The paraventriculo-infundibular corticotropin releasing factor (CRF) pathway as revealed by immunocytochemistry in long-term hypophysectomized or adrenalectomized rats

Regulatory Peptides, 5 (1983) 295-305 Elsevier Biomedical Press

295

The paraventriculo-infundibular corticotropin releasing factor (CRF) pathway as revealed by immunocytochemistry in long-term hypophysectomized or adrenalectomized rats I. Merchenthaler ~, S. Vigh b, p. Petrusz ~ and A.V. Schally b Department of Anatomy, University of North Carolina, Chapel Hill, NC 27514, and h Endocrine and Polypeptide Laboratory, Veterans Administration Hospital, and Department of Medicine, Tulane University School of Medicine, New Orleans, LA 70112, U.S.A. (Received 6 October 1982; revised manuscript received 16 December 1982; accepted for publication 17 December 1982)

Summary The immunocytochemicai localization of corticotropin releasing factor (CRF)containing pathways projecting from the paraventricular nucleus (PVN) to the external layer of the median eminence (ME) in long-term hypophysectomized or adrenalectomized rats is described. Immunocytochemistry was followed by silver intensification of the diaminobenzidine end-product. In comparison with untreated control rats, both hypophysectomy and adrenalectomy resulted in a dramatic increase in immunostaining of the CRF-containing perikarya and fibers, particularly those originating from the PVN and terminating in the ME. The staining was more intense in adrenalectomized than in hypophysectomized rats. The CRF-positive fibers emerging from the PVN form a medial, an intermediate and a lateral fiber pathway. The lateral and intermediate CRF tracts leave the dorsolateral part of the PVN and course laterally and medially of the fornix, respectively, then ventrally toward the optic tract. Just dorsal to the optic tract they turn in caudal direction and run parallel with and very close to the basal surface of the hypothalamus; individual fibers then turn medially to terminate in the external layer of the ME. Only a few fibers originate from the medial-ventral part of the PVN (medial pathway). These fibers run in ventral direction along the walls of the 3rd ventricle and terminate in the ME. Thus the majority of CRF fibers, similarly to other peptidergic systems, reach the medial basal hypothalamus from the anterolateral direction. hypothalamus; paraventricular nucleus; median eminence; pituitary Address for correspondence: Dr. I. Merchenthaler, Department of Anatomy, University of North Carolina, 108 Swing Building 217H, Chapel Hill, NC 27514, U.S.A. Phone: (919) 966-3077. 0167-0115/83/0000-0000/$03.00 © 1983 Elsevier Science Publishers

296

Introduction

The 41-residue peptide corticotropin releasing factor (CRF) has recently been isolated from ovine hypothalamic tissue [1] and has been shown to be a potent stimulator of the release of adrenocorticotropin (ACTH), fl-endorphin, and c~melanocyte stimulating hormone (~-MSH) from the pituitary [1-4]. Recent immunocytochemical studies in normal rats revealed a rich accumulation of CRF-immunoreactive terminals in the median eminence (ME) of the hypothalamus, suggesting the physiological role of this peptide in the regulation of A C T H release from the pituitary [5-8]. In order to demonstrate the cells of origin of this hypothalamic CRF system, it was necessary to pretreat rats with colchicine, a drug which blocks axoplasmic transport and results in the accumulation of synthetic products in the perikarya [5-8,10]. All immunocytochemical studies published to date [5-8,11] agree that the major group of cell bodies which give rise to the rich CRF innervation of the ME resides in the paraventricular nucleus (PVN) of the hypothalamus. However, neither the normal nor the colchicine-treated rat proved to be a satisfactory model in which the path of the axons from the PVN to the ME could be followed. In our search for such models, we found that long-term hypophysectomized or adrenalectomized rats were best suited for this purpose, permitting convincing immunoo cytochemical demonstration of CRF not only in cell bodies but in fibers and terminals as well. Based on these two models, the present report provides a detailed description of the course of the paraventriculo-infundibular CRF pathway in the rat.

Materials and Methods

Adult Sprague-Dawley rats weighing 250 g were purchased from A R S / S p r a g u e Dawley, Madison, Wisconsin. Five rats were adrenalectomized via the dorsal approach, and five others served as unoperated controls. Hypophysectomized rats of the same strain and of similar body weight were purchased from Hormone Assay Labs., Inc., Chicago, Illinois. The rats were maintained under constant temperature (24°C) and a 14-h light, 10-h dark cycle with the light on at 6:00 a.m. The drinking water of the hypophysectomized and adrenalectomized rats contained 0.9% NaC1 and 5% dextrose and was available ad libitum. Four weeks after surgery rats from each group were anesthetized with Nembutal (40 m g / k g i.p.), given a s.c. injection of procain (100 mg/kg) and perfused through the ascending aorta with 100 ml of 1% paraformaldehyde followed by 150 ml of 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). 30 min after perfusion the brains were carefully removed, cut into three pieces, submerged in 0.9% NaCl-buffered phosphate, 0.01 M, pH 7.5 (PBS), were sectioned in the coronal plane at 50/~m with a Lancer vibratome (Ted Pella, Irvine, California), and collected serially in glass vials containing PBS. The sections were kept in PBS for at least 1 h, then treated with ethanol (50% for 10 min~ 70% for 30 min, 50% for 10 min) to facilitate the penetration of antibodies, and rinsed again in PBS for 10 min. The sections were first exposed to 2% normal sheep serum (NSS)

297 in PBS for 10 min. Subsequently they were stained floating in glass vials according to the unlabeled antibody peroxidase-antiperoxidase (PAP) [12] method. Briefly, the following incubations were conducted: (1) Rabbit anti-CRF serum SV-22, 40-42 h, 4°C. Optimal dilution was 1 : 3000. The antiserum was produced and characterized by Vigh et al. [8]. (2) Sheep anti-rabbit T-globulin (S-ARGG), 1:100 (Antibodies Inc., Davis, California), 20 min at room temperature. (3) PAP (Sternberger-Meyer Immunocytochemicals, Jarrettsville, Maryland), 1 : 100, 20 min, room temperature. (4) 3,3-Diaminobenzidine tetrahydrochloride (DAB) (Aldrich Chemical Co., Milwaukee, Wisconsin), 40 mg/100 ml with 0.002% HzO 2 in 0.05 M pH 7.6 Tris buffer, containing 0.01 M imidazole [13], 10 min, room temperature. Each incubation was followed by appropriate washing. The method specificity of our immunostaining was tested by a series of increasing dilutions of the primary antiserum, resulting in a gradual decrease and eventual disappearance of the staining. The primary antiserum was characterized by both radioimmunoassay [8] and immunocytochemistry [7]. Briefly, the specificity of the antiserum was tested by staining adjacent sections with antiserum that had been absorbed with increasing quantities of synthetic CRF. The possible cross-reactivities of the primary antiserum were tested on adjacent sections stained after absorption with each of the following antigens: [Arg8]-vasopressin (AVP) (Calbiochem-Behring Co., La Jolla, California), oxytocin (OT) and et-MSH (Bachem Fine Chemicals, Marina Del Rey, California), vasoactive intestinal polypeptide (VIP) (synthesized in our laboratories), urotensin I (gift of Dr. K. Lederis) and angiotensin I and II (Beckman Bioproducts, Palo Alto, California). Additional tests of specificity are described in detail elsewhere [7]. After completing the DAB reaction the sections were treated with 20% perchloric acid overnight at room temperature, and washed in three changes of distilled water, 20 min each. A special physical developer containing mgNO 3 and formaldehyde was freshly prepared just before use as described [ 14,15]. The sections were immersed in the physical developer until the desired grade of intensification was achieved as seen under a dissecting microscope (usually 8-10 min). The reaction was terminated by immersing the sections in 1% acetic acid in distilled water for 10 min. Following a 20-min rinse in distilled water the sections were immersed in 0.2% yellow gold chloride (H3AuCI4) for 15-20 s to 'tone' the silver. Following a 10-min rinse in distilled water the sections were immersed in sodium thiosulphate for 5-10 min to remove unreduced silver. They were rinsed in distilled water for 10 min, collected in polyvinyl alcohol (5%) and mounted on slides. The specificity tests described earlier were conducted on both intensified and unintensified sections.

Results

All immunostaining with the anti-CRF serum was completely blocked by preincubation of the antiserum with synthetic CRF (7.5/~g//xl). Absorption with uroten-

298

sin I decreased the intensity of CRF staining but did not totally abolish the reaction. Absorption with AVP, OT, ct-MSH, VIP, or angiotensin I or II had no effect on the intensity and distribution of the staining. Dual localizations reported earlier [7] have demonstrated that the structures which contain immunoreactive CRF in the rat brain are distinct from those containing either immunoreactive neurophysin, AVP, OT, enkephalin, ~t-MSH, or VIP. Most of the perikarya with CRF-like immunoreactivity were seen in the medial and lateral parvocellular subdivisions of the PVN [cf. 16]. Smaller numbers of CRF-containing cells were present in the dorsal and anterior parvocellular as well as in the periventricular subdivisions. Axons with CRF-like immunoreactivity emerging from the dorsal and lateral parvocellular parts of the PVN (lateral and intermediate pathways) were seen as a curved fan of fibers leaving the ventrolateral aspect of the nucleus (Fig. 1). A small number of fibers from the periventricular p a r t of the nucleus (medial pathway) ran downward, parallel with and close to the walls of the 3rd ventricle. Just dorsal to the optic chiasm the fibers turned in caudal direction between the caudal portions of the suprachiasmatic nuclei; individual fibers then took a medial turn and seemed to terminate in the external layer of the ME (Fig. 1). The intermediate fibers passed laterally just below and medial to the fornix, while the lateral fibers passed above and lateral to the fornix (Figs. 1, 2 and 3). The periphery of the fan of fibers swept downward toward the supraoptic nucleus and then turned posteromedially above the caudal edge of the optic chiasm and optic tract to pass toward the ME (Figs. 1 and 6). Fibers approaching the ME from a lateral direction parallel with the basal

•

..

SM

Fig. 1. Schematic drawing summarizing the paraventriculo-infundibular CRF system of the rat. For detailed description see the text. ARC, arcuate nucleus; FX, fornix; OC, optic chiasm; OT, optic tract; PVN, paraventricular nucleus; SC, suprachiasmatic nucleus; SM, stria medullaris; VMH, ventromedial nucleus.

299

Figs. 2-6. (2) The central part of the paraventriculo-infundibular C R F system of the rat. CRF-containing perikarya are located in the PVN. The cell bodies are concentrated in the parvocellular subdivisions of the PVN and send their processes in either ventral or lateral directions forming a curved fan of fibers. Fibers originating from the PVN form a medial (cannot be seen in this figure), an intermediate (I) and a lateral (L) pathway. Long-term adrenalectomized rat, silver intensified section. V, 3rd ventricle; FX, fornix. × 125. (3) High magnification of the right half of Fig. 2. Bi- or multipolar cell bodies are concentrated in the medial and lateral parvocellular parts of the PVN. Perikarya in the dorsal-dorsolateral regions giving rise to the lateral fibers of the fan have a tendency to be oriented in the horizontal plane. × 165. (4) Middle portion of the fan of C R F fibers. The fibers sweep downward toward the optic chiasm. The dorsal edge of the chiasm can be seen at the bottom of the figure. Long-term adrenalectomized rat. Silver intensified section. × 165. (5) The distribution of CRF-immunopositive cell bodies in the PVN in long-term hypophysectomized rat. In the normal rat, these cells can only be stained after colchicine treatment. The 3rd ventricle is immediately to the right. Silver intensification. × 165. (6) The terminal part of the fan of C R F fibers. The fibers approach the ME from lateral direction parallel with and very close to the basal surface of the hypothalamus. Long-term hypophysectomized rat. Frontal section, silver intensification. The 3rd ventricle is immediately to the right. × 165.

300

Figs. 7-10. (7) Distribution of C R F immunoreactivity in the anterior region of the ME of a control rat. Darkly stained fibers and terminals are located close to the midline in the central regions of the ME. The majority of the fibers and terminals are seen in the external zone, only a few are located in the internal zone. × 165. (8) Distribution of C R F immunoreactivity in the anterior region of the ME of long-term hypophysectomized rat. A slight increase in immunostaining can be observed in both zones of the ME. × 165. (9) Distribution of C R F immunoreactivity in the anterior region of the ME of a long-term adrenalectomized rat. Increased immunostaining can be seen in both zones of the ME. x 165. (10) Fibers with immunoreactive C R F in the neurohypophysis of a long-term adrenalectomized rat. x 280.

surface of the hypothalamus were frequently observed as far caudally as the level of the pituitary stalk (Fig. 1). Beyond this level only a few CRF-containing fibers were detected in such position. CRF-containing cell bodies located in the ventromedial (periventricular) portion of the PVN tended to occupy a vertical position and gave rise to the medial CRF pathway. Cell bodies in the central regions of the PVN giving rise to the intermediate fibers had a tendency to be situated in an oblique plane (e.g., Figs. 2 and 3).

Figs. 11-16. (11) Cell bodies and fibers with CRF-like immunoreactivity in the bed nucleus of the stria terminalis of an intact rat. Frontal section. CPU, caudate nucleus-putamen. × 165. (12) Perikarya and neuronal processes with CRF-tike immunoreactivity in the bed nucleus of the stria terminalis of a long-term hypophysectomized rat (compare with Fig. 11). Frontal section from the same level as Fig. 1 I. x 165. (13) Immunoreactive C R F in cell bodies and fibers in the central nucleus of the amygdala of an intact rat. Frontal section, x 165. (14) The distribution of CRF-immunoreactive cell bodies and processes in the central nucleus of the amygdala of a long-term adrenalectomized rat (compare with Fig. 13). Majority of the perikarya are multipolar in shape and stain extensively. Frontal section from the same level as Fig. 13. x 165. (15) Immunoreactive C R F in the medial preoptic area of an intact rat (small arrows). The figure is taken from the region just under the anterior commissure (cannot be seen). Large

301

arrow shows the ependymal wall of the 3rd ventricle. Frontal section, x 567. (16) Immunoreactive C R F fibers in the medial preoptic area of a long-term hypophysectomized rat (compare with Fig. 15). Many fibers are located in a vertical-oblique position (arrows); however, in this plane most of them are running in oro-caudal direction and can be seen as dots representing cross-sections of fibers (*). Frontal section from the same level as Fig. 14. x226.

302 The majority of the CRF-immunoreactive perikarya were located in the dorsaldorsolateral regions of the PVN, gave rise to the lateral fibers of the fan, and many of them were oriented in the horizontal plane (Figs. 1, 2, 3, 5). Scattered CRF-positive cells were found among the fibers forming the fan (Fig. 1). Many immunoreactive perikarya were detected in the medial preoptic and periventricular preoptic nuclei, although their projections were not as easy to trace as those from the PVN. Fibers from the periventricular preoptic nuclei were seen to enter the fan apparently to terminate in the ME. In control rats many CRF-immunoreactive nerve fibers were observed in the ME. In the anterior regions of the ME they were found in the internal zone, while in the tuberal region the majority of the fibers and terminals were observed in the external zone and only a few were seen in the internal layer (Fig. 7). Many immunoreactive fibers were traced into the neural stalk but only a few were found in the posterior lobe of the pituitary gland. A slight increase in immunostaining was observed in the ME four weeks after hypophysectomy (Fig. 8). A dramatic increase in immunostaining was detected in the ME, the pituitary stalk, and the posterior lobe of the pituitary gland in long-term adrenalectomized animals (Figs. 9 and 10). In this group a dense accumulation of CRF immunoreactivity was observed not only in the external but also in the internal zone of the ME (Fig. 9). As a result of the hypophysectomy or adrenalectomy, increased CRF immunoreactivity was observed not only in the hypothalamus but throughout the brain. The degree of intensification was greatest in the PVN, followed by the bed nucleus of the stria terminalis (Figs. 11 and 12), the central nucleus of the amygdala (Figs. 13 and 14), and the preoptic area (Figs. 15 and 16). In addition, perikarya with CRF were observed in many brain regions where they could not be demonstrated in the normal rat without colchicine treatment, although the degree of intensification in these regions was less dramatic and more variable than in the four areas described above. Such regions included the neocortex, the olfactory bulb, the nucleus accumbens septi, the periaqueductal gray of the mesencephalon and pons, the locus coeruleus, the parabrachial nucleus, and the medial vestibular nucleus. Fibers with CRF were also seen in all these regions of the brain.

Discussion The results of this study show that the immunocytochemical staining pattern for C R F in the rat brain is dramatically altered four weeks after hypophysectomy or adrenalectomy as compared to the pattern seen in normal controls. In both of the experimental groups the staining becomes more intense and shows a distribution different from that found in normal rats. The changes are more expressed after adrenalectomy than after hypophysectomy and encompass CRF-containing perikarya, axons as well as terminals. All immunocytochemical studies published to date agree that CRF-containing perikarya stain very lightly or not at all in hypothalami of normal rats and can only be demonstrated after intraventricular injection of colchicine, a treatment known to block axonal transport of synthetic products from neuronal cell bodies [5-8,10]; four

303 weeks after hypophysectomy or adrenalectomy perikarya in the PVN and in other regions of the brain stained strongly for CRF (Figs. 2, 3, 5, 12, 14). Axons projecting from the PVN to the ME are practically unstainable in either normal or colchicinetreated rats [5-8,11]; in striking contrast, such axons in adrenalectomized or hypophysectomized rats stained strongly and clearly along their course from the PVN to the ME (Figs. 2, 3, 4, 6). Finally, the CRF-containing terminals in the external (palisade) zone of the ME stain well in normal rats but only marginally after colchicine treatment [6,7]; this staining was very intense four weeks after either hypophysectomy or adrenalectomy. In addition, CRF-containing fibers in the internal zone of the ME, which were scanty in normal controls (Fig. 7), appeared more numerous after hypophysectomy (Fig. 8). Fibers with immunoreactive CRF were also demonstrated in the neurohypophysis of adrenalectomized rats (Fig. 10). Although sprouting might have contributed to the intensified staining in terminal fields such as the ME, the overall changes in CRF-staining patterns observed after hypophysectomy or adrenalectomy are best explained by assuming that both operations resulted in an increase in the amount and a change in the distribution of CRF present within neurons which normally produce this peptide in their perikarya, transport it in their axons, and discharge it from their terminals. Naturally, the possible mechanisms of the increase differ in the two experimental models although the elimination of negative feedback effects is probably the fundamental cause in both cases. Corticosteroids are well known to exert negative feedback effects at both hypothalamic [17,18,28] and pituitary [19,20] levels. Direct (short feedback) effects of ACTH have also been described [9,21]. Hypophysectomy not only removes the principal target organ of CRF but also results in marked decrease in the levels of circulating corticosteroids and ACTH. The conclusions of earlier reports that the levels of bioassayable [22-24,29] and immunoassayable [25] CRF in the hypothalamus are substantially increased after hypophysectomy are consistent with our results. The results of bioassays of hypothalamic CRF content after adrenalectomy are contradictory [22,24,26]; this might be due to variable assay techniques and the presence of many interfering substances in the hypothalamic extracts (e.g., vasopressin, a-MSH, etc.). A recent study with specific radioimmunoassay for CRF indicated a 7-fold increase in adrenalectomized rats as compared to normal controls [25]. Our results with immunocytochemistry are consistent with this latter finding. The tendency for more intense immunostaining (present study) and higher levels of radioimmunoassayable CRF [25] in adrenalectomized than in hypophysectomized rats seems to implicate A C T H itself as an important factor in the regulation of the production and release of hypothalamic CRF (short feedback). Our findings indicate that the immunocytochemical staining of CRF-containing neurons in extrahypothalamic regiofls of the rat brain (e.g., Figs. l 1-16) is also intensified after adrenalectomy or hypophysectomy as compared to normal controls. The intensification is greatest - - besides the PVN - - in the bed nucleus of the stria terminalis, the central nucleus of the amygdala, and the preoptic area. A less dramatic and more variable increase is seen in regions more distant from the hypothalamus, such as the neocortex or several areas of the brain stem. The interpretation of these observations is difficult since it is not clear how and for what

304 p h y s i o l o g i c a l p u r p o s e A C T H a n d c o r t i c o s t e r o i d s w o u l d influence the activity of C R F - p r o d u c i n g n e u r o n s in regions of the b r a i n not directly involved in the c o n t r o l of a n t e r i o r p i t u i t a r y function. Similar ' g e n e r a l i z e d ' f e e d b a c k effects, as a rule, have n o t been seen with r e g a r d to o t h e r h y p o t h a l a m i c peptides, a l t h o u g h Sar et al. [27] d i d note a slight increase in e n k e p h a l i n staining in the rat b r a i n after h y p o p h y s e c tomy. T h e d r a m a t i c intensification of i m m u n o s t a i n i n g for C R F in the p a r a v e n t r i c u l o - i n f u n d i b u l a r system a n d in the central nucleus of the a m y g d a l a after a d r e n a l e c t o m y has i n d e p e n d e n t l y been o b s e r v e d in a n o t h e r l a b o r a t o r y (Dr. W . K . Paull, p e r s o n a l c o m m u n i c a t i o n ) . In conclusion, in this r e p o r t we describe the entire course of the p a r a v e n t r i c u l o i n f u n d i b u l a r C R F p a t h w a y which is readily d e m o n s t r a t e d b y i m m u n o c y t o c h e m i s t r y in l o n g - t e r m h y p o p h y s e c t o m i z e d or l o n g - t e r m a d r e n a l e c t o m i z e d rats. These experim e n t a l m o d e l s can be useful in further studies of C R F - c o n t a i n i n g n e u r o n a l structures, especially p a t h w a y s , in the rat brain. O u r results p r o v i d e s u p p o r t for the i m p o r t a n c e of direct f e e d b a c k from p i t u i t a r y to b r a i n b y A C T H . In a d d i t i o n , d e m o n s t r a t i n g that the 4 1 - a m i n o acid p e p t i d e C R F d e s c r i b e d b y Vale et al. [1] is subject to well-defined f e e d b a c k regulation, they suggest that this p e p t i d e might i n d e e d be one of the physiological factors regulating the release of A C T H from the pituitary.

Acknowledgements T h e a u t h o r s are grateful to Ms. Pia B.-M. L i n d s t r 6 m for expert technical help a n d to Dr. G 6 r c s for chemicals used in the silver method'. This w o r k was s u p p o r t e d b y U S P H S G r a n t s No. N S 14904 a n d A M 07467.

References 1 Vale, W., Spiess, J., Rivier, C. and Rivier, J., Characterization of a 41-residue ovine hypothalamic peptide that stimulates secretion of corticotropin and fl-endorphin, Science, 213 (1981) 1294-1387. 2 Proulx-Ferland, L., Labrie, F., Dumont, D. and Cote, J., Corticotropin-releasing factor stimulates secretion of melanocyte-stimulating hormone from the rat pituitary, Science, 217 (1982) 62-63. 3 Rivier, C., Brownstein, M.J., Spiess, J., Rivier, J. and Vale, W., In vivo corticotropin releasing factor induced secretion of adreno-corticotropin, fl-endorphin and corticosterone, Endocrinology. 110 (1982) 272-278. 4 Turkelson, C.M., Arimura, A., Culler, M.D., Fishback, J.G., Groot, K., Kanda, M., Luciano, M., Thomas, C.R., Chang, D., Chang, D.K. and Shimizu, M., In vivo and in vitro release of ACTH by synthetic CRF, Peptides, 2 (1982) 425-429. 5 Bloom, F.E., Battenberg, E.L.F., Rivier, J. and Vale, W., Corticotropin releasing factor (CRF): immunoreactive neurons and fibers in rat hypothalamus, Regul. Peptides, 4 (1982) 43-48. 6 Bugnon, C., Fellmann, D., Gouget, A. and Cardot, J., Corticoliberin in rat brain: immunocytochemical identification and localization of a novel neuroglandular system, Neurosci. Lett., 30 (1982) 25-30. 7 Merchenthaler, I., Vigh, S., Petrusz, P. and Schally, A.V., Immunocytochemical localization of corticotropin releasing factor (CRF) in the rat brain, Am. J. Anat., 165 (1982) 385-396. 8 Vigh, S., Merchenthaler, I., Torres-Aleman, I., Sueiras-Diaz, J., Coy, D.H., Petrusz, P. and Schally, A.V., Corticotropin releasing factor (CRF): immunocytochemical localization and radioimmunoassay (RIA), Life Sci., 31 (1982) 2441-2448.

305 9 Seiden, G. and Brodish, A., Physiological evidence for 'short-loop' feedback effects of ACTH on hypothalamic CRF, Neuroendocrinology, 8 ( 1971) 154-164. 10 Weisenberg, T.C., Borisy, G.G. and Taylor, E.W., The colchicine-binding protein of mammalian brain and its relation to microtubules, Biochemistry, 7 (1968) 4466-4479. 11 Paull, W.K., Sch~51er, J., Arimura, A., Meyers, C.A., Chang, J.K., Chang, D. and Shimizu, M., Immunocytochemical localization of CRF in the ovine hypothalamus, Peptides, 3 (1982) 183-191. 12 Sternberger, L.A., Immunocytochemistry, 2nd edition, John Wiley & Sons, New York, 1979. 13 Straus, W., Imidazole increases the sensitivity of the cytochemical reaction for peroxidase with diaminobenzidine at a neutral pH, J. Histochem. Cytochem., 30 (1982) 491-493. 14 Gallyas, F., Grrcs, T. and Merchenthaler, I., High-grade intensification of the end-product of the diaminobenzidine reaction for peroxidase histochemistry, J. Histochem. Cytochem., 30 (1982) 183-184. 15 Merchenthaler, I., G6rcs, T. and Petrusz, P., Silver intensification of the diaminobenzidine reaction product for peroxidase immunocytochemistry, J. Histochem. Cytochem., 30 (1982) 607. 16 Swanson, L.W. and Kuypers, H.G.J.M., The paraventricular nucleus of the hypothalamus: Cytoarchitectonic subdivisions and the organization of projections to the pituitary, dorsal vagal complex and spinal cord as demonstrated by retrograde fluorescence double labeling methods, J. Comp. Neurol., 194 (1980) 555-570. 17 Endrrczi, E., Liss~k, K. and Tekeres, M., Hormonal feedback regulation of pituitary-adrenocortical activity, Acta Physiol. Hung., 18 (1961) 291-299. 18 Simpson, C.W., Hamler, M. and Brodish, A., ACTH regulation of tissue-CRF, Neuroendocrinology, 31 (1980) 210-214. 19 Gonzales-Lugue, A., L'Age, M., Dhariwal, A.P.S. and Yates, F.E., Stimulation of corticotropin release by corticotropin releasing factor (CRF) or by vasopressin following intrapituitary infusions in unanesthetized dogs: Inhibition of the response by dexamethasone, Endocrinology, 86 (1970) 1134-1142. 20 Yates, F.E. and Maran, J.W., Stimulation and inhibition of adreno-corticotropin release. In E. Knobil and W. Sawyer (Eds.), Handbook of Physiology, Section 7: Endocrinology, Vol. IV. The Pituitary Gland and its Neuroendocrine Control. American Physiological Society, Washington, DC, 1974, pp. 364-404. 21 Motta, M., Mangili, G. and Martini, L., A "short" feedback loop in the control of ACTH secretion, Endocrinology, 77 (1965) 392-395. 22 Seiden, G. and Brodish, A., Persistence of a diurnal rhythm in hypothalamic corticotropin-releasing factor (CRF) in the absence of hormone feedback, Endocrinology, 90 (1972) 1401-1403. 23 Krieger, D.T., Liotta, A. and Brownstein, M.J., Corticotropin-releasing factor distribution in normal and Brattleboro rat brain, and effect of deafferentation, hypophysectomy and steroid treatment in normal animals, Endocrinology, 100 (1977) 227-237. 24 Vernikos-Danellis, J., Effect of stress, adrenalectomy, hypophysectomy and hydrocortisone on the corticotropin-releasing activity of the rat median eminence, Endocrinology, 76 (1965) 122-126. 25 Moldow, R.L. and Fischman, A.J., Hypothalamic CRF-like immunoreactivity in the rat after hypophysectomy or adrenalectomy, Peptides, 3 (1982) 143-147. 26 Hiroshige, T. and Sakakura, M., Circadian rhythm of corticotropin-releasing activity in the hypothalamus of normal and adrenalectomized rats, Neuroendocrinology, 7 (1971) 25-36. 27 Sar, M., Stumpf, W.E., Miller, R., Chang, K.-J. and Cuatrecasas, P., Immunohistochemical localization of enkephalin in rat brain and spinal cord, J. Comp. Neurol., 182 (1978) 17-38. 28 Smelik, P.G. and Sawyer, C.H., Effects, of implantation of cortisol into the brain stem or pituitary gland on the adrenal response to stress in the rabbit, Acta Endocrinol., 41 (1962) 561-570. 29 Takabe, K., Sakakura, M. and Mashimo, K., Continuance of diurnal rh2~thmicity of CRF activity in hypophysectomized rats, Endocrinology, 90 (1972) 1515-1520.