as well as in Britain. The associations are independent of adult lifestyle, and are not limited to cardiovascular disease. Chronic obstructive pulmonary disease11 and reduced bone mineral content12 13 are both associated with reduced growth in utero and during infancy. The major determinant of fetal growth is nutrition14-16 and the fetal origins hypothesis proposes that fetal undernutrition programmes the long term adverse sequelae of small size at birth. Cardiovascular disease, impaired glucose tolerance, chronic obstructive lung disease and bone loss are age related disorders, but the link between fetal nutrition and aging has scarcely been explored.

Postweaning and diet restriction studies

Nutrition is known to affect aging. The prolongation of lifespan by postweaning diet restriction was first shown in rodents in 191717 18 and the results have been replicated in many species, most recently primates.19 Several dietary manipulations with varying reductions in calorie and protein intake have been found effective.20 One study considered racial differences in renal and hepatic aging between Japanese and Caucasian people. Increased changes were seen in the Japanese histological specimens. It was suggested that the relatively low protein diet eaten by the Japanese was responsible for retarded development and accelerated aging.20 There has been just one controlled diet restriction study in humans.21 Healthy subjects over 65 years of age, living in a religious institution for the aged, were given their usual diet containing 2300 calories on odd days and 1000 calories from milk and fruit on even days. Control subjects continued with the usual diet every day. The experimental subjects spent significantly less time in an infirmary than controls and six of them died compared with 13 controls over three years of observation, although this difference was not significant.

In addition to prolonging median and maximum lifespan, nutritional interventions in animals delay aging processes and disorders and maintain function. The mechanisms remain unknown.22 23Initial suggestions that the effects were mediated through reduced growth after weaning were reconsidered when diet restriction was shown to have similar effects even when initiated in adult life. Other proposed mechanisms include an alteration in metabolic rate, free radical production, and protein turnover, but there is little evidence to support any of these.

Preweaning undernutrition and aging

The extensive investigation of postweaning diet restriction has not been accompanied by an equivalent body of work determining the effect of the same intervention in fetal life and infancy.24 The few gerontological studies of this issue in animals suggest the opposite effect. An early study in mice showed that a change in diet shortly after birth, sufficient to slow growth, resulted in a reduced lifespan.25 Studies in rats showed that a reduction in maternal diet in pregnancy produced offspring with an earlier age related decline in haemoglobin26 and shorter lifespan.27 Reduced nutrition in prenatal and early postnatal life has resulted in increased serum concentrations of age related hepatic and renal enzymes,28 and in the 1970s it was first proposed that very early diet restriction might be associated with accelerated aging in later life.29 This notion was not widely accepted in gerontological research of the time, however.

Recent studies have investigated the effects of maternal undernutrition on a range of physiological variables in the offspring. Low protein in the diet of pregnant rats results in offspring with permanently raised blood pressure.30 It also reduces pancreatic β cell mass and profoundly reduces islet vascularisation.31 A similar diet given for only three weeks after weaning permanently impairs the insulin response to glucose.32 The evidence from animal studies is therefore that nutrition has profound effects on lifespan and age related structural and functional changes. Diet restriction seems to have the opposite effects when operating in either fetal and early postnatal life or in later life. Undernutrition before weaning is associated with more rapid aging and decreased lifespan, while diet restriction in adult life is beneficial.

Fetal growth and aging

Previous studies in Hertfordshire have established that reduced growth in utero and during infancy is associated with an increase in age related disease,7 but we know little about associations with normal aging processes. A recent study in North Hertfordshire addressed this.33 Markers of aging in a number of different body systems, including the eye, ear, muscle and skin, were measured in a sample of men and women aged 64 to 74 years. Lower weight at 1 year was associated with increased lens opacity, worse hearing, reduced grip strength and thinner skin. These correlations were independent of social class. This study provides preliminary evidence that undernutrition and reduced growth in utero and during infancy leads to more rapid aging in certain systems.

Theories of aging

Aging theories can be divided into two main groups according to whether aging is viewed as genetically predetermined or as a cumulative response to events over time. Evidence for the role of genes in aging comes from the existence of species specific lifespans,34limited heritability of lifespan35 and the human progeroid syndromes where a simple gene defect is associated with phenotypic changes similar to aging which occur much earlier in life.36 A further distinction can be made as to whether aging is considered to have evolved as a beneficial process in its own right (adaptive theories) or as a byproduct of other processes (non-adaptive theories). One of the oldest adaptive theories of aging suggests that evolutionary pressure for aging genes would come from aging contributing to the fitness of the species by removing reproductively inactive aged individuals. Such an argument is circular, however. If aging did not occur there would be no need to remove old individuals.

Non-adaptive theories propose that natural selection for the genes involved has occurred for reasons other than to cause aging. The disposable soma theory37 suggests that aging is a result of the accumulation of defects in macromolecules and that these occur because of limited capacity for somatic maintenance and repair. Mathematical modelling can be used to determine the optimal allocation of nutritional resources between reproduction and repair to maximise survival of the species. This will be influenced by the degree of environmental danger to which the species is exposed. For example, species experiencing high levels of environmental danger such as predation are better served by early and rapid reproduction. Optimal allocation defined in this way results in less investment in repair than that required for perfect maintenance, and aging is the result. This theory suggests that repair is central to aging.

There is now experimental evidence to support a role for DNA repair in aging. There is a correlation between lifespan and efficiency of overall genomic DNA repair systems.38 DNA repair is also less efficient in cells from aging individuals or senescent cultures.39 This research has led to the recent identification of the defective gene in Werner’s syndrome, a premature aging condition. The gene product resembles helicase which is required for DNA repair.40

Repair of other long lived molecules may also be important. It has been suggested that aging is a result of acquired damage to all long lived macromolecules. For example, accumulated cross links may cause structural stabilisation and altered actions in intercellular structural proteins and collagen as well as DNA.41-43

Programming of repair processes

Failure of repair at the tissue level could be programmed. The relations of early growth and aging markers in the lens, cochlea, muscle and skin found in Hertfordshire may have occurred because these are all organs contain large proportions of long lived molecules or cells. The lens contains crystallins which once formed are never turned over.44 They are synthesised in the outer cortex of the lens and gradually move to the centre as new crystallins are formed around the outside of the lens throughout life. The centre of an adult lens, therefore, contains molecules formed in utero. The cochlea has long lived collagen molecules41 and hair cells which are not replaced.45 46 Muscle and skin contain collagen and elastin which also have a very slow turnover.41 Tissues containing a high proportion of long lived molecules or cells are likely to require particularly good repair systems as turnover is limited or non-existent. We postulate that early undernutrition results in impaired development of molecular and cellular repair mechanisms which may affect all tissues, but which are most critical in later life in those containing a high proportion of long lived components.

There is now good evidence that fetal growth and aging are related. This may be due to nutritional programming of aging and a possible mechanism is the impaired development of repair systems. This work now needs to be taken forward. Molecular studies of DNA repair capacity are planned, using markers such as mutation rates of specific genes, telomere length and DNA strand breaks. This will allow the postulated association between early growth and DNA repair to be investigated. The effects of preweaning diet restriction on growth, aging, and lifespan can be explored more fully in further animal studies where experimental manipulation of early diet is possible. Finally, follow up of the North Hertfordshire cohort will allow the rate of aging and its relation to early growth to be determined.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.