Responses of Anterior Pituitary Hormones and ...

Neuroendocrinology Editor-in-Chief:

S.M. McCann, Dallas. Tex.

Original Paper

Reprint Publisher: S. Karger AG. Basel Printed in Switzerland

Neuroendocrinology 1993;57:532-540

........................................................ Peter See-Jensen" Ulrich Knigge" Monique Garbarg" Andreas Kjœr" Agnes Rouleau" Flemming W Bacha Jean-Charles Schwartz" Jergen Warberg" a

b

Department of Medical Physiology, The Panum Institute, University of Copenhagen, Denmark; Unité de Neurobiologie et Pharmacologie, Unité 109, Institut National de la Santé et de la Recherche Médicale, Centre Paul-Broca, Paris, France

Responses of Anterior Pituitary Hormones and Hypothalamic Histamine to Blockade of Histamine Synthesis and to Selective Activation or Inactivation of Presynaptic Histamine H3 Receptors in Stressed Rats ................................................................................................. Abstract

.............................................. Key Words

Histamine Histamine synthesis a- Fluoromethylhistidine H3 receptors R(a)methylhistamine Thioperamide Prolactin ACTH ~-Endorphin Pituitary Hypothalamus Stress

The stress-induced release of anterior pituitary hormones and changes in hypothalamic content of histamine (HA) and its metabolite tele-methylHA (/meHA) were studied in male rats during inhibition of HA synthesis or activation or blockade of HA H3 receptors. Pretreatment with the HA synthesis inhibitor n-fluoromethylhistidine (a-FMH; 200 ug intracerebroventricularly (icv) at -120 min) or the specific H3 receptor agonist R(a)methylhistamine (RmHA; 10 mg/kg intraperitoneally (ip) at -180 and -60 min) inhibited by 30-80% the responses of prolactin (PRL), corticotropin (ACTH) and ~-endorphin (~-END) immunoreactivity to 1,2.5 or 5 min of restraint stress (p < 0.05-0.0 I), but had no effect on basal secretion of the hormones. The inhibitory effect of the H3 receptor agonist RmHA (10 mg/kg x 2) on the hormone response to 5 min of restraint stress was prevented by simultaneous ip administration of the H3 receptor antagonist thioperarnide. a- FMH reduced the hypothalamic content of HA 60% and that of t-meHA 30%, while RmHA had no effect on the HA content Restraint stress for 5 min did not affect the HA and t-meHA contents, which may be due to the short duration of stress exposure. Pretreatment with the H3 receptor antagonist thioperamide (5 or 10 mg/kg ip at -120 min) had no effect on basal or restraint stress-induced release of PRL, ACTH or ~-END, although the compound increased the hypothalamic content of t-meHA 2-fold. Insulin-induced (I lU/kg ip) hypoglycemia increased the plasma concentration of ACTH and ~-END 2- and 1.5-fold, respectively, and increased the hypothalamic content of r-rnel-LAalmost 2-fold, which indicates an increased turnover of neuronal HA. a-FMH or RmHA prevented the insulin-induced release of ACTH and ~-END. Insulin had no significant effect on PRL secretion. In conclusion, the findings that the response of PRL, ACTH and a-END to stress or insulin-induced hypoglycemia was inhibited by blockade of HA synthesis or activation of H3 receptors and that hypothalamic HA turnover increased in response to hypoglycemia further implicate a role of hypothalamic histaminergic neurons in the neuroendocrine regulation of pituitary hormone secretion.

Received: June 16, 1992 Accepted after revision: September 8, 1992

P. See-Jensen Department of Medical Physiology The Parium Institute, Building 12 University of Copenhagen, Blegdamsvej D K-2200 Copenhagen N (Denmark)

3e

Histamine (HA) acts as a neurotransmitter in the brain by activation of postsynaptic H I and H2 receptors [l, 2]. The histaminergic neurons originate in the posterior hypothalamus and project to other hypothalamic areas as well as other brain regions [3, 4]. HA exerts several actions in the hypothalamus and participates in the neuroendocrine regulation of pituitary hormone secretion [l, 5]. Histaminergic neurons appear to be involved in the mediation of the stress-induced release of prolactin (PRL), corticotropin (ACTH) and ~-endorphin (~-END), since the hormone responses to stress are diminished by inhibition of neuronal HA synthesis or blockade of postsynaptic HI or H2 receptors [6-8]. Recently, Schwartz and co-workers identified a presynaptic HA receptor in the brain. This receptor, which is termed a HA H3 receptor, is an autoreceptor localized to histaminergic nerve endings and perikarya and is pharmacologically distinct from HI and H2 receptors [9, 10].

Activation of H3 receptors by the specific agonist R(a)methylhistamine (RmHA) inhibits neuronal HA activity by inhibition of HA release and HA synthesis. In contrast, blockade of H3 receptors by the specific antagonist thioperamide prevents the effect of RmHA and, when administered alone thioperamide enhances neuronal HA activity in various brain areas, including the hypothalamus [11-13]. RmHA and thioperamide have recently been found to affect locomotor activity in mice and arousal mechanisms in cats [14, 15]. On the basis of these findings it is conceivable that activation of presynaptic H3 receptors by RmHA may lower the stress-induced release of pituitary hormones via decreased histaminergic activity in the hypothalamus. In contrast, blockade of presynaptic H3 receptors by thioperamide may - via increased histaminergic activity enhance the stress-induced release of pituitary hormones. Furthermore, thioperamide may abolish the possible inhibitory effect of RmHA on the stress-induced hormone release. To clarify further the role of histaminergic neurons in the neuroendocrine regulation of anterior pituitary hormone secretion, we investigated the effect of the HA synthesis inhibitor n-fluoromethylhistidine (aFMH) or the H3 receptor ligands RmHA and thioperamide on the release of PRL, ACTH and ~-END as stimulated by restraint stress or insulin-induced hypoglycemia in male rats. In addition, we studied the effect of restraint stress or insulin-induced hypoglycemia in combination with histaminergic compounds on the levels of HA and its metabolite tele-methylHA (tmeHA) in the hypothalamus.

Materials and Methods Animals Male rats of the Wistar strain (225-275 g) bred at the Panum Institute were housed under controlled conditions of temperature (22 ± 1°C), relative humidity (45-65%) and lighting (lights on 06.00-18.00 h daily). Compounds The compounds used were the histidine decarboxylase inhibitor a-FMH (gift from J. Kollonitsch, MSD Research Laboratories, West Point, Pa., USA) which blocks HA synthesis, the Hl receptor agonist RrnHA (gift from Laboratoire Bioproject, Paris, France), the Hl receptor antagonist thioperamide (gift from Laboratoire Bioproject), and insulin (Actrapid; NOVO, Copenhagen, Denmark). a-FMH and RrnHA were dissolved in saline while thioperamide was dissolved in methanol. Experimental Procedures Rats provided with a permanent metal cannula in a lateral ventricle of the brain (experiments 1and 3) or intact rats (experiments 2 and 3) were used. The permanent cannula was implanted during pentobarbital anesthesia (60 mg/kg intraperitoneally rip)) as previously described [6] approximately 1 week prior to experimentation. At the day of the experiment the cannula was extended by silastic tubing filled with a-FMH solution or saline. This permitted intracerebroventricular (icv) infusion of test substance without disturbing the animal. The animals were then left undisturbed for an adaptation period of at least 90 min prior to experimentation. In all experiments the animals were decapitated at 0 min. Experiment I: Effect of a-FM H on Restraint Stress-Induced Release of Pituitary Hormones and Hypothalamic Content of HA and t-meHA. The HA synthesis inhibitor a-FMH in a dose of200 ug or saline was infused iCY(total volume 5 ul; 2 ul/rnin) at -120 min. Restraint stress was performed by fixing the animal on its back for 0, l , 2.5 or 5 min. The animals were decapitated immediately after stress exposure. Experiment 2: Effect of RmHA and/or Thioperamide on Restraint Stress-Induced Release of Pituitary Hormones and Hypothalamic Content of HA or t-meHA. The Hl receptor agonist RrnHA in doses of2.5, 5 or 10 mg/kgwas injected ip(0.5 ml) at-180 and-60 min. Injection of saline served as control. Restraint stress was performed for 0, l , 2.5 or 5 min. The Hl receptor antagonist thioperamide (5 mg/kg) or methanol (control) was injected ip at -180 min and RrnHA (l0 mg/kg) was injected ip at -180 and -60 min. The rats were exposed to restraint stress for 5 min. Thioperamide in doses of 5 or 10 mg/kg or methanol was injected ip at -120 min. Restraint stress was performed for 0, 2.5 or 5 min. Experiment 3: Effect ofa-FMH or RmHA on Insulin/Hypoglycemia-Induced Release of Pituitary Hormones. After an overnight's fast, the animals were pretreated with the HA synthesis inhibitor a-FMH (200 ug iCYat -120 min) or the Hl receptor agonist RrnHA (10 mg/kg ip at -180 and -60 min). Saline administered iCYat -120 min or ip at -180 and -60 min served as control. Insulin (1 lU/kg) or saline was injected ip at -45 min.

533

Table 1. Effectof the HA synthesis inhibitor a- FMH (200 ug iCYat -120 min), the HJ receptor agonist RmHA (10 mg/kg ip at -180 and -60 min) or the HJ receptor antagonist thioperamide (5 mg/kg ip at -120 min) on the hypothalamic content of HA or the metabolite t-meHA in male rats during basal conditions or after 5 min of restraint stress. Saline (icv/ip) served as control solution for a-FMH and RmHA, and methanol served as control solution for thioperamide. Since no significance in these control groups were found, the data are pooled (n = 6-7 in each group).

Hypothalamic

content of HA and t-meHA

HA, ng/ g tissue control Saline/ methanol a-FMH RmHA Thioperarnide

b

487 177 448 533

± ± ± ±

t-meHA, ng/g tissue restraint

40' 29b 42 100

472 258 518 406

± ± ± ±

51 24b 35 33

control

restraint

142 ± 8 88 ± Ilb

143 ± 7 112 ± 6b

281 ± 19b

247 ± 13b

Mean ± SEM. P < 0.01 vs. saline/methanol.

Analysis of Hormones, Glucose, HA and t-meHA After decapitation of the rats, trunk blood was collected in tubes containing 500 lU heparin. The blood samples were centrifuged at 4 "C and plasma was stored at -20 "C until analyzed for PRL, ACTH and ~-END by specific radioimmunoassays [16, 17]. The least detectable quantity of PRL was 0.4 ug/I plasma and the intraand interassay coefficients of variation were 3.6 and 9.9%, respectively. The least detectable quantity of ACTH and ~-END were 1 and 4 pmol/l plasma, respectively. The intra- and interassay coefficients of variation were 4 and 5% for ACTH, respectively, and 8 and 15% for ~-END, respectively. The ~-END antiserum cross-reacted 100% with ~-lipotropin. Plasma glucose was measured by fluorometric determination using the glucose oxidase method [18]. The hypothalamus was immediately dissected according to the following boundaries: rostrally, the optic chiasm; caudally, behind the mamillary bodies, and laterally, the hypothalamic sulcus. The median eminence defined as a small area surrounding the origin of the pituitary stalk (weight approximately 0.4 mg) was removed from the hypothalamus in order to minimize the influence of its large content of mast cell HA on the analysis of hypothalamic neuronal HA. Each hypothalamus was homogenized in 2 x 250 ul o.l.N. perchloric acid containing 0.1 % EGTA. The homogenates were centrifuged at 4 ° C and the supernatant was lyophilized. The freezedried material was stored at -20 "C until analyzed for HA and tmeHA by radioimmunoassay as described [19]with slight modifications. Briefly, the dried samples were diluted with perchloric acid (0.4 N) and HA and t-meHA were derivatized with p-benzoquinone. The samples were then incubated with the [125I]iodinated tracer (HA or t-meHA-benzoquinone-leucine ['251]tyrosine) and an antiserum which was raised in rabbits against HA or t-meHA conjugated with bovine serum albumin via p-benzoquinone. The mixture was incubated for 16 h at 4 "C in swine anti-rabbit IgG-coated 96-well plates to allow the separation of bound from free radioactivity. The detection limits of the assays, defined as the concentration leading to 20% inhibition, were 3 and 5 pg for derivatized HA and t-meHA, respectively.

Restraint stress caused a time-dependent increase in the plasma concentrations of PRL, ACTH and ~-END in saline-pretreated control animals (fig. la, 2a, 3a). PRL increased 4- and 9-fold after 2.5 and 5 min stress, respectively (p < 0.01), while I min of restraint stress had no significant effect (fig. la). After I, 2.5 and 5 min of restraint stress ACTH increased 1.5-, 2- and almost 4-fold, respectively (p < 0.05-0.01; fig. 2a), while ~-END increased 2-, 3.5- and 6-fold, respectively (p < 0.0 I; fig. 3a). Since there was no significant difference in the hormone levels whether the rats were preinfused icv or ip with saline, the control values for icv and ip administration were pooled in figures 1-3. Inhibition of HA synthesis by pretreatment with aFMH 200 ug iCYhad no significant effect on the basal hormone secretion or the hormone response to I min of restraint stress. However, a-FMH reduced the PRL, ACTH and ~-END responses to 2.5 and 5 min of restraint stress by 50-80% (p < 0.01; fig. Ia, 2a, 3a). Restraint stress for 5 min had no significant effect on the hypothalamic content of HA or the metabolite tmeHA (table I). In control and restraint stressed animals, a-FMH diminished the hypothalamic HA content 65 and 45%, respectively (p < 0.01; table 1) and the t-meHA content 40 and 25%, respectively (p < 0.01; table 1).

Statistical Procedures Results are presented as the mean ± SEM and evaluated by one- or two-way analysis of variance, followed by Duncan's test for multiple comparisons. The limit of significance was p < 0.05.

Experiment 2: Effect of RmHA and/or Thioperamide on Restraint Stress-Induced Release of Pituitary Hormones and Hypothalamic Content of HA or t-meHA

Results

Experiment 1: Effect of a-FMH on Restraint StressInduced Release of Pituitary Hormones and Hypothalamic Content of HA and t-meHA

Pretreatment with the H3 receptor agonist RmHA (10 mg/kg ip at -180 and -60 min) had no significant effect

534

See-Jensen/ Knigge/Garbarg/ Kjœr/ Rouleau/ Bach/Schwartz/ Warberg

Histamine Hl Receptors and Pituitary Hormones

200

120 100 80

**

• Control (saline) .. a-FMH • RmHA

/

'§, :::L

_j

/

60

* _.f/

cr: a..

40

/1

/

/

160

a.

/

:c

IÜ

«

/

a

80

a

0 2.5

0

200

• Control (methanol) • Thioperamide 5 mg/kg .. Thioperamide 10 mg/kg

"" 160 "0 E ': 120

'§, 60 :::L

.r

5


:c

IÜ

40

«

20

b

2.5 Minutes of restraint stress

240

100

cr: a..

.-I

0

5

Minutes of restraint stress

80

.... .... .... **....

,f .... ....

40

-y 0

**

--

""E

"0 120

/

/

20

• Control (saline) .. a-FMH • RmHA

80 40

b

0 0

2.5

5


0 0

2.5

5


Fig. 1. a. b Effect of the HA synthesis inhibitor o.-FMH (200!lg icvat - [20 min) (a), the H3receptor agonist RmHA (10 mg/kg ip at -180 and -60 min) (a), or the H3 receptor antagonist thioperamide (5 or 10 mg/kg ip at -120 min) (b) on the PRL response to 0, 1,2.5 or 5 min of restraint stress. Saline (icv/ip) served as control for 0.FMH and RmHA, and methanol served as control for thioperamide. The results represent the mean ± SEM of 6-7 rats. *p < 0.05-0.01 vs. RmHA and o.-FMH; **p < 0.01 vs. o.-FMH and RmHA.

Fig. 2. a. b Effect of the HA synthesis inhibitor o.-FMH (a), the H3 receptor agonist RmHA (a), or the H3 receptor antagonist thioperamide (5 or ID mg/kg) (b) on the ACTH response to restraint stress. Further information is given in the legend to figure I. *p < 0.01 vs. RmHA; **p < 0.01 vs. o.-FMH and RmHA.

on basal hormone secretion but decreased the stress-induced release ofPRL, ACTH and ~-END (fig. la, 2a, 3a). After 2.5 and 5 min of restraint stress PRL was reduced 40 and 60%, respectively (p < 0.01; fig. 1a, whereas ACTH and ~-END were reduced 40-60 and 30-45%, respectively, at all times studied (p < 0.05-0.01; fig. 2a, 3a). Lower doses of RmHA (2.5 or 5 mg/kg ip at -180 and -60 min) had no significant effect on the restraint stress-induced release of these hormones (data not shown). RmHA (10 mg/kg x 2 ip) had no significant effect on the hypothalamic HA content (table 1). t-meHA was not

measured, since the antiserum directed towards t-meHA also binds to the methylated metabolite of RmHA. The inhibitory effect of RmHA (10 mg/kg x 2 ip) on the PRL, ACTH and ~-END response to 5 min of restraint stress was prevented by prior administration of the H3 receptor antagonist thioperamide (5 mg/kg ip at -180 min) (table 2). In rats pretreated with methanol (used as solvent for thioperamide) restraint stress caused a time-dependent increase in the plasma concentrations of PRL, ACTH and ~-END (p < 0.05-0.01; fig. lb, 2b, 3b). The pattern of secretion was similar to that found in saline-treated rats.

535

160 140 120 "0

E

Q_

ci

z

UJ

.,:,_

• Control (saline) .. ex-FMH • RmHA

........

100 80

*

60

....-

40

....-y

....- ....-

....-

** _A........

........

........

........

........

*** ........ !

*

100

80

"'" 60 "0

E Q_

20

a

I-

0

i-

0


5

~

40

160 140 120

"'" "0

100

Q_

80


20

E

ci

z .,:,_ I.J.J

a

0 --'--_ ......._""Sa ex-FMH RmHA

Sa

Saline

60

ex-FMH RmHA Insulin

40 20

b

*

40

0 0


5

Fig. 3. a, b Effect of the HA synthesis inhibitor a-FMH (a), the H3 receptor agonist RmHA (a), or the H3 receptor antagonist thioperamide (5 or 10 mg/kg) (b) on the ~-END response to restraint stress. Further information is given in the legend to figure I. *p < 0.05 vs. RmHA; **p < 0.05-0.01 vs. RmHA and a-FMH; ***p < 0.01 vs. a-FMH and RmHA.

30

**

**

"'""0E Q_

20

ci Z

I.J.J

.,:,_

10

Table 2. Effect of pretreatment with the H3 receptor antagonist thioperamide (5 mg/kg ip at -180 min) on the inhibitory effect of the H3 receptor agonist RmHA (10 mg/kg ip at -180 and -60 min) on the hormone response to restraint stress (-5 to 0 min). The rats were decapitated at 0 min (n = 6-7 in each group)

b

0 --'--_ ....... _""Sa ex-FMH RmHA Saline

Sa

ex-FMH RmHA Insulin

Restraint stress for 5 min

PRL ACTH ~-END

saline/methanol"

RmHA

thioperamide

90 ± 7b 186 ± Il 154 ± Il

47 ± se 127 ± 9c 95 ± 3e

97 ± 6 198 ± 14 180 ± 9

See text in table I. Mean ± SEM. p < 0.0 I vs. columns I and 3.

536

+ RmHA Fig. 4. a, b Effect of the HA synthesis inhibitor a-FMH or the Hl receptor agonist RmHA on the response of ACTH or ~-END to insulin-induced hypoglycemia. Insulin (I lU/kg ip) was administered at -45 min and the animals were decapitated at 0 min. The results represent the mean ± SEM of 6 rats. Sa = Saline. *p < 0.0 I vs. saline + saline; **p < 0.0 I vs. insulin + saline.

See-Jensen/ Knigge/ Garbarg/ Kjœr/ Rouleau/ Bach/Schwartzi Warberg


Pretreatment at -120 min with the H3 receptor antagonist thioperamide (5 or 10 mg/kg ip) had no effect on the stress-induced release of the hormones (fig. lb, 2b, 3b). Thioperamide had no significant effect on the hypothalamic content of HA, but increased the content of tmeHA 2-fold (p < 0.01; table 1). Experiment 3: Effect of a-FMH or RmHA on Insu/in/ Hypoglycemia-Induced Release of Pituitary Hormones Insulin (1 lU/kg) reduced plasma glucose from 5.69 ± 0.48 to 2.72 ± 0.13 mmol/l and increased the plasma concentrations of ACTH and ~-END about 2-fold (p < 0.01; fig. 4). Insulin-induced hypoglycemia had no significant effect on plasma PRL concentration (18 ± 4 vs. Il ± 3 ug/I). Inhibition of HA synthesis with a,-FMH (200 ug icv at -120 min) or activation of H3 receptors by RmHA (10 mg/kg ip at -180 and -60 min) abolished the responses of ACTH and ~-END to hypoglycemia (p < 0.01; fig. 4). Insulin had no effect on the hypothalamic HA content, but increased the content of t-meHA 2-fold (p < 0.01, fig. 5). a,-FMH reduced the content of HA and t-meHA in insulin-injected rats (p < 0.01 vs. saline + insulin; fig. 5).

600

500

400

!2!

œ c 300

-c:r:

2DD

100

0-'---'--Saline Saline a-FMH

a

Insulin

300

*

T

250

200

Discussion ~c We found that inhibition of HA synthesis by a,-FMH as well as activation of HA H3 receptors by RmHA reduced the response of PRL, ACTH and ~-END to restraint stress and of ACTH and ~-END to insulininduced hypoglycemia. The inhibitory effect of a,-FMH on the hormone response to stress is in accordance with previous observations [5, 8], and supports the contention that hypothalamic HA is involved in the hormonal stress response. The neuronal HA pool in the hypothalamus is reduced within a few hours after ip administration of a,-FMH [20] whereas the nonneuronal pool is only affected after several weeks of a,-FMH treatment [21], indicating that the effect of a,-FMH in the present study is exerted on the neuronal pool of HA. The concentrations of HA and tmeHA in the hypothalamus of control rats correspond to those previously found [19, 22]. However, in our study a,-FMH reduced the hypothalamic HA content by 65% while the t-meHA content was only reduced 40% in control rats. This is not in accordance with the findings of Oishi et al. [22], who found a larger decrease in t-meHA content than in HA content. The disagreement may be explained by difference in route of administration of a-

..f. 150 :r:

'"E

..!.

100

50

0 b

Saline Saline a-FMH Insulin

Fig. 5. The effect of insulin-induced hypoglycemia on the content of (a) HA and (b) t-meHA in the hypothalamus. The results represent the mean ± SEM of 6 rats. *p < 0.01 vs. saline control; **p < 0.01 vs. insulin + saline.

537

FMH and/or the interval between administration of uFMH and sacrifice. No changes in the hypothalamic content of HA or tmeHA was observed following 5 min of restraint stress, which seems to contradict an involvement of neuronal HA in the mediation of the hormone response to stress. However, the lack of effect may be related to the short duration of stress (5 min) when compared to the long TI/2 (almost 60 min) for hypothalamic HA [22]. The lack of effect seems unrelated to the stable steady state of neuronal HA (changes in synthesis equals changes in release), since restraint stress did not affect the HA and tmeHA levels in u-FMH-pretreated animals. In previous investigations discrepant results have been obtained, since it has been reported that stress either increased, reduced or had no effect on the HA content or turnover in the brain [23-26]. It has previously been shown that the H3 receptor agonist RmHA inhibits the release and synthesis of neuronal HA by activation of presynaptic H3 receptors [12, 13]. In addition, it has been reported that histaminergic neurons are involved in the mediation of the stress-induced release of PRL, ACTH and ~-END [6, 7]. Therefore, the present results suggest that the inhibitory effect of RmHA on retraint stress-induced release of PRL, ACTH and ~-END is caused by activation of presynaptic H3 receptors, leading to reduction of neuronal HA release and HA synthesis. This is further supported by our finding that the specific H3 receptor antagonist thioperamide prevented the inhibition exerted by RmHA on the hormone response to 5 min of restraint stress. Despite these results, we were unable to register any effect of RmHA on the hypothalamic content of HA, which may be related to the stable steady state or neuronal HA or to the inhibitory effect of RmHA on both HA release and synthesis. Furthermore, small changes in neuronal HA might be obscured by the presence of non-neuronal HA. However, it cannot be excluded that the effect of RmHA was mediated by activation of H3 heteroreceptors located on other aminergic neurons. The existence of cortical H3 heteroreceptors on presynaptic serotonergic or noradrenergic neurons have recently been implicated, since RmHA as well as HA inhibited the release of serotonin and noradrenaline from cortical slices, effects which were prevented by the H3 receptor antagonist thioperamide [27, 28]. The possibility that the inhibitory effect of RmHA is mediated via H3 heteroreceptors located on serotonergic or adrenergic neurons would seem to be in accordance with the fact that serotonin exerts stimulatory and norepinephrine exerts both inhibitory and stimulatory actions on the re-

538

lease of PRL, ACTH and ~-END [29-31]. Inhibition of serotonin and norepinephrine release by RmHA- or HAinduced activation of H3 heteroreceptors might therefore lead to a decreased hormone response to stress. However, interpretation of the data is intricate, since it has also been found that HA stimulates hypothalamic norepinephrine release by activation of HI or H2 receptors [32, 33] and that HA increases synaptosomal release of serotonin [34] and its turnover in the hypothalamus probably by activation of H2 receptors [35]. Furthermore, histaminergic stimulation of PRL secretion seems to involve serotonergic neurons [36]. Blockade of H3 receptors by the antagonist thioperamide prevented the inhibitory effect of the H3 receptor agonist RmHA on the restraint stress-induced release of PRL, ACTH and ~-END. Therefore, it was conceivable that administration of thioperamide prior to stress stimulation would enhance the hormone response by preventing the inhibitory effect of endogenously released HA on the presynaptic H3 receptors. However, this was not the case in spite of the fact that the dose of thioperamide used was capable of increasing the hypothalamic t-meHA content 2-fold. Since the effect of thioperamide is of long duration and appears within approximately 30 min [12] and since thioperamide was administered 120 min before stress stimulation, it is unlikely that the lack of effect on the hormone response to stress is explained by an inappropriate time schedule for the experiment. An explanation for the lack of effect of thioperamide might be that the amount of neuronal HA released locally after stress exposure is already exerting maximal histaminergic stimulation of hormone secretion and that further HA release as induced by thioperamide may have no additional effect on the hormone release. Another possibility might be that the H3 receptors involved are heteroreceptors located presynaptically on other aminergic neurons [27, 28] rather than histaminergic neurons. Since H3 heteroreceptors on some aminergic neurons may not be stimulated by endogenous histamine [28], this might explain why thioperamide administered alone in restraint-stressed animals had no effect. In contrast to our findings, it has recently been reported that thioperamide enhanced almost 100% the PRL response to morphine whereas RmHA had no effect on the response [37]. Since thioperamide and RmHA were administered in amounts comparable to those used in the present study, the discrepant observations seem unrelated to the doses used but rather related to differences in the involvement of HA in morphine- and stressinduced stimulation of PRL secretion.

See-J ensen/ Knigge/ Garbarg/ Kjœr/ Rouleau/Bach/Schwartz/ Warberg


Hypoglycemia induced by insulin stimulated the release of ACID and ~-END but had no significant effect on PRL secretion, which confirms previous observations [38-40]. At the same time, insulin, without affecting the HA level, increased the hypothalamic content of t-meHA almost 2-fold, indicating that neuronal HA is released by insulin-induced hypoglycemia. In accordance with this, it has previously been shown that the release of HA from hypothalamic tissue is increased by lowering the glucose concentration in the medium [41]. As found in the restraint stress experiments, the effect of insulin on ACID and ~-END secretion was prevented when the rats were pretreated with the HA synthesis inhibitor a-FMH or the H3 receptor agonist RmHA. Furthermore, pretreatment with a-FMH blocked the increase in hypothalamic tmeHA caused by insulin. These results strongly suggest that hypothalamic histaminergic neurons are involved in the mediation of the ACID and ~-END release caused by insulin-induced hypoglycemia. It has recently been shown that the H3 receptor agonist RmHA increased the release of ACID from the murine anterior pituitary tumor cell line AtT-20, an effect which was blocked by thioperamide but not by H I or H2 receptor antagonists [42]. However, it seems unlikely that the effect of the H3 receptor compounds seen in this study is caused by an effect at a pituitary level, since we found the opposite effect of RmHA compared to that reported by Clark et al. [42], since HA has no effect directly at the level of the normal pituitary gland [43, 44] and since tumor cells may express other receptors than normal cells. Furthermore, the effect observed in the previous study [42] is difficult to interpret, since it is delayed by several hours after the beginning of the incubation. Therefore, we suggest that the effect of the histaminergic system is exerted at a hypothalamic level.

We and others have suggested that histaminergic neurons are involved in the stress-induced release of PRL [5]. However, despite insulin-induced hypoglycemia increased ACID and ~-END secretion and the hypothalamic HA turnover, the stimulus did not affect PRL secretion significantly. We have no explanation for this lack of effect, although it might be due to the fact that PRL is tonically inhibited by hypothalamic dopamine, whereas the hypothalamic net effect on ACID and ~-END secretion is stimulatory, primarily exerted by CRH and vasopressin [29, 30, 31]. In summary, we found (1) that inhibition of neuronal HA synthesis by a-FMH and activation of inhibitory presynaptic H3 receptors by the agonist RmHA inhibited the response of PRL, ACID and ~-END to restraint stress and the ACID and ~-END response to insulin-induced hypoglycemia, and (2) that insulin-induced hypoglycemia increases the neuronal HA turnover in the hypothalamus. The results support previous reports implicating a role of hypothalamic histaminergic neurons in the neuroendocrine regulation of pituitary hormone secretion.

Acknowledgments This study was supported by grants from the Danish Medical Research Council; lb Henriksen's Foundation; the Velux Foundation; P. Carl Petersen's Foundation; the Danish Hospital Foundation for Medical Research, Regions of Copenhagen, The Faroe Islands and Greenland; Novo's Foundation; Nordic Insulin Foundation Committee; Gerda and Aage Haensch Foundation, and the Foundation for the Advancement of Medical Research. We thank Elsa Larsen, Jytte Oxbel and Jimmy Bidstrup for skilled technical assistance. The materials for RIA of prolactin and AClli were kindly provided by the National Hormone and Pituitary Program of the NIDDK.

................................................................................................................................................... References Schwartz J-C, Arrang J-M, Garbarg M, Pollard H, Ruat M: Histaminergic transmission in the mammalian brain. Physiol Rev 1991; 71:1-51. 2 Hough LB: Cellular localization and possible functions for brain histamine: Recent progress. Prog Neurobiol 1988;30:469-505. 3 Panula P, Pirvola U, Auvinen S, Airaksinen MS: Histamine-immunoreactive nerve fibers in the rat brain. Neuroscience 1989;28:585610.

4 Inagaki N, Yamatodani A, Ando-Yamamoto M, Tohyama M, Watanabe T, Wada H: Organization of histaminergic fibers in the rat brain. J Comp Neurol 1988;273:283-300. 5 Knigge U, Warberg J: The role of histamine in the neuroendocrine regulation of pituitary horrnone secretion. Acta Endocrinol (Copenh) 1991;124:609-619. 6 Knigge U, Matzen S, Warberg J: Histaminergic mediation of the stress-induced release of prolactin in male rats. Neuroendocrinology 1988;47:68-75.

7 Knigge U, Matzen S, Bach FW, Bang P, Warberg J: Involvement of histaminergic neurons in the stress-induced release of pro-opiomelanocortin-derived peptides in rats. Acta Endoerinol (Copenh) 1989;120:533- 559. 8 Knigge U, Matzen S, Hannibal T, Jorgensen H, Warberg J: Involvement of histamine in the mediation of the stress-induced release of alpha-melanocyte-stimulating horrnone in male rats. Neuroendocrinology 1991;54:646-652.

539

9 Arrang J-M, Garbarg M, Schwartz J-C: Autoinhibition of brain histamine release mediated by a novel class (H,) of histamine receptor. Nature 1983;302:832-836. 10 Arrang J-M, Garbarg M, Schwartz J-C: Autoinhibition of histamine synthesis mediated by presynaptic Hr-receptors. Neuroscience 1987;23:149-157. II Arrang J-M, Garbarg M, Lancelot J-C, Lecomte J-M, Pollard H, Robba M, Schunack W, Schwartz J-C: Highly potent and selective ligands for histamine Hi-receptors. Nature 1987;327:117-123. 12 Garbarg M, Trung Tuong MD, Gros C, Schwartz J-C: Effects of histamine Hy-receptor ligands on various biochemical indices of histaminergic neuron activity in rat brain. Eur J Pharmacol 1989;164:1-11. 13 Oishi R, Itoh Y, Nishibori M, Sadd K: Effects of the histamine Hr-agonist (R)-a-methylhistamine and the antagonist thioperamide on histamine metabolism in the mouse and rat brain. J Neurochem 1989;52:1388-1392. 14 Sakai N, Onodera K, Maeyama K, Yanai K, Watanabe T: Effects of thioperarnide, a histamine Hl receptor antagonist, on locomotor activity and brain histamine content in mast cell-deficient W/W mice. Life Sci 1991 ;48: 2397-2404. 15 Lin J-S, Sakai K, Vanni-Mercier G, Arrang JM, Garbarg M, Schwartz J-C, Jouvet M: Involvement of histaminergic neurons in arousal mechanisms demonstrated with Hi-receptor ligands in the cat. Brain Res 1990;523 :325330. 16 Wogensen L, Warberg J: Cyclic and diurnal alteration in the response of luteinizing hormone and prolactin to prostaglandin E, during the rat oestrus cycle. Acta Endocrinol (Copenh) 1984;106:30-37. 17 Knigge U, Bach FW, Matzen S, Bang P, Warberg J: Effect of histamine on the secretion of pro-opiomelanocortin derived peptides in rats. Acta Endocrinol (Copen h) 1988;119: 312-319. 18 Guilbault GG, Brignac PJ Jr., Juneau M: New substrates for the fluorometric determination of oxidative enzymes. Anal Chem 1968 ;80: 1256-1263. 19 Garbarg M, Pollard H, Trung Tuong MD, Schwartz J-c, Gros C: Sensitive radioimmunoassays for histamine and tele-methylhistamine in the brain. J Neurochem 1989 ;53: 1724-1730.

20 Garbarg M, Barbin G, Rodergas E, Schwartz JC: Inhibition of histamine synthesis in brain by a-fluoromethylhistidine, a new irreversible inhibitor: In vitro and in vivo studies. J Neurochem 1980;35:1045-1052. 21 Duggan DE, Hooke KF, Maycock AL: Inhibition of histamine synthesis in vitro and in vivo by S-a-fluoroemthylhistidine. Biochem Pharmacol 1984;33:4003-4009. 22 Oishi R, Nishibori M, Saeki K: Regional differences in the turnover of neuronal histamine in the rat brain. Life Sci 1984;34:691--699. 23 Taylor KM, Snyder SH: Brain histamine: Rapid apparent turnover altered by restraint and cold stress. Science 1971; 172: I 037-1 039. 24 Kobayashi RM, Kopin IJ: The effects of stress and environmental lighting on histamine in the rat brain. Brain Res 1974;74:356-359. 25 Verdière M, Rose C, Schwartz J-C: Turnover of cerebral histamine in a stressful situation. Brain Res 1977;129:107-119. 26 Yoshitomi I, Itoh Y, Oishi R, Saeki K: Brain histamine turnover enhanced by foots hock. Brain Res 1986;362:195-198. 27 Fink K, Schlicker E, Neise A, Gôthert M: Involvement of presynaptic Hl receptors in the inhibitory effect of histamine on serotonin release in the rat brain cortex. Naunyn Schmiedebergs Arch Pharmacol 1990;342: 513-519. 28 Schlicker E, Fink K, Hinterhaner M, Gôthert M: Inhibition of noradrenaline release in the rat brain cortex via presynaptic Hl receptors. Naunyn Schmiedebergs Arch Pharmacol 1989;340:633-638. 29 Tuomisto J, Mânnistô PT: Neurotransmitter regulation of anterior pituitary hormones. Pharmacol Rev 1985;37:249-332. 30 Plotsky PM, Cunningham ET, Widmaier EP: Catecholaminergic modulation of corticotropin-releasing factor and adrenocorticotropin secretion. Endocr Rev 1989; 10:437458. 31 Weiner RI, Findell PR, Kordon C: Role of classic and peptide neuromediators in the neuroendocrine regulation of LH and prolactin; in Knobil E, Neill J, et al (eds): The Physiology of Reproduction. New York, Raven Press, 1988, vol 37, pp 249-332. 32 Blandina P, Knott PJ, Leung LKH, Green JP: Stimulation of histamine H, receptor in rat hypothalamus releases endogenous norepinephrine. J Pharmacol Exp Ther 1989 ;249: 44-51. 33 Phillipu A, Bald M, Kraus A, Dietl H: In vivo release by histamine agonists and antagonists of endogenous catécholamines in the cat hypothalamus. Naunyn Schmiedebergs Arch PharmacoI1984;326:116-123.

34 Tuomisto J, Tuomisto L: Effects of histamine and histamine antagonists on the uptake and release of catecholamines and 5-HT in brain synaptosomes. Med Biol 1980;58 :33-37. 35 Pile A, Nowak JZ: Influence of histamine on the sertonergic system of rat brain. Eur J Pharmacol 1979;55 :269-272. 36 Knigge U, Sieimann I, Matzen S, Warberg J: Histaminergic regulation of prolactin secretion: .Involvement of serotoninergic neurons. Neuroendocrinology 1988;48 :527-533. 37 Netti C, Guidobono F, Sibilia V, Pagani F, Villa I, Pecile A: Effects of selective histamine Hr-receptor ligands on prolactin and growth hormone secretion in the rat. Agents Actions 1991 ;33: 147-149. 38 EI-Tayeb KM, Brubaker PL, Lickley HL, Cook E, Vranic M: Effect of opiate-receptor blockade on normoglycemic and hypoglycemic glucoregulation. Am J Physiol 1986;250:E236-E242. 39 Radosevich PM, Lacy DB, Brown LL, Williams PE, Abumrad NN: Effects of insulin- induced hypoglycemia on plasma and cerebrospinal fluid levels of ir-beta-endorphins, ACTH, cortisol. Brain Res 1988;458:325338. 40 Quabbe HJ, Bunge S, Walz T, Bratzke B: Plasma glucose and free fatty acids modulate the secretion of growth hormone, but not prolactin, in the rhesus and Java monkey. J Clin Endocrinol Metab 1990;70:908-915. 41 Nishibori M, Oishi R, Itoh Y, Saeki K: Glucose modulates the release of histamine from the mouse hypothalamus in vitro. J Neurochem 1986;47:1761-1767. 42 Clark MA, Korte A, Myers J, Egan RW: High affinity histamine Hj-receptors regulate ACTH release by AtT-20 cells. Eur J PharmacoI1992;210:31-35. 43 Hashimoto K, Yunoki S, Tekahara J, Ofuji T: ACTH release in pituitary cell cultures. Effect of neurogenic peptides and neurotransmitter substances on ACTH release induced by hypothalamic corticotropin-releasing factor. Endocrinol Jpn 1979;26:103-109. 44 Vermes I, Mulder GH, Smelik PG, lilders FJH: Differential control of ~-endorphin/~lipotropin secretion from anterior and intermediate lobes of the rat pituitary gland in vitro. Life Sci 1980;27:1761-1768.

540

Soe-Jensen/ Knigge/Garbarg/ Kjœr/ Rouleau/Bach/Schwartz/ Warberg

Histamine H3 Receptors and Pituitary Hormones