Developmental Changes in Ovine Corticotrophs in Vitro*

4 downloads 283 Views 404KB Size Report
chemical, RIA, and cell immunoblotting techniques to examine pop- ulations of corticotrophs ... further investigate corticotrophs during development, we used an ...
0013-7227/97/$03.00/0 Endocrinology Copyright © 1997 by The Endocrine Society

Vol. 138, No. 3 Printed in U.S.A.

Developmental Changes in Ovine Corticotrophs in Vitro* FRANK M. PEREZ, JEFFREY SCHWARTZ,

AND

JAMES C. ROSE

The Perinatal Research Laboratory (F.M.P., J.S., J.C.R.), Department of Physiology and Pharmacology (F.M.P., J.S., J.C.R.), and Department of Obstetrics and Gynecology (J.S., J.C.R.), The Bowman Gray School of Medicine, Winston-Salem, North Carolina 27157 ABSTRACT We recently reported that fetal sheep corticotrophs (ACTH-producing cells) at 108 6 5 d (days) of gestation are relatively more responsive to CRH than to AVP, whereas those at 139 6 0 d (term 5 145 d) and in the adult are more responsive to AVP. To further characterize these developmental changes, we used immunocytochemical, RIA, and cell immunoblotting techniques to examine populations of corticotrophs and individual cells. Immunocytochemical studies revealed that corticotroph frequency decreased from 22 6 1% of all pituitary cells at 100 d of gestation to 14 6 1% at 135 d and 9 6 0% in the adult. RIA measurements of ACTH secretion by cell populations showed that the response of corticotrophs to CRH diminished, whereas that to AVP increased during gestation and into adulthood. Cell blot analysis of individual corticotrophs identified two types of secretory responses (increases in the number of secreting cells and average amount of ACTH released per cell) to CRH or AVP that changed during fetal development. At 100 d of gestation, CRH in-

creased the proportion of secreting cells from 65 6 3% (no test agent) to approximately 90%; AVP exerted a negligible effect on the relative abundance of secreting cells. At 120 d of gestation, both secretagogues, alone or in combination, increased the proportion of secreting corticotrophs from 49 6 6% to about 85%. At 135 d of gestation and in the adult, AVP, alone or in combination with CRH, increased secreting corticotrophs from about 53 6 6% to about 80%. CRH alone exerted a nominal effect on the proportion of secreting cells. Additional analyses showed that, at 100 or 120 d of gestation, the average amount of ACTH secreted by individual corticotrophs did not change in response to CRH or AVP. However, near term and into adulthood, the average quantity of ACTH released from individual cells increased in response to these agents. Our findings suggest that maturational changes in fetal corticotrophs dictate whether their secretory response to CRH or AVP results from an increase in the proportion of cells secreting ACTH and (or) an increase in the average amount of hormone secreted by individual cells. (Endocrinology 138: 916 –921, 1997)

S

in terms of basal ACTH secretion and response to CRH and AVP.

UCCESSFUL preparation of the mammalian fetus for birth depends, in part, on timely activation of the hypothalamo-pituitary-adrenal (HPA) axis. As an integral part of this axis, corticotrophs (ACTH-producing cells) undergo both morphological and functional changes during HPA maturation. At least two types of cells (designated fetal and adult) are histologically distinguishable at 90 days (d) of gestation; beyond 130 d and into adulthood, the latter cell predominates (1, 2). During these periods, corticotroph frequency decreases in the adenohypophysis (3). These developmental changes in corticotroph morphology are accompanied by alterations in their response to hypothalamic regulatory factors. In this regard, we recently reported (4) that corticotrophs at 108 d of gestation are relatively more responsive to CRH than to AVP, whereas those at 135 d and in the adult are more responsive to AVP. These findings were obtained by measuring cumulative responses of cell populations to CRH and AVP in vitro. To further investigate corticotrophs during development, we used an immunoblotting technique to study individual cells. Other investigators (5–9) have used this method to characterize secretion of hormones and small peptides by single pituitary cells. We used cell immunoblotting in the present study to determine whether corticotrophs are heterogeneous

Materials and Methods Anterior pituitaries Anterior pituitary glands were obtained from adult pregnant ewes and their fetuses at 101 6 1, 121 6 1, and 135 6 4 d of gestation. A minimum of six animals was used for each group. Each ewe was deeply anesthetized with ketamine and pentobarbital before removal of the fetus by cesarean section. Each fetus was given a lethal injection of pentobarbital (iv) before removal of the pituitary. After this procedure, the ewe was given a lethal injection of potassium chloride (iv) before removing its pituitary. All procedures involving the animals were approved by the institutional animal care and use committee.

Preparation of pituitary cells Individual cells were prepared from adult and fetal anterior pituitaries according to a method described elsewhere (4). Minced tissue fragments were placed in HEPES-buffer containing 0.4% collagenase (Worthington Biochemical Corp., Freehold, NJ) and DNAase I (Sigma Chemicals, St. Louis, MO) and incubated for 2h at 37 C with gyratory shaking. After incubation, the fragments were passed gently through a flame-tapered glass pipet and the dispersed cells were washed in DMEM/Ham’s F12 containing 0.2% polypep. Cells were filtered to eliminate clumps and reaggregates and treated as described below.

Cell culture Adult and fetal sheep pituitary cells were divided into three groups: those cultured on glass cover slips, coated plastic surfaces, or a proteincapturing membrane (Immobilon, Millipore, Bedford, MA). Cells of the first group were used to measure the proportion of corticotrophs in the total cell population. Cells were cultured according to a modified method of Wilfinger (10). This procedure consisted of incubating 3.75 3 104 cells in a 50-ml droplet of medium on a glass cover slip (9 3 9 mm2;

Received September 9, 1996. Address all correspondence and requests for reprints to: Dr. Frank M. Perez, Assistant Professor, Department of Physiology and Pharmacology, The Bowman Gray School of Medicine, Medical Center Boulevard, Winston-Salem, North Carolina 27157. E-mail: [email protected]. * This work was supported by NIH Grant HD-11210.

916

ACTH SECRETION AND FETAL DEVELOPMENT Bellco Glass, Inc., Vineland, NJ) at the bottom of a well (24-well tissue culture plate; Corning, Corning, NY). After 1 h for the cells to attach to the glass, the well was filled with medium (1.0 ml) and the cells were incubated for an additional 1 h. After this step, the cells were fixed for 1 h, as described below, and processed by immunocytochemistry for ACTH. Cells of the second group were used to measure cumulative responses to hypothalamic stimulatory factors and cultured at a high cell density (1.5 3 105 cells/28 mm2) on poly-l-lysine-coated plastic surfaces (24-well tissue culture plate) according to the method of Perez (11). Cells were incubated for 2 h, washed once, and then treated with fresh medium containing vehicle (control), CRH (10 nm), AVP (100 nm), or both agents together. After 2 h, the medium was collected and centrifuged, and the supernatant solutions were stored at 220 C for ACTH assay. Cells of the third group were used to measure the response of individual corticotrophs to hypothalamic stimulatory factors. Cells were cultured on Immobilon; a droplet (100 ml) of medium containing cells (1.5 3 104) was applied to the membrane. After 15 min, CRH (10 nm, final concentration), AVP (100 nm), or both agents together (10 ml each) were added directly into the droplet. Medium alone (10 ml) was added to the control droplets. After 2 h, the cells, attached to the membrane, were fixed (1 h) with paraformaldehyde (4.0%) and processed by immunocytochemistry for ACTH.

ACTH immunocytochemistry Pituitary cells, attached to cover slips, were stained for ACTH according to a modified method of Denef et al. (12). Cells were fixed for 1 h with Bouin’s solution (1.0 ml), washed, and incubated overnight in TRIS/NaCl buffer containing primary antiserum (rabbit antisheep ACTH; 1:3,000). After this step, they were washed and incubated for 2 h with secondary antiserum (goat antirabbit IgG/HRP; 1:500). Immunoreaction product was developed with DAB and 0.006% H2O2; cells were counterstained with hematoxylin. Pituitary cells, attached to Immobilon, were immunostained for ACTH according to a modified method of Denef et al. (12). Cells were fixed with glutaraldehyde (2.5%), incubated with primary (1:1500) and secondary (1:500) antisera, reacted with DAB/H2O2, and counterstained with hematoxylin. Controls for all immunocytochemical staining procedures included: 1) replacement of the primary antiserum with nonimmune rabbit serum; 2) preadsorption of the primary antiserum with purified ACTH before its use; and 3) immunolocalization of purified ACTH on Immobilon. The ACTH antiserum that was used in these studies was prepared and described elsewhere (13).

ACTH cell analysis A minimum of 500 cells on cover slips or Immobilon membranes was included for analysis of each treatment by using a 40 3 objective and bright-field illumination. A serpentine surveying pattern was used to insure that each cell was counted only once. Cells on cover slips and Immobilon were characterized as either ACTH immunopositive or immunonegative. After determining that analysis of cells on cover slips or Immobilon yielded similar results, we used the former condition to measure the proportion of corticotrophs in the total cell population. Cells on Immobilon were used to determine the proportion of corticotrophs that secreted ACTH onto the membrane. Cells that were associated with ACTH on the membrane were counted as secretors. ACTH release by individual cells (see Fig. 7) was calculated by dividing the amount of hormone secreted by cell populations (measured by RIA) by the number of secretors (determined by immunoblotting).

917

together with the arithmetic sum of the net responses to each peptide alone.

Results Corticotroph frequency

The percentage of corticotrophs in the fetal adenohypophysis decreased during gestation and into adulthood (Fig. 1). Corticotroph frequency was 22 6 1%, 18 6 1%, and 14 6 1% at 100, 120, and 135 d of gestation, respectively. The frequency (9 6 0%) of corticotrophs in the adult did not differ from that of late gestation. Cell immunoblotting

Figure 2 shows a representative immunoblot of enzymatically-dispersed sheep pituitary cells. Cells that contained and released ACTH were readily identifiable on the basis of ACTH immunoreactivity. Individual corticotrophs stained intensely dark-brown and the amount of ACTH that was secreted on the membrane varied (as determined by computerized densitometry; data not shown). Immunonegative cells were stained only light-blue by hematoxylin. All cells that stained positively for secreted ACTH also were immunopositive for content. Immunoreactivity was not observed when the primary antiserum was replaced with nonimmune serum or when the primary antiserum was preadsorbed with ACTH (data not shown). Secretory profiles of individual corticotrophs

Studies of ACTH secretion by cell populations and individual cells showed that responses of corticotrophs to CRH and AVP changed during fetal development. In cells from fetuses at 100 d of gestation, CRH (10 nm) increased ACTH release in cell populations by 54 6 3% over control value; AVP-induced ACTH release (13 6 2%) was statistically indistinguishable from the control (Fig. 3, upper panel). These agents together increased ACTH release by 144 6 6% above the control value. These responses of 100-d cell populations to CRH and AVP were paralleled by changes in the proportion of individual corticotrophs that secreted ACTH. Approximately 65% of all corticotrophs secreted ACTH under basal (unstimulated) conditions (Fig. 3, lower panel). This percentage was increased

ACTH RIA and statistical tests Concentrations of ACTH were measured in duplicate by RIA as described elsewhere (13). The interassay and intraassay coefficients of variation were typically 8.0% or less. All data (shown as mean 6 sem) were analyzed by ANOVA and Fisher’s least-significant-difference tests and evaluated at the P # 0.05 level of significance. Synergism was tested by statistically comparing the net response of cells with CRH and AVP

FIG. 1. Effect of development on corticotroph frequency. Enzymatically dispersed cells were incubated on glass cover slips, fixed and processed by immunocytochemistry for ACTH. Data are from six to eight experimental replicates. Letters associated with individual bars summarize statistical analyses; bars with no letters in common are statistically different from each other.

918

ACTH SECRETION AND FETAL DEVELOPMENT

FIG. 2. ACTH secretion by individual sheep pituitary cells of an adult ewe. Enzymatically dispersed cells were incubated on Immobilon for 2 h. After incubation, the membrane and attached cells were stained by immunocytochemistry for ACTH and counterstained with hematoxylin The area defined by the white boundary is enlarged in the upper left hand corner. Large arrow, secreted ACTH; small arrow, corticotroph. (1200 3 magnification).

to 92 6 4% by CRH; in contrast, the proportion of corticotrophs responding to AVP (76 6 5%) was not different from the basal value. The combined effect of these agents on the proportion of secreting corticotrophs was similar to that of CRH alone. In cells from 120 d of gestation, CRH or AVP by themselves each increased ACTH release in cell populations by about 80% above the control value (Fig. 4, upper panel). The combined effect of these agents on ACTH release was similar to either one alone. Studies of individual corticotrophs showed similar responses to CRH and AVP (lower panel). Both agents, alone or in combination, increased the proportion of secreting corticotrophs from 49 6 6% (control value) to about 85% of the total ACTH cell population. In cell populations at 135 d of gestation, CRH did not significantly increase ACTH release; however, AVP increased ACTH release by 200 6 50% over the control value (Fig. 5, upper panel). The response of cells to these agents in combination (226 6 43%) resembled that of AVP alone. Single-cell studies showed a parallel secretory profile (lower panel): 53% 6 6% of all corticotrophs secreted ACTH under basal conditions. CRH had no effect on the percentage of secreting corticotrophs; AVP increased it to 82 6 4% These agents in combination increased the percentage of secreting cells to a value (77 6 6%) that was similar to that of AVP alone. In the adult, CRH increased ACTH release in cell populations by 144 6 7% over control value; AVP increased it by 210 6 7% (Fig. 6, upper panel). The combined effect of these agents on ACTH release was 554 6 9%. Single-cell studies showed that 56 6 4% of all corticotrophs secreted ACTH under basal conditions (lower panel). CRH alone had no effect

Endo • 1997 Vol 138 • No 3

FIG. 3. Secretory profile of corticotrophs at 100 d of gestation. Enzymatically dispersed cells were incubated (2 h) on plastic (upper panel) or Immobilon (lower panel). Conditioned medium from cells on plastic was assayed by RIA for ACTH. Cells on Immobilon were immunostained for ACTH, counterstained and counted microscopically. Data are expressed as a percentage of the total ACTH cell population from six experimental replicates. Letters associated with individual bars summarize statistical analyses; bars with no letters in common are statistically different from each other.

on the proportion of secreting cells; AVP increased it to 82 6 4%. There was no difference between the response of the cells to these agents in combination and that to AVP alone. Figure 7 shows the calculated average of ACTH release by individual corticotrophs as determined by expressing hormone secretion by cell populations per individual secreting corticotroph. This analysis showed that, at 100 and 120 d of gestation, CRH and (or) AVP did not alter ACTH secretion per secreting corticotroph. However, at 135 d of gestation, AVP alone increased ACTH release per secreting corticotroph by 240 6 80% over the control value and by 160 6 40% in combination with CRH. In the adult, CRH or AVP increased ACTH release by 121 6 10% and 107 6 10%, respectively; these agents were additive (292 6 10%) together. Discussion

The present studies show, for the first time, that the means by which corticotrophs respond to CRH or AVP, i.e. increases in the number of cells that secrete ACTH (Figs. 3– 6, lower panels) or increases in the amount of ACTH released by individual cells (Fig. 7), change during fetal development and in adulthood. Thus, maturation of the response of corticotrophs to CRH or AVP depends, in part, on changes among the population of ACTH-secreting cells that occur as a function of gestational age. This finding is consistent with other in vitro ontological studies (4) reporting that fetal sheep corticotrophs at 108 6 5 d of gestation respond to CRH or AVP, but by 139 d, they respond significantly only to AVP. Both types of secretory responses of sheep corticotrophs to

ACTH SECRETION AND FETAL DEVELOPMENT

FIG. 4. Secretory profile of corticotrophs at 120 d of gestation. Cells were treated and analyzed as described in Fig. 3. Data are from six experimental replicates. Letters associated with individual bars summarize statistical analyses; bars with no letters in common are statistically different from each other.

FIG. 5. Secretory profile of corticotrophs at 135 d of gestation. Cells were treated and analyzed as described in Fig. 3. Data are from seven experimental replicates. Letters associated with individual bars summarize statistical analyses; bars with no letters in common are statistically different from each other.

CRH (i.e. increases in numbers of secreting cells or secretory output per cell) have been reported also for adult rat pituitary cells. Leong and colleagues (14 –16) found that CRH alters both the number of ACTH-secreting cells and the amount of ACTH secreted by a fixed number of corticotrophs. Childs

919

FIG. 6. Secretory profile of corticotrophs from the adult ewe. Cells were treated and analyzed as described in Fig. 3. Data are from eight experimental replicates. Letters associated with individual bars summarize statistical analyses; bars with no letters in common are statistically different from each other.

and Burke (17) reported that CRH increases the percentage of secreting corticotrophs and speculated that the secretagogue recruits a subpopulation of quiescent cells to release ACTH. Our immunoblotting studies support this idea by identifying a subset of corticotrophs that had no associated secretory product on the membrane. Although it is possible that these quiescent cells released ACTH in amounts that fall below the detection limit of the cell blot assay, our results suggest that CRH stimulates a subpopulation of corticotrophs to release ACTH. This type of secretory response occurs at 100 and 120 d of gestation (Figs. 3 and 4). As the fetus matures, the secretory response changes to one that involves increased output of ACTH from already-secreting cells (Fig. 7). Developmental responses of sheep corticotrophs to AVP involved changes in both the proportion of secreting cells (Figs. 4 – 6) and amount of ACTH released per cell (Fig. 7). The former secretory response has not been observed in rat pituitary cells and might reflect species differences in the regulation of ACTH secretion (16). Increases in ACTH output by individual cells responding to AVP have been reported for rat corticotrophs (16). Developmental changes in secretory responses of corticotrophs to CRH and AVP reported herein coincide with increases in the total number of anterior pituitary cells, decreases in cell frequency (as shown in Fig. 1 and reported by others), and alterations in cell size (1–3). Whether these morphological changes correspond to differences in functional subtypes of corticotrophs remains unknown. However, our findings raise the possibility that maturation of the HPA axis involves alterations in subpopulations of secreting corticotrophs and amount of ACTH released by individual cells. Several mechanisms might be involved in changing the

920

ACTH SECRETION AND FETAL DEVELOPMENT

Endo • 1997 Vol 138 • No 3

FIG. 7. Effects of development on ACTH release by individual cells. Concentrations of ACTH secreted by cell populations (Figs. 3– 6; upper panels) and percentages of secreting corticotrophs (lower panels) were used to calculate the amount of ACTH released from individual cells. *, Differs from control (medium only) value within the group.

response of ovine corticotrophs to CRH and AVP during development. For example, the increased responsiveness of late gestational fetal and adult corticotrophs to AVP complements radioreceptor bindings studies of Shen et al. (18). These investigators reported that twice as many AVP receptors are present on sheep, compared with rat, anterior pituitary membranes, with both receptors showing similar affinities for the hormone. Similarly, diminishing responsiveness of late gestational corticotrophs to CRH (Fig. 5) is consistent with decreases in the number of CRH receptors near term (19). These findings raise the possibility that developmental changes in corticotroph responses, in part, are caused by expression of CRH and AVP receptors. The mechanism(s) controlling expression of these receptors during development of the fetal sheep is unclear but may involve cortisol and other steroids (20 –23). Additional mechanisms that might change corticotroph responses to CRH or AVP include alterations in signal transduction, cell-to-cell interactions, and differential release of these and other trophic factors by the hypothalamus (15, 24 –26). Concentrations of CRH and AVP used in these studies were selected on the basis of in vitro experiments with adult sheep anterior pituitary cells and in vivo measurements of concentrations in portal blood. We used concentrations that would produce measurable responses and be within a physiological range. In adult sheep, CRH (10 nm) and AVP (100 nm) induce the maximum secretory response from corticotrophs in vitro (27). In portal blood, AVP is usually found in excess to CRH at rest and in response to stress (28). We used the same concentrations of CRH and AVP, at all ages studied, to enable comparisons across ages. It remains possible that

higher concentrations of CRH could provoke responses from corticotrophs at ages where none were observed at the concentration used in this study. It is also possible that corticotrophs of pregnant ewes may respond to CRH and AVP differently from those of adult nonpregnant female or male sheep, but experience to date with tests of in vitro ACTH secretion suggests that this is unlikely. In summary, we have identified at least two modes by which sheep corticotrophs responded to CRH and AVP: increases in the number of secreting cells and quantity of ACTH released by single cells. The type of secretory response changed during fetal development and in adulthood. Our findings suggest that maturation of the HPA axis involves alterations in responses of corticotrophs to CRH and AVP. Acknowledgment The authors thank Ms. Regina Parker for her expert technical assistance.

References 1. Perry RA, Mulvogue HM, McMillen IC, Robinson PM 1985 Immunohistochemical localization of ACTH in the adult and fetal sheep pituitary. J Dev Physiol 7:397– 404 2. Antolovich C, Perry RA, Trahair JF, Silver M, Robinson PM 1989 The development of corticotrophs in the fetal sheep pars distalis: the effect of adrenalectomy or cortisol infusion. Endocrinology 124:1333–1339 3. Mulvogue HM, McMillen IC, Robinson PM, Perry RA 1986 Immuno-cytochemical localization of pro-g-MSH, g-MSH, ACTH and b-endorphin/b-lipotrophin in the fetal sheep pituitary: an ontogenic study. J Dev Physiol 8:355–368 4. Fora MA, Butler TG, Rose JC, Schwartz J 1996 Adrenocorticotropin secretion by fetal sheep anterior and intermediate lobe pituitary cells in vitro: effects of gestation and adrenalectomy. Endocrinology 137:3394 –3400 5. Kendall ME, Hymer WC 1987 Cell blotting: a new approach to quantify hormone secretion from individual rat pituitary cells. Endocrinology 121:2260 –2262

ACTH SECRETION AND FETAL DEVELOPMENT 6. Arita J, Kojima Y, Kimura F 1991 Identification by the sequential cell immunoblot assay of a subpopulation of rat dopamine-unresponsive lactotrophs. Endocrinology 128:1887–1894 7. Arita J, Kojima Y, Kimura F 1992 Lactotrophs secreting small amounts of prolactin reveal great responsiveness to thyrotropin-releasing hormone: analysis by the sequential cell immunoblot assay. Endocrinology 130:3167–3174 8. Arita J, Kouima Y, Mazawa S, Kimura F 1994 Somatotrophs and thyrotrophs in the rat anterior pituitary gland cosecrete substance P: analysis by the sandwich cell immunoblot assay. Neuroendocrinology 60:567–574 9. Arita J, Kojima Y, Kimura F 1993 Measurement of secretion of a small peptide at the single cell level by the cell immunoblot assay: thyroidectomy increases the number of substance P-secreting anterior pituitary cells. Endocrinology 132:2682–2688 10. Wilfinger WW, Larsen WJ, Downs TR, Wilbur DL 1984 An in vitro model for studies of intercellular communication in cultured rat anterior pituitary cells. Tissue Cell 16:483– 497 11. Perez FM 1993 Cell-to-cell interactions inhibit protein synthesis, GH, and LH release in male rat adenohypophyseal cells in vitro. Endocrine J 1:49 –55 12. Denef C, Hautekeet E, de Wolf A, Vanderschueren B 1978 Pituitary basophils from immature male and female rats: distribution of gonadotrophs and thyrotrophs as studied by unit gravity sedimentation. Endocrinology 103:724 –735 13. Castro MI, Valego TJ, Zehnder TJ, Rose JC 1993 Bioactive-to-immunoreactive ACTH activity changes with severity of stress in late-gestation ovine fetus. Am J Physiol 265: E68 –E73 14. Jia L, Canny BJ, Orth DN, Leong DA 1991 Distinct classes of corticotrophs mediate corticotropin-releasing hormone- and arginine vasopressin-stimulated adrenocorticotropin release. Endocrinology 128:197–203 15. Jia L, Canny BJ, Leong DA 1992 Paracrine communication regulates adrenocorticotropin secretion. Endocrinology 130:534 –539 16. Canny BJ, Jia L, Leong DA 1992 Corticotropin-releasing factor, but not arginine vasopressin, stimulates concentration-dependent increases in ACTH secretion from a single corticotrope. J Biol Chem 267:8325– 8329 17. Childs GV, Burke JA 1987 Use of the reverse hemolytic plaque assay to study

18. 19. 20. 21. 22. 23. 24. 25. 26.

27. 28.

921

the regulation of anterior lobe adrenocorticotropin (ACTH) secretion by ACTH-releasing factor, arginine vasopressin, angiotensin II, and glucocorticoids. Endocrinology 120:439 – 444 Shen PJ, Clarke IJ, Canny BJ, Funder JW, Smith AI 1990 Arginine vasopressin and corticotropin releasing factor: binding to ovine anterior pituitary membranes. Endocrinology 127:2085–2089 Lu F, Yang K, Challis RG 1990 Characteristics and developmental changes of corticotropin-releasing hormone-binding site in the fetal sheep anterior pituitary gland. J Endocrinol 130:223–229 Brooks AN, Challis JR 1988 Regulation of the hypothalamic-pituitary-adrenal axis in birth. Can J Physiol Pharmacol 66:1106 –1112 Brieu V, Cathiard AM, Naaman E, Saez JM, Durand P 1988 Development of the fetal pituitary-adrenal axis in the sheep. Reprod Nutr Dev 28:1129 –1144 Brooks AN, Currie IS, Thomas GB 1992 Neuroendocrine regulation of sheep fetuses. J Reprod Fertil 45:69 – 84 Pepe GJ, Albrecht ED 1995 Actions of placental and fetal adrenal steroid hormones in primate pregnancy. Endocr Rev 16:608 – 648 Schwartz J, Canny B, Vale W, Funder JW 1989 Intrapituitary cell-cell communication regulates ACTH secretion. Neuroendocrinology 50:716 –722 Schwartz J 1990 Evidence for intrapituitary intercellular control of adrenocorticotropin secretion. Mol Cel Endocrinol 68:77– 83 Clark I, Jessop D, Millar R, Bloom S, Lightman S, Coen CW, Lew R, Smith I 1993 Many peptides that are present in the external zone of the median eminence are not secreted into the hypophyseal portal blood of sheep. Neuroendocrinology 57:765–775 Kemppainen RJ, Clark TP, Sartin JL, Zerbe CA 1993 Hypothalamic peptide regulation of ACTH secretion from sheep pituitary. Am J Physiol 265: R840 –R845 Engler D, Pham T, Fullerton MJ, Ooi G, Funder JW, Clark IJ 1989 Studies of the secretion of corticotropin-releasing factor and arginine vasopressin into the hypophyseal-portal circulation of the conscious sheep. I. Effect of an audiovisual stimulus and insulin-induced hypoglycemia. Neuroendocrinology 49: 367–381