In the sheep fetus, pulmonary and renal concentrations of angiotensin-converting enzyme (ACE) increase towards term in parallel with the prepartum surges in plasma cortisol and tri-iodothyronine (T(3)). The ontogenic change in pulmonary ACE has been shown to be induced, at least in part, by cortisol but the role of the thyroid hormones is unknown. Therefore, this study investigated the effects of thyroid hormones on tissue ACE concentration in fetal sheep during late gestation. Pulmonary and renal ACE concentrations were measured in sheep fetuses after experimental manipulation of thyroid hormone status by fetal thyroidectomy and exogenous hormone infusion. In intact fetuses, pulmonary and renal ACE concentrations increased between 127-132 and 142-145 days of gestation (term 145 +/- 2 days), coincident with the prepartum rises in plasma cortisol and T(3). The ontogenic increment in pulmonary ACE concentration was abolished when the prepartum surge in T(3), but not cortisol, was prevented by fetal thyroidectomy. At 143-145 days, ACE concentration in the lungs and kidneys of the thyroidectomised fetuses were both lower than those in the intact fetuses. In intact fetuses at 127-132 days, pulmonary ACE was upregulated by intravenous infusions of either cortisol (2-3 mg/kg per day) or T(3) (8-12 microg/kg per day) for 5 days. Renal ACE was unaffected by cortisol or T(3) infusion. Therefore, thyroid hormones have an important role in the developmental control of pulmonary and renal ACE concentration in the sheep fetus towards term. In addition, the prepartum rise in plasma T(3) appears to mediate, in part, the maturational effect of cortisol on pulmonary ACE concentration.
AL Fowden, J Mapstone and AJ Forhead
The effects of thyroid hormone deficiency in utero on the fetal glucogenic capacity were investigated by measuring glucose production and hepatic levels of glycogen and gluconeogenic enzymes in normal sheep fetuses in the fed and fasted states during late gestation and in those made thyroid hormone deficient by fetal thyroidectomy (TX). In the fed state, fetal TX had no effect on glucose uptake, utilisation or production by the fetus. It also had no apparent effect on the glycogen content or activities of the key gluconeogenic enzymes in the fetal liver. In addition, fetal plasma concentrations of insulin, cortisol, adrenaline or noradrenaline were unaffected by fetal TX in the fed state. In contrast, the rates of fetal O(2) consumption and CO(2) production per kilogram fetal bodyweight were significantly lower in TX than in intact fetuses in the fed state (P<0.05). TX prevented fetal glucose production in response to maternal fasting for 48 h. It also abolished the normal decreases in the fetal glucose carbon oxidation fraction, the rate of CO(2) production from glucose carbon and in the fraction of the umbilical O(2) uptake used for glucose carbon oxidation that occur during fasting in intact fetuses. At the end of the fast, plasma noradrenaline concentrations and hepatic levels of glycogen, glucose 6-phosphatase, fructose diphosphatase and alanine aminotransferase were significantly lower in TX than in intact fetuses. These observations show that thyroid hormones are essential for glucogenesis in the sheep fetus during late gestation and suggest that these hormones act both on the hepatic glucogenic pathways and on the mechanisms activating glucogenesis in utero.
NB Holdstock, VL Allen, MR Bloomfield, CN Hales and AL Fowden
At birth, the endocrine pancreas must assume a glucoregulatory role if the neonate is to survive the transition from parenteral to enteral nutrition. In species like the horse, neonatal hypoglycaemia is common, which suggests that the glucoregulatory mechanisms are not always fully competent at birth. Hence, this study examined pancreatic beta cell function in newborn foals during nutritional adaptation over the first 10 days post partum. Over a 48 h period at three time intervals after birth (days 1-2, 5-6 and 9-10 post partum), the beta cell responses to suckling and to intravenous administration of glucose, arginine and saline were measured in seven normal pony foals. Basal plasma concentrations of proinsulin, but not insulin or glucose, increased significantly between days 1 and 10. Suckling caused a gradual increase in plasma glucose, which was accompanied by a significant increase in plasma insulin concentrations 15 min after the onset of suckling on days 5 and 9, but not day 1. There was no significant change in plasma proinsulin concentrations in response to suckling at any age. At all ages studied, glucose and arginine administration stimulated an increase in the plasma concentrations of insulin and proinsulin; these beta cell responses did not change significantly with postnatal age. The insulin responses to glucose were significantly greater than those of arginine at each time period. Glucose clearance was significantly slower on day 1 than subsequently. Proinsulin and glucose, but not insulin, concentrations decreased significantly after saline administration at all three ages. At each time period, there was a significant positive relationship between the plasma insulin and proinsulin concentrations, the slope of which was significantly shallower on days 1-2 than subsequently. These results show that equine beta cells are responsive to glucose and arginine and release both insulin and proinsulin during the immediate postnatal period. They also suggest that newborn foals may be insulin resistant on the first day after birth.
KA Clarke, JW Ward, AJ Forhead, DA Giussani and AL Fowden
The effect of fetal cortisol on the activity of the type 2 isoform of the enzyme, 11 beta-hydroxysteroid dehydrogenase (11 beta-HSD2), was examined in ovine placenta and fetal kidney by measuring tissue 11 beta-HSD2 activity during late gestation when endogenous fetal cortisol levels rise and after exogenous cortisol administration to immature fetuses before the prepartum cortisol surge. Placental 11 beta-HSD2 activity decreased between 128-132 days and term (approximately 145 days of gestation) in association with the normal prepartum increase in fetal plasma cortisol. Raising fetal cortisol levels to prepartum values in the immature fetus at 128--132 days of gestation reduced placental 11 beta-HSD2 activity to term values. In contrast, 11 beta-HSD2 activity in the fetal renal cortex was unaffected by gestational age or cortisol infusion. When all the data were combined, there was an inverse correlation between the log fetal plasma cortisol level at delivery and placental 11 beta-HSD2 activity, expressed both on a weight-specific basis and per mg placental protein. Fetal cortisol therefore appears to be a physiological regulator of placental, but not renal, 11 beta-HSD2 activity in fetal sheep during late gestation. These findings have important implications, not only for glucocorticoid exposure in utero, but also for the local actions of cortisol within the placental tissues that are involved in initiating parturition in the sheep.
AJ Forhead, JC Ousey, WR Allen and AL Fowden
This study examined the effects of intrauterine growth on insulin secretion and resistance in newborn foals. Embryo transfer between small pony and large Thoroughbred mares was used to produce four groups of foals with different birth weights (pony in pony n=7; pony in Thoroughbred n=7; Thoroughbred in Thoroughbred n=8; Thoroughbred in pony n=8). On day 2 after birth, glucose (0.5 g/kg) was administered intravenously to the foal and blood samples were taken for 2 h to determine plasma glucose and insulin concentrations. On day 3, insulin sensitivity was assessed by giving insulin (0.75 U/kg i.v.) and measuring the decrement in plasma glucose in the foals. There were no significant differences in insulin secretion, insulin sensitivity or glucose tolerance between the control and growth-retarded Thoroughbred foals. Overgrown pony foals delivered by Thoroughbred mares had higher basal insulin levels and greater beta cell responses to glucose than the other groups of foals. The relationship between plasma glucose and insulin was also significantly steeper in overgrown pony foals than in the other groups. Variations in intrauterine growth rate, therefore, affect postnatal insulin secretion in the horse. More specifically, it is overgrowth, not growth retardation in utero that alters equine beta cell function in the immediate neonatal period.
AI Fahmi, AJ Forhead, AL Fowden and JI Vandenberg
During development, the heart has to adapt to changes in shape, size and, at birth, to significant changes in arterial pressure. The orderly contraction of the heart is dependent on the coordinated expression of ion channels at appropriate densities in individual cardiac myocytes. The present study demonstrated that the expression of the alpha-subunit of the cardiac sodium channel, SCN5a, was high at mid gestation but then decreased until 10 days before birth before increasing again. Whereas the beta-subunit, SCN1b, gradually increased in expression towards partum, there was no detectable expression of SCN3b at any gestational time point. Fetal adrenalectomy prior to the normal prepartum surge in cortisol caused a reduction in expression of SCN1b and a 7.0 kb transcript of SCN5a, but not the major 8.5 kb transcript. Conversely, cortisol infusion into immature fetuses precociously increased expression levels of SCN1b and the SCN5a 7.0 kb transcript. The results show that cortisol regulates cardiac SCN gene expression in fetal sheep during late gestation. These findings could have implications for the aetiology of sudden infant death syndrome and for the intrauterine programming of adult cardiovascular disease.
A Mostyn, S Pearce, H Budge, M Elmes, AJ Forhead, AL Fowden, T Stephenson and ME Symonds
The present study examined the extent to which the late gestation rise in fetal plasma cortisol influenced adipose tIssue development in the fetus. The effect of cortisol on the abundance of adipose tIssue mitochondrial proteins on both the inner (i.e. uncoupling protein (UCP)1) and outer (i.e. voltage-dependent anion channel (VDAC)) mitochondrial membrane, together with the long and short forms of the prolactin receptor (PRLR) protein and leptin mRNA was determined. Perirenal adipose tIssue was sampled from ovine fetuses to which (i) cortisol (2-3 mg/day for 5 days) or saline was infused up to 127-130 days of gestation, and (ii) adrenalectomised and intact controls at between 142 and 145 days of gestation (term=148 days). UCP1 protein abundance was significantly lower in adrenalectomised fetuses compared with age-matched controls, and UCP1 was increased by cortisol infusion and with gestational age. Adrenalectomy reduced the concentration of the long form of PRLR, although this effect was only significant for the highest molecular weight isoform. In contrast, neither the short form of PRLR, VDAC protein abundance or leptin mRNA expression was significantly affected by gestational age or cortisol status. Fetal plasma triiodothyronine concentrations were increased by cortisol and with gestational age, an affect abolished by adrenalectomy. When all treatment groups were combined, both plasma cortisol and triiodothyronine concentrations were positively correlated with UCP1 protein abundance. In conclusion, an intact adrenal is necessary for the late gestation rise in UCP1 protein abundance but cortisol does not appear to have a major stimulatory role in promoting leptin expression in fetal adipose tIssue. It remains to be established whether effects on UCP1 protein are directly regulated by cortisol alone or mediated by other anabolic fetal hormones such as triiodothyronine.