Adrenal glands are essential for activation of glucogenesis during undernutrition in fetal sheep near term

Abstract

In adults, the adrenal glands are essential for the metabolic response to stress, but little is known about their role in fetal metabolism. This study examined the effects of adrenalectomizing fetal sheep on glucose and oxygen metabolism in utero in fed conditions and after maternal fasting for 48 h near term. Fetal adrenalectomy (AX) had little effect on the rates of glucose and oxygen metabolism by the fetus or uteroplacental tissues in fed conditions. Endogenous glucose production was negligible in both AX and intact, sham-operated fetuses in fed conditions. Maternal fasting reduced fetal glucose levels and umbilical glucose uptake in both groups of fetuses to a similar extent but activated glucose production only in the intact fetuses. The lack of fasting-induced glucogenesis in AX fetuses was accompanied by falls in fetal glucose ultilization and oxygen consumption not seen in intact controls. The circulating concentrations of cortisol and total catecholamines, and the hepatic glycogen content and activities of key gluconeogenic enzymes, were also less in AX than intact fetuses in fasted animals. Insulin concentrations were also lower in AX than intact fetuses in both nutritional states. Maternal glucose utilization and its distribution between the fetal, uteroplacental, and nonuterine maternal tissues were unaffected by fetal AX in both nutritional states. Ovine fetal adrenal glands, therefore, have little effect on basal rates of fetal glucose and oxygen metabolism but are essential for activating fetal glucogenesis in response to maternal fasting. They may also be involved in regulating insulin sensitivity in utero.

adrenal hormones are essential for survival during stressful conditions after birth (20, 21). They regulate multiple physiological systems and have a key role in glucose homeostasis and blood pressure control, even in normal, basal conditions (21). Postnatal deficiency of adrenocortical and adrenomedullary hormones induced by adrenalectomy leads to profound hypoglycemia and hypotension in several species, including rats and sheep (1, 6, 28, 31, 39). Adrenal insufficiency also impairs the response to common physiological challenges, such as cold exposure, fasting, and exercise, with adverse consequences for morbidity and mortality in the long term (1, 31, 39). However, the role of the adrenal glands in homeostasis before birth, particularly during conditions like undernutrition, is less well established.

Adrenalectomy of the sheep fetus is known to alter development of several fetal tissues during late gestation (29). It prevents the normal decline in fetal growth rate toward term and abolishes several of the morphological and functional changes in tissues, such as the lungs, liver, and gut, that are essential for neonatal survival (11, 14, 18, 38). It also prevents the onset of labor in the ewe (8). Adrenalectomized (AX) fetuses, therefore, tend to weigh more at term and are significantly heavier 10–15 days after normal term than intact, term controls (3, 4, 18). In addition, the normal ontogenic increase in fetal blood pressure is abolished in AX sheep fetuses with the result that fetal blood pressure is lower in AX than intact controls during the 5–10 days before normal term (36, 43). In contrast, fetal metabolite concentrations during the normal and extended periods of gestation appear to be unaffected by adrenalectomy of the sheep fetus (3). There are no significant differences in fetal blood Po₂ or plasma concentrations of glucose, lactate, fructose, urea, or α-amino nitrogen between AX and control fetuses during late gestation (3, 4, 18). However, prepartum deposition of glycogen in ovine fetal tissues, such as the liver, heart, and skeletal muscle, is adversely affected by fetal adrenalectomy and is known to depend on the adrenocortical hormone, cortisol (3, 4). Similarly, activation of glucogenesis in normal well-nourished sheep fetuses during the period immediately before birth is related to the circulating concentrations of both adrenocortical and adrenomedullary hormones (16). However, little is known about the role of adrenal hormones in regulating basal metabolism or the metabolic responses of the fetus to challenges, such as maternal undernutrition, during late gestation. Therefore, this study investigated the effects of adrenalectomizing the sheep fetus on the metabolism of glucose and oxygen by the fetal and uteroplacental tissues in late gestation and during a short period of maternal food deprivation close to normal term.

METHODS

Animals

A total of 10 Welsh Mountain ewes carrying single fetuses were used in this study. During the experimental period, the ewes were housed individually and maintained on concentrate (200 g/day; Beart, Stowbridge, Suffolk, UK) and hay and water ad libitum. One-half the daily ration of concentrates was fed at 0800 while the remainder was given at 1700. Food but not water was withheld for 18–24 h before surgery. All procedures were carried out under the Animals (Scientific Procedures) Act 1986.

Surgical Procedures

Under general anesthesia (1.5% halothane in a 5:1 mixture of O₂ and N₂O₂), fetuses were either adrenalectomized (AX, n = 5) or sham operated (intact controls, n = 5) at 115–118 days of gestation (normal term 145 ± 2 days) and then catheterized at a second operation 10–12 days later using the same anesthetic regime. Catheters were inserted in the uterine vein, umbilical vein, fetal dorsal aorta and caudal vena cava, and into the maternal aorta via a femoral artery of all animals as described previously (17). At the end of the experimental period, ewes and fetuses were given a lethal dose of anesthetic (200 mg/kg pentobarbitone sodium iv) before collection of fetal tissues in the fasted state.

Experimental Procedures

Blood samples (0.5–1.0 ml) were taken from the fetus and mother daily between 0900 and 1000 to monitor well-being. At least 6 days after vascular catheterization (136–138 days), measurements of fetal glucose and oxygen metabolism were made in the fed state in all 10 animals. Following this study, food but not water was withdrawn from the ewes, and a second set of measurements was made in the fasted state at 139–141 days of gestation when the animals had been without food for 48 h. In both fed and fasted animals, tritiated water (8 μCi/ml; Amersham International, Bucks, UK) and universally labeled [¹⁴C]glucose (10 μCi/ml in 0.09% NaCl wt/vol; ICN Biochemicals, High Wycombe, Bucks, UK) were infused together in the fetal caudal vena cava for 2–4 h at known rates between 0.08 and 0.09 ml/min after an initial priming dose (3–4 ml). Blood samples (3.5 ml) were taken simultaneously from the umbilical vein, fetal dorsal aorta, uterine vein, and maternal dorsal aorta before (0 min) and, when steady state had been established, at known times ∼120, 140, 160, and 180 min after beginning the infusion.

The simultaneous blood samples were analyzed immediately for blood pH, gas tensions, packed cell volume, and O₂ content (0.5 ml) and for labeled carbon dioxide (¹⁴CO₂) where appropriate (1.0 ml). The remainder of the sample (2 ml) was added to a chilled tube containing EDTA for subsequent analyses. An aliquot (0.5 ml) of the EDTA-treated blood was deproteinized with zinc sulfate (0.3 M) and barium hydroxide (0.3 M), and the supernatant was used for determination of both labeled and total concentrations of glucose. The remaining EDTA sample was centrifuged at 4°C, and the plasma was stored at −20°C until required for ³H₂O and hormone measurements. An additional aliquot of fetal arterial blood (1 ml) was taken at 0 min and placed in a chilled heparinized tube containing EGTA (5.0 μmol/ml blood) and glutathione (40 μmol/ml blood) for catecholamine assay.

At the end of the second study, the ewes and their fetuses were killed, and the fetuses and uteroplacental tissues were weighed. Samples of liver and kidney (5–10 g) were collected from all fetuses in the fasted state and frozen immediately in liquid nitrogen before storage at −80°C for the subsequent analyses of glycogen content and key gluconeogenic enzyme activities. The adrenal glands were weighed in the intact, control animals while the completeness of adrenalectomy was assessed in the AX fetuses. The position of all catheters was also verified at autopsy. Tissue and plasma samples were also obtained in the fed state from 11 additional fetuses (5 AX and 6 intact), which were killed as part of another study (15, 18) at the same gestational ages as the current cohort of fasted animals (139–141 days). No adrenal remnants were found in any of the AX fetuses.

Biochemical Analyses

The blood gas tensions, packed cell volume, O₂ content, and whole blood concentrations of glucose, [¹⁴C]glucose, ³H₂O, and ¹⁴CO₂ were measured in all five sets of simultaneous samples in the fed and fasted states. Blood O₂ content was calculated from the percentage of O₂ saturation and the hemoglobin concentrations measured using an OSM2 Hemoximeter (Radiometer, Copenhagen, Denmark) that had been calibrated for ovine blood. Blood pH and partial pressures of O₂ and CO₂ were measured using an ABL5 Radiometer and corrected for a fetal body temperature of 39°C.

Glucose concentrations were determined enzymatically in whole blood and plasma using a colorimetry assay (17) and an automated analyzer, respectively (2300 StatPlus; Yellow Springs Instruments, Yellow Springs, OH). Plasma ³H₂O concentrations were measured using scintillation counting and converted to blood concentrations using the packed cell volume as described previously (17, 22). Labeled glucose and CO₂ were determined using chemical methods published previously (22). Labeled glucose was separated from all other ¹⁴C-labeled products by anion exchange chromatography following preincubation with and without glucose oxidase (17, 23). The mean recovery of [¹⁴C]glucose from the anion exchange column was 99.7 ± 1.3% (n = 32). No corrections for glucose recovery were therefore made. In contrast, the mean recovery of ¹⁴CO₂ was 71.9 ± 0.6% (n = 22); hence, all blood ¹⁴CO₂ values have been corrected for recovery.

Plasma catecholamine concentrations were determined by high-pressure liquid chromatography using electrochemical detection (17). Recovery of isoprenaline added to the samples ranged from 63 to 97%; hence, all samples have been corrected for their respective recoveries. The limits of sensitivity of the method were 50 pg/ml for epinephrine and 30 pg/ml for norepinephrine. The interassay coefficients of variation for epinephrine and norepinephrine were 7.3 and 6.2%, respectively. Total catecholamine concentrations were calculated as the sum of the concentrations of epinephrine and norepinephrine in each sample. Plasma concentrations of insulin and cortisol were measured by radioimmunoassay validated for use with ovine plasma (9, 37). The interassay coefficients of variation for these two assays were 13.7 and 10.0%, respectively, while the minimum detectable quantity of hormone was 1.5 ng/ml for cortisol and 5.0 μU/ml for insulin. Only plasma cortisol concentrations were measured in the additional cohort of animals delivered in the fed state at 139–141 days.

Hepatic glycogen content and the activities of glucose-6-phosphatase (G-6-Pase, EC 3.1.3.9) and phosphoenolpyruvate carboxykinase (PEPCK, EC 4.1.1.32) in liver and kidney were assayed using established methods described in detail elsewhere (15, 19). Hormone concentrations and enzyme activities were measured in duplicate while all other biochemical analyses were measured in triplicate.

Calculations

All calculations were made using equations derived for steady-state kinetics (13, 23). Umbilical blood flow was measured using the ³H₂O steady-state diffusion technique and calculated using Eqs. 1 and 2 in Table 1. Net umbilical uptake of glucose and oxygen and net umbilical excretion rates of [¹⁴C]glucose and ¹⁴CO₂ were calculated by the Fick principle as the product of umbilical blood flow (ml/min) and the umbilical venous-arterial (uptake) or arterio-venous (excretion) concentration difference across the umbilical circulation (μmol/l). The net uterine uptake of glucose and oxygen and the uterine output of tracer glucose were measured in a similar manner using uterine blood flow and the corresponding concentration differences across the uterine circulation. Net uteroplacental consumption of glucose and oxygen was calculated as the differences between the uterine and umbilical uptakes.

Table 1.

Equations

Equations
(1) Umbilical blood flow (ml/min) = rate of loss of 3H2O by the umbilical circulation (dpm/min) ÷ umbilical arterio-venous concentration difference in blood 3H2O (dpm/ml).
(2) Uterine blood flow (ml/min) = rate of 3H2O uptake by the uterine circulation (dpm/min) ÷ uterine venous-arterial concentration difference in blood 3H2O (dpm/ml).
(3) Fetal glucose utilization (μmol/min) = net fetal tracer glucose uptake (dpm/min) ÷ fetal arterial glucose specific activity (dpm/μmol glucose)
(4) Net fetal tracer glucose uptake (dpm/min) = tracer glucose infusion rate (dpm/min) − net umbilical tracer glucose excretion rate (dpm/min).
(5) CO2 production from the oxidation of glucose carbon (μmol/min) = net umbilical 14CO2 excretion rate (dpm/min) ÷ fetal arterial glucose specific activity (dpm/μmol glucose carbon).
(6) Glucose carbon oxidation fraction (fraction of fetal glucose carbon utilization used for oxidation) = net umbilical 14CO2 excretion rate (dpm/min) ÷ net fetal tracer glucose infusion uptake (dpm/min).
(7) Fraction of O2 uptake used for oxidation of glucose carbon = amount of O2 used to oxidize fetal glucose carbon (μmol/min) ÷ net umbilical O2 uptake rate (μmol/min).
(8) Endogenous glucose production (μmol/min) = fetal glucose utilization (μmol/min) − umbilical glucose uptake (μmol/min).
(9) Glucose utilization by maternal nonuterine tissues (μmol/min) = net maternal tracer glucose uptake (dpm/min) ÷ maternal arterial specific activity (dpm/μmol).
(10) Total maternal glucose utilization (μmol/min) = net uterine glucose uptake (μmol/min) + glucose utilization by maternal nonuterine tissues (μmol/min).

The fetal rates of utilization and production of glucose, CO₂ production from glucose carbon, and the fraction of the net umbilical O₂ uptake used for glucose carbon oxidation by the fetus were calculated as shown in Table 1 (13, 17, 23). Oxidation of glucose carbon was measured as the rate of ¹⁴CO₂ production (Table 1). The glucose carbon oxidation fraction and the fraction of O₂ consumption used to oxidize glucose carbon in the fetus were then calculated where the amount of O₂ used to oxidize fetal glucose carbon is equal to the amount of CO₂ produced by this oxidative process (Table 1). Endogenous glucose production by the fetus was calculated from the rates of umbilical uptake and fetal utilization of glucose (Table 1). Finally, glucose utilization by nonuterine maternal tissues was calculated in the ewes by measuring the labeled and unlabeled glucose concentrations in maternal arterial and uterine venous blood (Eqs. 9 and 10 in Table 1). Because glucose uptake from the gastrointestinal tract is negligible in sheep (5), the total rate of glucose utilization by the maternal tissues is equivalent to the rate of glucose production by the ewe. Because studies were carried out in both the fed and fasted states, the values for [¹⁴C]glucose and ¹⁴CO₂ in the 0 min arterial, uterine, and umbilical venous samples of the fasted study were subtracted from the subsequent samples before calculation of the glucose metabolic rates. All fetal metabolic rates have been expressed per kilogram fetal body weight. No increase in fetal body weight was assumed to occur during the 48-h period of maternal food withdrawal.

Statistical Analyses

Steady state was defined as <10% variation of values around the mean for each sampling period, with no consistent trend for the absolute values to increase or decrease with time. Mean values ± SE have been used throughout. Statistical analyses were made using Sigmastat (Jandel Scientific, Chicago, IL). Comparison of metabolic rates between treatments and nutritional states were made using two-way ANOVA and unpaired and paired t-tests, as appropriate. For all statistical analyses, significance was accepted when P < 0.05.

RESULTS

Effects of Fetal Adrenalectomy on Tissue Biometry

Fetal adrenalectomy had no effect on the body weight, crown rump length, or ponderal index of the fetuses at delivery at 139–141 days of gestation (Table 2). There was also no significant difference in the total weight of the uteroplacental tissues or of the placentomes alone between AX and intact, sham-operated fetuses at delivery (Table 2). No adrenal remnants were found in any of the AX fetuses (Table 2). Adrenal weight of the sham-operated controls (Table 1) was similar to that seen previously in intact fetuses at the same gestational age (3).

Table 2.

CRL, ponderal index, and weights of the total uteroplacental tissues, placentomes, fetus, and adrenals of intact and adrenalectomized sheep fetuses at delivery at 139–141 days of gestation

	Weight
	Uteroplacental tissues, g*	Placentomes, g	Fetus, g	Adrenals, mg	CRL, cm	Ponderal Index, kg/m3
Intact	1,641 ± 199	262 ± 23	3,435 ± 242	425 ± 49	50.6 ± 0.7	26.4 ± 1.5
Adrenalectomized	1,341 ± 135	310 ± 30	3,532 ± 277	ND	50.8 ± 1.8	27.0 ± 1.9

Values are means ± SE; n = 5 animals in each group. CRL, crown rump length.

^*Combined weight of uterus, placentomes, and fetal membranes. ND, not detected.

Effects of Fetal Adrenalectomy on Metabolite and Hormone Concentrations

Fetal adrenalectomy also had no effect on the blood gas status of the fetus in either the fed or fasted state: mean values of blood pH, Po₂, Pco₂, O₂ saturation, and O₂ content were not significantly different between AX and intact fetuses in either nutritional state (Table 3). In addition, there was no change in fetal blood gas status with maternal food withdrawal for 48 h in either group of animals (Table 3). Glucose concentrations in maternal and fetal blood were not significantly different between AX and intact animals in either the fed or fasted states and decreased to a similar extent during the 48-h period of fasting (Table 4). The transplacental concentration gradient in plasma glucose was similar in AX and intact fetuses irrespective of nutritional state and fell to the same extent during maternal fasting in the two groups of animals (Table 3).

Table 3.

Blood gas status

	Intact		Adrenalectomized
	Fed	Fasted	Fed	Fasted
pH	7.374 ± 0.017	7.366 ± 0.018	7.340 ± 0.013	7.332 ± 0.008
Po2, mmHg	21.7 ± 1.6	20.4 ± 0.3	20.2 ± 0.7	21.9 ± 0.7
Pco2, mmHg	50.7 ± 1.9	47.1 ± 1.4	53.9 ± 2.3	51.8 ± 1.1
O2 saturation, %	60.1 ± 2.2	62.0 ± 2.5	54.6 ± 2.9	55.6 ± 1.5
O2 content, mmol/l	3.75 ± 0.22	3.92 ± 0.20	3.32 ± 0.20	3.62 ± 0.24
Hemoglobin, g/100 g	10.1 ± 0.5	10.2 ± 0.4	10.0 ± 0.5	10.5 ± 0.5
PCV, %	33.5 ± 1.4	34.0 ± 1.8	31.1 ± 1.3	34.1 ± 2.0

Values are means ± SE; n = 5 animals in each group. Shown are values of pH, Po₂, Pco₂, O₂ saturation, O₂ content, hemoglobin content, and packed cell volume (PCV) in arterial blood of intact and adrenalectomized sheep fetuses in fed conditions at 136–138 days and in the fasted state after maternal food withdrawal for 48 h at 139–141 days in the same animals.

Table 4.

Metabolite and hormone concentrations

	Intact		Adrenalectomized
	Fed	Fasted	Fed	Fasted
Blood glucose, mmol/l
Fetus	0.86 ± 0.08	0.53 ± 0.04*	0.70 ± 0.10	0.36 ± 0.09*
Mother	2.34 ± 0.13	1.18 ± 0.12*	2.17 ± 0.11	1.22 ± 0.12*
Plasma glucose, mmol/l
Fetus	0.99 ± 0.09	0.64 ± 0.05*	0.88 ± 0.11	0.46 ± 0.10*
Mother	3.07 ± 0.12	1.83 ± 0.13*	2.87 ± 0.15	1.68 ± 0.18*
Transplacental	2.08 ± 0.16	1.19 ± 0.10*	2.02 ± 0.20	1.23 ± 0.13*
Cortisol, ng/ml	12.8 ± 2.7	41.0 ± 9.9*	7.7 ± 1.0	7.7 ± 0.9†
Insulin, μU/ml	22.6 ± 2.4	14.0 ± 1.6*	10.9 ± 1.5†	7.2 ± 0.7*†
Epinephrine, pg/ml	42 ± 24	62 ± 32	30 ± 9	34 ± 14
Norepinephrine, pg/ml	196 ± 54	864 ± 211*	260 ± 43	428 ± 41
Total catecholamines, pg/ml	258 ± 74	928 ± 185*	290 ± 42	462 ±55*†

Values are means ± SE; n = 5 animals in each group. Shown are concentrations of glucose in maternal and fetal arterial blood and plasma and of cortisol, insulin, epinephrine, norepinephrine, and total catecholamines in fetal arterial plasma together with the transplacental plasma glucose concentration gradient in intact and adrenalectomized fetuses in fed conditions at 136–138 days and after maternal fasting for 48 h at 139–141 days in the same animals.

^*Significantly different from the value in the fed state in the same group of fetuses (P < 0.05, paired t-test).

^†Significantly different from the value in intact fetuses in the same nutritional state (P < 0.05, ANOVA).

Plasma cortisol concentrations were not significantly different between AX and intact fetuses in the fed state at 136–138 days but were significantly lower in AX than intact fetuses at 139–141 days after maternal fasting for 48 h (Table 4). The plasma cortisol concentration in the additional cohort of intact fetuses delivered in the fed state at 139–141 days (25.5 ± 4.6 ng/ml, n = 6) was lower than the value found in the current cohort of animals sampled in the fasted state at the same gestational age, but not significantly so (P > 0.05, Table 4). There was also no significant difference between the plasma cortisol concentration in the AX fetuses of the fed (6.8 ± 0.7 ng/ml, n = 5) and fasted cohorts at 139–141 days (P > 0.05, Table 4).

In the current cohort of animals, plasma insulin concentrations were significantly lower in AX than intact fetuses in both the fed and fasted states and were reduced significantly by maternal fasting in both groups of animals (Table 4). The mean decrement in fetal plasma insulin during maternal fasting was similar in AX (−3.7 ± 1.0 μU/ml, n = 5) and intact (−7.1 ± 2.1 μU/ml, n = 5, P > 0.05) fetuses. There were no significant differences in the plasma concentrations of epinephrine and norepinephrine between AX and intact fetuses in either nutritional state (Table 4). Plasma epinephrine concentrations were unaffected by maternal fasting in both groups of fetuses, whereas plasma norepinephrine concentrations were higher in the fasted than fed state in intact but not AX fetuses (Table 4). Total catecholamine concentrations were similar in the two groups of fetuses in the fed state and, although higher after 48 h of fasting in both groups, mean values in the fasted state were significantly lower in AX than intact fetuses (Table 4).

Effects of Fetal adrenalectomy on Glucose and Oxygen Metabolism by the Fetus

Fed state.

Fetal adrenalectomy had no significant effect on the basal rates of umbilical blood flow and O₂ uptake; mean values were similar in the two groups of fetuses in fed conditions (Table 5). There were also no significant differences in the rates of umbilical glucose uptake, glucose utilization, or CO₂ production from glucose carbon between AX and intact fetuses in the fed state (Fig. 1). Neither was the glucose oxidation fraction nor were the fraction of umbilical O₂ uptake used for glucose carbon oxidation significantly different between the two groups of fetuses (Table 5). Endogenous glucose production was negligible in both groups of fetuses in the fed state (Fig. 1C).

Table 5.

Oxygen metabolism

	Intact		Adrenalectomized
	Fed	Fasted	Fed	Fasted
Blood flow
Umbilical
ml/min	592 ± 38	570 ± 47	617 ± 82	541 ± 75
ml·min−1·kg−1‡	174 ± 13	170 ± 21	174 ± 19	161 ± 11
Uterine
ml/min	1,224 ± 151	1,206 ± 119	1,472 ± 182	1,310 ± 11
ml·min−1·kg−1†	796 ± 129	794 ± 136	1,140 ± 251	1,082 ± 211
Umbilical O2 uptake
Rate, μmol·min−1·kg−1‡	266 ± 10	255 ± 27	297 ± 19	249 ± 16*
Fraction	0.294 ± 0.023	0.179 ± 0.034*	0.228 ± 0.023	0.123 ± 0.023*
Glucose oxidation fraction	0.469 ± 0.077	0.398 ± 0.039	0.405 ± 0.031	0.393 ± 0.053
Total O2 uptake, ml/min
Uterus	1,344 ± 41	1,221 ± 51	1,611 ± 174	1,660 ± 132
Uteroplacental tissues	591 ± 104	418 ± 100	560 ± 180	672 ± 44
Fetus	842 ± 56	803 ± 92	1,158 ± 132	864 ± 114*
Uterine O2 uptake, %
Uteroplacental tissues	37.1 ± 3.7	33.8 ± 5.9	33.4 ± 8.1	44.3 ± 2.3
Fetus	62.9 ± 3.4	66.2 ± 6.8	66.6 ± 8.9	55.6 ± 2.6

Values are means ± SE; n = 5 animals in each group. Shown are values of absolute and weight-specific umbilical and uterine blood flow, weight-specific umbilical oxygen uptake, the fraction of umbilical oxygen uptake used for glucose carbon oxidation by the fetus, the glucose oxidation fraction, the total uterine, uteroplacental, and fetal uptake of oxygen, and the distribution of the total uterine oxygen uptake between the uteroplacental and fetal tissues in intact and adrenalectomized fetuses in fed conditions at 136–138 days and after maternal fasting for 48 h at 139–141 days in the same animals.

^*Significantly different from the value in the fed state in the same group of fetuses (P < 0.05, paired t-test).

^‡Per kg fetal body weight.

^†Per kg total weight of uterus, placentomes, and fetal membranes.

Fig. 1.

Mean ± SE rates of umbilical glucose uptake (A), glucose utilization (B), endogenous glucose production (C), and CO₂ production (D) from glucose carbon in fed conditions (open bars) and after maternal fasting for 48 h (gray bars) and the mean change in rate between the fed and fasted states (black bars) in intact and adrenalectomized (AX) fetuses in late gestation (n = 5 in each group). †Significant change between the fed and fasted states (P < 0.05, paired t-test). *Significant rate of endogenous glucose production (P < 0.05, t-test).

Fasted state.

Maternal fasting for 48 h significantly reduced the rate of umbilical glucose uptake in both AX and intact fetuses; mean decrements were similar in the two groups (Fig. 1A). In AX but not intact fetuses, this occurred in parallel with a significant reduction in the rate of glucose utilization (Fig. 1B). The rate of endogenous glucose production, therefore, remained negligible in AX fetuses but occurred at a significant rate in intact fetuses after 48 h of maternal fasting (Fig. 1C). The rate of CO₂ production from glucose carbon decreased to a similar extent during fasting in both groups of fetuses (Fig. 1D). This was accompanied by a reduction in the fraction of the umbilical O₂ uptake used for oxidation of glucose carbon in both groups of fetuses and by a decreased rate of umbilical O₂ uptake in the AX but not intact fetuses (Table 5). During fasting, the fall in the weight-specific rate of oxygen consumption was significant in AX fetuses (−48 ± 16 μmol·min⁻¹·kg⁻¹, n = 5, P < 0.05) but not in intact controls (−11 ± 28 μmol·min⁻¹·kg⁻¹, n = 5, P > 0.05). In neither group of fetuses was the glucose oxidation fraction affected by maternal fasting (Table 5) nor did umbilical blood flow alter with maternal food withdrawal (Table 5). Hepatic glycogen content and hepatic, but not renal, activities of G-6-Pase and PEPCK were significantly lower in AX than intact fetuses in the fasted state at delivery at 139–141 days of gestation (Fig. 2). Hepatic glycogen content was lower while hepatic but not renal activities of G-6-Pase and PEPCK were higher in the fasted than fed state in both groups of fetuses at 139–141 days (Fig. 2).

Fig. 2.

Mean ± SE values of glycogen content and activities of glucose-6-phosphatase (G-6-Pase) and phosphoenolpyruvate carboxykinase (PEPCK) in the liver and kidney of intact (open bars) and AX (hatched bars) fetuses in the fed state (intact, n = 6 and AX = 5) and after 48 h of maternal fasting at 139–141 days of gestation (n = 5 in both groups). *Significantly different from the corresponding value in the intact fetuses (P < 0.05, Student's t-test). †Significantly different from the corresponding value in the same group of fetuses in the fed state (P < 0.05, Student's t-test).

Effects of Fetal Adrenalectomy on Oxygen and Glucose Metabolism by the Uteroplacental Tissues and Nonuterine Maternal Tissues

Compared with intact controls, fetal adrenalectomy had no apparent effect on uterine blood flow or the uptake of oxygen by the uterus and uteroplacental tissues; mean rates were similar in the two groups of animals, irrespective of nutritional state (Table 5). There were also no significant differences in the rates of glucose uptake by the uterus and uteroplacental tissues between AX and intact fetuses in either nutritional state (Fig. 3). Total maternal glucose utilization and its percentage distribution between the fetal, uteroplacental, and nonuterine maternal tissues were also similar in ewes with AX and intact fetuses in both nutritional states (Fig. 3). Fasting had no effect on uterine blood flow or oxygen uptake in either group of animals (Table 5). There was a trend for increased uteroplacental O₂ consumption in AX but not intact fetuses in response to maternal fasting, but this did not reach statistical significance (Table 5). In contrast, fasting reduced total maternal glucose utilization and glucose uptake by the nonuterine maternal tissues to a similar extent in the two groups of animals (Fig. 3). Uteroplacental glucose consumption also decreased significantly with fasting in the intact but not AX fetuses (Fig. 3). There was no change in the percentage distribution of maternal glucose production between the fetal, uteroplacental, and nonuterine maternal tissues with maternal food withdrawal for 48 h in either group of animals (Fig. 3).

Fig. 3.

Mean ± SE rates of total maternal glucose utilization (whole bar) and its distribution between the nonuterine maternal tissues (gray bar), the uteroplacental tissues (stippled bar), and the fetal tissues (open bar) in ewes with intact and AX fetuses in fed conditions and after fasting for 48 h. The figures within the bars show the mean percentage (±SE) distribution of the total rate of maternal utilization set at 100% between the different tissues. *Significant difference in the absolute rate of glucose utilization by the whole ewe (above column) or the specific tissue (to the right of the column) from the values seen in the fed state (P < 0.05, paired t-test).

DISCUSSION

The results demonstrate that the fetal adrenal glands have little apparent effect on the basal rates of fetal glucose and oxygen metabolism in well-nourished ewes but are essential for activation of fetal glucogenesis in response to short-term maternal fasting close to term. This inability of AX sheep fetuses to induce glucogenesis during undernutrition was associated with lower circulating concentrations of cortisol and total catecholamines and with reductions in the hepatic glycogen content and activities of key gluconeogenic enzymes, relative to intact, sham-operated controls. It was also accompanied by a reduced rate of oxygen consumption. Insulin concentrations were also lower in AX than intact fetuses in both the fed and fasted states, despite the similarity in fetal glycemia in the two groups of animals. These observations show that, like the adult adrenals, the fetal adrenal glands have an important role in the metabolic response to stressful conditions in utero. In addition, they suggest that adrenal secretions may be involved in regulating fetal insulin sensitivity, even in well-nourished conditions during late gestation.

The lack of glucogenesis in response to maternal fasting in AX fetuses may be due to the combined deficiencies of adrenocortical and adrenomedullary hormones, since both cortisol and total catecholamine concentrations were low in AX relative to intact fetuses after maternal fasting for 48 h. Cortisol is known to activate glucose production by the liver in intact fetuses near term (42). It also increases the circulating concentration of gluconeogenic amino acids, such as alanine, in fetal sheep during short-term infusion (33, 41). In the longer term, cortisol has also been shown to enhance glycogen deposition in several tissues of intact and AX fetuses during late gestation and to increase the activities of G-6-Pase and PEPCK in the liver and kidney of intact fetuses during the last 10–15 days of gestation (4, 15). Certainly, the low hepatic activities of G-6-Pase and PEPCK are likely to be an important contributory factor to the lack of endogenous glucose production by the AX fetus after maternal fasting for 48 h. Similar defects in fasting-induced glucogenesis associated with low hepatic glycogen and gluconeogenic enzyme activities have been observed in late gestation in sheep fetuses made cortisol deficient by fetal hypophysectomy (12). However, the finding that hepatic G-6-Pase and PEPCK activities were higher in the fasted than fed state in both AX and intact fetuses suggests that the increasing level of cortisol is not the only factor activating gluconeogenic enzymes in the ovine fetal liver.

Like cortisol, epinephrine and norepinephrine have been shown to stimulate hepatic glucose output in intact sheep fetuses during short-term infusions close to term (2). At high doses, catecholamine infusions also cause fetal hyperglycemia, although this only occurs at catechholamine concentrations higher than those seen in the current study (35). Basal levels of catecholamines were unaffected by fetal adrenalectomy and were within the range of values reported previously for AX and intact fetuses (7, 16, 35, 40). Similarly, basal catecholamine concentrations were unaffected by short-term chemical sympathectomy and by more long-term adrenal denervation or demedullation of fetal sheep (25, 34). Collectively, these observations suggest that the circulating catecholamine and norepinephrine, in particular, can also be derived from the peripheral nervous system and/or the para-aortic and other extra-adrenal chromaffin tissue. Fetal adrenalectomy reduced but did not completely abolish the catecholaminergic response to maternal fasting in the present study, a finding in common with the response of AX sheep fetuses to hypoxia reported previously (40). In intact fetuses of well-fed ewes during late gestation, significant glucogenesis is observed at total fetal catecholamine concentrations similar to those seen in the AX fetuses after maternal fasting for 48 h (16). Fetal cortisol concentrations, therefore, appear critical in determining fetal glucogenesis. Indeed, previous studies of intact fetuses during fed conditions have shown that endogenous glucose production only rises with physiological increases in the total catecholamine concentration when fetal cortisol concentrations exceed 18 ng/ml (16), a value well above that seen in the AX fetuses in the present study in either the fed or fasted state. In part, the dependence of catecholamine-induced glucogenesis on the cortisol level is due to the cortisol-induced upregulation of hepatic gluconeogenic enzyme activities, but it may also reflect the action of cortisol in increasing β-adrenoreceptor abundance in the fetal liver near term (2, 10). Taken together, these observations suggest than cortisol deficiency may be the primary cause of the absence of glucogenesis in response to maternal fasting after AX of the sheep fetus.

This lack of fasting-induced glucogenesis in the AX fetus led to a significant reduction in its rate of glucose utilization as its umbilical supply of glucose declined. In contrast, in intact fetuses, the decreased glucose supply during maternal fasting was ameliorated, in part, by the endogenous production of glucose and, hence, there was no significant reduction in the rate of fetal glucose utilization. Oxidation of glucose carbon decreased during fasting to the same extent in both AX and intact fetuses as did fetal glycemia, consistent with previous observations that the glucose oxidation is determined primarily by the fetal glucose level (22). Despite the decrease in glucose oxidation, oxygen consumption by the intact fetuses was unaffected by maternal fasting so other substrates must be oxidized in increased amounts to account for the normal rate of oxidative metabolism. A number of previous studies have shown that amino acids act as alternative oxidative substrates when glucose availability is limited in the sheep fetus (26, 30). In the AX fetuses, oxidative metabolism fell by 15–10% with the decreases in glucose utilization and oxidation during maternal fasting. Consequently, recruitment of alternative substrates for oxidative metabolism during fetal hypoglycemia may be compromised by deficiency of adrenal hormones in keeping with the known role of cortisol in stimulating protein catabolism and leucine oxidation in fetal sheep during late gestation (32). Thus the low cortisol levels in AX fetuses after 48 h of maternal fasting may not only limit upregulation of the glucogenic pathways but also impair the provision of amino acids for fetal gluconeogenesis and oxidative metabolism. The maintenance of glucose uptake and the tendency for increased O₂ consumption by the uteroplacental tissues of AX fetuses during fasting may also reflect the low fetal cortisol levels or, alternatively, the differences in frequency distribution of the different placentome types between AX and intact fetuses at this stage of gestation (44).

Insulin concentrations were low in AX compared with intact fetuses in both the fed and fasted states, despite no significant differences in fetal glycemia. Low insulin levels have also been observed in sheep fetuses after specific demedullation of the adrenal, which suggests that deficiency of the adrenomedullary secretions rather than cortisol may be the more important factor in determining the hypoinsulinemia seen after complete fetal adrenalectomy (25). These secretions are unlikely to be the catecholamines, since epinephrine and norepinephrine suppress, not enhance, fetal insulin secretion, and their basal concentrations were unaltered in the AX fetuses (2, 9, 27). Despite their low insulin levels, AX fetuses maintained rates of glucose metabolism within the normal range in fed conditions. Because insulin is known to regulate glucose utilization in fetal sheep by enhancing glucose uptake in fetal tissues (13), the current findings suggest that either tissue sensitivity to insulin must be increased or there are other insulin-independent mechanisms of stimulating glucose utilization in the AX fetus. Preliminary observations of the insulin-signaling pathways in skeletal muscle from AX fetuses in the fed state indicate that there are no changes in the abundance of the insulin-sensitive glucose transporter, GLUT4, or of the receptors for insulin and insulin-like growth factor I, relative to the values in intact fetuses (24). Further studies are, therefore, needed to determine whether the insulin sensitivity of fetal glucose metabolism is altered during late gestation by removal of the fetal adrenal glands.

In summary, fetal adrenalectomy altered the fetal metabolic response to maternal fasting but had no apparent effect on the basal rates of fetal glucose and oxygen metabolism in the fed state, despite concomitant hypoinsulinemia. The changes in the fetal metabolic response to fasting induced by fetal adrenalectomy were closely related to the low cortisol and catecholamine levels and were not accompanied by any major differences in uteroplacental metabolism of glucose and oxygen or in the distribution of the uterine uptake of these substances between the fetal and uteroplacental tissues. The poor glucogenic capacity of the AX fetus will limit its ability to withstand hypoglycemia and other stressful challenges during late gestation. Adrenal hormones are, therefore, essential for the adaptive changes in fetal metabolism that ensure survival during adverse nutritional and other conditions in utero.

GRANTS

We are indebted to the Wellcome Trust and the Biotechnology and Biological Sciences Research Council for financial support.

DISCLOSURES

No conflicts of interest are declared by the authors.