- Email: [email protected]
Administration of corticosterone to pregnant adrenalectomized dams does not alter the hypothalamic-pituitary-adrenal axis' activity of the offspring
MOLECULAR
AND CELLULAR
3,118-123
NEUROSCIENCES
(1992)
Administration of Corticosterone to Pregnant Adrenalectomized Dams Does Not Alter the Hypothalamic-Pituitary-Adrenal Axis’ Activity of the Offspring SOON LEE The
Clayton
Foundation
Laboratories
for Peptide
Biology,
Received
AND
CATHERINE RIVIER
The Salk
Institute,
for publication
We have previously suggested a possible role of increased maternal corticosterone (CORT) secretion in mediating the endocrine effects of prenatal alcohol administration. We tested this hypothesis by adrenalectomizing pregnant rats and exposing them to high levels of CORT that mimicked those previously measured in intact dams exposed to alcohol. CORT (28 mg/day) or placebo pellets were placed subcutaneously in adrenalectomized (ADX) dams during Days 8-14 of gestation. At 21 days of age, the offspring were either decapitated under nonstressed conditions or exposed to mild, inescapable electroshocks (1.5 mA; 1 s duration; 2.5/min) over a lo-min period and then sacrificed. Plasma ACTH and CORT levels were measured by radioimmunoassay. There were no measurable differences between basal or stress-induced ACTH secretion of pups born to ADXCORT or ADX-placebo dams. In contrast, the offspring of ADX-CORT dams demonstrated significantly (P G 0.01; P d 0.05) lower CORT secretion when exposed to the shocks despite unaltered basal levels. Measurement of CRF expression by in situ hybridization methodology indicated no difference between CRF biosynthesis of both groups of pups. These results do not support the hypothesis that the increased stress-induced ACTH release of rats born to intact dams exposed to alcohol is primarily mediated by increased maternal corticosterone levels. 0 1992 Academic
Press, Inc.
INTRODUCTION Maternal consumption of ethanol is widely recognized as a frequent cause of mental retardation, abnormal growth development, and behavioral deficits in children (l-4). In the rat, prenatal exposure to alcohol causes an increased activity of the HPA axis in response to stress in the offspring (5-7). This change, which is specifically observed following exposure of pregnant dams to alcohol during the second week of gestation, is accompanied by significantly elevated CRF mRNA levels in the hypo-
Copyright All rights
$3.00 1992 by Academic Press, of reproduction in any form
118
1044-7431/92
0
Inc. reserved.
10010
November
North
Torrey
Pines
Road,
La Jolla,
California
92037
1, 1991
thalamus of the 21-day-old pups (8). At present, the mechanisms through which exposure of pregnant mothers to alcohol causes these endocrine changes in their offspring have not been determined. However, our observation that this increased stress-induced ACTH secretion and augmented brain CRF biosynthesis weFe not present in the offspring of adrenalectomized (ADX) dams suggested increased adrenal maternal activity as a possible modulator of the effect of prenatal alcohol exposure. The present studies were therefore undertaken to test this hypothesis. MATERIALS
AND METHODS
Animals Female Sprague-Dawley rats and male breeders were obtained from Charles River Breeding Laboratories (Wilmington, MA). After mating, the females were checked for vaginal plugs twice daily. The presence of a plug was considered Gestational Day 0. Pregnant rats were adrenalectomized via a dorsal approach under methoxyfluorane anesthesia and then replaced with either CORT (200 mg/7 days) or placebo (Innovative Research, Toledo, OH) SCpellets from Days 8 to 14 of gestation. The pellets were custom-made so that corticosterone was delivered over a 7-day period only. All ADX dams received a 0.9% saline solution to drink and their diet (Purina Lab Chow) was supplemented with fresh oranges daily. The room was temperature controlled with a 12-h light/l2-h dark cycle. On Day 12 of gestation, blood samples were obtained by tail bleeding. Delivery occurred after 22-23 days of gestation in both ADX rats. The pups were weighed at birth, pooled, and randomized within each group. Foster mothers were provided to all pups. At 21 days of age, groups of pups were placed in shock cages. Three hours later, they were exposed to mild, intermittent electroshocks (1.5 mA, 1 s duration, 25 times) for 10 min. The electric shocks were delivered to the grid floors of plexiglass boxes by a Coulbourn shocker (Coul-
EXOGENOUS
CORT
TREATMENT
bourn Instruments, Lehigh Valley, PA) controlled by an Apple 2E computer chamber (9). These pups were then quickly removed from the shockers and decapitated, and trunk blood was collected. Control pups were separated from their mothers, but otherwise left undisturbed in a cage. They were decapitated rapidly following removal from their boxes, at the time of sacrifice of the stressed animals. Blood samples were centrifuged and plasma was frozen at -20°C until assayed. All protocols were approved by the Salk Institute Animal Use and Care Committee.
Hormone Assays Plasma ACTH (10) and CORT (11) were determined by radioimmunoassay (RIA). The ACTH RIA has a sensitivity of 30-40 pg/ml and intraassay and interassay coefficients of variation of 7 and 15%, respectively. The limit of detection of the CORT RIA is 5-9 rig/ml plasma, and the intraassay and interassay coefficients of variation are 6 and 11%, respectively.
CRF Expression Four pups born to ADX-CORT or ADX-placebo dams were deeply anesthetized and perfused with saline followed by 4% paraformaldehyde in 0.1 M borate buffer. Brains were placed overnight at 4°C in fixative with 10% sucrose added. Frozen sections (30 pm) were cut on a sliding microtome, stored in antifreeze solution (30% ethylene glycol, 20% glycerol, 50% NaPO.& mounted onto gelatin- and poly-L-lysine-coated slides, and air-dried. The hybridization protocol has been described previously (12). Prior to hybridization, sections were dried under vacuum overnight, digested by proteinase K (10 pg/ ml) for 20 min at 37°C acetylated, and dehydrated. The probe was dissolved in a buffer containing 50% formamide, 10% dextran, 1X Denhart’s solution, 1 r&f EDTA (pH 8.0), 10 n&f Tris (pH 8.0), 30 mM NaCl, 0.5 mg/ml yeast transfer RNA, and 10 n&f dithiothreitol (DTT). After being vacuum-dried, 1 X lo6 dpm of probe in 100 ~1 of hybridization solution was applied to each slide and sealed under coverslips. Hybridizations were performed at 60°C overnight. The coverslips were then removed, and the slides were rinsed in four 5-min changes of 4X SSC (1X SSC: 15 mM trisodium citrate buffer, pH 7.0/0.15 M NaCl) at room temperature. They were digested by RNase A (20 pg/ml) for 30 min followed by two 5-min changes of 2X SSC, 10 min 1X SSC, and 10 min 0.5X SSC at room temperature and 30 min 0.1X SSC at 55°C. They were then dehydrated by 50% (3 min), 70% (3 min), 95% (3 min), 100% (3 min) twice EtOH, and air-dried. These sections were exposed to X-ray film for 2 days, then dipped in NTB2 nuclear emulsion (1:l with HzO, Kodak), exposed for 7 days, and developed. The sections were counterstained with thionin.
AND
HPA
119
AXIS
Prepro-CRFcRNA probe (gift from Dr. K. Mayo) was synthesized by SP-6 polymerase. The pGEM-4 plasmid containing the CRFcDNA insert was linealized with HindIII. Radioactive cRNA copies were synthesized by incubation of 36 mM Tris at pH 7.5, 0.1 mg of linealized plasmid in 6 mM MgC$, 2 mM spermidine, 8 mM DTT, 25 mM GTP/ATP/CTP, [w~~S]UTP, 1 U RNasin (Promega), and 10 U SP-6 polymerase (Promega) for 60 min at 37°C. The specific activity of the probe was about 1.0 X 10’ dpm/mg. As a control for nonspecific labeling, a sense CRF probe (same sequence as CRF mRNA) generated by T7 polymerase was used on some adjacent sections from experimental animals, and no specific signal was detected.
Quantitative
Analysis
Quantification was done in nuclear emulsion-dipped slides as previously described (8). The sections containing the dorsal aspect of the medial parvocellular subdivision of the paraventricular nucleus (PVN), where hypophysiotropic CRF cells are concentrated, were identified. The number of grains were counted using a microscopy utilizing an eyepiece with a superimposed grid for the measuring area. The background values, measured by counting the grains over deep layers of the cerebral cortex, were subtracted from the grain numbers measured in the PVN.
Statistical Analysis The data were first analyzed by analysis of variance, followed by the multiple comparison tests of Duncan. Criteria for significance was set at P < 0.05. Results are presented as the mean f SEM. The unit of statistical analysis is the individual pup. Several experiments were pooled for the results presented in Figs. 1 and 2. RESULTS
Effect of CORT Replacement on the Weights of the Offspring Body weight was measured on Days 1 and 21 of age (Table 1). The birth weight of female and male pups was significantly (P < 0.01) lower in pups born to ADX dams replaced with placebo pellets than in those born to CORTreplaced mothers. This difference persisted at Day 21.
Plasma CORT Levels in ADX and Intact Dams during the Second Week of Gestation Table 2 presents CORT values of ADX-CORT-replaced rats and, for the sake of comparison, those of intact, EtOH-exposed dams used in our previous study which were statistically comparable (8). There was no significant difference between plasma CORT levels of control intact dams and ADX-placebo-replaced dams.
120
LEE
TABLE Effect of CORT Treatment in Grams) of the Offspring
Note. ADX, adrenalectomized; ’ P d 0.01 from ADX-placebo * P d 0.001 from ADX-placebo
RIVIER
1
400 1
on the Body Weight (Expressed at Birth and 21 Days of Age
Treatment ADX dams Placebo-F CORT-F ADX dams Placebo-M CORT-M
AND
1 day
21 days
4.8 k 0.3 5.7 + 0.2”
33.5 f 0.8 36.6 + 0.7”
4.9 + 0.2 6.4 + 0.2*
34.5 k 0.9 39.4 + 0.6”
F, females; dams. dams.
i-
q
PLACEBO
5
W
CORT
g
300-
M, males.
FEMALE
+--
Effect of CORT Treatment during the Second Week of Gestation on ACTH and CORT Secretion of Pups Born to ADX Dams
MALE
Basal
_+
w
t-EMALE
+
MALE
Stress
---+
FIG. 1. Effect of CORT treatment on plasma ACTH secretion under basal or stress conditions. Twenty-one-day-old pups were born to adrenalectomized (ADX) mothers exposed to placebo or CORT pellets (200 mg/7 days) during the second week of gestation and left undisturbed for 3 h. One group of pups was decapitated immediately (basal). Another group of pups was exposed to mild, intermittent electroshocks for 10 min and decapitated immediately (stress). Each bar represents mean ACTH levels + SEM of 20-30 rats.
There were no significant differences between basal or stress-induced plasma ACTH levels of pups born to ADX-placebo or ADX-CORT dams (Fig. 1). Basal CORT levels were also comparable between male and female pups of the two experimental groups (Fig. 2). In contrast, both female (P < 0.01) and male (P < 0.05) offspring of ADX-CORT mothers exhibited significantly lower CORT secretion than pups born to ADXplacebo rats (Fig. 2).
PVN) between pups born to ADX dams replaced with placebo pellets and pups born to CORT-replaced ADX mothers.
Effect of CORT Replacement on Hypothalamic CRF Expression
We have previously shown that prenatal exposure to EtOH increased stress-induced ACTH secretion and augmented brain CRF biosynthesis in 21-day-old offspring of intact dams and that pups born to adrenalec-
Expression by in situ hybridization or measurement of CRF mRNA levels (Figs. 3 and 4) did not indicate any significant difference (based on the number of silver grains distributed over the parvocellular division of the
TABLE
2
Levels of Circulating Corticosterone in ADX Dams either Replaced with CORT or Placebo Pellet and Intact Dams Exposed to EtOH during the Second Week of Gestation Treatment
Corticosterone
ADX dams Placebo CORT Intact dams Control EtOH Note. ADX, adrenalectomized. ’ P < 0.01 vs ADX-placebo * P < 0.01 vs control dams. cd P > 0.05.
DISCUSSION
&
PLACEBO
W
CORT
T
I
I-+
-IL
(rig/ml)
”
6.6 + 2.0d 147.4 -c 11.0”” 8.9 -+ 3.4d 138.1 + 37.7*$
dams.
0
FEMALE
+ FIG. 2.
MALE
Basal
-+
MALE
FEMALE
+
Stress
+
Effect of CORT treatment on plasma CORT secretion under basal or stress conditions. Twenty-one-day-old pups were born to ADX mothers replaced with either CORT (200 mg/7 days) or placebo pellets during the second week of gestation, left undisturbed for 3 h, and decapitated immediately (basal) or after lo-min intermittent electroshocks (stress). Each bar represents mean CORT levels + SEM of 20-30 rats. **p G 0.01; *P s 0.05.
EXOGENOUS
FIG. replaced
3. Dark-field photograph with either placebo pellets
CORT
TREATMENT
AND
HPA
(X320) of the parvocellular division of the paraventricular nucleus (A) or CORT pellets (B) during the second week of gestation.
tomized mothers similarly exposed to ethanol failed to exhibit this increased HPA axis activity (8). This suggested an important mediating role of adrenal factors in the offspring’s responsiveness to prenatal alcohol exposure. Specifically, the observation of increased corticosterone levels in intact pregnant dams exposed to alcohol led us to hypothesize that this increase might mediate the altered endocrine status of their offspring. However, as shown by the present studies, exogenous
121
AXIS
of 21-day-old
pups
born
from
ADX
mothers
corticosterone replacement, at doses similar to those measured in intact, alcohol-exposed dams, did not mimic the effect of prenatal EtOH exposure. The interactions between the maternal and the fetal HPA axis are complex. Alcohol crosses the placenta readily (13,14) and can influence the fetal brain. However, the fetal HPA axis is not functional until the middle of the third week of gestation (15-17). It is thus improbab’le that alcohol, which was administered only until
LEE
AND
RIVIER
and/or effects of alcohol unrelated remains to be determined.
to adrenal function,
ACKNOWLEDGMENTS This research was supported by NIAAA Grant AA06420 to C.R. and Training Grant AA07456 to S.L. This research was conducted in part by the Clayton Foundation for Research California Division. C.R. is a Clayton Foundation investigator. We are grateful to B. Jewell, C. Gonzalez, S. Henson, Y. Haas, S. Johnson, and S. Smith for excellent technical assistance, and B. Coyne for secretarial assistance.
PLACEBO
REFERENCES
CORT
FIG. 4. Bar graph summarizes the effect of CORT treatment on the number of grains counted in the PVN of 2l-day-old pups born from ADX dams replaced with either placebo or CORT pellet. Each bar represents the mean f SEM of 4 animals in each group.
1.
2. 3.
Day 14 of gestation, could exert effects on the embryo’s PVN similar to those observed in adult animals. It is also necessary to emphasize that the increased CRF expression we measured following prenatal alcohol exposure was observed in 21-day-old pups. While these pups showed increased responsiveness of their HPA axis in response to shocks, younger animals are reported to exhibit a suppressed stress response (18-20). It is therefore possible that the changes which we observed (8) represent compensatory mechanisms caused by an earlier decrease in HPA axis activity, possibly resulting from alcohol-induced changes in adrenal and/or pituitary responsiveness (18). If this is the case, exogenous corticosterone may not mimic the effect of alcohol, as increased levels of this steroid represent only part of the HPA axis’ response to the drug. There is also evidence that in the developing fetus, CRF gene expression may be insensitive to corticosteroid feedback (21). If correct, this finding would indicate that alcohol primarily acts through mediators other than corticoids. While many such mediators are likely candidates, it is important to remember that adrenals also secrete secretagogues such as catecholamines, which exert powerful effects on the HPA axis. Furthermore, corticosterone treatment causes an alteration in brain catecholamines directly and in turn has a modulating role on the HPA axis (22). The potential role of these compounds as mediators of the endocrine influence of prenatal alcohol exposure, and the consequence of their removal in ADX dams, has not been evaluated. In conclusion, we have observed that exogenous corticoid treatment of adrenalectomized dams does not mimic the effect of prenatal EtOH exposure on the HPA axis’ activity of the offspring. Whether this reflects a relative insensitivity of the embryos’ HPA axis to steroid feedback, the influence of adrenal factors other than corticosterone,
Abel, E. L. (1978). Effects of ethanol on pregnant rats and their offspring. Psychopharmacology 67: 5-11. Cogan, D. C., L. E. Cohen, and G. Sparman (1983). Effects of gestational alcohol on the development of neonatal reflexes in the rat. Neurobehnv. Toxicol. Teratol. 5: 517-522. Streissguth, A. P. (1984). The behavioral teratology of alcohol: Performance, behavioral, and intellectual deficits in prenatally exposed children. In Alcohol and Brain Development (J. West, Eds.), pp. 3-44. Oxford, New York.
4. Jones,
5. 6. 7.
K. L., D. W. Smith, C. N. Ulleland, and A. P. Streissguth (1973). Patterns of malformation in offspring of chronic alcoholic mothers. Lancet 1: 1267-1271. Taylor, A. N., B. J. Branch, S. H. Liu, and N. Kokka (1982). Longterm effects of fetal ethanol exposure on pituitary-adrenal responses to stress. Pharmacol. B&hem. Behuv. 16: 585-589. Weinberg, J., and P. V. Gallo (1982). Prenatal ethanol exposure: Pituitary-adrenal activity in pregnant dams and offspring. Neurobehav. Toricol. Teratol. 4: 515-520. Taylor, A. N., B. J. Branch, L. R. Nelson, L. A. Lane, and R. E. Poland (1986). Prenatal ethanol and ontogeny of pituitary-adrenal responses to ethanol and morphine. Alcohol 3: 255-259.
8.
Lee, S. Y., T. Imaki, W. Vale, and C. R. Rivier (1990). Effect of prenatal exposure to ethanol on the activity of hypothalamicpituitary-adrenal axis of the offspring Importance of the time of exposure to ethanol and possible modulating mechanisms. Mol. Cell. Neurosci. 1: 168-177.
9.
Rivier, stress Res.
C., and W. Vale (1988). Interaction on ACTH and @-endorphin secretion.
between ethanol Alcohol. Clin.
and Exp.
12:206-210.
10.
Orth, D. (1979). Adrenocorticotrophic Methods of Hormone Radioimmuncussay Eds.), 2nd ed., pp. 245-284. Academic
11.
Gwosdow-Cohen, A., C. L. Chen, and E. L. Besch (1982). munoassay (RIA) of serum corticosterone in rats. Proc. Biol. Med. 170:29-34.
12.
Simmons, D. M., J. L. Arriza, and L. W. Swanson (1989). A complete protocol for in situ hybridization of messenger RNAs in brain and other tissues with radiolabeled single-stranded RNA probes. J. Histotechnol. 12: 169-181. Eguchi, Y. (1969). Interrelationships between the fetal and maternal hypophyseal-adrenal axes in rats and mice. In Physiology and Pathology of Adaptation Mechanics (E. Bajusz, Eds.), pp. 3-27. Pergamon Press, New York.
13.
hormone (ACTH). In (B. Jaffe and H. Behrman, Press, New York. RadioimSot. Exp.
14.
Weinberg, J., and S. Bezio (1987). Alcohol-induced tuitary-adrenal activity during pregnancy. Alcohol.
changes in piClin. Exp. Res.
15.
Eguchi, Y., 0. Hirai, Y. Morikawa, and Y. Hashimoto (1973). ical time in the hypothalamic control of the pituitary-adrenal
11:274-280. Critsys-
EXOGENOUS
16.
17.
18. 19.
CORT
TREATMENT
tern in fetal rats: Observations in fetuses subjected to hypervitaminosis A and hypothalamic destruction. Endocrinology 93: l-11. Yamamoto, M., Y. Eguchi, M. Asari, and Y. Kano (1983). Developmental changes in fetal adrenal hypertrophy following maternal bilateral and fetal unilateral adrenalectomy at different stages of gestation in the rat. Biol. Neon&e 43: 43-49. Bugnon, C., D. Fellmann, A. Gouget, and J. Cardot (1982). Ontogeny of the corticoliberin neuroglandular system in rat brain. Nature 298: 159-161. Weinberg, J. (1989). Prenatal ethanolexposure alters adrenocortical development of offspring. Alcohol. Clin. Exp. Res. 13: 73-83. Levine, S., and G. D. Coover (1976). Environmental control of suppression of the pituitary-adrenal system. Physiol. Behav. 17: 35-37.
AND
HPA
AXIS
123
20.
Levine, S. (1967). Maternal and environmental adrenocortical response to stress in weanling 258-260.
influences on the rats. Science 156:
21.
Jingami, H., S. Matsukura, S. Numa, and H. Imura (1985). Effects of adrenalectomy and dexamethasone administration on the level of prepro-corticotropin-releasing factor messenger ribonucleic acid (mRNA) in the hypothalamus and adrenocorticotropin/&Lipotropin precursor mRNA in the pituitary in rate. Endocrinology 117: 1314-1320.
22.
Jhanwar-Uniyal, M., K. J. Renner, M. T. Bailo, V. N. Luine, and S. F. Leibowitz (1989). Corticosterone-dependent alterations in utilization of catecholaminee in discrete areas of rat brain. Brain Res. 500: 247-255.
AND CELLULAR
3,118-123
NEUROSCIENCES
(1992)
Administration of Corticosterone to Pregnant Adrenalectomized Dams Does Not Alter the Hypothalamic-Pituitary-Adrenal Axis’ Activity of the Offspring SOON LEE The
Clayton
Foundation
Laboratories
for Peptide
Biology,
Received
AND
CATHERINE RIVIER
The Salk
Institute,
for publication
We have previously suggested a possible role of increased maternal corticosterone (CORT) secretion in mediating the endocrine effects of prenatal alcohol administration. We tested this hypothesis by adrenalectomizing pregnant rats and exposing them to high levels of CORT that mimicked those previously measured in intact dams exposed to alcohol. CORT (28 mg/day) or placebo pellets were placed subcutaneously in adrenalectomized (ADX) dams during Days 8-14 of gestation. At 21 days of age, the offspring were either decapitated under nonstressed conditions or exposed to mild, inescapable electroshocks (1.5 mA; 1 s duration; 2.5/min) over a lo-min period and then sacrificed. Plasma ACTH and CORT levels were measured by radioimmunoassay. There were no measurable differences between basal or stress-induced ACTH secretion of pups born to ADXCORT or ADX-placebo dams. In contrast, the offspring of ADX-CORT dams demonstrated significantly (P G 0.01; P d 0.05) lower CORT secretion when exposed to the shocks despite unaltered basal levels. Measurement of CRF expression by in situ hybridization methodology indicated no difference between CRF biosynthesis of both groups of pups. These results do not support the hypothesis that the increased stress-induced ACTH release of rats born to intact dams exposed to alcohol is primarily mediated by increased maternal corticosterone levels. 0 1992 Academic
Press, Inc.
INTRODUCTION Maternal consumption of ethanol is widely recognized as a frequent cause of mental retardation, abnormal growth development, and behavioral deficits in children (l-4). In the rat, prenatal exposure to alcohol causes an increased activity of the HPA axis in response to stress in the offspring (5-7). This change, which is specifically observed following exposure of pregnant dams to alcohol during the second week of gestation, is accompanied by significantly elevated CRF mRNA levels in the hypo-
Copyright All rights
$3.00 1992 by Academic Press, of reproduction in any form
118
1044-7431/92
0
Inc. reserved.
10010
November
North
Torrey
Pines
Road,
La Jolla,
California
92037
1, 1991
thalamus of the 21-day-old pups (8). At present, the mechanisms through which exposure of pregnant mothers to alcohol causes these endocrine changes in their offspring have not been determined. However, our observation that this increased stress-induced ACTH secretion and augmented brain CRF biosynthesis weFe not present in the offspring of adrenalectomized (ADX) dams suggested increased adrenal maternal activity as a possible modulator of the effect of prenatal alcohol exposure. The present studies were therefore undertaken to test this hypothesis. MATERIALS
AND METHODS
Animals Female Sprague-Dawley rats and male breeders were obtained from Charles River Breeding Laboratories (Wilmington, MA). After mating, the females were checked for vaginal plugs twice daily. The presence of a plug was considered Gestational Day 0. Pregnant rats were adrenalectomized via a dorsal approach under methoxyfluorane anesthesia and then replaced with either CORT (200 mg/7 days) or placebo (Innovative Research, Toledo, OH) SCpellets from Days 8 to 14 of gestation. The pellets were custom-made so that corticosterone was delivered over a 7-day period only. All ADX dams received a 0.9% saline solution to drink and their diet (Purina Lab Chow) was supplemented with fresh oranges daily. The room was temperature controlled with a 12-h light/l2-h dark cycle. On Day 12 of gestation, blood samples were obtained by tail bleeding. Delivery occurred after 22-23 days of gestation in both ADX rats. The pups were weighed at birth, pooled, and randomized within each group. Foster mothers were provided to all pups. At 21 days of age, groups of pups were placed in shock cages. Three hours later, they were exposed to mild, intermittent electroshocks (1.5 mA, 1 s duration, 25 times) for 10 min. The electric shocks were delivered to the grid floors of plexiglass boxes by a Coulbourn shocker (Coul-
EXOGENOUS
CORT
TREATMENT
bourn Instruments, Lehigh Valley, PA) controlled by an Apple 2E computer chamber (9). These pups were then quickly removed from the shockers and decapitated, and trunk blood was collected. Control pups were separated from their mothers, but otherwise left undisturbed in a cage. They were decapitated rapidly following removal from their boxes, at the time of sacrifice of the stressed animals. Blood samples were centrifuged and plasma was frozen at -20°C until assayed. All protocols were approved by the Salk Institute Animal Use and Care Committee.
Hormone Assays Plasma ACTH (10) and CORT (11) were determined by radioimmunoassay (RIA). The ACTH RIA has a sensitivity of 30-40 pg/ml and intraassay and interassay coefficients of variation of 7 and 15%, respectively. The limit of detection of the CORT RIA is 5-9 rig/ml plasma, and the intraassay and interassay coefficients of variation are 6 and 11%, respectively.
CRF Expression Four pups born to ADX-CORT or ADX-placebo dams were deeply anesthetized and perfused with saline followed by 4% paraformaldehyde in 0.1 M borate buffer. Brains were placed overnight at 4°C in fixative with 10% sucrose added. Frozen sections (30 pm) were cut on a sliding microtome, stored in antifreeze solution (30% ethylene glycol, 20% glycerol, 50% NaPO.& mounted onto gelatin- and poly-L-lysine-coated slides, and air-dried. The hybridization protocol has been described previously (12). Prior to hybridization, sections were dried under vacuum overnight, digested by proteinase K (10 pg/ ml) for 20 min at 37°C acetylated, and dehydrated. The probe was dissolved in a buffer containing 50% formamide, 10% dextran, 1X Denhart’s solution, 1 r&f EDTA (pH 8.0), 10 n&f Tris (pH 8.0), 30 mM NaCl, 0.5 mg/ml yeast transfer RNA, and 10 n&f dithiothreitol (DTT). After being vacuum-dried, 1 X lo6 dpm of probe in 100 ~1 of hybridization solution was applied to each slide and sealed under coverslips. Hybridizations were performed at 60°C overnight. The coverslips were then removed, and the slides were rinsed in four 5-min changes of 4X SSC (1X SSC: 15 mM trisodium citrate buffer, pH 7.0/0.15 M NaCl) at room temperature. They were digested by RNase A (20 pg/ml) for 30 min followed by two 5-min changes of 2X SSC, 10 min 1X SSC, and 10 min 0.5X SSC at room temperature and 30 min 0.1X SSC at 55°C. They were then dehydrated by 50% (3 min), 70% (3 min), 95% (3 min), 100% (3 min) twice EtOH, and air-dried. These sections were exposed to X-ray film for 2 days, then dipped in NTB2 nuclear emulsion (1:l with HzO, Kodak), exposed for 7 days, and developed. The sections were counterstained with thionin.
AND
HPA
119
AXIS
Prepro-CRFcRNA probe (gift from Dr. K. Mayo) was synthesized by SP-6 polymerase. The pGEM-4 plasmid containing the CRFcDNA insert was linealized with HindIII. Radioactive cRNA copies were synthesized by incubation of 36 mM Tris at pH 7.5, 0.1 mg of linealized plasmid in 6 mM MgC$, 2 mM spermidine, 8 mM DTT, 25 mM GTP/ATP/CTP, [w~~S]UTP, 1 U RNasin (Promega), and 10 U SP-6 polymerase (Promega) for 60 min at 37°C. The specific activity of the probe was about 1.0 X 10’ dpm/mg. As a control for nonspecific labeling, a sense CRF probe (same sequence as CRF mRNA) generated by T7 polymerase was used on some adjacent sections from experimental animals, and no specific signal was detected.
Quantitative
Analysis
Quantification was done in nuclear emulsion-dipped slides as previously described (8). The sections containing the dorsal aspect of the medial parvocellular subdivision of the paraventricular nucleus (PVN), where hypophysiotropic CRF cells are concentrated, were identified. The number of grains were counted using a microscopy utilizing an eyepiece with a superimposed grid for the measuring area. The background values, measured by counting the grains over deep layers of the cerebral cortex, were subtracted from the grain numbers measured in the PVN.
Statistical Analysis The data were first analyzed by analysis of variance, followed by the multiple comparison tests of Duncan. Criteria for significance was set at P < 0.05. Results are presented as the mean f SEM. The unit of statistical analysis is the individual pup. Several experiments were pooled for the results presented in Figs. 1 and 2. RESULTS
Effect of CORT Replacement on the Weights of the Offspring Body weight was measured on Days 1 and 21 of age (Table 1). The birth weight of female and male pups was significantly (P < 0.01) lower in pups born to ADX dams replaced with placebo pellets than in those born to CORTreplaced mothers. This difference persisted at Day 21.
Plasma CORT Levels in ADX and Intact Dams during the Second Week of Gestation Table 2 presents CORT values of ADX-CORT-replaced rats and, for the sake of comparison, those of intact, EtOH-exposed dams used in our previous study which were statistically comparable (8). There was no significant difference between plasma CORT levels of control intact dams and ADX-placebo-replaced dams.
120
LEE
TABLE Effect of CORT Treatment in Grams) of the Offspring
Note. ADX, adrenalectomized; ’ P d 0.01 from ADX-placebo * P d 0.001 from ADX-placebo
RIVIER
1
400 1
on the Body Weight (Expressed at Birth and 21 Days of Age
Treatment ADX dams Placebo-F CORT-F ADX dams Placebo-M CORT-M
AND
1 day
21 days
4.8 k 0.3 5.7 + 0.2”
33.5 f 0.8 36.6 + 0.7”
4.9 + 0.2 6.4 + 0.2*
34.5 k 0.9 39.4 + 0.6”
F, females; dams. dams.
i-
q
PLACEBO
5
W
CORT
g
300-
M, males.
FEMALE
+--
Effect of CORT Treatment during the Second Week of Gestation on ACTH and CORT Secretion of Pups Born to ADX Dams
MALE
Basal
_+
w
t-EMALE
+
MALE
Stress
---+
FIG. 1. Effect of CORT treatment on plasma ACTH secretion under basal or stress conditions. Twenty-one-day-old pups were born to adrenalectomized (ADX) mothers exposed to placebo or CORT pellets (200 mg/7 days) during the second week of gestation and left undisturbed for 3 h. One group of pups was decapitated immediately (basal). Another group of pups was exposed to mild, intermittent electroshocks for 10 min and decapitated immediately (stress). Each bar represents mean ACTH levels + SEM of 20-30 rats.
There were no significant differences between basal or stress-induced plasma ACTH levels of pups born to ADX-placebo or ADX-CORT dams (Fig. 1). Basal CORT levels were also comparable between male and female pups of the two experimental groups (Fig. 2). In contrast, both female (P < 0.01) and male (P < 0.05) offspring of ADX-CORT mothers exhibited significantly lower CORT secretion than pups born to ADXplacebo rats (Fig. 2).
PVN) between pups born to ADX dams replaced with placebo pellets and pups born to CORT-replaced ADX mothers.
Effect of CORT Replacement on Hypothalamic CRF Expression
We have previously shown that prenatal exposure to EtOH increased stress-induced ACTH secretion and augmented brain CRF biosynthesis in 21-day-old offspring of intact dams and that pups born to adrenalec-
Expression by in situ hybridization or measurement of CRF mRNA levels (Figs. 3 and 4) did not indicate any significant difference (based on the number of silver grains distributed over the parvocellular division of the
TABLE
2
Levels of Circulating Corticosterone in ADX Dams either Replaced with CORT or Placebo Pellet and Intact Dams Exposed to EtOH during the Second Week of Gestation Treatment
Corticosterone
ADX dams Placebo CORT Intact dams Control EtOH Note. ADX, adrenalectomized. ’ P < 0.01 vs ADX-placebo * P < 0.01 vs control dams. cd P > 0.05.
DISCUSSION
&
PLACEBO
W
CORT
T
I
I-+
-IL
(rig/ml)
”
6.6 + 2.0d 147.4 -c 11.0”” 8.9 -+ 3.4d 138.1 + 37.7*$
dams.
0
FEMALE
+ FIG. 2.
MALE
Basal
-+
MALE
FEMALE
+
Stress
+
Effect of CORT treatment on plasma CORT secretion under basal or stress conditions. Twenty-one-day-old pups were born to ADX mothers replaced with either CORT (200 mg/7 days) or placebo pellets during the second week of gestation, left undisturbed for 3 h, and decapitated immediately (basal) or after lo-min intermittent electroshocks (stress). Each bar represents mean CORT levels + SEM of 20-30 rats. **p G 0.01; *P s 0.05.
EXOGENOUS
FIG. replaced
3. Dark-field photograph with either placebo pellets
CORT
TREATMENT
AND
HPA
(X320) of the parvocellular division of the paraventricular nucleus (A) or CORT pellets (B) during the second week of gestation.
tomized mothers similarly exposed to ethanol failed to exhibit this increased HPA axis activity (8). This suggested an important mediating role of adrenal factors in the offspring’s responsiveness to prenatal alcohol exposure. Specifically, the observation of increased corticosterone levels in intact pregnant dams exposed to alcohol led us to hypothesize that this increase might mediate the altered endocrine status of their offspring. However, as shown by the present studies, exogenous
121
AXIS
of 21-day-old
pups
born
from
ADX
mothers
corticosterone replacement, at doses similar to those measured in intact, alcohol-exposed dams, did not mimic the effect of prenatal EtOH exposure. The interactions between the maternal and the fetal HPA axis are complex. Alcohol crosses the placenta readily (13,14) and can influence the fetal brain. However, the fetal HPA axis is not functional until the middle of the third week of gestation (15-17). It is thus improbab’le that alcohol, which was administered only until
LEE
AND
RIVIER
and/or effects of alcohol unrelated remains to be determined.
to adrenal function,
ACKNOWLEDGMENTS This research was supported by NIAAA Grant AA06420 to C.R. and Training Grant AA07456 to S.L. This research was conducted in part by the Clayton Foundation for Research California Division. C.R. is a Clayton Foundation investigator. We are grateful to B. Jewell, C. Gonzalez, S. Henson, Y. Haas, S. Johnson, and S. Smith for excellent technical assistance, and B. Coyne for secretarial assistance.
PLACEBO
REFERENCES
CORT
FIG. 4. Bar graph summarizes the effect of CORT treatment on the number of grains counted in the PVN of 2l-day-old pups born from ADX dams replaced with either placebo or CORT pellet. Each bar represents the mean f SEM of 4 animals in each group.
1.
2. 3.
Day 14 of gestation, could exert effects on the embryo’s PVN similar to those observed in adult animals. It is also necessary to emphasize that the increased CRF expression we measured following prenatal alcohol exposure was observed in 21-day-old pups. While these pups showed increased responsiveness of their HPA axis in response to shocks, younger animals are reported to exhibit a suppressed stress response (18-20). It is therefore possible that the changes which we observed (8) represent compensatory mechanisms caused by an earlier decrease in HPA axis activity, possibly resulting from alcohol-induced changes in adrenal and/or pituitary responsiveness (18). If this is the case, exogenous corticosterone may not mimic the effect of alcohol, as increased levels of this steroid represent only part of the HPA axis’ response to the drug. There is also evidence that in the developing fetus, CRF gene expression may be insensitive to corticosteroid feedback (21). If correct, this finding would indicate that alcohol primarily acts through mediators other than corticoids. While many such mediators are likely candidates, it is important to remember that adrenals also secrete secretagogues such as catecholamines, which exert powerful effects on the HPA axis. Furthermore, corticosterone treatment causes an alteration in brain catecholamines directly and in turn has a modulating role on the HPA axis (22). The potential role of these compounds as mediators of the endocrine influence of prenatal alcohol exposure, and the consequence of their removal in ADX dams, has not been evaluated. In conclusion, we have observed that exogenous corticoid treatment of adrenalectomized dams does not mimic the effect of prenatal EtOH exposure on the HPA axis’ activity of the offspring. Whether this reflects a relative insensitivity of the embryos’ HPA axis to steroid feedback, the influence of adrenal factors other than corticosterone,
Abel, E. L. (1978). Effects of ethanol on pregnant rats and their offspring. Psychopharmacology 67: 5-11. Cogan, D. C., L. E. Cohen, and G. Sparman (1983). Effects of gestational alcohol on the development of neonatal reflexes in the rat. Neurobehnv. Toxicol. Teratol. 5: 517-522. Streissguth, A. P. (1984). The behavioral teratology of alcohol: Performance, behavioral, and intellectual deficits in prenatally exposed children. In Alcohol and Brain Development (J. West, Eds.), pp. 3-44. Oxford, New York.
4. Jones,
5. 6. 7.
K. L., D. W. Smith, C. N. Ulleland, and A. P. Streissguth (1973). Patterns of malformation in offspring of chronic alcoholic mothers. Lancet 1: 1267-1271. Taylor, A. N., B. J. Branch, S. H. Liu, and N. Kokka (1982). Longterm effects of fetal ethanol exposure on pituitary-adrenal responses to stress. Pharmacol. B&hem. Behuv. 16: 585-589. Weinberg, J., and P. V. Gallo (1982). Prenatal ethanol exposure: Pituitary-adrenal activity in pregnant dams and offspring. Neurobehav. Toricol. Teratol. 4: 515-520. Taylor, A. N., B. J. Branch, L. R. Nelson, L. A. Lane, and R. E. Poland (1986). Prenatal ethanol and ontogeny of pituitary-adrenal responses to ethanol and morphine. Alcohol 3: 255-259.
8.
Lee, S. Y., T. Imaki, W. Vale, and C. R. Rivier (1990). Effect of prenatal exposure to ethanol on the activity of hypothalamicpituitary-adrenal axis of the offspring Importance of the time of exposure to ethanol and possible modulating mechanisms. Mol. Cell. Neurosci. 1: 168-177.
9.
Rivier, stress Res.
C., and W. Vale (1988). Interaction on ACTH and @-endorphin secretion.
between ethanol Alcohol. Clin.
and Exp.
12:206-210.
10.
Orth, D. (1979). Adrenocorticotrophic Methods of Hormone Radioimmuncussay Eds.), 2nd ed., pp. 245-284. Academic
11.
Gwosdow-Cohen, A., C. L. Chen, and E. L. Besch (1982). munoassay (RIA) of serum corticosterone in rats. Proc. Biol. Med. 170:29-34.
12.
Simmons, D. M., J. L. Arriza, and L. W. Swanson (1989). A complete protocol for in situ hybridization of messenger RNAs in brain and other tissues with radiolabeled single-stranded RNA probes. J. Histotechnol. 12: 169-181. Eguchi, Y. (1969). Interrelationships between the fetal and maternal hypophyseal-adrenal axes in rats and mice. In Physiology and Pathology of Adaptation Mechanics (E. Bajusz, Eds.), pp. 3-27. Pergamon Press, New York.
13.
hormone (ACTH). In (B. Jaffe and H. Behrman, Press, New York. RadioimSot. Exp.
14.
Weinberg, J., and S. Bezio (1987). Alcohol-induced tuitary-adrenal activity during pregnancy. Alcohol.
changes in piClin. Exp. Res.
15.
Eguchi, Y., 0. Hirai, Y. Morikawa, and Y. Hashimoto (1973). ical time in the hypothalamic control of the pituitary-adrenal
11:274-280. Critsys-
EXOGENOUS
16.
17.
18. 19.
CORT
TREATMENT
tern in fetal rats: Observations in fetuses subjected to hypervitaminosis A and hypothalamic destruction. Endocrinology 93: l-11. Yamamoto, M., Y. Eguchi, M. Asari, and Y. Kano (1983). Developmental changes in fetal adrenal hypertrophy following maternal bilateral and fetal unilateral adrenalectomy at different stages of gestation in the rat. Biol. Neon&e 43: 43-49. Bugnon, C., D. Fellmann, A. Gouget, and J. Cardot (1982). Ontogeny of the corticoliberin neuroglandular system in rat brain. Nature 298: 159-161. Weinberg, J. (1989). Prenatal ethanolexposure alters adrenocortical development of offspring. Alcohol. Clin. Exp. Res. 13: 73-83. Levine, S., and G. D. Coover (1976). Environmental control of suppression of the pituitary-adrenal system. Physiol. Behav. 17: 35-37.
AND
HPA
AXIS
123
20.
Levine, S. (1967). Maternal and environmental adrenocortical response to stress in weanling 258-260.
influences on the rats. Science 156:
21.
Jingami, H., S. Matsukura, S. Numa, and H. Imura (1985). Effects of adrenalectomy and dexamethasone administration on the level of prepro-corticotropin-releasing factor messenger ribonucleic acid (mRNA) in the hypothalamus and adrenocorticotropin/&Lipotropin precursor mRNA in the pituitary in rate. Endocrinology 117: 1314-1320.
22.
Jhanwar-Uniyal, M., K. J. Renner, M. T. Bailo, V. N. Luine, and S. F. Leibowitz (1989). Corticosterone-dependent alterations in utilization of catecholaminee in discrete areas of rat brain. Brain Res. 500: 247-255.