INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

PEPTIDES DERIVED FROM postranslational modification of preproadrenomedullin have been demonstrated to exert a variety of biological effects, many related to fluid and electrolyte homeostasis and cardiovascular function (13). Indeed, at least two such peptides, adrenomedullin (AM) and proadrenomedullin NH₂-terminal 20 peptide (PAMP), lower blood pressure when administered peripherally. The hypotensive effects of AM are exerted at least in part by direct activation of nitric oxide synthase (3) in the vasculature. On the other hand, hypotensive actions of PAMP result from the ability of the peptide to act presynaptically to inhibit norepinephrine release from sympathetic nerve terminals that innervate the blood vessels (14) and, possibly in a paracrine or autocrine fashion, to inhibit cholinergic activation of catecholamine release in the adrenal medulla (6). Central nervous system (8, 10) and pituitary (11) actions of AM have been demonstrated, potentially reflecting the action of endogenous peptide produced in brain (12). The diuretic and natriuretic actions of AM in kidney (5, 17) are complemented by effects of the peptide in brain to inhibit water drinking (8) and salt appetite (10). Passive immunoneutralization studies have suggested the physiological relevance of the latter effect (10).

Surprisingly, it has been reported that the hypotensive action of AM in the periphery is not matched by a similar effect in brain (16). Instead, at least in anesthetized animals, administration of the peptide into the cerebroventricular fluid resulted in an elevation of blood pressure. In the studies described here, we attempted to determine the effects of AM and, for the first time, PAMP on blood pressure in conscious, unrestrained rats after central administration. Our results indicate that both AM and PAMP exert hypertensive actions when administered centrally in conscious animals. Furthermore, these effects are mediated via activation of the -adrenergic division of the sympathetic nervous system and, in the case of PAMP, are due at least in part to recruitment of endogenous ANG II.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Adult male rats (Harlan Sprague Dawley, 200-250 g) were housed individually in laminar flow, isolation units and provided tap water and lab chow ad libitum. All surgical (including anesthesia) and handling protocols were approved by the Institutional Animal Care and Use Committee. Under tribromoethanol anesthesia (2.5% in saline, 1 ml/100 g body wt, Aldrich Chemical, Rouses Point, NY) animals were placed in a stereotaxic device and the skull was exposed via a scalp incision (1-2 cm). An indwelling, stainless steel cannula (23 gauge, 18 mm) was lowered through a burr hole in the skull into the right lateral cerebroventricle (relative to interaural line: anterior +6.2, lateral 0.6, vertical +7.0). The cannula was plugged with stainless steel wire (25 gauge) and fixed in place with dental screws and cement. The incision was sutured closed, and animals were allowed to recover to presurgery body weight (a minimum of 5 days). On the day of experimentation, rats were again anesthetized with tribromoethanol and the left external carotid artery was exposed through a surgically prepared incision (1-2 cm). A cannula (PE-50, filled with 0.9% NaCl containing 250 U/ml heparin) was inserted, tunneled subcutaneously to exit on the dorsum of the neck, and sutured in place. The incision was sutured closed, and animals were allowed to recover for 3 h. At that time the exteriorized cannula was flushed with heparinized saline and attached to a pressure transducer (Digi-Med Blood Pressure Analyzer, Micro-Med, Louisville, KY) and blood pressure was monitored during a 1-h stabilization period. The hypertensive effects of centrally administered AM and PAMP observed in this animal model were also observed in similarly cannulated rats that had been allowed 24 h of recovery after carotid cannulation and in rats bearing chronically implanted (abdominal aorta) radiotelemeters (data not shown). Peptides were dissolved in 0.9% NaCl and injected intracerebroventricularly via the indwelling cerebroventricular cannula in a total volume of 2 µl over 30 s. The animals were unrestrained during these injections. Saline injections (intracerebroventricular) were conducted in each animal to assure the absence of injection artifact. ANG II injections (100 pmol icv) were employed as positive controls for cannula patency, because this dose of peptide results not only in an increase in blood pressure but also a readily verified behavioral response (dipsogenesis). Doses of AM chosen for testing (44, 88, 176, and 352 pmol) exceed those used by us (8, 10) in our previous behavioral studies (44 and 88 pmol), but at the highest dose approach those demonstrated to be hypertensive in anesthetized rats (16). All peptides were the rat sequences and were purchased from Phoenix Pharmaceuticals (Mt. View, CA). Phentolamine mesylate and L-phenylephrine hydrochloride were obtained from Sigma Chemical (St. Louis, MO). Data were analyzed by paired t-test comparing 5-min means before and after injection of test substances. An outcome with a probability of P < 0.05 was considered significant.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Experiment 1. In a randomized order, rats received intracerebroventricular injections of 100 pmol ANG II and 176 or 352 pmol AM, separated by baseline stabilization periods of at least 20 min. On completion of intracerebroventricular testing, five of these animals received via the carotid cannula bolus injections separated by recovery periods of at least 20 min of both test doses of AM in 0.1 ml 0.9% NaCl. Mean arterial blood pressure (MAP; mmHg) was recorded online and reported at 1-min intervals, which reflect the average MAP over the final 15 s of the sampling period. As expected (Fig. 1), intracerebroventricular injection of 100 pmol ANG II resulted in a significant (P < 0.001) elevation in MAP. Significant elevation (P < 0.001) in MAP was also observed after intracerebroventricular injection of 176 or 352 pmol AM. This is most clearly observed when the data were expressed as means of the five 1-min sampling periods before and immediately after intracerebroventricular injections (Table 1). By comparison, these doses of AM, when administered via the carotid cannula directly into the peripheral circulation, resulted in significant depression of MAP (Table 1). Lower doses of AM (44 and 88 pmol) did not significantly alter MAP, regardless of route (intravenous or intracerebroventricular) of administration (data not shown).

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Effect of intracerebroventricular administration of ANG II or adrenomedullin (AM) on mean arterial blood pressure (MAP) in conscious rats. Numbers in parentheses indicate group size. Data presented indicate pressures monitored at 1-min intervals before (5 min) and immediately after (10 min) peptide injection (means ± SE).

Experiment 2. To compare the effects of AM with those of the other identified vasoactive peptide derived from preproadrenomedullin, rats received in randomized order intracerebroventricular injections of 100 pmol ANG II, 1 or 0.1 nmol PAMP, saline, and 176 pmol AM. Again, in these animals significant elevations in MAP followed peptide administration (Fig. 2). When the data were expressed as means of the 5-min sampling periods before and after injections (Table 2), PAMP, like ANG II, was found to evoke a highly significant elevation in MAP (P < 0.001). A lower dose of PAMP (0.01 nmol) did not significantly alter MAP in these conscious rats (data not shown). Although also significant (P < 0.01), the magnitude of MAP elevation in response to 176 pmol AM was smaller than that observed after a similar (100 pmol) dose of PAMP.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Effect of intracerebroventricular administration of ANG II, proadrenomedullin NH2-terminal 20 peptide (PAMP), AM, or saline vehicle (Sal) on MAP in conscious rats. Numbers in parentheses indicate group size. Data presented indicate pressures monitored at 1-min intervals (means ± SE).

Experiment 3. To determine the role of the sympathetic nervous system in the hypertensive effects of intracerebroventricularly administered ANG II, AM, and PAMP, injections were performed in two separate groups of animals before and after peripheral -adrenergic blockade with phentolamine. The dose chosen (10 µg/kg body wt, administered via the carotid cannula) was determined in preliminary studies to completely suppress the pressor effect of phenylephrine (5 µg/kg body wt). This dose of phentolamine significantly reduced blood pressure in all animals. The pressor effects of 100 pmol ANG II and 300 pmol AM (Table 3) were completely abolished by -adrenergic blockade. In a second group of animals, again the hypertensive effect of 100 pmol ANG II and also that of 300 pmol PAMP (Table 3) were abolished by phentolamine pretreatment.

View this table: [in this window] [in a new window]   Table 3.   -Adrenergic blockade abrogates the hypertensive actions of centrally administered ANG II (100 pmol) and the preproadrenomedullin-derived peptide AM (300 pmol) and PAMP (300 pmol)

Experiment 4. Because PAMP and AM act centrally to elevate MAP in a fashion similar to that of ANG II, another group of animals was injected intracerebroventricularly with 100 pmol ANG II and 300 pmol PAMP (in random order) before and after blockade of central ANG II receptors with saralasin (10 µg icv). The dose chosen was established in preliminary studies to block the dipsogenic response of conscious animals to intracerebroventricular injection of 100 pmol ANG II, but not to exert locomotor or pressor effects. Additionally, this dose of saralasin did not, by itself, significantly alter baseline blood pressure. The effect of this blocking dose of saralasin was reversible (data not shown), because 90 min after saralasin treatment, intracerebroventricular administration of 100 pmol ANG II stimulated water drinking and elevated MAP. Therefore injections of ANG II and PAMP during ANG II receptor blockade occurred within 40 min of saralasin injection. Before saralasin treatment, both ANG II and PAMP elicited pressor responses in conscious rats (Table 4). As expected, after blockade of central ANG II receptors, intracerebroventricular injection of 100 pmol ANG II failed to elevate MAP. Similarly, the pressor response to 300 nmol PAMP was absent after ANG II receptor blockade.

Experiment 5. AM shares significant structural homology with CGRP, and some of the actions of AM can be blocked by the CGRP antagonist CGRP-(837). To determine whether the hypertensive effects of intracerebroventricularly administered AM are exerted via the CGRP receptor, peptide injections were performed in two separate groups of animals before and after blockade of central CGRP receptors by intracerebroventricular injection of CGRP-(837). The CGRP antagonist did not significantly alter baseline blood pressure at the dose employed. In the first group of animals, 100 pmol ANG II and 300 pmol PAMP were injected before and after administration of the CGRP antagonist (Table 5). Significant hypertensive effects of both peptides were observed both in the absence and presence of CGRP-(837). Similarly, blockade of the CGRP receptor with CGRP-(837) did not prevent the hypertensive effects of 100 pmol ANG II or 300 pmol AM in the second series of animals (Table 5).

View this table: [in this window] [in a new window]   Table 5.   Blockade of the CGRP receptor with 3 nmol CGRP-(837) fails to prevent the hypertensive actions of centrally administered ANG II (100 pmol) and the preproadrenomedullin-derived peptides, AM (300 pmol) and PAMP (300 pmol)

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The AM gene is transcribed and the gene product translated (10) within the central nervous system (CNS). There, at least one of the postranslational products of preproadrenomedullin has been demonstrated to exert behavioral actions that complement actions of the peptide in the periphery. Indeed, AM exerts diuretic and natriuretic actions in the kidney (5, 17) and inhibits both water drinking (8) and salt appetite (10) by actions in the CNS. The action of AM in pituitary gland to inhibit adrenocorticotropin secretion (11) matches its action in adrenal gland (18). However, not all the actions of AM in the periphery appear coordinated to contribute to elimination of fluid and electrolytes or unloading of vascular pressure. Significant, direct ionotropic and chronotropic effects have been demonstrated (15). Indeed, in clinical situations of cardiac failure and volume overload, the elevations in circulating AM have been hypothesized (4, 7) to be a reflection of compensatory mechanisms designed to protect renal and cardiovascular function. This too may be the physiological relevance of the central hypertensive effects of these two products of the same gene. One could envision central cardioprotective effects of AM and PAMP that counterbalance the peripheral hypotensive actions. Of course our data in no way indicate, and we do not imply, that central and peripheral releases of the two peptides are synchronized; however, it is attractive to speculate that, just as the direct cardiostimulatory effects of AM in the heart may offset the hypotensive actions in the vessels and therefore protect against vascular collapse, so too the CNS actions may buffer the peripheral hypotensive effects of the peptides.

We have demonstrated for the first time that PAMP, which has been identified to act on nerves in the periphery to elicit hypotensive effects (14), exerts significant hypertensive actions in conscious animals by a central action to stimulate -adrenergic, sympathetic function. A similar, central mechanism is also suggested for AM, a peptide already demonstrated to exert behavioral effects in conscious animals (8, 10). We hypothesize that PAMP and AM act centrally in a fashion similar to ANG II, activating descending fiber pathways that stimulate intermediolateral cell column function. This hypothesis is supported by the observation that, in addition to increasing blood pressure after intracerebroventricular administration, both AM (300 pmol; 5 min preinjection: 344 ± 24 beats/min; postinjection: 373 ± 22; P < 0.05) and PAMP (300 pmol; preinjection: 350 ± 26; postinjection: 375 ± 20; P < 0.05) stimulated significant increases in heart rate in these conscious, unrestrained animals.

Although we cannot speculate on the site of action of PAMP and AM after lateral cerebroventricular injection, the rapidity of effect suggests a site within the forebrain. The peptides may be acting, however, at more distant sites. Direct neural actions of AM in brain slices taken from the medulla have been demonstrated by Allen and Ferguson (1). Those effects have cardiovascular consequences, because in the anesthetized rat microinjection of AM into the area postrema resulted in elevated blood pressure (2).

Of particular interest is the ability of ANG II receptor blockade to abrogate the hypertensive effect of PAMP. This strongly suggests that either PAMP acts via the ANG II receptor or that the cellular effects of PAMP are to eventually recruit neural networks employing ANG II as the predominant communicator of cellular activity. We do not assert that PAMP interacts directly with the ANG II receptor, in fact, simultaneous administration of submaximal doses of both PAMP and ANG II did not result in more than additive effects on MAP (data not shown).

Finally, whereas at least some of the peripheral effects of AM have been attributed to an interaction of the peptide with the CGRP receptor, we do not feel this to be the case in our model. This opinion is based on our observations that similar intracerebroventricular injections of CGRP, in log doses ranging from 0.01 to 1.0 nmol, failed to significantly alter MAP in conscious rats (data not shown). Furthermore, pretreatment of animals intracerebroventricularly with 3 nmol of the CGRP antagonist CGRP-(837) failed to block the hypertensive effects of both 300 pmol PAMP or AM and 100 pmol ANG II (Table 5). This dose of CGRP antagonist can block the anorexic effect of 300 pmol CGRP (data not shown). In conclusion, we demonstrate with these data that both peptides derived from postranslational processing of preproadrenomedullin exert significant hypertensive effects when administered centrally in conscious rats. The effects are abrogated by -adrenergic blockade, suggesting activation by the peptides of descending neural networks controlling preganglionic cell function in the sympathetic system. Additionally, the effects of PAMP and AM not only mirror those of ANG II, but also likely involve the recruitment of endogenous ANG II systems.

Perspectives