Developmental plasticity is the property of a given genotype to produce different phenotypes in response to the environmental conditions experienced during development. Chordates have two basic modes of development, direct and indirect. Direct development (mode of humans) was derived evolutionarily from indirect development (mode of many amphibians), the major difference being the presence of a larval stage with indirect development; larvae undergo metamorphosis to the juvenile adult. In amphibians, environmental conditions experienced during the larval stage can lead to extreme plasticity in behaviour, morphology and the timing of metamorphosis and can cause variation in adult phenotypic expression (carry-over effects, or developmental programming). Hormones of the neuroendocrine stress axis play pivotal roles in mediating environmental effects on animal development. Stress hormones, produced in response to a deteriorating larval habitat, accelerate amphibian metamorphosis; in mammals, stress hormones hasten the onset of parturition and play an important role in pre-term birth caused by intra-uterine stress. While stress hormones can promote survival in a deteriorating larval or intra-uterine habitat, costs may be incurred, such as reduced growth and size at metamorphosis or birth. Furthermore, exposure to elevated stress hormones during the tadpole or foetal stage can cause permanent neurological changes, leading to altered physiology and behaviour later in life. The actions of stress hormones in animal development are evolutionarily conserved, and therefore amphibians can serve as important model organisms for research on the mechanisms of developmental plasticity.

Abundant epidemiologic evidence from the developed world now shows that more rapid weight gain during the first half of infancy predicts later obesity and cardio-metabolic risk. In countries in transition, in which stunting is still prevalent, distinguishing the effects of gain in weight from linear growth remains a challenge. Moreover, very few studies to date have incorporated body composition measures during infancy, which is key to understanding determinants of infant weight gain that also predict later obesity. In addition to infant feeding type, potential determinants include the perinatal endocrine milieu. Animal and emerging human data raise the possibility that ensuring adequate leptin exposure to the growing fetus may regulate energy balance as the infant grows. Understanding these pathways, as well as examining the balance between cardiovascular and cognitive effects in both term and preterm infants, will point the way toward effective interventions to alter infant growth to prevent later obesity.

The associations between fetal environment and diabetes risk are likely mediated by the offspring’s diminished pancreatic β-cell function or reduced insulin sensitivity. Our ability to distinguish between these mechanisms is impeded by the lack of markers describing an individual’s gestational environment. Fingerprints, however, are permanently fixed in the first half of gestation, and increased values of a dermatoglyphic marker that contrasts fingerprint ridge counts between the thumbs and fifth fingers (Md15) have been linked to type 2 diabetes. Among 763 adults without known diabetes from the Dutch Hunger Winter Families Study, we tested the associations of Md15 with homeostatic assessment indices of β-cell function (HOMA-b) and insulin sensitivity (QUICKI). For either outcome index, linear regression models included terms for Md15 tertiles and for maternal history of diabetes as reported by each participant. All models were corrected for sibling pairs and adjusted for age, sex and famine exposures. Increased Md15 was associated with decreased HOMA-b (P = 0.03 for Md15 tertile 3 v. 1) but not with QUICKI. In contrast, maternal history of diabetes was associated with decreased QUICKI (P < 0.001) but not with HOMA-b. Birth weight (available for 520 participants) was positively associated with increased QUICKI (P = 0.04 for birth weight tertile 3 v. 1) but not with HOMA-b. Fingerprint Md15, maternal history of diabetes and birth weight may help to identify specific defects in the control of adult glucose metabolism. Research into the environment associated with Md15 variation may suggest prenatal strategies for optimizing β-cell function in adult life.

The biological mechanisms in the association between maternal body mass index (BMI) and birth weight are not well understood, but are likely to involve maternal plasma glucose levels and nutrient transport across the placenta, both important modulators of fetal growth. Adipose tissue contributes to circulating levels of interleukins that may affect glucose metabolism and possibly also placental transport of nutrients. We investigated possible mediating roles of Interleukin 6 (IL-6) and Interleukin 1 Receptor antagonist (IL-1Ra) in 208 pregnant women. Known and hypothesized dependencies between BMI in early pregnancy and fasting glucose, IL-1Ra and IL-6 at gestational weeks 30–32, and birth weight were specified in a path diagram. Standardized regression coefficients, expressing direct, indirect and total effects, were estimated by Bayesian path analysis. Mean (s.d.) BMI was 24.9 kg/m2 (4.2) and mean (s.d.) birth weight 3748 g (454). The total effect of BMI on birth weight was 0.24 (95% credibility interval (CrI) [0.12, 0.36]). The direct effect of IL-1Ra on birth weight was not statistically significant, but significant effects of BMI on IL-1Ra (0.61, 95% CrI [0.51, 0.72]), of IL-1Ra on fasting glucose (0.17, 95% CrI [0.01, 0.34]) and of fasting glucose on birth weight (0.14, 95% CrI [0.01, 0.27]) implied an indirect pathway from BMI via IL-1Ra on birth weight. Approximately 20% of the effect of BMI on birth weight was mediated through IL-1Ra. For IL-6, no such effects were found. Our results indicate that IL-1Ra may be a mediator in the association between BMI and birth weight.

Low birth weight is related to increased risk of developing cardiovascular disease and type 2 diabetes in adult life. Since obesity is closely associated with type 2 diabetes and cardiovascular disease, the relationship between size at birth and adult anthropometry is of interest as a mediator of the relationship between birth weight and metabolic diseases. The aim of this study was, therefore, to examine the effect of size at birth and prematurity on measures of adult anthropometry taking adult socio-economic status and lifestyle variables into account. Midwife records with information on mother’s age and parity as well as weight, length and maturity at birth were traced in 4744 Danes born between 1939 and 1970. Measures of adult anthropometry (weight, height, body mass index (BMI), waist circumference, hip circumference and waist/hip ratio) had previously been recorded together with information on socio-economic factors, lifestyle and parental diabetes status. Mother’s age, parity and diabetes status were associated with offspring birth weight. Size at birth was positively associated with adult height and weight, but only weakly associated with BMI and not associated with waist/hip ratio when adjusted for socio-economic and lifestyle factors. Infants born preterm were less growth restricted at birth and grew to be taller and heavier compared with term infants born small for gestational age. Altogether, this study does not find evidence that obesity or a central fat distribution is mediating the relationship between low birth weight and risk of cardiovascular disease or type 2 diabetes in later life.

Foetal development may permanently affect muscle function. Indian newborns have a low mean birthweight, predominantly due to low lean tissue and muscle mass. We aimed to examine the relationship of birthweight, and arm muscle area (AMA) at birth and post-natal growth to handgrip strength in Indian children. Grip strength was measured in 574 children aged 9 years, who had detailed anthropometry at birth and every 6–12 months post-natally. Mean (standard deviation (s.d.)) birthweight was 2863 (446) g. At 9 years, the children were short (mean height s.d. −0.6) and light (mean weight s.d. −1.1) compared with the World Health Organization growth reference. Mean (s.d.) grip strength was 12.7 (2.2) kg (boys) and 11.0 (2.0) kg (girls). Weight, length and AMA at birth, but not skinfold measurements at birth, were positively related to 9-year grip strength (β = 0.40 kg/s.d. increase in birthweight, P < 0.001; and β = 0.41 kg/s.d. increase in AMA, P < 0.001). Grip strength was positively related to 9-year height, body mass index and AMA and to gains in these measurements from birth to 2 years, 2–5 years and 5–9 years (P < 0.001 for all). The associations between birth size and grip strength were attenuated but remained statistically significant for AMA after adjusting for 9-year size. We conclude that larger overall size and muscle mass at birth are associated with greater muscle strength in childhood, and that this is mediated mainly through greater post-natal size. Poorer muscle development in utero is associated with reduced childhood muscle strength.

Rates of weight gain in infancy and early childhood can influence later neurocognitive, metabolic and cardiovascular health. We studied the relationship of weight gain during infancy and early childhood to intelligence quotient (IQ), blood pressure (BP) and body mass index (BMI) at age 9 in children born with very low birth weight (VLBW). Sixty-five children born prematurely with VLBW were followed longitudinally and at 9 years IQ, BP and BMI were measured. The mean weight z-scores at birth, neonatal intensive care discharge, 1 year corrected for prematurity, 5 and 9 years were −0.17, −2.09, −1.3, −0.68 and 0.06, respectively. Weight gain during infancy (discharge to 1 year corrected for prematurity) and early childhood (1 year corrected age to 5 years) was expressed as rate of change in weight, rate of change in weight z-score and interval change in weight z-score. In multiple regression analyses that adjusted for race, gender, maternal education, antenatal steroids, birth weight z-score, major intracranial lesions on ultrasound and chronic lung disease, rates of weight gain in infancy and early childhood were predictive of BMI z-score at 9 years, regression coefficients (95% confidence intervals); 0.19 (0.02, 0.36) and 0.37 (0.11, 0.63), respectively, expressed as change in BMI z-score per 10 g/week weight increase. Rates of weight gain were not predictive of systolic BP z-score, Verbal IQ or Performance IQ. In VLBW infants, more rapid weight gain during infancy, and especially early childhood, is associated with higher BMI at school age.

The aim of this study was to investigate the assumed positive association of pre-existent and pregnancy-induced hypertension with the offspring’s weight and length gain in the first 14 months of life. We studied 3994 pregnant women and their offspring in a prospective community-based cohort study, starting between 2003 and 2004 (Amsterdam Born Children and their Development, ABCD study). Questionnaires obtaining information about hypertension during pregnancy were completed, and this was complemented with additional information from the obstetric caregiver. Anthropometry of the offspring was followed during the first 14 months of life. Main outcome measures were presence or absence of growth acceleration in weight or length (normal: ΔSDS ⩽ 0.67 v. growth acceleration: ΔSDS > 0.67). The relation between hypertension during pregnancy and weight and length gain was addressed by logistic regression analyses. We found that pre-existent hypertension was related to growth acceleration in weight and length. After correction for birth weight and pregnancy duration, the effect remained significant for growth acceleration in weight (OR 1.89; 95% CI 1.21–2.97; P < 0.01). Pregnancy-induced hypertension showed similar results, although correction for birth weight and pregnancy duration rendered the associations non-significant. In conclusion, infants of women with pre-existent hypertension during pregnancy more frequently have growth acceleration in weight and length, and yet the mechanisms acting on postnatal growth appear to be different.