INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

MAGNOCELLULAR NEURONS within the supraoptic and paraventricular nuclei of the hypothalamus, along with their axonal projections via the infundibular stalk to the neurohypophysis, form the final link in the neural network that controls arginine vasopressin (AVP) secretion. This neurosecretory system receives an elaborate array of neural inputs, including a dense catecholaminergic innervation that is predominantly noradrenergic, but also includes a dopaminergic component (3, 6).

These dopaminergic pathways influence AVP secretory rate. Most of the evidence supports a stimulatory action by dopamine (7, 15, 20, 31), but data showing no change or inhibition also exist (23, 25, 27). Specifically, intracerebroventricular injection of dopamine (7, 15, 20) or direct injection of dopamine into supraoptic or paraventricular nuclei (17) in euvolemic or water-loaded rats provokes antidiuresis and increases plasma AVP levels. This rise in AVP is inhibited by nonselective dopamine receptor antagonism (7).

Dopamine has been implicated in the osmotic release of AVP in rats (31, 40). In addition, Brooks and Claybaugh (2) have shown that haloperidol blocks the increase in plasma AVP associated with infusion of angiotensin II in dehydrated dogs, consistent with facilitation of a nonosmotic mechanism for AVP release. Stimulation of AT₁ angiotensin receptors within the brain potentiates dopamine release (11). The central loci involved in AVP release are replete with AT₁ receptors (14).

Both D₁ and D₂ receptor subtypes have been identified in the hypothalamo-neurohypophysial system (1, 5). For the most part, D₁ receptor activation has been associated with stimulation and D₂ receptor with inhibition of AVP release (28). Most recently, D_1A receptor mRNA has been detected and D_1A receptors have been identified on neurophysin-expressing cells from the supraoptic nucleus in culture (8, 18). However, selective dopamine agonists and antagonists have been used in only a limited number of studies on AVP secretion (31, 32, 41). Although noradrenergic modulation of angiotensin II-stimulated AVP release has been studied (24, 37), the dopamine receptor subtype mediating the response to angiotensin II has not been identified.

The present studies used rat hypothalamo-neurohypophysial explants to characterize pharmacologically the dopamine receptor subtype that induces AVP release and to identify the selective dopaminergic actions that regulate angiotensin II-stimulated AVP secretion.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Experiments were performed on male Long-Evans rats weighing 125-150 g. The rats were housed at constant temperature with a 12:12-h light-dark cycle. They were given free access to water and standard rodent chow. All procedures were reviewed and approved by the institutional Animal Investigation Committee and were in compliance with the National Institutes of Health guidelines.

Dissection and culture of explants. Explants of the hypothalamo-neurohypophysial system were dissected as previously described (30, 34). Each explant contained supraoptic nuclei with intact axonal projections to the neural lobe. The arcuate, suprachiasmatic, preoptic, and ventromedial nuclei were present as well as the organum vasculosum of the lamina terminalis and the intermediate lobe. The paraventricular nuclei and subfornical organ were absent.

Sampling procedures. Protocols were performed 48 h after dissection and follow the sampling techniques reported earlier (34). All experiments were performed in the presence of the peptidase inhibitor bacitracin, 0.05 mg/ml to prevent degradation of released AVP (30). Immediately before the protocol, ascorbic acid was added to each well to bring the final concentration to 10 µM to retard the oxidation of dopamine and its analogs.

Protocols. The following protocols were performed: 1) dose-response curves for 0.01-10 µM dopamine (nonselective D₁/D₂ agonist), 0.01-10 µM SKF-38393 (selective D₁ agonist), and 0.01 nM to 1 µM apomorphine (D₂  D₁ agonist); 2) nonselective dopaminergic antagonism of dopamine-induced AVP release by 1 µM haloperidol; 3) antagonism of maximally stimulating doses of each agonist (1 µM angiotensin II, 5 µM dopamine, 1 µM SKF-38393, and 0.1 µM apomorphine) using the selective D₁ antagonist, 1 µM SCH-23390, or the D₂ antagonist, 1.5 µM domperidone, and the nonselective adrenergic antagonist, 1 µM phenoxybenzamine; 4) angiotensin receptor blockade of each agonist with the nonselective antagonist 1 µM saralasin, the AT₁ antagonist, 0.5 µM losartan, or the AT₂ antagonist, 0.5 µM CGP-42112A; 5) antagonism of 1 µM SKF-38393 or 1 µM angiotensin II with 1 µM SCH-23390 or the selective 1 µM ₁-prazosin alone or in combination; 6) the combination of submaximally stimulating concentrations of 0.05 µM dopamine or 0.01 µM SKF-38393 with a subthreshold dose of 0.01 µM angiotensin II; and 7) dopaminergic or angiotensinergic receptor inhibition of the combined effect of submaximal angiotensin II and dopamine agonists. Doses of the antagonists were chosen based on reported values for inhibition constants.

Analytic methods. AVP content of the medium was measured by radioimmunoassay using methods reported earlier from our laboratory (30). All standards and samples were assayed in duplicate; samples were diluted 1:100 with buffer and assayed directly. Standards were prepared with purified AVP (Ferring, Malmo, Sweden). The tracer was [¹²⁵I]iodotyrosyl AVP (Amersham, Arlington Heights, IL). Anti-AVP serum no. 2849 was used at a final dilution of 1:3.6 × 10⁵. Assay buffer contained 0.15 M sodium phosphate, 0.01 M sodium EDTA, 0.1 g/100 ml sodium azide, and 0.1 g/100 ml bovine serum albumin (ICN-Miles, Irvine, CA) at pH 7.4. Assay sensitivity and cross-reactivities have been reported previously (30).

Statistics. All comparisons between basal and test period release rates were performed on the absolute release rates. Because basal AVP release rates varied among the groups of explants, the release of AVP during the test hour was normalized as the percentage of AVP released during the preceding basal hour for the same explant. Each explant acted as its own control.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Overall basal AVP release rate at 48 h was 79 ± 2 pg · explant¹ · h¹ (n = 418). Basal medium osmolality was 305 ± 1 mosmol/kgH₂O and did not change by the end of the test period (307 ± 2 mosmol/kgH₂O). AVP release did not vary during time control experiments with standard medium administered during both basal and test periods [basal, 100 ± 25 vs. test, 115 ± 6% · explant¹ · h¹ (P > 0.05, n = 7)].

Dose-response curves for the dopamine agonists are shown in Fig. 1. AVP release was significantly increased by concentrations of dopamine or SKF-38393 >1 µM or apomorphine >10 nM. Nonselective dopaminergic receptor blockade with haloperidol inhibited dopamine-induced AVP secretory activity (Fig. 2). The vehicle for haloperidol, 0.0038% methanol, did not alter AVP release [basal (no methanol) 100 ± 35 vs. test (methanol vehicle) 128 ± 28% · explant¹ · h¹]. The dose-response relationship to dopamine was unchanged by the presence of methanol [compare Figs. 1 (without methanol) and 2 (with methanol)]. Haloperidol alone did not alter basal AVP secretory rate [basal, 100 ± 31 vs. haloperidol, 86 ± 20% · explant¹ · h¹ (n = 8; P > 0.05)].

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Test period arginine vasopressin (AVP) secretory rate by explants in response to increasing doses of dopamine (; n = 6, 6, 6, and 7), SKF-38393 (; n = 6, 6, 6, 8, 6, and 6), and apomorphine (; n = 6, 7, 8, 8, 6, and 6). Values are means ± SE. *P < 0.05 vs. basal release rate by paired t-test. [Agonist], agonist concentration.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   AVP release by explants in response to dopamine (; n = 6, 8, 8, and 7) or dopamine in the presence of 1 µM haloperidol (; n = 8, 6, 7, and 7). Methanol, the vehicle for haloperidol, was present during the test period in all groups. Data are means ± SE. *P < 0.05 vs. basal by paired t-test. +P < 0.05 vs. same concentration of dopamine ([Dopamine]) alone by ANOVA.

Figure 3 shows the effect of the selective D₁ and D₂ dopamine receptor antagonists and the nonselective adrenergic blocker on dopamine-induced AVP release. Significant inhibition occurred only with SCH-23390. The D₁ dopaminergic blocker also blocked AVP release by maximally stimulating doses of SKF-38393 or apomorphine, as depicted in Table 1. Domperidone also failed to inhibit AVP release by 1 µM SKF-38393 (369 ± 41) vs. SKF-38393 with domperidone [397 ± 72% · explant¹ · h¹ (P > 0.05, n = 6 and 6, respectively)] or 0.1 µM apomorphine (392 ± 67) vs. apomorphine with domperidone [309 ± 33% · explant¹ · h¹ (P > 0.05, n = 6 and 7, respectively)]. SCH-23390 alone did not change AVP release [basal, 100 ± 21 vs. SCH-23390, 165 ± 25% · explant¹ · h¹ (P > 0.05, n = 7)] nor did domperidone [basal, 100 ± 10 vs. domperidone, 104 ± 15% · explant¹ · h¹ (P > 0.05, n = 6)].

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Antagonism of dopamine (DA, 5 µM, n = 6)-induced AVP release with SCH-23390 (1 µM SCH; n = 7), domperidone (1.5 µM DMD, n = 6), or phenoxybenzamine (1 µM PBZ, n = 6). Basal (open bars) and test period (filled bars) release rates are shown. Data are means ± SE. *P < 0.05 vs. preceding basal period by paired t-test. +P < 0.025 vs. DA test period by ANOVA.

D₁ antagonism prevented the AVP secretory response to a maximally stimulating dose of angiotensin II as well as the response to the D₁ agonists. D₂ antagonism did not alter the effect of any of the agonists (Fig. 4). Figure 5 shows that ₁-adrenergic blockade with prazosin did not inhibit AVP release to either SKF-38393 or angiotensin II. No significant additional inhibition occurred with the combined D₁ dopaminergic and ₁-adrenergic antagonists. In contrast, AT₁ receptor blockade inhibited only angiotensin II-stimulated but not dopamine- or SKF-38393-induced AVP release. AT₂ receptor inhibition did not prevent AVP secretion (Fig. 6). Given alone, neither losartan nor CGP-42112A changed basal AVP secretory rate [82 ± 27 and 60 ± 19% · explant¹ · h¹, respectively (P > 0.05 vs. basal)].

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Selective dopamine receptor antagonism with 1 µM SCH-23390 (SCH) or with 1.5 µM domperidone (DMD) of AVP release induced by 5 µM dopamine (filled bars), 1 µM SKF-38393 (open bars), or 1 µM angiotensin II (crosshatched bars). Values are means ± SE; n = 6 for each of the nine groups except for angiotensin II alone where n = 8. *P < 0.05 vs. basal release rate by paired t-test. +P < 0.05 vs. agonist alone by ANOVA.

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Antagonism of SKF-38393 (1 µM; open bars)- or angiotensin II (1 µM; crosshatched bars)-induced AVP release by 1 µM SCH-23390 (SCH) or 1 µM prazosin (PZN). Values are means ± SE; n = 6 for each group except angiotensin II with both SCH-23390 and prazosin where n = 8. *P < 0.025 vs. basal release rate by paired t-test. **P < 0.05 and +P < 0.01 vs. agonist alone by ANOVA.

View larger version (28K): [in this window] [in a new window]   Fig. 6.   Selective angiotensin AT1 and AT2 antagonism with 0.5 µM losartan (LOS) or 0.5 µM CGP-42112A (CGP) of AVP release induced by 5 µM dopamine (open bars), 1 µM SKF-38393 (filled bars), or 1 µM angiotensin II (crosshatched bars). Values are means ± SE; n = 8 for each group except dopamine alone and SKF-38393 with CGP-42112A where n = 7. *P < 0.025 vs. basal release rate by paired t-test. +P < 0.001 vs. agonist alone by ANOVA.

A submaximal concentration of either dopamine or SKF-38393 in combination with a subthreshold dose of angiotensin II prompted a significant release of AVP. Stimulation by the combined agonists was blocked by both D₁ and AT₁ inhibitors as well as nonselective angiotensin receptor antagonism (Table 2).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The present study reports four major findings. First, dopamine and its agonists increase basal AVP release dose dependently via D₁ receptor activation. Second, angiotensin II-induced AVP secretion occurs via an AT₁ receptor. Third, the AVP secretory response to angiotensin II is blocked by selective D₁ receptor antagonism, but dopamine-stimulated AVP release is not altered by angiotensin II receptor inhibition. Fourth, D₁ agonism potentiates the stimulatory effect of a subthreshold dose of angiotensin II.

Before attempting to assess the effect of dopamine on angiotensin II secretion by hypothalamo-neurohypophysial explants, it was necessary to characterize the response of these explants to dopamine. Previous studies reported that intracerebroventricular injections as well as direct injections of dopamine into the supraoptic and paraventricular nuclei increased plasma AVP and resulted in antidiuresis in euvolemic, water-loaded rats (7, 15, 17, 20), but some investigators observed suppression or no change in AVP secretion with dopamine (4, 39). The variability of the response to dopamine was highly dependent on the initial activity of the neurohypophysial system. Dopamine had little, if any, effect on AVP release when hormone levels were already elevated, but when AVP levels were low, as in water-loaded rats, dopamine increased AVP (7, 31). In our explants, where the confounding effects of osmolality, extracellular volume, systemic arterial pressure, or anesthesia can be controlled or eliminated, dopamine stimulated AVP secretion.

Our findings clearly implicate a D₁ receptor as responsible for the increase in AVP. Earlier studies were compelled to rely on nonselective ligands, and stimulation of other catecholaminergic receptors, such as -adrenergic receptors, often was not excluded (7, 15, 17, 25, 26). The present studies extend observations made in perifused explants and in cultured neurons from supraoptic nucleus by Sladek and colleagues (18, 32) by showing the concentration dependency and the D₁ selectivity of the AVP stimulatory response.

Dopamine fibers terminate synaptically on magnocellular neuron cell bodies within the supraoptic nucleus (3, 6). These cells express D_1A mRNA and exhibit D₁ receptors on the cell membrane (8, 18). Dopaminergic fibers also project from the arcuate nucleus to the neurointermediate lobe and end in close proximity to the AVP neurosecretory endings (3, 17, 23). Experiments in these explants do not distinguish between these sites of action. D₁ receptors at either or both of these loci could potentially have been activated. Data exist showing that, at least at the level of the neural lobe, D₁ agonists facilitate electrically evoked AVP release (25, 26).

Numerous studies have shown that angiotensin II stimulates AVP secretion (30, 31, 33). AT₁ receptors identified within hypothalamic loci have been implicated in AVP release (19, 22, 24). Qadri et al. (24) have shown that microinjection of losartan into supraoptic nuclei reduced AVP release after angiotensin II was administered intracerebroventricularly, but AVP secretion was still twofold higher than baseline levels. In the present experiments, losartan inhibited AVP secretion by the explants to basal levels. A plausible explanation for this disparity is that the explants do not contain the subfornical organ or the paraventricular nuclei. Thus angiotensin II in the medium was only able to act upon AT₁ receptors on the magnocellular cell body itself or on neuronal pathways within the explant (33). Angiotensin II administered in vivo may have acted on the circumventricular organs to elicit AVP release not only via well-characterized angiotensinergic pathways (12) but also via ₁-adrenergic pathways (37).

In dogs, nonselective dopamine receptor antagonism prevented AVP release in response to intravenous angiotensin II (2). Our data indicate that a D₁ dopamine receptor mechanism is responsible. Clearly, when the D₁ antagonist was present, AVP secretion in response to a maximally stimulating concentration of angiotensin II was reduced by 84% to a level that did not differ from the basal release rate. Within the substantia nigra, AT₁ receptor activation potentiated dopamine release from nerve terminals (11). A similar scheme in the hypothalamo-neurohypophysial system would result in angiotensin II inducing dopamine release from nerve terminals within the explant. Dopamine, in turn, would act upon D₁ receptors on the magnocellular cell body, along the axon, or on the neural lobe. In the absence of D₁ receptor activation, as seen with D₁ receptor antagonism, angiotensin II stimulation of the magnocellular neuron fails to release AVP. This does not imply that a dopaminergic neuron or a D₁ receptor mediates angiotensin II actions. These findings do provide evidence, however, that D₁ input at either the cell body or the axon of the magnocellular neuron is required for AT₁ activation to stimulate AVP release (Fig. 7).

View larger version (20K): [in this window] [in a new window]   Fig. 7.   Schematic illustration of the putative D1 dopamine receptors (D1) on vasopressinergic magnocellular neurons within the supraoptic nucleus and on axonal nerve endings within the neural lobe of hypothalamo-neurohypophysial explants. Dopamine (DA) acts on postsynaptic D1 receptors on either the cell body or axonal projections of the magnocellular neuron. Stimulation of the D1 receptors depolarizes the magnocellular neuron releasing AVP into the circulation. Angiotensin II (ANG II) acts via AT1 (AT1) receptors to release dopamine from nerve endings within the supraoptic nucleus. Angiotensin II can also directly stimulate AT1 receptors on the magnocellular neuron cell body. Because blockade of D1 dopamine receptors prevents angiotensin II-induced AVP release, D1 input at either the cell body or the axon is needed for AVP secretion. (The subfornical organ is not contained within the explant.)

The involvement of an ₁-adrenergic pathway in supraoptic nucleus mediating the angiotensin II effect has also been suggested; however, AVP levels were reduced only by 50% when prazosin was microinjected bilaterally (24). The residual AVP concentrations after ₁-adrenergic blockade remained at least threefold higher than baseline levels reported using the same paradigm. Veltmar et al. (37) have shown that the effect of ₁- and ₂-adrenergic antagonism was more pronounced on AVP release prompted by angiotensin II within the paraventricular nuclei, which are not contained within the hypothalamo-neurohypophysial explants. In fact, ₁-adrenergic blockade did not change AVP release to angiotensin II by the explants. The small additional decrement that prazosin elicited when used in conjunction with SCH-23390 was not significant. Thus our findings suggest that dopamine plays a major role in mediating AVP release stimulated by angiotensin II acting at the supraoptic nucleus.

The D₁ dopamine receptor is linked to adenylate cyclase (13, 16). Several investigators have also observed that D₁ dopamine receptor activation increases cytosolic free calcium (9, 36). Although influx of extracellular calcium is not required, cytoplasmic calcium concentration is an important determinant of the response to dopamine in vasopressinergic magnocellular neurons (29). In contrast, AT₁ receptor signaling occurs by activation of the phosphoinositide cascade in the central nervous system (10, 21, 35). Because coadministration of subthreshold concentrations of D₁ agonists and angiotensin II potentiated AVP secretion, it appears that simultaneous occupation of D₁ and AT₁ receptors at low levels on the magnocellular neurons leads to sufficient activation of these separate pathways so as to elicit a rise in intracellular calcium sufficient to provoke a large secretory response. Such an interaction between dopamine and angiotensin II to augment intracellular calcium could quite conceivably occur at the nerve terminal. If the interaction is at the level of the cell body, the mechanisms for modulating the propagation of action potentials to the neural lobe may be more varied and complex. For example, it could involve the participation of excitatory and/or inhibitory amino acid receptors via activation of protein kinase A or protein kinase C by dopamine and angiotensin II. Finally, an action at the level of the soma does not necessarily exclude an action at the level of the axon or nerve terminal.

Taken together, the present data provide evidence that D₁ dopamine receptor activation concentration dependently increases AVP secretion. The findings further support the concept that activation of a D₁ dopaminergic pathway is required for angiotensin II-induced AVP release within the hypothalamo-neurohypophysial explant. We can speculate that the need for dual inputs would allow for integration and fine tuning of AVP in the whole organism. Finally, the potentiation of AVP secretion by combined subthreshold D₁ and AT₁ agonism is consistent with sufficient activation of apparently distinct intracellular pathways to increase intracellular calcium to a level that prompts secretion. Additional studies are warranted to localize the site of dopaminergic modulation of the angiotensin II response more precisely, such as magnocellular cell body versus neurointermediate lobe.