INTRODUCTION

Top Abstract Introduction Materials Results Discussion References

SYNTHETIC RAT adrenomedullin (rADM), consisting of 50 amino acid residues, has been shown to exert a potent vasorelaxant effect and shares structural homology with the calcitonin gene-related peptide (CGRP) (24). Intravenous administration of ADM causes hypotension associated with increases in heart rate (HR) and renal sympathetic nerve activity (RSNA) in awake rabbits (8). Immunoreactive ADM (ir-ADM) and ADM mRNA have been detected not only in the peripheral tissue but also in the brain of human and rat (12, 19, 23). Thus it has been postulated that central ADM might play some role in the central nervous system (CNS). Increases in blood pressure (BP) and sympathetic nerve activity have been reported in response to intracerebroventricular administration of ADM in anesthetized rats (27). Because anesthesia is known to profoundly affect the cardiovascular and autonomic nervous systems (17), we examined the effects of intracerebroventricular administration of rADM on BP, HR, and RSNA in conscious unrestrained rats. We also investigated the possible involvements of the arterial baroreceptors, human CGRP receptor antagonist hCGRP-(837) (2) and putative human ADM receptor antagonist hADM-(2252) (5), in central ADM-induced responses.

	METHODS AND MATERIALS

Top Abstract Introduction Materials Results Discussion References

Animal preparation and data collection. Wistar male rats weighing 350-450 g were implanted with a lateral cerebroventricular cannula under anesthesia by intraperitoneal injection of pentobarbital sodium (50 mg/kg). A 24-gauge stainless steel guide cannula (length 19 mm) was lowered 2.5 mm from the cortex and placed 1 mm above the left lateral cerebroventricle through a burr hole stereotaxically located 0.8 mm posterior and 1.5 mm lateral to the bregma. The guide cannula was fixed to the skull with four screws and dental cement. Both an acute elevation in mean arterial pressure (MAP) over 20 mmHg and a persistent (at least 10 min) water drinking response to intracerebroventricular administration of 10 pmol ANG II were considered as indicators of cannula patency and proper placement in the ventricular system. Approximately 1 wk later, again under pentobarbital anesthesia (50 mg/kg ip), SP-31 tubing heat-coupled to SP-50 and PE-50 catheters was inserted into the abdominal aorta and the inferior vena cava for measurement of BP and for intravenous administration of drugs, respectively. The arterial catheter, filled with heparinized (10 U/ml) saline solution, was connected to a Statham pressure transducer (Gould, Saddle Brook, NJ) to monitor BP, and the venous catheter was sealed. HR was monitored with a cardiotachometer (model 1321; San-Ei, Tokyo, Japan) triggered by an electrocardiogram signal that was recorded via subcutaneous electrodes implanted into the chest. The sinoaortic baroreceptors were denervated using Krieger's method as previously described (20). The effectiveness of sinoarotic denervation (SAD) was confirmed by abolishment of bradycardia and sympathoinhibitory responses to phenylephrine (8 µg/rat iv).

Recording of RSNA. The left renal nerve was carefully dissected via a retroperitoneal approach and freed from its surrounding tissues under stereoscopic microscopy. The nerve was placed on a bipolar electrode made from Teflon-coated wire (Cooner Wire; Chatsworth, CA) and covered with silicone rubber (Semicosil 902A and B cement; Wacker Chemicals East Asia, Tokyo, Japan). Spike potentials, which were amplified (Biophysioamplifier AVB-9; Nihon Kohden, Tokyo, Japan) and filtered (50-1,000 Hz), were monitored on a storage oscilloscope (model VC-9A; Nihon Kohden) and continuously recorded on a magnetic tape recorder (TEAC, Tokyo, Japan). Impulses were then fed into a pulse counter (MET-1100; Nihon Kohden), the output of which was digitized and printed as a histogram and recorded simultaneously with BP and HR on a thermal rectigraph (San-Ei). Tapes were later played back, and the waveforms of RSNA were integrated after full-wave rectification using an amplitude analyzer (series 5500; Concurrent, Fort Lauderdale, FL) with the sample-hold function reset to baseline by an internal timer set at 5 s. Absolute values for integrated RSNA were corrected before data analysis by subtracting the residual electrical output (background noise level) recorded from the integrator after intravenous injection of phenylephrine (8 µg/rat).

Experimental protocol. All experiments were performed in conscious, freely moving rats 1-7 days after surgery. After BP, HR, and RSNA stabilized, 5 µl of vehicle (physiological saline solution) or synthetic rADM (Peptide Institute, Osaka, Japan) dissolved in physiological saline solution was manually injected intracerebroventricularly over 10 s through an infusion cannula (30-gauge stainless steel tubing) connected to a 25-µl microsyringe in intact and SAD conscious rats. This injection was made by inserting an infusion cannula that extended 1 mm beyond the tip of the guide cannula. rADM in doses of 0.1 and 1.0 nmol/kg employed in the experiment was selected on the basis of previous reports in anesthetized rats (14, 24, 27), and 0.5 nmol/kg was chosen as a midvalue between 0.1 and 1.0 nmol/kg.

Under similar experimental conditions, the effect of pretreatment with hCGRP-(837) (Peptide Institute) or hADM-(2252) (Peptide Institute) on ADM-induced changes in MAP, HR, and RSNA was investigated. Either hCGRP-(837) (2.0 nmol/kg) or hADM-(2252) (20 nmol/kg) dissolved in 5 µl of saline or vehicle (5 µl of saline alone) was injected intracerebroventricularly, and 10 min later another intracerebroventricular injection of 1.0 nmol/kg rADM (5 µl) was given. hCGRP-(837) in a dose of 2.0 nmol/kg was selected via the previous report (27). As we found in preliminary experiments, the same dose (2.0 nmol/kg) of hADM-(2252) as hCGRP-(837) was ineffective, so we increased the dose of hADM-(2252) to 20 nmol/kg. After each experiment, Pontamine sky blue was injected to verify the correct placement of the intracerebroventricular cannula tip.

Statistics. Data were analyzed using two-way analysis of variance for repeated measurements. If statistically significant effects were found, Student-Newman-Keuls test was applied to the differences between groups. P value <0.05 was considered to be significant. All data are presented as means ± SE.

	RESULTS

Top Abstract Introduction Materials Results Discussion References

The effects of varying the amount of rADM in the intracerebroventricular administration on MAP, HR, and RSNA in conscious rats are shown in Fig. 1. When rADM (1.0 nmol/kg, n = 11) was injected intracerebroventricularly, MAP and HR increased, whereas RSNA decreased and then gradually increased. MAP started to rise immediately after the injection, reaching peak values at ~10 min. An increase in HR occurred at 7 min after the injection, with a peak increase at 20-25 min. The RSNA troughed at ~8 min after the injection and peaked at 30-35 min. In contrast, intracerebroventricular administration of 0.1 (n = 10) or 0.5 (n = 9) nmol/kg rADM did not provoke a significant increase in MAP and produced a lower increase in HR than the administration of 1.0 nmol/kg rADM. A decrease in RSNA for the first 15 min after the injection without the late increase in RSNA was observed after administration of the two lower doses of rADM. Vehicle (saline, n = 12) did not elicit any effects on MAP, HR, or RSNA (Fig. 1). Leakage of intracerebroventricular rADM into the systemic circulation was ruled out because intravenous administration of rADM (1.0 nmol/kg) caused quite the opposite BP response to intracerebroventricular rADM at the same dose in conscious rats, i.e., a decrease in BP (data not shown).

View larger version (31K): [in this window] [in a new window]   Fig. 1.   Time courses of changes in mean arterial pressure (MAP), heart rate (HR), and renal sympathetic nerve activity (RSNA) after intracerebroventricular administration of rat adrenomedullin (rADM) (0.1, 0.5, and 1.0 nmol/kg) and vehicle (saline) in conscious rats. Arrow shows time (0 min) of injection of rADM or saline solution. bpm, Beats/min. All data represent means ± SE; n is no. of animals. * P < 0.05 vs. vehicle.

Inasmuch as the early inhibition of RSNA corresponded to the initial pressor phase, we next examined the possible involvement of the arterial baroreceptor reflex in the ADM-induced responses using SAD rats. SAD treatment virtually abolished the phenylephrine-induced decreases in HR/MAP (from 6.6 ± 0.5 to 0.7 ± 0.3 beats · min · ¹ · mmHg¹; P < 0.01) and RSNA/MAP (from 5.8 ± 0.6 to 0.5 ± 0.3 %/mmHg; P < 0.01). When 1.0 nmol/kg icv rADM was injected into SAD rats (n = 5), augmented increases in MAP, HR, and the late phase of RSNA were observed, but the early decrease in RSNA was not abolished (Fig. 2).

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Time courses of changes in MAP, HR, and RSNA after intracerebroventricular rADM (1.0 nmol/kg) and vehicle (saline) in sinoaortic-denervated rats. Arrow shows time (0 min) of injection of rADM or saline solution. Results are expressed as means ± SE; n is no. of animals. * P < 0.05 vs. vehicle.

hCGRP-(837) and hADM-(2252) are considered to be receptor antagonists of CGRP and ADM, respectively (2, 5). The effects of intracerebroventricular administration of these antagonists on the changes in MAP, HR, and RSNA evoked by intracerebroventricular rADM are shown in Figs. 3 and 4. Pretreatments with both hCGRP-(837) (2.0 nmol/kg icv, n = 7) (Fig. 3) and hADM-(2252) (20 nmol/kg icv, n = 7) (Fig. 4) significantly attenuated rADM (1.0 nmol/kg icv)-induced changes in MAP, HR, and RSNA. We noted that the increases in MAP, HR, and the late phase of RSNA were more potently abolished by hCGRP-(837) than by hADM-(2252), but only the early decrease in RSNA was significantly suppressed with hADM-(2252). The intracerebroventricular administration of each antagonist alone did not induce any significant changes in these parameters (data not shown).

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Time courses of changes in MAP, HR, and RSNA after intracerebroventricular rADM (1.0 nmol/kg). Pretreatment with intracerebroventricular human calcitonin gene-related peptide-(837) [hCGRP-(837)] (2.0 nmol/kg) or vehicle (saline) was given 10 min before the administration of rADM. Arrow shows time (0 min) of injection of rADM solution. Values are means ± SE; n is no. of animals. * P < 0.05 vs. vehicle-pretreated group.

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Time courses of changes in MAP, HR, and RSNA after intracerebroventricular rADM (1.0 nmol/kg). Pretreatment with intracerebroventricular human ADM-(2252) [hADM-(2252)] (20 nmol/kg) or vehicle (saline) was given 10 min before the administration of rADM. Arrow shows time (0 min) of injection of rADM solution. Values are means ± SE; n is no. of animals. * P < 0.05 vs. vehicle-pretreated group.

	DISCUSSION

Top Abstract Introduction Materials Results Discussion References

This is the first report showing the effects of central ADM on sympathetic nerve activity recorded directly in conscious unrestrained rats. In the present study, intracerebroventricular administration of rADM in doses of 0.1 and 0.5 nmol/kg induced only tachycardia and an early decrease in RSNA. In contrast, our observations that central rADM in a dose of 1.0 nmol/kg provoked hypertension and the late increase in RSNA are nearly consistent with the observations of Takahashi et al. (27), who showed in anesthetized rats that hADM (1.0 and 3.0 nmol/kg) centrally induced an increase in preganglionic sympathetic discharge accompanied by a vasopressor response. Thus it was revealed that central ADM induced different responses according to the doses administered. Interestingly, we found a biphasic change in RSNA in conscious rats that has never been noted in anesthetized rats, even under similar doses of centrally administered ADM. We recommend using conscious animals when intracerebroventricular ADM is given.

To explore whether the early inhibition of RSNA was due to the arterial baroreceptor reflex after an increase in MAP, we used SAD rats. In SAD rats, the increases in MAP, HR, and the late phase of RSNA were elicited excessively, but the early decrease in RSNA was approximately the same as in the intact rats. Accordingly, this indicates that the early inhibition of RSNA is not mediated by the arterial baroreceptors.

The present finding that central ADM elicited changes in BP, HR, and RSNA in different temporal and directional manners along with reports that the ADM receptor gene is detected not only in the peripheral areas but also in several brain sites (18) led us to hypothesize that ADM in the CNS might have plural receptors and action mechanisms. Based on reports that vasopressor responses of central ADM are abolished by pretreatment with hCGRP-(837) in anesthetized rats (27) and that the COOH-terminal fragment hADM-(2252) might be an antagonist for vascular ADM receptors (5), we studied the effects of two different putative receptor antagonists, hCGRP-(837) and hADM-(2252), on the central ADM-induced responses. We observed that both antagonists significantly suppressed the central actions of ADM, indicating that ADM works through at least the ADM and CGRP receptors in the CNS. In addition, it has been previously reported that specific CGRP binding sites are abundantly identified in the human and rat CNS (9, 31); that intracerebroventricular administration of CGRP causes increases in BP, HR, and norepinephrine release (7, 21); and that ADM shares the same receptor with CGRP in the peripheral vasculature system (6, 10). Therefore, the inhibition of RSNA evoked by lower doses of ADM and tachycardia, in part, might be mediated through the ADM receptor alone. The increases in BP, HR, and RSNA via higher doses of ADM might depend chiefly on binding with the CGRP receptor, whereas the early inhibition of RSNA may not. Consequently, it is likely that at least two different receptors for ADM, which are activated by ADM dose dependently, are located in the brain and that ADM has a more potent affinity for the ADM receptor than for the CGRP receptor in the CNS.

Systemic ADM is considered to play some role in pathological conditions, such as hypertension, renal failure, congestive heart failure, and sepsis, all of which are associated with elevated circulating levels of ADM (11, 13, 16, 30). Recently, it has been reported that ir-ADM was detected in the incubated medium of carcinoma cells derived from the epithelium of the choroid plexus, and both ir-ADM concentrations in the medium and the expression levels of ADM mRNA in carcinoma cells were significantly increased by cytokines, which are produced in excess in certain disease conditions (28). Moreover, ir-ADM in the cerebrospinal fluid (CSF) of patients with neurological disease has been detected at approximately the same concentration as the serum level (29). Hence, central ADM may also play an important role in cardiovascular and sympathetic regulation through increased secretion into the CSF under pathological conditions as well as systemic ADM. Certainly, each antagonist alone did not affect basal parameters in the present study. However, it seems to be likely that the intracerebroventricular administration of antagonist alone induces significant changes, especially under endogenous ADM-stimulated or augmented conditions.

It is well known that the renal sympathetic nerves play important roles in the homeostasis of body fluids and the circulatory system; diuresis and natriuresis are elicited through the inhibition of RSNA in cases of hypertension or volume overload (3). It has been reported that peripheral ADM elicits diuresis and natriuresis through a direct renal action (4, 15, 23), and central ADM produces inhibitory effects on water intake, salt appetite, and arginine vasopressin (AVP) release (22, 25, 34). Therefore, central ADM-induced inhibition of RSNA observed in the present study may, at least in part, have amplified diuretic and natriuretic actions in the kidney induced by peripheral ADM.

Actually, we do not know the exact action sites of ADM in the CNS. However, as it has been shown that ADM-immunoreactive neurons are found in the paraventricular nucleus (PVN) and supraoptic nucleus (SON) of the hypothalamus (26, 32), and the electrical activity of neurons within the area postrema (AP) of the medulla are directly affected by ADM in brain slice preparations (1), the PVN, SON, and AP may be involved in the central action of ADM. Further studies are required to clarify the physiological and pathophysiological significance of central ADM, particularly concerning its regulatory and action mechanisms.

In conclusion, the present study provides the first evidence that central administration of ADM in conscious rats causes tachycardia and a decrease in RSNA without hypertension. These findings suggest that ADM may play some role as a neuropeptide in the central regulation of the heart and kidneys through at least the ADM and CGRP receptors.

Perspectives