0013-7227/90/1265-2527$02.00/0 Endocrinology Copyright© 1990 by The Endocrine Society

Vol. 126, No.- 5 Printed in U.S.A.

Prolactin Response to Immobilization Stress and Hemorrhage: The Effect of Hypothalamic Deafferentations and Posterior Pituitary Denervation J. JURCOVICOVA, R. KVETNANSKY, M. DOBRAKOVOVA, D. JEZOVA, A. KISS, AND G. B. MAKARA Institute of Experimental Endocrinology, Centre of Physiological Sciences, Slovak Academy of Sciences, and the First Department of Neurology, Medical School, Comenius University (A.K.), Bratislava, Czechoslovakia; and the Institute of Experimental Medicine, Hungarian Academy of Sciences (G.B.M.), Budapest, Hungary

ABSTRACT. The roles of posterior and anterolateral connections to the mediobasal hypothalamus (MBH) as well as innervation of the posterior pituitary in the PRL response to immobilization (IMO) and hemorrhage (HEM) were studied by means of surgical isolation, performed 6-9 days before stress exposure. Male rats bearing indwelling tail artery cannulae subjected to 120-min IMO reached peak PRL secretion in 5-20 min. HEM of 25% elicited a significant rise of PRL levels. A posterior cut in the MBH, performed without damaging the serotonergic fibers from the brain stem, attenuated the PRL response to 25% HEM, whereas the PRL elevation due to IMO remained unaffected.

T

HE RELEASE of PRL from the anterior pituitary under various circumstances is controlled by multiple stimulatory and inhibitory factors (1-3). In intact male rats the stress-induced release of PRL is triggered by activation of putative PRL-releasing factors (4-6), probably originating in the hypothalamus anterolaterally to the mediobasal hypothalamus (MBH) (7, 8). The caudal area of the MBH was also shown to be involved in PRL stimulation by ether stress (9). There is increasing evidence indicating that besides the median eminence (ME) the posterior pituitary (PP) is also a source of stimulators and inhibitors of PRL release. Posterior lobe extracts inhibited PRL secretion in vitro (10), and direct dopaminergic inhibitory effect was shown in vivo (11). The neurohypophysial hormones are known to affect PRL release; vasopressin (AVP) stimulated PRL secretion in male rats and in ectopic pituitaries located under the kidney capsule (12). On the other hand, AVP antagonized novelty stress-induced PRL secretion in male rats (13). Oxytocin (OT) was Received October 23,1989. Address all correspondence and requests for reprints to: J. Jurcovicova, Ph.D., Institute of Experimental Endocrinology, Centre of Physiological Sciences, Slovak Academy of Sciences, Vlarska 3, 833 06 Bratislava, Czechoslovakia.

An anterolateral cut around the MBH eliminated both the IMOand HEM-induced stimulation of PRL. Posterior lobe denervation reduced by about 27% the PRL response to IMO and eliminated the response to HEM. These results suggest the following conclusions. The neural structures located posteriorly to the MBH are involved in the transfer of signals triggering PRL secretion due to hypovolemia. Intact anterolateral pathways to the MBH and stalk-median eminence region are essential for the PRL-releasing activity under both stimuli. The posterior lobe may be an important link in the PRL stress response in male rats. {Endocrinology 126: 2527-2533, 1990)

found to s t i m u l a t e P R L secretion in vitro a n d in

vivo,

with an obvious physiological preponderance in female rats during lactation (14, 15). In male rats, OT was not involved in PRL stimulation by ether (16), and it antagonized the novelty stress effect (13, 17). Recent findings of Ben-Jonathan's group demonstrated that posterior lobectomy blocked suckling-induced PRL release (18) as well as the PRL surge on proestrus (19). These researchers described and characterized a PRL-releasing factor in the posterior pituitary (PP) independent of OT (20, 21). In the present study we attempted to elucidate the relative roles of anterolateral and posterior connections to the hypothalamus as well as the role of the neural lobe in the release of PRL in male rats in response to two different stress stimuli: immobilization (IMO) and hemorrhage (HEM). The experiments presented here show that deafferentation of the posterior hypothalamus reduced the PRL response to HEM, but failed to affect the response to IMO, an anterolateral cut around the MBH (ALC) eliminated the response to both HEM and IMO, and denervation of the PP reduced IMO-induced PRL release and abolished the effect of HEM. Preliminary results of this work were presented at the 31st Congress of IUPS (22). 2527

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 20:21 For personal use only. No other uses without permission. . All rights reserved.

HYPOTHALAMIC CUTS, POSTERIOR LOBE DENERVATION, PRL

2528

Materials and Methods

Endo • 1990 Vol 126-No 5

Cannulation

Altogether 150 male albino Wistar rats of a specific pathogen-free colony (Velaz, Prague, Czechoslovakia), weighing 312 ± 10 g, were used. They were housed under controlled conditions in the animal room (temperature, 24 C; lights on between 0600-1800 h) with free access to pelleted rat chow and tap water and were kept there for at least 1 week before onset of the experiments.

The rats were anesthetized with pentobarbital (50 mg/kg, ip) 20-24 h before the experiments, and a polyethylene tube (P.S. 0.5 x 1.0; VEB MLW Halberstadt, East Germany) was inserted into the tail artery. The free end of the cannula was drawn under the skin to the back of the neck and exteriorized to the top of the cage through a stainless steel spring fixed to the skin. The tube was flushed with heparinized saline (150 IU/0.5 ml). After the cannulation each rat was housed in an individual cage.

Surgery and histology

Experimental procedure

For all surgical interventions the rats were anesthetized with sodium pentobarbital (50 mg/kg; SPOFA, Prague, Czechoslovakia) and subjected to the following operations using a Halasztype knife with a 1.8 mm radius and height. ALC, the area deafferented, contained the ventromedial, arcuate, and premammillary nuclei, and caudally it was open to the mammillary region. A posterior cut (PC) was placed in the mammillary region behind the medial portion of the hypothalamus. It interrupted only the posterior connections of the hypothalamus (23). For denervation of the PP, an L-shaped rotating wire knife bent to about 70° to the shaft with the blade pointing caudally was used. The rats were fixed in a David Kopf (Tunaga, CA) apparatus in the position 5 mm nose down. At a point 4.6 mm behind the bregma the knife was lowered in the midline through the sagittal sinus to 9 mm below the level of the top of the skull, and the blade was turned +60° to the left. It was further lowered to the bone of the base of the brain. Then, it was lifted by 0.6 mm and turned to -60° to the other side, lowered by 0.2 mm, turned to +60°, lowered by 0.2 mm, and turned to -60° until the bone was touched. It was then brought back to the midline and removed from the brain (24). This procedure interrupted the nerve fibers in the stalk by pressing them against the bone. In the sham-operated animals the knife was lowered into the brain to the level of the thalamus without rotating the blade. After the operation the rats with PC or posterior lobe denervation (PLD) and a representative number of sham-operated animals were placed into individual cages to measure their water consumption. To check the completeness of the surgery, the hypothalami and pituitaries were embedded in paraffin and serially sectioned in the coronal plane. The sections were stained and evaluated for neurosecretory material (23). Only rats with ALC reaching the basal surface of the hypothalamus were taken into consideration. Of the rats with PC, only those without any compression or crush of the stalk or PP were included in the experiments. Of the PLD animals we selected rats with intact anterior pituitaries free of necrosis and without neurosecretory material in the neural lobe. Several brains after PC were used for immunostaining of serotonin in the area immediately surrounding the transsection. Two or 10 days after surgery immunostaining of selected freely floating cryostat sections was performed using avidin-biotin complex immunostaining reagents, as described previously (25). Serotonin antibodies (Immuno-Nuclear, Stillwater, MN) in a 1:1000 dilution were used.

The experiments were performed between 0800-1100 h, 6-9 days after surgery. All external noises or any other disturbance of the animals were avoided. In one type of experiments PC, ALC, and PLD rats as well as sham-operated controls were subjected to IMO, performed by taping the limbs to metal mounts attached to a plastic board. The duration of IMO was 120 min. Blood was withdrawn in 1-ml portions before IMO and at 5, 20, and 120 min of IMO. After each blood sample, 1 ml physiological saline was injected as a replacement for the lost volume. In the other type of experiment, HEM was used as the stress stimulus in rats after PC, ALC, or PLD and in sham-operated controls. Twenty-five percent of the estimated blood volume (26) was removed by spontaneous bleeding from the tail artery. Blood was collected in 1-ml portions at the beginning of the bleeding (control value), then at 12.5% and 25% HEM (including the samples), and 30 min after HEM. No saline or other liquid was injected during the procedure. The blood was collected into plastic tubes and centrifuged at 4 C, and plasma was stored at -20 C until assayed.

Animals

Hormone assay Plasma PRL was measured by RIA using the material kindly provided by the National Pituitary Agency, NIDDK (Bethesda, MD). The coefficient of intraassay variability was about 6%, and interassay variability was about 11%. All samples were measured in duplicate, and samples from a particular experimental set were measured in the same assay. The PRL concentrations are expressed in terms of the NIDDK rat PRL RP-2 or NIDDK rat PRL RP-3 standards. Statistical evaluation The statistical significance was established by two-way analysis of variance, followed by Dunnett's test (27) when comparing sequential changes within one group, and Dunn's test (28) when comparing the differences between groups. To compare the areas under the PRL curves the nonparametrical test of Mann-Whitney was used.

Results In all experiments IMO evoked a typical biphasic PRL response, as described previously (29, 30); the rapid rise to the peak was followed by a decline in spite of continuing IMO. Blood loss of 25% elicited a clear PRL release,

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 20:21 For personal use only. No other uses without permission. . All rights reserved.

HYPOTHALAMIC CUTS, POSTERIOR LOBE DENERVATION, PRL

2529

while 12.5% blood loss did not affect the initial PRL material in the neural lobe of the sham-operated rats, levels (Figs. 1, 2, and 5). The duration of HEM ranged whereas in the neural lobe of the rats after PLD there from 21 ± 2 to 27 ± 4 min in the different series without was no stainable neurosecretory material. One week after being influenced by any of the surgical procedures used. PLD the AVP content of the neural lobe dropped to less No significant correlation was found between the durathan 5% of the control value, and the vasopressin content tion of bleeding and PRL responses. in the stalk-median eminence increased 3-fold (Makara, Figure 1 shows PRL responses to IMO or HEM after G. B., in preparation). In Fig. 4 is depicted the 6-day the PC or sham operation. The mean volume of water profile of water consumption in the rats after PLD or consumed per day did not differ between the two groups. sham operation. The highest water consumption was Activation of PRL induced by IMO was significant (P < observed on the morning after surgery, i.e. approximately 0.01) over the time course studied in both sham-operated 12 h after recovery from anesthesia. During the following rats and rats after PC, without any differences between 24 h the volume of consumed water decreased sharply the groups (Fig. 1, right panel). PRL levels in response and then remained at a constant but significantly higher to HEM are depicted in the left panel of Fig. 1. The level than that in the sham-operated animals until day HEM-associated PRL release was significantly reduced 6. The water consumption of sham-operated rats was by PC, though not totally inhibited. Thirty minutes after significantly lowered immediately after the operation, completion of the bleeding plasma PRL levels returned became normalized within the next day, and then reto initial values in both groups. mained stable. The right panel of Fig. 5 shows that Because of the well known importance of serotonergic denervation of the PP attenuated the PRL response to fibers for PRL release any possible damage to the asIMO. After PLD the PRL levels increased significantly, cending serotonergic fibers by PC was checked. The but the peak increase was lower than that in the shamposterior isolation of the hypothalamus did not exhibit operated immobilized rats at both 5 and 20 min of IMO. any sign of transection of efferent serotonergic pathways The area under the PRL curve in the group of shamfrom the dorsal raphe or other brain stem serotonergic operated animals was 5556 ± 370 ng/min-ml (35 rats), nuclei. Two and 10 days after transection no fibers with and in the group after PLD it was 4051 ± 604 ng/minretrograde or anterograde accumulation of serotoninml (18 rats; P < 0.01). HEM-induced changes are deimmunostained material were seen around the cut. picted in the left panel. The increased PRL release obIn the next experimental series we evaluated the reserved in sham-operated rats was eliminated by PLD. sponse of PRL to IMO and HEM after the ALC (Fig. 2). This group did not exhibit any PRL response. ALC alone did not affect the prestress PRL levels, but the response to IMO was suppressed (Fig. 2, right panel). Discussion A similar effect of ALC was observed in HEM-induced Immobilization and HEM, both potent stimuli for PRL release. This group failed to produce any change PRL release but different in nature, appear to stimulate during the interval studied, whereas sham-operated aniPRL secretion in different ways. While IMO, which is mals showed a pronounced response to 25% HEM (Fig. an extreme stimulus of physical and emotional stress, 2, left panel). elicited a prompt peak of PRL, the effect of HEM was In the last experimental series we studied the secretion dependent on its magnitude. Other stress hormones, such of PRL induced by IMO and HEM after denervation of as ACTH and peripheral catecholamines in rats (31) and the PP. Figure 3 shows an abundant neurosecretory ng.ml

FIG. 1. Left panel, Plasma PRL levels in rats in response to 12.5% and 25% HEM and 30 min later in sham-operated rats (n = 15) and rats with PC (n = 25). C, Control. Right panel, PRL response to 120-min IMO in sham-operated (n = 27) and PC (n = 17) animals. Values represent the mean ± SEM. Statistical significance vs. initial values: +, P < 0.05; ++ , P < 0.01; and between the groups: **, P