Fetal programming: from gene to functional systems ‐ an overview

The Journal of Physiology supported a Symposium entitled ‘Fetal programming: from gene to functional systems’ during the Fetal Physiology Group Meeting at the 49th Meeting of the Society for Gynecologic Investigation in March 2002 in Los Angeles. The purpose of the Symposium was to address the concept that epigenetic factors in the intrauterine environment have a profound effect on the trajectory of prenatal development. Since 1989 when David Barker showed an inverse relationship between birth weight and mortality due to adult ischaemic heart disease, the role of prenatal programming as a determinant of adult diseases has become increasingly clear. Programming is now known to be an important underlying feature of many systemic adult diseases including coronary heart disease, hypertension, insulin resistance syndromes and osteoporosis. While epidemiological evidence for the role of early developmental growth patterns leading to specific adult disease outcomes has continued to strengthen, experimental evidence from animal models has brought new and compelling support for this important determinant of lifetime health. As discussed in this issue, controlled animal-based studies are beginning to provide insights into the molecular, cellular and systemic mechanisms that contribute to the several different manifestations of fetal programming. While it is clear that fetal programming is not simply the effect of genes passed on through generations, it is just as clear that the phenotype of offspring subjected to a variety of intrauterine challenges is determined by lifelong gene expression patterns set into motion during critical windows of development and that in some instances these phenotypes may pass on their effects to future generations. This issue of The Journal of Physiology contains five reviews by experts who were invited to speak at the Symposium and eight papers individually submitted for review by investigators working in this burgeoning area of integrative physiology. The Symposium was opened by Professor Nicholas Hales from Cambridge University who has provided an excellent and succinct summary of the background to the concept of the fetal origins of adult disease (Hales & Ozanne, 2003). He presents an overview of the metabolic syndrome, the thrifty phenotype and the role of oxidative damage and telomere repair in determining long term disease in rat models. Professor Hales addresses the critical issue of postnatal catch-up growth following adverse prenatal nutritional supply. There is now overwhelming evidence that attempts to restore overall body dimensions and organ weights may run counter to the best interests of the long term health of the individual organism. Van Assche and colleagues from the Catholic University of Leuven address the mechanisms whereby pancreatic development may be compromised (Holemans et al. 2003). As result, subsequent lifetime regulation of glucose and insulin metabolism may be disturbed and the individual will be predisposed to the development of type II diabetes. A critical and provocative part of this story is the capacity for the transgenerational passage of the consequences of the maternal diabetic state during pregnancy. Female offspring of rats with diabetes during pregnancy are themselves more prone to gestational diabetes. If these female offspring become diabetic during their own pregnancies, a transgenerational chain can be set up. Philip Shaul from SouthWestern Medical Center at Dallas discusses the interactions between lipids and the powerful vasodilatory nitric oxide synthase (NOS) system (Shaul, 2003). Oxidized low density lipoproteins disrupt NOS activity by displacing the enzyme from cell surface caveoli. High density lipoprotein reverses this displacement effect. As a result the interplay of lipids during development and alterations in the factors that control their expression may play a key role in atherogenesis. Shaul's review considers remarkable evidence on the early origins of atherosclerosis, previously considered to be a disease of much later onset. The review by Wolf Reik and colleagues addresses the regulation of placental supply and demand (Reik et al. 2003). Much of the early epidemiological and animal research on fetal programming focused on the effects of poor nutrition during pregnancy. While factors such as vascular determinants of placental uptake and transfer of nutrients have been extensively investigated over the past 30 years, especially in sheep and rodent models, genetic factors have been somewhat ignored. Reik rectifies this omission by considering the key role of imprinting of genes that alter placental function and fetal growth. Improvements in fetal diagnosis have been driven by high definition ultrasonography and a better understanding of genetics, embryology and fetal metabolism. Earlier diagnosis of many conditions such as diaphragmatic hernia and cardiac malformations is now possible. Left untreated these intrauterine abnormalities, while themselves not necessarily lethal, can alter development in ways that produce secondary morbidity that is harmful to long term health. Correction of the deformity in utero removes this potential for long term adverse sequelae. Professor Flake takes a broad view of fetal surgery to include the possibility of gene delivery to correct problems (Flake, 2003). The paper by Davis and colleagues evaluates the effects of experimentally induced fetal anaemia in the fetal sheep model (Davis et al. 2003). The findings that fetal anaemia alters coronary conductance and results in permanent remodelling of the coronary vascular tree is a compelling example of fetal programming. Following prenatal exposure to repeated elevations of maternal glucocorticoids in the final third of gestation, Molnar and colleagues (2003) demonstrate increased peripheral vascular sensitivity to both vasodilator and vasoconstrictor agonists in five month old sheep. These studies support the concept that fetal programming involves both primary alterations and secondary compensations. When Wallace and Hammond conducted their classic experiments in the 1930s crossing shire horses and Shetland ponies to demonstrate the maternal constraint to growth, they did not control for the genotype of mother and sire. Giussani and colleagues have elegantly avoided that practical and theoretical limitation by conducting embryo transfer between thoroughbred horses and ponies (Giussani et al. 2003). They confirm the general principles delineated by Hammond and demonstrate that postnatal cardiovascular function in yet another mammal, the equine species, is modified by the uterine environment experienced prenatally. Whether programming of female fetuses affects their own subsequent pregnancies, and hence potentially their own offspring is a fundamental question of major importance to our health. Torrens and colleagues demonstrate that when female offspring of under-nourished pregnant rats themselves become pregnant they demonstrate impaired vasodilator responses in resistance but not conduit arteries (Torrens et al. 2003). This finding enhances the possibility of transgenerational effects similar to those described in the reviews of Hales and Van Assche (Hales & Ozanne, 2003; Holemans et al. 2003). Limesand and Hay study the effects of altered glycaemia status in the chronically instrumented fetal sheep model (Limesand & Hay, 2003). They demonstrate that prolonged fetal hypoglycaemia programmes fetal pancreatic insulin secretion. They postulate that changes such as these may play a role in type II diabetes in later life. Using fetal glucose clamp techniques combined with embryo transfer in the pregnant sheep Wallace and colleagues investigated the effects of altered dietary intake on placental glucose transfer capacity (Wallace et al. 2003). They make the important observation that placental size is the major limitation to placental glucose transfer. Poore and Fowden present data relating postnatal pituitary adrenal function through the first year of life to birth weight of piglets (Poore & Fowden et al. 2003). Thinness at birth is associated with elevated adrenal responsiveness to hypoglycaemia at 12 months of age. This finding is consistent with the concept of prenatal programming of the set point of pituitary adrenal function. Fetal exposure to glucocorticoids at higher than physiological levels occurs during maternal stress, fetal compromise and iatrogenic glucocorticoid administration. Antonow-Schlorke and colleagues follow up their demonstrations of glucocorticoid-induced changes in fetal brain development in the sheep with parallel studies in the non-human primate (Antonow-Schlorke et al. 2003). They demonstrate decreases in fetal cerebral neuronal cytoskeletal microtubule associated proteins and the presynaptic marker protein synaptophysin following administration of synthetic glucocorticoids to the pregnant baboon. This study demonstrates that general principles of fetal effects demonstrated in other animal models can be extrapolated to primates. Once again, we thank The Journal of Physiology for supporting this Symposium and area of research with this issue of the journal. It brings together critical new findings across this field for a thorough examination of current knowledge and for a critical discussion of unexplored territory in the new world of programming. The invited reviews (published in the order they were given at the Symposium) and investigator-initiated primary papers collected together in this issue of The Journal of Physiology together cover a broad range of topics that describe the current state of the art and provide a launching pad for new and exciting forays into what is one of the most rapidly growing areas of biology that attempts to unite nature and nurture.