INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

IN HEALTHY MAMMALS atrial natriuretic peptide (ANP) is synthesized, stored, and secreted most abundantly by cardiac atrial myocytes (38). Secretion occurs predominantly in response to stress on the atrial wall (17). In some settings, the peptide has a pronounced natriuretic effect, and this used to be taken as the hallmark of ANP action. Recent studies in ANP knockouts or transgenics that overexpress the peptide have revealed that ANP is not essential for normal salt balance, even on a high-salt diet (16). However, it exerts a persistent hypotensive effect in transgenic mice in whom increased plasma ANP levels, 5- to 8-fold above normal, are associated with a 25-mmHg decrease in mean arterial blood pressure (ABP) (36). Similarly, mice lacking the pro-ANP gene have a persistent hypertension (16). These blood pressure effects do not arise from ANP-mediated changes in sodium balance (16). They are due to changes in peripheral resistance (2, 23), although isolated peripheral resistance vessels show little response to direct application of ANP (6, 27).

Biologically active receptors for ANP (ANP-A or GC-A receptors) have been localized to large vessels (7, 37), but, with the exception of some vasodilatation in the renal artery and the rabbit facial vein (7), their activation appears to have little effect on whole body vascular resistance. Isolated resistance vessels also show no dilatation after locally applied ANP, except when they have been preconstricted by -adrenergic agents (6, 27). It is puzzling, therefore, that transgenic mice with a chronic increase in plasma ANP should show life-long hypotension that is due to decreased peripheral resistance (2) and that mice lacking the pro-ANP gene show chronically elevated mean ABP that is also due to changes in total peripheral resistance (23).

We recently sought to determine whether ANP influences either tissue content of the endothelial factors, nitric oxide, endothelin 1 or C-type natriuretic peptide (CNP), or their contribution to blood pressure maintenance. Our results showed no differences among ANP knockouts, ANP transgenics, or wild-type mice (22). Having, therefore, ruled out altered endothelial biology as a major explanation for chronic blood pressure changes in ANP knockouts or transgenics and having seen no differences in their basal plasma renin activity (8), we address in this report the question whether the vasopressin system may be significantly involved in maintenance of a hypertensive state in ANP knockouts.

Vasopressin is synthesized in the cell bodies of magnocellular neurons of the supraoptic and paraventricular nuclei of the hypothalamus. It is transported in axons that project to the posterior lobe of the pituitary. Its release is inhibited by neural input from atrial and arterial stretch receptors (11) and is promoted by several afferents from peripheral sensors or central nervous nuclei (11). Vasopressin release is also promoted by ₁-adrenergic agonists, such as noradrenaline secreted from fibers originating from cell groups in the ventrolateral medulla, or nicotinic agonists, such as acetylcholine from cholinergic neurons adjacent to the supraoptic nucleus (11). Of particular importance are the reports that show angiotensin II-mediated potentiation of vasopressin release as well as suppression of vasopressin release by angiotensin II antagonists (33). These observations are significant because of the known reciprocal relationship between ANP and angiotensin II in the regulation of blood pressure and sympathetic outflow (35) as well as the known inhibitory influence of ANP on vasopressin release (25, 30). Yamada et al. (42) have demonstrated a specific inhibitory action of ANP on angiotensin II-stimulated vasopressin release. These observations led us to hypothesize that the life-long hypertension in ANP knockout mice is caused, at least in part, by disinhibition of angiotensin II-mediated vasopressin release. The experiments described here were designed to answer three questions relevant to the hypertension of ANP-deleted mice. 1) Is lack of central ANP involved? 2) Is central angiotensin II involved? 3) Is peripheral vasopressin involved?

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animals. Experiments were conducted in anesthetized (Inactin, 100 mg/kg body wt) F₂ homozygous ANP gene-disrupted mice (/) and the corresponding wild type (+/+) littermates. They were 20-24 wk old and weighed between 20 and 35 g. The ANP gene-disruption model is fully described by John et al. (15). In the mutant (/) mice, plasma and atrial ANP levels are undetectable by radioimmunoassay, and ANP-specific atrial granules are absent (15). Our colonies were begun from breeding pairs of both types of mice. The genotype of each animal was confirmed by Southern blot analysis of EcoR I-digest genomic DNA extracted from tail tissue (15).

Surgical preparation. Once surgical-plane anesthesia was established, the mice were tracheotomized with polyethylene tubing (PE-240; pulled to appropriate diameter) and both the right carotid artery and right external jugular vein were cannulated with PE-50 tubing that had a short length of PE-10 glued to the vessel insertion end. Then the animal was carefully turned so as to rest on its stomach, was placed in a stereotaxic frame, and a small hole was drilled through the skull near bregma so that a 28-gauge stainless steel guide cannula could be epoxy cemented into place 0.2 mm posterior and 1 cm left laterally to bregma and lowered 2.5 mm into the brain. This placed the cannula tip into the left lateral ventricle at the site at which the interventricular foramen to the third ventricle is at its widest (9). Correct placement was ascertained at the end of each experiment by injection of 10 µl of methylene blue (1%). We accepted only animals in which the dye, visualized under a low-power microscope, remained confined to the brain ventricular system.

Protocols. Experimental protocols were begun 30 min after completion of all surgical preparations. Baseline carotid ABP was recorded for 10 min. Then a background intravenous injection was given, and 15 min later, the first of two intracerebroventricular injections (ICV-1) was given (Table 1), followed 5 min later by the second (ICV-2). Ten minutes after ICV-2, the intracerebroventricular injection of methylene blue was given and the experiment was terminated. There were five groups of mice. Each consisted of 10 (/) and 10 (+/+), and they were distinguished from one another by the different injections as shown in Table 1.

Drugs. Twenty-eight-amino acid rat ANP was purchased from Peninsula Laboratories (Belmont, CA). Sigma Chemical (St. Louis, MO) supplied the vasopressin antagonist [-mercapto-, -cyclopentamethylenepropionyl¹, O-Me-Tyr², Val⁴, Arg⁸]-vasopressin (Manning compound), and losartan, a nonpeptide antagonist of angiotensin II AT₁ receptors, was generously donated by DuPont-Merck Pharmaceutical.

Drug dosages. The losartan dosage (10 µg icv) was chosen on the basis of generally employed, maximally effective dosages (5, 10, 14). Intracerebroventricular dosages of ANP have ranged from 0.12 (1) to 5 µg (42). We chose 0.5 µg because it was the lowest dose that led to measurable decreases in ABP in the most susceptible animals (/). The dose of arginine vasopressin (AVP) V₁ antagonist was determined in pilot experiments. We chose the lowest dose that gave the greatest fall in mean ABP.

Data collection and analysis. ABP was recorded from the carotid artery cannula with an Electromedics microdisplacement transducer (MS20BA07ADS) attached to a BioPac data collection and analysis system (BioPac Systems, Goleta, CA). Heart rate was calculated by the BioPac from the pressure record. Baseline values were taken as the average over the initial 10-min period. The effects of intracerebroventricular injections on blood pressure are reported as changes after ICV-1 or ICV-2 and were calculated as follows. The initial value was taken as the average of the 60-s period preceding injection. The response value was taken as the average over a 60-s period that was centered around the respective peak response. The effects of intravenous injections are also reported as changes. The relevant periods were a 60-s interval immediately before ICV-1 (called "after" in Fig. 3) and an interval of equal duration immediately before the intravenous injection. Between-group comparisons were done by one-way ANOVA, followed by a Bonferroni-corrected t-test when post hoc comparisons were warranted. Adherence to the one-way ANOVA assumption of normally distributed values was tested by applying the Kolmogorov-Smirnov test, and the assumption of equal variances was tested by a Levene median test. Statistical significance was defined as P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

ANP-deficient mice were significantly hypertensive with respect to their wild-type littermates (112 ± 8 vs. 94 ± 8 mmHg; means ± SE) but showed no significant difference in basal heart rate (389 ± 20 vs. 428 ± 19 beats/min). Heart rate responses to any of the injections performed in the course of the experiments did not differ between (/) and (+/+) and are, therefore, not mentioned further. Intracerebroventricular ANP or losartan had significant hypotensive effects only in ANP-deleted (/) mice (Figs. 1 and 2). The onset of these effects was quick (<2 min) and persisted throughout the postinjection period. Both strains showed significant decreases in mean ABP after intravenous injection of vasopressin V₁-receptor antagonist (Fig. 3), suggesting involvement of vasopressin in the maintenance of ABP in both (+/+) and (/). However, contrary to our hypothesis, there was no significant difference between them (Fig. 3). Finally, intracerebroventricular injection of ANP caused no significant change in ABP when both central AT₁ and peripheral V₁ receptors had been inhibited by a preceding injection of the appropriate antagonist (Fig. 4, comparing groups 5 and 4 after ICV-2).

View larger version (8K): [in this window] [in a new window]   Fig. 1.   Change in mean arterial blood pressure (ABP) observed in F2 homozygous atrial natriuretic peptide (ANP) gene-disrupted mice (/) and corresponding wild type (+/+) after intracerebroventricular injection of physiological saline (5 µl) or ANP (500 ng in 5 µl). Data are shown as means (±SE) of a continuous 60-s record, centered about the peak response in each animal. Baseline was taken as mean of 60-s interval preceding injection; * P < 0.05. , change in; ns, not significant.

View larger version (8K): [in this window] [in a new window]   Fig. 2.   ABP observed in (+/+) and (/) after intracerebroventricular injection of physiological saline (5 µl) or AT1-receptor antagonist losartan. Means ± SE; * P < 0.05.

View larger version (8K): [in this window] [in a new window]   Fig. 3.   ABP observed in (+/+) and (/) after intravenous injection of physiological saline or vasopressin V1-receptor antagonist AVPX. Means ± SE; * P < 0.05.

View larger version (8K): [in this window] [in a new window]   Fig. 4.   ABP observed after intracerebroventricular injection of physiological saline or 500 ng ANP in (+/+) and (/) mice that had previously received both vasopressin V1-receptor antagonist [-mercapto-,-cyclopentamethylenepropionyl1,O-Me-Tyr2,Val4,Arg8]-vasopressin (Manning compound) and AT1-receptor antagonist losartan. Means ± SE; * P < 0.05.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The involvement of ANP in the central neuromodulation of cardiovascular function was first suggested by Thoren and colleagues (39) and has subsequently been supported by others who have demonstrated that ANP immunoreactivity and binding sites are found in brain structures closely involved in the regulation of ABP, particularly in hypothalamic nuclei, median eminence, septal areas, the AV3V region, and the circumventricular organs (13, 24, 31). This suggests that locally derived as well as plasma-borne natriuretic peptides could exert modulating influences on central cardiovascular regulatory sites. Our present findings (Fig. 1) confirm the report of others that central ANP fails to elicit significant depressor effects at low doses in normal animals (4, 32). On the other hand, the finding that hypotension can be produced in normals by intracerebroventricular doses as high as 5 µg (21) or by a dose of only 0.5 µg in ANP-depleted mice (Fig. 1) implies that ANP is involved in hypotensive responses under some circumstances. We speculate that low doses of exogenous ANP fail to show an additional effect whenever most central receptor sites are already occupied by native ANP. In the absence of native ANP, small amounts of exogenous peptide lead to hypotension (/) (Fig. 1). The mechanisms by which central ANP exerts its effects are unknown.

The left lateral ventricle is part of the cerebral ventricular system, and substances injected into it will eventually diffuse throughout the cerebrospinal fluid. Its most prominent neighbors are the thalamus and the caudate nucleus of the basal ganglia (28). Its neighbors of potential cardiovascular significance are the paraventricular nucleus, a major site of vasopressin synthesis (18), and two circumventricular organs, subfornical organ and organum vasculosum of the lamina terminalis, both of which are important loci for the modulation of drinking behavior by angiotensin II (41) as well as the modulation, by ANP, of angiotensin II-mediated neuronal excitation (12), water intake (19), or vasopressin release (25, 34, 42). In view of reports that central angiotensin II may be involved as an intermediary of central ANP actions (3, 32) and that ANP and angiotensin II show reciprocal brain expression in their respective receptors in genetic hypertension (29), we tested first whether (/) showed elevated central angiotensin II AT₁-receptor activation. These receptors are believed to be responsible for the hemodynamic effects of angiotensin II, and they predominate in the hypothalamus (26) and other brain areas with significant influence over cardiovascular control (20). Figure 2 shows that central AT₁ activation was a significant part of blood pressure maintenance in (/), but not (+/+) mice. However, enhanced peripheral vasopressin-receptor activation was not involved in (/) hypertension in these experiments (Fig. 3). Both strains showed significant decreases in ABP immediately after vasopressin V₁-receptor antagonism (Fig. 3), but, contrary to our hypothesis, (/) animals did not show a significantly greater hypotension after V₁ antagonist than did (+/+). Correct interpretation of the role of vasopressin in the hypertension of ANP knockouts may have been complicated by the use of anesthesia in our experiments. Thus, consistent with the demonstration by Yamada et al. (42) in conscious rats, conscious ANP (/) mice might have elevated plasma vasopressin levels. In addition, it is possible that the release of vasopressin that occurs with some anesthetic agents (40) might have been enhanced in ANP (+/+) compared with ANP (/). This potential clouding of the role of vasopressin can be removed only if the experiments could be conducted in conscious mice.

Finally, we tested whether intracerebroventricular ANP exerted a hypotensive effect above and beyond that which was mediated via central AT₁ receptors or peripheral V₁ receptors. ANP injected intracerebroventricularly into mice that had previously received a V₁ antagonist intravenously and an AT₁ antagonist intracerebroventricularly caused no significant fall in ABP in either (/) or (+/+) (Fig. 4).

In summary, the results of this study suggest that ABP in ANP-deleted mice is in part maintained by peripheral vasopressin, just as ABP in wild-type controls is partly maintained by that agent. The permanently elevated blood pressure in ANP knockouts is associated with increased central nervous angiotensin II AT₁-receptor activation. Its subsequent peripheral mediator is more likely to be enhanced sympathetic nervous activity (23) than elevated vasopressin.

Perspectives