help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2005-1496
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
147/4/1904    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by McArthur, S.
Right arrow Articles by Gillies, G. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by McArthur, S.
Right arrow Articles by Gillies, G. E.
Endocrinology Vol. 147, No. 4 1904-1915
Copyright © 2006 by The Endocrine Society

Perinatal Glucocorticoid Treatment Disrupts the Hypothalamo-Lactotroph Axis in Adult Female, But Not Male, Rats

S. McArthur, Z.-L. Siddique, H. C. Christian, G. Capone, E. Theogaraj, C. D. John, S. F. Smith, J. F. Morris, J. C. Buckingham and G. E. Gillies

Department of Cellular and Molecular Neuroscience, Division of Neuroscience and Mental Health, Imperial College (S.M., Z.-L.S., E.T., C.D.J., J.C.B., G.E.G.), London W12 0NN, United Kingdom; Respiratory Health Services Research Group, National Heart and Lung Institute, Imperial College (S.F.S.), London W6 8RF, United Kingdom; Department of Human Anatomy and Genetics, University of Oxford (H.C.C., J.F.M.), Oxford OX1 3QX, United Kingdom; and Dipartemento di Scienze Farmaceutiche, Universita di Modena e Reggio Emilia (G.C.), 41100 Modena, Italy

Address all correspondence and requests for reprints to: Dr. Glenda E. Gillies, Department of Cellular and Molecular Neuroscience, Division of Neuroscience and Mental Health, Imperial College London, Hammersmith Campus, Du Cane Road, London W12 0NN, United Kingdom. E-mail: g.gillies{at}imperial.ac.uk.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Conclusions
 References
 
This study aimed to test the hypothesis that the tuberoinfundibular dopaminergic neurons of the arcuate nucleus and/or the lactotroph cells of the anterior pituitary gland are key targets for the programming effects of perinatal glucocorticoids (GCs). Dexamethasone was administered noninvasively to fetal or neonatal rats via the mothers’ drinking water (1 µg/ml) on embryonic d 16–19 or neonatal d 1–7, and control animals received normal drinking water. At 68 d of age, the numbers of tyrosine hydroxylase-positive (TH+) cells in the arcuate nucleus and morphometric parameters of pituitary lactotrophs were analyzed. In control animals, striking sex differences in TH+ cell numbers, lactotroph cell size, and pituitary prolactin content were observed. Both pre- and neonatal GC treatment regimens were without effect in adult male rats, but in females, the overriding effect was to abolish the sex differences by reducing arcuate TH+ cell numbers (pre- and neonatal treatments) and reducing lactotroph cell size and pituitary prolactin content (prenatal treatment only) without changing lactotroph cell numbers. Changes in circulating prolactin levels represented a net effect of hypothalamic and pituitary alterations that exhibited independent critical windows of susceptibility to perinatal GC treatments. The dopaminergic neurons of the hypothalamic periventricular nucleus and the pituitary somatotroph populations were not significantly affected by either treatment regimen in either sex. These data show that the adult female hypothalamo-lactotroph axis is profoundly affected by perinatal exposure to GCs, which disrupts the tonic inhibitory tuberoinfundibular dopaminergic pathway and changes lactotroph morphology and prolactin levels in the pituitary and circulation. These findings provide new evidence for a long-term disruption in prolactin-dependent homeostasis in females, but not males, after inappropriate GC exposure in perinatal life.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Conclusions
 References
 
GLUCOCORTICOID (GC) HORMONES produced by the adrenal cortex are essential for the development of many tissues and organs. There are, however, a number of circumstances in which the developing organism is exposed to elevated levels of GCs, including stress to the pregnant or nursing mother as well as in cases of premature birth, where the use of synthetic GCs to mature the lung is widespread. Growing evidence in many species, including humans, rodents, and domestic animals (1, 2, 3) suggests that such inappropriately elevated levels of GCs during critical developmental periods can have long-lasting deleterious effects on the offspring (1, 2, 3). For example, prenatal treatment with the synthetic GC, dexamethasone, or maternal restraint stress during the last week of pregnancy produce similar metabolic, cardiovascular, and behavioral changes in the adult offspring (2, 4). In the search for an underlying mechanism for these programming effects, particular attention has focused on the permanent changes in endocrine and metabolic processes that predispose to the development of chronic diseases in adulthood. Specifically, the hypothalamo-pituitary-adrenal (HPA) axis has been identified as an endocrine axis that is highly susceptible to programming (1, 2, 3, 5). In the adult rat, for example, GC treatment and exposure to stressors during the last week of pregnancy similarly reduce GC receptor (GR) expression in brain regions regulating negative feedback within the HPA stress axis (4). The associated increase in adrenal GC secretion is thought to represent a key factor in the development of insulin resistance and hypertension after exposure to adverse early life events in rodents and humans (6, 7).

Surprisingly, the potential for early life GC exposure to program endocrine axes other than the HPA axis has received relatively little attention. Although best known for its actions on the mammary gland, prolactin, like the GCs, is a pleiotropic hormone with over 300 identified physiological functions. It plays an important homeostatic role, including regulation of immune and cardiovascular responses, osmotic balance, reproduction, and sexual behaviors (8), all of which may be compromised by early adverse events and GC exposure (2, 5, 9, 10, 11). We propose, therefore, that disruption of the developing hypothalamo-lactotroph axis might represent another important mechanism that underlies the pathophysiological changes resulting from early elevations in GC levels.

To test this hypothesis, the first aim of the present study was to establish whether the hypothalamic dopaminergic populations that provide the major, tonic inhibitory regulation of prolactin secretion are targets for the programming effects of GCs. These comprise the tuberoinfundibular dopaminergic (TIDA) neurons located throughout the arcuate nucleus (A12 cell group) that release dopamine (DA) into the hypothalamo-pituitary portal circulation (8, 12). A minor portion of the arcuate DA neurons, the tuberohypophyseal dopaminergic pathway, terminate in the neural and intermediate lobes of the rat pituitary gland, but may also regulate prolactin release, because their DA can reach the anterior lobe via the short portal vessels (8). For comparison, we have also investigated the DA neurons of the periventricular nucleus (PeN; A14 cell group) that project to the intermediate lobe of the hypophysis (the PHDA system). This population is thought not to play a major role in regulating prolactin release, although, like the TIDA and tuberohypophyseal dopaminergic neurons, they appear to be responsive to prolactin (13). In the case of the HPA axis, the effects of GC treatment in late pregnancy, which result in axis hyperreactivity in adult rat offspring (2, 3, 4), appear to contrast with reports of a reduction in basal and stress-induced HPA activation in adulthood after GC treatment during the first week of life (14, 15). Moreover, the profile of GC-induced effects in the female differs from that in the male (3, 4). Therefore, in view of the significance of sex and the window of exposure to GCs during development, our studies included male and female rats that were exposed either prenatally or neonatally to the synthetic GC, dexamethasone. Because the parenteral route of steroid administration constitutes a stress and is likely to trigger the release of endogenous GCs as well as other mediators, we have used our noninvasive method for delivery of dexamethasone via the maternal drinking water (16, 17) and determined its effects on adult hypothalamic DA cell numbers and nuclear volume. Using similar methods, we have recently shown that the size of the midbrain DA populations in the substantia nigra pars compacta (A9) and the ventral tegmental area (A10), is highly sensitive to perinatal GC programming (16). Our recent work has also identified the anterior pituitary gland as an important site for GC programming of corticotroph cell numbers, GC sensitivity, and folliculostellate (FS) cell morphology (17). Therefore, another aim of the present study was to explore the effects of pre- and neonatal dexamethasone treatment on lactotroph cell profiles and morphology as well as prolactin levels in the pituitary and circulation of adult male and female rats. For comparison, the impact of perinatal GC treatments on somatotroph populations was also assessed.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Conclusions
 References
 
Animals
All animal work was carried out under license in accordance with the United Kingdom Animals (Scientific Procedures) Act of 1986. Adult male and female Sprague Dawley rats were purchased for breeding (Harlan Olac, Blackthorn, Beicester, UK) and housed in the Comparative Biology Unit at Charing Cross Hospital (Faculty of Medicine, Imperial College, London, UK) under controlled lighting (lights on at 0800 h, off at 2000 h), temperature (21–23 C), and humidity (63%), with standard rat chow and drinking water (except as described below) provided ad libitum. On arrival, male and female rats, caged separately, were acclimatized to their new environment for 1 wk, after which groups of two female rats and one male rat were housed together overnight. The presence of vaginal plugs the following morning confirmed mating, and pregnancy was confirmed approximately 6 d later by palpation. The timed pregnant rats were housed five per cage until gestational d 15 (GD 15), when they were caged singly in preparation for giving birth. From GD 19/20, pregnant rats were monitored several times a day, and the day of birth was designated d 0. Offspring were weaned at 3 wk, and from then on, male and female animals were housed separately in standard, wire-topped cages in groups of five per cage and were allowed to grow to young adulthood with no additional treatment. At 68 ± 2 d of age, animals were killed by decapitation between 0900 and 1000 h to minimize effects associated with circadian rhythms. Studies in female rats have shown that at this time of day the concentrations of DA and its major metabolite, 3,4-dihydroxyphenylacetic acid, in the anterior pituitary gland and, hence, the activity in the TIDA pathway are constant throughout the 4-d estrous cycle (18). Therefore, female rats were used at random stages of the cycle to avoid the possible effects of handling that collection of cervical smears would introduce into the experimental design.

Dexamethasone treatment regimens
Numerous studies of the impact of perinatal GC treatment on adult physiology have administered the drug by injection of the pregnant dam or newborn pups, but in preliminary studies we found that the injection stress per se altered levels of DA in the adult brain (Gillies, G. E., and S. McArthur, unpublished observations). Therefore, using our recently established method (17), we took advantage of the oral bioavailability of dexamethasone and its ability to cross the placenta (19) and enter the milk (20) for its delivery to the developing pups. Dexamethasone sodium phosphate (Faulding Pharmaceuticals Plc., Royal Leamington Spa, UK) was delivered noninvasively by addition of the drug to the drinking water of pregnant or nursing dams between embryonic d 16–19 (E16–E19) or neonatal d 1–7 (P1–P7) at a concentration of 1 µg/ml. Three litters were used for each of the dexamethasone treatment groups and the control group (normal drinking water) and males and females (n = 8/group) were randomly selected from these litters for the analyses described below. The treatment periods used were chosen to model the stages of human brain development when GCs may be used therapeutically to promote fetal lung maturation in cases of threatened premature birth (16, 17). Hence, E16–E19 reflects the developmental stage at which the earliest premature infants receive GCs, whereas P1–P7 mirrors the period in which less premature infants are treated (21). Based on the volume of water the dams drank, maternal consumption was 50 ± 8 µg/d in late gestation and 54 ± 3 µg/d for the first week postpartum. We were unable to make direct measurements of circulating levels of dexamethasone in the fetal and neonatal rat pups, but, as explained in detail previously (17), we have estimated that plasma concentrations are in the region of 40 or 15 ng/ml, respectively, based on maternal drug intake, body mass (mother and developing young), and available pharmacokinetic data (volume of distribution, maternal to fetal plasma gradient, milk to plasma gradient, and average neonatal fluid intake). Taking into account the many assumptions in our calculations as well as individual variations in fluid intake, it would appear that exposure to dexamethasone is of a similar order of magnitude in the two treatment groups (17). Because studies of receptor binding and antiinflammatory activity suggest that the GC potency of dexamethasone is, respectively, 1 or 2 orders of magnitude greater than that of corticosterone (22), it would appear that the GC levels attained might be in the region of those that prevail after stress-induced activation of the maternal HPA axis (17).

TH immunohistochemistry
After decapitation, brains were rapidly removed (n = 8/treatment group) and immersed in 4% formaldehyde in 0.1 M PBS (pH 7.4) for 1 wk, cryoprotected in 20% sucrose in PBS, and frozen at –80 C. Coronal slices of the hypothalamus (20 µm) were cut to include both the arcuate nuclei (between approximately –2.3 and –3.6 mm relative to bregma) and the periventricular nuclei (between approximately –1.4 and –2.3 mm relative to bregma) using a cryostat (Bright Instruments Ltd., Huntingdon, UK) maintained at –22 C and were stored in antifreeze solution [0.1 M NaH2PO4·H2O, 0.05 M Na2HPO4, 0.15 mm NaCl, 50% (vol/vol) ethanediol, 1% (wt/vol) polyvinylpyrrolidone, and 0.1% NaN3] at –20 C until immunostaining for TH as described previously (16). Briefly, sections were rinsed in PBS before incubation for 1 h at 22 C in 20% normal goat serum (NGS) in PBS (Serotec, Oxford, UK) to saturate nonspecific binding sites. Sections were permeabilized with 0.05% Triton X-100 in 1% NGS for 5 min before being incubated overnight at 22 C under constant agitation with rabbit antirat TH primary antibody (Chemicon, Chandlers Ford, UK) diluted 1:2000 in 1% NGS. After three rinses in 1% NGS, cells were visualized using an ABC Vectorstain kit (Vector Laboratories, Peterborough, UK), 0.025% 3,3'-diaminobenzidine tetrahydrochloride, and 0.01% hydrogen peroxide. Sections were mounted on gelatin-coated microscope slides and were allowed to air-dry before being coverslipped.

The TH-immunoreactive (TH-IR) cells within the arcuate and periventricular nuclei were counted using an image analysis software package (Image ProPlus 4.5, Media Cybernetics, Finchampstead, UK) as described previously (16, 23). A digital image (magnification, x100) of each section was captured using a CoolSNAP-Pro camera (Roper Scientific, Marlow, UK) attached to an Eclipse E800 microscope (Nikon UK Ltd., Kingston-upon-Thames, UK). To ensure objectivity, images were coded and stored in a manner that rendered the person performing the cell analysis unaware of the treatment groups. Images were projected onto a personal computer monitor, and cumulative manual counts were made of TH-IR cells in each hemisphere in every second 20-µm coronal section of the hypothalamus, encompassing the full extent of TH immunoreactivity in both the arcuate nucleus (20–22 sections counted/animal) and periventricular nuclei (10–11 sections counted/animal; see Fig. 1Go). Counts were then summated and doubled (to account for the total number of sections) to give an estimate of the total TH-IR population in each nucleus per animal, and the individual values were pooled to give mean values for treatment groups (n = 8 animals/group). This profile-based counting technique for serial reconstruction has proven the most appropriate method when cell numbers are relatively low (24). Furthermore, this method, which we have employed previously (16), counts across the whole of the coronal plane of the arcuate and periventricular nuclei, where TH immunohistochemistry delineates the boundaries of the dopaminergic nuclei. This provides a large sampling window, enabling highly accurate assessment in the x, y plane, comparing favorably with three-dimensional counting methods (25). The volumes of the arcuate and periventricular nuclei, as delineated by the presence of TH-IR cells, were also estimated, using the method of Cavalieri et al. (26). Briefly, after projection of an image of the section at x40 magnification on a personal computer monitor, as described above, the cross-sectional area occupied by TH-IR cells for each nucleus in every second section was measured and multiplied by the thickness of the tissue section, with allowance made for tissue shrinkage (estimated as 3.3 ± 1.7% by measurement with an electronic microcator) (27). These values were summated and doubled (to account for measurement of alternate sections) to estimate volumes for each animal, and values were pooled to calculate the overall group means (n = 8). Again, to ensure objectivity, the measurements of cross-sectional area were conducted blind to treatment group.


Figure 1
View larger version (73K):
[in this window]
[in a new window]
  		FIG. 1. Representative photomicrographs of the arcuate nucleus (A) and the PeN (B) of the hypothalamus (magnification, x40) of tissue sections immunostained for TH.

 
Hormone determinations
After decapitation (n = 8/treatment group), trunk blood was collected, allowed to clot on ice, and centrifuged at 1500 x g for 10 min at 4 C. The serum fraction was retained and frozen at –80 C for later measurement of prolactin content by RIA. Immediately after decapitation, anterior pituitary glands were removed and snap-frozen at –40 C. The tissue was then homogenized by sonication (15 sec, 10 µm amplitude, 4 C; Soniprep 150, Sanyo, UK) in homogenization buffer [0.1 M PBS, 1 mm EDTA, 0.2% (wt/vol) phenylmethylsulfonylfluoride, and 1% (vol/vol) aprotinin (pH 7.4)] and centrifuged at 15,000 x g for 10 min at 4 C, whereupon supernatants were removed and stored at –80 C until required for prolactin and GH measurements by RIA. The prolactin assay was performed using a rabbit antirat prolactin antibody (National Hormone and Pituitary Program, Torrance, CA) with a donkey antirabbit IgG precipitating antibody (IDS, Tyne & Wear, UK) for separation (28). Pituitary samples were diluted 1:20 (male) and 1:40 (female) in assay buffer (0.01 M PBS and 0.5% BSA, pH 7.4), whereas serum samples were assayed undiluted. The limit of detection was 1.95 ng/ml. Pituitary GH content was measured using a monkey antirat GH primary antibody (National Hormone and Pituitary Program) with a 1:1 mixture of heat-inactivated horse serum (Sigma-Aldrich Corp., Poole, UK) and 0.9% NaCl for separation (29). The limit of detection was 1.25 ng/ml. Pituitary samples were diluted 1:4000 in assay buffer [0.01 M PBS and 0.5% BSA (pH 7.4)]; serum levels of GH were not analyzed due to the difficulty in obtaining meaningful data from single-sample analysis, given the highly periodic nature of pituitary GH secretion. To eliminate interassay variation and enable direct comparisons of treatments within and between sexes, all pituitary extracts were included in a single prolactin or GH assay, as were the serum samples.

Electron microscopic analysis
After decapitation, anterior pituitaries (n = 4/treatment group) were fixed in 3% paraformaldehyde, 0.05% (vol/vol) glutaraldehyde (VWR International Ltd., Lutterworth, UK) in PBS (Oxoid Chemicals Ltd., Basingstoke, UK; pH 7.2) for 4 h at 22 C before being processed and embedded into LR Gold resin as previously described (30). Sections were cut using a Reichart-Jung Ultracut microtome (Nussloch, Germany) and were prepared either for Immunogold labeling of prolactin by mounting onto Formvar-coated mesh nickel grids (Agar Scientific Ltd., Stanstead, UK) or for immunofluorescent analysis of lactotroph numbers. To determine the specificity of perinatal treatments, a parallel set of sections was processed in a similar manner for Immunogold or immunofluorescence labeling of GH in somatotrophs.

Immunogold labeling for morphological analysis of lactotrophs or somatotrophs
Ultrathin sections (50–80 nm) mounted on mesh grids were incubated for 2 h with anti-prolactin (1:5000; National Hormone and Pituitary Program) or for 2 h with rabbit antirat GH (1:5000; National Hormone and Pituitary Program) and for 1 h with protein A-15 nm gold complex (Biocell, Cardiff, UK), then lightly stained with uranyl acetate and lead citrate (31). The specificity of antibody staining was confirmed by use of preadsorbed antibody (100-fold excess of recombinant prolactin or GH) and albumin-containing buffer in place of the primary antiserum. Sections (50–80 nm) were viewed with a JEM-1010 transmission microscope (JEOL USA, Inc., Peabody, MA). On the basis of their secretory granule populations (shape, electron density, size, and distribution; see Fig. 2Go for typical examples) and Immunogold labeling (31), a total of 10 lactotrophs and 10 somatotrophs were selected per animal (n = 4/treatment group) from three sections taken systematically at different levels of the pituitary gland according to a systematic random procedure (32). Electron micrographs were taken at a magnification of x4000, and each cell was scanned and analyzed using Image ProPlus 4.5 software (Media Cybernetics, Finchampstead, UK) to measure granule number, nuclear area, total cell area, cytoplasmic area (total cell area – nuclear area), and granule density (granule number/cytoplasmic area). Images were analyzed in such a way as to render the person performing the analysis unaware of the treatment groups. Mean values for each cell type were calculated for each animal, and these were combined to give an overall mean for each treatment group.


Figure 2
View larger version (91K):
[in this window]
[in a new window]
  		FIG. 2. Representative electron micrographs (magnification, x4000) of a lactotroph (left) and a somatotroph (right) from control female animals.

 
Immunofluorescent analysis of lactotroph and somatotroph densities
Semithin (1 µm) LR Gold sections were incubated with either rabbit antiprolactin (1:5000; National Hormone and Pituitary Program) or rabbit antirat GH (1:5000; National Hormone and Pituitary Program), followed by antirabbit IgG biotinylated secondary antibody (Vector Laboratories, Inc., Burlingame, CA) and visualized using fluorescein avidin D (Vector Laboratories, Inc.) at 1:250 in PBS. Sections were mounted in Vectashield mounting medium with propidium iodide to counterstain cell nuclei and were analyzed using a BX51 microscope (Olympus UK Ltd., Middlesex, UK) and Axiovision 4.2 imaging software (Carl Zeiss Ltd., Herts, UK). The proportions of lactotrophs and somatotrophs were quantified by counting the number of prolactin- and GH-immunopositive cells and the total number of nucleated cells, dividing the prolactin- and GH-positive cell number by the total number of nucleated cells and converting to a percentage. Cells were counted in a fixed area (0.25 mm2) of four complete sections taken from different depths of the tissue block for each animal, and the mean values for each animal were pooled to give the group means (n = 4 animals/group). The total number of cells per section was not significantly different between the treatment groups.

Statistical analysis
Statistical analysis was carried out using Jandel SigmaStat 2.0 software (Jandel Corp., San Ramon, CA). Preliminary analysis was undertaken to show that data were normally distributed (Kolmogorov-Smirnov test) and that variances were equal (Levene Median test). Subsequent analysis was carried out by two-way ANOVA, with post hoc analysis performed using Bonferroni’s corrected Student’s t test. In all cases, differences were considered to be significant at P < 0.05.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Conclusions
 References
 
Exposure of fetal or newborn rats to dexamethasone via the maternal drinking water had no significant effect on adult body weight in either male (control, 318.5 ± 1.9 g; prenatal exposure, 323.6 ± 8.0 g; neonatal exposure, 317.9 ± 9.7 g) or female (control, 209.1 ± 6.1 g; prenatal exposure, 215.6 ± 5.1 g; neonatal exposure, 215.8 ± 4.5 g) animals.

TH-IR cell numbers
A significant and sexually dimorphic effect of perinatal dexamethasone treatment was seen in the number of TH-IR cells in the arcuate nucleus of rats in adulthood (F2,38 = 9.04; P < 0.001), as shown in Fig. 3AGo. In females, both prenatal and neonatal dexamethasone exposures resulted in a significant decrease in adult cell numbers compared with control animals (P < 0.05, by Bonferroni’s t test), whereas males were unaffected by either treatment regimen. Furthermore, a significant gender difference was apparent (F1,38 = 32.903; P < 0.001), with females possessing significantly greater numbers of TH-IR cells in the arcuate nucleus than males in both the control and neonatally treated groups, but not in the prenatal treatment group (P < 0.05, by Bonferroni’s t test). Unlike the arcuate nucleus, neither the female nor the male PeN exhibited significant changes in TH-IR cell number after either perinatal treatment regimen (Fig. 3BGo).


Figure 3
View larger version (27K):
[in this window]
[in a new window]
  		FIG. 3. Effects of prenatal (E16–E19) or neonatal (P1–7) dexamethasone treatment (1 µg/ml in the maternal drinking water) on TH-IR neuronal number in the arcuate nucleus (A) and the hypothalamic PeN (B) of male ({square}) and female (Figure 3) rats in adulthood. Controls received normal drinking water. Data are the mean ± SEM (n = 8 for all groups). *, P < 0.05, male vs. female; +, P < 0.05 vs. control group of the same gender.

 
Arcuate and periventricular nuclear volumes
Although the total volumes of the arcuate nucleus and PeN were unaffected by perinatal glucocorticoid exposure, a significant main effect of gender was observed in the volume of both the arcuate nucleus (F1,34 = 16.002; P < 0.001) and the PeN (F1,29 = 6.719; P = 0.015), with the volume of both nuclei being greater in females than in males (P < 0.05 in both cases, by Bonferroni’s t test; see Table 1Go).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Effects of prenatal (E16–19) or neonatal (P1–7) dexamethasone treatment (1 µg/ml in the maternal drinking water) on the volume of the arcuate nucleus and periventricular nucleus of the hypothalamus in male and female rats in adulthood

 
Pituitary hormone content
Significant effects of both sex (F1,40 = 26.216; P < 0.001) and dexamethasone treatment (F2,40 = 9.257; P < 0.001) were seen on the level of prolactin in the anterior pituitary gland, with a significant interaction between the two factors (F2,40 = 3.602; P = 0.036). As shown in Fig. 4AGo, prolactin content in the adult anterior pituitary gland was unaffected by prenatal or neonatal GC treatment in males or by neonatal treatment in females. Prenatal GC exposure, however, markedly reduced pituitary prolactin content in females. The data show, also, that control females had approximately 6 times more prolactin in their pituitary gland than males (P < 0.05, by Bonferroni’s t test), and this sex difference was significantly attenuated after prenatal steroid treatment. Adult male and female anterior pituitary gland weights were not significantly affected by early dexamethasone treatment.


Figure 4
View larger version (24K):
[in this window]
[in a new window]
  		FIG. 4. Effects of prenatal (E16–E19) or neonatal (P1–P7) dexamethasone treatment (1 µg/ml in the maternal drinking water) on prolactin content (A) and GH content (B) in the anterior pituitary of male ({square}) and female (Figure 4) rats in adulthood. Controls received normal drinking water. Data are the mean ± SEM (n = 8 for all groups). *, P < 0.05, male vs. female; +, P < 0.05 vs. control group of the same gender.

 
A significant main effect of sex was seen in the GH content of the anterior pituitary (F1,40 = 26.037; P < 0.001), with males having significantly greater levels of the protein than females (P < 0.05, by Bonferroni’s t test), as shown in Fig. 4BGo. GH content was, however, unaffected by either prenatal or neonatal dexamethasone exposure in either males or females.

Pituitary morphology
As shown in Fig. 5AGo, neither prenatal nor neonatal dexamethasone exposure altered the proportion of total cells that were lactotrophs in either sex, but the results revealed a significantly greater percentage of lactotrophs in females compared with males. Figure 5BGo shows that the adult male and female pituitary somatotroph proportion was also unaffected by the perinatal treatment regimens, and unlike lactotrophs, there was no sex difference in the percentage of cells of this type.


Figure 5
View larger version (28K):
[in this window]
[in a new window]
  		FIG. 5. Effects of prenatal (E16–E19) or neonatal (P1–P7) dexamethasone treatment (1 µg/ml in the maternal drinking water) on the proportions of lactotroph (A) and somatotroph (B) cell numbers expressed as a percentage of the total number of nucleated cells in the anterior pituitary of male ({square}) and female (Figure 5) rats in adulthood. Controls received normal drinking water. Data are the mean ± SEM (n = 4 for all groups). *, P < 0.05, male vs. female.

 
In rodents, lactotrophs can be divided into two major morphological subtypes using electron microscopy. Type I cells contain relatively small numbers of large (>300 nm diameter) polymorphic electron-dense granules, and type II contain more numerous smaller (<200 nm diameter), spherical, electron-dense granules (33, 34). Analysis of type I adult lactotroph populations showed that total cell area, cytoplasmic area, nuclear area, and granule density were unaffected by either prenatal or neonatal GC exposure in males or by neonatal exposure in females (Fig. 6Go). In contrast, prenatal exposure to dexamethasone exerted sex-specific effects on total cell area (F2,17 = 4.453; P = 0.028) and cytoplasmic area (F2,17 = 7.285; P = 0.005) by reducing these parameters, but not nuclear area, in adult females (Fig. 6Go, A–C; P < 0.05, by Bonferroni’s t test); granule density remained unchanged (Fig. 6DGo). The data in Fig. 6Go also reveal inherent sex differences in the control animals, with a significantly greater total cell area in females compared with males (F1,17 = 28.482; P < 0.001), which appears largely attributable to an increase in cytoplasmic cross-sectional area (F1,17 = 46.765; P < 0.001) rather than nuclear area. In contrast, the density of secretory granules within the cytoplasm was greater in males than in females (F1,17 = 11.128; P = 0.004). Type II lactotrophs exhibited similar inherent sex differences in morphological parameters in control animals. However, the reduction in cytoplasmic area resulting from prenatal dexamethasone treatment in females was less pronounced than that for the type I cells and failed to significantly affect total cell area (Fig. 7Go).


Figure 6
View larger version (53K):
[in this window]
[in a new window]
  		FIG. 6. Effects of prenatal (E16–E19) or neonatal (P1–P7) dexamethasone treatment (1 µg/ml in the maternal drinking water) on total cell area (A), cytoplasmic area (B), nuclear area (C), and secretory granule density (D) of type I lactotrophs from the anterior pituitary of male ({square}) and female (Figure 6) rats in adulthood. Controls received normal drinking water. Data are the mean ± SEM (n = 4 for all groups). *, P < 0.05, male vs. female; +, P < 0.05 vs. control group of the same gender.

 

Figure 7
View larger version (49K):
[in this window]
[in a new window]
  		FIG. 7. Effects of prenatal (E16–E19) or neonatal (P1–P7) dexamethasone treatment (1 µg/ml in the maternal drinking water) on total cell area (A), cytoplasmic area (B), nuclear area (C), and secretory granule density (D) of type II lactotrophs from the anterior pituitary of male ({square}) and female (Figure 7) rats in adulthood. Controls received normal drinking water. Data are the mean ± SEM (n = 4 for all groups). *, P < 0.05 male vs. female; +, P < 0.05 vs. control group of the same gender.

 
Pituitary somatotroph morphology
Figure 8Go shows that none of the somatotroph morphometric parameters was significantly affected by either prenatal or neonatal dexamethasone treatment. The data, however, reveal a male-dominant sex difference in total cell area (Fig. 8AGo) due to a greater cytoplasmic area (Fig. 8BGo). There were no sex differences in nuclear area (Fig. 8CGo) and granule density per somatotroph (Fig. 8DGo).


Figure 8
View larger version (52K):
[in this window]
[in a new window]
  		FIG. 8. Effects of prenatal (E16–E19) or neonatal (P1–P7) dexamethasone treatment (1 µg/ml in the maternal drinking water) on total cell area (A), cytoplasmic area (B), nuclear area (C), and secretory granule density (D) of somatotrophs from the anterior pituitary of male ({square}) and female (Figure 8) rats in adulthood. Controls received normal drinking water. Data are the mean ± SEM (n = 4 for all groups). *, P < 0.05, male vs. female.

 
Serum prolactin concentrations
As reported by others (35), we found that basal serum prolactin levels were not significantly different between control male and female rats (Fig. 9Go) despite considerable differences in pituitary prolactin content. However, a sexually dimorphic effect of perinatal dexamethasone exposure on serum prolactin levels was evident (F2,40 = 4.025; P = 0.026). Although GC had no effect on circulating prolactin levels in adult males, levels in females were significantly increased by neonatal, but not prenatal, dexamethasone exposure (P < 0.05, by Bonferroni’s t test).


Figure 9
View larger version (29K):
[in this window]
[in a new window]
  		FIG. 9. Effects of prenatal (E16–E19) or neonatal (P1–P7) dexamethasone treatment (1 µg/ml in the maternal drinking water) on circulating prolactin levels in male ({square}) and female (Figure 9) rats in adulthood. Controls received normal drinking water. Data are the mean ± SEM (n = 8 for all groups). +, P < 0.05 vs. control group of the same gender.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Conclusions
 References
 
This study has identified a striking sex difference in the hypothalamo-lactotroph axis in terms of arcuate DA cell numbers and lactotroph morphology as well as prolactin content. The results also show, for the first time, that perinatal dexamethasone treatment produces long-term changes in this axis in female rats, but not in males, to defeminize or masculinize the female profile. Pituitary susceptibility, however, is restricted to the prenatal period, whereas hypothalamic susceptibility encompasses both the pre- and neonatal periods. These findings extend the concept that disruption of neuroendocrine function is a consequence of early life exposure to GC hormones and support the idea of independent, tissue-specific windows of GC vulnerability during development. In addition, they provide new evidence for sex differences in susceptibility to early programming events that may underpin sexually dimorphic disturbances in endocrine homeostatic mechanisms and, hence, altered physiology and disease susceptibility in adulthood.

Hypothalamic DA populations
At the level of the hypothalamus, it is well documented that females exhibit a higher basal activity in their TIDA system (36, 37, 38). In agreement with this we are the first to report, to the best of our knowledge, that the arcuate DA neuronal number in control females is significantly greater (by ~75%) than that in males, whereas there were no significant sex differences in the PeN. Notably, our results show that prenatal GC treatment abolished, and neonatal treatment markedly reduced, the sex differences in adult TIDA cell numbers by reducing them in the female. Because the overall volume of the nucleus was not affected by perinatal GC treatment, this would appear to be a genuine reduction in TH-IR cell density. Although difficult to measure directly, such a loss of DA neurons might reasonably be expected to result in a fall in the level of DA in the hypophyseal-portal system and a reduced inhibitory drive to the pituitary lactotrophs. In contrast to the effects in the arcuate nucleus, we have recently shown that identical perinatal GC treatments markedly increased TH-IR cell numbers in the adult midbrain substantia nigra pars compacta and ventral tegmental area in both sexes (16). Together with our observation that TH-IR cell numbers in the hypothalamic PeN were not significantly affected by perinatal GC treatments, these findings demonstrate enduring, population-specific as well as sex-specific influences on central DA populations.

Although the mechanisms that govern adult DA neuronal number in the arcuate nucleus are unknown, our GC treatments span important phases of their development. The prenatal regimen (E16–E19) began just after the birth of the diencephalic DA neurons (E14–E15) (39), and their numbers continue to increase into the neonatal period, until P21 in mice (40), which encompasses our neonatal regimen (P1–P7). Because GCs have potent influences on neurogenesis, neural migration, survival, and maturation (41, 42), and GRs have been detected in the arcuate nucleus as early as E15 (43), it is possible that dexamethasone may influence these processes directly in the female brain. It is not clear why the male arcuate DA populations are not similarly reduced in number, but the fact that our perinatal GC treatments also span a critical period for sexual differentiation of the brain is likely to be important in this respect. Specifically, a transient rise in testicular testosterone production on E18–E19 and again during the first 7–10 d of life in the rat is widely recognized to be critical for imprinting a male circuitry in the brain, which, otherwise, defaults to a female or neutral course (44, 45). To the best of our knowledge, there are no reports that this mechanism is responsible for generating sex differences in TIDA cell numbers. However, the organizational actions of the perinatal testosterone surge are thought to suppress DA cell numbers in the hypothalamic anteroventral periventricular nucleus, which, as we show here for the arcuate nucleus, are also significantly reduced in males compared with females (46). We propose, therefore, that due to the presence of endogenous testosterone perinatally in the male, an additional reduction in TIDA neurons by dexamethasone treatment may not be possible.

In addition to direct actions during development, it should be considered that the changes we found in the adult TIDA populations could be secondary to other changes induced by perinatal dexamethasone treatment. In particular, in adult animals, TIDA neurons are responsive to physiological levels of GCs (47, 48), and it is well established that perinatal exposure to GCs permanently alters the responsiveness of the HPA axis and the prevailing levels of GCs (2). These effects are, however, marked in males as well as females, and the pre- and neonatal GC influences appear to be qualitatively different (2, 3, 6, 14, 49). Thus, it is unlikely that these contrasting time- and sex-dependent effects of prenatal and neonatal GC treatments on HPA activity can be causative factors for the consistent effects of our pre- and neonatal dexamethasone treatment regimens that affect only female TIDA cell numbers. Moreover, evidence suggests that although prevailing levels of GCs in adult rats appear to influence the level of TH expression, the number of cells expressing TH does not change (50). The balance of evidence would thus favor a primary effect of GCs on the developing neurons.

Because pituitary hormones, especially prolactin, appear to be important for the maintenance of A12 TH-IR cells (40, 51), it could also be argued that changes in TIDA cell numbers may be secondary to GC-induced changes in prevailing levels of prolactin. However, this would not be compatible with our finding that in females treated neonatally with dexamethasone, TIDA cell number fell, whereas prevailing adult levels of prolactin rose. Prenatal GC treatment similarly affected TIDA cell numbers, but not prolactin levels. Our data therefore support our premise that the primary effect is within the developing hypothalamus. In support of this, it has been shown that in a mouse model of isolated prolactin deficiency, TIDA cell numbers are normal (52).

Lactotroph morphology and prolactin levels
In agreement with other reports (53, 54), we found a significantly greater proportion of lactotrophs and a greater prolactin content in the normal adult female pituitary compared with male glands. In addition, our data show that the cytoplasmic area of female type I and II lactotrophs is double that of males, and even though granule density is slightly greater in males (~25%), it would appear that the larger cell size in females makes a fundamental contribution to the sex differences in pituitary prolactin content. The present study also showed, for the first time, that GC exposure between E16–E19, but not P1–P7, abolished sex differences in prolactin content in the adult rat pituitary by reducing it in females without affecting males. Similarly, sex differences in cytoplasmic and total cell area in type I lactotrophs were abolished by prenatal, but not neonatal, GC treatment due to a 25–30% reduction in females, but not males. In contrast, cytoplasmic and total cell areas of type II lactotrophs remained sexually dimorphic after prenatal GC treatment, which had no significant effect on total cell area despite a small reduction in cytoplasmic area. Although the functional significance of the morphological heterogeneity of the lactotroph population is unclear, there is a notable sex difference in the abundance of the two main subtypes. In the male pituitary gland, type I and II lactotrophs are present with equal abundance, and they each represent about 12% of the total secretory population in the adult rat, whereas in females type I lactotrophs are more prevalent than type II, constituting approximately 35% and 7% of the total adult secretory population, respectively (34). Together, these observations suggest that the greater sensitivity to dexamethasone treatment between E16–E19 of type I lactotrophs, which are more abundant in females, may underlie the sexually dimorphic effect of prenatal steroid exposure on lactotroph morphology and pituitary prolactin content. These observations, therefore, extend our evidence that perinatal GC treatments influence pituitary morphology and that these effects are more prominent with the prenatal treatment regimen (17).

Because somatotrophs were unaffected by perinatal GC exposure (see below), the data discussed above suggest that E16–E19 represents a unique and critical period for sexual differentiation of the lactotroph population, which can be compromised by GC exposure. Our data do not allow us to determine when the sex differences in pituitary lactotroph morphology and prolactin content first appear or why only females during the prenatal period are susceptible to the disruptive effects of GC programming. However, evidence suggests that in male rats, it is the masculinizing effects of the perinatal testosterone surge that set the adult male level of pituitary prolactin (55). Sex differences in the prevailing levels of sex steroid hormones early in development are, therefore, likely factors that influence lactotroph susceptibility to dexamethasone treatment. Other studies suggest that sex differences in pituitary lactotroph number and prolactin content emerge at puberty (56, 57). It is possible, therefore, that the programming or organizational actions of prenatal GCs may disrupt the ability of the female lactotrophs to respond to the activational effects of the pubertal/adult sex hormone environment.

Although the factors that govern lactotroph cell size are not known, studies of the rat show that the timing of our treatment regimens (E16–E19 and P1–P7) encompassed notable developmental landmarks in lactotroph differentiation. For example, pituitary prolactin mRNA and protein levels are normally first detectable on E17–E17.5 (58, 59), and the numbers of cells synthesizing and expressing prolactin increase most markedly from E18–E19 to P10 (58). Moreover, evidence from fetal pituitary organ cultures (60) and in vivo suppression of corticosteroidogenesis during gestation (61) indicate a suppressive role of GCs on lactotroph cell number during development. However, our finding that the proportion of lactotrophs in the pituitaries of adult male and female rats are not subject to enduring effects of GC exposure during development supports the idea that the steroid’s influence on lactotroph cell numbers during development are due more to an effect on the timing of emergence of the phenotype rather than to an effect on the proliferation of prolactin progenitors, which would be more likely to alter the final numbers of committed lactotrophs (57, 60). Interestingly, our recent work has shown that perinatal GC treatment alters the morphology of FS cells, which play an important role in the development, growth, and differentiation of the anterior pituitary gland (17, 62). These cells also produce important paracrine mediators, such as IL-6 and annexin 1, that influence secretory cell function and growth, and we have shown that perinatal GC treatment impairs the ability of FS cells to sustain annexin 1-dependent GC regulation of ACTH release (17). Because annexin 1 also mediates some effects of GCs on lactotrophs (28), which do not appear to express GRs, at least in the adult pituitary gland (63), the GR-expressing FS cells are potential attractive targets for mediating some of the developmental influences of GCs on the lactotroph population.

As reported by others (35), we found no significant sex difference in basal circulating prolactin levels, which, in view of the marked differences in pituitary prolactin content and cell size in males and females, indicates significant sex differences in the dynamics of prolactin secretion. In male rats, adult serum prolactin levels were unaffected by perinatal dexamethasone treatments, which accords with an absence of effects at the hypothalamic and pituitary levels. In contrast, in females, influences on serum prolactin levels need to be interpreted in the light of combined long-term effects of perinatal GCs within the TIDA and lactotroph populations, which have different critical windows of susceptibility. Therefore, although the reduction in TIDA cell number in adult female rats after prenatal dexamethasone treatment might predict a reduction in hypothalamic inhibitory tone on prolactin release and a rise in circulating prolactin levels, prenatal dexamethasone treatment also had effects at the pituitary level that might predict reduced prolactin release (disrupted lactotroph morphology and reduced pituitary prolactin content). The net effect could therefore account for the unchanged serum prolactin levels in the offspring treated prenatally with dexamethasone. In contrast, the reduction in adult TIDA cell number after neonatal dexamethasone treatment is accompanied by normal lactotroph morphology and pituitary prolactin content in female rats, so the increase in circulating prolactin found in this treatment group is consistent with a reduction in hypothalamic inhibitory tone.

Somatotroph morphology and GH levels
In accord with other reports, we found that GH content was greater in adult male pituitaries (53, 54). Our morphological data allow us to attribute this to a greater cytoplasmic area and an increased total number of secretory granules per somatotroph in males compared with females and not with a sex difference in somatotroph cell number, as suggested in other reports (55, 64). In the present study, however, we used sensitive immunocytochemical methods to identify the hormone-producing cells directly, whereas others have identified individual cells on the basis that that they are actively secreting GH, using the hemolytic plaque assay (64), or expressing GH mRNA (55). It is possible, therefore, that cell-counting methods based on secretory or synthetic activity may not be as accurate in discerning sex differences in cell numbers as methods that rely on detecting stored hormone. Methodological differences might thus account for these differing observations, possibly due to an underestimation of somatotroph number in females.

Like that of lactotrophs, somatotroph development is highly active during the perinatal period, with GH mRNA and protein in the pituitary also appearing on E17–E17.5 (59), and GH content and GH secretory cell number increasing markedly from E18 (65). It would appear, however, that somatotroph development is remarkably resistant to interference by perinatal GCs, because the proportion and morphology of pituitary somatotrophs and the pituitary GH content in male and female rats were entirely unresponsive to either pre- or neonatal GC treatment regimens. These treatments also failed to alter body weight (this study) or expression levels of somatostatin mRNA in hypothalamic periventricular neurons, cells that are major regulators of GH secretion (Warne, J., G. E. Gillies, and J. C. Buckingham, unpublished observations). Together, our findings indicate a relative insensitivity of the hypothalamo-pituitary-GH axis to GC programming. Although it is well documented that perinatal GC treatment or stress can inhibit fetal/neonatal growth in many species, there is often catch-up growth (66). Our present results do not allow us to say whether the GH axis could be temporarily affected and hence contribute to early growth problems. However, the precise role that GH plays in regulating fetal/neonatal growth is subject to debate (67), and it is feasible that the influence of perinatal GCs on nutrient supply, which is of primary importance for early growth, might be more critical than any effect on GH. In contrast to the resistance of both lactotroph and somatotroph cell types in male rats to GC programming effects, we have recently shown that dexamethasone treatment regimens identical with those used in this study caused a marked reduction in adult pituitary corticotroph numbers (17). Together, these findings demonstrate the unique responsiveness of pituitary cell types to perinatal GC exposure.


   Conclusions
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Conclusions
References
 
In summary, we have shown that in female rats, but not in males, brief exposure to GCs using a noninvasive method of administration during late gestation or early neonatal life markedly reduced the size of the DA population in the adult arcuate nucleus. Profound effects at the pituitary level were also seen on lactotroph morphology and prolactin content, but only with the prenatal treatment regimen, indicating independent critical windows of susceptibility to GC programming of hypothalamic and pituitary targets. These changes would be expected to compromise activity within the hypothalamo-lactotroph axis, and future studies involving provocative challenges to the system are planned to investigate the functional significance of the hypothalamic and/or pituitary disruptions. Of note, however, neonatal dexamethasone treatment significantly raised circulating prolactin levels, a condition that has been linked with reproductive disorders (68), autoimmune disease (69), and cancer (70). Our data therefore provide substantial new evidence that extends the premise that inappropriate exposure to GCs in early life can lead to disordered neuroendocrine function in adulthood and supports the novel view that sex-specific disturbances in prolactin-dependent homeostasis may influence disease susceptibility and reproductive function in adulthood. It is important to note also that our dexamethasone treatment regimen (~150 µg/kg·d) provides a dosage of the same order of magnitude as the standard dose of 5 mg, im (~ 50–75 µg/kg, depending on body weight), used in perinatal medicine. Although the value of a single course of GC treatment to promote lung maturation in cases of threatened premature delivery is unchallenged, our findings add mechanistic evidence to the growing concern that the increasing use of repeated does of GCs during the neonatal as well as the antenatal period is not without consequences later in life (71, 72). Although the pharmacological profile of dexamethasone is not equivalent to that of corticosterone, our results also raise the possibility that the female hypothalamo-lactotroph axis may be susceptible to the programming effects of stress-induced elevations of endogenous GCs.


   Acknowledgments
 
We are very grateful to Mr. Colin Rantle and Miss Devinder Mehet for their expert technical assistance.


   Footnotes
 
This work was supported by the Wellcome Trust and the Biotechnology and Biological Sciences Research Council, UK.

S.M., Z.-L. S., H.C., G.C., E.T., C.J., S.F., J.M., J.B., G.G. have nothing to declare.

First Published Online January 26, 2006

Abbreviations: DA, Dopamine; E, embryonic day; FS, folliculostellate; GC, glucocorticoid; GD, gestational day; GR, glucocorticoid receptor; HPA, hypothalamo-pituitary-adrenal; NGS, normal goat serum; P, postnatal day; PeN, periventricular nucleus; TH+, tyrosine hydroxylase positive; TH-IR, tyrosine hydroxylase immunoreactive; TIDA, tuberoinfundibular dopaminergic.

Received November 23, 2005.

Accepted for publication January 13, 2006.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Conclusions
 References
 

  1. Ward AM, Syddall HE, Wood PJ, Chrousos GP, Phillips DI 2004 Fetal programming of the hypothalamic-pituitary-adrenal (HPA) axis: low birth weight and central HPA regulation. J Clin Endocrinol Metab 89:1227–1233[Abstract/Free Full Text]
  2. Seckl JR, Meaney MJ 2004 Glucocorticoid programming. Ann NY Acad Sci 1032:63–84[Abstract/Free Full Text]
  3. Owen D, Andrews MH, Matthews SG 2005 Maternal adversity, glucocorticoids and programming of neuroendocrine function and behaviour. Neurosci Biobehav Rev 29:209–226[CrossRef][Medline]
  4. Welberg LA, Seckl JR 2001 Prenatal stress, glucocorticoids and the programming of the brain. J Neuroendocrinol 13:113–128[CrossRef][Medline]
  5. Phillips DI 1998 Birth weight and the future development of diabetes. A review of the evidence. Diabetes Care 21(Suppl 2):B150–B155
  6. Welberg LA, Seckl JR, Holmes MC 2001 Prenatal glucocorticoid programming of brain corticosteroid receptors and corticotrophin-releasing hormone: possible implications for behaviour. Neuroscience 104:71–79[CrossRef][Medline]
  7. Benediktsson R, Lindsay RS, Noble J, Seckl JR, Edwards CR 1993 Glucocorticoid exposure in utero: new model for adult hypertension. Lancet 341:339–341[CrossRef][Medline]
  8. Freeman ME, Kanyicska B, Lerant A, Nagy G 2000 Prolactin: structure, function, and regulation of secretion. Physiol Rev 80:1523–1631[Abstract/Free Full Text]
  9. Igosheva N, Klimova O, Anishchenko T, Glover V 2004 Prenatal stress alters cardiovascular responses in adult rats. J Physiol 557:273–285[Abstract/Free Full Text]
  10. Kaiser S, Sachser N 2005 The effects of prenatal social stress on behaviour: mechanisms and function. Neurosci Biobehav Rev 29:283–294[CrossRef][Medline]
  11. Shanks N, Lightman SL 2001 The maternal-neonatal neuro-immune interface: are there long-term implications for inflammatory or stress-related disease? J Clin Invest 108:1567–1573[CrossRef][Medline]
  12. Ben-Jonathan N, Hnasko R 2001 Dopamine as a prolactin (PRL) inhibitor. Endocr Rev 22:724–763[Abstract/Free Full Text]
  13. DeMaria JE, Lerant AA, Freeman ME 1999 Prolactin activates all three populations of hypothalamic neuroendocrine dopaminergic neurons in ovariectomized rats. Brain Res 837:236–241[CrossRef][Medline]
  14. Kamphuis PJ, Bakker JM, Broekhoven MH, Kunne C, Croiset G, Lentjes EG, Tilders FJ, van Bel F, Wiegant VM 2002 Enhanced glucocorticoid feedback inhibition of hypothalamo-pituitary-adrenal responses to stress in adult rats neonatally treated with dexamethasone. Neuroendocrinology 76:158–169[CrossRef][Medline]
  15. Felszeghy K, Bagdy G, Nyakas C 2000 Blunted pituitary-adrenocortical stress response in adult rats following neonatal dexamethasone treatment. J Neuroendocrinol 12:1014–1021[CrossRef][Medline]
  16. McArthur S, McHale E, Dalley JW, Buckingham JC, Gillies GE 2005 Altered mesencephalic dopaminergic populations in adulthood as a consequence of brief perinatal glucocorticoid exposure. J Neuroendocrinol 17:475–482[CrossRef][Medline]
  17. Theogaraj E, John CD, Christian HC, Morris JF, Smith SF, Buckingham JC 2005 Perinatal glucocorticoid treatment produces molecular, functional, and morphological changes in the anterior pituitary gland of the adult male rat. Endocrinology 146:4804–4813[Abstract/Free Full Text]
  18. DeMaria JE, Livingstone JD, Freeman ME 1998 Characterization of the dopaminergic input to the pituitary gland throughout the estrous cycle of the rat. Neuroendocrinology 67:377–383[CrossRef][Medline]
  19. Funkhouser JD, Peevy KJ, Mockridge PB, Hughes ER 1978 Distribution of dexamethasone between mother and fetus after maternal administration. Pediatr Res 12:1053–1056[Medline]
  20. Pearlman WH 1983 Glucocorticoids in milk: a review. Endocrinol Exp 17:165–174[Medline]
  21. Bayer SA, Altman J, Russo RJ, Zhang X 1993 Timetables of neurogenesis in the human brain based on experimentally determined patterns in the rat. Neurotoxicology 14:83–144[Medline]
  22. Rang HP, Dale MM, Ritter JM 1999 Pharmacology. 4th ed. London: Churchill Livingstone
  23. Murray HE, Pillai AV, McArthur SR, Razvi N, Datla KP, Dexter DT, Gillies GE 2003 Dose- and sex-dependent effects of the neurotoxin 6-hydroxydopamine on the nigrostriatal dopaminergic pathway of adult rats: differential actions of estrogen in males and females. Neuroscience 116:213–222[CrossRef][Medline]
  24. Hart AM, Terenghi G 2004 Frozen-section fluorescence microscopy and stereology in the quantification of neuronal death within dorsal root ganglia. J Mol Histol 35:565–580[CrossRef][Medline]
  25. Benes FM, Lange N 2001 Two-dimensional versus three-dimensional cell counting: a practical perspective. Trends Neurosci 24:11–17[CrossRef][Medline]
  26. Cavalieri U, Tammaro AE, Quadri A, Baldoli C, Bonaccorso O 1966 A new approach to the pathogenesis of osteoporosis. G Gerontol 14:1187–1189[Medline]
  27. Bonthius DJ, McKim R, Koele L, Harb H, Karacay B, Mahoney J, Pantazis NJ 2004 Use of frozen sections to determine neuronal number in the murine hippocampus and neocortex using the optical disector and optical fractionator. Brain Res Brain Res Protoc 14:45–57[CrossRef][Medline]
  28. Taylor AD, Cowell A-M, Flower RJ, Buckingham JC 1995 Dexamethasone suppresses the release of prolactin from the rat anterior pituitary gland by lipocortin 1 dependent and independent mechanisms. Neuroendocrinology 62:530–542[Medline]
  29. Carmignac DF, Robinson IC 1990 Growth hormone (GH) secretion in the dwarf rat: release, clearance and responsiveness to GH-releasing factor and somatostatin. J Endocrinol 127:69–75[Abstract]
  30. Traverso V, Christian HC, Morris JF, Buckingham JC 1999 Lipocortin 1 (annexin 1): a candidate paracrine agent localized in pituitary folliculo-stellate cells. Endocrinology 140:4311–4319[Abstract/Free Full Text]
  31. Nakane PK 1975 Identification of anterior pituitary cells by immunogold labelling and immunoelectron microscopy. In: Tixier-Vidal A, Farquhar M, eds. The anterior pituitary gland. New York: Academic Press; 134–158
  32. Cruz-Orive LM, Weibel ER 1990 Recent stereological methods for cell biology: a brief survey. Am J Physiol 258:L148–L156
  33. Nogami H, Yoshimura F 1982 Fine structural criteria of prolactin cells identified immunohistochemically in the male rat. Anat Rec 202:261–274[CrossRef][Medline]
  34. Huerta-Ocampo I, Christian HC, Thompson NM, El-Kasti MM, Wells T 2005 The intermediate lactotroph: a morphologically distinct, ghrelin-responsive pituitary cell in the dwarf (dw/dw) rat. Endocrinology 146:5012–5023[Abstract/Free Full Text]
  35. Rey-Roldan EB, Lux-Lantos V, Chamson-Reig A, Libertun C 1997 In vivo interaction of baclofen, TRH and serotonin on PRL and TSH secretion in the developing and adult male and female rats. Life Sci 61:2283–2290[CrossRef][Medline]
  36. Cheung S, Will YM, Hentschel K, Moore KE, Lookingland KJ 1997 Role of gonadal steroids in determining sexual differences in expression of Fos-related antigens in tyrosine hydroxylase-immunoreactive neurons in subdivisions of the hypothalamic arcuate nucleus. Endocrinology 138:3804–3810[Abstract/Free Full Text]
  37. Lookingland KJ, Jarry HD, Moore KE 1987 The metabolism of dopamine in the median eminence reflects the activity of tuberoinfundibular neurons. Brain Res 419:303–310[CrossRef][Medline]
  38. Arbogast LA, Voogt JL 1991 Ontogeny of tyrosine hydroxylase mRNA signal levels in central dopaminergic neurons: development of a gender difference in the arcuate nuclei. Brain Res Dev Brain Res 63:151–161[Medline]
  39. Balan IS, Ugrumov MV, Borisova NA, Calas A, Pilgrim C, Reisert I, Thibault J 1996 Birthdates of the tyrosine hydroxylase immunoreactive neurons in the hypothalamus of male and female rats. Neuroendocrinology 64:405–411[Medline]
  40. Phelps CJ, Vaccarella MY, Romero MI, Hurley DL 1994 Postnatal reduction in number of hypothalamic tuberoinfundibular dopaminergic neurons in prolactin-deficient dwarf mice. Neuroendocrinology 59:189–196[Medline]
  41. Abraham IM, Harkany T, Horvath KM, Luiten PG 2001 Action of glucocorticoids on survival of nerve cells: promoting neurodegeneration or neuroprotection? J Neuroendocrinol 13:749–760[CrossRef][Medline]
  42. Gould E, Cameron HA 1996 Regulation of neuronal birth, migration and death in the rat dentate gyrus. Dev Neurosci 18:22–35[Medline]
  43. Cintra A, Solfrini V, Bunnemann B, Okret S, Bortolotti F, Gusaffson JA, Fuxe K 1993 Prenatal development of glucocorticoid receptor gene expression and immunoreactivity in the rat brain and pituitary gland: a combined in situ hybridization and immunocytochemical analysis. Neuroendocrinology 57:1133–1147[Medline]
  44. Weisz J, Ward IL 1980 Plasma testosterone and progesterone titers of pregnant rats, their male and female fetuses, and neonatal offspring. Endocrinology 106:306–316[Abstract]
  45. Simerly RB 2005 Wired on hormones: endocrine regulation of hypothalamic development. Curr Opin Neurobiol 15:81–85[CrossRef][Medline]
  46. Simerly RB 1989 Hormonal control of the development and regulation of tyrosine hydroxylase expression within a sexually dimorphic population of dopaminergic cells in the hypothalamus. Brain Res Mol Brain Res 6:297–310[Medline]
  47. Demarest KT, Moore KE, Riegle GD 1985 Acute restraint stress decreases tuberoinfundibular dopaminergic neuronal activity: evidence for a differential response in male versus female rats. Neuroendocrinology 41:504–510[Medline]
  48. Harfstrand A, Fuxe K, Cintra A, Agnati LF, Zini I, Wikstrom AC, Okret S, Yu ZY, Goldstein M, Steinbusch H, Verhofstad A, Gustafsson J-A 1986 Glucocorticoid receptor immunoreactivity in monoaminergic neurons of rat brain. Proc Natl Acad Sci USA 83:9779–9783[Abstract/Free Full Text]
  49. Catalani A, Marinelli M, Scaccianoce S, Nicolai R, Muscolo LA, Porcu A, Koranyi L, Piazza PV, Angelucci L 1993 Progeny of mothers drinking corticosterone during lactation has lower stress-induced corticosterone secretion and better cognitive performance. Brain Res 624:209–215[CrossRef][Medline]
  50. Fossom LH, Sterling CR, Tank AW 1992 Regulation of tyrosine hydroxylase gene transcription rate and tyrosine hydroxylase mRNA stability by cyclic AMP and glucocorticoid. Mol Pharmacol 42:898–908[Abstract]
  51. Phelps CJ 2004 Postnatal regression of hypothalamic dopaminergic neurons in prolactin-deficient Snell dwarf mice. Endocrinology 145:5656–5664[Abstract/Free Full Text]