Predictive adaptive responses and human evolution

https://doi.org/10.1016/j.tree.2005.08.001Get rights and content

The importance of a single genotype being able to produce different phenotypes in different environments (phenotypic plasticity) is widely recognized in evolutionary theory and its adaptive significance is clear. In most cases, the developing organism responds to an environmental cue by producing a selectively and immediately appropriate phenotype. One subset of phenotypic responses to environmental stimuli, however, does not necessarily provide an immediate selective advantage. Rather, these kinds of responses, which we call ‘predictive adaptive responses’ (PARs), act primarily to improve fitness at a later stage of development. We argue that PARs have had an important role in human evolution, and that their recognition and interpretation has major significance for public health.

Introduction

Clinical and epidemiological research has identified numerous links between measures of early human development and the incidence of adult disease [1]. Most work has linked an impaired fetal environment, as reflected by small size at birth, to a greater risk of coronary heart disease and non-insulin dependent diabetes in middle-aged and elderly people 2, 3, 4. This relationship has been termed the ‘fetal (or developmental) origins of adult disease’ [2] because it suggests that events in early development irreversibly alter components of homeostasis in such a way that, when amplified or challenged by postnatal environmental factors, disease becomes manifest. Hales and Barker 1, 5 proposed this phenomenon to be the accidental effect of what they termed the ‘thrifty phenotype hypothesis’, an immediate fetal adaptation to altered nutrient supply that is mediated in part by insulin deficiency and/or resistance (insulin being a major fetal growth promoting hormone) for survival in a deficient intrauterine environment. These changes leave the growth-retarded fetus to cope with the consequences, which depend on whether the postnatal environment is nutritionally rich or poor. Experimental studies have readily replicated a relationship between early life experience and adult metabolic and cardiovascular function in a variety of mammal species [6]: the breadth of this demonstration suggests that some general biological process underpins it. Hales and Barker 1, 5 recognized that there might be postnatal advantage for a ‘thrifty’ phenotype in a nutritionally deprived environment. We 6, 7, 8 and others [9] have proposed that these observations can be better interpreted as part of a broader set of developmental and evolutionary strategies that we have termed ‘predictive adaptive responses’ (PARs) 6, 8. Placing these strategies in this perspective has important implications for understanding changing patterns of human disease.

Section snippets

Predictive adaptive responses

We define PARs as a form of developmental plasticity that evolved as adaptive responses to environmental cues acting early in the life cycle, but where the advantage of the induced phenotype is primarily manifest in a later phase of the life cycle. The cue affects the processes of developmental plasticity and thus induces changes in the developmental trajectory of form and function such that the organism presets its physiology in expectation of that physiology matching its future environment.

Developmental responses

When considering effects of the environment on development, it is important to distinguish between responses that might be adaptive later in life from those that (i) merely disrupt development in a pathological manner (e.g. those causing fetal malformation) leaving the organism to ‘cope’ with the consequences [28]; and (ii) involve immediate adaptation for fetal survival (e.g. fetal growth retardation) having an incidental advantage later. Some responses to environmental stress could be

A general model of PARs

The presence of many similar PARs in diverse groups of organisms suggests that the capacity to induce PARs is adaptive [8]. This assumption implies that there has been an advantage in retaining and refining PARs because fetal predictive responses have generally been appropriate for the postnatal environment, and this choice has conferred a selective advantage on the bearers. In our view, it is the ability to mount a PAR itself that constitutes an adaptation, not only the phenotypes that it

Maternal constraint, normal variation in fetal growth and PARs

Variation in postnatal growth has a major genetic component, whereas fetal growth is more environmentally sensitive. This difference is revealed, for example, in the far greater correlation between siblings in adult height than in birth size [41]. The environment of the fetus is created by its mother and, once developed, by the placenta. Maternal factors must dominate in determining fetal growth because the mammalian fetus will not survive if it outgrows the pelvic canal of its mother. Thus, it

Conclusion and future directions

PARs are a form of phenotypic plasticity with delayed selective benefits seen in many species. We argue that they have been retained in humans because they conferred survival advantage in the poorer nutritional and high-energy expenditure environment of our ancestral hominids. PARs are induced by environmental cues in development, utilizing the normal processes of maternal constraint as part of the mechanism of providing environmental cues to the fetus. In our evolutionary past, it was

Acknowledgements

We thank Patrick Bateson for many discussions. Comments from the referees also led to a significant improvement in the article. M.A.H. is supported by the British Heart Foundation.

References (60)

  • V. Lummaa et al.

    Early development, survival and reproduction in humans

    Trends Ecol. Evol.

    (2002)
  • A.C. Ravelli

    Glucose tolerance in adults after prenatal exposure to famine

    Lancet

    (1998)
  • S.E. Ozanne et al.

    Early programming of glucose-insulin metabolism

    Trends Endocrinol. Metab.

    (2002)
  • C.N. Hales et al.

    The thrifty phenotype hypothesis

    Br. Med. Bull.

    (2001)
  • D.J.P. Barker

    Mothers, Babies and Health in Later Life

    (1998)
  • C.A. Newsome

    Is birthweight related to later glucose and insulin metabolism? Systematic review

    Diabet. Med.

    (2003)
  • C.N. Hales et al.

    Type 2 (non-insulin dependent) diabetes mellitus: the thrifty phenotype hypothesis

    Diabetologia

    (1992)
  • P.D. Gluckman et al.

    Living with the past: evolution, development, and patterns of disease

    Science

    (2004)
  • P.D. Gluckman et al.

    The Fetal Matrix; Evolution, Development and Disease

    (2004)
  • P. Bateson

    Fetal experience and good adult design

    Int. J. Epidemiol.

    (2001)
  • M.J. West-Eberhard

    Developmental Plasticity and Evolution

    (2003)
  • M. Pigliucci

    Phenotypic Plasticity: Beyond Nature and Nurture

    (2001)
  • N.A. Moran

    The evolutionary maintenance of alternative phenotypes

    Am. Nat.

    (1992)
  • S.E. Sultan et al.

    Metapopulation structure favors plasticity over local adaptation

    Am. Nat.

    (2002)
  • S.W. Applebaum et al.

    Density-dependent physiological phase in insects

    Annu. Rev. Entomol.

    (1999)
  • S.J. Simpson

    A behavioural analysis of phase change in the desert locus

    Biol. Rev.

    (1999)
  • S.J. Simpson

    Gregarious behavior in desert locus is evoked by touching their back legs

    Proc. Natl. Acad. Sci. U. S. A.

    (2001)
  • T.M. Lee et al.

    Vole infant development is influenced perinatally by maternal photoperiodic history

    Am. J. Physiol.

    (1988)
  • A. Kawahata et al.

    Some observations on sweating of the Aino

    Jpn. J. Physiol.

    (1951)
  • M.J. Meaney

    Maternal care, gene expression, and the transmission of individual differences in stress reactivity across generations

    Annu. Rev. Neurosci.

    (2001)
  • Cited by (0)

    View full text