Effects on rat fetuses of intrauterine injections of insulin

Programming of cardiometabolic health: the role of maternal and fetal hyperinsulinaemia

Obesity and gestational diabetes during pregnancy have multiple short- and long-term consequences for both mother and child. One common feature of pregnancies complicated by maternal obesity and gestational diabetes is maternal hyperinsulinaemia, which has effects on the mother and her adaptation to pregnancy. Even though insulin does not cross the placenta insulin can act on the placenta as well affecting placental growth, angiogenesis and lipid metabolism. Obese and gestational diabetic pregnancies are often characterised by maternal hyperglycaemia resulting in exposure of the fetus to high levels of glucose, which freely crosses the placenta. This leads to stimulation of fetal ß-cells and insulin secretion in the fetus. Fetal hyperglycaemia/hyperinsulinaemia has been shown to cause multiple complications in fetal development, such as altered growth trajectories, impaired neuronal and cardiac development and early exhaustion of the pancreas. These changes could increase the susceptibility of the offspring to develop cardiometabolic diseases later in life. In this review, we aim to summarize and review the mechanisms by which maternal and fetal hyperinsulinaemia impact on (i) maternal health during pregnancy; (ii) placental and fetal development; (iii) offspring energy homeostasis and long-term cardiometabolic health; (iv) how interventions can alleviate these effects.

Adaptive responses to maternal nutrient restriction alter placental transport in ewes

INTRODUCTION

Maternal nutrient partitioning, uteroplacental blood flow, transporter activity, and fetoplacental metabolism mediate nutrient delivery to the fetus. Inadequate availability or delivery of nutrients results in intrauterine growth restriction (IUGR), a leading cause of neonatal morbidity and mortality. Maternal nutrient restriction can result in IUGR, but only in an unforeseeable subset of individuals.

METHODS

To elucidate potential mechanisms regulating fetal nutrient availability, singleton sheep pregnancies were generated by embryo transfer. Pregnant ewes received either a 50% NRC (NR; n = 24) or 100% NRC (n = 7) diet from gestational Day 35 until necropsy on Day 125. Maternal weight did not correlate with fetal weight; therefore, the six heaviest (NR Non-IUGR) and five lightest (NR IUGR) fetuses from nutrient-restricted ewes, and seven 100% NRC fetuses, were compared to investigate differences in nutrient availability.

RESULTS

Insulin, multiple amino acids, and their metabolites, were reduced in fetal circulation of NR IUGR compared to NR Non-IUGR and 100% NRC pregnancies. In contrast, glucose in fetal fluids was not different between groups. There was a nearly two-fold reduction in placentome volume and fetal/maternal interface length in NR IUGR compared to NR Non-IUGR and 100% NRC pregnancies. Changes in amino acid concentrations were associated with altered expression of cationic (SLC7A2, SLC7A6, and SLC7A7) and large neutral (SLC38A2) amino acid transporters in placentomes.

DISCUSSION

Results establish a novel approach to study placental adaptation to maternal undernutrition in sheep and support the hypothesis that amino acids and polyamines are critical mediators of placental and fetal growth in sheep.

Insulin deficiency alters the metabolic and endocrine responses to undernutrition in fetal sheep near term

Insulin deficiency affects the adult metabolic response to undernutrition, but its effects on the fetal response to maternal undernutrition remain unknown. This study examined the effects of maternal fasting for 48 h in late gestation on the metabolism of fetal sheep made insulin deficient by pancreatectomy (PX). The endocrine and metabolic responses to maternal fasting differed between intact, sham-operated and PX fetuses, despite a similar degree of hypoglycemia. Compared with intact fetuses, there was no increase in the plasma concentrations of cortisol or norepinephrine in PX fetuses during maternal fasting. In contrast, there was a significant fasting-induced rise in plasma epinephrine concentrations in PX but not intact fetuses. Umbilical glucose uptake decreased to a similar extent in both groups of fasted animals but was associated with a significant fall in glucose carbon oxidation only in intact fetuses. Pancreatectomized but not intact fetuses lowered their oxygen consumption rate by 15-20% during maternal fasting in association with increased uteroplacental oxygen consumption. Distribution of uterine oxygen uptake between the uteroplacental and fetal tissues therefore differed with fasting only in PX fetuses. Both groups of fetuses produced glucose endogenously after maternal fasting for 48 h, which prevented any significant fall in the rate of fetal glucose utilization. In intact but not PX fetuses, fasting-induced glucogenesis was accompanied by a lower hepatic glycogen content. Chronic insulin deficiency in fetal sheep therefore leads to changes in the counterregulatory endocrine response to hypoglycemia and an altered metabolic strategy in dealing with nutrient restriction in utero.

IGF-I and insulin regulate eIF4F formation by different mechanisms in muscle and liver in the ovine fetus

The mechanisms by which insulin-like growth factor I (IGF-I) and insulin regulate eukaryotic initiation factor (eIF)4F formation were examined in the ovine fetus. Insulin infusion increased phosphorylation of eIF4E-binding protein (4E-BP1) in muscle and liver. IGF-I infusion did not alter 4E-BP1 phosphorylation in liver. In muscle, IGF-I increased 4E-BP1 phosphorylation by 27%; the percentage in the gamma-form in the IGF-I group was significantly lower than that in the insulin group. In liver, only IGF-I increased eIF4G. Both IGF-I and insulin increased eIF4E. eIF4G binding in muscle, but only insulin decreased the amount of 4E-BP1 associated with eIF4E. In liver, only IGF-I increased eIF4E. eIF4G binding. Insulin increased the phosphorylation of p70 S6 kinase (p70(S6k)) in both muscle and liver and protein kinase B (PKB/Akt) in muscle, two indicative signal proteins in the phosphatidylinositol (PI) 3-kinase pathway. IGF-I increased PKB/Akt phosphorylation in muscle but had no effect on p70(S6k) phosphorylation in muscle or liver. We conclude that insulin and IGF-I modulate eIF4F formation; however, the two hormones have different regulatory mechanisms. Insulin increases phosphorylation of 4E-BP1 and eIF4E. eIF4G binding in muscle, whereas IGF-I regulates eIF4F formation by increasing total eIF4G. Insulin, but not IGF-I, decreased 4E-BP1 content associated with eIF4E. Insulin regulates translation initiation via the PI 3-kinase-p70(S6k) pathway, whereas IGF-I does so mainly via mechanisms independent of the PI 3-kinase-p70(S6k) pathway.

Regulation of fetal amino acid metabolism: substrate or hormonal regulation?

Insulin is regarded as the primary fetal growth-promoting hormone, but direct in vivo experimental data supporting this conjecture are sparse. Data obtained from studies in in vivo, chronically catheterized fetal lambs under a variety of experimental circumstances demonstrate that glucose availability is the primary modulator of fetal protein accretion, via its ability to diminish amino acid catabolism. The ovine fetus is shown to be resistant to insulin-induced suppression of proteolysis, relative to the adult. Data from studies in the human premature infant show that the findings in the ovine fetus are similar to those in the ex utero premature human.

Increased fetal glucose concentration decreases ovine fetal leucine oxidation independent of insulin

Fetal leucine oxidation rate is elevated during fasting of the ewe. Euglycemic hyperinsulinemia causes the leucine oxidation rate to decline. However, it is unclear whether this is a direct effect of insulin or is secondary to increased insulin-mediated glucose utilization. To better delineate the mechanism of decreased oxidation, we suppressed fetal insulin secretion by somatostatin infusion. Glucose was infused at a variable rate to achieve glucose concentrations 125 and 150% of basal. Leucine rate of appearance (Ra) was determined by infusion of [15N, 1-13C]leucine. Fraction of leucine appearance oxidized was determined by [1-14C]leucine infusion and determination of fetal 14CO2 excretion. Each fetus was studied during ad libitum maternal feeding and after a 5-day complete maternal fast. Changes were noted in fetal leucine oxidation, which declined from 8.4 +/- 1.2 to 5.0 +/- 0.8 mumol/min in the fed state during glucose infusion. Basal leucine oxidation was elevated during fasting (11 +/- 1.5 mumol/min, P < 0.05) and declined to 8.0 +/- 1.4 mumol/min during glucose infusion (P = 0.056). Leucine carbon Ra was unchanged by fasting and by glucose infusion; leucine nitrogen Ra declined in the fed state only. Leucine oxidation was inversely correlated with glucose concentration (oxidation = 12-0.26 x glucose concentration, r = 0.42, P = 0.004). Leucine oxidation was not correlated with insulin concentration (r = 0.2). Changes in fetal glucose concentration may alter the pattern of utilization of essential amino acids, independent of changes in insulin and insulin-mediated glucose utilization rate.

Insulin injection in the fetal rat: accelerated intrauterine growth and altered fetal and neonatal glucose homeostasis

Fetal hyperinsulinemia is a well-known correlate of accelerated fetal growth; the consequences of fetal hyperinsulinemia upon fetal and neonatal glucoregulation are less well understood. We injected rat fetuses of a litter on day 18 of gestation with either 5 units of long acting insulin (I) or 154 mmol/L NaCl. Twelve hours after injection, the wet and dry mass of total body and liver of I fetuses significantly exceeded that of controls. At birth (day 21.5), newborn I pups weighed 5.86 +/- .08 g, and controls, 5.48 +/- .05 g, (P less than .001). On day 18, within one hour of injection, fetal plasma insulin concentrations were significantly elevated and remained so for 24 hours. Mothers of I fetuses had significant elevations of plasma insulin at 1, 3, and 6 hours, and they developed transient hypoglycemia. Plasma glucose concentrations in I fetuses were significantly diminished at 1, 3, and 6 hours and then achieved control levels by 12 hours. Fetal hypoglycemia resulted from an apparent direct effect of insulin upon fetal tissue and from the maternal hypoglycemia. Hypoglycemic I fetuses demonstrated a sluggish alpha-cell response; they failed to increase plasma glucagon one hour after insulin injection. Values were significantly increased three hours after injection. At birth, I pups became hypoglycemic relative to controls. This was, in part, due to their significantly elevated plasma insulin concentrations at 120 and 240 minutes (120 minutes, 43.8 +/- 8 v 17.5 +/- 6 microU/mL, P less than .001). Plasma glucagon was significantly increased in I pups at 240 minutes.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of metabolic factors in the diabetic state on the in vitro development of preimplantation mouse embryos

The effects of insulin, glucagon, beta-hydroxybutyrate, and acetoacetate on the in vitro development of preimplantation mouse embryos were studied. In controls, 24% of blastocysts failed to develop successfully when grown for 72 h in Eagle's medium supplemented with 10% fetal calf serum. Insulin at concentrations of 1.0 and 2.0 IU/ml of culture medium interfered with development in 62-63% of the blastocysts. Preimplantation embryos showed a threshold pattern in their reaction to glucagon: its addition in concentrations of 0.0015 mM (5 micrograms/ml) did not significantly inhibit blastocyst development, while concentrations of 0.003 mM (10 micrograms/ml) inhibited 70% of blastocysts. The embryotoxic effects of ketone bodies were manifested only in relatively high doses. beta-hydroxybutyrate was embryotoxic at concentrations greater than 5 mg/ml, and its effects were dose dependent: 48 mM (6 mg/ml) inhibited 45% of blastocysts, while 80 mM (10 mg/ml) arrested 87% of embryos from further development. Acetoacetate at concentrations of 0.1 mM (10 micrograms/ml) inhibited the development of 50% of the blastocysts, and its effects were not dose dependent: concentrations of 1 mM (100 micrograms/ml) inhibited development in 63% of the embryos. The combination of the diabetic metabolic factors in relatively low concentrations was highly embryotoxic, especially when accompanied by hyperglycemia.

Skeletal growth of fetuses from streptozotocin diabetic rat mothers: in vivo and in vitro studies

For largely unknown reasons severe or moderate diabetes of pregnant rats results in pronounced fetal growth retardation. Therefore, some skeletal growth parameters of fetal rats from streptozotocin diabetic mothers were studied in vivo and in vitro. Two days post conception rats were intravenously injected with 65 mg/kg body weight streptozotocin. On day 20 post conception 8 normal and 8 diabetic rat mothers received 5 mu Ci 3-H thymidine intraperitoneally. One day later the experiments were terminated. Fetal body weight and body length were significantly (p less than 0.05-0.001) reduced in the hyperglycaemic rats compared to normal rats, as was the thymidine incorporation into rib cartilage (p less than 0.02). In the cell culture colony formation from isolated chondrocytes of normal and hyperglycaemic fetuses was determined. Proinsulin, insulin (62.5-250 ng/ml), insulin-like growth factor I and II (6.25-25 ng/ml) significantly (p less than 0.05-0.001) augmented colony formation in a dose-dependent manner, with the somatomedins being 8 times more effective than proinsulin or insulin. Isolated chondrocytes from hyperglycaemic compared to normal fetuses formed significantly (p less than 0.05-0.001) fewer colonies in the basal state and in response to all 4 hormones. The results confirm the growth retardation of fetuses from diabetic rat mothers. A reduced responsiveness of chondrocytes from hyperglycaemic fetuses to various growth factors could be demonstrated as compared to cells from normal fetuses.

Postnatal growth and fatty acid synthesis in overgrown rat pups induced by fetal hyperinsulinemia

Fetal hyperinsulinemia in the rat results in increased body weight, lipid content, and enhanced lipogenesis in liver and carcass. The purpose of our study was to determine whether the macrosomia and enhancement of fatty acid (FA) synthesis and/or content persisted postnatally in this animal model. Fetal hyperinsulinemia was produced in Sprague-Dawley rats by injecting fetuses with 2 units of insulin at 20.5 days of gestation. Alternate pups in the same litter were injected with saline. Pups were delivered surgically at 22.5 days of gestation, were weighed daily and sacrificed on day 15. FA content and synthesis rates of liver and skeletal muscle were measured. We found: (1) At birth, insulin-treated pups were 12% heavier than saline littermates, (5.88 +/- 0.14 g v 5.26 +/- 0.14 g, P less than .01); and (2) The enhanced growth associated with prenatal insulin treatment persisted during the suckling period, ie, compared with saline-treated controls, insulin pups were 15.7% heavier at 15 days of age (P less than .01); growth velocity of insulin pups, beginning on day 3, significantly exceeded that of control pups (P less than .05). FA contents of liver and muscle in insulin pups, (62.6 +/- 5.7 mumol/g and 62.7 +/- 13.2 mumol/g) were significantly greater (P less than .05) than in saline littermates (45.1 +/- 5.6 mumol/g and 30.2 +/- 4.7 mumol/g, respectively). We conclude that.(ABSTRACT TRUNCATED AT 250 WORDS)

Untoward effects of pharmacological doses of insulin in early chick embryos: through which receptor are they mediated?

The teratogenic effect of insulin in early vertebrate embryos is controversial and the mechanisms involved are unknown. We studied the effects of pharmacological doses of insulin in chick embryos during the period of differentiation. We compared the effects of insulin with two proinsulins, desoctapeptide-insulin and multiplication-stimulating activity, peptides that have little insulin-like metabolic activity while they have significant growth effects. Chick embryos at 46 h of development were injected with the different peptides. At 96 h the mortality and abnormal growth elicited by the peptides were dose-dependent. Considering the indices of lethality (LD50) and affected embryos (ED50) as 100% for insulin, proinsulin was 59-66% as potent as insulin, desoctapeptide-insulin 2-6% and multiplication-stimulating activity 176-204%. In the surviving embryos, insulin (5 micrograms, decreased DNA, RNA and protein content by 49%, 40% and 48% respectively compared with controls. The effects of insulin were not corrected by simultaneous glucose injections. These data suggest that insulin, at pharmacological doses, interferes with embryo development through a non-metabolic pathway, probably via a growth-type receptor.

Effects of insulin and hydrocortisone on lung tissue phosphatidyl choline and disaturated phosphatidyl choline in fetal rabbits in vivo

To determine the effects of insulin and hydrocortisone on lung tissue surfactant, the fetuses of New Zealand White rabbits were injected with insulin, saline (0.154 mol/l), insulin+hydrocortisone, insulin+saline or saline+hydrocortisone, or were sham-operated on day 27 of gestation. Twenty-four hours later (on day 28 of gestation) delivery was accomplished by Caesarian hysterectomy. The lung tissues were analysed for phosphatidyl choline and disaturated phosphatidyl choline. Both were higher in fetuses injected with insulin than in sham-operated control animals. The phosphatidyl choline and disaturated phosphatidyl choline were also higher in the fetuses injected with insulin+hydrocortisone than in fetuses injected with insulin or hydrocortisone alone. These results suggest that insulin increases the phosphatidyl choline and disaturated phosphatidyl choline content in lung tissue in fetal rabbits in vivo, and that in the presence of hydrocortisone, insulin appears to have an additive effect.

Raised plasma somatomedin activity and cartilage metabolic activity (35S sulphate uptake in vitro) in the fetus of the mildly diabetic pregnant rat

A mildly diabetic state was induced in pregnant rats following treatment with streptozotocin the day after mating. On day 21 of gestation, these rats had a lower plasma insulin (55 +/- 9 versus 107 +/- 23 mU/l for control rats; p less than 0.05, mean +/- SEM) and a reduced pancreatic area occupied by insulin-containing cells compared with control animals (0.40 +/- 0.04 versus 1.03 +/- 0.08%; p less than 0.001), but hyperglycaemia was not apparent. Fetuses from mildly diabetic animals were longer but not heavier than those from control rats. Plasma somatomedin activity measured by fetal rat cartilage bioassay was higher in fetuses from mildly diabetic rats (1.12 +/- 0.07 versus: 0.74 +/- 0.05 U/ml for control fetuses; p less than 0.001) as was cartilage metabolic activity in basal culture medium (35S sulphate uptake) (1883 +/- 141 versus 1473 +/- 104 c.p.m./mg for control rats; p less than 0.05), but plasma insulin levels and the pancreatic area occupied by insulin-containing cells did not differ between the two groups of fetuses. Fetal plasma somatomedin activity, measured by fetal cartilage assay, showed a significant positive correlation with both body weight and length. It is concluded that by day 21 of gestation a small body overgrowth had occurred in the fetus of the mildly diabetic rat and this was associated with an increase in plasma somatomedin activity, but not with any abnormality of circulating insulin levels or volume density of B cells in the pancreatic islets.

The relationship of body and placental weight to plasma levels of insulin and other hormones during development in fetal rabbits

The relationship of body and placental weight with plasma levels of insulin, glucose, glucagon, glucocorticoids and somatomedins in fetal rabbits between 22 days of gestation and term has been investigated. At 22, 24, 26, 28 and 30 days gestation, body and placental weights were strongly correlated and were also significantly correlated with log transformed plasma insulin concentrations in individual fetuses from nulliparous does. At 30 days gestation, mean plasma insulin levels, as well as body and placental weights, were significantly elevated in fetuses from multiparous does (p less than 0.05) and in fetuses whose growth had been increased by surgical reduction in litter size on day 9 of pregnancy (p less than 0.001). Log insulin concentrations in fetuses from multiparous does and in litter-reduced fetuses were also significantly correlated with bodyweight. The correlation coefficient for log insulin and bodyweight in all individual rabbit fetuses at 30 days gestation was 0.69 (p less than 0.001, n = 116). Fetal log plasma glucagon concentrations at 26 and 28 days gestation were negatively correlated with body weight, but were not significantly related at 30 days gestation. Neither glucose nor glucocorticoid concentrations were significantly related to bodyweight in individual fetuses. Plasma somatomedin activity in litter-reduced fetuses at 30 days gestation, was significantly higher (p less than 0.02) than in normal fetuses, but was not significantly correlated with fetal body weight. The observation of strong positive correlation between insulin and bodyweight throughout the last third of gestation reinforces the belief that fetal insulin plays an important role in the regulation of both placental and fetal growth.