Adrenomedullin (ADM) acts in a site-specific manner within autonomic centers of the brain to modulate mean arterial pressure (MAP). To determine the role of ADM in the pontine autonomic center, the lateral parabrachial nucleus (LPBN), we used urethane-anesthetized adult Sprague-Dawley male rats to test the hypothesis that ADM increases MAP at this site through glutamate- and nitric oxide (NO)-dependent mechanisms. ADM microinjected into the LPBN increased MAP in a dose-dependent manner. The pressor effect of ADM (0.01 pmol) had a peak value of 11.9 ± 1.9 mmHg at 2 min and lasted for 7 min. We demonstrated that ADM's effect is receptor mediated by blocking the effect with the ADM receptor antagonist, ADM22-52. We showed that glutamate mediates ADM's pressor response, as this response was blocked using coinjections of ADM with dizolcipine hydrogen maleate or 6-cyano-7-nitroquinoxaline-2,3-dione, N-methyl-d-aspartate (NMDA) and non-NMDA glutamate receptor antagonists, respectively. We tested the roles of NO with coinjections of ADM with either N5-(1-iminoethyl)-l-ornithine or 7-nitroindazole monosodium salt, nonspecific and neuronal NO synthase (NOS) inhibitors, respectively; both inhibitors blocked ADM's pressor effect. Finally, we studied the role of calcium influx in ADM's pressor effect, as intracellular calcium is important in both glutamate and NO neurotransmission. ADM's effect was blocked when nifedipine, an L-type calcium channel blocker, was coinjected with ADM into the LPBN. This study is the first to show that ADM acts in the LPBN to increase MAP through mechanisms dependent on activation of ionotropic glutamate receptors, neuronal and endothelial NOS-mediated NO synthesis, and L-type calcium channel activation.
- brain autonomic centers
- mean arterial pressure
- glutamate receptors
- nitric oxide synthase
discovered in 1993 (20), adrenomedullin (ADM) is a vasoactive peptide that modulates mean arterial blood pressure (MAP) through both central (36, 38, 45, 49) and peripheral (19, 32) mechanisms. Peripherally, ADM acts as a potent hypotensive agent (20). Because plasma levels of ADM are higher in human hypertensive patients, it has been suggested that ADM is involved in the physiopathology of hypertension (14, 29, 30). Hypertension affects one in four adults (44) and is highly correlated with the incidence of stroke and cardiovascular-related deaths. Therefore, it is important to understand the roles of ADM in MAP regulation.
Although ADM acts as a vasodilator in the periphery (12, 28), microinjections of ADM into various regions of the central nervous system (CNS) either decrease (49) or increase (36, 38, 45) MAP in rats, illustrating that ADM's actions in the CNS are site specific. Microinjections of ADM into the rostral ventrolateral medulla of anesthetized rats had a pressor, baroreflex-inhibiting effect (45). Similarly, microinjections of ADM into either the cerebral ventricles of conscious rats (36, 38) or into the area postrema/nucleus of the solitary tract (NTS) of anesthetized rats (1) increased MAP. Conversely, ADM induced a hypotensive effect when injected into the hypothalamic paraventricular nucleus (41, 49). These results suggest that ADM acts in a pleiotropic manner in the CNS to regulate MAP and that ADM's roles must be separately defined in individual cardiovascular centers.
The lateral parabrachial nucleus (LPBN), an important autonomic nervous center involved in MAP homeostasis, is interconnected with other autonomic cardiovascular centers, including the NTS (15), the hypothalamic paraventricular nucleus (22), and the rostral ventrolateral medulla (13), and contains neurons that express ADM (39) and ADM receptors (ADMR) (43). In vivo studies performed in rats (6, 16, 18) and cats (2, 7) suggested that the LPBN receives nervous signals related to changes in MAP, illustrated with c-fos staining or electrophysiological recordings. Glutamate plays an important role in mediating cardiovascular responses in the LPBN, as microinjections of glutamate in anesthetized rats increased both MAP and heart rate (HR) (24). Moreover, stimulation of cardiac sympathetic afferents in the anesthetized cat activated glutamatergic and neuronal nitric oxide synthase (NOS)-containing neurons in the LPBN (7). Finally, psychological stress in conscious rats activated nitric oxide (NO)-producing neurons in LPBN (23). Together, these studies show that NO and glutamate in the LPBN mediate changes in MAP and that ADM may act in the LPBN to affect MAP.
ADM acts through two specific metabotropic receptors that are widely distributed in the brain (42, 43). These receptors are formed by the calcitonin-receptor-like-receptor and receptor activity-modifying proteins 2 or 3 (10, 25). ADM binds to its receptors and activates intracellular signaling pathways (27, 31, 40, 48), which further increase the intracellular calcium concentration (46–48) to stimulate NO production (21). Our findings support the role of NO and glutamate in mediating at least some of ADM's central effects on MAP in the rostral ventrolateral medulla and paraventricular nucleus of anesthetized rats, as inhibition of NO synthesis or blockade of glutamate ionotropic receptors in these centers abolished ADM's effects (45, 49). Furthermore, our in vitro studies confirm that ADM stimulates NO production in SK-N-SH human neuroblastoma cells (48) and in primary rat hypothalamic neuron cultures (47). ADM also regulates the activity of L-type calcium channels to facilitate calcium entry into the cell (48) to potentially contribute to a greater activation of glutamate NMDA receptors (4, 35) and to stimulation of NO synthesis (21). Although these findings address the cellular effects of ADM in other autonomic centers of the brain, the interactions among these elements in mediating the effects of ADM in the LPBN remain to be elucidated.
We used urethane-anesthetized male rats to test the hypothesis that ADM acts in the LPBN to increase MAP and HR through glutamate- and/or NO-mediated mechanisms. To determine whether these effects are ADMR mediated, we attempted to block them with the ADMR-specific antagonist, ADM22-52. Finally, to define the roles of glutamate, NO, and calcium-mediated mechanisms in ADM's effect on MAP, we coinjected ADM with glutamate receptor antagonists, NOS inhibitors, or an L-type calcium channel blocker into the LPBN.
MATERIALS AND METHODS
Adult male Sprague-Dawley rats (350–400 g) were purchased from the Biological Animal Center, University of Alberta, housed in a 12:12-h light-dark cycle at 22°C, and given food and water ad libitum. All surgical and handling protocols were approved by the local Animal Welfare Committee.
Preparation of animals.
Rats were anesthetized with urethane (1.75 g/kg ip; Sigma Chemical, St. Louis, MO); additional anesthetic was given as needed during each experiment. Body temperature was monitored with a rectal thermometer and maintained at 37°C with a heating pad.
The left femoral artery was cannulated using polyethylene (PE)-50 tubing (Becton Dickinson, Sparks, MD), and mean arterial blood pressure (MAP) and heart rate (HR) were measured using a pressure transducer connected to the computer-based acquisition system DI-150 RS (DATAQ Instruments, Akron, OH).
Rats were secured into a stereotaxic device (David Kopf Instruments, Tujunga, CA), and a guide cannula (C315G; Plastics One, Wallingford, CT) was lowered into the left LPBN according to the coordinates: −0.7–1.3 mm posterior, 2.0–2.4 mm lateral, and 2.7–3.3 mm dorsal to the interaural zero (34). An internal cannula (C315I; Plastics One) was connected to a syringe filled with pharmacological agents via PE-50 tubing and was inserted into the guide cannula. Drugs were injected into the LPBN using an electronic infusion pump (Harvard Apparatus, Holliston, MA). All injections were made in a volume of 100 nl over 1 min.
The pharmacological agents used in this study were ADM (0.001–0.1 pmol; American Peptide, Sunnyvale, CA), ADM22-52 (0.01 pmol; American Peptide) (45), dizolcipine hydrogen maleate (MK-801, 500 pmol; Sigma Chemical) (9, 45), 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 50 pmol; Sigma Chemical) (37), 7-nitroindazole monosodium salt (7-NiNa, 0.05 pmol; AG Scientific, San Diego, CA) (2), N5-(1-iminoethyl)-l-ornithine (l-NIO, 200 pmol; Tocris Cookson, Ellisville, MO) and nifedipine (1 pmol and 0.1 pmol, Sigma Chemical). The doses of the pharmacological agents were chosen on the basis of our previous studies (45) and were tested alone in separate groups of control experiments to determine the concentration that would not significantly change MAP and HR (Table 1).
MAP and HR were measured for 20–30 min to obtain baseline values. Increasing doses of ADM (0.001–0.1 pmol) were unilaterally microinjected in separate groups of experiments to determine the dose-dependent effects of ADM on MAP and HR. ADM at 0.01 pmol was used in subsequent experiments, as it was the lowest dose to elicit consistent increases in MAP compared with the controls. Saline, various drugs alone, or a combination of ADM (0.01 pmol) with these drugs (in saline) were unilaterally injected into the left LPBN (n ≥ 6 for each group). MAP was recorded for at least 1 h after the microinjections. Bilateral microinjections were used for experiments that employed both nifedipine and ADM, according to the following sequence: injection of nifedipine (0.1 pmol), followed 15 min later by ADM (0.01 pmol) into the ipsilateral LPBN; after a further 15 min, ADM (0.01 pmol) was injected into the contralateral LPBN. At the end of each experiment, 300 nl of 1% Evans blue were injected for the histological verification of the injection site.
At the end of experiments, rats were killed by decapitation. The brains were removed and fixed in 4% paraformaldehyde for 2 or 3 days, then frozen at −80°C. Coronal brain stem sections of 60 μm were cut using a cryostat (−20°C), thaw-mounted onto slides, and stained with neutral red (Allied Chemical, New York, NY). Slides were examined under a microscope, and accurate microinjections were confirmed by the presence of Evans blue within the LPBN.
All results are expressed as means ± SE. The temporal effects of various treatments on AP and HR were assessed at 1-min intervals and compared among groups using the two-way ANOVA with repeated measures followed by the Student-Newman-Keuls test for post hoc comparison of individual means. P < 0.05 indicated statistical significance.
Microinjections of ADM into the LPBN increase MAP through an ADM receptor-mediated mechanism.
ADM (0.001, 0.01, and 0.1 pmol) microinjected into the LPBN (Fig. 1) increased MAP in a dose-dependent manner (Fig. 2). Microinjections of ADM placed adjacent to the LPBN did not elicit statistically significant changes in MAP, showing that ADM's effects are specific to the LPBN. ADM (0.01 pmol) in the LPBN elicited a significant increase in MAP beginning in the first minute after microinjection. The pressor effect of ADM lasted for ∼7 min with the maximal increase in MAP occurring at 2 min postinjection (11.9 ± 1.9 mmHg, Fig. 2). HR was not significantly influenced by any of the ADM microinjections.
The ADM receptor antagonist, ADM22-52 (0.01 pmol), coinjected with ADM (0.01 pmol), blocked the ADM-induced increase in MAP, showing that ADM acts through its receptors in the LPBN to increase MAP (Fig. 3). ADM22-52 alone (0.01 pmol) did not affect the MAP when microinjected into the LPBN (Fig. 3 and Table 1).
Effects of ADM on MAP are mediated by L-glutamate ionotropic receptors in the LPBN.
The effects of ADM on MAP in the LPBN were blocked when ADM (0.01 pmol) was coinjected with MK-801 (500 pmol), a selective NMDA receptor antagonist (50 pmol; Fig. 4). Coinjection of ADM with the selective non-NMDA receptor antagonist CNQX also blocked the ADM-induced increase in MAP (Fig. 4). MK-801 (500 pmol) or CNQX (50 pmol) microinjected alone into LPBN did not affect MAP (Table 1). These results suggest that ADM acts in the LPBN to increase MAP through a mechanism dependent on the activation of ionotropic glutamate receptors.
Effects of ADM in the LPBN on MAP are mediated by NO produced by nNOS.
The increase in MAP was blocked when ADM (0.01 pmol) was coinjected into the LPBN with l-NIO (200 pmol), a nonspecific NO synthase inhibitor (Fig. 5). Furthermore, the neuronal NO synthase inhibitor, 7-NiNa (0.05 pmol), coinjected with ADM (0.01 pmol) into the LPBN, blocked the ADM-induced increase in MAP (Fig. 5). Neither l-NIO (200 pmol) nor 7-NiNa (0.05 pmol), microinjected alone into LPBN, elicited a significant change in MAP compared with saline controls (Table 1). These data show that the ADM-induced increase in MAP, which occurs after microinjection into LPBN, depends on NO produced by neuronal NOS.
The pressor effect of ADM in the LPBN is mediated by a mechanism dependent on activity of L-type calcium-channels.
The pressor effect of ADM in the LPBN was blocked when ADM (0.01 pmol) was injected into the ipsilateral LPBN 15 min after the L-type calcium-channel blocker, nifedipine (0.1 pmol) (Fig. 6). Nifedipine at 0.1 pmol was chosen in these experiments because it did not significantly affect MAP when injected alone (Table 1). To confirm that nifedipine specifically blocked ADM's effects on MAP, ADM (0.01 pmol) was then microinjected into the contralateral LPBN, in the absence of nifedipine. These contralateral injections of ADM consistently elicited increases in MAP that were not significantly different from those described for single ADM (0.01 pmol) injections (Figs. 2 and 6). These results demonstrate that the pressor effect of ADM is mediated by intracellular calcium.
In our study, we show for the first time that ADM increases MAP in a dose-dependent manner when microinjected into the LPBN, a major cardiovascular center in the brain. The pressor effect of ADM is receptor mediated, as the specific ADMR antagonist, ADM22-52, blocked this effect. Moreover, ADM's pressor effect depends on the activation of glutamate receptors because it was blocked when ADM was coinjected with either NMDA or non-NMDA receptor antagonists. The pressor effect of ADM in the LPBN is also mediated by NO, as it was abolished when NO synthesis was blocked in the LPBN. Finally, ADM's effect is dependent on intracellular calcium as nifedipine, a specific L-type calcium channel blocker, abolished the pressor effect when microinjected into the LPBN before ADM. Together, our data show that ADM increases MAP in the LPBN through mechanisms dependent on glutamate, NO, and L-type calcium channel activity.
The concentration of ADM (0.01 pmol) used in our study is similar to the concentrations found to exert a pressor effect in the rostral ventrolateral medulla (45) and a depressor effect in the hypothalamic paraventricular nucleus (49). We demonstrated that ADM's effects are specific to the LPBN, as no pressor effects were obtained when ADM was injected outside the LPBN. In addition, our data show that ADM's actions in the LPBN are mediated by ADMR, as ADM22-52, an ADMR-specific antagonist, blocked ADM's effect when both agents were injected together.
In the present study, we show that ADM acts in the LPBN through a glutamate receptor-dependent mechanism, as we were able to block ADM's effect by coinjecting ADM with either NMDA (MK-801) or non-NMDA (CNQX) glutamate receptor antagonists. Other studies support a role for glutamate receptors in the LPBN-generated MAP changes. For example, extracellular recordings from LPBN neurons following electrical stimulation of the NTS showed excitatory responses, mediated through both NMDA and non-NMDA glutamate receptors (17). In addition, the increase in MAP initiated by electrical stimulation of the dorsal periaqueductal gray of anesthetized rats was blocked by bilateral injections of kynurenic acid, an NMDA receptor antagonist, into the LPBN (11). Studies in anesthetized rats suggest that glutamate receptors also mediate the effects of ADM in the rostral ventrolateral medulla and paraventricular nucleus (45, 49). Thus, our data show that ADM in the LPBN increases MAP by activating glutamate receptors and suggest that the mechanisms for ADM's effects on MAP are similar in several central autonomic nuclei.
Other studies have also demonstrated roles for glutamate in mediating pressor responses in the LPBN. In anesthetized rats, injections of glutamate into the LPBN generated pressor responses (5). In other studies using anesthetized cats, glutamate microinjections in the LPBN elicited pressor responses but had no effects on HR (8), and epicardial application of bradykinin to the left ventricle of the heart activated glutamatergic neurons in LPBN, shown with immunohistochemical localization of c-fos and vesicular glutamate transporter 3 (7). In decerebrate rabbits, glutamate injected into the LPBN-induced consistent increases in MAP and HR (33). Together, these data confirm a role for glutamate in mediating pressor responses in the LPBN.
Our results establish that ADM's pressor effect in the LPBN is also dependent on NO produced by nNOS, as inhibition of NO synthesis with either l-NIO or 7-NiNa (nonspecific NOS and nNOS inhibitors, respectively) effectively blocked this effect. ADM's role in potentiating NO synthesis is also suggested by our in vitro studies showing that ADM stimulates both NO release from human neuroblastoma cells and NO production from primary rat hypothalamic neurons (47, 48). These data are further supported by studies suggesting that NO mediates the ADM-induced cardiovascular changes in the hypothalamic paraventricular nucleus and rostral ventrolateral medulla (45, 49). In addition, as discussed above, ventricular stimulation with bradykinin activated nNOS-containing neurons in the LPBN of anesthetized cats (7). Moreover, in conscious rats, stress led to c-fos activation in NO-producing neurons of the LPBN (23). Therefore, central NO appears to be involved in regulation of autonomic functions, including the blood pressure in a number of central sites, including the LPBN. Currently, the sequence of NO-dependent events, mediating ADM's pressor effect in the LPBN, downstream of NO synthesis, is not clear. Future studies using guanylate cyclase inhibitors would establish whether NO itself, rather than other proteins posttranslationally modified through NO-induced S-nitrosylation, is the primary mediator.
At the present time, the sequence of interactions among ADM, glutamate, and NO in neurons of the LPBN remains unclear. Others have shown that NMDA receptors and nNOS are functionally linked at the postsynaptic level, in a complex dependent on the protein, PSD-95 (3). It has also been shown that glutamate stimulates NO synthesis in a calcium-dependent manner and that NO then acts through a feedback mechanism to modulate the activity of presynaptic neurons; in this way, NO is thought to stimulate the release of neurotransmitters, including glutamate (26). We have shown that L-type calcium channels are required for the stimulation of NO production by ADM in neuroblastoma cells (48), and we show here that ADM's pressor effect in the LPBN is dependent on L-type calcium channels.
Perspectives and Significance
Our study provides data important for understanding ADM's central actions in MAP regulation. We have used site-specific microinjections of ADM in anesthetized rats to show that ADM acts in the LPBN to increase MAP through complex mechanisms dependent on glutamate receptor activation, NO synthesis, and L-type calcium channel activation. Additional studies in conscious animals will be needed to fully establish the physiological significance of ADM signaling in the LPBN. Future research is also required, not only to define the nature of ADM-induced cellular interactions between glutamate and NO, but to explore the possibility that dysfunctions in ADM signaling in the LPBN contribute to the development of hypertension.
This work was supported by the Heart and Stroke Foundation of Alberta/Northwest Territories/Nunavut.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society