Modulation of stress-induced and stimulated hyperprolactinemia with the group II metabotropic glutamate receptor selective agonist, LY379268

Neuropharmacology 43 (2002) 799–808 www.elsevier.com/locate/neuropharm

Modulation of stress-induced and stimulated hyperprolactinemia with the group II metabotropic glutamate receptor selective agonist, LY379268 M.P. Johnson *, M. Chamberlain Neuroscience Department, Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, IN 46285, USA Received 21 January 2002; received in revised form 21 June 2002; accepted 12 July 2002

Abstract It is well recognized that glutamate is an integral excitatory neurotransmitter in the neuroendocrine control of several hormonal factors. While the ability of pharmacological agents acting at ionotropic glutamate receptors to modulate the levels of serum prolactin levels has been investigated, there have been few reports of the effects mediated by the G-protein coupled, metabotropic glutamate (mGlu) receptors. The present work was undertaken to investigate the role of the Group II mGlu receptors, mGlu2 and mGlu3 in the regulation of serum polactin levels. LY379268, a Group II selective agonist, did not alter basal levels of circulating prolactin in young (36–40 day old) male rats. However, when an immobilization stress-induced hyperprolactinemia was examined, 10 mg/kg s.c. of LY379268 significantly lowered serum prolactin levels. Similarly, pretreatment with LY379268 was able to reverse the hyperprolactinemia induced with the catecholamine synthesis inhibitor, a-methyl-p-tyrosine (aMPT). This inhibition of hyperprolactinemia could be prevented by pretreatment with LY341495, a Group II mGlu receptor antagonist. The Group II antagonist alone had no effect on either basal nor stimulated prolactin levels. The agonist LY379268 was able to prevent the transient hyperprolactinemia associated with stimulation of serotonin 5-HT2A receptors by 2,5-dimethoxy-4-iodoamphetamine (DOI), but did not alter the high levels of circulating prolactin induced with the D2 antagonist, haloperidol. When treatment with LY379268 was delayed until 1 h after aMPT, a time demonstrated to show a full effect of aMPT on serum prolactin levels, the Group II agonist was similarly able to reverse hyperprolactinemia, suggesting LY379268 did not act by preventing the partial catecholamine depletion by aMPT. Similarly, high doses of amphetamine, a dopamine (DA) releaser, were able to reverse the aMPT-induced hyperprolactinemia, consistent with sufficient levels of dopamine remaining after aMPT treatment to modulate prolactin levels. LY379268 did not alter the hyperprolactinemia seen in estrogen-primed, ovariectomized female rats. Taken together the results indicate that stimulation of mGlu2/3 has an indirect inhibitory action on pituitary prolactin release. It is speculated that disinhibition of tubero-infundibular DA release by presynaptic Group II mGlu receptors located on inhibitory inputs to the arcuate hypothalamic nucleus is a possible explanation for the findings.  2002 Elsevier Science Ltd. All rights reserved. Keywords: mGlu2/3; alpha-methyl-p-tyrosine; LY379268; LY341495; DOI; GABA

1. Introduction Prolactin is an anterior pituitary hormone required for endocrine control of lactogenesis in females. However, in both males and females, prolactin also plays a key role in activated T-cell immune responses and shows a circadian rhythmic-like response to inactivity. Further-

*

Corresponding author. +1-317-277-6581; fax: +1-317-276-5546. E-mail address: [email protected] (M.P. Johnson).

more, along with glucocorticoids, prolactin is increased as part of the normal stress responses in mammals (Conn and Freeman, 2000). Thus, prolactin plasma levels are tightly controlled in both male and females. It is well accepted that prolactin is under tonic inhibitory control by one or more prolactin release inhibitory factors (PIF). The evidence to date suggests that one major PIF is dopamine (DA), which is released into the hypophysial portal vasculature from the tubero-infundibular tract. The neuronal cell bodies that contribute to the tuberoinfundibular tract reside within the arcuate and periven-

0028-3908/02/$ - see front matter.  2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 8 - 3 9 0 8 ( 0 2 ) 0 0 1 4 2 - 9

800

M.P. Johnson, M. Chamberlain / Neuropharmacology 43 (2002) 799–808

tricular nuclei of the hypothalamus (Ben-Jonathan, 1985). Thus, DA releasing agents, such as amphetamine, will decrease plasma prolactin level in rats (Demarest et al., 1984; Horowski and Graf, 1976). In contrast, partial depletion of DA with catecholamine synthesis inhibitors (e.g. a-methyl-p-tyrosine methyl ester or aMPT), lesions of the arcuate nucleus, or severing the tubero-infundibular tract will increase prolactin levels (Ben-Jonathan, 1985; Horowski and Graf, 1976; Moore et al., 1980). The well documented finding that D2-antagonists result in pronounced and persistent hyperprolactinemia further demonstrates that it is stimulation of the D2 receptor on pituitary lactotropes cells that mediates the tonic inhibition of prolactin release (Demarest et al., 1984; Horowski and Graf, 1976). Both aMPT and D2 antagonists such as haloperidol have been found to have similar effects both in rodents and in humans (Rubin et al., 1976). DA release from the tubero-infundibular tract is thought to be controlled both directly and indirectly by inhibitory and excitatory projections into the hypothalamic nuclei. Evidence to date suggests that a major excitatory input regulating the arcuate nucleus is glutamatergic. For example, the glutamatergic agonist, AMPA (alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid), decreases serum prolactin levels while the AMPA/kainate receptor antagonist NBQX (2,3-dihydro6-nitro-7-sulphamoyl-benzo(f)quinoxaline) induces a hyperprolactinemia (Gonzalez et al., 1999a, b; Gonzalez et al., 2000). In contrast, the N-methyl-d-aspartate (NMDA) family of glutamatergic ionotropic receptors, has the opposite effect. It has been shown that NMDA mediates an increase in serum prolactin while the NMDA channel blockers such as MK-801 will most often decrease prolactin levels (Aguilar et al., 1997; Login, 1990; Pechnick et al., 1989; Wagner et al., 1993). This implies a complex neuronal circuitry where glutamatergic manipulations can have state-dependent and receptor subtype-specific actions on neuroendocrine control of prolactin release. In addition to the ionotropic glutamate receptors there are a number of G-protein coupled glutamate receptors, the metabotropic glutamate receptors (mGlu) (Schoepp et al., 1999). However, little is known about metabotropic glutamate receptor modulation of prolactin release. There is evidence indicating that mGlu1, 5, 2, 3, and 7 are all present in the arcuate and periventricular hypothalamic nucleus (Ohishi et al., 1995; Ohishi et al., 1993b; Petralia et al., 1996; Van den Pol, 1994; Van den Pol et al., 1994, 1995). While mGlu1 and 5 have been largely found to be stimulatory (such as intracellular calcium release), mGlu2, 3 and 7 are primarily inhibitory (such as the inhibition of adenylate cyclase). In the later case, there is evidence for presynaptic autoreceptor actions (inhibition of glutamate release) and heteroreceptor actions (inhibition of other neurotransmitter release)

(for review see Cartmell and Schoepp, 2000). Thus, it was hypothesized that treatment with mGlu2/3 agonists might alter the levels of circulating prolactin levels.. LY379268 ((-)-2-oxa-4-aminobicyclo[3.1.0] hexane-4,6dicarboxylate), a selective systemically-active mGlu2/3 agonists (Monn et al., 1999), was utilized as a pharmacological tool in these experiments to test that hypothesis. 2. Methods 2.1. Animal subjects and pharmaceutical agents a-Methyl-p-tyrosine methyl ester HCl (aMPT), 2,5dimethoxy-4-iodoamphetamine HCl (DOI), d-amphetamine sulfate and haloperidol were purchased from Sigma/RBI (St. Louis, MO). LY341495 (2S-2-amino-2(1S,2S-2-carboxycycloprop-1-yl)-3-(xanthy-9yl)propanoic acid) was provided by Tocris (Ballwin, MO) and LY379268 was synthesized at Lilly Research Laboratories (Indianapolis, IN). Young male Sprague Dawley (Harlan Indianapolis IN) rats weighing 130–150 gm (N=6–7/treatment group) were group housed for at least four days prior to experimentation with ad lib access to food and water. Experiments were conducted in the morning (i.e. prior to 12:00 noon). During and immediately prior to experimentation efforts were taken to minimize stress by avoiding such things as cage cleaning, feeding, loud noises, and exposure to female rats during the experiment. Animals were injected subcutaneously (s.c.) with drugs or vehicle (sterile water) at the doses indicated (2 ml/kg injection volume) and/or submitted to immobilization stress by placing them in tapered plastic film tubes (decapicones from Braintree Scientific, Braintree, MA). All drug solutions were checked to ensure a neutral pH prior to injection. In one set of experiments ovarriectomized estrogenprimed female rats were examined. Ovariectomized female rats (200–250 g, 65+ days, surgery preformed by vender 7 days prior to shipping) were obtained from Taconic Farms (Germantown, NY). Oestrus phase was induced by implanting subcutaneous 21-day release beta-estadiol slow release pellets (Innovative Research of America, Sarasota, FL) 7 days prior to the experiment. It has been demonstrated this will result in an adaptive change in the estrous cycle such that a prolactin-surge occurs in the afternoon. Thus, these experiments were run in the afternoon in order to examine the drugmediate effects on the prolactin surge of the oestrus cycle. All other experimental conditions were as described above. 2.2. Treatment paradigms and plasma sample collection With all drug injections, animals were rapidly injected using a minimal degree of restraint and returned to their

M.P. Johnson, M. Chamberlain / Neuropharmacology 43 (2002) 799–808

home-cages. Following a typically 30 min drug pretreatment, some animals were subjected to immobilization stress for 15 to 30 min prior to sacrificing. Control groups (unstressed and/or vehicle treated) were left undisturbed in their home cage after injection but sacrificed at an equivalent time post any pretreatment within an experiment to compensate for any circadian rhythm changes in circulating prolactin levels. At the times indicated post the final treatment animals were sacrificed by decapitation and trunk blood collected in EDTA containing borosilicate tubes. Following centrifugation for 15 min at 1000g at 4°C, plasma was isolated and stored at ⫺4°C until assayed. 2.3. Plasma prolactin determinations Prolactin levels were determined using a colorimetric enzyme immunoassay kit from Amersham Life Sciences Inc (Arlington Heights, IL) following the protocol provided. Briefly, blanks, samples or standards were incubated at room temperature for 3 h with anti-serum. Conjugated prolactin was added and the incubation continued for another 30 min prior to four washes with 10 mM phosphate buffered saline pH 7.4 containing 0.2% tween and 0.01% thimerosal. As provided in the purchased kit, the ‘Amdex’ solution (horseradish peroxidase or alkaline phosphatase-based) was added and incubated at room temperature for 30 min. The plate was then washed as before and substrate added and the color change monitored over the next 30 min. When the range of color formation for the standard curve was optimal, the reaction was terminated with 1 M sulfuric acid before reading at 450 mm. Intra-assay variability between duplicate samples was typically less than 5%. However, inter-assay variability in the basal levels of vehicle treated animals was higher (up to a 3- to 5-fold change between some experiments). Thus, care was taken to ensure that the appropriate treatment controls were included in each separate experiment, and all the samples measured within a single assay plate or evenly distributed between multiple plates where necessary. In order to facilitate understanding when contrasting certain of the reported drug effects, results are presented as a percentage of the mean in the vehicle treatment group within that experimental set of animals. 2.4. Statistical analysis Statistical analysis was carried out using the program JMP (SAS Institute, Cary, NC). Typically, an ANOVA or two-way ANOVA was carried out and, if significant, post-hoc comparisons made using a predefined contrast analysis.

801

3. Results 3.1. Modulation of basal and stress-induced prolactin levels by LY379268 The mGlu 2/3 agonist, LY379268, did not produce any alteration in basal levels of plasma prolactin (Figs. 1B, 2A, 3B, 4B). Immobilization stress resulted in a significant increase in plasma prolactin levels with a peak effect seen at 15 min (Fig. 1A, Two-way ANOVA F ⫽ 22.10, p ⬍ 0.0001 for the effect of time). LY379268 (10 mg/kg, s.c.) prevented the stress-induced hyperprolactinemia following both 15 and 30 min of immobilization (two-way ANOVA F ⫽ 16.24, p ⬍ 0.0001 for the interaction of time and drug factors). This effect was dose-dependent in that lower doses of LY379268 were not effective in blocking the stressinduced increase in prolactin levels following 15 min of immobilization (Fig. 1B) 3.2. LY379268 does not alter the oestrus cycleinduced prolactin surge in female rats Ovariectomized rats were implanted with slow-release beta-estradiol pellets as described in the methods in order to consistently induce an afternoon prolactin surge that mimics that seen during the normal oestrus cycle. LY379268 (0.1–10 mg/kg, s.c.) was injected into ovariectomized, estrogen-primed rats and 1 h later sacrificed. As seen in Fig. 1C, at no dose of LY379268 was there a significant alteration of the high plasma prolactin levels seen in these animals. 3.3. LY379268 prevents aMPT-induced hyperprolactinemia Administration of aMPT (250 mg/kg, s.c.) produced a 9-fold increase in plasma prolactin levels (Fig. 2A). Pretreatment with LY379268 (0.1–10 mg/kg, s.c.) resulted in a dose-dependent decrease in the aMPTinduced hyperprolactinemia (p ⬍ 0.02 vs. aMPT, ANOVA F ⫽ 3.34 followed by contrast analysis) (Fig. 2A). LY341495 has been shown to be a selective, systemically active, antagonist of the mGlu2/3 receptors (Kingston et al., 1998). Administration of the mGlu2/3 antagonist alone had no significant effects on either basal or aMPT-induced prolactin levels (Fig. 2B). Pretreatment with the antagonist LY341495 before the agonist LY379268 prevented the mGlu2/3-agonist decrease in aMPT-induced hyperprolactinemia (p ⬍ 0.01, ANOVA F ⫽ 4.91 followed by contrast analysis), consistent with a mGlu2/3 receptor-mediated action. 3.4. Reversal of aMPT-induced hyperprolactinemia with post-treatment of amphetamine or LY379268 To examine whether sufficient levels of tubero-infundibular DA remained after treatment with aMPT to

802

M.P. Johnson, M. Chamberlain / Neuropharmacology 43 (2002) 799–808

Fig. 1. Plasma Prolactin levels in rats following immobilization stress and/or treatment with LY379268. Trunk blood was collected at the time indicated and plasma prolactin levels determined as described in the methods section. Values are mean ± S.E.M. for 6–7 rats per group. A. Male Sprague-Dawley rats (36–40 days old) were injected subcutaneously with 10 mg/kg LY379268 or vehicle 30 min prior to the onset of stress. Rats were subjected to immobilization stress for 0.5, 15 or 30 min prior to sacrifice. Two-way ANOVA (F ⫽ 22.10) indicated a significant interaction between time and treatment (p ⬍ 0.0001). *Indicates significantly different than matched time vehicle/stress group ( p ⬍ 0.001, two-way ANOVA followed by contrast analysis). B. Rats were pretreated with 0.1–10 mg/kg LY379268 30 mins. prior to immobilization stress and sacrificed 15 min later. LY379268 (10 mg/kg) without stress was not different than vehicle. *Indicates significantly different than stress alone group (p ⬍ 0.01, ANOVA F ⫽ 4.37 followed by contrast analysis). C. Female rats (65+ days) that had undergone ovariectomy and estradiol treatment for 7 days (see Methods) were treated with vehicle or LY379268 (0.1–10 mg/kg) and sacrificed 1 h later. An ANOVA followed by contrast analysis indicated none of the LY379268 treated groups were significantly different than vehicle (p ⬎ 0.10).

Fig. 2. Plasma Prolactin levels in male rats following treatment with LY379268 and aMPT. Male rats (36–40 days old) were injected subcutaneously as indicated below and sacrificed 2 h after aMPT treatment. Trunk blood was collected and plasma prolactin levels determined as described in the methods section. Values are mean ± S.E.M. for 6–7 rats per group. A. Rats were pretreated with 0.1–10 mg/kg LY379268 30 min prior to 250 mg/kg aMPT and sacrificed 2 h later. LY379268 (10 mg/kg) alone was not different than vehicle. *Indicates significantly different than matched aMPT (p ⬍ 0.02, ANOVA F ⫽ 3.34 followed by contrast analysis). Values normalized to the vehicle groups average plasma prolactin levels determined on the same assay plate ( 7.1 ± 2.6 ng / ml). B. Vehicle or LY341495 (3 mg/kg) was pretreated 15 min prior to treatment with 10 mg/kg LY379268. Thirty minutes after LY379268 a 250 mg/kg aMPT subcutaneous injection of aMPT was given and sacrificed 2 h later. Brackets indicate the p value for the indicated contrast analysis following a significant ANOVA. Values normalized to the vehicle groups average plasma prolactin levels determined on the same assay plate (57.6 ± 4.8 ng /ml).

effectively modulate prolactin levels, the DA releaser amphetamine was given after DA depletion and plasma prolactin levels determined. As seen in Fig. 3A, high doses of amphetamine (7.5 and 10 mg/kg, s.c.), but not a lower often behaviorally active dose (3 mg/kg), were able to reverse the hyperprolactinemia induced by the tyrosine hydroxylase inhibitor, aMPT (p ⬍ 0.005 vs. aMPT, ANOVA F ⫽ 11.25 followed by contrast analysis).

M.P. Johnson, M. Chamberlain / Neuropharmacology 43 (2002) 799–808

Fig. 3. Plasma Prolactin levels in male rats following treatment with amphetamine or LY379268 and aMPT. Male Sprague-Dawley rats (36–40 days old) were injected subcutaneously with vehicle, amphetamine, LY379268, and/or aMPT and sacrificed at the time indicated. Trunk blood was collected and serum prolactin levels determined as described in the methods section. Values are mean ± S.E.M. for 6–7 rats per group. A. Rats were pretreated with 250 mg/kg aMPT 1 h prior to 3–10 mg/kg amphetamine and sacrificed 1 h later (i.e. 2 h post aMPT treatment). Amphetamine (10 mg/kg) alone was not different than vehicle control. *Indicates significantly different than aMPT control (p ⬍ 0.005, ANOVA F ⫽ 11.25 followed by contrast analysis). Values normalized to the vehicle groups average plasma prolactin levels determined on the same assay plate (52.4 ± 4.1 ng / ml). B. Rats were pretreated with 250 mg/kg aMPT 1 h prior to 0.1–10 mg/kg LY379268 and sacrificed 1 h later (i.e. 2 h post aMPT treatment). LY379268 (10 mg/kg) alone was not different than vehicle control. * Indicates significantly different than aMPT control (p ⬍ 0.002, ANOVA F ⫽ 10.71 followed by contrast analysis). Values normalized to the vehicle groups average plasma prolactin levels determined on the same assay plate (15.7 ± 4.5 ng / ml).

In a similar manner, we investigated whether the mGlu2/3 agonists could reverse aMPT-induced hyperprolactinemia even after the DA synthesis inhibitor had presumably lowered the levels of hypothalamic DA. As seen in Fig. 3B, post-treatment with LY379268 1 h after aMPT significantly reversed the DA depletion-induced hyperprolactinemia. As in the previous experiments, a dose of 10 mg/kg s.c. LY379268 was required to see a

803

Fig. 4. Plasma Prolactin levels in male rats following treatment with LY379268 and haloperidol or DOI. Male Sprague-Dawley rats (36– 40 days old) were injected subcutaneously with vehicle, LY379268, haloperidol and/or DOI and sacrificed at the time indicated. Trunk blood was collected and plasma prolactin levels determined as described in the methods section. Values are mean ± S.E.M. for 6–7 rats per group. A. Rats were pretreated with 10 mg/kg LY379268 30 min prior to 0.75 mg/kg haloperidol and sacrificed 2 h later. Brackets and associated p values represent contrast analysis following a statistically significant ANOVA (p ⬍ 0.0001, F ⫽ 43.3). Values normalized to the vehicle groups average plasma prolactin levels determined on the same assay plate (24.3 ± 5.6 ng / ml). B. LY379268 (0.1–10 mg/kg) was pretreated 30 mins. prior to administration of 12 mg/kg DOI and sacrificed 15 min later. LY379268 (10 mg/kg) alone was not different than vehicle. *Indicates significantly different than DOI alone group ( p ⬍ 0.02, ANOVA F ⫽ 6.04 followed by contrast analysis). Values normalized to the vehicle groups average plasma prolactin levels determined on the same assay plate (65.1 ± 17.2 ng / ml).

significant decrease in plasma prolactin (p ⬍ 0.002 vs. aMPT, ANOVA F ⫽ 10.73 followed by contrast analysis). 3.5. Differential effects of LY379268 on pharmacologically-induced hyperprolactinemia Administration of haloperidol (0.75 mg/kg, s.c.), a dopamine D2 class antagonist, produced a 7-fold

804

M.P. Johnson, M. Chamberlain / Neuropharmacology 43 (2002) 799–808

increase in plasma prolactin levels, consistent with previous demonstrations of D2 antagonist-mediated hyperprolacinemia (Demarest et al., 1984; Horowski and Graf, 1976). Pretreatment with LY379268 (10 mg/kg, s.c.) did not significantly alter the haloperidol-induced hyperprolactinemia (Fig. 4A). Administration of DOI (12 mg/kg s.c.) produced a 7fold increase in plasma prolactin levels (Fig. 4B). Pretreatment with LY379268 (0.1, 1.0, 10 mg/kg, s.c.) produced a dose-dependent blockade of the DOI-induced hyperprolactinemia (p ⬍ 0.02 vs. DOI, ANOVA F ⫽ 6.04 followed by contrast analysis). 4. Discussion It is clear from the data presented in Fig. 1 that the mGlu2/3 receptor-selective agonist, LY379268, can prevent the hyperprolactinemia that results from immobilization stress in young male rats. This was evident with a pretreatment of the mGlu agonist prior to either a 15 min or 30 min exposure to stress and was clearly dose related, showing a significant effect after a 10 mg/kg dose but not at lower doses. One obvious possible explanation for an inhibition of a stress-induced hyperprolactinemia is that LY379268 acts as an anxiolytic thus limiting the normal stress response regulated through higher brain functions. Certainly there is evidence to suggest that Group II mGlu receptor activation might be anxiolytic. For instance LY354740, a close structural analog of LY379268, has been reported to have anxiolytic-like activity in the elevated plus maze, fear-potentiated startle, and in a model of panic attacks (Helton et al., 1998; Monn et al., 1997; Shekhar and Keim, 2000). Furthermore, the inability of LY379268 in this report to significantly alter basal levels of prolactin in young males or to significantly alter the prolactin-surge seen in estrogen-primed, ovariectomized female rats could also be interpreted as consistent with a ‘anxiolytic’ mechanism hypothesis. However, glutamate is known to be a major excitatory neurotransmitter within the neuroendocrine system, modulating directly and indirectly several pituitarysecreted endocrine factors (Abbud and Smith, 1993; Aguilar et al., 1997; Parker and Crowley, 1993; Zelena et al., 1999). Thus, an alternative explanation that should be considered is LY37926, through activation of Group II mGlu receptors, might have the ability to alter hyperprolactinemia by a means other than altering an animals ‘perception’ of an anxious stimulus. Conceivably this might include direct inhibition of prolactin-secreting lactotrophs and/or modulation of the release of one or more prolactin releasing factors, such as DA. Thus, we began to investigate the mechanism by which systemic LY379268 might act to limit stress-induced hyperprolactinemia, by examining various responses to pharmacologically-induced hyperprolactinemia in young male rats.

Previous work has clearly suggested glutamate neurotransmission, through ionotropic glutamate receptors, plays a key role in the modulation of the hypothalamic arcuate nucleus. For instance, AMPA has state-dependent effects illustrated by a decrease in basal prolactin levels in young male and female rats, but does not appear to alter the hyperprolactinemia-induced by pretreatment with D2 antagonists (Gonzalez et al., 1999b). Also, the AMPA receptor antagonist NBQX caused a delayed increase in prolactin levels while AMPA had no direct effects in vitro on hemi-pituitaries. Thus, Gonzalez and co-workers concluded that AMPA acts indirectly to decrease prolactin levels by stimulating tubero-infundibular DA release within the arcuate nucleus (Gonzalez et al., 1999b, 2000). The potential mechanism of actions for another ionotropic channel receptor family, the NMDA receptors, are more difficult to ascertain since they have been associated with both a direct effect on pituitary lactotropes in vitro (Bhat et al., 1995; Login, 1990) and an indirect control of prolactin in vivo that is age- and cycle-dependent (Wagner et al., 1993). However, it has been shown that NMDA channel blockers such as MK-801 decrease prolactin via an increase in DA release (Aguilar et al., 1997; Pechnick et al., 1989; Wagner et al., 1993). The present work clearly illustrates that LY379268, a subtype-selective mGlu2/3 receptor agonist (Monn et al., 1999), can decrease the hyperprolactinemia under some but not all conditions examined. Specifically, similar doses of LY379268 appear to prevent or reverse the elevated prolactin levels seen in male rats resulting from immobilization stress, inhibition of catecholamine synthesis with aMPT, or treatment with indirect acting agents such as the 5-HT2A agonist DOI. In contrast, this mGlu2/3 agonist does not alter basal prolactin levels in male rats. Furthermore, neither the haloperidol-induced hyperprolactinemia nor the prolactin-surge seen in estrogen-primed ovariectomized rats was significantly altered by LY379268 treatment. As with NMDA and AMPA, this state-dependent decrease in plasma prolactin levels seen with mGlu2/3 stimulation suggests an indirect mechanism for LY379268. Certainly one possible indirect means to influence prolactin levels could be a mGlu2/3-mediated decreases in putative circulating prolactin releasing factors. This possibility should certainly be investigated further as the scientific understanding of any prolactin releasing factors increases. However, it seems as likely that, like the ionotropic glutamate receptors, mGlu2/3 activation modulates plasma prolactin levels by increasing tuberoinfundibular DA release into the median eminence. One argument against an DA-mediated mechanism is that there are reports of low to moderate staining in the pituitary anterior lobe with an mGlu2/3 specific antibody (Petralia et al., 1996), suggesting a possible direct inhibition of prolactin release from lactotrophs. However,

M.P. Johnson, M. Chamberlain / Neuropharmacology 43 (2002) 799–808

LY379268 failed to alter the hyperprolactinemia induced with a D2 antagonist, that acts directly to block the DA inhibitory action on pituitary lactotrophs. Furthermore, a direct inhibitory action on lactropophs might be expected to more dramatically alter basal prolactin levels or the elevated levels associated with proestrus in female rats, which the mGlu2/3 agonists did not. Rather, the mGlu2/3 agonist only modulated hyperprolactinemia induced indirectly by altering DA release, e.g. the partial depletion of catecholamines with aMPT. The ability of the selective mGlu2/3 antagonist LY341495 (Kingston et al., 1998) to prevent the effects of LY379268, and the dose-related effects with activity at efficacious doses in other animal models linked to Group II activity (Bond et al., 1999, 2000; Cai et al., 1999; Cartmell et al., 1999, 2000a, b, c) strongly suggests a mGlu2/3 receptor-mediated activity. However, the mechanism by which activation of these typically inhibitory receptors increases the release of tuberoinfundibular DA may not be immediately evident. Certainly one plausible explanation that is consistent with both the present data and literature reports would be that mGlu2 and/or 3 negatively modulate the release of gamma-aminobutyric acid (GABA) within the arcuate nucleus (Fig. 5).

805

It is clear that several mGlu receptors, including the Group II mGlu receptors, are expressed in the arcuate nucleus of the hypothalamus (Ohishi et al., 1995, 1993b; Petralia et al., 1996; Van den Pol, 1994; Van den Pol et al., 1994, 1995). A Group II-selective (i.e. mGlu2 and 3 receptors) antibody shows moderate labeling within the arcuate nucleus (Petralia et al., 1996), although it has been reported that there is relatively little staining with an mGlu2-selective antibody (Ohishi et al., 1998). Most recently Tamaru and co-workers (Tamaru et al., 2001) helped to clarify the relative distribution of Group II mGlu receptors, by examining a somewhat preferring mGlu3 antibody and contrasting results in wild type and mGlu2 knock-out mice. Tamaru found that mGlu3 was present within a number of hypothalamic nuclei including the arcuate and the entire periventircular nucleus while mGlu2 is evident only in the anteroventral portion of the periventricular nucleus. Similarly, mRNA for the mGlu3 receptor (Ohishi et al., 1993b) has been found within the periventricular nucleus while no mGlu2 mRNA was apparent (Ohishi et al., 1993a). Taken together, this suggests that there are mGlu3 receptors within the arcuate and periventricular hypothalamic nuclei. Futhermore, the presence of mRNA might suggest presynaptically localized Group II mGlu receptors

Fig. 5. Proposed mechanism by which mGlu2/3 agonists effect the hypothalamic-tubero-infundibular-pituitary system. Dopamine released into the hypophysial portal system stimulates D2 receptors located on lactotrophs within the anterior pituitary. In the absence of this D2-mediated dopaminergic tone, lactotrophs will release prolactin into the general circulation. The soma of the DA containing axons are located within certain nuclei of the hypothalamus, such as the arcuate nucleus, receiving both excitatory (glutamatergic) and inhibitory (GABAergic) inputs. Serotonergic agents such as the 5-HT2A agonist DOI increase plasma prolactin levels by increasing inhibitory tone on dopamergic neurons and those decrease release of the inhibitory DA into the hypophysial portal system. One possible explanation for the current results is that mGlu2/3 (most likely mGlu3) receptors are located presynaptically on GABAergic projections within the arcuate nucleus. Stimulation by the agonist LY379268, would result in a decrease in the GABAergic tone on tubero-infundibular DA neurons and thus increase the release of DA. The increased DA resulting in an inhibition of prolactin release from the anterior pituitary.

806

M.P. Johnson, M. Chamberlain / Neuropharmacology 43 (2002) 799–808

on short projection axons or interneurons feeding into the arcuate nucleus. Certainly, mGlu3 has been implicated in the presynaptic modulation of GABAergic axons from reticular thalamic nuclei (Salt and Turner, 1998) and positively identified using double-labeling techniques on the presynaptic terminals of GABAergic synapses (Tamaru et al., 2001) in the past. There is much evidence suggesting GABAergic input onto the DA neurons in the arcuate nucleus mediates the degree of DA release in the median eminence. This is evidenced by the fact that glutamate decarboxylase-positive (an enzyme used in the synthesis of GABA) terminals contact dopaminergic neurons in the periventricular-arcuate nucleus (Tappaz et al., 1985). Also GABAergic agonists infused into this nucleus result in increased prolactin levels (Ferreira et al., 1998). Furthermore, the GABAa antagonist SR95531 decreased plasma prolactin and increased median eminence 3,4-dihydroxyphenylacetic acid (DOPAC, a metabolite of released DA) (Wagner et al., 1994). One implication of this antagonist effect is that some inhibitory GABAergic tone on postsynaptic GABAa receptors regulating DA arcuate neurons under basal conditions. Similarly, GABAergic tone has been implicated in the photoperiodic regulation of prolactin and in the feedback inhibition induced by hyperprolactinemia (Boissin-Agasse et al., 1991). Thus, the propensity of evidence suggests tubero-infundibular DA release is under control from GABAergic input and alterations in this system can alter the levels of circulating prolactin. Furthermore there is substantial precedence for presynaptic mGlu receptors acting to inhibit neurotransmitter release in other brain regions. For instance, mGlu2 has been localized to asymmetrical (probably GABAergic) terminals in the basal ganglia and hippocampus (Liu et al., 1998; Shigemoto et al., 1997). Furthermore, Salt and Turner (1998) have reported that there is a mGlu2/3 receptor-mediated decrease in inhibitory sensory input within the ventrobasal thalamic nuclei. Also, it has been found repeatedly that activation of mGlu2/3 receptors results in decreases in not only glutamate but also GABA release (for review see Cartmell and Schoepp (2000) in microdialysis and in vitro slice release studies in several brain nuclei. Thus, a common feature of the Group II mGlu receptors appears to be their action as presynaptic neurotransmitter release inhibitors. The ability of LY379268 to alter the DOI-induced transient hyperprolactinemia is also consistent with a GABAergic-mediated modulation of prolactin levels. It has been suggested that Serotonin (5-HT) acts indirectly on DA-containing arcuate nucleus neurons through the modulation of GABA release. This is evidenced by the ability of 5-HT releasing agents (i.e. fenfluramine) and 5-HT2a agonists (DOI) to not only increase prolactin levels but also increase the synthesis of GABA within

the arcuate nucleus (Afione et al., 1990; Coccaro et al., 1996; Fuller et al., 1976; Liang and Pan, 2000; Quattrone et al., 1979; Scarduelli et al., 1985). Both of these actions were blocked by pretreatment with the 5-HT2a antagonist, ritanserin (Coccaro et al., 1996). Thus, the ability of LY379268 to inhibit DOI-induced hyperprolactinemia may be a reflection of an inhibition of 5-HT2a-stimulated GABA release (Fig. 5). Given the localization of Group II mGlu receptors mentioned above, the data suggesting an indirect mechanism in mGlu2/3 agonists actions presented here, the importance of the GABAergic system in the control of prolactin levels, and the precedence for inhibition of GABA release with this Group of mGlu receptors, it is plausible that mGlu2 and/or 3 are located presynaptically on GABAergic interneurons within the arcuate nucleus. Thus, activation by mGlu2/3 agonist might be expected to decrease the inhibitory tone on tuberoinfundibular DA neurons, increasing the levels of medial eminence DA and as such limiting the plasma levels of prolactin. At first glance, one argument against this hypothesis might be the ability of LY379268 to reverse the DAdepletion induced hyperprolactinemia following aMPT (a catecholamine synthesis inhibitor) treatment. Indeed, if the levels of dopamine were completely depleted by this treatment of aMPT then one would not expect that any indirect mechanism that is dependent upon regulating DA release would prevent aMPT’s action. Yet, it has been shown that this dosing paradigm with aMPT does not completely eliminate the DA within certain brain nuclei (Salhoff and Schoepp, personal communication). Most importantly, in the present work high doses of amphetamine could reverse aMPT hyperprolactinemia, even when dosing with the DA releasing agent is delayed until a maximal effect of aMPT was evident. One interpretation is that sufficient levels of hypothalamic DA remain, following partial depletion of catecholamines, to be released by amphetamine and thus, decrease prolactin levels. Certainly this would explain the need for higher doses of amphetamine than are required for many behavioral effects. Furthermore, the fact that a similar post-treatment with LY379268 or amphetamine is as efficacious as a pretreatment with the mGlu2/3 agonist is consistent with LY379268 acting through a DA-dependent mechanism despite its ability to reverse aMPT-induced hyperprolactinemia. It should be noted that one possibly important aspect of the present study is that the mGlu2/3 agonist clearly does not induce hyperprolactinemia itself. Previous reports from work in rodents have suggested that mGlu2/3 agonists might be efficacious antipsychotics, since they alter prefrontal cortex DA release and can reverse some of the behavioral actions of hallucinogens such as phenylcyclidine (PCP) (Cartmell et al., 1999, 2000a, b, d; Moghaddam and Adams, 1998). A number of DA antagonists, such as haloperidol, are classified as

M.P. Johnson, M. Chamberlain / Neuropharmacology 43 (2002) 799–808

typical antipsychotics. These agents, although clearly efficacious against at least some of the symptoms of psychosis, induce a significant and sustained hyperprolactinemia that can limit compliance and the clinical utility of these agents. Thus, if the present work in rodents with the mGlu2/3 agonists translates to similar findings in clinical trials, the mGlu2/3 agonists would not be anticipated to induce any side effects associated with hyperprolactinemia. To summarize, these results show a state-dependent mGlu2/3-mediated decrease in hyperprolactinemia. The present results suggest a mechanism for the inhibition of stress-induced hyperprolactinemia that is independent of any anxiolytic actions or a direct inhibition of lactotrophs within the pituitary. Rather, these results are more consistent with the hypothesis that LY379268 indirectly decreases hyperprolactinemia by means of regulating tubero-infundibular DA release. One plausible explanation is that treatment with LY379268 involves a modulation of arcuate nucleus DA neurons by inhibiting GABA release onto these neurons. A more definitive test of this hypothesis will have to await more selective and systemically active pharmacological tools for the metabotropic glutamate receptors, as well as a better understanding of the actions of mGlu2/3 receptors within the various areas of control for the neuroendocrine system.

Acknowledgements The authors would like to thank Laura Nisenbaum for helpful discussions of the results and editorial comments of the manuscript.

References Abbud, R., Smith, M.S., 1993. Altered luteinizing hormone and prolactin responses to excitatory amino acids during lactation. Neuroendocrinology 58, 454–464. Afione, S., Duvilanski, B., Seilicovich, A., Lasaga, M., Diaz, M.C., Debeljuk, L., 1990. Effects of serotonin on the hypothalamic-pituitary GABAergic system. Brain Research Bulletin 25, 245–249. Aguilar, E., Tena-Sempere, M., Aguilar, R., Gonzalez, D., Pinilla, L., 1997. Interactions between N-methyl-D-aspartate, nitric oxide and serotonin in the control of prolactin secretion in prepubertal male rats. European Journal of Endocrinology 137, 99–106. Ben-Jonathan, N., 1985. Dopamine: a prolactin-inhibiting hormone. Endocrine Review 6, 564–589. Bhat, G.K., Mahesh, V.B., Chu, Z.W., Chorich, L.P., Zamorano, P.L., Brann, D.W., 1995. Localization of the N-methyl-D-aspartate R1 receptor subunit in specific anterior pituitary hormone cell types of the female rat. Neuroendocrinology 62, 178–186. Boissin-Agasse, L., Tappaz, M., Roch, G., Gril, C., Boissin, J., 1991. Morphofunctional evidence for the involvement of hypothalamic dopaminergic and GABAergic neurons in the mechanisms of photoperiod-dependent prolactin release in the mink. Neuroendocrinology 53, 537–542. Bond, A., Jones, N.M., Hicks, C.A., Whiffin, G.M., Ward, M.A.,

807

O’Neill, M.F., Kingston, A.E., Monn, J.A., Ornstein, P.L., Schoepp, D.D., Lodge, D., O’Neill, M.J., 2000. Neuroprotective effects of LY379268, a selective mGlu2/3 receptor agonist: investigations into possible mechanism of action in vivo. Journal of Pharmacology and Experimental Therapeutics 294, 800–809. Bond, A., Ragumoorthy, N., Monn, J.A., Hicks, C.A., Ward, M.A., Lodge, D., O’Neill, M.J., 1999. LY379268, a potent and selective Group II metabotropic glutamate receptor agonist, is neuroprotective in gerbil global, but not focal, cerebral ischaemia. Neuroscience Letters 273, 191–194. Cai, Z., Xiao, F., Fratkin, J.D., Rhodes, P.G., 1999. Protection of neonatal rat brain from hypoxic-ischemic injury by LY379268, a Group II metabotropic glutamate receptor agonist. Neuroreport 10, 3927–3931. Cartmell, J., Monn, J.A., Schoepp, D.D., 1999. The metabotropic glutamate 2/3 receptor agonists LY354740 and LY379268 selectively attenuate phencyclidine versus d-amphetamine motor behaviors in rats. Journal of Pharmacology & Experimental Therapeutics 291, 161–170. Cartmell, J., Monn, J.A., Schoepp, D.D., 2000a. Attenuation of specific PCP-evoked behaviors by the potent mGlu2/3 receptor agonist, LY379268 and comparison with the atypical antipsychotic, clozapine. Psychopharmacology (Berl) 148, 423–429. Cartmell, J., Monn, J.A., Schoepp, D.D., 2000b. The mGlu(2/3) receptor agonist LY379268 selectively blocks amphetamine ambulations and rearing. European Journal of Pharmacology 400, 221–224. Cartmell, J., Monn, J.A., Schoepp, D.D., 2000c. Tolerance to the motor impairment, but not to the reversal of PCP-induced motor activities by oral administration of the mGlu2/3 receptor agonist, LY379268. Naunyn Schmiedebergs Arch Pharmacol 361, 39–46. Cartmell, J., Perry, K.W., Salhoff, C.R., Monn, J.A., Schoepp, D.D., 2000d. The potent, selective mGlu2/3 receptor agonist LY379268 increases extracellular levels of dopamine,3,4-dihydroxyphenylacetic acid, homovanillic acid, and 5-hydroxyindole-3-acetic acid in the medial prefrontal cortex of the freely moving rat. Journal of Neurochemistry 75, 1147–1154. Cartmell, J., Schoepp, D.D., 2000. Regulation of neurotransmitter release by metabotropic glutamate receptors. Journal of Neurochemistry 75, 889–907. Coccaro, E.F., Kavoussi, R.J., Oakes, M., Cooper, T.B., Hauger, R., 1996. 5-HT2a/2c receptor blockade by amesergide fully attenuates prolactin response to d-fenfluramine challenge in physically healthy human subjects. Psychopharmacology (Berl) 126, 24–30. Conn, P., Freeman, M. (Eds.), 2000. Neuroendocrinology in Physiology and Medicine. Totowa, New Jersey, Humana Press. Demarest, K.T., Riegle, G.D., Moore, K.E., 1984. Pharmacological manipulation of anterior pituitary dopamine content in the male rat: relationship to serum prolactin concentration and lysosomal enzyme activity. Endocrinology 115, 493–500. Ferreira, S.A., Browning, D.A., Scott, C.J., Kuehl, D.E., Jackson, G.L., 1998. Effect of infusing gamma-aminobutyric acid receptor agonists and antagonists into the medial preoptic area and ventromedial hypothalamus on prolactin secretion in male sheep. Endocrine 9, 303–312. Fuller, R.W., Perry, K.W., Clemens, J.A., 1976. Elevation of 3, 4dihydroxyphenylacetic acid concentrations in rat brain and stimulation of prolactin secretion by fenfluramine: evidence for antagonism at dopamine receptor sites. Journal of Pharmacy and Pharmacology 28, 643–644. Gonzalez, L.C., Pinilla, L., Tena-Sempere, M., Aguilar, E., 1999a. Regulation of growth hormone secretion by alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid receptors in infantile, prepubertal, and adult male rats. Endocrinology 140, 1279–1284. Gonzalez, L.C., Pinilla, L., Tena-Sempere, M., Aguilar, E., 1999b. Role of alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid receptors in the control of prolactin, growth hormone and gonadotropin secretion in prepubertal rats. Journal of Endocrinology 162, 417–424. Gonzalez, L.C., Pinilla, L., Tena-Sempere, M., Aguilar, E., 2000. Regulation of prolactin secretion by alpha-amino-3-hydroxy-5-

808

M.P. Johnson, M. Chamberlain / Neuropharmacology 43 (2002) 799–808

methylisoxazole-4-propionic acid receptors in male rats. Journal of Endocrinology 166, 669–675. Helton, D.R., Tizzano, J.P., Monn, J.A., Schoepp, D.D., Kallman, M.J., 1998. Anxiolytic and side-effect profile of LY354740: a potent, highly selective, orally active agonist for Group II metabotropic glutamate receptors. Journal of Pharmacology and Experimental Therapeutics 284, 651–660. Horowski, R., Graf, H.J., 1976. Influence of dopaminergic agonists and antagonists on serum prolactin concentrations in the rat. Neuroendocrinology 22, 273–286. Kingston, A.E., Ornstein, P.L., Wright, R.A., Johnson, N.G., Mayne, N.G., Burnett, J.P., Belagaje, R., Wu, S., Schoepp, D.D., 1998. LY341495 is a nanomolar potent and selective antagonist of Group II metabotropic glutamate receptors. Neuropharmacology 37, 1–12. Liang, S.L., Pan, J.T., 2000. An endogenous serotonergic rhythm acting on 5-HT(2A) receptors may be involved in the diurnal changes in tuberoinfundibular dopaminergic neuronal activity and prolactin secretion in female rats. Neuroendocrinology 72, 11–19. Liu, X.B., Munoz, A., Jones, E.G., 1998. Changes in subcellular localization of metabotropic glutamate receptor subtypes during postnatal development of mouse thalamus. Journal of Comparative Neurology 395, 450–465. Login, I.S., 1990. Direct stimulation of pituitary prolactin release by glutamate. Life Science 47, 2269–2275. Moghaddam, B., Adams, B.W., 1998. Reversal of phencyclidine effects by a Group II metabotropic glutamate receptor agonist in rats. Science 281, 1349–1352. Monn, J.A., Valli, M.J., Massey, S.M., Hansen, M.M., Kress, T.J., Wepsiec, J.P., Harkness, A.R., Grutsch, J.L. Jr, Wright, R.A., Johnson, B.G., Andis, S.L., Kingston, A., Tomlinson, R., Lewis, R., Griffey, K.R., Tizzano, J.P., Schoepp, D.D., 1999. Synthesis, pharmacological characterization, and molecular modeling of heterobicyclic amino acids related to (+)-2-aminobicyclo[3.1.0] hexane-2,6dicarboxylic acid (LY354740): identification of two new potent, selective, and systemically active agonists for Group II metabotropic glutamate receptors. Journal of Medical Chemistry 42, 1027–1040. Monn, J.A., Valli, M.J., Massey, S.M., Wright, R.A., Salhoff, C.R., Johnson, B.G., Howe, T., Alt, C.A., Rhodes, G.A., Robey, R.L., Griffey, K.R., Tizzano, J.P., Kallman, M.J., Helton, D.R., Schoepp, D.D., 1997. Design, synthesis, and pharmacological characterization of (+)-2-aminobicyclo[3.1.0]hexane-2,6-dicarboxylic acid (LY354740): a potent, selective, and orally active Group 2 metabotropic glutamate receptor agonist possessing anticonvulsant and anxiolytic properties. Journal of Medicinal Chemistry 40, 528–537. Moore, K.E., Demarest, K.T., Johnston, C.A., 1980. Influence of prolactin on dopaminergic neuronal systems in the hypothalamus. Federation Proceedings 39, 2912–2916. Ohishi, H., Akazawa, C., Shigemoto, R., Nakanishi, S., Mizuno, N., 1995. Distributions of the mRNAs for L-2-amino-4-phosphonobutyratesensitive metabotropic glutamate receptors, mGluR4 and mGluR7, in the rat brain. Journal of Comparative Neurology 360, 555–570. Ohishi, H., Neki, A., Mizuno, N., 1998. Distribution of a metabotropic glutamate receptor, mGluR2, in the central nervous system of the rat and mouse: an immunohistochemical study with a monoclonal antibody. Neuroscience Research 30, 65–82. Ohishi, H., Shigemoto, R., Nakanishi, S., Mizuno, N., 1993a. Distribution of the messenger RNA for a metabotropic glutamate receptor, mGluR2, in the central nervous system of the rat. Neuroscience 53, 1009–1018. Ohishi, H., Shigemoto, R., Nakanishi, S., Mizuno, N., 1993b. Distribution of the mRNA for a metabotropic glutamate receptor (mGluR3) in the rat brain: an in situ hybridization study. Journal of Comparative Neurology 335, 252–266. Parker, S.L., Crowley, W.R., 1993. Stimulation of oxytocin release in the lactating rat by central excitatory amino acid mechanisms: evidence for specific involvement of R,S-alpha-amino-3-hydroxy-

5-methylisoxazole-4-propionic acid-sensitive glutamate receptors. Endocrinology 133, 2847–2854. Pechnick, R.N., George, R., Poland, R.E., 1989. Characterization of the effects of the acute and repeated administration of MK-801 on the release of adrenocorticotropin, corticosterone and prolactin in the rat. European Journal of Pharmacology 164, 257–263. Petralia, R.S., Wang, Y.X., Niedzielski, A.S., Wenthold, R.J., 1996. The metabotropic glutamate receptors, mGluR2 and mGluR3, show unique postsynaptic, presynaptic and glial localizations. Neuroscience 71, 949–976. Quattrone, A., Schettini, G., Di Renzo, G.F., Preziosi, P., 1979. Pharmacological evidence of an interaction between serotonergic and dopaminergic neurons in the control of prolactin secretion in male rats. Archives Internationales de Pharmacodynamie et de Therapie 238, 42–49. Rubin, R.T., Poland, R.E., O’Connor, D., Gouin, P.R., Tower, B.B., 1976. Selective neuroendocrine effects of low-dose haloperidol in normal adult men. Psychopharmacologia 47, 135–140. Salt, T.E., Turner, J.P., 1998. Modulation of sensory inhibition in the ventrobasal thalamus via activation of Group II metabotropic glutamate receptors by 2R,4R-aminopyrrolidine-2,4-dicarboxylate. Experimental Brain Research 121, 181–185. Scarduelli, C., Mattei, A.M., Brambilla, G., Zavaglia, C., Adelasco, P., Cavioni, V., Ferrari, C., 1985. Effect of fenfluramine oral administration on serum prolactin levels in healthy and hyperprolactinemic women. Gynecological and Obstetric Investigation 19, 92–96. Schoepp, D.D., Jane, D.E., Monn, J.A., 1999. Pharmacological agents acting at subtypes of metabotropic glutamate receptors. Neuropharmacology 38, 1431–1476. Shekhar, A., Keim, S.R., 2000. LY354740, a potent Group II metabotropic glutamate receptor agonist prevents lactate-induced panic-like response in panic-prone rats. Neuropharmacology 39, 1139–1146. Shigemoto, R., Kinoshita, A., Wada, E., Nomura, S., Ohishi, H., Takada, M., Flor, P.J., Neki, A., Abe, T., Nakanishi, S., Mizuno, N., 1997. Differential presynaptic localization of metabotropic glutamate receptor subtypes in the rat hippocampus. Journal of Neuroscience 17, 7503–7522. Tamaru, Y., Nomura, S., Mizuno, N., Shigemoto, R., 2001. Distribution of metabotropic glutamate receptor mGluR3 in the mouse CNS: Differential localization relative to pre- and postsynaptic sites. Neuroscience 106, 481–503. Tappaz, M.L., Bosler, O., Paut, L., Berod, A., 1985. Glutamate decarboxylase-immunoreactive boutons in synaptic contacts with hypothalamic dopaminergic cells: a light and electron microscopy study combining immunocytochemistry and radioautography. Neuroscience 16, 111–122. Van den Pol, A.N., 1994. Metabotropic glutamate receptor mGluR1 distribution and ultrastructural localization in hypothalamus. Journal of Comparative Neurology 349, 615–632. Van den Pol, A.N., Kogelman, L., Ghosh, P., Liljelund, P., Blackstone, C., 1994. Developmental regulation of the hypothalamic metabotropic glutamate receptor mGluR1. Journal of Neuroscience 14, 3816–3834. Van den Pol, A.N., Romano, C., Ghosh, P., 1995. Metabotropic glutamate receptor mGluR5 subcellular distribution and developmental expression in hypothalamus. Journal of Comparative Neurology 362, 134–150. Wagner, E.J., Goudreau, J.L., Moore, K.E., Lookingland, K.J., 1994. GABAergic regulation of tuberoinfundibular dopaminergic neurons in the male rat. Brain Research Bulletin 659, 194–200. Wagner, E.J., Moore, K.E., Lookingland, K.J., 1993. Sexual differences in N-methyl-D-aspartate receptor-mediated regulation of tuberoinfundibular dopaminergic neurons in the rat. Brain Research Bulletin 611, 139–146. Zelena, D., Makara, G.B., Jezova, D., 1999. Simultaneous blockade of two glutamate receptor subtypes (NMDA and AMPA) results in stressor-specific inhibition of prolactin and corticotropin release. Neuroendocrinology 69, 316–323.