Divergent changes in plasma ACTH and pituitary POMC
mRNA after cortisol administration to late-gestation
ovine fetus
T. M. JEFFRAY,1–3 S. G. MATTHEWS,1 G. L. HAMMOND,2,4 AND J. R. G. CHALLIS1–3
of Physiology and Obstetrics and Gynaecology, University of Toronto,
Toronto M2S 1A8; 2Medical Research Council Group in Fetal and Neonatal Health and
Development, 3Lawson Research Institute and Departments of Obstetrics and Gynaecology,
and Physiology, and 4London Regional Cancer Centre, Victoria Hospital,
Departments of Obstetrics and Gynaecology, Biochemistry, and Oncology,
University of Western Ontario, London N6A 5C1, Ontario, Canada
1Departments
corticosteroid-binding globulin; free cortisol; adrenocorticotropic hormone; proopiomelanocortin
IN FETAL SHEEP, basal plasma concentrations of immunoreactive (ir) adrenocorticotropic hormone (ACTH) and
cortisol (32) and pituitary proopiomelanocortin (POMC)
expression (29) rise concomitantly in late gestation.
This occurs despite demonstrable negative-feedback
effects of glucocorticoids on stress-induced (1) or corticotropin-releasing hormone (CRH)-stimulated pituitaryadrenal function (31), suggesting that there may be a
decrease in cortisol negative feedback on the hypothalamus and pituitary at this time. Apostolakis et al. (3)
previously showed that administration of cortisol to the
ovine fetus at 134 days of gestation altered ACTH
pulsatility, increasing ACTH pulse peak and nadir
values (3). However, the mechanisms whereby intrafetal
cortisol has a positive effect on plasma ACTH concentrations during late gestation is not clear. Therefore we
infused cortisol to fetal sheep, beginning before the
prepartum rise in endogenous cortisol, in amounts that
would reproduce plasma cortisol concentrations similar
to those near term to determine the effects on plasma
ACTH and to examine the underlying mechanisms of
any changes.
The plasma concentration of corticosteroid-binding
globulin (CBG), the high-affinity binding protein for
cortisol, also rises during the last third of gestation (4,
7), reflecting an increase in its synthesis in the fetal
liver (7). We have previously showed that CBG biosynthesis was stimulated by exogenous cortisol (5, 23) and
reduced after bilateral adrenalectomy of the ovine fetus
(23). It was therefore suggested that the rise in cortisol
stimulated CBG and would in turn help to maintain a
low free cortisol concentration in plasma despite elevations in the total (free 1 bound) cortisol concentration
(4, 11). We reasoned that a rise in CBG would result in
decreased free cortisol (11, 26), which in turn would
diminish the negative-feedback effects of cortisol on the
pituitary, resulting in an increase in pituitary POMC
mRNA levels and ACTH output. To examine the underlying mechanisms of the ACTH response to exogenous
cortisol, we measured levels of POMC mRNA in different regions of the pituitary and CBG mRNA in the liver
at various times during an intrafetal cortisol infusion.
Therefore the overall hypothesis of the present study
was that intrafetal cortisol administration during late
gestation would increase hepatic CBG synthesis and
circulating CBG levels, thereby reducing free cortisol
concentrations in plasma. In turn, this would reduce
the negative-feedback effects of cortisol on pituitary
POMC expression and irACTH output, resulting in an
increase in circulating irACTH concentrations, despite
elevated total cortisol concentrations in plasma. In this
manner, we would approximate the plasma profiles of
ACTH and cortisol observed in the fetal sheep near
term.
METHODS
Animals. Surgery was performed under general anesthesia, on mixed breed ewes, at days 119–122 of gestation (full
term is 145–147 days). The techniques used have been
0193-1849/98 $5.00 Copyright r 1998 the American Physiological Society
E417
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on June 17, 2017
Jeffray, T. M., S. G. Matthews, G. L. Hammond, and
J. R. G. Challis. Divergent changes in plasma ACTH and
pituitary POMC mRNA after cortisol administration to lategestation ovine fetus. Am. J. Physiol. 274 (Endocrinol. Metab.
37): E417–E425, 1998.—Plasma concentrations of cortisol
and adrenocorticotropic hormone (ACTH) rise in the lategestation sheep fetus at approximately the same time as
there is an increase in the plasma levels of corticosteroidbinding globulin (CBG). We hypothesized that intrafetal
cortisol infusion during late pregnancy would stimulate an
increase in fetal plasma CBG, which in turn would bind
cortisol and diminish glucocorticoid negative-feedback regulation of the fetal pituitary, leading to an increase in plasma
ACTH concentrations. Cortisol was infused into chronically
catheterized fetal sheep beginning at 126.1 6 0.5 days of
gestation and continued for 96 h. Control fetuses were infused
with saline. In cortisol-infused fetuses, the plasma cortisol
concentrations rose significantly from control levels (4.4 6 0.6
ng/ml) to 19.3 6 3.1 ng/ml within 24 h and remained
significantly elevated throughout the infusion period. Plasma
immunoreactive (ir) ACTH concentrations were significantly
elevated in cortisol-infused fetuses within 24–48 h and
remained significantly higher than in controls throughout the
96-h experimental period. Plasma free cortisol concentrations
increased 10-fold and remained significantly elevated in
cortisol-infused animals, despite a rise in plasma corticosteroid-binding capacity. Levels of pituitary proopiomelanocortin (POMC) mRNA in the fetal pars distalis and pars intermedia were 96 and 38% lower, respectively, after 96 h of cortisol
infusion. Therefore physiological elevations of plasma cortisol, in the late-gestation ovine fetus, lead to increases in mean
plasma irACTH concentrations, but this is not associated
with increases in fetal pituitary POMC mRNA levels.
E418
EFFECTS OF CORTISOL ON FETAL PITUITARY FUNCTION
(32). The intra- and interassay coefficients of variation were
9 and 13%, respectively, and the mean assay sensitivity was
6.5 pg/ml. This ACTH antibody cross-reacts ,0.01% with
a-melanocyte-stimulating hormone (MSH), b-MSH, b-endorphin, and b-lipotropin (b-LPH; Incstar) and does not recognize pro-ACTH or POMC (kindly provided by Dr. J. Schwartz,
Bowman Gray School of Medicine, Winston-Salem, NC). The
antibody recognized .95% immunoreactivity corresponding
to ACTH-(1—39) when samples of fetal sheep plasma in
normoxemia or during hypoxemia were assayed after highperformance liquid chromatography separation of ACTHrelated peptides (12). Plasma cortisol concentrations were
quantified by RIA after extraction with diethyl ether. The
antibody characteristics and assay validation for measurement of cortisol in fetal sheep plasma have been described
previously (14). The combined intra- and interassay coefficient of variation was 12%. The percentage of free cortisol in
duplicate samples of fetal plasma was measured by the
centrifugal ultrafiltration-dialysis technique developed and
described by Hammond et al. (20). All of the samples were
measured in a single assay. Fetal plasma CBG levels were
measured as corticosteroid binding capacity (CBC) determined using the saturation binding assay of Ballard et al. (4),
with modifications described previously (13). The withinassay coefficient of variation was ,10%.
Northern blot analysis of CBG mRNA. We previously
showed that the liver is the major site of CBG biosynthesis in
fetal sheep (7). Total cellular RNA was extracted from the
fetal liver, electrophoresed, and blotted as described previously (6). The blotting membranes were probed using an
ovine CBG cDNA (6). After autoradiographic exposure, the
blots were stripped and reprobed with a cDNA to mouse 18S
rRNA to allow correction for variations in gel loading and
transfer. The relative optical densities (ROD) were determined using computerized image analysis (Imaging Research, St. Catharines, ON, Canada). Results are expressed
as the ratio of the ROD of the CBG mRNA to 18S rRNA
hybridization signals.
In situ hybridization of pituitary POMC mRNA. Frozen
pituitaries were sectioned (15-µm coronal sections) using a
cryostat (Tissue-Tek, Miles Canada, Etobicoke, ON, Canada),
mounted onto poly-L-lysine (Sigma Chemical, St. Louis, MO)coated slides and fixed in 4% paraformaldehyde (28). Pituitary sections were incubated with a 35S-labeled, 45-mer
oligonucleotide complementary to bases 711–756 of the porcine POMC gene (19) as previously described in detail (37).
The slides were exposed to autoradiographic film (XAR 5,
Kodak) for 2 h at room temperature to determine the hybridization signal for POMC mRNA levels in the pars intermedia
and then reexposed for 4 days to measure the levels of POMC
mRNA in the pars distalis. The two exposure times were
necessary for the hybridization signal to be within the linear
range of the film for the two regions of the pituitary. Linearity
was established by the simultaneous exposure of the film to
14C standards (27). The autoradiograms were then analyzed
using computerized image analysis (Imaging Research). Results are expressed as ROD for a minimum of 15 pituitary
sections per animal. A control 45-mer sense oligonucleotide
was also synthesized, and no signal was observed when it was
hybridized with pituitary sections (10). The fetal pituitary
sections were then coated with Ilford K5 liquid emulsion,
developed and counterstained with Carazzi’s hematoxylin to
identify nuclei. The silver grain deposits were visualized
using light microscopy.
Immunohistochemistry. Immunohistochemical detection of
irACTH was performed on 15-µm frozen pituitary sections
prepared as described above for in situ hybridization. A
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on June 17, 2017
described previously (14). Briefly, a midline incision was
made in the ewe’s lower abdomen to expose the uterus, which
was then opened, and the fetal head was exteriorized. Polyvinyl catheters, filled with sterile heparinized saline, were
inserted into a fetal carotid artery and jugular vein. Catheters were also placed into the amniotic cavity and a maternal
femoral vein. Uterine electromyographic (EMG) leads (Cooner
Wire, Chatsworth, CA) were attached to the uterus to monitor
myometrial electrical activity. Prophylactic antibiotics were
administered at the time of surgery and continued for 3 days
postoperatively as described previously (14).
Blood samples were collected daily, and pH, PCO2, and PO2
were measured using an ABL-5 blood gas analyzer (Radiometer, Copenhagen, Denmark). The protocols were approved by
the Animal Care Committees of St. Joseph’s Health Centre,
University of Western Ontario, and University of Toronto in
accordance with the guidelines of the Canadian Council on
Animal Care.
Experimental protocols. Animals were allowed a minimum
of 5 days to recover after surgery before experimentation
began. Starting on days 124–129, fetuses received an intravenous infusion of either cortisol (11b,17,21-trihydroxy-4pregnene-3,20-dione, Steraloids, Wilton, NH; 5 µg/min, n 5 7)
or an equal volume of saline (3 ml/h, with 2% vol/vol ethanol,
n 5 6) for 96 h. Fetal arterial blood samples (4 ml) were
collected into chilled, heparinized syringes every 8 h beginning 24 h before the start of infusion and continuing throughout the experiment. Samples were centrifuged immediately
at 1,500 g for 10 min at 4°C, and the plasma was stored at
220°C until analysis. Additional animals were infused for 12
(n 5 4) or 24 h (n 5 4) to examine changes in POMC and CBG
mRNA levels associated with the initial phase of the rise in
plasma cortisol concentrations. At the conclusion of the
infusion periods (12, 24, or 96 h), the animals were killed with
an overdose of 24% pentobarbital sodium (Euthanyl, MTC
Pharmaceuticals, Cambridge, ON, Canada), and fetal tissues
were collected quickly. Fetal pituitaries were frozen on dry ice
for in situ hybridization and immunohistochemistry. A portion of the right lobe of the liver was frozen rapidly in liquid
nitrogen for Northern blot analysis.
Fetal blood gases and uterine activity. Fetal blood pressure
and amniotic fluid pressure were measured continuously
using Statham pressure transducers (P23XL, Spectramed,
Oxnard, CA) and displayed on a chart recorder (model 78D,
Grass Instrument). Cortisol has previously been shown to
induce hypertension in the ovine fetus (17). Therefore blood
pressure was monitored to determine whether there was any
association between mean arterial pressure and changes in
plasma hormone values. Mean fetal blood pressure was
calculated as 0.4 3 (systolic pressure 2 diastolic pressure) 1
diastolic pressure 2 amniotic fluid pressure. Blood pressure
was measured at five different time points every hour, and the
mean was calculated. The daily mean was then calculated
from these hourly values. Because intrafetal administration
of cortisol can induce premature parturition in sheep (25),
uterine activity was monitored continuously to ensure that
any changes in plasma irACTH concentrations were not
attributable to the process of labor. Uterine EMG activity was
measured using a Grass wide-band AC preamplifier (Grass
model 78D) and was recorded as the number of episodes of
low-amplitude activity lasting .5 min, referred to as ‘‘contractures’’ (30), and the number of contractions (activity lasting
0.5–1 min) per 2-h period (21).
Measurement of plasma ACTH, cortisol, free cortisol, and
CBG. Plasma irACTH concentrations were measured by a
commercial radioimmunoassay (RIA) kit (Incstar, Stillwater,
MN) that was validated previously for use in the fetal sheep
E419
EFFECTS OF CORTISOL ON FETAL PITUITARY FUNCTION
times of treatment were assessed by Student-Newman-Keuls
multiple range tests. Values for hepatic CBG mRNA levels,
free cortisol concentrations, and number of corticotrophs
staining positively for irACTH were not distributed normally
and were therefore assessed by the Kruskal-Wallis analysis of
variance followed by Dunn’s test or the Mann-Whitney ranksum test (Sigmastat, Jandel Scientific, CA).
RESULTS
polyclonal antibody to human ACTH-(1—24) (Dako, Carpinteria, CA) was used in conjunction with avidin-biotin-peroxidase reagents from the Vectastain ABC kit (Vector Laboratories, Burlingame, CA), as previously described (33). The
ACTH antibody has been characterized extensively, and the
antigenic site has been shown to be between amino acids 18
and 24 (22). This ACTH antibody cross-reacts ,1% with
a-MSH, b-MSH, and b-LPH (22). The number of immunopositive cells within 10 fields (475 µm 3 350 µm 5 1.7 3 105 µm2 )
was counted for each animal. Adjacent sections were incubated with the primary antibody in the presence of an excess
of antigen [human ACTH-(1—24)] to provide negative controls.
Data analysis. Blood pressure and plasma hormone concentrations were measured in samples collected every 8 h, and
the daily mean value was calculated for each animal. These
values are reported as means 6 SE for the number of animals
stated. The maximal and minimal changes in irACTH concentrations were calculated from the three plasma samples
collected each day, relative to the mean plasma irACTH value
during the control period, for individual animals and reported
as means 6 SE. Changes in plasma cortisol, CBG, irACTH,
maximal and minimal change in irACTH, mean arterial pressure, and uterine activity were analyzed by two-way analysis
of variance corrected for repeated measures. Statistical significance was determined as P # 0.05. The effects of individual
Table 1. Fetal arterial blood gases, mean arterial pressure, and uterine activity (per 2-h period) during first 24 h
before and first and last 24 h of infusion if saline or cortisol
Saline
pH
PO2, mmHg
PCO2, mmHg
O2 saturation, %
MAP, mmHg
Contractures, no./2 h
Contractions, no./2 h
Cortisol
224 to 0 h
0–24 h
72–96 h
224 to 0 h
0–24 h
72–96 h
7.36 6 0.01
22.2 6 0.7
53.5 6 0.8
60.2 6 2.5
40.3 6 0.4
2.6 6 0.2
0.6 6 0.1
7.34 6 0.01
20.65 6 0.9
55.6 6 0.9
56.6 6 4.0
41.6 6 0.5
2.7 6 0.2
0.7 6 0.1
7.35 6 0.01
21.41 6 0.6
54.9 6 1.4
56.0 6 2.3
41.5 6 0.9
1.9 6 0.1
0.4 6 0.1
7.38 6 0.01
21.8 6 1.0
60.2 6 2.5
60.8 6 2.5
42.6 6 1.0
2.7 6 0.1
1.0 6 0.2
7.37 6 0.01
22.1 6 1.3
56.6 6 4.0
61.4 6 3.5
48.7 6 0.7*
3.3 6 0.1
1.0 6 0.2
7.36 6 0.01
24.4 6 1.7
56.0 6 2.3
65.5 6 3.5
50.9 6 1.6*
4.0 6 0.3*
3.3 6 0.7*
Values are means 6 SE; for blood gases and saline, n 5 5; for cortisol, n 5 7; for mean arterial pressure (MAP) and uterine activity, n 5 3.
Contracture, uterine activity lasting .5 min. Contraction, uterine activity lasting ,1 min and .30 s. *P , 0.05.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on June 17, 2017
Fig. 1. Plasma cortisol concentrations before and during a 96-h
cortisol (filled bars, n 5 7) or saline (open bars, n 5 6) infusion,
starting at time 0, to fetal sheep in late gestation. Values are means 6
SE. * P , 0.05. Cortisol infusion increased plasma cortisol concentrations significantly within 24 h [2-way repeated-measures analysis of
variance (ANOVA) followed by Student-Newman-Keuls] to values
similar to those for full-term animals.
Plasma cortisol concentrations. The daily mean concentration of cortisol in fetal plasma rose during cortisol infusion from basal values of 4.4 6 0.6 to 19.3 6 3.1
ng/ml within 24 h (n 5 6, P , 0.05; Fig. 1). The mean
plasma cortisol concentration rose progressively to a
maximal concentration of 38.6 6 2.7 ng/ml, which is
similar to that seen in the ovine fetus near term (32).
Plasma cortisol concentrations in the control animals
did not change significantly throughout the infusion
period.
Fetal blood gases and uterine activity. Fetal arterial
PO2, PCO2, O2 saturation, and pH were unchanged
throughout the study in both cortisol- and salinetreated animals (Table 1). Mean arterial pressure
increased significantly within the first 24 h of cortisol
administration from 42.6 6 1.0 to 48.7 6 0.7 mmHg and
remained at this level throughout the infusion period.
There was no significant change in mean arterial
pressure in the saline-infused fetuses (Table 1). The
number of contractures or contractions per 2-h interval
did not change significantly in the saline-treated fetuses. Similarly, during intrafetal cortisol infusion,
uterine activity was not altered during the first 72 h (at
48–72 h, mean number of contractures 5 3.4 6 0.2 and
contractions 5 1.0 6 0.2 per 2-h period). However, there
was an increase in uterine contractility during the final
24 h of intrafetal cortisol infusion (Table 1).
Effects of cortisol on plasma irACTH. Plasma irACTH
concentrations rose significantly in the cortisol-treated
fetuses from values of 25.4 6 2.7 pg/ml during the
control period (Fig. 2) to a significantly elevated level of
37.8 6 3.9 pg/ml at 24–48 h (Fig. 2). Fetal ACTH values
remained elevated throughout the cortisol infusion
period (n 5 7, P , 0.05, effect of time F 5 7.9 and
cortisol F 5 9.6; Fig. 2). In the saline-infused animals,
plasma ACTH concentrations did not change significantly from control values of 27.0 6 1.7 pg/ml (n 5 6;
Fig. 2).
E420
EFFECTS OF CORTISOL ON FETAL PITUITARY FUNCTION
ACTH is secreted in pulses; therefore plasma irACTH
concentrations were variable between samples within
individual fetuses. The maximal and minimal changes
in plasma concentrations of irACTH for the three
samples collected each day were calculated relative to
the initial 24-h control period for individual animals.
The maximal change in irACTH was significantly
higher at 24–48 h of cortisol infusion and remained
significantly elevated throughout the infusion compared with saline-treated animals (Table 2). There was
no difference in the minimal change in irACTH for
either of the infusion groups throughout the experiment (data not shown).
Effects of cortisol on CBG biosynthesis and secretion.
Plasma CBG levels were similar in both the cortisol
(34.6 6 2.1 ng/ml) and saline-treated animals (27.5 6
3.0 ng/ml) during the control period (Fig. 3). The
plasma CBG levels of the cortisol-treated animals were
significantly elevated (51.1 6 3.2 ng/ml) by 48–72 h of
infusion and remained elevated at 72–96 h (P , 0.05,
effect of time F 5 6.0 and treatment F 5 46.3).
Northern blot analysis of RNA from the fetal liver
identified a single CBG transcript of 1.8 kb (Fig. 4A).
Cortisol treatment elevated levels of hepatic CBG
mRNA (P , 0.05, Kruskal-Wallis analysis of variance).
After 96 h, hepatic CBG mRNA levels were signifi-
Fig. 3. Plasma corticosteroid-binding capacity (CBC) before and
during a 96-h cortisol (filled bars, n 5 7) and saline (open bars, n 5 6)
infusion, starting at time 0, to fetal sheep in late gestation. Values are
means 6 SE. * P , 0.05 (repeated-measures 2-way ANOVA followed
by Student-Newman-Keuls).
cantly higher (P # 0.05, Mann-Whitney rank-sum test)
in the cortisol-infused animals.
Plasma free cortisol levels. The percentage of free
cortisol in plasma increased threefold, from 6.3 6 0.5 to
16.8 6 6.2% at 8 h of cortisol infusion (Fig. 5A), but by
72 h the percentage of free cortisol was not significantly
different from that of saline-infused animals. However,
the absolute plasma concentration of free cortisol rose
within 8 h and remained significantly elevated in
cortisol-treated animals throughout the 96-h infusion
(Fig. 5B). There was no change in the percentage of free
cortisol in the plasma of saline-treated animals throughout the course of the experiment, and the absolute free
Table 2. Mean maximal change in plasma irACTH
compared with average plasma irACTH concentrations
during 24-h control period
Mean
Plasma
irACTH,
pg/ml
n
Control
Maximal Change in Plasma irACTH, pg/ml
0–24 h
24–48 h
48–72 h
72–96 h
Saline 6 27.0 6 2.3 8.7 6 2.0 7.7 6 2.9 13.7 6 5.5 11.4 6 4.8
Cortisol 7 25.8 6 4.1 11.1 6 4.4 20.7 6 6.4* 23.7 6 5.4* 23.7 6 4.7*
Values are means 6 SE. * P , 0.05. ir, Immunoreactive.
Fig. 4. Effect of cortisol on hepatic corticosteroid-binding globulin
(CBG) expression. A: Northern blot analysis of CBG mRNA in livers
of individual fetuses treated with cortisol or saline for 12 (n 5 4), 24
(n 5 4), or 96 h (saline n 5 6 and cortisol n 5 7). A cDNA probe to 18S
rRNA was used to control for amount of RNA analyzed. B: histograms
show ratio of relative optical densities (ROD) of CBG mRNA: 18S
rRNA for each of cortisol (filled bars) and saline (open bars) infusion
groups. Values are means 6 SE. * P , 0.05 (Mann-Whitney U test).
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on June 17, 2017
Fig. 2. Plasma immunoreactive (ir) ACTH concentrations before and
during a 96-h infusion of cortisol (filled bars, n 5 7) or saline (open
bars, n 5 6), starting at time 0, to fetal sheep in late gestation. Values
are means 6 SE. * P , 0.05 (for repeated-measures 2-way ANOVA
followed by Student-Newman-Keuls). Cortisol infusion significantly
increased mean irACTH concentrations by 24–48 h, and irACTH
remained significantly elevated throughout infusion compared with
saline controls.
EFFECTS OF CORTISOL ON FETAL PITUITARY FUNCTION
E421
Fig. 5. Changes in percentage of free cortisol (A) and total free cortisol (B) concentrations in cortisol (filled bars)- or saline (open bars)treated animals. Values are means 6 SE; n 5 6 in each group, except
at 96 h, n 5 3 cortisol-infused and n 5 4 control animals. * P , 0.05.
cortisol concentration did not change significantly from
mean values of 0.3 6 0.1 ng/ml (n 5 3; Fig. 5B).
Effect of cortisol on levels of pituitary POMC mRNA
and irACTH. Levels of POMC mRNA in the pars
distalis of cortisol-treated animals were lower (Fig. 6C)
than those in saline control animals (Fig. 6A) after 96 h
of infusion. POMC mRNA levels in the pars intermedia
were also decreased after 96 h of cortisol infusion (Fig.
6D) compared with the saline-treated controls (Fig. 6B).
POMC mRNA levels were quantified by computerized image analysis. There was no significant change in
levels of POMC mRNA within the pars distalis or the
pars intermedia after 12 h of cortisol treatment, but the
levels had fallen to 4% of controls by 96 h (Fig. 7).
High-resolution analysis using liquid silver emulsion
autoradiography confirmed this finding by the near
absence of silver grain deposits in the pars distalis after
96 h of cortisol infusion (Fig. 8C), compared with the
abundance of silver grain deposits seen in the salinetreated fetuses (Fig. 8A). POMC mRNA levels in the
pars intermedia of the cortisol-treated animals were
unchanged at the end of the first 24 h of infusion, but
the levels had decreased by 38% compared with saline
controls after 96 h of infusion (Fig. 7). There was no
apparent change in the number of silver grains deposited between groups (Fig. 8, E and G).
The intensity of irACTH peptide staining within the
pars distalis did not appear different after 96 h of
Fig. 6. Autoradiograms of coronal pituitary sections after in situ hybridization using a 35Slabeled proopiomelanocortin (POMC) oligonucleotide. Fetuses treated for 96 h with saline
(A and B) or cortisol (C and D). A and C were
exposed for 4 days to allow analysis of POMC
mRNA in pars distalis. B and D were exposed
for 2 h to allow analysis of POMC mRNA in pars
intermedia (signal in pars distalis is barely
evident at this time). Scale bar: 470 µm for A
and C; 222 µm for B and D.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on June 17, 2017
Fig. 7. Histogram illustrating changes in POMC mRNA levels in
pars distalis and pars intermedia after 96 h of cortisol (filled bars) or
saline (open bars) treatment. Values are expressed as ROD of
autoradiograms and expressed as means 6 SE. * P , 0.05 (StudentNewman-Keuls).
EFFECTS OF CORTISOL ON FETAL PITUITARY FUNCTION
E422
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on June 17, 2017
EFFECTS OF CORTISOL ON FETAL PITUITARY FUNCTION
cortisol treatment (Fig. 8D), compared with the salineinfused animals (Fig. 8B). However, the number of
irACTH-positive corticotrophs in the pars distalis was
significantly reduced (a decrease of 13.9%) from 218.6 6
12.5 immunopositive cells per 1.7 3 105 µm2 in the
saline controls to 188.2 6 15.1 immunopositive cells per
1.7 3 105 µm2 after 96 h of cortisol infusion. There was
no diminution in the number of immunopositive cells in
the pars intermedia after 96 h of cortisol infusion (Fig.
8H ) compared with control (Fig. 8F). Adjacent sections
were incubated with antibody preabsorbed with an
excess of human ACTH-(1—24), and there was no
staining present (data not shown; see also Ref. 22).
DISCUSSION
rise in ACTH is not hypoxia mediated. In addition,
there was no change in the plasma irACTH concentrations of the saline-treated fetuses, verifying that the
rise in ACTH is not an effect of the sampling protocol
employed.
It has previously been reported that a 96-h cortisol
infusion to the ovine fetus at 134 days of gestation
affects pulsatility of irACTH, increasing pulse peak and
nadir (3). We examined the maximal and minimal
changes in irACTH and compared these with the mean
irACTH concentration during the control period. The
maximal change in irACTH from the three plasma
samples collected each day was significantly elevated
during the cortisol infusion compared with that of
control fetuses. However, we did not employ a frequent
sampling protocol necessary to establish whether the
cortisol infusion administered in these animals altered
ACTH pulse frequency over the course of the experiment.
It has been demonstrated that plasma CBG levels
rise within 2–4 days of cortisol infusion (3, 5). In the
present study, plasma CBG levels rose in response to
cortisol, becoming significantly greater than those in
controls at 48–72 h of infusion. This was associated
with an increase in hepatic CBG mRNA levels. Lowdose cortisol infusion to the ovine fetus at 100 days of
gestation (for 100 h), or administration of dexamethasone at 130 days of gestation (for 96 h), increased CBG
biosynthesis and secretion and also altered the pattern
of CBG glycosylation (5, 6). Changes in CBG glycoforms
may increase the half-life of CBG in the circulation, and
this may account for the earlier rise in CBC in plasma
than in steady-state levels of hepatic CBG mRNA
determined in the present study.
The percentage of free cortisol in plasma rose within
8 h of cortisol infusion and then returned to control
levels by 72 h of infusion. However, the absolute
concentrations of free cortisol remained elevated
throughout the experiment. This suggests that CBG is
effective at maintaining the percentage of free cortisol
in fetal circulation but does not control the absolute
concentration of free cortisol during periods of rapidly
increasing plasma cortisol concentrations. This reflects
the CBG response to increasing plasma cortisol concentrations in the sheep fetus at term. Plasma CBG and
cortisol concentrations rise in parallel during late
gestation, and low free cortisol concentrations are
effectively maintained until the last 5 days of pregnancy (4, 7). Although CBG did not appear to be
effective in decreasing the negative-feedback effects of
the rapidly increasing plasma cortisol concentrations,
this does not preclude a role for circulating CBG in
modifying feedback control of the prepartum rise in
plasma ACTH levels. In addition, we showed earlier
that the fetal sheep pituitary synthesizes CBG (7),
which could alter local feedback mechanisms. However,
Fig. 8. Photographs of emulsion autoradiograms localizing POMC mRNA within pars distalis (A and C) and pars
intermedia (E and G) and irACTH peptide levels in pars distalis (B and D) and pars intermedia (F and H) of ovine
fetus after a 96-h infusion of cortisol (C, D, G, and H) or saline (A, B, E, and F). PI, pars intermedia; PD, pars
distalis. Scale bar: 50 µm.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on June 17, 2017
We have shown that, in the late-gestation ovine
fetus, mean plasma irACTH concentrations rise significantly in response to physiological elevations in plasma
cortisol concentrations. However, this does not appear
to be due to a decrease in cortisol negative feedback,
since plasma free cortisol concentrations remained
elevated despite an increase in plasma CBG levels. The
high concentration of plasma free cortisol suppressed
pituitary POMC mRNA levels in both the pars distalis
and pars intermedia, and the number of irACTH immunopositive cells within the pituitary pars distalis was
also reduced after 96 h of cortisol infusion. However,
the number of immunopositive cells within the pars
intermedia did not appear to be appreciably altered in
the presence of elevated circulating free cortisol.
We believe that the increase in plasma irACTH
predominantly reflects changes in ACTH-(1—39). The
ACTH antibody used to measure fetal ACTH concentrations has been validated extensively and does not show
cross-reactivity with the higher molecular weight
ACTH-related peptides. We suggest that the increase
in irACTH reflects an increase in ACTH output in
response to the infusion of cortisol, although we cannot
exclude the possibility that the metabolic clearance
rate of circulating ACTH is reduced by cortisol infusion.
Mean fetal arterial pressure increased by ,7 mmHg
within the first 24 h of cortisol infusion and remained at
this level throughout the experiment. This is similar to
the increase reported during a 48-h cortisol infusion to
the midgestation ovine fetus (17). However, the hypertensive effects of cortisol do not elicit an immediate
ACTH response and seem very unlikely to be causal in
the later rise in plasma irACTH concentrations. Although uterine activity does increase slightly within
the last 24 h of cortisol infusion, this event is preceded
by the ACTH rise; therefore the initial rise in plasma
irACTH cannot be labor induced. The increase in
plasma irACTH at 72–96 h may be associated with an
increase in uterine activity; however, there was no
change in fetal arterial PO2, which indicates that the
E423
E424
EFFECTS OF CORTISOL ON FETAL PITUITARY FUNCTION
mented the release of bioactive ACTH (8). The
pituitaries collected for the present study were slowfrozen for in situ hybridization, and therefore the
cellular morphology could not be examined effectively.
It remains possible that the rise in plasma irACTH in
response to the cortisol infusion could be a result of
AVP-stimulated ACTH release and/or a change in the
processing of POMC such that ACTH-(1—39) is preferentially released from POMC remaining in the anterior
pituitary corticotrophs. Alternatively, we speculate that
changes in prohormone convertase activities (39) in the
pars distalis and pars intermedia of the fetal pituitary
might lead to increased proportions of ACTH-(1—39)
secreted or that circulating ACTH might be derived
from alternate sources, including the lung and placenta
(15, 16, 24). At present, the effects of cortisol on these
alternative sites of POMC production and processing
have yet to be elucidated.
We thank Jac Homan and Drs. Mildred Ramirez, Geert Braems,
and Ed Berdusco for their input and assistance.
This work was supported by the Medical Research Council (MRC)
of Canada (Group in Fetal and Neonatal Health and Development;
G. L. Hammond and J. R. G. Challis). S. G. Matthews was supported
by an MRC fellowship.
Address for reprint requests: T. M. Jeffray, Dept. of Physiology,
Medical Science Bldg., University of Toronto, 1 King’s College Circle,
Toronto, ON, Canada M2S 1A8.
Received 16 September 1997; accepted in final form 18 November
1997.
REFERENCES
1. Akagi, K., E. T. M. Berdusco, and J. R. G. Challis. Cortisol
inhibits ACTH but not the AVP response to hypoxaemia in fetal
lambs at days 123–128 of gestation. J. Dev. Physiol. 14: 319–324,
1990.
2. Antolovich, G. C., R. A. Perry, J. F. Trahair, M. Silver, and
P. M. Robinson. The development of corticotrophs in the fetal
sheep pars distalis: the effect of adrenalectomy or cortisol
infusion. Endocrinology 124: 1333–1339, 1989.
3. Apostolakis, E. M., L. D. Longo, and S. M. Yellon. Cortisol
feedback regulation of pulsatile ACTH secretion in fetal sheep
during late gestation. Am. J. Physiol. 267 (Endocrinol. Metab.
30): E521–E527, 1994.
4. Ballard, P. L., J. A. Kitterman, R. D. Bland, R. I. Clyman,
P. D. Gluckman, A. C. G. Platzker, S. L. Kaplan, and M. M.
Grumbach. Ontogeny and regulation of corticosteroid binding
globulin capacity in plasma of fetal and newborn lambs. Endocrinology 110: 359–366, 1982.
5. Berdusco, E. T. M., G. L. Hammond, R. A. Jacobs, A. Grolla,
K. Akagi, D. Langlois, and J. R. G. Challis. Glucocorticoidinduced increase in plasma corticosteroid-binding globulin levels
in fetal sheep is associated with increased biosynthesis and
alterations in glycosylation. Endocrinology 132: 2001–2008, 1993.
6. Berdusco, E. T. M., W. K. Milne, and J. R. G. Challis.
Low-dose cortisol infusion increases plasma corticosteroidbinding globulin (CBG) and the amount of hepatic CBG mRNA in
fetal sheep on day 100 of gestation. J. Endocrinol. 140: 425–430,
1994.
7. Berdusco, E. T. M., K. Yang, G. L. Hammond, and J. R. G.
Challis. Corticosteroid-binding globulin (CBG) production by
hepatic and extra-hepatic sites in the ovine fetus; effects of CBG
on glucocorticoid negative feedback on pituitary cells in vitro. J.
Endocrinol. 146: 121–130, 1995.
8. Brieu, V., and P. Durand. Changes in the ratio of bioactive to
immunoreactive adrenocorticotropin-like activity released by
pituitary cells from ovine fetuses and lambs. Endocrinology 120:
936–940, 1987.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on June 17, 2017
we did not determine changes in pituitary CBG biosynthesis in the current study.
POMC mRNA levels in the pars intermedia and pars
distalis were not affected by 12 h of cortisol treatment,
as has been demonstrated previously (28). However,
after 96 h of cortisol infusion, POMC mRNA levels were
suppressed in both the pars intermedia and pars
distalis. In addition, the number of irACTH-positive
cells within the pars distalis was decreased after 96 h of
cortisol treatment, suggesting that the elevated plasma
free cortisol concentrations are exerting negativefeedback effects on the corticotrophs of the pars distalis. Nonetheless, it is possible that cortisol affects the
rate of POMC translation or ACTH secretion. However,
within the pars intermedia there was no apparent
change in irACTH-positive cells or the intensity of
irACTH staining. Pars intermedia corticotrophs from
fetal sheep secrete ACTH-(1—39) in vitro (18). In
addition, attenuation of the endogenous cortisol rise, by
bilateral fetal adrenalectomy, does not alter the basal
ACTH-(1—39) output from subsequently cultured pars
intermedia cells (18). The pars intermedia may therefore provide an additional source of ACTH. In the
presence of high plasma concentrations of cortisol,
there may be an alteration of POMC processing within
the pars intermedia, resulting in an increase in circulating ACTH-(1—39). Alternatively, the pars intermedia
may secrete large molecular weight POMC products
into the circulation (34), which are processed to ACTH(1—39) under the influence of elevated cortisol at other
sites.
Schwartz et al. (35) have examined the regulation of
pituitary ACTH secretion in vitro and have shown that
there are subpopulations of corticotrophs within the
pars distalis which are differentially regulated. These
are the CRH-sensitive corticotroph, which secretes
ACTH in response to CRH via an increase in adenosine
38,58-cyclic monophosphate, and the arginine vasopressin (AVP)-responsive corticotroph, which stimulates
ACTH release by a second pathway likely involving
inositol trisphosphate (38). In vitro studies indicate
that glucocorticoids do not inhibit ACTH-(1—39) secretion and significantly increased ACTH precursor release from the AVP-responsive corticotrophs (35). A
direct effect of cortisol on fetal corticotrophs has also
been described. Antolovich et al. (2) characterized a
change in the morphology of the corticotrophs from
ovine fetuses treated with cortisol from a ‘‘fetal-’’ to an
‘‘adult-type’’ cell. This morphological change occurs
normally during late gestation but did not occur in
adrenalectomized fetuses (2). Histological maturation
of the corticotroph may also be associated with altered
processing of POMC to produce ACTH-(1—39) preferentially (2). ACTH-(1—39) is the predominant POMC
product of the adult corticotroph (36). In support of this
possibility, Brieu and Durand found that fetal ovine
corticotroph cells treated with cortisol (4 days) in vitro
decreased the output of total ACTH peptides but increased the relative proportion of immunoreactive material that coeluted with ACTH-(1—39) (9) and aug-
EFFECTS OF CORTISOL ON FETAL PITUITARY FUNCTION
25. Liggins, G. C. Premature parturition after infusion of corticotrophin or cortisol into foetal lambs. J. Endocrinol. 43: 323–329,
1968.
26. Liggins, G. C., R. J. Fairclough, S. A. Grieves, J. Z. Kendall,
and B. S. Knox. The mechanism of initiation of parturition in
the ewe. Recent Prog. Horm. Res. 29: 111–150, 1973.
27. Matthews, S. G., and J. R. G. Challis. Regulation of CRH and
AVP mRNA in the developing ovine hypothalamus: effects of
stress and glucocorticoids. Am. J. Physiol. 268 (Endocrinol.
Metab. 31): E1096–E1107, 1995.
28. Matthews, S. G., and J. R. G. Challis. Levels of proopiomelanocortin and prolactin mRNA in the fetal sheep pituitary following hypoxaemia and glucocorticoid treatment in late
gestation. J. Endocrinol. 147: 139–146, 1994.
29. Matthews, S. G., X. Han, F. Lu, and J. R. G. Challis.
Developmental changes in the distribution of pro-opiomelanocortin and prolactin mRNA in the pituitary of the ovine fetus and
lamb. J. Mol. Endocrinol. 13: 175–185, 1994.
30. Nathanielsz, P. W., S. Ratter, A. L. Thomas, L. Rees, and
P. M. B. Jack. The role and regulation of ACTH in the fetal
sheep. In: The Fetus and Birth. Amsterdam: Elsevier, 1976, p.
73–91. (Ciba Found. Symp. 47)
31. Rose, J. C., B. Y. Hargrave, P. M. Dix, P. J. Meis, M. LaFave,
and B. Torpe. Corticotropin-releasing factor-induced adrenocorticotropic hormone release in the sheep fetus: blockade by
cortisol. Am. J. Obstet. Gynecol. 151: 1128–1133, 1985.
32. Norman, L. J., S. J. Lye, M. E. Wlodek, and J. R. G. Challis.
Changes in pituitary responses to synthetic ovine corticotrophin
releasing factor in fetal sheep. Can. J. Physiol. Pharmacol. 63:
1398–1403, 1985.
33. Riley, S. C., J. C. Walton, J. M. Herlick, and J. R. G. Challis.
The localization and distribution of corticotropin-releasing hormone in the human placenta and fetal membranes throughout
gestation. J. Clin. Endocrinol. Metab. 72: 1001–1007, 1991.
34. Saphier, P. W., B. P. Glynn, R. J. Woods, D. A. Whepherd,
M. K. Jeacock, and P. J. Lowry. Elevated levels of N-terminal
pro-opiomelanocortin peptides in fetal sheep plasma may contribute to fetal adrenal gland development and the pre-parturient
cortisol surge. Endocrinology 133: 1459–1461, 1993.
35. Schwartz, J., P. Ash, V. Ford, H. Raff, S. Crosby, and A.
White. Secretion of adrenocorticortrophin (ACTH) and ACTH
precursors in ovine anterior pituitary cells: action of corticotrophin-releasing hormone, arginine vasopressin and glucocorticoids. J. Endocrinol. 140: 189–195, 1994.
36. Silman, R. E., D. Holland, T. Chard, P. J. Lowry, J. Hope,
L. H. Rees, A. Thomas, and P. W. Nathanielsz. Adrenocorticotrophin-related peptides in adult and foetal sheep pituitary
glands. J. Endocrinol. 81: 19–34, 1979.
37. Sirinathsinghji, D. J. S., B. J. Morris, W. Wisden, A.
Northrop, S. P. Hunt, and S. B. Dunnett. Gene expression in
striatal grafts. I. Cellular localization of neurotransmitter
mRNAs. Neuroscience 34: 675–686, 1990.
38. Todd, K., and S. L. Lightman. Vasopressin activation of
phosphatidylinositol metabolism in rat anterior pituitary in vitro
and its modification by changes in the hypothalamo-pituitaryadrenal axis. Neuroendocrinology 45: 212–218, 1987.
39. Zheng, M., R. D. Streck, R. E. M. Scott, N. G. Seidah, and
J. E. Pintar. The developmental expression in rat of proteases
furin, PC1, PC2, and carboxypeptidase E: implications for early
maturation of proteolytic processing capacity. J. Neurosci. 14:
4656–4673, 1994.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on June 17, 2017
9. Brieu, V., and P. Durand. Adrenocorticotropic hormone released by pituitary cells from ovine fetuses and lambs. Neuroendocrinology 49: 300–308, 1989.
10. Broad, K. D., K. M. Kendrick, D. J. S. Sirinathsinghji, and
E. B. Keverne. Changes in pro-opiomelanocortin and preproenkephalin mRNA levels in the ovine brain during pregnancy,
parturition and lactation and in response to estrogen and
progesterone. J. Neuroendocrinol. 5: 711–719, 1993.
11. Challis, J. R. G., and A. N. Brooks. Maturation and activation
of hypothalamic-pituitary-adrenal function in fetal sheep. Endocr. Rev. 10: 182–204, 1989.
12. Challis, J. R. G., L. Fraher, J. Oosterhuis, S. E. White, and
A. D. Bocking. Fetal and maternal endocrine responses to
prolonged reductions in uterine blood flow in pregnant sheep.
Am. J. Obstet. Gynecol. 160: 926–932, 1989.
13. Challis, J. R. G., E. A. Nancekievell, and S. J. Lye. Possible
role of cortisol in the stimulation of cortisol-binding capacity in
the plasma of fetal sheep. Endocrinology 116: 1139–1144, 1985.
14. Challis, J. R. G., J. E. Patrick, J. Cross, J. Workewych, E.
Manchester, and S. Power. Short-term fluctuations in the
concentration of cortisol and progesterone in fetal plasma, maternal plasma, and amniotic and allantoic fluids from sheep during
late pregnancy. Can. J. Physiol. Pharmacol. 59: 261–267, 1981.
15. Cudd, T. A., and C. E. Wood. Secretion and clearance of
immunoreactive ACTH by fetal lung. Am. J. Physiol. 268 (Endocrinol. Metab. 31): E845–E848, 1995.
16. Deol, H. K., S. G. Matthews, and J. R. G. Challis. Levels and
distribution of POMC mRNA and CRH mRNA and irACTH and
irCRH in the developing ovine lung (Abstract). Prog. Annu. Meet.
Endocr. Soc. 77th Washington DC 1995, p. 425.
17. Dodic, M., and E. M. Wintour. Effects of prolonged (48h)
infusion of cortisol on blood pressure, renal function and fetal
fluids in the immature ovine foetus. Clin. Exp. Pharmacol.
Physiol. 21: 971–980, 1994.
18. Fora, M. A., T. G. Butler, J. C. Rose, and J. Schwartz.
Adrenocorticotropin secretion by fetal sheep anterior and intermediate lobe pituitary cell in vitro: effects of gestation and
adrenalectomy. Endocrinology 137: 3394–3400, 1996.
19. Gossard, F. J., A. C. Y. Chang, and S. N. Cohen. Sequence of
cDNA encoding porcine pro-opiomelanocortin. Biochim. Biophys.
Acta 866: 68–74, 1986.
20. Hammond, G. L., J. A. Nisker, L. A. Jones, and P. K. Siiteri.
Estimation of the percentage of free steroid in undiluted serum
by centrifugal ultrafiltration-dialysis. J. Biol. Chem. 255: 5023–
5026, 1980.
21. Harding, R., E. R. Poore, A. Bailey, G. D. Thorburn, C. A.
Jansen, and P. W. Nathanielsz. Electromyographic activity of
the nonpregnant and pregnant sheep uterus. Am. J. Obstet.
Gynecol. 142: 448–457, 1982.
22. Jacobs, R. A., J. Oosterhuis, D. G. Porter, D. K. Lobb, A. A.
Yuzpe, and J. R. G. Challis. Immunoreactive adrenocorticotrophin is present in the ovary and in particular the oocyte of
several mammalian species. J. Reprod. Fertil. 91: 285–291, 1991.
23. Jeffray, T. M., E. T. M. Berdusco, M. J. Wallace, A. Fowden,
and J. R. G. Challis. Effects of incremental cortisol and
adrenalectomy on plasma corticosteroid-binding capacity in fetal
sheep. Can. J. Physiol. Pharmacol. 273: 1568–1573, 1995.
24. Keller-Wood, M., and C. E. Wood. Does the ovine placenta
secrete ACTH under normoxic or hypoxic conditions? Am. J.
Physiol. 260 (Regulatory Integrative Comp. Physiol. 29): R389–
R395, 1991.
E425