Insulin-induced alterations in amino acid metabolism in the fetal lamb

To investigate the role of insulin in modulation of fetal amino acid metabolism, insulin infusions were performed in 10 chronically-catheterized fetal lambs. Fetal insulin infusion caused a dose related fall in the arterial blood concentrations of 13 of 15 amino acids studied as well as a 15-25% decrease in total amino acid concentration. Fetal lambs exhibited a biphasic response of umbilical total amino acid uptake when compared to fetal blood insulin concentration, i.e., at achieved fetal insulin concentrations less than 100 microU/ml, umbilical uptake of 9 specific amino acids as well as summed amino acid uptake from the umbilical circulation were depressed, but at insulin concentrations of 100-350 microU/ml, amino acid uptakes were similar to or above control values. Insulin infusion also caused a drastic diminution in the rate of fetal urea excretion. These findings suggest that insulin acts in the fetus to depress amino acid catabolism, thus altering amino acid extraction and uptake. Depressed protein catabolism with or without enhanced amino acid uptake would have the theoretical effect of stimulation of net protein synthesis with a shift toward use of nonprotein substrates for energy purposes.     

Fetal insulin and placental 3-O-methyl glucose clearance in near-term sheep

It is difficult, if not impossible, to measure the placental transfer of glucose directly because of placental glucose consumption and the low A-V glucose difference across the sheep placenta. We have approached the problem of quantifying placental hexose transfer by using a nonmetabolized glucose analogue (3-O-methyl glucose) which shares the glucose transport system. We have measured the clearance by using a multisample technique permitting least squares linear computing to avoid the errors implicit in the Fick principle. The placental clearance of 3-O-methyl glucose was measured in the control condition and after the administration of insulin to the fetal circulation. A glucose clamp technique was used to maintain constant transplacental glucose concentrations throughout the duration of the experiment. A control series was performed in which the only intervention was the infusion of normal saline. In these experiments the maternal and fetal glucose concentrations remained constant as did the volume of distribution of 3-O-methyl glucose in the fetus. The maternal insulin concentration remained constant and fetal insulin concentration changed from 11 +/- 2 microU/ml to 355 +/- 51 microU/ml (P less than 0.01). In the face of this large increase in fetal plasma insulin, there was no change in the placental clearance of 3-O-methyl glucose. In the control condition the clearance was 14.1 +/- 1.0 ml/min per kg and this was 13.8 +/- 1.0 ml/min per kg in the high insulin condition. Fetal insulin may change placental glucose flux by decreasing fetal plasma glucose concentrations but does not do so by changing the activity of the glucose transport system.     

Failure of hyperketonemia to alter basal and insulin-mediated glucose metabolism in man

The effect of physiologic elevations of plasma hydroxybutyrate induced by the infusion of sodium D,L-beta-hydroxybutyrate (15 mumol X kg-1 X min-1) on carbohydrate metabolism was examined with the euglycemic insulin clamp technique in nine healthy volunteers. Plasma insulin concentration was acutely raised and maintained at 126 +/- 6 microU/ml and plasma glucose was held constant at the fasting level by a variable glucose infusion. Glucose uptake of 6.53 +/- 0.80 mg X kg-1 X min-1 was unchanged by hyperketonemia when compared with an intraindividual control study using saline instead of beta-OH-butyrate infusion (6.26 +/- 0.59 mg X kg-1 X min-1). In studies, in which the degree of metabolic alkalosis accompanying butyrate infusion was mimicked by the continuous administration of bicarbonate, glucose uptake was also unaffected (6.25 +/- 0.45 mg X kg-1 X min-1). Furthermore, hyperketonemia had no effect on basal glucose production or the suppression of hepatic glucose production following hyperinsulinemia. It is concluded that moderate elevations in plasma beta-hydroxy-butyrate do not alter hepatic or peripheral glucose metabolism.     

Effect of insulin, prostaglandin E1 and uptake inhibitors on glucose transport in the perfused guinea-pig placenta

The effects of insulin, prostaglandin E1 (PGE1) and uptake inhibitors on unidirectional D-glucose influx at brush border (maternal) and basal (fetal) sides of the guinea-pig syncytotrophoblast were investigated in the intact, perfused guinea-pig placenta by rapid, paired-tracer dilution. Experiments were performed in either an in situ preparation artificially perfused through the umbilical vessels (intact maternal circulation) or in the fully isolated dually-perfused placenta in which both interfaces were studied simultaneously. Kinetic characterization of unidirectional D-glucose influx gave apparent Km values (mean +/- SEM) at maternal and fetal sides of 70 +/- 6 and 87 +/- 16 mM respectively; corresponding Vmax values were 53 +/- 3 and 82 +/- 6 mumol min-1g-1. At the fetal side (singly-perfused placenta) cytochalasin B (50 microM), ethylidene-D-glucose (100 mM) and PGE1 (1 microM) partially inhibited D-glucose uptake whereas cortisol (50 microM) and progesterone (100 microM) had no effect. Abolition of the sodium gradient across the fetal interface did not modulate the kinetics of influx. In the presence of 150 mu units ml-1 insulin (dually-perfused placenta), unidirectional uptake into the trophoblast and transplacental D-[3H]glucose transfer were unaltered. In contrast, prostaglandin E1 (1 microM) markedly reduced the Km and Vmax for D-glucose at both interfaces and the inhibitory effect was reflected in a reduction in specific transplacental D-glucose transfer. Further experiments showed that the isolated placenta releases prostaglandins (PGE; PGF2 alpha) into both circulations. Bilateral insulin perfusion did not affect either lactate release by the placenta or rapid metabolism of D-[14C]glucose to [3H]lactate (usually less than 10% effluent [14C]lactate in 5 min). An asymmetric degradation of exogenous insulin was observed in the dually-perfused placenta: uterine venous samples contained 24 +/- 7 microunits ml-1 immunoreactive insulin when compared to the arterial concentration (151 +/- 3 microU ml-1 perfusate) while no change was measureable in the fetal circulation within the same time period (152 +/- 5 microU ml-1). This asymmetry was confirmed in experiments employing [125I]insulin. These results demonstrate that glucose transport in the intact guinea-pig placenta occurs by a sodium-independent, cytochalasin B-inhibitable system which is insulin-insensitive. Prostaglandin E1 appeared to be a potent transport inhibitor which suggests that prostaglandins may be involved in the 'down' regulation of placental glucose transport in vivo.     

The relationship of umbilical glucose uptake to uterine blood flow

In 30 experiments performed on 5 pregnant sheep, the rate of glucose transfer from the placenta to fetus via the umbilical circulation was measured while varying uterine blood flow by means of a cuff-type occluder and while maintaining a constant maternal glucose concentration by means of a 'glucose clamp'. Over the range of uterine blood flows obtained, there was no significant effect on the simultaneously measured umbilical blood flow. Fetal glucose uptake and arterial glucose concentration remained normal as the uterine blood flow rate decreased from 600 to 300 ml per min per kg of fetus. At blood flow rates less than 300 ml.min-1.kg-1, the fetal glucose uptake decreased and became negative in one instance while the arterial glucose concentration became variable and markedly increased in 2 animals. This increase in fetal glucose concentration was associated with a decrease in the uterine oxygen delivery rate, a decrease in fetal oxygen content and a decrease in fetal oxygen uptake. These observations support the concept that fetal glucose metabolism is altered by severe hypoxia and demonstrate that there is little effect of uterine blood flow on fetal glucose uptake in the normal physiological range.     

Fetal CO2 kinetics

Knowledge of CO2 kinetics in the fetus is important for the design and interpretation of fetal metabolic studies that use carbon-labelled tracers. To study fetal CO2 kinetics, four fetal sheep were infused at constant rate with NaH14CO3 to simulate a constant rate of fetal 14CO2 production from the metabolism of a 14C-labelled substrate. Uterine and umbilical blood flows, and concentrations of 14CO2 and total CO2 in umbilical arterial and venous blood and in uterine arterial and venous blood were measured. During steady state, the excretion of 14CO2 via the umbilical circulation was 99.6 +/- 1.0 (SEM)% of the NaH14CO3 infusion rate. The irreversible disposal rate of CO2 molecules from the fetal CO2 pool was approximately 5 times greater than the metabolic production of CO2 by the fetus. This evidence demonstrates that measurements of fetal 14CO2 excretion via the umbilical circulation can provide an accurate measurement of fetal 14CO2 production and that the exchange rate of CO2 molecules between placenta and fetal blood is much greater than the net rate of excretion of CO2 molecules from fetus to placenta.     

The placenta releases branched-chain keto acids into the umbilical and uterine circulations in the pregnant sheep

There was net uptake of branched-chain keto acids by the fetus from the umbilical circulation. Mean fetal uptake of the 3 keto acids 2-keto isovalerate, 2-keto isocaproate and 2-keto methylvalerate was 1.8 mumol/min per kg of fetus. The concentrations in the umbilical vein for these keto acids were 10.9 +/- 3.8 microM (mean +/- SD: 2-keto isovalerate), 19.7 +/- 6.1 microM (2-keto isocaproate) and 14.8 +/- 5.3 microM (2-keto methylvalerate) respectively. The coefficients of extraction for the same keto acids were 17.2%, 16.8% and 11.9% respectively. Fetal uptakes (both mumol/min and mumol/min per kg fetus) were positively correlated with umbilical supply. There were concentration gradients across the placenta, with fetal concentration: maternal concentration ratios of 3.3 +/- 1.5 for 2-keto isovalerate, 2.1 +/- 0.8 for 2-keto isocaproate and 1.3 +/- 0.6 for 2-keto methylvalerate. The net release of 2-keto acids into the umbilical circulation may conserve the carbon skeleton of branched-chain amino acids for fetal metabolism and growth. In the uterine circulation there was not a consistent pattern of release from or uptake by the uteroplacental tissues. It is suggested that branched-chain keto acids may contribute to fetal growth or energy metabolism.     

Insulin induced fetal hypoglycaemia and fetal and maternal plasma prostaglandin concentrations in sheep in late gestation.

Fetal hypoglycaemia consequent on food withdrawal for 48 h in sheep in late pregnancy is accompanied by an increase in fetal PGE2 plasma concentrations and myometrial contractility. To assess the contribution of fetal hypoglycaemia and related cellular glucopenia in the increased production of fetal PGE2 we studied the effect of 48 h insulin infusion to the fetus. Fetal whole blood glucose was lowered from 19 +/- 2 to 9 +/- 1 mg.dl-1. This experimental regimen maintains glucose availability to those fetal cells in which insulin increases glucose uptake. Fetal umbilical venous and femoral arterial PGE2 concentrations and umbilical veno-arterial PGE2 difference were unchanged, but maternal uterine veno-arterial difference for PGFM increased during the insulin induced fetal hypoglycaemia. Myometrial activity was also unchanged. We conclude that the increased fetal PGE concentration previously reported during food withdrawal is due to a deficiency of glucose to specific insulin dependent cells within vascular beds served by the fetal cardiovascular system. In addition, the findings suggest a need for a supply of glucose of fetal origin for cells that are responsible for increased PGFM concentrations in the maternal uteroplacental circulation.     

Maternal insulin and placental 3-O-methyl glucose transport

The effects of insulin in the maternal circulation on the placental clearance of 3-O-methyl glucose were investigated in 7 animals in the presence of a constant maternal glucose concentration. While maternal insulin concentration changed from 12 +/- 4 to 175 +/- 33 mu Units/ml, the placental clearance remained constant at 16.2 +/- 1.2 (control) and 15 +/- 1.3 ml/min per kg fetus under the influence of the insulin. To test the secondary hypothesis that in the control condition the hexose transport system was saturated, we performed a further series of experiments in 6 fasted animals. In these animals the control maternal plasma insulin concentration was 2 +/- 0.3 mu Units/ml and after the infusion of insulin it increased to 562 +/- 26 mu Units/ml. Under conditions of constant maternal and fetal plasma glucose concentrations, this massive elevation of plasma insulin did not change the placental clearance of 3MeG which was 15.2 +/- 1.6 in the control condition and 13.3 +/- ml/min per kg under the influence of high insulin. We conclude that maternal insulin ranging from 2 mu Units/ml to supraphysiologic doses does not effect a physiologically significant change in placental hexose transfer. Placental glucose transfer can probably therefore, be changed only be changing the concentration of glucose in the maternal and fetal plasma.     

Varying brain insulin concentrations differentially regulate the fetal brain insulin receptor

We investigated the downregulating effect of varying states (physiologic and pharmacologic) of systemic and intracranial hyperinsulinism on the 28 to 30 day fetal rabbit brain insulin receptor. Alloxan-induced maternal diabetes (n = 5) produced mild fetal hyperinsulinemia (D) (plasma insulin concentrations = 59.80 +/- 8.10 microU/ml, control = 26.25 +/- 3.70; p less than 0.01), whereas systemic administration (IMI) of 1.0 U (n = 4) and 2.0 U (n = 4) of insulin to the fetus resulted in moderate (103.13 +/- 34.63 microU/ml) and severe (288.3 +/- 51 microU/ml) fetal hyperinsulinemia respectively. All three states of systemic hyperinsulinemia neither altered the fetal brain insulin content nor the brain insulin receptor number and affinity. 0.01 U (n = 4) of intracranial insulin administration (ICI) increased the brain insulin content four-fold (p less than 0.01) but did not alter the brain insulin receptor number or affinity. 0.1 (n = 5) and 2.0 U (n = 7) of intracranial insulin increased the brain insulin content to supraphysiologic concentrations (p less than 0.01) and decreased the fetal brain insulin receptor number (p less than 0.01), the affinity remaining constant. We conclude that 1) regardless of the ability of insulin to cross the blood brain barrier, the downregulation of the brain insulin receptor is insulin dose-dependent and 2) the downregulation of the fetal brain insulin receptor is not a physiologic but a pharmacologic effect of insulin.     

Placental transfer of glucose

The rates of glucose transfer from maternal blood to pregnant uterus and from placenta to fetus were measured in eight sheep at spontaneously occurring glucose concentrations (control state) and while the fetus, the mother, or both were receiving a constant infusion of glucose. In addition two fetuses received insulin infusions. In the control state the net glucose flux from placenta to fetus was only 27 +/- 2.6% (SEM) of the net flux from the uterine circulation to the pregnant uterus. An empirical equation describing the relationship between placental glucose transfer and arterial plasma glucose concentrations was derived from the data and compared with equations constructed on the basis of methematical models of placental function. This analysis indicates that: (1) placental glucose transfer is mediated by carriers with Km approximately equal to 70 mg/dl; (2) the rate of glucose transfer from mother to fetus is limited primarily by the transport characteristics and glucose consumption rate of the placenta; (3) under normal conditions of placental perfusion, glucose transfer is approximately 15% less than it would be if placental blood flows were infinitely large.     

Effect of hyperinsulinemia on amino acid utilization in the ovine fetus

We studied the effect of an acute 4-h period of hyperinsulinemia (H) on net utilization rates (AAUR(net)) of 21 amino acids (AA) in 17 studies performed in 13 late-gestation fetal sheep by use of a novel fetal hyperinsulinemic-euglycemic-euaminoacidemic clamp. During H [84 +/- 12 (SE) microU/ml H, 15 +/- 2 microU/ml control (C), P < 0. 00001], euglycemia was maintained by glucose clamp (19 +/- 0.05 micromol/ml H, 1.19 +/- 0.04 micromol/ml C), and euaminoacidemia (mean 4.1 +/- 3.3% increase for all amino acid concentrations [AA], nonsignificantly different from zero) was maintained with a mixed amino acid solution adjusted to keep lysine concentration constant and other [AA] near C values. H produced a 63.7% increase in AAUR(net) (3.29 +/- 0.66 micromol. min(-1). kg(-1) H, 2.01 +/- 0.55 micromol. min(-1). kg(-1) C, P < 0.001), accounting for a 60.1% increase in fetal nitrogen uptake rate (2,064 +/- 108 mg. day(-1). kg(-1) H, 1,289 +/- 73 mg. day(-1). kg(-1) C, P < 0.001). Mean AA clearance rate (AAUR(net)/[AA]) increased by 64.5 +/- 18.9% (P < 0. 001). Thus acute physiological H increases net amino acid and nitrogen utilization rates in the ovine fetus independent of plasma glucose and [AA].     

The effect of elevation of maternal plasma catecholamines on the fetus and placenta of the pregnant sheep

To study the effects of reduced uterine blood flow on fetal and placental metabolism, adrenaline has been infused at physiological doses (0.5 microgram/min per kg) into the circulation of the pregnant sheep. This gives a reduction of about one third of uterine blood flow at days 120-143 of pregnancy, but causes no significant change in umbilical blood flow. In contrast to the effects of constricting the uterine artery to reduce blood flow to a similar degree, placental oxygen consumption was reduced and that, together with a large increase in lactate production, indicated the placenta became hypoxic. The fetal blood gas status and hence oxygen consumption was not affected significantly. A consistent arterio-venous difference for glucose across the umbilical or uterine circulations was not detected unless the uterine blood flow was comparatively high. Glucose balance across the uterus showed a close linear relationship with uterine blood flow and more particularly with the supply of glucose to the uterus. There was clear evidence for glucose uptake by the placenta and fetus and also glucose output by both. The latter was more common when uterine blood flow was comparatively low or reduced by adrenaline infusion. The results are consistent with the concept that glucose supply has to be maintained to the placenta even at the expense of fetal stores, although lactate can substitute if there is enhanced output because of fetal hypoxia. They indicate that placental mobilisation of glycogen can lead to a net output of glucose to the mother. The manner of communicating to the fetus changes in placental state that occur during maternal adrenaline infusion is not clear. However towards the end of the 60 min infusion, elevation of fetal plasma adrenaline, probably resulting from a breakdown of the placental permeability barrier, may be an important signal.     

Hyperinsulinemia compensates for infection-induced impairment in net hepatic glucose uptake during TPN

In animals receiving total parenteral nutrition (TPN), infection impairs net hepatic glucose uptake (NHGU) by 40% and induces mild hyperinsulinemia. In the normal animal, the majority of the glucose taken up by the liver is diverted to lactate, but in the infected state, lactate release is curtailed. Because of the hyperinsulinemia and reduced NHGU, more glucose is utilized by peripheral tissues. Our aims were to determine the role of infection-induced hyperinsulinemia in 1) limiting the fall in NHGU and hepatic lactate release and 2) increasing the proportion of glucose disposed of by peripheral tissues. Chronically catheterized dogs received TPN for 5 days via the inferior vena cava. On day 3, a fibrin clot with a nonlethal dose of E. coli was placed into the peritoneal cavity; sham dogs received a sterile clot. On day 5, somatostatin was infused to prevent endogenous pancreatic hormone secretion, and insulin and glucagon were replaced at rates matching incoming hormone concentrations observed previously in sham or infected dogs. The TPN-derived glucose infusion was adjusted to maintain a constant arterial plasma glucose level of approximately 120 mg/dl. after a basal blood sampling period, the insulin infusion rate was either maintained constant (infected time control, Hi-Ins, n = 6; sham time control, Sham, n = 6) or decreased (infected + reduced insulin, Lo-Ins; n = 6) for 180 min to levels seen in noninfected dogs (from 23 +/- 2 to 12 +/- 1 microU/ml). Reduction of insulin to noninfected levels decreased NHGU by 1.4 +/- 0.5 mg x kg(-1) x min(-1) (P < 0.05) and nonhepatic glucose utilization by 4.8 +/- 0.8 mg x kg(-1) x min(-1) (P < 0.01). The fall in NHGU was caused by a decline in HGU (Delta-0.6 +/- 0.4 mg x kg(-1) x min(-1)) and a concomitant increase in hepatic glucose production (HGP, Delta0.8 +/- 0.5 mg x kg(-1) x min(-1)); net hepatic lactate release was not altered. Hyperinsulinemia that accompanies infection 1) primarily diverts glucose carbon to peripheral tissues, 2) limits the fall in NHGU by enhancing HGU and suppressing HGP, and 3) does not enhance hepatic lactate release, thus favoring hepatic glucose storage. Compensatory hyperinsulinemia plays a critical role in facilitating hepatic and peripheral glucose disposal during an infection.     

Impaired insulin secretion after intravenous glucose in neonatal rhesus monkeys that had been chronically hyperinsulinemic in utero.

Chronic hyperinsulinemia in the fetal rhesus monkey results in fetal macrosomia without change in fetal plasma glucose concentration. After 18 days of hyperinsulinemia, fetuses were delivered by cesarean section, at which time experimental animals had significantly (P less than 0.05) elevated umbilical artery plasma insulin concentrations of 2039 +/- 854 pM compared with 129 +/- 72 pM. Plasma immunoreactive C peptide (IRCP) was significantly reduced to 39 +/- 17 pM compared with 286 +/- 134 pM. Eight hours after the insulin-delivering pumps were removed, plasma glucose, insulin, and IRCP were the same in both the experimental and control groups. At this time, 0.5 g glucose/kg was given intravenously and insulin and IRCP secretion was measured over a 1-hr period. The secretion, as assessed by integrating the incremental response of both insulin and IRCP, was significantly (P less than 0.05) lower by 80% in the experimental animals compared with the controls. Our data show that experimentally produced in utero euglycemic hyperinsulinemia in the fetal rhesus monkey produces a defect in the glucose-mediated insulin secretory mechanism that is detectable in the neonatal period even when hyperinsulinemia is no longer present. This study provides more support for the concept that fuel/hormone-mediated fetal teratogenesis may explain some of the fetopathy of the infant of the diabetic mother.     

Placental Glucose Transfer: A Human In Vivo Study

ObjectivesThe placental transfer of nutrients is influenced by maternal metabolic state, placenta function and fetal demands. Human in vivo studies of this interplay are scarce and challenging. We aimed to establish a method to study placental nutrient transfer in humans. Focusing on glucose, we tested a hypothesis that maternal glucose concentrations and uteroplacental arterio-venous difference (reflecting maternal supply) determines the fetal venous-arterial glucose difference (reflecting fetal consumption).MethodsCross-sectional in vivo study of 40 healthy women with uncomplicated term pregnancies undergoing planned caesarean section. Glucose and insulin were measured in plasma from maternal and fetal sides of the placenta, at the incoming (radial artery and umbilical vein) and outgoing vessels (uterine vein and umbilical artery).ResultsThere were significant mean (SD) uteroplacental arterio-venous 0.29 (0.23) mmol/L and fetal venous-arterial 0.38 (0.31) mmol/L glucose differences. The transplacental maternal-fetal glucose gradient was 1.22 (0.42) mmol/L. The maternal arterial glucose concentration was correlated to the fetal venous glucose concentration (r = 0.86, p<0.001), but not to the fetal venous-arterial glucose difference. The uteroplacental arterio-venous glucose difference was neither correlated to the level of glucose in the umbilical vein, nor fetal venous-arterial glucose difference. The maternal-fetal gradient was correlated to fetal venous-arterial glucose difference (r = 0.8, p<0.001) and the glucose concentration in the umbilical artery (r = −0.45, p = 0.004). Glucose and insulin concentrations were correlated in the mother (r = 0.52, p = 0.001), but not significantly in the fetus. We found no significant correlation between maternal and fetal insulin values.ConclusionsWe did not find a relation between indicators of maternal glucose supply and the fetal venous-arterial glucose difference. Our findings indicate that the maternal-fetal glucose gradient is significantly influenced by the fetal venous-arterial difference and not merely dependent on maternal glucose concentration or the arterio-venous difference on the maternal side of the placenta.     

Metabolism of glucose by fetus and placenta of sheep. The effects of normal fluctuations in uterine blood flow

The metabolism by the fetus and placenta of [2-3H, U-14C]glucose infused into fetal sheep has been studied. Uptake of glucose from the fetus by the placenta and transfer to the ewe, as well as placental metabolism of glucose to fructose and lactate have been quantified. About two-thirds of the glucose removed from the fetal circulation was taken up by placenta. Less than 15% of this passed back into the maternal circulation, the remainder was converted, at roughly equivalent rates, into lactate and fructose, most of which was transferred back to the fetus. It seems likely that little of this glucose is oxidised by the placenta. This data indicates that there are substrate cycles between the placenta and fetus, one possible function of which is to limit fetal glucose loss back to the mother; lactate and fructose have limited placental permeability. At uterine blood flow rates in the middle of the normal range net glucose uptake by the placenta from the maternal circulation was about 7-fold higher than that from the fetus. About 20% of this was transported to the fetus, 50% was oxidised and much of the remainder converted to lactate and transferred back to the ewe. Labelling patterns in fructose and lactate make it unlikely that this placental pool of glucose mixes freely with that derived from uptake from the fetus. Net movement of glucose across the placenta is markedly influenced by fluctuations in uterine blood flow over the normal range of 500-3000 ml/min. At low flow rates there is net output of glucose from the fetus to the placenta, and in some instances from the placenta to the ewe, i.e. there is evidence of net utero-placental production of glucose to the ewe separate from output by the fetus. There is a close linear relationship between uterine glucose supply (maternal arterial concentration x uterine blood flow) and net balance across the placenta. As uterine supply of glucose falls there is increased uptake by the placenta of glucose from the fetal circulation and corresponding enhanced recycling of fructose and lactate to the fetus. This production of fructose and lactate by the placenta may function to reduce glucose loss from the fetus to the ewe. Hence at high rates of placental uptake of glucose from the fetus placental production of lactate and particularly fructose may approach saturation and allow significant backflow of glucose from the fetus to the ewe. Under these conditions glucose uptake may in part sustain placental oxygen consumption.     

Effects of maternal fasting on hepatic gluconeogenesis and glucose metabolism in fetal lambs.

In unstressed, normoglycaemic fetal lambs, the liver produces little glucose, and gluconeogenesis is insignificant. Indirect measurements have suggested that the fetus may produce glucose endogenously during hypoglycaemia induced by prolonged maternal starvation. In eight fetal lambs we directly measured total and radiolabelled substrate concentration differences across the liver to determine whether the fetal liver produces glucose after four days of fasting-induced hypoglycaemia. Simultaneously we measured umbilical glucose uptake and fetal glucose utilization. Glucose concentrations in ewes (1.78 +/- 0.44 mmol.-1) and fetuses (0.61 +/- 0.17 mmol.l-1) were decreased. Fetal glucose utilization rate (21.7 +/- 8.9 mumol.min-1.kg-1) was not significantly different from umbilical glucose uptake (17.2 +/- 8.9 mumol.min-1.kg-1). Hepatic glucose production (8.9 +/- 17.2 mumol.min-1.100 g-1) and gluconeogenesis (6.1 +/- 4.4 mumol.min-1.100 g-1) were present, but could account for only 13% and 8% of fetal glucose requirements, respectively. To determine whether glucose output by the fetal liver was limited by substrate availability, we infused lactate, acetate, and acetone into the umbilical veins of four fasted animals, increasing hepatic substrate delivery. Hepatic glucose output did not increase during infusion of gluconeogenic substrates, indicating that substrate availability did not limit gluconeogenesis. We conclude that the gluconeogenic pathway is intact in late-gestation fetal lambs and that the fetal liver is capable of gluconeogenesis. However, the primary change in fetal metabolism during maternal starvation is the reduction in fetal glucose utilization, obviating the need for substantial hepatic glucose production. The factors stimulating this modest increase in fetal hepatic glucose production remain to be elucidated.     

Acute supplementation of amino acids increases net protein accretion in IUGR fetal sheep

Placental insufficiency decreases fetal amino acid uptake from the placenta, plasma insulin concentrations, and protein accretion, thus compromising normal fetal growth trajectory. We tested whether acute supplementation of amino acids or insulin into the fetus with intrauterine growth restriction (IUGR) would increase net fetal protein accretion rates. Late-gestation IUGR and control (CON) fetal sheep received acute, 3-h infusions of amino acids (with euinsulinemia), insulin (with euglycemia and euaminoacidemia), or saline. Fetal leucine metabolism was measured under steady-state conditions followed by a fetal muscle biopsy to quantify insulin signaling. In CON, increasing amino acid delivery rates to the fetus by 100% increased leucine oxidation rates by 100%. In IUGR, amino acid infusion completely suppressed fetal protein breakdown rates but increased leucine oxidation rate by only 25%, resulting in increased protein accretion rates by 150%. Acute insulin infusion, however, had very little effect on amino acid delivery rates, fetal leucine disposal rates, or fetal protein accretion rates in CON or IUGR fetuses despite robust signaling of the fetal skeletal muscle insulin-signaling cascade. These results indicate that, when amino acids are given directly into the fetal circulation independently of changes in insulin concentrations, IUGR fetal sheep have suppressed protein breakdown rates, thus increasing net fetal protein accretion.     

Stimulation of glucose uptake in skeletal muscle using bolus or continuous insulin delivery

We investigated glucose uptake in the non-cyclically perfused rat hindlimb in response to continuous infusion (CI) or bolus injection (BI) of insulin. Ten mM glucose was infused at 3 ml/min, venous glucose was monitored at two minute intervals, and glucose uptake was calculated on the basis of arteriovenous-difference and expressed as micron/min/100 g body wt. Insulin BI given every ten minutes equaled the amount of insulin given by CI for ten minutes. Insulin doses of 1500, 3000, 6000, and 45,000 microU/30 min showed no significant difference between the two modes of delivery in either onset of stimulation or maximal stimulation of glucose uptake. At the lowest insulin dose tested (1500 microU/30 min) neither BI nor CI stimulated glucose uptake above the control of 1.849 micron/min/100 g. A dose response curve for glucose uptake was obtained using insulin boluses ranging from 2000 to 20,000 microU. Insulin uptake by the muscle was always greater when insulin was administered CI. Net disappearance of immunoreactive insulin over the entire 30 minutes of perfusion was 29.4 +/- 2.6% for CI but only 7.1 +/- 1.6% for BI. Thus in the perfused rat hindlimb, stimulation of glucose uptake in skeletal muscle is comparable with BI and CI delivery of insulin but insulin uptake by the muscle is several-fold greater with CI delivery.