European Journal of Endocrinology (2002) 147 815–824 ISSN 0804-4643

EXPERIMENTAL STUDY Serotonergic involvement in stress-induced and oxytocin secretion Henrik Jørgensen1, Ulrich Knigge1,2, Andreas Kjær1,3 and Jørgen Warberg1 1 Department of Medical Physiology, The Panum Institute, 2 Department of Surgery C and 3 Department of Clinical Physiology and Nuclear Medicine, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark (Correspondence should be addressed to Henrik Jørgensen, Department of Medical Physiology, The Panum Institute 12.3.21, University of Copenhagen, Blegdamsvej 3, DK-2200N, Denmark; Email: hsj@mfi.ku.dk)

Abstract Objective: To investigate the involvement of serotonin (5-hydroxytryptamine – 5-HT) receptors in mediation of stress-induced arginine vasopressin (AVP) and oxytocin (OT) secretion in male rats. Design: Experiments on laboratory rats with control groups. Methods: Different stress paradigms were applied after pretreatment with intracerebroventricular infusion of saline or different 5-HT antagonists. Results: Restraint stress (5 min), hypotensive hemorrhage or dehydration for 24 h increased AVP secretion fivefold and OT secretion threefold. Swim stress for 3 min had no effect on AVP secretion, but increased OT secretion threefold. Ether vapor or hypoglycemia had no effect on AVP or OT secretion. The restraint stress-induced AVP response was inhibited by pretreatment with the 5-HT2Aþ2C antagonists (KET) and LY-53857 (LY) and the 5-HT3þ4 antagonist ICS- 205930 (ICS), whereas the 5-HT1A antagonist WAY-100635 (WAY) had no effect. The OT response to restraint stress was inhibited by WAY, KET and LY but not by ICS. KET and LY inhibited OT response to dehydration, and LY inhibited OT response to hemorrhage. Neither of the antagonists affected AVP responses to dehydration or hemorrhage, nor the swim stress-induced OT response. Conclusion: 5-HT2A, 5-HT2C and possibly 5-HT3 and 5-HT4 receptors, but not 5-HT1A receptors, are involved in the restraint stress-induced AVP secretion. 5-HT does not seem to be involved in the dehy- dration- or hemorrhage-induced AVP response. The restraint stress-induced OT response seems to be mediated via 5-HT1A, 5-HT2A and 5-HT2C receptors. The dehydration and hemorrhage-induced OT responses are at least mediated by the 5-HT2A and 5-HT2C receptors. The 5-HT3 and 5-HT4 receptors are not involved in stress-induced OT secretion.

European Journal of Endocrinology 147 815–824

Introduction hypoglycemia or exercise increase AVP secretion (9, 10), whereas predominant psychological Arginine-vasopressin (AVP) and oxytocin (OT) are such as restraint stress, ether-vapor stress or forced synthesized in the magnocellular neurons of the hypo- swimming have minor or no effect (6, 11–13). How- thalamic supraoptic nucleus (SON) and paraventricular ever, recent studies indicate that this is not necessarily nucleus (PVN). Following axonal transport, AVP and true, since repeated was found to OT are released from nerve terminals in the neurohypo- increase the concentration of AVP in the PVN (14) or physis upon stimulation. Various types of psychological the median eminence (15) or gene expression of AVP or physical stressors have been shown to stimulate the in the PVN (16). Together, these findings indicate that hypothalamo–neurohypophysial system and to both types of stress activate the hypothalamo–neuro- increase the secretion of AVP and OT (1–3). However, hypophysial system, which may not necessarily be the relative release of AVP and OT depends on the reflected in elevated peripheral levels (17, specific applied (4–6). Two decades ago it 18). was considered that OT but not AVP was a stress hor- Several seem to be involved in the mone in the rat (7, 8). This has since been modified, mediation of the stress-induced release of AVP and OT, but controversies about the AVP response to different since administration of acetylcholinergic, stressors still exist. It has been proposed that primary and aminergic receptor antagonists and administration physical stressors such as hypovolemia, hemorrhage, of GABA and opiate receptor agonists inhibit the

q 2002 Society of the European Journal of Endocrinology Online version via http://www.eje.org

Downloaded from Bioscientifica.com at 10/02/2021 01:07:08AM via free access 816 H Jørgensen and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2002) 147 neurohypophysial hormone response to various stres- reduce or eliminate all unnecessary stress or pain to sors (19–23). In addition, serotonin (5-hydroxytrypta- the animals. mine – 5-HT) seems to be involved in the mediation of the basal as well as the stress-induced secretion of the Drugs neurohypophysial (24, 25). The effect of 5-HT on the secretion of the neurohypophysial hor- The following 5-HT receptor antagonists were used: mones is exerted via centrally located receptors, of the 5-HT1A receptor antagonist WAY-100635, N-tert- which primarily 5-HT2C receptors are known to be butyl-3-(4-(2-methoxyphenyl) piperazine-1-yl)-2-phenyl- involved in the release of AVP, and 5-HT1A and 5-HT2 propionamide (WAY), gift from Lundbeck Inc., Valby, receptors in OT release (26–28). Recently, we reported Denmark; the 5-HT2A(þ2C) receptor antagonist that the 5-HT4 receptor might also be involved in the ketanserin, 3-[2-[4-(4-fluorobenzoyl)-1-piperidinyl]ethyl]- 5-HT-induced secretion of AVP and OT, whereas the 2,4(1H,3H)-quinazolinedione tartrate (KET), RBI, 5-HT3 receptor seems to be of minor importance (29). Natick, MA, USA; the 5-HT2C(þ2A) receptor antagonist We have previously found that 5-HT1, 5-HT2 and LY-53857, 6-methyl-1-(-methyl ethyl)-ergoline-8b- 5-HT3 receptors are involved in stress-induced release carboxylic acid 2-hydroxy-1-methyl propyl ester mal- of prolactin (30) and that 5-HT1, 5-HT2 and possibly eate (LY), RBI; and the 5-HT3þ4 receptor antagonist the 5-HT4 receptors seem to be involved in the stress- ICS-205930, endo-8-methyl-8-axabiocylol [3.2.1]oct- induced adrenocorticotropin (ACTH) secretion (31). 3-ol indol-3-yl-carboxylate hydrochloride (ICS; tropise- Therefore, the aim of the present study was to inves- trone), RBI. The receptor affinities of the antagonists tigate (1) which types of stress affect AVP and OT are given in Table 1. secretion, (2) whether the serotonergic system partici- Insulin was purchased from Actrapid, Novo Nordisk, pates in the mediation of the release of the neurohypo- Bagsværd Denmark. All drugs were dissolved in saline physial hormones in response to different stress and adjusted to pH 6.8–7.5, except for KET, which paradigms, and (3) which 5-HT receptors are involved was dissolved in 0.9% saline acidified with 0.05% HCl in this response. and adjusted to pH 6.8 with 0.05 mol/l NaOH.

Experimental procedures Materials and methods General procedures For intracerebroventricular (i.c.v.) Animal procedures administration of 5-HT antagonists, a permanent stain- less steel cannula was implanted in the right lateral Male Wistar rats (250–325 g), bred at the Panum cerebral ventricle through a drilled hole in the skull. Institute, were used in all experiments. The rats were The coordinates were: P ¼ 0:0; L ¼ 1:5; H ¼ 24:0 housed under controlled temperature ð22^1 8CÞ and relative to bregma. The cannula was closed with a humidity (80%), in cages of four animals before and platinum obturator. Compounds were infused intracer- one per cage after surgery. A cycle of 12 h light: 12 h ebroventricularly over a period of 2.5 min in a volume darkness with lights on at 0600 was used. The rats of 5 ml. All operations were performed during pentobar- had free access to laboratory chow and tap water, bital anesthesia (66 mg/kg i.p.) 4–5 days prior to the when not dehydrated. The experimental protocol was experiments. On the day of the experiments the animals in accordance with, and accepted by, the European were brought to the laboratory and allowed to de-stress Communities Council directive of 24 November 1986 in their cage 3 h prior to injections. After 90 min i.v. and the Danish Ministry of Justice, Board for Animal catheters and i.c.v. cannulas were extended by poly- Experimental Research. The animal experiments endea- propylene plastic tubes permitting infusion of test vored to reduce the overall number of animals and to substances without disturbing the rats. The antagonists

Table 1 The 5-HT receptor antagonists used with the respective receptor affinities; values in pKA or pKi. Shaded areas indicate the primary type of receptor specificity. A pKi above 7.0 is considered to indicate specificity for that receptor. Numbers in parentheses are reference numbers.

5-HT1A 5-HT1B 5-HT2A 5-HT2C 5-HT3 5-HT4 5-HT5 5-HT7 WAY (64) 5-HT1A 8.9 ,7.0 ,7 ,7 ,7–– KET 5-HT2A+2C 5.9 (65) 5.7 (65) 8.7 (66) 7.2 (66) ,4 (67) – 4.8 6.7 LY 5-HT2A+2C 6.4 (68) 5.5 (68) 7.7 (66) 8.3 (66) – – – – ICS 5-HT3+4 5.3 (68) 4.6 (68) 8.5 (67) 6.2 (67)

www.eje.org

Downloaded from Bioscientifica.com at 10/02/2021 01:07:08AM via free access EUROPEAN JOURNAL OF ENDOCRINOLOGY (2002) 147 5-HT and stress-induced AVP 817 were infused intracerebroventricularly at 215 min, unless anesthesia for the i.c.v. cannula. On the morning otherwise stated. The time for administration and the of intervention the pumps were filled with either doses of antagonists were chosen on the basis of pre- saline or one of the following 5-HT antagonists: KET vious experiments, with the antagonists administered (5-HT2Aþ2C), LY (5-HT2Cþ2A) or ICS (5-HT3þ4), and both intraperitoneally and intracerebroventricularly connected to the i.c.v. cannula. Euhydrated rats (30, 32, 33). served as the control. At the end of the 24-h period, In all experiments, blood samples were obtained by blood samples were obtained as described above. decapitation of rats at 0 min, which was scheduled There were eight rats in each treatment group. between 1000 and 1200 h. Blood samples were col- lected in chilled polyethylene tubes containing 50 ml aprotinin (Trasylol, 20 000 kIU/ml; Bayer, Leverkusen, Experiment 4. Effect of 5-HT antagonists on Germany) and 100 ml 0.5 mol/l EDTA and were kept hemorrhage-induced AVP or OT secretion Four to on ice until centrifuged at 4 8C. The plasma was sepa- five days prior to the experiment a silicone catheter rated into polyethylene tubes and stored at 220 8C was implanted in the right jugular vein as previously until hormone analysis could be carried out. described (33), and an i.c.v. cannula was implanted as described above. Rats were pretreated intracerebro- Experiment 1. Effect of stress on AVP or OT ventricularly with saline, LY (50 nmol) or ICS secretion Rats were exposed to one of the following (10 nmol) at 220 min. At 210 min, 3.0 ml of blood seven different stressors. Restraint stress was performed was drawn from the vein catheter over a period of by fixing the animal on its back in a supine position for 2 min. The rats (six in each group) were decapitated 5 min. Cold-swim stress was performed by allowing the at 0 min. rat to swim in cold (4 8C), deep water in a plastic bowl for 3 min followed by a 2-min period to allow it to shake Experiment 5. Effect of 5-HT antagonists on AVP or the skin dry. Dehydration was performed for 24 h start- OT secretion induced by cold-swim stress Groups of ing at noon. The length of dehydration chosen has pre- seven or eight rats were exposed to cold-swim stress for viously been shown to cause significant AVP release 3 min from 25to22 min and decapitated at 0 min, (20) and did not lead to any visible discomfort for the after pretreatment (intracerebroventricularly at animals. The Danish Government Animal Research 215 min) with saline or the following 5-HT antagonist: Control Committee, Ministry of Justice, has approved the 5-HT1A antagonist WAY (10 nmol); the 5-HT2Aþ2C dehydration for up to 48 h as ethically acceptable. antagonist KET (50 nmol); the 5-HT2Cþ2A antagonist Hemorrhage stress was performed by withdrawal of LY (50 nmol); or the 5-HT3þ4 antagonist ICS 3.0 ml venous blood (approximately 20% of the blood (10 nmol). volume) from 210 to 28 min. Ether stress was per- formed by placing the rat in an ether-vapor filled glass beaker for 5 min from 25 to 0 min. Hypoglycemia Hormone analysis stress was achieved by i.p. injection of 3 IU/kg rapid- working insulin preparation at 245 min. The rats AVP and OT were analyzed by RIA in extracted plasma were decapitated at 0 min. Each experiment consisted by means of C18 Sep-Pak cartridges as previously of eight rats. described (20). The least detectable quantity for AVP and OT was 0.1–0.3 pmol/l plasma and 4–6 pmol/l Experiment 2. Effect of 5-HT antagonists on respectively. Intra- and interassay coefficients were 10 restraint stress-induced AVP or OT secretion Rats and 15% respectively for both assays. were exposed to restraint stress from 25to0min after pretreatment with either saline or one of the fol- lowing 5-HT antagonists (i.c.v. infusion at 215 min): Osmolality analysis the 5-HT1A antagonist WAY (10 nmol); the 5- Measurement of plasma osmolality was performed on HT2Aþ2C antagonist KET (50 nmol); the 5-HT2Cþ2A plasma samples in triplicate by freezing-point antagonist LY (50 nmol); or the 5-HT3þ4 antagonist depression (Advanced Microosmometer 3MO, ICS (10 nmol). The seven rats in each group were Advanced Instruments Inc., Needham Heights, MA, decapitated at 0 min. USA) by a previously described method (34). Experiment 3. Effect of 5-HT antagonists on dehydration-induced AVP or OT secretion Rats Statistical tests were exposed to 24 h of dehydration. One group of rats was continuously infused intracerebroventricularly Results are presented as the mean ^ S.E.M. and evalu- with 5-HT antagonists via a microosmotic pump ated by ANOVA followed by Newman–Keul’s test for (Alzet, Cupertino, CA, USA; infusion rate 2 ml/h equal- multiple comparisons when appropriate. The level of izing 3 nmol/h), placed intraperitoneally during the significance was set at P , 0:05:

www.eje.org

Downloaded from Bioscientifica.com at 10/02/2021 01:07:08AM via free access 818 H Jørgensen and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2002) 147

Results the 5-HT1A antagonist WAY had no effect (Fig. 2a). WAY inhibited the OT response to restraint stress Experiment 1. Effect of stress on AVP or OT almost 80% (P , 0:01; Fig. 2b). KET and LY reduced secretion the OT response to restraint stress about 60% Five minutes of restraint stress, 24 h of dehydration (P , 0:01; Fig. 2b), whereas ICS had no significant and 3.0 ml hemorrhage increased plasma AVP level effect. six-, five- and fourfold respectively (P , 0:01; Fig. 1a), whereas swim, ether, endotoxin or hypoglycemia stress had no effect (Fig. 1a). Restraint, cold-swim, dehydration and hemorrhage stress increased plasma OT level 3.5-, 2-, 3- and Experiment 3. Effect of 5-HT antagonists on 2-fold respectively (P , 0:01; P , 0:05; P , 0:01 dehydration-induced AVP or OT secretion and P , 0:05 respectively, Fig. 1b), whereas ether or Twenty-four hours of dehydration increased plasma hypoglycemia stress had no effect (Fig. 1b). osmolality 4% from 288^1:2 mosmol=l to 300^ 1:2 mosmol=l ðP , 0:05Þ; AVP secretion twofold ðP , Þ ð Þ Experiment 2. Effect of 5-HT antagonists on 0:05 ; and OT secretion almost fourfold P , 0:05 restraint stress-induced AVP or OT secretion (Fig. 3a and 3b). Infusion of 5-HT antagonists had no effect on the dehydration induced AVP response The restraint stress-induced AVP response was inhib- (Fig. 3a). However, continuous i.c.v. infusion of the ited almost 70% by pretreatment with the 5-HT2Aþ2C 5-HT2Cþ2A antagonist LY or the 5-HT2A antagonist antagonist KET, the 5-HT2Cþ2A antagonist LY and the KET inhibited the OT response 75% (P , 0:01; Fig. 5-HT3þ4 antagonist ICS (P , 0:01; Fig. 2a), whereas 3b), whereas the 5-HT3þ4 antagonist ICS had no effect.

Figure 1 Effect of different stressors on plasma level of AVP (a) or OT (b). Restraint stress (RS; 5 min), dehydration (Dehy; 24 h), hypotensive hemorrhage (Hemo; 3.0 ml at 210 min), cold swim (Swim; 3 min in 4 8C cold water), ether stress (Ether; 5 min) or hypoglycemia (Hypo; i.p. 3 IU insulin at 245 min). Rats were decapitated at 0 min. Data represent means ^ S.E.M. of eight rats. ## P , 0:01 and # P , 0:05 compared with control rats.

www.eje.org

Downloaded from Bioscientifica.com at 10/02/2021 01:07:08AM via free access EUROPEAN JOURNAL OF ENDOCRINOLOGY (2002) 147 5-HT and stress-induced AVP 819

Figure 2 Effect of i.c.v. pretreatment at 215 min with saline, the 5-HT1A antagonist WAY (10 nmol), the 5-HT2Aþ2C antagonists KET or LY (50 nmol), or the 5-HT3þ4 antagon- ist ICS (10 nmol) on restraint stress-induced AVP (a) and OT (b) response. Rats were decapitated at 0 min. Data represent means ^ S.E.M. of seven or eight rats; ## P , 0:01 compared with control; *P , 0:05 and **P , 0:01 compared with saline-treated restraint stressed rats.

Experiment 4. Effect of 5-HT antagonists on Discussion hemorrhage-induced AVP or OT secretion Withdrawal of 3.0 ml of blood from the jugular vein The present experiments show that physical as well as increased AVP and OT secretion seven- and twofold psychological stress paradigms stimulate the secretion respectively (P , 0:05; Fig. 4a and P , 0:01; Fig. 4b, of the neurohypophysial hormones AVP and OT, since respectively). Pretreatment with the 5-HT antagonist had no effect on hemorrhage-induced AVP secretion. OT secretion was inhibited 50% by LY, while ICS had Table 2 Effect of i.c.v. infusion of different 5-HT antagonists on the AVP and OT response to 3 min of cold-swim stress from 25 no effect (P , 0:05; Fig. 4b). to 22 min, and decapitation at 0 min. Values are in pmol/l expressed as means of seven or eight rats ^ S.E.M. *P , 0:05 versus control.

Experiment 5. Effect of 5-HT antagonists on Stress Antagonist AVP OT AVP or OT secretion induced by cold-swim stress Control Saline 1.9^0.2 11^1.0 Swim Saline 1.5^0.2 19^2.1* Three minutes of cold-swim stress had no effect on AVP Swim WAY (5-HT1A)1.5^0.1 23^2.3* ^ ^ secretion, but increased OT secretion twofold (Table 2, Swim KET (5-HT2A)1.90.2 21 1.7* Swim LY (5-HT2C+2A)1.6^0.1 26^1.5* Fig. 1). None of the 5-HT antagonists affected this Swim ICS (5-HT )2.4^0.2 25^3* response (Table 2). 3+4

www.eje.org

Downloaded from Bioscientifica.com at 10/02/2021 01:07:08AM via free access 820 H Jørgensen and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2002) 147

Figure 3 Effect of continuous i.c.v. infusion (0.3 nmol/h) of saline, KET or LY (5-HT2A+2C) or ICS (5-HT3+4) during a 24-h dehydration period on the AVP (a) and OT (b) response. Rats were decapitated at the end of the dehydration period (0 h). Data represent means ^S.E.M. of eight rats; ## P , 0:01 compared with control; **P , 0:01 compared with saline-treated dehydrated rats. dehydration and hemorrhage (physical stressors), the neurohypophysial hormones illustrates the flexi- as well as restraint stress (psychological stressor) bility of the hypothalamo–neurohypophysial system, increased the secretion of both hormones. However, which is suitable to achieve a maximal overall positive some of the applied stress paradigms, such as ether effect on the target organs under different physiological and hypoglycemia, did not affect the secretion of conditions, such as bleeding, labor, hypovolemia, etc. either of the two hormones. The stress response is (5). A heterogeneous hormone response in respect to probably an interaction between the neuroendocrine AVP and OT has also been found after immobilization, system, the sympathetic nervous system and the forced swimming, ether stress, hemorrhage, endotoxin target organs, resulting in activation of different neur- and osmotic loading (4–6, 36). onal pathways and release of specific hormones (35). The present experiments show that involvement of Stress in itself releases various neurotransmitters, 5-HT and the effect of 5-HT receptor blockade on the which may have either stimulating (catecholamines, AVP or OT responses to different stress paradigms 5-HT, ) or inhibiting (GABA) effect on are not general, but seem to be as important as the hypothalamic and pituitary hormone release (19). involvement of other neurotransmitters, such as nor- The secretion of the two neurohypophysial hormones epinephrine and histamine (20, 37, 38). The present seem to be differentially affected by stress, since cold- finding of an increased level of plasma AVP after swim stress, which is considered to be combined physi- 5min of restraint stress is in accordance with two cal and psychological stressor, stimulated OT secretion, studies (5, 7), but in contrast to the a previous finding, but not AVP secretion. The independent regulation of probably due to a short period of stress (1 min) in that

www.eje.org

Downloaded from Bioscientifica.com at 10/02/2021 01:07:08AM via free access EUROPEAN JOURNAL OF ENDOCRINOLOGY (2002) 147 5-HT and stress-induced AVP 821

Figure 4 Effect of i.c.v. pretreatment at 220 min with saline, LY (5-HT2Aþ2C;50nmol) or ICS (5-HT3þ4; 10 nmol) on the hemorrhage (3.0 ml venous blood at 210 min) induced AVP (a) and OT (b) response. Rats were decapitated at 0 min. Data represent means ^ S.E.M. of six rats; # P , 0:05 and ##P , 0:01 compared with control; *P , 0:05 compared with saline-treated hemorrhage-stressed rats. study (4). The stimulating effect of restraint stress on whereas a 3–5-min period of manual restraint, as plasma OT is in agreement with our own and others used in this experiment, is sufficient to increase previous findings (4, 39, 40). Therefore, it seems that peripheral plasma levels of AVP (5, 7). both AVP and OT are stimulated by restraint stress. A Serotonergic antagonists with affinity for 5-HT2, single prolonged period of restraint stress (30– 5-HT3 and 5-HT4 receptors inhibit the AVP response 150 min) has been found to increase AVP mRNA in to restraint stress, whereas 5-HT1A and 5-HT2 antagon- the medial parvocellular part but not in the magno- ists reduce the OT response to this stressor. There are no cellular part of the PVN (41, 42). However, an available comparable data, but in unstressed rats, the increased synthesis of AVP in the parvocellular PVN AVP response to serotonergic stimulation is mediated does not inevitably induce an increased plasma level via 5-HT2A and 5-HT2C receptors (26, 28, 44, 45), of AVP in the peripheral circulation, since AVP pro- while 5-HT1A and 5-HT2 receptors are involved in the duced in the parvocellular part of the PVN is released 5-HT-induced OT response via the PVN (46–49). primarily into the pituitary portal circulation. The We found that dehydration is a potent physiological pattern emerging from the previous and the present stimulus for both AVP and OT secretion, which con- findings indicate that single ultra short-term restraint firms previous findings (9, 50). An involvement of the stress (1–2 min) performed by mechanical fixation or serotonergic system in AVP response to dehydration is restraint in a plastic glass box, does not seem to be possible but does not seem likely, since none of the sufficient to increase either AVP mRNA in the hypo- antagonists (5-HT2A, 5-HT2C, 5-HT3þ4) used in the pre- thalamic PVN or AVP in peripheral plasma (36, 43), sent experiments had any effect on this response. An

www.eje.org

Downloaded from Bioscientifica.com at 10/02/2021 01:07:08AM via free access 822 H Jørgensen and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2002) 147 involvement of other receptors cannot be excluded. The Although relatively few in the PVN, they are more possibility that the receptor blockade was insufficient is abundant in the parvocellular than in the magno- limited, since we used continuous i.c.v. infusion of com- cellular part and are primarily projecting to OT pounds via an osmotic pump. In addition, using an neurons, whereas they are very few and scattered at identical experimental design, we previously found AVP neurons in the PVN (60, 61). The few 5-HT that the AVP response to dehydration was inhibited fibers in the SON are found to be in close contact by histamine receptor antagonists (20). Furthermore, with AVP neurons, but sparse at the OT neurons (62). an increase in AVP secretion, induced by osmotic chal- The involvement of different serotonergic receptors lenge with 500 mmol NaCl/l, was not affected by either in the mediation of the stress-induced AVP and OT the 5-HT2 antagonist ritanserin or the 5-HT2Aþ2C responses is found to be differentiated for, for example, agonist DOI (51). blockade of 5-HT1A receptors inhibited restraint stress- Hypotensive hemorrhage-induced by acute depletion induced OT, but not AVP secretion. This finding is in of approximately 20% of the total blood volume accordance with previous findings of a different potently increased AVP and OT in peripheral plasma involvement of 5-HT receptors in AVP and OT secretion in our study. This is in accordance with previous find- (28, 63). The reason for this dissimilarity may be ings, which showed that a 15–30% acute reduction absence of 5-HT1A receptors on AVP neurons. The pre- in blood volume increased AVP (9, 52–54), and OT sent study has not investigated the specific localization concentration in peripheral (5, 50, 55) and pituitary of the interaction between the 5-HT system and the portal plasma (56), and increased AVP content and neurohypophysial system. Further studies, for example, mRNA in the hypothalamus (52, 57, 58). The OT with specific anatomical lesions, are required to response to hemorrhage seems to be mediated via elucidate this. 5-HT2 receptors, since the response was inhibited by In conclusion, we found that various stress para- pretreatment with the 5-HT2Cþ2A antagonist LY. The digms affect AVP and OT secretion differently and AVP response also tended to be inhibited, but this was that the serotonergic system is involved differently not significant. in these responses. The restraint stress-induced AVP Cold-swim stress stimulated OT but not AVP and OT responses are at least mediated through secretion, which is in agreement with previous obser- 5-HT2Aþ2C and 5-HT4 receptors, and via 5-HT1A and vations (6, 59). In the present experiments, rats were 5-HT2Aþ2C receptors respectively. The dehydration- exposed to deep cold water (4 8C for 2 min). The and hemorrhage-induced OT response is in part increase in OT secretion could be due either to the mediated through the 5-HT2Aþ2C receptors, whereas immersion in cold water or the swimming exercise, or the AVP responses to these stressors do not seem to both. Exposure to cold environment (4 8C for 30 min) involve the investigated 5-HT receptors. Swim stress has previously been found to increase OT levels (5), increased only OT secretion, but the investigated 5-HT and single or repeated forced swim stress (20 8Cfor receptors are not involved in this response. Ether and 10 min) increased both AVP and OT in microdialysates hypoglycemia stress had no effect on AVP or OT from both the SON and the PVN, but only OT in periph- secretion. eral plasma (59). The OT response induced by cold- swim stress does not seem to involve the investigated 5-HT receptors. Acknowledgements Five minutes of ether stress did not affect AVP or OT secretion, which conflicts with previous findings both This study was supported by grants from the Danish of ourselves and of others (4, 5, 36, 39). A reason for Medical Research Council; NOVO’s Foundation; The the missing response to ether stress may be the stress- Lundbeck Foundation; Nordic Insulin Foundation Com- time schedule, since it was found that 1 min of ether mittee; The Danish Hospital Foundation for Medical stress increased AVP or OT secretion twofold, measured Research, Regions of Copenhagen, The Faroe Islands 1 min after the end of the stress period, whereas the and Greenland; and the Danish Medical Association hormone response disappeared 4 min after the stress Research Fund. We thank Elsa Larsen and Jytte period was discontinued (4). In accordance with most Oxbøll for skilled technical assistance. The materials of the previous studies (5, 39), we found no effect on for RIA of AVP and OT were kindly provided by the insulin-induced hypoglycemia stress on AVP or OT National Hormone and Pituitary Program of the secretion. NIDDK. The present findings demonstrate that 5-HT is differ- entially involved in the stress-induced secretion of AVP and OT. One explanation might be that various stress paradigms activate cortical, limbic and hypothalamic References areas differently, and these areas in addition have a het- 1 Udelsman R & Chrousos GP. Hormonal responses to surgical erogeneous distribution of 5-HT neurons. 5-HT nerve stress. Advances in Experimental Medical Biology 1988 245 fibers are localized differently within the hypothalamus. 265–272.

www.eje.org

Downloaded from Bioscientifica.com at 10/02/2021 01:07:08AM via free access EUROPEAN JOURNAL OF ENDOCRINOLOGY (2002) 147 5-HT and stress-induced AVP 823

2 Van de Kar LD, Richardson Morton KD & Rittenhouse PA. Stress: vasopressin and oxytocin. Progress in Neurobiology 1991 36 neuroendocrine and pharmacological mechanisms. Methods and 131–169. Achievements in Experimental Pathology 1991 14 133–173. 23 Reichlin S. Neuroendocrinology. In Textbook of Endocrinology,pp 3 Chrousos GP & Gold PW. The concepts of stress and stress system 165–248. Eds JD Wilson, DW Foster, HM Kronenberg & PR disorders. Overview of physical and behavioral homeostasis. Larsen. Philadelphia: W B Saunders, 1998. JAMA 1992 267 1244–1252. 24 Iovino M & Steardo L. Effect of substances influencing brain sero- 4 Hashimoto K, Murakami K, Takao T, Makino S, Sugawara M & tonergic transmission on plasma vasopressin levels in the rat. Ota Z. Effect of acute ether or restraint stress on plasma cortico- European Journal of Pharmacology 1985 113 99–103. tropin-releasing hormone, vasopressin and oxytocin levels in 25 Van de Kar LD. Neuroendocrine pharmacology of serotonergic the rat. Acta Medicina Okayama 1989 43 161–167. neurons. Annual Review of Pharmacology and Toxicology 1991 31 5 Kasting NW. Simultaneous and independent release of vasopres- 289–320. sin and oxytocin in the rat. Canadian Journal of Physiology and 26 Li Q, Brownfield MS, Battaglia G, Cabrera TM, Levy AD, Ritten- Pharmacology 1988 66 22–26. house PA et al. Long-term treatment with the 6 Lang RE, Heil JW, Ganten D, Hermann K, Unger T & Rascher W. fluoxetine and potentiates endocrine responses Oxytocin unlike vasopressin is a stress hormone in the rat. to the serotonin agonists 6-chloro-2-[1-piperazinyl]-pyrazine Neuroendocrinology 1983 37 314–316. (MK-212) and (þ2)-1-(2,5-dimethoxy-4-iodophenyl)-2-amino- 7 Husain MK, Manger WM, Rock TW, Weiss RJ & Frantz AG. Vaso- propane HCl (DOI). Journal of Pharmacology and Experimental pressin release due to manual restraint in the rat: role of body Therapeutics 1993 266 836–844. compression and comparison with other stressful stimuli. 27 Uvnas MK, Hillegaart V, Alster P & Ahlenius S. Effects of 5-HT Endocrinology 1979 104 641–644. agonists, selective for different receptor subtypes, on oxytocin, 8 Landgraf R, Malkinson TJ, Veale WL, Lederis K & Pittman QJ. CCK, gastrin and somatostatin plasma levels in the rat. Neuro- Vasopressin and oxytocin in rat brain in response to prostaglan- pharmacology 1996 35 1635–1640. din fever. American Journal of Physiology 1990 259 28 Bagdy G, Sved AF, Murphy DL & Szemeredi K. Pharmacological R1056–R1062. characterization of serotonin receptor subtypes involved in vaso- 9 Fyhrquist F, Tikkanen I & Linkola J. Plasma vasopressin concen- pressin and plasma renin activity responses to serotonin agonists. tration and renin in the rat: effect of hydration and hemorrhage. European Journal of Pharmacology 1992 210 285–289. Acta Physiologica Scandinavia 1981 113 507–510. 29 Jorgensen H, Knigge U, Kjaer A & Warberg J. Stress-induced 10 Baylis PH & Robertson GL. Rat vasopressin response to insulin- neurohypophysial hormone secretion is partly mediated by induced hypoglycemia. Endocrinology 1980 107 1975–1979. serotonergic neurons. In International Union of Pharmacology IV, 11 Yagi K & Onaka T. Suppressive vasopressin response to emotional Fourth IUPHAR sattelite meeting on serotonin, Rotterdam, pp stress: the neuroactive substance that may be involved. Annals of 112, 1998. the New York Academy of Sciences 1993 689 685–688. 30 Jorgensen H, Knigge U & Warberg J. Effect of selective serotonin 12 Keil LC & Severs WB. Reduction in plasma vasopressin levels of receptor agonists on prolactin secretion in male rats. Neuroendo- dehydrated rats following acute stress. Endocrinology 1977 100 crinology 1993 57 401–407. 30–38. 31 Jorgensen H, Knigge U, Kjaer A, Vadsholt T & Warberg J. Seroto- 13 Gibbs DM. Vasopressin and oxytocin: hypothalamic modulators of nergic involvement in stress-induced ACTH release. Brain Research the stress response: a review. Psychoneuroendocrinology 1986 11 1998 811 10–20. 131–139. 32 Jorgensen H, Knigge U & Warberg J. Effect of serotonin 5-HT1, 5- 14 Wotjak CT, Kubota M, Liebsch G, Montkowski A, Holsboer F, HT2, and 5-HT3 receptor antagonists on the prolactin response to Neumann I et al. Release of vasopressin within the rat paraven- restraint and ether stress. Neuroendocrinology 1992 56 371–377. tricular nucleus in response to emotional stress: a novel mechan- 33 Jorgensen H, Knigge U, Kjaer A & Warberg J. Adrenocorticotropic ism of regulating adrenocorticotropic hormone secretion? Journal hormone secretion in rats induced by stimulation with serotoner- of Neuroscience 1996 16 7725–7732. gic compounds. Journal of Neuroendocrinology 1999 11 283–290. 15 de Goeij DC, Dijkstra H & Tilders FJ. Chronic psychosocial stress 34 Kjœr A, Knigge U, Jørgensen H & Warberg J. Dehydration-induced enhances vasopressin, but not corticotropin-releasing factor, in vasopressin secretion in humans: involvement of the histaminer- the external zone of the median eminence of male rats: relation- gic system. American Journal of Physiology and Endocrinology and ship to subordinate status. Endocrinology 1992 131 847–853. Metabolism 2000 279 E1305–E1310. 16 Ma XM, Lightman SL & Aguilera G. Vasopressin and cortico- 35 Bohus B, Benus RF, Fokkema DS, Koolhass JM, Nyakas C, van tropin-releasing hormone gene responses to novel stress in rats Oortmerssen GA et al. Neuroendocrine states and behavioral adapted to repeated restraint. Endocrinology 1999 140 and physiological stress responses. In Progress in Brain Research, 3623–3632. pp 57–70. Eds ER De Kloet, VM Wiegant & D de Wied. Amster- 17 Engelmann M, Ebner K, Landgraf R, Holsboer F & Wotjak CT. dam: Elsevier, 1987. Emotional stress triggers intrahypothalamic but not peripheral 36 Gibbs DM. Dissociation of oxytocin, vasopressin and corticotropin release of oxytocin in male rats. Journal of Neuroendocrinology secretion during different types of stress. Life Sciences 1984 35 1999 11 867–872. 487–491. 18 Engelmann M, Wotjak CT, Ebner K & Landgraf R. Behavioural 37 Onaka T. Catecholaminergic mechanisms underlying neuro- impact of intraseptally released vasopressin and oxytocin in hypophysial hormone responses to unconditioned or conditioned rats. Experimental Physiology 2000 85 125S–130S. aversive stimuli in rats. Experimental Physiology 2000 85 19 Van de Kar LD & Blair ML. Forebrain pathways mediating stress- 101S–110S. induced hormone secretion. Frontiers of Neuroendocrinology 1999 38 Itoi K, Helmreich DL, Lopez-Figueroa MO & Watson SJ. Differen- 20 1–48. tial regulation of corticotropin-releasing hormone and vasopres- 20 Kjaer A, Knigge U, Rouleau A, Garbarg M & Warberg J. Dehy- sin gene transcription in the hypothalamus by . dration-induced release of vasopressin involves activation of hypo- Journal of Neuroscience 1999 19 5464–5472. thalamic histaminergic neurons. Endocrinology 1994 135 39 Kjaer A, Knigge U & Warberg J. Involvement of oxytocin in 675–681. histamine- and stress-induced ACTH and prolactin secretion. 21 Stratakis CA & Chrousos GP. Neuroendocrinology and pathophy- Neuroendocrinology 1995 61 704–713. siology of the stress system. Annals of the New York Academy of 40 Jezova D, Michajlovskij N, Kvetnansky R & Makara GB. Para- Sciences 1995 771 1–18. ventricular and supraoptic nuclei of the hypothalamus are not 22 Renaud LP & Bourque CW. Neurophysiology and neurophar- equally important for oxytocin release during stress. Neuroendo- macology of hypothalamic magnocellular neurons secreting crinology 1993 57 776–781.

www.eje.org

Downloaded from Bioscientifica.com at 10/02/2021 01:07:08AM via free access 824 H Jørgensen and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2002) 147

41 Aubry JM, Bartanusz V, Jezova D, Belin D & Kiss JZ. Single stress 55 Dunning BE, Verbalis JG & Fawcett CP. Evidence for participation induces long-lasting elevations in vasopressin mRNA levels in of the neurohypophysial hormones in the hyperglucagonemic CRF hypophysiotrophic neurones, but repeated stress is required response to hemorrhage in the rat. Neuroendocrinology 1985 41 to modify AVP immunoreactivity. Journal of Neuroendocrinology 385–389. 1999 11 377–384. 56 Plotsky PM, Bruhn TO & Vale W. Evidence for multifactor regu- 42 Herman JP. In situ hybridization analysis of vasopressin gene tran- lation of the adrenocorticotropin secretory response to hemo- scription in the paraventricular and supraoptic nuclei of the rat: dynamic stimuli. Endocrinology 1985 116 633–639. regulation by stress and glucocorticoids. Journal of Comparative 57 Ota M, Crofton JT & Share L. Hemorrhage-induced vasopressin Neurology 1995 363 15–27. release in the paraventricular nucleus measured by in vivo micro- 43 Makino S, Smith MA & Gold PW. Increased expression of dialysis. Brain Research 1994 658 49–54. corticotropin-releasing hormone and vasopressin messenger 58 Demotes-Mainard J, Chauveau J, Rodriguez F, Vincent JD & ribonucleic acid (mRNA) in the hypothalamic paraventricular Poulain DA. Septal release of vasopressin in response to osmotic, nucleus during repeated stress: association with reduction in hypovolemic and electrical stimulation in rats. Brain Research glucocorticoid receptor mRNA levels. Endocrinology 1995 136 1986 381 314–321. 3299–3309. 59 Wotjak CT, Ganster J, Kohl G, Holsboer F, Landgraf R & 44 Brownfield MS, Greathouse J, Lorens SA, Armstrong J, Urban JH & Engelmann M. Dissociated central and peripheral release of vaso- Van de Kar LD. Neuropharmacological characterization of seroto- pressin, but not oxytocin, in response to repeated swim stress: ninergic stimulation of vasopressin secretion in conscious rats. new insights into the secretory capacities of peptidergic neurons. Neuroendocrinology 1988 47 277–283. Neuroscience 1998 85 1209–1222. 45 Bagdy G, Calogero AE, Szemeredi K, Gomez MT, Murphy DL, 60 Sawchenko PE, Swanson LW, Steinbusch HW & Verhofstad AA. Chrousos GP et al. Beta-endorphin responses to different serotonin The distribution and cells of origin of serotonergic inputs to the agonists: involvement of corticotropin-releasing hormone, vaso- paraventricular and supraoptic nuclei of the rat. Brain Research pressin and direct pituitary action. Brain Research 1990 537 1983 277 355–360. 227–232. 61 Sawchenko PE & Swanson LW. Immunohistochemical identifi- 46 Van de Kar LD, Rittenhouse PA, Li Q, Levy AD & Brownfield MS. cation of neurons in the paraventricular nucleus of the hypo- Hypothalamic paraventricular, but not supraoptic neurons, med- thalamus that project to the medulla or to the spinal cord in iate the serotonergic stimulation of oxytocin secretion. Brain the rat. Journal of Comparative Neurology 1982 205 260–272. Research Bulletin 1995 36 45–50. 62 Ferris CF, Melloni JR, Koppel G, Perry KW, Fuller RW & Delville Y. 47 Van de Kar LD, Javed A, Zhang Y, Serres F, Raap DK & Gray TS. Vasopressin/serotonin interactions in the anterior hypothalamus 5-HT2A receptors stimulate ACTH, corticosterone, oxytocin, control aggressive behavior in golden hamsters. Journal of renin, and prolactin release and activate hypothalamic CRF and Neuroscience 1997 17 4331–4340. oxytocin-expressing cells. Journal of Neuroscience 2001 21 63 Li Q, Levy AD, Cabrera TM, Brownfield MS, Battaglia G & Van de 3572–3579. Kar LD. Long-term fluoxetine, but not desipramine, inhibits the 48 Bagdy G. Role of the hypothalamic paraventricular nucleus in ACTH and oxytocin responses to the 5-HT1A agonist, 8-OH- 5-HT1A, 5-HT2A and 5-HT2C receptor-mediated oxytocin, pro- DPAT, in male rats. Brain Research 1993 630 148–156. lactin and ACTH/corticosterone responses. Behavioural Brain 64 Fletcher A, Forster EA, Bill DJ, Brown G, Cliffe IA, Hartley JE et al. Research 1996 73 277–280. Electrophysiological, biochemical, neurohormonal and beha- 49 Li Q, Brownfield MS, Levy AD, Battaglia G, Cabrera TM & Van de vioural studies with WAY-100635, a potent, selective and silent Kar LD. Attenuation of hormone responses to the 5-HT1A agonist 5-HT1A receptor antagonist. Behavioural Brain Research 1996 ipsapirone by long-term treatment with fluoxetine, but not desi- 73 337–353. pramine, in male rats. Biological Psychiatry 1994 36 300–308. 65 Hoyer D. 5-Hydroxytryptamine receptors and effector coupling 50 Summy-Long JY, Miller DS, Rosella-Dampman LM, Hartman RD & mechanisms in peripheral tissues. In Peripheral Actions of 5-HT, Emmert SE. A functional role for opioid peptides in the differential pp 72–99. Ed. J Fozard. Oxford: Oxford University Press, 1989. secretion of vasopressin and oxytocin. Brain Research 1984 309 66 Schreiber R, Brocco M, Audinot V, Gobert A, Veiga S & Millan MJ. 362–366. (1-(2,5-dimethoxy-4 iodophenyl)-2-aminopropane)-induced head- 51 Faull CM, Charlton JA, Butler TJ & Baylis PH. The effect of acute twitches in the rat are mediated by 5-hydroxytryptamine (5- pharmacological manipulation of central serotonin neurotrans- HT) 2A receptors: modulation by novel 5-HT2A/2C antagonists, mission on osmoregulated secretion of arginine vasopressin in D1 antagonists and 5-HT1A agonists. Journal of Pharmacology and the rat. Journal of Endocrinology 1993 139 77–87. Experimental Therapeutics 1995 273 101–112. 52 Shoji M, Kimura T, Kawarabayasi Y, Ota K, Inoue M, Yamamoto T 67 Hoyer D, Clarke DE, Fozard JR, Hartig PR, Martin GR, Mylechar- et al. Effects of acute hypotensive hemorrhage on arginine vaso- ane EJ et al. International Union of Pharmacology classification of pressin gene transcription in the rat brain. Neuroendocrinology receptors for 5-hydroxytryptamine (serotonin). Pharmacology 1993 58 630–636. Reviews 1994 46 157–203. 53 Burnard DM, Pittman QJ & Veale WL. Increased motor disturb- 68 Hoyer D. Functional correlates of serotonin 5-HT1 recognition ances in response to arginine vasopressin following hemorrhage sites. Journal of Receptor Research 1988 8 59–81. or hypertonic saline: evidence for central AVP release in rats. Brain Research 1983 273 59–65. 54 Sapru HN, Krieger AJ & Haldar J. Vasopressin release in the spon- taneously hypertensive rat. Research Communications in Chemical Received 6 March 2002 Pathology and Pharmacology 1977 16 195–198. Accepted 22 July 2002

www.eje.org

Downloaded from Bioscientifica.com at 10/02/2021 01:07:08AM via free access