Reprint Series 4 May 1984, Volume 224, pp. 452-459 SCIENCE Stress Hormones: Their Interaction and Regulation Julius Axelrod and Terry D. Reisine Copyright 0 1984 by the American Association for the Advancement of Science Stress Hormones: Their Interaction and Regulation Julius Axelrod and Terry D. Reisine The constancy of the "milieu inter- ieur" is the condition of a free and independent existence. -CLAUDE BERNARD (I) The body responds to increased physi- cal or psychological demands by releas- ing adrenocorticotropin (ACTH) from the anterior pituitary, glucocorticoids from the adrenal cortex, epinephrine from the adrenal medulla, and norepi- nephrine from sympathetic nerves. These hormones serve to adapt the body to stressors ranging from the mildly psy- chological to the intensely physical by affecting cardiovascular, energy-produc- ing, and immune systems. It was the 19th-century physiologist Claude Ber- nard who recognized the importance of adaptive mechanisms with one of the most cogent statements (cited above) framed by a biological scientist (1). Wal- ter Cannon referred to the complex bio- logical responses necessary to maintain a steady state in the body as homeostasis (2). In a series of landmark experiments during the early part of the 20th century, Cannon recognized the importance of the sympathomedullary system in react- ing to stressful events evoked by acute 452 physical or psychobiological stressors (3). He observed that the tissues liberate a humoral agent which he termed "sym- pathin." This was later identified as epi- nephrine (adrenaline) and norepine@- rine (noradrenaline) (4). In 1936 Selye reported that diverse noxious agents cause an enlargement of the adrenal cortex as a consequence of the "stress syndrome" (5). During the following three decades many investiga- tors observed that a variety of stressful events cause a release of ACTH from the anterior pituitary (6). The secreted ACTH stimulates the synthesis of corti- costeroids in the adrenal cortex. The elevated corticosteroid levels in plasma then inhibit the further release of ACTH from the pituitary. In a series of elegant experiments, Harris demonstrated that the release of ACTH from the pituitary is regulated by a corticotropin-releasing factor (CRF) from the hypothalamus (7). The CRF synthesized in the hypothala- mus reaches the pituitary by a private portal blood supply. It then stimulates the secretion of ACTH from the pitu- itary. After a long period of intensive investigations CRF was isolated and pu- rified, and its structure was character- ized as a 41 amino acid peptide by Vale and co-workers (8). CRF was thought to be the major if not the sole means of releasing ACTH from the pituitary. Re- cent experiments indicate that ACTH can also be released and regulated by catecholamines and other hormones. Catecholamines, Glucocorticoids, and Sympathoadreno Activity A variety of stressors cause an in- creased activity of the sympathetic ner- vous system and adrenal medulla (2). This activity results in a discharge of epinephrine and norepinephrine into the blood stream and changes in the activity of enzymes that synthesize catechol- amines and in the concentrations of nor- epinephrine and epinephrine in the brain. With prolonged stress, marked compen- satory changes in the activity of the catecholamine biosynthetic enzymes ty- rosine hydroxylase, dopamine P-hydrox- ylase, and phenylethanolamine N-meth- yltransferase (PNMT) occur. These changes in enzyme activity are regulated to varying degrees by glucocorticoids, ACTH, and neuronal activity. When mice are subjected to psychoso- cial stressors through competition for food and living space, they show in- creases in blood pressure, adrenal weight, and catecholamine concentra- tions in the adrenal medulla (9). The biosynthetic enzymes tyrosine hydroxy- lase and PNMT are both increased in mice experiencing excessive social stim- ulation. Forced immobilization of rats The authors are members of the Section on Phar- macology, Laboratory of Clinical Science, National Institute of Mental Health, Bethesda, Maryland 20205. SCIENCE, VOL. 224 with the use of a model that combines emotional stress (escape reaction) and physical stress (muscle work) activates both the sympathetic adrenal medullary and adrenocortical systems and the pitu- itary gland (IO). The activities of tyrosine hydroxylase, dopamine P-hydroxylase, and PNMT are increased in this type of stress (II). Stress such as that associated with swimming, electroshock, and insu- lin also elevate these biosynthetic en- zymes (II). Forced immobilization of rats can cause an increase in the plasma epinephrine and norepinephrine concen- trations (12). Mild stressors such as opening a cage door or handling a rat produces an eightfold increase in plasma epinpehrine concentrations. stress the activity of tyrosine hydroxy- these enzymes. As in the case of PNMT, lase is rapidly elevated in the adrenal removal of the pituitary gland of the rat medulla without any change in the num- results in a gradual reduction of tyrosine ber of enzyme molecules (16). In chronic hydroxylase and dopamine P-hydroxy- stress, however, the activities of tyro- lase activity in the adrenal medulla (22). sine hydroxylase (17) and dopamine p- *In hypophysectomized rats subjected to hydroxylase are increased (18) in sympa- immobilization stress the reduced dopa- thetic nerves and the adrenal medulla as mine P-hydroxylase activity in the adre- a result of an increase in the number of nal medulla can be restored by ACTH enzyme molecules (16). If the pregangli- and dexamethasone. From these experi- The adrenal gland, important in stress reactions, consists of at least two ana- tomically and chemically distinct struc- tures: an inner medullary area containing catecholamine-producing chromaffin cells and an outer cortical region in which glucocorticoids are synthesized. In most mammals, the adrenal cortex and medulla are contiguous and the main catecholamine produced is epinephrine. In some species, however, the chromaf- fin tissue is separated from the adrenal cortex and the predominant catechol- amine found is norepinephrine (IS). This suggested that a substance produced in adrenal cortical tissue might gain access to the medulla and regulate the conver- sion of norepinephrine to epinephrine. Such substances were thought to be glu- cocorticoids whose synthesis is stimulat- ed by ACTH. Whether or not giucocorti- coids and indirectly ACTH modify the formation of epinephrine in the adrenal medulla was determined by subjecting animals to hypophysectomy and, several days later, measuring the activity of PNMT, the epinephrine-forming enzyme in the medulla (14). The activity of PNMT showed marked decrease. Fur- thermore, the repeated administration of ACTH or the potent glucocorticoid dexamethasone to hypophysectomized rats restored PNMT activity to almost normal values. PNMT activity in the adrenal medulla is much lower in hy- pophysectomized rats subjected to im- mobilization stress (15), but this de- crease can be prevented or reversed by treatment with ACTH or glucocorti- coids. These experiments indicated the interrelationship of ACTH in the pitu- itary and glucocorticoids in the adrenal cortex in effecting the synthesis of epi- nephrine (Fig. 1). Summary. Stress stimulates several adaptive hormonal responses. Prominent among these responses are the secretion of catecholamines from the adrenal medulla, corticosteroids from the adrenal cortex, and adrenocorticotropin from the anterior pituitary. A number of complex interactions are involved in the regulation of these hormones. Glucocorticoids regulate catecholamine biosynthesis in the adrenal medulla and catecholamines stimulate adrenocorticotropin release from the anterior pituitary. In addition, other hormones, including corticotropin-releasing factor, vasoac- tive intestinal peptide, and arginine vasopressin stimulate while the corticosteroids and somatostatin inhibit adrenocorticotropin secretion. Together these agents appear to determine the complex physiologic responses to a variety of stressors. - onic innervation to sympathetic nerves of the superior cervical ganglia of the rat are cut, the enzyme activity in cell bod- ies and nerve terminals is not increased, indicating that this mode of compensa- tory regulation of the enzyme is a trans- synaptic event (19). The transsynaptic induction of tyrosine hydroxylase ap- pears to be due to stimulation of acetyl- choline nicotinic receptors (20). In- creased stimulation of sympathetic nerves also elevates the activity of dopa- mine P-hydroxylase (18) and to a smaller extent PNMT (21). Cutting the presynap- tic nerves will block the increase of dopamine P-hydroxylase and PNMT, in- dicating a transsynaptic induction of ments it appears that the stress-induced increase in tyrosine hydroxylase is due mainly to neuronal activity, whereas do- pamine P-hydroxylase is affected by both nerve activity and the pituitary ad- renal axis and PNMT is controlled main- ly by ACTH and glucocorticoids (Fig. 1). By means of immunocytochemistry and double labeling with radioisotopes, the splanchnic nerve innervating the adrenal medulla has been shown to induce de novo synthesis of dopamine P-hydroxy- lase and PNMT while glucocorticoids inhibit degradation of these enzymes (23). Rats subjected to a variety of stressors show considerably reduced concentra- Fig. 1. Regulation of catecholamine bio- synthesis in the adre- nal medulla. The expression of cate- cholamines in the ad- renal medulla is regu- lated at a number of different steus in the biosynthetic- path- Anterior pituitary Af way. Tyrosine hy- droxvlase (TH) activi- ty, ihe rate limiting step in dopamine syn- thesis, is affected by nerve activity and to a minor extent by glu- cocorticoids of the adrenal cortex. Dopa- mine P-hydroxylase (DBH) activity is reg- ulated by nerve activ- itv and elucocorti- Nerve The activity of tyrosine hydroxylase, a rate-limiting enzyme in catecholamine biosynthesis, can increase considerably in response to stressors (II). In acute coids. Pheiylethanol- amine N-methyltrans- Adrenal gland ferase (PNMT), the enzyme that converts norepinephrine to epinephrine, is predominantly regulated by glucocorticoids and to a small degree by nerve activity. Glucocorticoid synthesis is stimulated by ACTH which is released from the anterior pituitary. 4 MAY 1984 453 ni tions of brain norepinephrine (24). By means of a technique for the precise dissection of small areas of the brain (25) it was observed that the stress-induced depletion of norepinephrine occurs in specific brain nuclei such as the nucleus tractus solitarius, and the arcuate, peri- ventricular, and ventromedial hypotha- lamic nuclei (26). Epinephrine-contain- ing neurons are also present in the brain and localized in the brainstem (27). Im- mobilization stress causes a selective depletion of epinephrine in nucleus trac- tus solitarius, locus ceruleus, and para- ventricular and arcuate nuclei (28). Neu- rons containing CRF have been found in hypothalamic regions such as the para- ventricular nucleus (29). Recent studies have shown that inhibiting brain PNMT activity enhances CRF immunoreactiv- ity in the paraventricular nucleus, sug- gesting that epinephrine may regulate the activity of CRF-containing neurons (30). Such an interaction between adrenergic and CRF-containing neurons may be im- portant in mediating stress-related re- sponses. Although a variety of stressors in- crease the catecholamine concentrations in urine (31), which can serve as a mea- sure of stress, catecholamine concentra- tions in plasma are generally considered Table 1. Multireceptor control of cyclic AMP formation and ACTH release in mouse pituitary tumor cells. An EDSo value is the concentration of drug at which 50 percent of the maximum sttmulation occurs; an ICsO value is the concentration of drug at which 50 percent of the maximum inhibition occurs. N.D., not determined. Additions (references) ACTH release Cyclic AMP Stimulants CRF (90) P-Adrenergic agonists (48) (-)-Isoproterenol (+)-Isoproterenol (- )-Epinephrine (-)-Norepinephrine VIP (31) . . Forskolin (51, 91) Cholera toxin (61) Potassium (92) Calcium ionophores A23 187 (92) Ionomycin (92) Inhibitors Dexamethasone* CRF (90) lsoproterenol (48) VIP (58) Forskolin (51) Somatostatint CRF (93) Isoproterenol (93) VIP (93) Forskolin (93) Potassium (64, 92) Melittin (61) A23187 (61) Ionomycin (92) X537A (52, 92) Receptor binding Somatostatin (64) PZ-Adrenergic (48) Maximum response (percent of basal) EDso (nM) Maximum response (percent of basal) EDso (nM 300 to 400 4 to 5 500 2 200 to 400 I 600 50 100 N.D. 100 N.D. 200 to 400 50 600 100 200 to 400 200 400 1,000 250 1 250 1 500 1,ooO 4,000 10,000 500 N.D. 4,000 N.D. 400 N.D. 100 N.D. 400 N.D. 100 N.D. 400 5mo N.D. N.D. Percentage inhibition of stimulated response Go (nM) Percentage inhibition of stimulated response Go (nM) 100 I 0 N.D. 100 I 0 N.D. 60 IO 0 N.D. 90 5 0 N.D. 100 1 80 to 90 0.1 100 1 80 to 90 0.1 100 1 80 to 90 0.1 100 1 80 to 90 0.1 100 0.4 80 to 90 0.1 0 N.D. N.D. N.D. 0 N.D. N.D. N.D. 50 N.D. N.D. N.D. 100 N.D. N.D. N.D. Dissociation constant (nM) Maximum binding (fmole/mg protein) I.7 11 141 64 I "Cells were treated with IO-`M dexamethasone. The cells were washed and then dexamethasone and various stimulants were added. ACTH released into the medium and intracellular cyclic AMP content were measured after such manipulations. YSomatostatin and various stimulants were added simultaneously to the cells. 454 to be a more precise measure of the stress-induced activation of the sympa- thetic medullary system. Until recently it was difficult to obtain a reliable measure of plasma norepinephrine and epineph- rine because of their extremely low con- centrations. However, with the introduc- tion of specific and highly sensitive ra- dioenzymatic assays for catecholamines it has become possible to determine their concentrations (32) during basal condi- tions and stressful situations in humans. In general, plasma norepinephrine levels reflect the activity of the sympathetic nerves while epinephrine is a measure of secretion from the adrenal medulla. Pos- tural changes can cause two- to threefold elevations in plasma norepinephrine but negligible changes in epinephrine (33). Public speaking results in a twofold in- crease in plasma epinephrine and a 50 percent increase in norepinephrine (34). The reverse is true in performing mental arithmetic (3.5). During harassment, type A individuals (coronary prone) have a greater elevation of plasma epinephrine than type B subjects (noncoronary prone) (36). Depressed subjects show increased basal levels of plasma norepi- nephrine and epinephrine which has been related to their degree of anxiety (33. Propranolol, a P-adrenoreceptor blocking drug, relieves somatic symp- toms (tachycardia, tremors) or acute panic state in anxious persons but ap- pears to be of little use in other forms of anxiety (38). Physical, cold, and thermal stress can cause moderate to marked elevation of plasma catecholamine (39). Exercise also results in a rise in plasma catecholamines (40) that depends on the duration and severity of the exercise. Stress due to surgery, hemorrhage, myo- cardial infarction, hypoglycemia, and hypoxia causes considerable increases in plasma catecholamines (41). Hypoglyce- mia induced by insulin or 2-deoxyglu- case evokes the greatest increase in plas- ma catecholamines, especially epineph- tine. The high levels of catecholamine achieved in hypoglycemia and myocardi- al infarction may be enough to stimulate B-adrenoreceptors on the anterior pitu- itary. Studies in our laboratory have revealed that insulin stress in rats stimu- lates ACTH release from the pituitary (42). This effect is blocked by proprano- 101 and since insulin stress raises plasma catecholamine, the rise in ACTH release appears to be due to the direct action of epinephrine on the pituitary. It would be of clinical interest to examine the effect of propranolol on the plasma concentra- tions of ACTH and glucocorticoids in patients with hypoglycemia. SCIENCE, VOL. 224 Multireceptor Release of ACTH from Anterior Pituitary Adrenocorticotropin is synthesized and released from the anterior pituitary. The availability of synthetic CRF (8, 43) and of a mouse anterior pituitary cell line (AtT-20/D16-16) that secretes ACTH made it possible to study the mechanism of release of ACTH and the involvement of catecholamines, glucocorticoids, and other hormones in this secretory pro-' cess. Recent investigators of ACTH se- cretion from the pituitary have usually used primary cultures of the rat adeno- hypophysis (44). Although much useful information has been obtained with such preparations, the heterogeneity of the cell types and the low density of the ACTH secreting cells (2 to 3 percent of the total cell population) in the anterior pituitary has limited the characterization of factors directly controlling ACTH re- lease. The AtT-20 cell line was used previously to examine the processing of ACTH from its precursor protein pro- opiomelanotropin (POMC) as well as the storage and secretion of ACTH and p- endorphin (45). This cell line appears to be homogeneous with regard to cell type and, in contrast to primary anterior pitu- itary cultures, predominantly releases hormones of the POMC family of pep- tides. Normal anterior pituitary cells re- spond to synthetic CRF by releasing ACTH and P-endorphin (43). AtT-20 cells also secrete immunoreactive ACTH and P-endorphin in response to CRF (Table 1). Analogs of CRF show the same order of potency in releasing ACTH from AtT-20 cells as observed in normal corticotrophs and, as shown in intact animals and primary cultures of the pituitary, glucocorticoids block the CRF-stimulated release of ACTH in the tumor cells. These findings prompted the use of AtT-20 cells as a model for investi- gating the cellular and molecular mecha- nisms that regulate ACTH secretion from the anterior pituitary. There were previous indications that catecholamines can stimulate ACTH re- lease in vivo and in vitro (46). The com- plexities involved in using the intact ani- mal and primary cultures made it difficult to interpret the precise mechanism whereby catecholamines induced this re- lease. Norepinephrine stimulated ACTH release from AtT-20 cells in a calcium- dependent manner (47). Ligand binding studies with tritiated dihydroalprenolol, a P-adrenoreceptor antagonist, indicated the presence of a P-adrenoreceptor on AtT-20 cell membranes which binds cat- 4 MAY 1984 echolaminergic agents with high affinity (48). Isoproterenol, a P-adrenoreceptor agonist, as well as epinephrine induced a potent and stereoselective increase of ACTH release from mouse tumor cells which was calcium-dependent and blocked by the P-adrenoreceptor antago- nist propranolol(48). Two subtypes of p- adrenoreceptors are known, PI and p2 (49). &-Adrenoreceptors are most sensi- tive to epinephrine whereas p,-adrenore- ceptors are equally responsive to epi- nephrine and norepinephrine. Pharmaco- logic characterization showed that p2- receptors are present on AtT-20 cells and could mediate the release of ACTH (48). The presence of pZ- but not p,-adrenore- ceptors has been reported in the anterior pituitary, but whether they are located on a specific cell type has not been Outside Glucocortlcoids Gi*+ 1 Catecholamlnes VIP I I CRF I determined (50). As in the case of CRF, dexamethasone (Table 1 and Fig. 2) blocks catecholamine-evoked secretion of ACTH from AtT-20 cells (51). The longer these cells have been incubated with glucocorticoids the greater the de- gree of inhibition of ACTH release. In anterior pituitary membranes CRF also stimulates adenylate cyclase activi- ty (44). CRF increases the formation of adenosine 3',5'-monophosphate (cyclic AMP) in primary cultures and activates cyclic AMP-dependent protein kinase, an intracellular effector of cyclic AMP (52). CRF also stimulates adenylate cy- clase and cyclic AMP-dependent protein kinase activity in AtT-20 cells (53). Sup- port for a critical role of cyclic AMP in regulating ACTH release comes from studies with forskolin. This diterpene, SRIF R K+ I ?I I Inside I Cyclic AMP I 0 0 ri 0 Ca*+ Nongranular ACTH / Outside 4 + Fig. 2. Molecular mechanisms involved in ACTH release. The release of ACTH can be stimulated (+) by various hormones such as CRF. catecholamines acting on P-adrenoreceptors (PI?), or a,-adrenoreceptors (not shown), VIP, or vasopressin (not shown). Each agonist acts on separate and specific receptors. The hormone-induced secretion of ACTH involves a multitude of intracellular second messengers. Secretagogues can activate adenylate cyclase (AC) to form cyclic AMP. A guanine nucleotide stimulatory protein (N,) is required for hormone activation of AC. Cyclic AMP activates protein kinase (PK) which catalyses the phosphorylation (P,) of a protein substrate (KS). The phosphorylated KS may induce ACTH synthesis (Syn ACTS) or the release of granular ACTH (in the circle). The secretion of nongranular ACTH appears to be regulated differently from granular ACTH. Hormones and membrane depolarization, induced by extracellular potassium (K') may stimulate calcium (Ca'+) influx or mobilization from intracellular compartments. Other intracellular events (1) induced by secretagogues may involve changes in phospholipid methylation, protein carboxymethylation, phosphatidylinositol turnover, C-protein kinase, and glycosyltransferase activity. ACTH release can be inhibited (-) by at least two hormones. Glucocorticoids can inhibit ACTH synthesis or release. Somatostatin (SRIF) blocks the activation of AC or Ca" mobilization. k guanine nucleotide inhibitory protein (NJ mediates SRIF's inhibition of AC. Coupling of N, with SRIF receptors leads to either a direct blockage of the catalytic subunit of AC or an inhibition of N, activity. 455 which may directly activate adenylate cyclase (.54), stimulates with equal po- tency ACTH release (Table 1 and Fig. 2) and cyclic AMP-dependent protein ki- nase as well as the phosphorylation of several distinct proteins in AtT-20 cells (53). It is possible that these endogenous proteins mediate the effect of cyclic AMP on ACTH release. Isoproterenol also stimulates adenylate cyclase activi- ty (Fig. 2) and cyclic AMP-dependent protein kinase in AtT-20 cells, indicating that cyclic AMP may mediate the effect of P-adrenoreceptor agonists on ACTH release (53). The increase in ACTH se- cretion induced by isoproterenol is blocked by calcium antagonists such as verapamil or by the absence of calcium in the extracellular medium (51). The activation of adenylate cyclase by iso- proterenol is not prevented by calcium ionophore blocking agents nor is the stimulation of cyclic AMP-dependent protein kinase altered by the depletion of extracellular calcium (53). This suggests that the actions of Ca*+ in the release of ACTH occur at a step distal to kinase activation by cyclic AMP. The secretion of ACTH is stimulated by increasing extracellular calcium concentrations. Furthermore, isoproterenol and cate- cholamines change the membrane po- tential of AtT-20 cells and increase the frequency of action potentials on these cells by a calcium-dependent mechanism (55). Since catecholamine-evoked ACTH release from AtT-20 cells is also dependent on the presence of calcium, there are probably at least two intracellu- lar mechanisms involving calcium or cy- clic AMP that participate in corticotro- pin secretion. pituitary (57), also evokes the secretion of ACTH from AtT-20 cells (Table I) in a dose-dependent manner (51). This pep- tide increases cyclic AMP accumulation in these cells as well as cyclic AMP- dependent protein kinase activity (53). VIP also stimulates ACTH release from human anterior pituitary tumor cells (58). The effect of VIP on ACTH secre- tion, like that of other hormones, is blocked by glucocorticoids (51). Argi- nine-vasopressin is also found in hypo- thalamic neurons and was one of the first hormones proposed to have CRF-like actions. Vasopressin increases the secre- tion of ACTH from primary cultures of the anterior pituitary (59). This releasing action is not mediated by cyclic AMP but may involve some other intracellular ef- fector system (56). Inhibition of ACTH Release In addition to the P-adrenoreceptor control of ACTH release, an at-ad- renergic-like mechanism may also be involved (56). In primary cultures of the anterior pituitary, ACTH release stimu- lated by epinephrine and norepinephrine is blocked by the potent and selective aI- adrenergic receptor antagonist, prazo- tin. Dopamine and serotonin antagonists also block epinephrine-induced ACTH secretion. at-Receptor agonists such as phenylephrine and methoxyamine are poor stimulators of ACTH release, sug- gesting the presence of a nonclassical a,- receptor mediated release of ACTH which does not involve activation of adenylate cyclase. Besides synthetic CRF and catechol- amines, other hormones have been found to stimulate ACTH release. Vaso- active intestinal peptide (VIP), present in the hypothalamus and known to stimu- late prolactin release from the anterior Glucocorticoids consistently block basal and stimulate ACTH release from the anterior pituitary in the intact animal as well as cell preparations in vitro (Ta- ble 1). Glucocorticoids may act through several mechanisms to inhibit ACTH se- cretion (60). Long-term treatment of ani- mals or AtT-20 cells with dexametha- sone reduces ACTH messenger RNA activity, indicating an inhibition of ACTH synthesis at some pretransla- tional site. The reduction of ACTH mes- senger RNA production induced by glu- cocorticoids correlates with their intra- cellular glucocorticoid receptor binding activity. Short-term treatment (2 hours) of AtT-20 cells with dexamethasont seems predominantly to effect hormone- and cyclic AMP-stimulated ACTH re- lease rather than ACTH synthesis, since the rise in ACTH release promoted by the calcium ionophore A23187 is not reduced by brief glucocorticoid treat- ment (53). Treatment of AtT-20 cells with dexamethasone for short intervals, while inhibiting isoproterenol-, CRF-, and forskolin-stimulated ACTH release (Table 1 and Fig. 2), does not affect the activation of cyclic AMP-dependent protein kinase by these secretagogues (53). The ability of these secretagogues to stimulate cyclic AMP accumulation is not affected by short- or long-term treat- ment with dexamethasone. The inability of glucocorticoids to block CRF-stimu- lated cyclic AMP formation was also reported in primary cultures of the ante- rior pituitary (44). Preliminary studies have shown that the enzyme phospholi- pase Az may be involved in releasing ACTH since melittin, an activator of this 456 enzyme, stimulates ACTH secretion (61). Glucocorticoids have been ob- served in some cell systems to induce the synthesis of lipomodulin, a protein that inhibits phospholipase A2 activity (62). If glucocorticoids can rapidly induce either the synthesis or the mobilization of this protein, then the rapid inhibitory effects of dexamethasone on ACTH release may be related to a blockade of phospholi- pase A2 activity. Thus, inhibition of phospholipase A2 activity may serve as another mechanism by which this class of steroids act in the regulation of ACTH secretion. Somatostatin (SRIF) also inhibits ACTH release in AtT-20 cells (Table 1). This 14 amino acid peptide is of hypotha- lamic origin and is known to block the secretion of growth hormone, prolactin, and thyroid-stimulating hormone from the anterior pituitary (63). AtT-20 cells have SRIF receptors which, when stimu- lated, cause a reduction in ACTH secre- tion (61, 64) evoked by potassium, CRF, isoproterenol, VIP, cholera toxin, or forskolin. SRIF reduces the ability of these secretagogues to increase cyclic AMP accumulation (61) (Table 1). These observations suggest that SRIF can block ACTH release by inhibiting the activation of adenylate cyclase. SRIF also inhibits forskolin-stimulated cyclic AMP formation in cyc- variants of S49 lymphoma cells that are deficient in the guanine nucleotide stimulatory protein (NJ required for most hormones to acti- vate adenylate cyclase (65). From these data it was proposed that SRIF acted through a guanine nucleotide inhibitory protein (Ni) to reduce adenylate cyclase activity (61, 65) as well as by a stimula- tion of guanosine triphosphatase activity (65). A useful agent in studying the man- ner by which hormones inhibit adenylate cyclase is a toxin derived from the nacte- rium Bordetella pertussis. This toxin in- duces the adenosine diphosphate (ADP)- ribosylation of a 41,000-dalton protein believed to be Ni (66). The toxin also blocks the inhibitory effects of hormones and guanine nucleotides on adenylate cyclase in many tissues (66). The inhibi- tion of growth hormone release by SRIF from primary cultures of the anterior pituitary is also blocked by the toxin (67). In membranes of AtT-20 cells, Bor- detella pertussis toxin induces the ADP- ribosylation of a 41,000-dalton protein and also prevents the inhibitory effect of SRIF on forskolin-, CRF-, or isoprotere- nol-stimulated cyclic AMP formation and ACTH release (68). These findings suggest that SRIF can act through Ni to inhibit hormone-induced ACTH release. SCIENCE, VOL. 224 Interactions of ACTH-Releasing Factors The ACTH secretagogues can act indi- vidually or in concert to regulate the release of ACTH. Hypothalamic ex- tracts are more potent in releasing ACTH than any secretatogue alone. These extracts appear to contain several different CRF-like factors. In primary cultures of the anterior pituitary, vaso- pressin added together with synthetic CRF induced a greater release of ACTH than CRF alone, indicating that vaso- pressin can potentiate the action of CRF (59, 69). Although vasopressin does not alter cyclic AMP accumulation in the anterior pituitary, it causes a fourfold potentiation in the stimulation of cyclic AMP synthesis by CRF, suggesting that vasopressin improves the efficiency of coupling between CRF receptors and adenylate cyclase. Epinephrine, by acti- vating at-adrenoreceptors in the anterior pituitary, also potentiates synthetic CRF's stimulation of ACTH release and, like vasopressin, enhances the cyclic AMP response to CRF (70). These find- ings indicate a synergism between vaso- pressin, a,-adrenergic agonists, and syn- thetic CRF in releasing ACTH. &- Adrenoreceptor agonists and CRF also interact to regulate ACTH release. When CRF and isoproterenol are added togeth- er, the increase in ACTH secretion is less than additive, suggesting that these secretagogues act through a common mechanism (51). This intracellular mech- anism is distal to cyclic AMP accumula- tion since the combined application of CRF and isoproterenol produce additive effects on cyclic AMP formation. VIP, another ACTH secretagogue, appears to release ACTH through a process inde- pendent of CRF or P-adrenergic agonists (50). VIP, together, with isoproterenol or CRF, causes an additive increase in both ACTH secretion and cyclic AMP pro- duction. These findings indicate that CRF, isoproterenol, and VIP may act on separate compartments of adenylate cy- clase as well as other second messenger systems (Fig. 2). Desensitization Although hormones can induce rapid and pronounced responses from cells, their persistent presence can induce de- sensitization (71). Corticotrophs become refractory to CRF after prolonged expo- sure to this peptide (72). This desensiti- zation is manifest as a reduced maximal ability of CRF to stimulate both cyclic AMP formation and ACTH release. 4 MAY 1984 ACTH content is not affected by CRF treatment, indicating that cells are not depleted of the peptide hormone. Fors- kolin-stimulated cyclic AMP accumula- tion or ACTH release is not reduced by CRF treatment, suggesting that both adenylate cyclase and the intracellular mechanisms mediating stimulus-secre- tion coupling are unaffected in the desen- sitized cells. Thus, either CRF receptors are lowered in density or their coupling to adenylate cyclase is impaired. Studies in which anterior pituitary membranes are exposed to `251-labeled CRF indicate that desensitization involves the loss of CRF receptors (73). Adrenalectomy, a procedure that abolishes the glucocorti- coid feedback inhibition of CRF release in the hypothalamus, markedly de- creased CRF receptor binding in the pituitary 4 to 6 days after surgery. The density of these sites returns almost to normal after treatment with dexametha- sane. Thus, CRF receptors can be down- regulated and this may explain the de- sensitization observed in primary cul- tures. Recent studies have also shown that the effect of CRF on ACTH release in vivo is reduced after prior treatment of animals with CRF, suggesting that de- sensitization can occur under physiolog- ic conditions (74). Vasopressin not only potentiates CRF-stimulated ACTH release but can also increase the ability of CRF to desen- sitize its own receptor (75). Treatment of primary cultures of the anterior pituitary with a fixed concentration of arginine- vasopressin and varying amounts of CRF reduces the amount of CRF needed to desensitize its receptors. Thus, vaso-' pressin and CRF act synergistically to release ACTH and regulate CRF recep- tors. P-Adrenoreceptors in many cell types are also readily desensitized (76). This was also found to be the case for mouse pituitary tumor cells. Treatment of AtT- 20 cells with isoproterenol results in a marked reduction of cyclic AMP forma- tion and release of ACTH after restimu- lation with the catecholamines (77). A decreased binding of [3H]dihydroalpren- 0101 becomes apparent after 20 hours of treatment with isoproterenol. The re- duced binding is associated with a de- creased density of j3-adrenoreceptors but no change in receptor ligand affinity. The desensitization of the cyclic AMP accu- mulation and ACTH secretion responses were observed before there was a de- crease in receptor density. These find- ings suggest that the desensitization of the P-adrenoreceptor is a two-step pro- cess. The first is rapid in onset and shows a reduced capacity of catechol- amines to elevate cyclic AMP accumula- tion and ACTH release. The second step is slower and is associated with a loss of p-adrenoreceptors from cell membranes (down regulation). The rapid desensitization of the p- adrenoreceptors to stimulation of cyclic AMP synthesis and ACTH release with- out changes in receptor density could be due to the uncoupling of the receptor from the adenylate cyclase complex, de- creased activity of adenylate cyclase, or changes in the ACTH secretory process. These possibilities were examined by first treating AtT-20 cells with isoproter- enol to reduce the responsiveness to cyclic AMP elevation and ACTH secre- tion by about 50 percent (77). The cells were then treated with forskolin to di- rectly stimulate adenylate cyclase. The generation of cyclic AMP and release of ACTH in forskolin-treated cells were the same as that of the fully sensitized cells. This experiment indicates that during the early desensitization of the P-adreno- receptors, the adenylate cyclase and ACTH secretory mechanism is normal and that the desensitization is due to an uncoupling of the receptor from adenyl- ate cyclase. P-Adrenoreceptor desensitization is homologous (only desensitized to its own receptor) since CRF- and VIP-stim- ulated cyclic AMP accumulation and ACTH release were unaffected by cate- cholamine treatment of AtT-20 cells. CRF receptors on normal corticotrophs are also regulated independently of cate- cholamine receptors (72, 75). The inde- pendent nature of the desensitization of these receptors indicates that the mecha- nisms involved are specific for each re- ceptor. Such a property would allow corticotrophs to respond to some stimuli despite the loss of responsiveness to other CRF-like substances. Somatostatin can also regulate the sensitivity of its own receptor (78). Prior exposure of mouse anterior pituitary cells to SRIF lessens SRIF's antagonism of CRF-, VIP-, isoproterenol-, and fors- kolin-stimulated cyclic AMP accumula- tion and ACTH release. Treatment with SRIF increases the formation of cyclic AMP in response to forskolin in these cells. This increase is delayed in onset, slow to recover, and is blocked by the protein synthesis inhibitor cyclohexa- mide. This suggests that prolonged treat- ment of AtT-20 cells with SRIF desensi- tizes SRIF receptors and causes a com- pensatory sensitization of adenylate cy- clase through a process requiring protein synthesis. In cultures of brain cells. pro- 457 Fig. 3. Multihormonal control of ACTH re- lease. ACTH release from the anterior pitu- itary is regulated by a number of different hormones. CRF and vasopressin, which are present in hypothalamic neurons, reach the anterior pituitary by a portal system and stim- ulate (+) ACTH release. The secretion of the releasing factors may be inhibited (-) by an epinephrine input to the hypothalamus. Hy- pothalamic VIP stimulates ACTH release from both human and mouse tumor cells suggesting a possible role of this peptide as a Epinephrine CRF-like substance. Neuronal norepineph- I+) rine or epinephrine released from the adrenal medulla can also evoke ACTH release. ACTH stimulates the synthesis of glucocort- coids in the adrenal cortex. Glucocorticoids can stimulate the synthesis of epinephrine or inhibit ACTH release. Glucocorticoids can either act directly on the anterior pituitary to inhibit the formation of messenger RNA for ACTH or inhibit ACTH release. In addition, these steroids can act in the hypothalamus to effect CRF release (not shown). Somatostatin (SRIF) can block the evoked release of ACTH from tumor cells and reduces the plasma concentrations of ACTH in Nelson's syndrome (94, suggesting a role for this peptide in ACTH secretion. longed application of SRIF produced an adaptive response that was manifest as a reduced ability of SRIF to stimulate cell firing activity (79). Continued treatment of anterior pituitary cells with SRIF re- duces the subsequent ability of the pep- tide to inhibit the release of growth hor- mone and thyroid-stimulating hormone (TSH) (80). These findings indicate that SRIF receptors on many cell types can be self-regulated. A model detailing the multireceptor regulation in ACTH re- lease in AtT-20 cells is shown in Fig. 2. The Multireceptor Release of ACTH in vivo The studies showing that ACTH re- lease in vitro is stimulated by multiple factors raises the question of whether the release of ACTH in vivo is under multi- hormonal control. The injection of syn- thetic ovine CRF in rats causes an imme- diate increase in plasma ACTH concen- trations (81). This stimulation is dose- dependent and is neutralized by anti- bodies to CRF. The ACTH release induced by acute ether stress is partially blocked by CRF antibodies, indicating that molecules with immunologic charac- teristics similar to synthetic CRF are stress mediators (81). The lack of total blockage of stress-evoked ACTH release in rats by the CRF antibodies suggests that hormones other than CRF are in- volved in promoting the release of ACTH in vivo. Vasopressin also causes the release of ACTH in the intact rat. This stimulation is dose-dependent and prevented by a vasopressin antagonist (81, 82). The physiologic condition of the animal has an important role in the stimulation of ACTH secretion by vasopressin. Ani- 458 mals anesthetized with neuroleptics, opi- ates, and Nembutal (conditions that block CRF release) respond to vasopres- sin with a smaller elevation in ACTH release than that found in awake, freely moving animals. Immunoneutralization of CRF in nonanesthetized rats also low- ers stimulation of ACTH release by vasopressin. suggesting a dependence on CRF for the ACTH releasing action of vasopressin. Vasopressin potentiates CRF-stimulated ACTH release in anes- thetized rats, indicating that vasopressin and CRF act in a synergistic manner to regulate ACTH release in vivo as well as in vitro. Catecholamines also appear to stimu- late ACTH release in vivo by a direct action on the anterior pituitary. Periph- eral injections of epinephrine increase plasma concentrations of ACTH in in'- tact rats (46). The increase in plasma ACTH caused by either epinephrine or (-)-isoproterenol is stereospecifically blocked by propranolol, suggesting that P-adrenoreceptors are linked to the re- lease of ACTH in vivo (46). P-Adrenore- ceptor agonists have been proposed to stimulate ACTH release in vivo by act- ing through "central mechanisms." since hypothalamic lesions in female rats prevented the increase in plasma ACTH induced by (-)-isoproterenol (82, 83). The process by which catecholamines gain access to the brain to initiate these central effects is not known. Previous work has demonstrated that catechol- amines do not cross the blood-brain bar- rier (84). In contrast to these studies, however, Mezey et al. (85) found that (-)-isoproterenol stimulated ACTH re- lease from male rats in which the hypo- thalamus was separated from the pitu- itary by either stalk-transection or le- sions of the median eminence. The effect of isoproterenol was blocked by pro- pranolol but not by the selective PI- adrenoreceptor antagonist practolol. Sal- mefamol, a PI-adrenoreceptor agonist, also stimulated ACTH release in stalk- sectioned animals, indicating that p2- adrenoreceptors can mediate the stimu- lation of ACTH release by catechol- amines in vivo. Isoproterenol-stimulated ACTH release in stalk-transected ani- mals is blocked by dexamethasone, sug- gesting that the ACTH release induced by P-adrenoreceptor agonists originates from the anterior pituitary. These find- ings are consistent with the previous studies of Fortier (86) and McDermott et al. (87) who used pituitary transplants to examine the direct action of epinephrine on an ACTH-mediated response. In these studies, the anterior pituitary was placed into the anterior chamber of the eye of hypophysectomized rats. Injec- tion of small quantities of epinephrine into the eye reduced the level of circulat- ing white blood cells (eosinophilia) which is believed to accompany an in- crease of ACTH release. Similar injec- tions of epinephrine into the other eye did not produce this response. These data as well as the findings in stalk- transected animals indicate that epineph- rine can act directly on the anterior pitu- itary possibly by way of P-adrenorecep- tors to stimulate ACTH release. Conclusion Studies conducted in vitro and in vivo indicate that the release of the stress hormone ACTH is controlled by com- plex regulatory mechanisms. Multiple factors such as CRF, vasopressin. cate- cholamines, and conceivably other hor- mones stimulate ACTH release by di- rectly acting on the anterior pituitary (Fig. 3). Glucocorticoids and possibly SRIF may inhibit the secretion of ACTH also by a direct action on anterior pitu- itary. Various hormones can indirectly control ACTH release by acting on cen- tral locations to modify the secretion of these releasing or inhibiting factors. In addition, corticotrophs become refrac- tory to hormones that stimulate or inhibit ACTH release. In most cases, the phys- iologic significance of the desensitization of hormone receptors on corticotrophs is not known. However, down regulation of glucocorticoid receptors in cortico- trophs (88) may be responsible for the reduced ability of dexamethasone to lower plasma ACTH concentrations in some depressed patients (89). The release of ACTH is the final out- come of the interplay among the hypo- SCIENCE.VOL. 224 thalamus, adrenal cortex, adrenal medul- la, and possibly other organs. Depending on the type of stress experienced, it is likely that a number of different hor- mones may singly or in combination af- fect the amount and duration of ACTH secreted. The interaction of the various stress mediators may act to fine-tune the responsiveness of ACTH-secreting cells. References and Notes I. C. Bernard, Les Phenomenes de /a Vie (Li- brairie J-B Bailliere et Fds, Paris. 1878), vol. I. p. 879. 2. W. B. Cannon, Physiol. Rev. 9, 399 (1929); The Wisdom of rhe Body (Norton. New York. 1939). 3. __ and J. E. Uridil. Am. J. Physiol. 58, 353 (1921). 4. 5. 6. i: 9. IO. II. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. L. S. van Euler. Pharmacol. Rev. 6, 15 (1954). H. Selye, Narure (London) 138. 32 (1936). The stress syndrome was subsequently called the general adaptation syndrome by Selye. This syndrome proceeds in three stages: (i) the alarm reaction, (ii) the stage of resistance. and (iii) the stage of exhaustion. F. E. Yates and J. W. Maran. in Handbook of Physiology. R. 0. Creep and E. B. Astwood. Eds. (American Physiological Society, Wash- ington, D.C., 1974). vol. 4. section 7. part 2. p. 367. G. W. Harris, Physiol. Rev. 28, 139 (1948). W. Vale, J. Spiess, C. Rivier, J. Rivier, Science 213, 1394 (1981). I. Axelrod. R. A. Mueller. J. P. Henry, P. M. Stephens, Nature (London) 225, 1059 (1980); J. P. Henry, P. M. Stephens, J. Axelrod, R. A. Mueller, Psychosum. Med. 33, 227 (1971). R. Kvetnansky and L. Mikulaj, Endocrinology 87, 738 (1970). R. Kvetnansky. V. K. Weise. I. I. Kopin, ibid.. p. 744; R. Kvetnansky, G. P. Gewirtz. V. K. Weise, I. J. Kopin. MO/. Pharmacol. 7. 81 (1971); Am. J. Physiol. 220, 9zS (1971). K&etnansky ef al., Endocrmology 103. 1868 R. E. Coupland, J. Endocrine/. 9, 194 (1953). R. J. Wurtman and J. Axelrod, Science 150, 1464 (1965): J. Bio/. Chem. 241, 2301 (1966). R. Kvetnansky. G. P. Gewirtz, V. K. Weise. I. J. Kopin, Endocrinology 87, 1323 (1970). T. H. Joh, D. H. Park. D. J. Reis, Proc. Nat/. Acad. Sri. U.S.A. 75, 4744 (1978). R. A. Mueller, H. Thoenen, J. Axelrod. Science 163. 468 (1969); J. Pharmucol. Exp. T&r. 169. 74 (1969). H. Thoenen, R. A. Mueller. J. Axelrod, Nnrure (London) 221, 1264 (1969); P. B. Molinoff, W. S. Brimijoin, R. M. Weinshilboum, J. Axelrod, Proc. Narl. Acad. Sri. U.S.A. 66, 453 (1970). H. Thoenen, R. A. Mueller. J. Axelrod, J. Pharmacol. Exp. Thu. 169, 249 (1969). R. A. Mueller, H. Thoenen. J. Axelrod, Eur. J. Pharmacol. 10. 51 (1970). H. Thoenen, R. A. Mueller. J. Axelrod. Bio- them. Pharmacol. 19, 669 (1970). G. P. Gewirtz. R. Kvetnansky. V. K. Weise, I. J. Kopin, Mol. Pharmacol. 7. 163 (1971). R. D. Ciaranello. G. F. Wooten, J. Axelrod. J. Biol. Chem. 250, 3204 (1975). J. D. Barchas and D. X. Freedman, B&hem. Pharmaco/. 17. 1232 (1963): E. W. Maynert and R. Levy, J. Phnrmarol. Exp. Thu. 143, 90 (1964); M. Zigmond and J. Harvey, J. Neuro- Vise. Relar. 31, 373 (1970). M. Palkovits. Brain Res. 59, 449 (1973). J. M. Saavedra, R. Kvetnansky. I. J. Kopin, ibid. 160, 271 (1979); J. M. Saavedra, N~uroen- drocrinology 35, 396 (1982). J. M. Saavedra. M. Palkovits, M. J. Brownstein, J. Axelrod, Nature (London) 248,695 (1974): T. Hiikfelt, K. Fuxe. M. Goldstein, 0. Johanson, Brain Res. 66, 235 (1974). R. Kvetnanskv. I. J. Konin. J. M. Saavedra. Brain Res. 155,`387 (19783. F. E. Bloom, J. Battenberg, J. Rivier. W. Vale, Regal. Peprides 4. 43 (1982); W. K. Paul1 et al.. Peplides 1. 183 (1982); S. Cummings, R. Elde, I. Eils, A. Lindael, J. Neurosci. 3, 1355 (1983). 30. E. Mezey ef a/., in preparation. 31. G. Bloom. U. van Euler. M. Frankenhaeser. Acta Physiol. Stand. 58, 77 (1963): L. Levi, Psychosom. Med. 27, 80 (1965). 32. P. G. Passon and J. D. Peuler, Anal. Biochem. 51, 618 (1973); V. K. Weise and I. J. Kopin, Life. SC;. 19, 1673 (1976); M. Da Prada and G. Zurcher, ibid., p. I I61 ; J. D. Peuler and G. A. Johnson, ibid. 21, 625 (1977). 33. C. R. Lake, M. G. Ziegler, I. J. Kopin, Life SC;. 18. 1315 (1976); D. E. Cryer, J. V. Santiago, S. D. Shah. J. C/in. Endocrinol. Merab. 39. 1025 119741 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. , , J. C. Dimsdale and J. Moss, J. Am. Med. Assoc. 243, 340 (1980). A. Leblanc, J. Cote, J. Jobin, F. Labrie, J. Appl. Physiol. 47. 1207 (1979). D. C. Glass er al., Psychophysiology 17, 453 (1980). W. J. Louis, A. E. Doyle, S. N. Anavekar, C/in. Sci. Mol. Med. 48, 239 (1975). I. Tyrer and M. H. Leder. Br. J. C/in. Pharma- col. 1, 387 (1974); I. F. Heisler and D. De Francisco, Am. J. Psychiarry 133, 1389 (1976). Y. D. Kim. C. R. Lake. D. E. Lees, W. H. Schuette, J. M. Bull. V. K. Weise. I. J. Kopin, Am. J. Physiol. 237. H570 (1979): C. R. Lake and M. G. Zeigler. Circulation 57. 774 (1978). H. Galbo, J. J. Hoist, N. J. Christensen, J. ADD/. Phvsiol. 38. 70 (1975): N. J. Christensen ef ai.; Dia6eres 28 (sippI. `i), 58 (1979); N. J. Christensen and 0. Brandsborg, Eur. J. C/in. Invesr. 3. 299 (1973). C. R. Benedict and D. G. Graham-Smith, Q. J. Med. U. Ser. 185. I (1978): J. B. Halter. A. E. Pfug. 6. Porte, /.`&I. E&xrinol. Met&l 45, 936 (1977); N. J. Christensen, K. G. M. M. Alberti. 0. Brandsbore. Eur. J. C/in. Invest. 3. -. 299 (1973). E. Mezey. T. D. Reisine, J. Axelrod, in prepara- tion. J. Spiess, J. Rivier, C. River, W. Vale, Proc. Nat/. Acad. Sri. 1J.S.A. 78. 6517 (1981). F. Labrie ef al., Science -h6, lob? (lb82); V. Giguiere et al., Proc. Narl. Acad. Sci. U.S.A. 79. 3466 11982). R. Main; and' ti. Eipper. J. Biol. Chem. 251, 4115 (1976); J. Roberts, M. Phillips, P. Rosa, E. Herbert. Biochemistry 17.3609 (1978); S. Sabol, Arch. Biochem. Biophys. 203, 37 (1980). W. Vale and C. Rivier, Fed. Proc. Fed. Am. Sot. Exp. Biol. 36,2094 (1977): F. Berkenbosch. I. Vermes, R. Binnekade, F. Tilders. Life Sci. 29. 2249 (1981): F. Tilders. F. Berkenbosch. P. G. Smelik. Enbocrinology 110, 114 (1982). R. Mains and B. Eipper, J. Cell Biol. 89, 21 (1981). T. D. Reisine, S. Heisler, V. Y. H. Hook, J. Axelrod, J. Neurosci. 3. 725 (1983). H. R. Furchgott. in Curecholamines, H. Blaschko and E. Muscholl, Eds. (Springer-Ver- lag, Berlin, 1972). pp. 283-335. S. L. Petrovick, J. K. McDonald, G. 0. Snyder, S. M. McCann, Brain Res. 261, 249 (1983). * T. D. Reisine. S. Heisler. V. Y. H. Hook. J. Axelrod, Biochem. Biophys. Res. Commun. 108, 125) (1982). G8A~udera ef al., J. Biol. Chem. 258. 8039 ~____,. K. Miyazki. T. D. Reisine, J. Kebabian, in preparation. K. Seamon and J. Daly, J. Biol. Chem. 256,9799 (1981). k. Suprenant, J. Cell Biol. 95, 559 (1982); M. Adler et al., Proc. Nat/. Acad. Sci. U.S.A. 80, 2086 (1983). V. Giguere, J. Cote, F. Labrie. Endocrinology 109. 757 11981). w. kotsztejn hr u/., Neuroendocrinology 31, 282 (1980). D. Oliva, S. Nicosia, A. Spada, G. Giannattasio, Eur. J. Pharmacol. 83, 101 (1982). V. Giguere and F. Labrie, Endocrinology 111, 1752 (1982); W. Vale, J. Vaughan, M. Smith, G. Yamamoto. I. Rivier, C. Rivier. ibid. 113, 112 (1983). S. Nakanishi. T. Kita. S. Taii, H. Imura, S. Numa. Proc. Nat/. Acad. Sci. U.S.A. 74, 3283 (1977); M. Nakamura, S. Nakanishi, S. Sueoka, H. Imura, S. Numa, Eur. 1. Eiochem. 86, 61 (1978); H. Watanabe, W. Nicholson, D. Orth, Endocrinology 93, 411 (1973). 61 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85 86. 87 88 E S. Heisler, T. D. Reisine. V. Y. H. Hook, J. Axelrod. Proc. Natl. Acad. Sci. U.S.A. 79.6502 (1982). F. Hirata. E. Schiffmann, K. Venkatasubraman- ian. D. Salomon, J. Axelrod, ibid. 77, 2533 ~t~~~~i F. Hirata, J. Eiol. Chem. 256, 7730 ?. Biazeau. J. Rivier, W. Vale, R. Guillemin, Endocrinology 94, 184 (1974). A. Schonbmnn and A. Tashjian, J. Biol. Chem. 255, 190 (1980); U. Richardson and A. Schon- brunn, Endocrinology 108, 281 (1981). G. Schultz and C. Staunek, Naunyn-Schmiede- berg's Arch. Pharmacol. 322 (suppl.), R5 (1983); K. Jakobs. K. Aktories. G. Schultz, Narure (London) 303. 177 (1983). T. Katada and M. Ui, Proc. Nat/. Acad. SC;. U.S.A. 79, 3129 (1982); J. Eiol. Chem. 256, 8310 (1981); J. Hildebrandt, R. Sekura, J. Codina. R. 1. Yengar. C. Manclark, L. Birnbaumer, Nature (London) 302, 706 (1983). M. Cronin, A. Rogol, G. Myers, E. Hewlett, Endocrinology 113, 209 (1983). T. D. Reisine, Y-L. Zhang, R. Sekura, Biochem. Biophys. Res. Commun. 115. 794 (1983); in preparation. F. Yates ef al., Endocrinology 88, 3 (1971); G. Gillies, E. Linton, P. Lowry. Nature (London) 299,355 (198l);C. RivierandW. Vale, Endocri- nology 113, 939 (1983). V. Gieuere and F. Labrie. E&hem. Bioohvs. Res. ?ommun. 110, 456 (1483). ` , K. Catt. J. Harwood, G. Aquilera, M. Dufau. Nature (London] 280, 109 (1979). T. D. Reisine and A. Hoffman, Biochem. Biophys. Res. Commun.' 111, 919 (1983). P. Wynn. G. Aguilera, J. Morell, K. Catt. ibid. 110, 602 (1983). C. Rivier and W. Vale, Endocrinology 113, 1422 (1983). A. Hoffman, G. Ceda, T. D. Reisine. in prepara- tion. J. W. Kebabian, M. Zatz, J. A. Romero, J. Axelrod. Proc. NatI. Acad. Sri. U.S.A. 72, 3725 (1975): Y. F. Su. T. Harden. J. Perkins. J. Biol. Chem. 254, 38 (1979). S. Heisler. T. D. Reisine. J. Axelrod, Biochem. Biophys. Res. Commun. 111, 112 (1983); T. D. Reisine and S. Heisler. J. Pharmacol. E.rp. Ther. 227. 107 (1983). T. D. Reisine and J. Axelrod. Endocrinology 113, 81 I (1983). J. Delfs and M. Dichter, J. Neurosci. 3. 1176 (1983). M. Smith and W. Vale, Endocrinology 106, (suppl.). 261A (1980). C. Rivier. J. Rivier, W. Vale, Science 218. 377 (1982): C. Rivier. M. J. Brownstein. J. Soiess. I. Rivie;, W. Vale,`Endocrino/ogy 110, 27i (198'2); C. Riwer and W. Vale, ibid. 113. 939 (1983). W. Knepel, K. Benner, G. Hertting. Eur. J. Pharmacol. 81. 645 (1982). 1. Vermes, F. Berkenbosch, F. Tilder. P. Sme- lik, Neurosci. LetI. 27, 89 (1981). H. Weil-Malherbe. 1. Axelrod, R. Tomchick. Science 129. 1226 (1959): H. Weil-Malherbe, L. G. Whitby. J. Axelrod. J. Neurochem. 8. 55 (1961). E. Mezey. T. D. Reisine, M. Palkovits. M. J. Brownstein. J. Axelrod. Proc. Narl. Acod. Sci. U.S.A. 80. 6728 (1983). C. Fortier, J. C/in. Endocrinol. 11, 751 11951). W. McDermott, E. Fry, J. Brobeck. C. Long. Yule J. Eiol. Med. 23, 52 (1950). F. Svec and M. Rudis, J. Bio/. Chem. 256, 5984 (1981). B. Carroll. Br. J. Psychiatry 140. 292 (1982). V. Y. H. Hook. S. Heisler, S. Sabol, J. Axelrod, Biochem. Biophys. Res. Commun. 106. 1361 (1982). T. D. Reisine and J. Takahashi, J. Neurosci., in press; S. Heisler and T. D. Reisine, J. Neuro- them., in press. them., in press. U. Richardson, Endocrinology 113, 62 (1983). U. Richardson, Endocrinology 113, 62 (1983). T. D. Reisine. J. Pharmacoi. Exp. Thu., in T. D. Reisine. J. Pharmacoi. Exp. Thu., in press; T. D. Reisine, unpublished results. press; T. D. Reisine, unpublished results. J. Tyrrell, M. Lorenzi, J. Gehrich, P. Forshon, J. Tyrrell, M. Lorenzi, J. Gehrich, P. Forshon, J. Clin. Endocrinol. Metab. 40. I I25 (1975). J. Clin. Endocrinol. Metab. 40. I I25 (1975). We thank S. Sabol for providing the AtT-20 cells We thank S. Sabol for providing the AtT-20 cells and V. Hook, S. Heisler. and E. Mezey for and V. Hook, S. Heisler. and E. Mezey for collaboration in these studies. We also thank M. collaboration in these studies. We also thank M. Brownstein, M. Dratman, and M. Zatr for help- Brownstein, M. Dratman, and M. Zatr for help- ful suggestions. ful suggestions. 4 MAY 1984 459