Main

During late gestation, the fetal pituitary-adrenal axis plays a major role in the response of the fetus to acute stress and in the timing of parturition in the sheep (term = 145 ± 3 d of gestation)(1). Acute episodes of fetal hypoxemia are associated with increased mRNA levels of corticotrophin releasing hormone mRNA in the fetal hypothalamus, and of POMC, the ACTH precursor, in the fetal pituitary and in increased circulating ACTH and cortisol concentrations(2–5) Fetal plasma concentrations of ir ACTH and cortisol also increase consistently in response to acute episodes of hemorrhage and hypoglycemia during late pregnancy(4, 6).

Although parturition in the sheep is dependent on the prepartum increase in fetal plasma cortisol, it is less clear whether the capacity of the fetal pituitary to synthesize and secrete ACTH alters before delivery. There are conflicting reports of an increase(7–9) or a decrease(10, 11) in the levels of POMC mRNA in the fetal sheep pituitary during the 10 d preceding delivery. These differences may in part be a consequence of the different gestational ages surveyed and the limited number of pituitaries collected at each gestational age range in these separate studies.

ACTH is secreted in a range of precursor forms from the fetal pituitary(12) and is present in the fetal circulation in high(POMC or Pro-ACTH) and low (ACTH 1-39) molecular weight forms(13, 14). Although the concentrations of ir ACTH in the fetal circulation have been shown to increase or remain unchanged(3, 15, 16) during late gestation, the anterior lobe of the fetal sheep pituitary secretes relatively more bioactive ACTH 1-39 in the week before birth, at a time when the fetal adrenal output of cortisol is increasing(12). The factors involved in triggering the activation of the pituitary adrenal axis before normal delivery are unknown. However, it is known that the onset of parturition is accelerated by chronic stress in utero(1). There are some limited data from cordocentesis studies in human pregnancies and from experimental studies of placental restriction in the sheep, which show that placental insufficiency is associated with chronic fetal hypoxemia, hypoglycemia, and increased fetal cortisol concentrations(17, 18). We have therefore hypothesized that chronic stress in late gestation accelerates the normal ontogenic changes in the ACTH synthetic capacity of the fetal pituitary and circulating ir ACTH, ACTH 1-39, and cortisol concentrations in the 10-15 d preceding delivery. We have experimentally restricted placental growth and hence substrate delivery to the fetus and determined the effect on fetal pituitary POMC mRNA levels and on circulating ir ACTH, ACTH 1-39, and cortisol concentrations during late gestation.

METHODS

Animals and Surgery

All procedures were approved by the Adelaide University Standing Committee on Ethics in Animal Experimentation. Nineteen pregnant merino ewes were used in this study. In nine ewes endometrial caruncles were removed before conception as previously described(18) to reduce the number of placental cotyledons, thus restricting placental and fetal size. All surgery was performed under aseptic conditions. General anesthesia was induced by an i.v. injection of sodium thiopentone (1.25 g, Boehringer lngelheim, NSW, Australia) and maintained with 3-4% halothane in oxygen. The ewes were kept under observation for 4-7 d postsurgery, then returned to the farm (PR group). After a minimum of 10 wk postsurgery, the ewes entered a mating program, and pregnancies were confirmed by ultrasound at approximately 60 d of gestation. At between 102 and 112 d of gestation, catheters were placed in the fetal and maternal carotid artery and jugular vein, and in the amniotic sac of PR ewes(n = 9) and control ewes (n = 10). Fetal catheters were exteriorized through an incision in the ewe's flank, and all catheters were filled with heparinized saline. All sheep received a 2-mL intramuscular injection of antibiotics (procaine penicillin, 250 mg/mL; dihydrostreptomycin, 250 mg/mL; procaine hydrochloride, 20 mg/mL) after the induction of anesthesia. The ewes were housed in individual pens in animal holding rooms, with a 12-h light/dark lighting regimen, and fed once daily between 0900 and 1300 h with water ad libitum.

Experimental Protocols

Fetal blood sampling. Fetal arterial (3.5 mL) and maternal carotid artery (5 mL) blood samples were collected into heparinized tubes at least three times per week from 104 to 140 d of gestation. Plasma samples were prepared from fetal arterial blood by centrifugation at 1500 ×g for 10 min at 4 °C and stored at -20 °C for assay of ACTH, cortisol, and glucose.

Additional arterial blood (0.5 mL) was collected from the fetus and ewe, and whole blood pH, Po2, Pco2, oxygen saturation, and hemoglobin content were measured using an ABL 550 acid base analyzer and the OSM2 hemoximeter (Radiometer, Copenhagan, Denmark).

Pituitary collection. At 140 or 141 d of gestation, all ewes were killed by i.v. overdose of sodium pentobarbitone. Fetal sheep were delivered by hysterotomy and weighed, and the fetal adrenal and pituitary glands were rapidly dissected and weighed. In six fetal sheep in the PR group and in six control fetal sheep, the anterior lobe of the pituitary was dissected free from the neurointermediate lobe, and the anterior lobes were snap frozen and stored at -80 °C before extraction of the total RNA.

Measurement of POMC mRNA. A 400-bp EcoRI fragment of ovine POMC cDNA was radiolabeled using [32P]deoxycytidine triphosphate(3000 Ci/mmol, Bresatec Pty, Ltd., Adelaide, SA, Australia) and a random priming kit (Pharmacia, North Ryde, NSW, Australia) to a specific activity of 109 cpm/μg or greater. An antisense 30-base oligomer to part of the 18 S rRNA was end-labeled with [γ-32P]ATP (4000 Ci/mmol, Bresatec) by T4 polynucleotide kinase (Pharmacia).

Total RNA was extracted from six anterior pituitaries in the PR group, and six anterior pituitaries in the control group. Pituitaries were homogenized in 4 M guanidine hydrochloride, and after ultracentrifugation through a 5.7 M cesium chloride cushion, the purity and concentration of the nucleic acids were quantitated by spectrophotometric measurement as described previously(10, 11).

Pituitary total RNA samples (10 μg/lane) were separated on a 1% agarose gel containing 2.2 M formaldehyde, and Northern blot analysis(19) was used to quantitate the abundance of POMC mRNA in pituitary total RNA. After hybridization with DNA probes, membranes were exposed to Fuji phosphorimage plates and Kodak X-AR film. The POMC probe was then stripped from the membranes by washing in 0.1 × SSC (1 × SSC: 0.15 M NaCl, 15 mM sodium citrate, pH 7.0), 0.5% SDS for 10 min at 80 °C. The consistency of lane loading was verified by hybridization with radiolabeled 18 S rRNA oligo probe. The signals present on the phosphorimage plates were quantitated on a Fuji-BAS 1000 phosphorimager machine.

Glucose measurement. Plasma glucose concentrations were determined by enzymatic analysis using hexokinase and glucose-6-phosphate dehydrogenase and measuring the formation of NADH photometrically at 340 nm(COBAS MIRA automated analyzing system). The intraassay coefficient of variation was 1.4%.

ACTH RIA. The ir ACTH concentrations in fetal sheep plasma (PR group, n = 40 samples; control group, n = 66 samples) were measured by RIA using an ICN Biomedicals kit (ICN Biomedicals, Seven Hills, NSW, Australia). This assay measures ACTH 1-39 and other molecular weight forms of ACTH derived from the ACTH precursor, POMC. The sensitivity of the assay was 7 pg/mL, and the rabbit anti-human ACTH 1-39 had a cross-reactivity of <0.1% with β-endorphin, α-melanocyte-stimulating hormone,α-lipotrophin, and β-lipotrophin. The interassay coefficient of variation was 14.6%, and the intraassay coefficient of variation was<10%.

ACTH 1-39 immunoradiometric assay. ACTH 1-39 was measured in a total of 20 plasma samples collected from five fetal sheep in each of the PR and control groups at two gestational age ranges (124-128 d and 134-141 d). ACTH 1-39 was measured using an immunoradiometric assay which uses one radiolabeled MAb specific for the sequence ACTH 10-18 and a second MAb which is coupled to Sephacryl S-300 as a solid phase and is specific for the sequence ACTH 25-39. The assay cross-reacts <1% with POMC and Pro-ACTH at the concentrations measured, and there is no cross-reaction with fragments of ACTH(20).

Cortisol RIA. Cortisol was measured in a total of 117 plasma samples (PR, n = 46; control, n = 71). Total cortisol concentrations were measured in fetal sheep plasma by RIA using an Orion Diagnostica kit (Orion Diagnostica, Turku, Finland). Before assay, cortisol was extracted from the plasma using dichloromethane as previously described(21). The efficiency of recovery of 125I-cortisol from fetal plasma using this extraction procedure was at least 90%. The sensitivity of the assay was 0.39 nmol/L, and the cross-reactivity of the rabbit anti-cortisol was <1% with cortisone and 17-hydroxyprogesterone, and<0.01% with pregnenolone, aldosterone, progesterone, and estradiol.

Statistical Analyses

Fetal morphometry, blood gas, and hormonal data are given as mean ± SEM. The fetal body, adrenal, and pituitary weights in the PR and control groups were compared using an unpaired t test. There were 181 arterial blood samples (PR; n = 87 in 9 sheep: control; n= 94 in 10 sheep) collected for the measurement of blood gas variables. Arterial Po2 and plasma glucose concentrations for each fetus were grouped in 5-d age blocks between 105 and 140 d. These values were then compared in the two experimental groups across gestation using a two-way analysis of variance with repeated measures. Mean gestational values for arterial Pco2, pH, O2 saturation, and hemoglobin concentration(g/dL) were calculated for each fetus in the PR and control groups between 104 and 140 d of gestation and compared using an unpaired t test. An unpaired t test was also used to compare the relative abundance of anterior pituitary POMC mRNA:18 S RNA in the PR and control groups at 140 d of gestation.

The hormonal data were logarithmically transformed when necessary to reduce heterogeneity of variance. Plasma ir ACTH, ACTH 1-39, and cortisol concentrations were compared using two-way analysis of variance with repeated measures with treatment (PR versus control) and gestational age as the specified factors. When there was significant interaction between treatment and gestational age for a measured parameter, the data were split on the basis of the interacting factor and reanalyzed. Duncan's multiple range test was used to identify differences between mean values. A probability of less than 5% (i.e. p < 0.05) was taken to be significant for all statistical analyses.

RESULTS

Fetal weights, plasma glucose, and arterial blood gas status. Surgical removal of the majority of endometrial caruncles decreased(p < 0.0001) placental weight (PR group, 171 ± 31 g; control group, 534 ± 46 g) and fetal body weight (p < 0.001) at 140 d of gestation (PR group, 2.98 ± 0.27 kg; control group, 4.46 ± 0.24 kg). The combined adrenal weight was not different between the PR group (343 ± 25 mg, n = 9) and the control group (356± 18 mg, n = 9). Combined adrenal weight expressed in relation to fetal body weight was, however, higher (p < 0.005) in the PR group (130 ± 10 mg/kg) than in the controls (80 ± 1 mg/kg). In contrast, fetal pituitary weight was lower in the PR group (80± 10 mg, n = 8) than in the control group (130 ± 10 mg, n = 9), whereas fetal pituitary weight expressed in relation to body weight was not different between the two groups (PR and control groups; 30 ± 1 mg/kg).

Fetal arterial Po2 and O2 saturation were also reduced(p < 0.001) in the PR group (2.0 ± 0.1 kPa and 42.8± 1.1%) when compared with control animals (3.1 ± 0.1 kPa and 66.4 ± 0.9%, respectively). The arterial hemoglobin concentration and Pco2 were significantly elevated (p < 0.05) in the PR fetal sheep (11.8 ± 0.2 g/dL and 6.7 ± 0.1 kPa) compared with the control group (10.1 ± 0.1 g/dL and 6.2 ± 0.1 kPa, respectively). There was no difference, however, in the fetal arterial pH across the gestational range of the study between the two groups (PR and controls, 7.36 ± 0.01). Plasma glucose concentrations were lower(p < 0.04) in the PR group (1.15 ± 0.12 mmol/L,n = 9) than in control fetal sheep (1.44 ± 0.11 mmol/L,n = 10) throughout late gestation.

Fetal arterial Po2 and plasma glucose concentrations decreased(p < 0.001) with increasing gestational age in both the PR and control groups and were lower at all ages after 117 d when compared with before 116 d of gestation (Fig. 1).

Figure 1
figure 1

Fetal arterial Po2 (top panel), plasma glucose (middle panel), and cortisol (bottom panel) concentrations (mean ± SEM) in control (clear histograms) and PR (solid histograms) groups between 112 and 141 d of gestation.

Full size image

Anterior pituitary POMC mRNA levels. A single mRNA transcript of 1.1 kb hybridized with the POMC cDNA probe (Fig. 2). The ratio of POMC mRNA:18 S rRNA was lower in the PR group (0.49 ± 0.05, n = 6) when compared with the control group (0.80 ± 0.12, n = 6; p < 0.05) at 140 d of gestation(Fig. 3).

Figure 2
figure 2

Autoradiograph of a Northern blot after hybridization of radiolabeled POMC cDNA and 18 S oligonucleotide probes with RNA extracted from individual anterior pituitaries from the PR and control groups.

Full size image
Figure 3
figure 3

POMC mRNA relative to 18 S RNA (mean ± SEM) in the anterior pituitaries of fetal sheep from the control and PR groups. The mean POMC mRNA:18 S RNA ratio in the PR group was significantly lower(p < 0.05) than that in the control group.

Full size image

Plasma ACTH 1-39 and ir ACTH concentrations. Plasma concentrations of ACTH 1-39 increased (p < 0.006) after 134 d of gestation in both the PR and control groups. Placental restriction did not alter plasma ACTH 1-39 concentrations at either gestational age range(Table 1).

Table 1 Fetal plasma ACTH 1-39 and ir ACTH concentrations(mean ± SEM) in PR and control groups between 112 and 141 d of gestation
Full size table

There was also no difference in the plasma concentrations of ir ACTH between the PR and control groups between 112 and 141 d of gestation. The mean plasma ACTH concentrations across this gestational age range were 117.5± 27 pg/mL (n = 9) and 99.3 ± 19 pg/mL (n = 10) in the PR and control groups, respectively. There was also no change with gestational age in plasma ir ACTH concentrations in either group in late gestation (Table 1).

Plasma ir ACTH concentrations correlated negatively with arterial Po2 (p < 0.03, r = -0.366) in the PR but not the control group between 112 and 141 d of gestation (Fig. 4). Plasma ir ACTH did not correlate with glucose concentrations in the PR or control groups across the gestational range of the study.

Figure 4
figure 4

The relationship between plasma ir ACTH concentrations and fetal arterial Po2 values in fetal sheep in the PR group between 112 and 141 d of gestation (the solid line represents the line of best fit derived by linear regression analysis).

Full size image

Plasma cortisol concentrations. Fetal plasma cortisol increased(p < 0.001) with increasing gestational age in both PR and control groups (Fig. 1). Plasma cortisol concentrations were higher (p < 0.05) between 132 and 141 d of gestation than between 112 and 126 d of gestation in the PR and control groups. Plasma cortisol concentrations were also higher (p < 0.02) in the PR than in the control group between 127 and 141 d of gestation(Fig. 1). Plasma cortisol did not correlate with arterial Po2 or plasma glucose concentrations in either the PR or control fetal sheep.

DISCUSSION

We have demonstrated that POMC mRNA in the fetal pituitary was decreased after chronic placental restriction. We have also found that, whereas experimental restriction of placental function did not alter plasma concentrations of ACTH 1-39 during the last 15 d of gestation, placental restriction was associated with an increase in fetal adrenal weight and in plasma cortisol concentrations in late pregnancy.

It is particularly interesting that experimental restriction of placental function was associated with a significant decrease in pituitary POMC mRNA levels. It is likely that the steady state levels of POMC mRNA in the fetal anterior pituitary represent the balance between stimulatory hypothalamic and inhibitory adrenal influences acting in late gestation. It has previously been demonstrated that POMC mRNA levels markedly increase after bilateral fetal adrenalectomy(22, 23). One possibility, therefore, is that the lower levels of POMC mRNA in the placental restriction group may be the result of the negative feedback effects of the higher cortisol concentrations present in this group. It also appears from this and our previous studies that there is no direct relationship between changes in the steady state POMC mRNA levels in the fetal pituitary and the circulating levels of ACTH 1-39 or ir ACTH before delivery(10, 11).

We found that there was a similar increase in the plasma concentrations of ACTH 1-39 in the growth-restricted and control groups in late gestation. We have previously shown that there is a relative increase in the output of ACTH 1-39 compared with ACTH precursors from the perifused fetal sheep pituitary after 136 d of gestation(12), and it is therefore likely that the increase in circulating ACTH 1-39 is a consequence of a change in the processing of POMC in the anterior pituitary in the 2 wk preceding delivery. From the present study, it appears that chronic fetal hypoxemia or hypoglycemia do not alter the timing or magnitude of the age-related change in posttranslational processing of POMC in the fetal corticotrophs. It has been suggested that this change may be related to alterations in the morphologic characteristics of the fetal corticotrophs which occur in the fetal sheep pituitary at the same stage of gestation(24).

Although experimental restriction of placental function resulted in chronic fetal hypoxemia and hypoglycemia, there were no differences in the circulating concentrations of either ACTH 1-39 or ir ACTH between the growth-restricted and control animals throughout late gestation. It was interesting, however, that there was an inverse correlation between Po2 and fetal ir ACTH concentrations in the PR group, which was not present in the control animals.

The lack of a relationship between arterial Po2 and ir ACTH in the control animals presumably reflects the narrower range of Po2 values present in this group. It may be that chronic hypoxemia acts to either increase pituitary secretion or decrease fetal clearance of ACTH. Although there were no differences in ACTH concentrations between the two fetal groups, adrenal weight and plasma cortisol concentrations were higher in the growth-restricted fetal sheep than in their control counterparts after 127 d of gestation. One previous study on the effects of chronic hypoxemia induced by reduction of uteroplacental blood flow for 24 h reported, however, that fetal ACTH concentrations were elevated only at 2 h after the onset of hypoxemia and then returned to baseline values(25). In contrast, fetal cortisol concentrations were increased by 2 h after the onset of hypoxemia induced by uteroplacental occlusion and remained elevated throughout the 24-h period. Cordocentesis studies in human fetuses that were small for gestational age have also found that plasma cortisol concentrations are higher and plasma ACTH concentrations are lower than in normally grown fetuses at 18-38 wk of gestation(1).

The data from the present and these previous studies clearly suggest that the effects of chronic hypoxemia on the fetal hypothalamo-pituitary-adrenal axis differ from those of acute hypoxemia. It has been shown in the adult rat that the main effects of a range of chronic stressors in the hypothalamo-pituitary-adrenal axis are an increase in adrenal weight and plasma corticosteroid concentrations, an increased sensitivity of the adrenal to ACTH, maintained ACTH concentrations, and a diminished central sensitivity to glucocorticoid feedback(26). It is possible, therefore, that the hypothalamo-pituitary-adrenal axis in the growth-restricted fetus is operating at a new central set point, which results in maintained ACTH concentrations, increased adrenal sensitivity to ACTH, and increased adrenal weight and corticosteroid output. A further possibility is that there are additional placental endocrine signals in growth-restricted fetuses, such as an increase in placental prostaglandin E2 production(25), which may act independently to increase the output of cortisol from the fetal adrenal.

In summary, therefore, we have used a model of chronic restriction of placental growth and function to investigate the impact of chronic stress on the fetal pituitary-adrenal axis. We have demonstrated that POMC mRNA levels in the anterior pituitary after 135 d of gestation are suppressed by chronic placental restriction. We have also found that although experimental restriction of placental function did not alter plasma concentrations of ACTH 1-39 during the last 15 d of gestation, placental restriction increased fetal adrenal weight and plasma cortisol concentrations in late pregnancy. These changes may imply that the hypothalamo-pituitary-adrenal axis is operating at a new central set point in the growth-restricted fetus and highlight the central role of this axis in the physiologic adaptations, which allow the fetus to survive prolonged intrauterine stress and to initiate and prepare for the transition to extrauterine life.